Merge pull request 'adapter: support OpenAI streaming (SSE) in /v1/chat/completions' (#1 ) from feat/adapter-streaming into main

adapter: support OpenAI streaming (SSE) in /v1/chat/completions
OpenCode's @ai-sdk/openai-compatible sends stream:true and would render an empty response because the adapter always returned a single non-streaming chat.completion JSON body. Now when stream:true, emit OpenAI-compatible SSE chat.completion.chunk events (role, content, [DONE]) so streaming clients render text. Non-streaming path unchanged. Adds a local EchoServer test that exercises the streaming route end-to-end.
2026-09-12 12:51:34 +02:00 · 2026-09-12 10:50:47 +00:00 · 2026-09-11 18:44:45 +00:00 · 2026-09-11 15:36:15 +00:00 · 2026-09-10 14:35:28 +00:00 · 2026-09-10 07:21:42 +00:00
404 changed files with 1181 additions and 91273 deletions
@@ -1 +0,0 @@
-lib/elixir/test/elixir/fixtures/*.txt text eol=lf
@@ -1,17 +1,11 @@
-/.eunit/*
-/.release
-/lib/*/ebin/*
-/lib/*/tmp
-/lib/*/test/tmp
-/lib/elixir/src/elixir.app.src
-/lib/elixir/src/*_lexer.erl
-/lib/elixir/src/*_parser.erl
-/lib/elixir/test/ebin
-/deps/*
-/ebin
-/rel/elixir
+.formatter.exs
+/_build/
+/cover/
+/deps/
+/doc/
+/.fetch
 erl_crash.dump
-.dialyzer_plt
-.dialyzer.base_plt
-.*.swp
-docs
+*.ez
+n8n_openai_adapter-*.tar
+/tmp/
+/result
@@ -1,9 +0,0 @@
-language: erlang
-script: "make compile && make test"
-notifications:
-  irc: "irc.freenode.org#elixir-lang"
-  recipients:
-    - jose.valim@plataformatec.com.br
-    - yrashk@gmail.com
-otp_release:
-  - R16B
@@ -1,651 +0,0 @@
-# v0.11.0 (2013-11-02)
-
-* Enhancements
-  * [Code] Eval now returns variables from other contexts
-  * [Dict] Document and enforce all dicts use the match operator (`===`) when checking for keys
-  * [Enum] Add `Enum.slice/2` with a range
-  * [Enum] Document and enforce `Enum.member?/2` to use the match operator (`===`)
-  * [IEx] Split `IEx.Evaluator` from `IEx.Server` to allow custom evaluators
-  * [IEx] Add support for `IEx.pry` which halts a given process for inspection
-  * [IO] Add specs and allow some IO APIs to receive any data that implements `String.Chars`
-  * [Kernel] Improve stacktraces on command line interfaces
-  * [Kernel] Sigils can now handle balanced tokens as in `%s(f(o)o)`
-  * [Kernel] Emit warnings when an alias is not used
-  * [Macro] Add `Macro.pipe/3` and `Macro.unpipe/1` for building pipelines
-  * [Mix] Allow umbrella children to share dependencies between them
-  * [Mix] Allow mix to be escriptize'd
-  * [Mix] Speed mix projects compilation by relying on more manifests information
-  * [Protocol] Protocols now provide `impl_for/1` and `impl_for!/1` functions which receive a structure and returns its respective implementation, otherwise returns nil or an error
-  * [Set] Document and enforce all sets use the match operator (`===`) when checking for keys
-  * [String] Update to Unicode 6.3.0
-  * [String] Add `Enum.slice/2` with a range
-
-* Bug fixes
-  * [Exception] Ensure `defexception` fields can be set dynamically
-  * [Kernel] Guarantee aliases hygiene is respected when the current module name is not known upfront
-  * [Kernel] `Kernel.access/2` no longer flattens lists
-  * [Mix] Ensure cyclic dependencies are properly handled
-  * [String] Implement the extended grapheme cluster algorithm for `String` operations
-
-* Deprecations
-  * [Kernel] `pid_to_list/1`, `list_to_pid/1`, `binary_to_atom/2`, `binary_to_existing_atom/2` and `atom_to_binary/2` are deprecated in favor of their counterparts in the `:erlang` module
-  * [Kernel] `insert_elem/3` and `delete_elem/2` are deprecated in favor of `Tuple.insert_at/3` and `Tuple.delete_at/2`
-  * [Kernel] Use of `in` inside matches (as in `x in [1,2,3] -> x`) is deprecated in favor of the guard syntax (`x when x in [1,2,3]`)
-  * [Macro] `Macro.expand_all/2` is deprecated
-  * [Protocol] `@only` and `@except` in protocols are now deprecated
-  * [Protocol] Protocols no longer fallback to `Any` out of the box (this functionality needs to be explicitly enabled by setting `@fallback_to_any` to true)
-  * [String] `String.to_integer/1` and `String.to_float/1` are deprecated in favor of `Integer.parse/1` and `Float.parse/1`
-
-* Backwards incompatible changes
-  * [CLI] Reading `.elixirrc` has been dropped in favor of setting env vars
-  * [Kernel] `Kernel.access/2` now expects the second argument to be a compile time list
-  * [Kernel] `fn -> end` quoted expression is no longer wrapped in a `do` keyword
-  * [Kernel] Quoted variables from the same module must be explicitly shared. Previously, if a function returned `quote do: a = 1`, another function from the same module could access it as `quote do: a`. This has been fixed and the variables must be explicitly shared with `var!(a, __MODULE__)`
-  * [Mix] Umbrella apps now treat children apps as dependencies. This means all dependencies will be checked out in the umbrela `deps` directory. On upgrade, child apps need to point to the umbrella project by setting `deps_path: "../../deps_path", lockfile: "../../mix.lock"` in their project config
-  * [Process] `Process.group_leader/2` args have been reversed so the "subject" comes first
-  * [Protocol] Protocol no longer dispatches to `Number`, but to `Integer` and `Float`
-
-# v0.10.3 (2013-10-02)
-
-* Enhancements
-  * [Enum] Add `Enum.take_every/2`
-  * [IEx] IEx now respects signals sent from the Ctrl+G menu
-  * [Kernel] Allow documentation for types with `@typedoc`
-  * [Mix] Allow apps to be selected in umbrella projects
-  * [Record] Generated record functions `new` and `update` also take options with strings as keys
-  * [Stream] Add `Stream.unfold/1`
-
-* Bug fixes
-  * [Dict] Fix a bug when a HashDict was marked as equal when one was actually a subset of the other
-  * [EEx] Solve issue where `do` blocks inside templates were not properly aligned
-  * [ExUnit] Improve checks and have better error reports on poorly aligned doctests
-  * [Kernel] Fix handling of multiple heredocs on the same line
-  * [Kernel] Provide better error messages for match, guard and quoting errors
-  * [Kernel] Make `Kernel.raise/2` a macro to avoid messing up stacktraces
-  * [Kernel] Ensure `&()` works on quoted blocks with only one expression
-  * [Mix] Address an issue where a dependency was not compiled in the proper order when specified in different projects
-  * [Mix] Ensure `compile: false` is a valid mechanism for disabling the compilation of dependencies
-  * [Regex] Fix bug on `Regex.scan/3` when capturing groups and the regex has no groups
-  * [String] Fix a bug with `String.split/2` when given an empty pattern
-  * [Typespec] Guarantee typespecs error reports point to the proper line
-
-* Deprecations
-  * [Kernel] The previous partial application syntax (without the `&` operator) has now been deprecated
-  * [Regex] `Regex.captures/3` is deprecated in favor of `Regex.named_captures/3`
-  * [String] `String.valid_codepoint?/1` is deprecated in favor of pattern matching with `<<_ :: utf8 >>`
-
-* Backwards incompatible changes
-  * [IEx] The `r/0` helper has been removed as it caused surprising behaviour when many modules with dependencies were accumulated
-  * [Mix] `Mix.Version` was renamed to `Version`
-  * [Mix] `File.IteratorError` was renamed to `IO.StreamError`
-  * [Mix] `mix new` now defaults to the `--sup` option, use `--bare` to get the previous behaviour
-
-# v0.10.2 (2013-09-03)
-
-* Enhancements
-  * [CLI] Add `--verbose` to elixirc, which now is non-verbose by default
-  * [Dict] Add `Dict.Behaviour` as a convenience to create your own dictionaries
-  * [Enum] Add `Enum.split/2`, `Enum.reduce/2`, `Enum.flat_map/2`, `Enum.chunks/2`, `Enum.chunks/4`, `Enum.chunks_by/2`, `Enum.concat/1` and `Enum.concat/2`
-  * [Enum] Support negative indices in `Enum.at/fetch/fetch!`
-  * [ExUnit] Show failures on CLIFormatter as soon as they pop up
-  * [IEx] Allow for strings in `h` helper
-  * [IEx] Helpers `r` and `c` can handle erlang sources
-  * [Integer] Add `odd?/1` and `even?/1`
-  * [IO] Added support to specifying a number of bytes to stream to `IO.stream`, `IO.binstream`, `File.stream!` and `File.binstream!`
-  * [Kernel] Include file and line on error report for overriding an existing function/macro
-  * [Kernel] Convert external functions into quoted expressions. This allows record fields to contain functions as long as they point to an `&Mod.fun/arity`
-  * [Kernel] Allow `foo?` and `bar!` as valid variable names
-  * [List] Add `List.replace_at/3`
-  * [Macro] Improve printing of the access protocol on `Macro.to_string/1`
-  * [Macro] Add `Macro.to_string/2` to support annotations on the converted string
-  * [Mix] Automatically recompile a project if the Elixir version changes
-  * [Path] Add `Path.relative_to_cwd/2`
-  * [Regex] Allow erlang `re` options when compiling Elixir regexes
-  * [Stream] Add `Stream.concat/1`, `Stream.concat/2` and `Stream.flat_map/2`
-  * [String] Add regex pattern support to `String.replace/3`
-  * [String] Add `String.ljust/2`, `String.rjust/2`, `String.ljust/3` and `String.rjust/3`
-  * [URI] `URI.parse/1` supports IPv6 addresses
-
-* Bug fixes
-  * [Behaviour] Do not compile behaviour docs if docs are disabled on compilation
-  * [ExUnit] Doctests no longer eat too much space and provides detailed reports for poorly indented lines
-  * [File] Fix a bug where `File.touch(file, datetime)` was not setting the proper datetime when the file did not exist
-  * [Kernel] Limit `inspect` results to 50 items by default to avoid printing too much data
-  * [Kernel] Return a readable error on oversized atoms
-  * [Kernel] Allow functions ending with `?` or `!` to be captured
-  * [Kernel] Fix default shutdown of child supervisors to `:infinity`
-  * [Kernel] Fix regression when calling a function/macro ending with bang, followed by `do/end` blocks
-  * [List] Fix bug on `List.insert_at/3` that added the item at the wrong position for negative indexes
-  * [Macro] `Macro.escape/2` can now escape improper lists
-  * [Mix] Fix `Mix.Version` matching on pre-release info
-  * [Mix] Ensure `watch_exts` trigger full recompilation on change with `mix compile`
-  * [Mix] Fix regression on `mix clean --all`
-  * [String] `String.strip/2` now supports removing unicode characters
-  * [String] `String.slice/3` still returns the proper result when there is no length to be extracted
-  * [System] `System.get_env/0` now returns a list of tuples as previously advertised
-
-* Deprecations
-  * [Dict] `Dict.update/3` is deprecated in favor of `Dict.update!/3`
-  * [Enum] `Enum.min/2` and `Enum.max/2` are deprecated in favor of `Enum.min_by/2` and `Enum.max_by/2`
-  * [Enum] `Enum.join/2` and `Enum.map_join/3` with a char list are deprecated
-  * [IO] `IO.stream(device)` and `IO.binstream(device)` are deprecated in favor of `IO.stream(device, :line)` and `IO.binstream(device, :line)`
-  * [Kernel] `list_to_binary/1`, `binary_to_list/1` and `binary_to_list/3` are deprecated in favor of `String.from_char_list!/1` and `String.to_char_list!/1` for characters and `:binary.list_to_bin/1`, `:binary.bin_to_list/1` and `:binary.bin_to_list/3` for bytes
-  * [Kernel] `to_binary/1` is deprecated in favor of `to_string/1`
-  * [Kernel] Deprecate `def/4` and friends in favor of `def/2` with unquote and friends
-  * [Kernel] Deprecate `%b` and `%B` in favor of `%s` and `%S`
-  * [List] `List.concat/2` is deprecated in favor of `Enum.concat/2`
-  * [Macro] `Macro.unescape_binary/1` and `Macro.unescape_binary/2` are deprecated in favor of `Macro.unescape_string/1` and `Macro.unescape_string/2`
-  * [Mix] `:umbrella` option for umbrella paths has been deprecated in favor of `:in_umbrella`
-
-* Backwards incompatible changes
-  * [IO] IO functions now only accept iolists as arguments
-  * [Kernel] `Binary.Chars` was renamed to `String.Chars`
-  * [Kernel] The previous ambiguous import syntax `import :functions, Foo` was removed in favor of `import Foo, only: :functions`
-  * [OptionParser] `parse` and `parse_head` now returns a tuple with three elements instead of two
-
-# v0.10.1 (2013-08-03)
-
-* Enhancements
-  * [Behaviour] Add support for `defmacrocallback/1`
-  * [Enum] Add `Enum.shuffle/1`
-  * [ExUnit] The `:trace` option now also reports run time for each test
-  * [ExUnit] Add support for `:color` to enable/disable ANSI coloring
-  * [IEx] Add the `clear` helper to clear the screen.
-  * [Kernel] Add the capture operator `&`
-  * [Kernel] Add support for `GenFSM.Behaviour`
-  * [Kernel] Functions now points to the module and function they were defined when inspected
-  * [Kernel] A documentation attached to a function that is never defined now prints warnings
-  * [List] Add `List.keysort/2`
-  * [Mix] `:test_helper` project configuration did not affect `mix test` and was therefore removed. A `test/test_helper.exs` file is still necessary albeit it doesn't need to be automatically required in each test file
-  * [Mix] Add manifests for yecc, leex and Erlang compilers, making it easier to detect dependencies in between compilers and providing a more useful clean behaviour
-  * [Mix] `mix help` now outputs information about the default mix task
-  * [Mix] Add `--no-deps-check` option to `mix run`, `mix compile` and friends to not check dependency status
-  * [Mix] Add support for `MIX_GIT_FORCE_HTTPS` system environment that forces HTTPS for known providers, useful when the regular git port is blocked. This configuration does not affect the `mix.lock` results
-  * [Mix] Allow coverage tool to be pluggable via the `:test_coverage` configuration
-  * [Mix] Add `mix cmd` as a convenience to run a command recursively in child apps in an umbrella application
-  * [Mix] Support `umbrella: true` in dependencies as a convenience for setting up umbrella path deps
-  * [Mix] `mix run` now behaves closer to the `elixir` command and properly mangles the ARGV
-  * [String] Add `Regex.scan/3` now supports capturing groups
-  * [String] Add `String.reverse/1`
-
-* Bug fixes
-  * [Behaviour] Ensure callbacks are stored in the definition order
-  * [CLI] Speed up boot time on Elixir .bat files
-  * [IEx] Reduce cases where IEx parser can get stuck
-  * [Kernel] Improve error messages when the use of an operator has no effect
-  * [Kernel] Fix a bug where warnings were not being generated when imported macros conflicted with local functions or macros
-  * [Kernel] Document that `on_definition` can only be a function as it is evaluated inside the function context
-  * [Kernel] Ensure `%w` sigils with no interpolation are fully expanded at compile time
-  * [Mix] `mix deps.update`, `mix deps.clean` and `mix deps.unlock` no longer change all dependencies unless `--all` is given
-  * [Mix] Always run ` mix loadpaths` on `mix app.start`, even if `--no-compile` is given
-  * [OptionParser] Do not add boolean flags to the end result if they were not given
-  * [OptionParser] Do not parse non-boolean flags as booleans when true or false are given
-  * [OptionParser] Ensure `:keep` and `:integer`|`:float` can be given together as options
-  * [OptionParser] Ensure `--no-flag` sets `:flag` to false when `:flag` is a registered boolean switch
-
-* Deprecations
-  * [Kernel] `function(Mod.fun/arity)` and `function(fun/arity)` are deprecated in favor of `&Mod.fun/arity` and `&fun/arity`
-  * [Kernel] `function/3` is deprecated in favor of `Module.function/3`
-  * [Kernel] `Kernel.ParallelCompiler` now receives a set of callbacks instead of a single one
-  * [Mix] `:test_coverage` option now expect keywords arguments and the `--cover` flag is now treated as a boolean
-
-* Backwards incompatible changes
-  * [Regex] `Regex.scan/3` now always returns a list of lists, normalizing the result, instead of list with mixed lists and binaries
-  * [System] `System.halt/2` was removed since the current Erlang implementation of such function is bugged
-
-# v0.10.0 (2013-07-15)
-
-* Enhancements
-  * [ExUnit] Support `trace: true` option which gives detailed reporting on test runs
-  * [HashDict] Optimize `HashDict` to store pairs in a cons cell reducing storage per key by half
-  * [Kernel] Add pretty printing support for inspect
-  * [Kernel] Add document algebra library used as the foundation for pretty printing
-  * [Kernel] Add `defrecordp/3` that enables specifying the first element of the tuple
-  * [Kernel] Add the `Set` API and a hash based implementation via `HashSet`
-  * [Kernel] Add `Stream` as composable, lazy-enumerables
-  * [Mix] `mix archive` now includes the version of the generated archive
-  * [Mix] Mix now requires explicit dependency overriding to be given with `override: true`
-  * [Mix] Projects can now define an `:elixir` key to outline supported Elixir versions
-  * [Typespec] Improve error messages to contain file, line and the typespec itself
-
-* Bug fixes
-  * [CLI] Elixir can now run on Unix directories with `:` in its path
-  * [Kernel] `match?/2` does not leak variables to outer scope
-  * [Kernel] Keep `head|tail` format when splicing at the tail
-  * [Kernel] Ensure variables defined in the module body are not passed to callbacks
-  * [Mix] On dependencies conflict, show from where each source is coming from
-  * [Mix] Empty projects no longer leave empty ebin files on `mix compile`
-  * [Module] Calling `Module.register_attribute/3` no longer automatically changes it to persisted or accumulated
-
-* Deprecations
-  * [Enum] Receiving the index of iteration in `Enum.map/2` and `Enum.each/2` is deprecated in favor of `Stream.with_index/1`
-  * [File] `File.iterator/1` and `File.biniterator/1` are deprecated in favor of `IO.stream/1` and `IO.binstream/1`
-  * [File] `File.iterator!/2` and `File.biniterator!/2` are deprecated in favor of `File.stream!/2` and `File.binstream!/2`
-  * [Kernel] Deprecate recently added `quote binding: ...` in favor of the clearer `quote bind_quoted: ...`
-  * [Kernel] Deprecate `Kernel.float/1` in favor of a explicit conversion
-  * [Mix] Deprecate `mix run EXPR` in favor of `mix run -e EXPR`
-  * [Record] `Record.__index__/2` deprecated in favor of `Record.__record__(:index, key)`
-
-* Backwards incompatible changes
-  * [Kernel] The `Binary.Inspect` protocol has been renamed to `Inspect`
-  * [Kernel] Tighten up the grammar rules regarding parentheses omission, previously the examples below would compile but now they raise an error message:
-
-            do_something 1, is_list [], 3
-            [1, is_atom :foo, 3]
-
-  * [Module] Calling `Module.register_attribute/3` no longer automatically changes it to persisted or accumulated
-  * [Record] First element of a record via `defrecordp` is now the `defrecordp` name and no longer the current atom
-  * [URI] Remove custom URI parsers in favor of `URI.default_port/2`
-
-# v0.9.3 (2013-06-23)
-
-* Enhancements
-  * [File] Add `File.chgrp`, `File.chmod` and `File.chown`
-  * [Kernel] Add `--warnings-as-errors` to Elixir's compiler options
-  * [Kernel] Print warnings to stderr
-  * [Kernel] Warn on undefined module attributes
-  * [Kernel] Emit warning for `x in []` in guards
-  * [Kernel] Add `binding/0` and `binding/1` for retrieving bindings
-  * [Kernel] `quote` now allows a binding as an option
-  * [Macro] Add `Macro.expand_once/2` and `Macro.expand_all/2`
-  * [Mix] Implement `Mix.Version` for basic versioning semantics
-  * [Mix] Support creation and installation of archives (.ez files)
-  * [Mix] `github: ...` shortcut now uses the faster `git` schema instead of `https`
-  * [Record] Allow types to be given to `defrecordp`
-
-* Bug fixes
-  * [Kernel] The elixir executable on Windows now supports the same options as the UNIX one
-  * [Kernel] Improve error messages on default clauses clash
-  * [Kernel] `__MODULE__.Foo` now returns `Foo` when outside of a Module
-  * [Kernel] Improve error messages when default clauses from different definitions collide
-  * [Kernel] `^x` variables should always refer to the value before the expression
-  * [Kernel] Allow `(x, y) when z` in function clauses and try expressions
-  * [Mix] Mix now properly evaluates rebar scripts
-
-* Deprecations
-  * [Code] `Code.string_to_ast/1` has been deprecated in favor of `Code.string_to_quoted/1`
-  * [Macro] `Macro.to_binary/1` has been deprecated in favor of `Macro.to_string/1`
-  * [Typespec] Deprecate `(fun(...) -> ...)` in favor of `(... -> ...)`
-
-* Backwards incompatible changes
-  * [Bitwise] Precedence of operators used by the Bitwise module were changed, check `elixir_parser.yrl` for more information
-  * [File] `rm_rf` and `cp_r` now returns a tuple with three elements on failures
-  * [Kernel] The quoted representation for `->` clauses changed from a tuple with two elements to a tuple with three elements to support metadata
-  * [Kernel] Sigils now dispatch to `sigil_$` instead of `__$__` where `$` is the sigil caracter
-  * [Macro] `Macro.expand/2` now expands until final form. Although this is backwards incompatible, it is very likely you do not need to change your code, since expansion until its final form is recommended, particularly if you are expecting an atom out of it
-  * [Mix] No longer support beam files on `mix local`
-
-# v0.9.2 (2013-06-13)
-
-* Enhancements
-  * [ExUnit] `capture_io` now captures prompt by default
-  * [Mix] Automatically import git dependencies from Rebar
-  * [Mix] Support for dependencies directly from the umbrella application
-  * [Regex] Add `Regex.escape`
-  * [String] Add `String.contains?`
-  * [URI] Implement `Binary.Chars` (aka `to_binary`) for `URI.Info`
-
-* Bug fixes
-  * [HashDict] Ensure HashDict uses exact match throughout its implementation
-  * [IEx] Do not interpret ANSI codes in IEx results
-  * [IEx] Ensure `--cookie` is set before accessing remote shell
-  * [Kernel] Do not ignore nil when dispatching protocols to avoid infinite loops
-  * [Mix] Fix usage of shell expressions in `Mix.Shell.cmd`
-  * [Mix] Start the application by default on escripts
-
-* Deprecations
-  * [Regex] `Regex.index/2` is deprecated in favor `Regex.run/3`
-  * [Kernel] `super` no longer supports implicit arguments
-
-* Backwards incompatible changes
-  * [Kernel] The `=~` operator now returns true or false instead of an index
-
-# v0.9.1 (2013-05-30)
-
-* Enhancements
-  * [IEx] Limit the number of entries kept in history and allow it to be configured
-  * [Kernel] Add `String.start_with?` and `String.end_with?`
-  * [Typespec] Allow keywords, e.g. `[foo: integer, bar: boolean | module]`, in typespecs
-
-* Bug fixes
-  * [Dict] `Enum.to_list` and `Dict.to_list` now return the same results for dicts
-  * [IEx] Enable shell customization via the `IEx.Options` module
-  * [Kernel] Fix a bug where `unquote_splicing` did not work on the left side of a stab op
-  * [Kernel] Unused functions with cyclic dependencies are now also warned as unused
-  * [Mix] Fix a bug where `mix deps.get` was not retrieving nested dependencies
-  * [Record] Fix a bug where nested records cannot be defined
-  * [Record] Fix a bug where a record named Record cannot be defined
-
-# v0.9.0 (2013-05-23)
-
-* Enhancements
-  * [ExUnit] `ExUnit.CaptureIO` now accepts an input to be used during capture
-  * [IEx] Add support for .iex files that are loaded during shell's boot process
-  * [IEx] Add `import_file/1` helper
-
-* Backwards incompatible changes
-  * [Enum] `Enum.Iterator` was replaced by the more composable and functional `Enumerable` protocol which supports reductions
-  * [File] `File.iterator/1` and `File.biniterator/1` have been removed in favor of the safe `File.iterator!/1` and `File.biniterator!/1` ones
-  * [Kernel] Erlang R15 is no longer supported
-  * [Kernel] Elixir modules are now represented as `Elixir.ModuleName` (using `.` instead of `-` as separator)
-
-# v0.8.3 (2013-05-22)
-
-* Enhancements
-  * [CLI] Flags `-p` and `-pr` fails if pattern match no files
-  * [CLI] Support `--hidden` and `--cookie` flags for distributed Erlang
-  * [Enum] Add `Enum.to_list/1`, `Enum.member?/2`, `Enum.uniq/2`, `Enum.max/1`, `Enum.max/2`, `Enum.min/1` and `Enum.min/2`
-  * [ExUnit] Add `ExUnit.CaptureIO` for IO capturing during tests
-  * [ExUnit] Consider load time on ExUnit time reports
-  * [IEx] Support `ls` with colored output
-  * [IEx] Add `#iex:break` to break incomplete expressions
-  * [Kernel] Add `Enum.at`, `Enum.fetch` and `Enum.fetch!`
-  * [Kernel] Add `String.to_integer` and `String.to_float`
-  * [Kernel] Add `Dict.take`, `Dict.drop`, `Dict.split`, `Dict.pop` and `Dict.fetch!`
-  * [Kernel] Many optimizations for code compilation
-  * [Kernel] `in` can be used with right side expression outside guards
-  * [Kernel] Add `Node.get_cookie/0` and `Node.set_cookie/2`
-  * [Kernel] Add `__DIR__`
-  * [Kernel] Expand macros and attributes on quote, import, alias and require
-  * [Kernel] Improve warnings related to default arguments
-  * [Keyword] Add `Keyword.delete_first/2`
-  * [Mix] Add `local.rebar` to download a local copy of rebar, and change `deps.compile` to use it if needed
-  * [Mix] Support umbrella applications
-  * [Mix] Load beam files available at `MIX_PATH` on CLI usage
-  * [String] Add `String.valid?` and `String.valid_character?`
-
-* Bug fixes
-  * [ExUnit] Handle exit messages from in ExUnit
-  * [ExUnit] Failures on ExUnit's setup_all now invalidates all tests
-  * [Kernel] Ensure we don't splice keyword args unecessarily
-  * [Kernel] Private functions used by private macros no longer emit an unused warning
-  * [Kernel] Ensure Elixir won't trip on empty receive blocks
-  * [Kernel] `String.slice` now returns an empty string when out of range by 1
-  * [Mix] Generate manifest files after compilation to avoid depending on directory timestamps and to remove unused .beam files
-  * [Path] `Path.expand/2` now correctly expands `~` in the second argument
-  * [Regex] Fix badmatch with `Regex.captures(%r/(.)/g, "cat")`
-  * [URI] Downcase host and scheme and URIs
-
-* Deprecations
-  * [Code] `Code.eval` is deprecated in favor of `Code.eval_string`
-  * [Exception] `Exception.format_entry` is deprecated in favor of `Exception.format_stacktrace_entry`
-  * [ExUnit] `assert left inlist right` is deprecated in favor of `assert left in right`
-  * [IO] `IO.getb` is deprecated in favor of `IO.getn`
-  * [List] `List.member?/2` is deprecated in favor of `Enum.member?/2`
-  * [Kernel] `var_context` in quote was deprecated in favor of `context`
-  * [Kernel] `Enum.at!` and `Dict.get!` is deprecated in favor of `Enum.fetch!` and `Dict.fetch!`
-
-* Backwards incompatible changes
-  * [Dict] `List.Dict` was moved to `ListDict`
-  * [IO] `IO.gets`, `IO.getn` and friends now return binaries when reading from stdio
-  * [Kernel] Precedence of `|>` has changed to lower to support constructs like `1..5 |> Enum.to_list`
-  * [Mix] `mix escriptize` now receives arguments as binaries
-
-# v0.8.2 (2013-04-20)
-
-* Enhancements
-  * [ExUnit] Use ANSI escape codes in CLI output
-  * [ExUnit] Include suite run time on CLI results
-  * [ExUnit] Add support to doctests, allowing test cases to be generated from code samples
-  * [File] Add `File.ls` and `File.ls!`
-  * [IEx] Support `pwd` and `cd` helpers
-  * [Kernel] Better error reporting for invalid bitstring generators
-  * [Kernel] Improve meta-programming by allowing `unquote` on `def/2`, `defp/2`, `defmacro/2` and `defmacrop/2`
-  * [Kernel] Add support to R16B new functions: `insert_elem/3` and `delete_elem/2`
-  * [Kernel] Import conflicts are now lazily handled. If two modules import the same functions, it will fail only if the function is invoked
-  * [Mix] Support `--cover` on mix test and `test_coverage` on Mixfiles
-  * [Record] Each record now provides `Record.options` with the options supported by its `new` and `update` functions
-
-* Bug fixes
-  * [Binary] inspect no longer escapes standalone hash `#`
-  * [IEx] The `h` helper can now retrieve docs for special forms
-  * [Kernel] Record optimizations were not being triggered in functions inside the record module
-  * [Kernel] Aliases defined inside macros should be carried over
-  * [Kernel] Fix a bug where nested records could not use the Record[] syntax
-  * [Path] Fix a bug on `Path.expand` when expanding paths starting with `~`
-
-* Deprecations
-  * [Kernel] `setelem/3` is deprecated in favor of `set_elem/3`
-  * [Kernel] `function(:is_atom, 1)` is deprecated in favor of `function(is_atom/1)`
-
-* Backwards incompatible changes
-  * [Kernel] `unquote` now only applies to the closest quote. If your code contains a quote that contains another quote that calls unquote, it will no longer work. Use `Macro.escape` instead and pass your quoted contents up in steps, for example:
-
-            quote do
-              quote do: unquote(x)
-            end
-
-      should become:
-
-            quote do
-              unquote(Macro.escape(x))
-            end
-
-# v0.8.1 (2013-02-17)
-
-* Enhancements
-  * [ExUnit] Tests can now receive metadata set on setup/teardown callbacks
-  * [ExUnit] Add support to ExUnit.CaseTemplate to share callbacks in between test cases
-  * [IO] Add `IO.ANSI` to make it easy to write ANSI escape codes
-  * [Kernel] Better support for Unicode lists
-  * [Kernel] Reduce variables footprint in `case`/`receive` clauses
-  * [Kernel] Disable native compilation when on_load attributes is present to work around an Erlang bug
-  * [Macro] `Macro.expand` also considers macros from the current `__ENV__` module
-  * [Mix] Improve support for compilation of `.erl` files
-  * [Mix] Add support for compilation of `.yrl` and `.xrl` files
-  * [OptionParser] Switches are now overridden by default but can be kept in order if chosen
-  * [Typespec] Better error reporting for invalid typespecs
-
-* Bug fixes
-  * [Mix] Allow Mix projects to be generated with just one letter
-
-* Backwards incompatible changes
-  * [Kernel] `before_compile` and `after_compile` callbacks now receive the environment as first argument instead of the module
-
-* Deprecations
-  * [ExUnit] Explicitly defined test/setup/teardown functions are deprecated
-  * [Kernel] Tidy up and clean `quote` API
-  * [Kernel] Old `:local.(args)` syntax is deprecated
-  * [Process] `Process.self` is deprecated in favor `Kernel.self`
-
-# v0.8.0 (2013-01-28)
-
-* Enhancements
-  * [Binary] Support `<< "string" :: utf8 >>` as in Erlang
-  * [Binary] Support `\a` escape character in binaries
-  * [Binary] Support syntax shortcut for specifying size in bit syntax
-  * [CLI] Support `--app` option to start an application and its dependencies
-  * [Dict] Support `put_new` in `Dict` and `Keyword`
-  * [Dict] Add `ListDict` and a faster `HashDict` implementation
-  * [ExUnit] ExUnit now supports multiple runs in the same process
-  * [ExUnit] Failures in ExUnit now shows a tailored stacktrace
-  * [ExUnit] Introduce `ExUnit.ExpectationError` to provide better error messages
-  * [Kernel] Introduce `Application.Behaviour` to define application module callbacks
-  * [Kernel] Introduce `Supervisor.Behaviour` to define supervisors callbacks
-  * [Kernel] More optimizations were added to Record handling
-  * [Kernel] `?\x` and `?\` are now supported ways to retrieve a codepoint
-  * [Kernel] Octal numbers can now be defined as `0777`
-  * [Kernel] Improve macros hygiene regarding variables, aliases and imports
-  * [Mix] Mix now starts the current application before run, iex, test and friends
-  * [Mix] Mix now provides basic support for compiling `.erl` files
-  * [Mix] `mix escriptize` only generates escript if necessary and accept `--force` and `--no-compile` as options
-  * [Path] Introduce `Path` module to hold filesystem paths related functions
-  * [String] Add `String.capitalize` and `String.slice`
-  * [System] Add `System.tmp_dir`, `System.cwd` and `System.user_home`
-
-* Bug fixes
-  * [Kernel] `import` with `only` accepts functions starting with underscore
-  * [String] `String.first` and `String.last` return nil for empty binaries
-  * [String] `String.rstrip` and `String.lstrip` now verify if argument is a binary
-  * [Typespec] Support `...` inside typespec's lists
-
-* Backwards incompatible changes
-  * [Kernel] The AST now allows metadata to be attached to each node. This means the second item in the AST is no longer an integer (representing the line), but a keywords list. Code that relies on the line information from AST or that manually generate AST nodes need to be properly updated
-
-* Deprecations
-  * [Dict] Deprecate `Binary.Dict` and `OrdDict` in favor of `HashDict` and `ListDict`
-  * [File] Deprecate path related functions in favor of the module `Path`
-  * [Kernel] The `/>` operator has been deprecated in favor of `|>`
-  * [Mix] `Mix.Project.sources` is deprecated in favor of `Mix.Project.config_files`
-  * [Mix] `mix iex` is no longer functional, please use `iex -S mix`
-  * [OptionParser] `:flags` option was deprecated in favor of `:switches` to support many types
-
-# v0.7.2 (2012-12-04)
-
-* Enhancements
-  * [CLI] `--debug-info` is now true by default
-  * [ExUnit] Make ExUnit exit happen in two steps allowing developers to add custom `at_exit` hooks
-  * [IEx] Many improvements to helpers functions `h/1`, `s/1` and others
-  * [Kernel] Functions defined with `fn` can now handle many clauses
-  * [Kernel] Raise an error if clauses with different arities are defined in the same function
-  * [Kernel] `function` macro now accepts arguments in `M.f/a` and `f/a` formats
-  * [Macro] Improvements to `Macro.to_binary`
-  * [Mix] Mix now echoes the output as it comes when executing external commands such as git or rebar
-  * [Mix] Mix now validates `application` callback's values
-  * [Record] Record accessors are now optimized and can be up to 6x faster in some cases
-  * [String] Support `\xXX` and `\x{HEX}` escape sequences in strings, char lists and regexes
-
-* Bug fixes
-  * [Bootstrap] Compiling Elixir source no longer fails if environment variables contain utf-8 entries
-  * [IEx] IEx will now wait for all command line options to be processed before starting
-  * [Kernel] Ensure proper stacktraces when showing deprecations
-
-* Deprecations
-  * [Enum] `Enum.qsort` is deprecated in favor of `Enum.sort`
-  * [List] `List.sort` and `List.uniq` have been deprecated in favor of their `Enum` counterparts
-  * [Record] Default-based generated functions are deprecated
-  * [Typespec] Enhancements and deprecations to the `@spec/@callback` and the fun type syntax
-
-# v0.7.1 (2012-11-18)
-
-* Enhancements
-  * [IEx] Only show documented functions and also show docs for default generated functions
-  * [IO] Add `IO.binread`, `IO.binwrite` and `IO.binreadline` to handle raw binary file operations
-  * [ExUnit] Add support for user configuration at `HOME/.ex_unit.exs`
-  * [ExUnit] Add support for custom formatters via a well-defined behaviour
-  * [Kernel] Add support for `defrecordp`
-  * [Kernel] Improved dialyzer support
-  * [Kernel] Improved error messages when creating functions with aliases names
-  * [Mix] Improve SCM behaviour to allow more robust integration
-  * [Mix] Changing deps information on `mix.exs` forces users to fetch new dependencies
-  * [Mix] Support (parallel) requires on mix run
-  * [Mix] Support `-q` when running tests to compile only changed files
-  * [String] Support `String.downcase` and `String.upcase` according to Unicode 6.2.0
-  * [String] Add support for graphemes in `String.length`, `String.at` and others
-  * [Typespec] Support `@opaque` as attribute
-  * [Typespec] Define a default type `t` for protocols and records
-  * [Typespec] Add support for the access protocol in typespecs
-
-* Bug fixes
-  * [Kernel] Fix an issue where variables inside clauses remained unassigned
-  * [Kernel] Ensure `defoverridable` functions can be referred in many clauses
-  * [Kernel] Allow keywords as function names when following a dot (useful when integrating with erlang libraries)
-  * [File] File is opened by default on binary mode instead of utf-8
-
-* Deprecations
-  * [Behaviour] `defcallback/1` is deprecated in favor of `defcallback/2` which matches erlang `@callbacks`
-  * [Enum] `Enum.times` is deprecated in favor of using ranges
-  * [System] `halt` moved to `System` module
-
-# v0.7.0 (2012-10-20)
-
-* Enhancements
-  * [Behaviour] Add Behaviour with a simple callback DSL to define callbacks
-  * [Binary] Add a Dict binary that converts its keys to binaries on insertion
-  * [Binary] Optimize `Binary.Inspect` and improve inspect for floats
-  * [CLI] Support `--detached` option
-  * [Code] `Code.string_to_ast` supports `:existing_atoms_only` as an option in order to guarantee no new atoms is generated when parsing the code
-  * [EEx] Support `<%%` and `<%#` tags
-  * [ExUnit] Support `after_spawn` callbacks which are invoked after each process is spawned
-  * [ExUnit] Support context data in `setup_all`, `setup`, `teardown` and `teardown_all` callbacks
-  * [IEx] Support `after_spawn` callbacks which are invoked after each process is spawned
-  * [Kernel] Better error messages when invalid options are given to `import`, `alias` or `require`
-  * [Kernel] Allow partial application on literals, for example: `{ &1, &2 }` to build tuples or `[&1|&2]` to build cons cells
-  * [Kernel] Added `integer_to_binary` and `binary_to_integer`
-  * [Kernel] Added `float_to_binary` and `binary_to_float`
-  * [Kernel] Many improvements to `unquote` and `unquote_splicing`. For example, `unquote(foo).unquote(bar)(args)` is supported and no longer need to be written via `apply`
-  * [Keyword] Keyword list is no longer ordered according to Erlang terms but the order in which they are specified
-  * [List] Add `List.keyreplace` and `List.keystore`
-  * [Macro]  Support `Macro.safe_term` which returns `:ok` if an expression does not execute code and is made only of raw data types
-  * [Mix] Add support for environments - the current environment can be set via `MIX_ENV`
-  * [Mix] Add support for handling and fetching dependencies' dependencies
-  * [Module] Support module creation via `Module.create`
-  * [Range] Support decreasing ranges
-  * [Record] Improvements to the Record API, added `Record.defmacros`
-  * [Regex] Add `:return` option to `Regex.run` and `Regex.scan`
-  * [String] Add a String module responsible for handling UTf-8 binaries
-
-* Bug fixes
-  * [File] `File.cp` and `File.cp_r` now preserves the file's mode
-  * [IEx] Fix a bug where printing to `:stdio` on `IEx` was causing it to hang
-  * [Macro] Fix a bug where quoted expressions were not behaving the same as their non-quoted counterparts
-  * [Mix] `mix deps.get [DEPS]` now only gets the specified dependencies
-  * [Mix] Mix now exits with status 1 in case of failures
-  * [Protocol] Avoid false positives on protocol dispatch (a bug caused the dispatch to be triggered to an invalid protocol)
-
-* Backwards incompatible changes
-  * [ExUnit] `setup` and `teardown` callbacks now receives the test name as second argument
-  * [Kernel] Raw function definition with `def/4`, `defp/4`, `defmacro/4`, `defmacrop/4` now evaluates all arguments. The previous behaviour was accidental and did not properly evaluate all arguments
-  * [Kernel] Change tuple-related (`elem` and `setelem`), Enum functions (`find_index`, `nth!` and `times`) and List functions (List.key*) to zero-index
-
-* Deprecations
-  * [Code] `Code.require_file` and `Code.load_file` now expect the full name as argument
-  * [Enum] `List.reverse/1` and `List.zip/2` were moved to `Enum`
-  * [GenServer] Rename `GenServer.Behavior` to `GenServer.Behaviour`
-  * [Kernel] Bitstring syntax now uses `::` instead of `|`
-  * [Kernel] `Erlang.` syntax is deprecated in favor of simply using atoms
-  * [Module] `Module.read_attribute` and `Module.add_attribute` deprecated in favor of `Module.get_attribute` and `Module.put_attribute` which mimics Dict API
-
-# v0.6.0 (2012-08-01)
-
-* Backwards incompatible changes
-  * [Kernel] Compile files now follow `Elixir-ModuleName` convention to solve issues with Erlang embedded mode. This removes the `__MAIN__` pseudo-variable as modules are now located inside `Elixir` namespace
-  * [Kernel] `__using__` callback triggered by `use` now receives just one argument. Caller information can be accessed via macros using `__CALLER__`
-  * [Kernel] Comprehensions syntax changed to be more compatible with Erlang behavior
-  * [Kernel] loop and recur are removed in favor of recursion with named functions
-  * [Module] Removed data functions in favor of unifying the attributes API
-
-* Deprecations
-  * [Access] The semantics of the access protocol were reduced from a broad query API to simple data structure key-based access
-  * [ExUnit] Some assertions are deprecated in favor of simply using `assert()`
-  * [File] `File.read_info` is deprecated in favor of `File.stat`
-  * [IO] `IO.print` is deprecated in favor of `IO.write`
-  * [Kernel] Deprecate `__LINE__` and `__FUNCTION__` in favor of `__ENV__.line` and `__ENV__.function`
-  * [Kernel] Deprecate `in_guard` in favor of `__CALLER__.in_guard?`
-  * [Kernel] `refer` is deprecated in favor of `alias`
-  * [Module] `Module.add_compile_callback(module, target, callback)` is deprecated in favor of `Module.put_attribute(module, :before_compile, { target, callback })`
-  * [Module] `Module.function_defined?` is deprecated in favor of `Module.defines?`
-  * [Module] `Module.defined_functions` is deprecated in favor of `Module.definitions_in`
-
-* Enhancements
-  * [Enum] Enhance Enum protocol to support `Enum.count`
-  * [Enum] Optimize functions when a list is given as collection
-  * [Enum] Add `find_index`, `nth!` and others
-  * [ExUnit] Support setup and teardown callbacks
-  * [IEx] IEx now provides autocomplete if the OS supports tty
-  * [IEx] IEx now supports remsh
-  * [IEx] Elixir now defaults to compile with documentation and `d` can be used in IEx to print modules and functions documentation
-  * [IEx] Functions `c` and `m` are available in IEx to compile and print available module information. Functions `h` and `v` are available to show history and print previous commands values
-  * [IO/File] Many improvements to `File` and `IO` modules
-  * [Kernel] Operator `!` is now allowed in guard clauses
-  * [Kernel] Introduce operator `=~` for regular expression matches
-  * [Kernel] Compiled docs now include the function signature
-  * [Kernel] `defmodule` do not start a new variable scope, this improves meta-programming capabilities
-  * [Kernel] quote special form now supports line and unquote as options
-  * [Kernel] Document the macro `@` and allow attributes to be read inside functions
-  * [Kernel] Add support to the `%R` sigil. The same as `%r`, but without interpolation or escaping. Both implementations were also optimized to generate the regex at compilation time
-  * [Kernel] Add `__ENV__` which returns a `Macro.Env` record with information about the compilation environment
-  * [Kernel] Add `__CALLER__` inside macros which returns a `Macro.Env` record with information about the calling site
-  * [Macro] Add `Macro.expand`, useful for debugging what a macro expands to
-  * [Mix] First Mix public release
-  * [Module] Add support to `@before_compile` and `@after_compile` callbacks. The first receives the module name while the latter receives the module name and its object code
-  * [OptionParser] Make OptionParser public, add support to flags and improved switch parsing
-  * [Range] Add a Range module with support to `in` operator (`x in 1..3`) and iterators
-  * [Record] Allow `Record[_: value]` to set a default value to all records fields, as in Erlang
-  * [Record] Records now provide a `to_keywords` function
-  * [Regex] Back references are now properly supported
-  * [System] Add `System.find_executable`
-
-# v0.5.0 (2012-05-24)
-
-* First official release
@@ -1,255 +0,0 @@
-# Contributing to Elixir
-
-Please take a moment to review this document in order to make the contribution
-process easy and effective for everyone involved!
-
-## Using the issue tracker
-
-Use the issues tracker for:
-
-* [bug reports](#bugs-reports)
-* [submitting pull requests](#pull-requests)
-
-Please **do not** use the issues tracker for personal support requests nor feature requests. Support requests should be send to:
-
-* [the elixir-talk mailing list](http://groups.google.com/group/elixir-lang-talk)
-* [Stack Overflow](http://stackoverflow.com/questions/ask?tags=elixir)
-* [#elixir-lang](irc://chat.freenode.net/elixir-lang)
-
-Feature requests can be discussed on [the elixir-core mailing list](http://groups.google.com/group/elixir-lang-core).
-
-We do our best to keep the issues tracker tidy and organized, making it useful
-for everyone. For example, we classify open issues per application and perceived
-difficulty of the issue, making it easier for developers to
-[contribute to Elixir](#contributing).
-
-## Bug reports
-
-A bug is a _demonstrable problem_ that is caused by the code in the repository.
-Good bug reports are extremely helpful - thank you!
-
-Guidelines for bug reports:
-
-1. **Use the GitHub issue search** &mdash; check if the issue has already been
-   reported.
-
-2. **Check if the issue has been fixed** &mdash; try to reproduce it using the
-   latest `master` or development branch in the repository.
-
-3. **Isolate the problem** &mdash; ideally create a reduced test
-   case.
-
-A good bug report shouldn't leave others needing to chase you up for more
-information. Please try to be as detailed as possible in your report. What is
-your environment? What steps will reproduce the issue? What version of Erlang
-and Elixir experience the problem? What would you expect to be the outcome?
-All these details will help people to fix any potential bugs.
-
-Example:
-
-> Short and descriptive example bug report title
->
-> A summary of the issue and the environment in which it occurs. If suitable,
-> include the steps required to reproduce the bug.
->
-> 1. This is the first step
-> 2. This is the second step
-> 3. Further steps, etc.
->
-> `<url>` - a link to the reduced test case (e.g. a GitHub Gist)
->
-> Any other information you want to share that is relevant to the issue being
-> reported. This might include the lines of code that you have identified as
-> causing the bug, and potential solutions (and your opinions on their
-> merits).
-
-## Feature requests
-
-Feature requests are welcome and should be discussed on [the elixir-core mailing list](http://groups.google.com/group/elixir-lang-core). But take a moment to find
-out whether your idea fits with the scope and aims of the project. It's up to *you*
-to make a strong case to convince the project's developers of the merits of this
-feature. Please provide as much detail and context as possible.
-
-## Contributing
-
-We incentivate everyone to contribute to Elixir and help us tackle
-existing issues! To do so, there are a few things you need to know
-about the code. First, Elixir code is divided in applications inside
-the `lib` folder:
-
-* `elixir` - Contains Elixir's kernel and stdlib
-
-* `eex` - Template engine that allows you to embed Elixir
-
-* `ex_unit` - Simple test framework that ships with Elixir
-
-* `iex` — IEx, Elixir's interactive shell
-
-* `mix` — Elixir's build tool
-
-You can run all tests in the root directory with `make test` and you can
-also run tests for a specific framework `make test_#{NAME}`, for example,
-`make test_ex_unit`.
-
-From time to time, your tests may fail in an existing Elixir checkout and
-may require a clean start by running `make clean compile`. You can always
-check [the official build status on Travis-CI](https://travis-ci.org/elixir-lang/elixir).
-
-With tests running and passing, you are ready to contribute to Elixir and
-send your pull requests.
-
-## Contributing Documentation
-
-Code documentation (`@doc`, `@moduledoc`, `@typedoc`) has a special convention:
-the first paragraph is considered to be a short summary.
-
-For functions, macros and callbacks say what it will do. For example write
-something like:
-
-```elixir
-@doc """
-Return only those elements for which `fun` is true.
-
-...
-"""
-def filter(collection, fun) ...
-```
-
-For modules, records, protocols and types say what it is. For example write
-something like:
-
-```elixir
-defrecord File.Stat, [...] do
-  @moduledoc """
-  Information about a file.
-
-  ...
-  """
-end
-```
-
-Keep in mind that the first paragraph might show up in a summary somewhere, long
-texts in the first paragraph create very ugly summaries. As a rule of thumb
-anything longer than 80 characters is too long.
-
-Try to keep unneccesary details out of the first paragraph, it's only there to
-give a user a quick idea of what the documented "thing" does/is. The rest of the
-documentation string can contain the details, for example when a value and when
-`nil` is returned.
-
-If possible include examples, preferably in a form that works with doctests. For
-example:
-
-```elixir
-@doc """
-Return only those elements for which `fun` is true.
-
-## Examples
-
-    iex> Enum.filter([1, 2, 3], fn(x) -> rem(x, 2) == 0 end)
-    [2]
-
-"""
-def filter(collection, fun) ...
-```
-
-This makes it easy to test the examples so that they don't go stale and examples
-are often a great help in explaining what a function does. 
-
-## Pull requests
-
-Good pull requests - patches, improvements, new features - are a fantastic
-help. They should remain focused in scope and avoid containing unrelated
-commits.
-
-**IMPORTANT**: By submitting a patch, you agree that your work will be
-licensed under the license used by the project.
-
-If you have any significant pull request in mind (e.g. implementing features,
-refactoring code, porting to a different language), **please ask first**
-otherwise you risk spending a lot of time working on something that the
-project's developers might not want to merge into the project.
-
-Please adhere to the coding conventions in the project (indentation,
-accurate comments, etc.) and don't forget to add your own tests and
-documentation. When working with git, we recommend the following process
-in order to craft an excellent pull request:
-
-1. [Fork](http://help.github.com/fork-a-repo/) the project, clone your fork,
-   and configure the remotes:
-
-   ```bash
-   # Clone your fork of the repo into the current directory
-   git clone https://github.com/<your-username>/elixir
-   # Navigate to the newly cloned directory
-   cd elixir
-   # Assign the original repo to a remote called "upstream"
-   git remote add upstream https://github.com/elixir-lang/elixir
-   ```
-
-2. If you cloned a while ago, get the latest changes from upstream:
-
-   ```bash
-   git checkout master
-   git pull upstream master
-   ```
-
-3. Create a new topic branch (off of `master`) to contain your feature, change,
-   or fix.
-
-   **IMPORTANT**: Making changes in `master` is discouraged. You should always
-   keep your local `master` in sync with upstream `master` and make your
-   changes in topic branches.
-
-   ```bash
-   git checkout -b <topic-branch-name>
-   ```
-
-4. Commit your changes in logical chunks. Keep your commit messages organized,
-   with a short description in the first line and more detailed information on
-   the following lines. Feel free to use Git's
-   [interactive rebase](https://help.github.com/articles/interactive-rebase)
-   feature to tidy up your commits before making them public.
-
-5. Make sure all the tests are still passing.
-
-   ```bash
-   make test
-   ```
-
-   This command will compile the code in your branch and use that
-   version of Elixir to run the tests. This is needed to ensure your changes can
-   pass all the tests.
-
-6. Push your topic branch up to your fork:
-
-   ```bash
-   git push origin <topic-branch-name>
-   ```
-
-7. [Open a Pull Request](https://help.github.com/articles/using-pull-requests/)
-    with a clear title and description.
-
-8. If you haven't updated your pull request for a while, you should consider
-   rebasing on master and resolving any conflicts.
-
-   **IMPORTANT**: _Never ever_ merge upstream `master` into your branches. You
-   should always `git rebase` on `master` to bring your changes up to date when
-   necessary.
-
-   ```bash
-   git checkout master
-   git pull upstream master
-   git checkout <your-topic-branch>
-   git rebase master
-   ```
-
-We have saved some excellent pull requests we have received in the past in case
-you are looking for some examples:
-
-* https://github.com/elixir-lang/elixir/pull/992
-* https://github.com/elixir-lang/elixir/pull/1041
-* https://github.com/elixir-lang/elixir/pull/1058
-* https://github.com/elixir-lang/elixir/pull/1059
-
-Thank you for your contributions!
@@ -1,8 +0,0 @@
-LEGAL NOTICE INFORMATION
------------------------
-
-All the files in this distribution are covered under either Elixir's
-license (see the file LICENSE) except the files mentioned below that
-contains sections that are under Erlang's License (EPL):
-
-lib/elixir/src/elixir_parser.erl (generated by build scripts)
@@ -1,13 +0,0 @@
-Copyright 2012-2013 Plataformatec.
-
-  Licensed under the Apache License, Version 2.0 (the "License");
-  you may not use this file except in compliance with the License.
-  You may obtain a copy of the License at
-
-      http://www.apache.org/licenses/LICENSE-2.0
-
-  Unless required by applicable law or agreed to in writing, software
-  distributed under the License is distributed on an "AS IS" BASIS,
-  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-  See the License for the specific language governing permissions and
-  limitations under the License.
@@ -1,164 +0,0 @@
-REBAR := "$(shell echo `pwd`/rebar)"
-ELIXIRC := bin/elixirc --verbose --ignore-module-conflict $(ELIXIRC_OPTS)
-ERLC := erlc -I lib/elixir/include
-ERL := erl -I lib/elixir/include -noshell -pa lib/elixir/ebin
-VERSION := $(strip $(shell cat VERSION))
-Q := @
-PREFIX := /usr/local
-LIBDIR := lib
-INSTALL = install
-INSTALL_DIR = $(INSTALL) -m755 -d
-INSTALL_DATA = $(INSTALL) -m644
-INSTALL_PROGRAM = $(INSTALL) -m755
-
-.PHONY: install compile erlang elixir dialyze test clean docs release_docs release_zip release_erl
-.NOTPARALLEL: compile
-
-#==> Templates
-
-define APP_TEMPLATE
-$(1): lib/$(1)/ebin/Elixir.$(2).beam lib/$(1)/ebin/$(1).app
-
-lib/$(1)/ebin/$(1).app: lib/$(1)/mix.exs
-	$(Q) cd lib/$(1) && ../../bin/elixir -e "Mix.Server.start_link(:dev)" -r mix.exs -e "Mix.Task.run('compile.app')"
-
-lib/$(1)/ebin/Elixir.$(2).beam: $(wildcard lib/$(1)/lib/*.ex) $(wildcard lib/$(1)/lib/*/*.ex) $(wildcard lib/$(1)/lib/*/*/*.ex)
-	@ echo "==> $(1) (compile)"
-	@ rm -rf lib/$(1)/ebin
-	$(Q) $$(ELIXIRC) "lib/$(1)/lib/**/*.ex" -o lib/$(1)/ebin
-
-test_$(1): $(1)
-	@ echo "==> $(1) (exunit)"
-	$(Q) cd lib/$(1) && ../../bin/elixir -r "test/test_helper.exs" -pr "test/**/*_test.exs";
-endef
-
-#==> Compilation tasks
-
-KERNEL:=lib/elixir/ebin/Elixir.Kernel.beam
-UNICODE:=lib/elixir/ebin/Elixir.String.Unicode.beam
-
-default: compile
-
-compile: lib/elixir/src/elixir.app.src erlang elixir
-
-lib/elixir/src/elixir.app.src: src/elixir.app.src
-	$(Q) rm -rf lib/elixir/src/elixir.app.src
-	$(Q) cp src/elixir.app.src lib/elixir/src/elixir.app.src
-
-erlang:
-	$(Q) cd lib/elixir && $(REBAR) compile
-
-# Since Mix depends on EEx and EEx depends on
-# Mix, we first compile EEx without the .app
-# file, then mix and then compile eex fully
-elixir: kernel lib/eex/ebin/Elixir.EEx.beam mix ex_unit eex iex
-
-kernel: $(KERNEL) VERSION
-$(KERNEL): lib/elixir/lib/*.ex lib/elixir/lib/*/*.ex
-	$(Q) if [ ! -f $(KERNEL) ]; then                       \
-		echo "==> bootstrap (compile)";                 \
-		$(ERL) -s elixir_compiler core -s erlang halt;  \
-	fi
-	@ echo "==> kernel (compile)";
-	$(Q) $(ELIXIRC) "lib/elixir/lib/**/*.ex" -o lib/elixir/ebin;
-	$(Q) $(MAKE) unicode
-	$(Q) rm -rf lib/elixir/ebin/elixir.app
-	$(Q) cd lib/elixir && $(REBAR) compile
-
-unicode: $(UNICODE)
-$(UNICODE): lib/elixir/priv/unicode.ex lib/elixir/priv/UnicodeData.txt lib/elixir/priv/GraphemeBreakProperty.txt
-	@ echo "==> unicode (compile)";
-	@ echo "This step can take up to a minute to compile in order to embed the Unicode database"
-	$(Q) $(ELIXIRC) lib/elixir/priv/unicode.ex -o lib/elixir/ebin --no-debug-info;
-
-$(eval $(call APP_TEMPLATE,ex_unit,ExUnit))
-$(eval $(call APP_TEMPLATE,eex,EEx))
-$(eval $(call APP_TEMPLATE,mix,Mix))
-$(eval $(call APP_TEMPLATE,iex,IEx))
-
-install: compile
-	@ echo "==> elixir (install)"
-	for dir in lib/*; do \
-		$(INSTALL_DIR) "$(DESTDIR)$(PREFIX)/$(LIBDIR)/elixir/$$dir/ebin"; \
-		$(INSTALL_DATA) $$dir/ebin/* "$(DESTDIR)$(PREFIX)/$(LIBDIR)/elixir/$$dir/ebin"; \
-	done
-	$(INSTALL_DIR) "$(DESTDIR)$(PREFIX)/$(LIBDIR)/elixir/bin"
-	$(INSTALL_PROGRAM) $(filter-out %.bat, $(wildcard bin/*)) "$(DESTDIR)$(PREFIX)/$(LIBDIR)/elixir/bin"
-	$(INSTALL_DIR) "$(DESTDIR)$(PREFIX)/bin"
-	for file in "$(DESTDIR)$(PREFIX)"/$(LIBDIR)/elixir/bin/* ; do \
-		ln -sf "../$(LIBDIR)/elixir/bin/$${file##*/}" "$(DESTDIR)$(PREFIX)/bin/" ; \
-	done
-
-clean:
-	cd lib/elixir && $(REBAR) clean
-	rm -rf ebin
-	rm -rf lib/*/ebin
-	rm -rf lib/*/tmp
-	rm -rf lib/mix/test/fixtures/git_repo
-	rm -rf lib/mix/test/fixtures/deps_on_git_repo
-	rm -rf lib/mix/test/fixtures/git_rebar
-	rm -rf lib/elixir/src/elixir.app.src
-
-clean_exbeam:
-	$(Q) rm -f lib/*/ebin/Elixir.*.beam
-
-#==> Release tasks
-
-SOURCE_REF = $(shell head="$$(git rev-parse HEAD)" tag="$$(git tag --points-at $$head | tail -1)" ; echo "$${tag:-$$head}\c")
-
-docs: compile ../ex_doc/bin/ex_doc
-	mkdir -p ebin
-	rm -rf docs
-	cp -R -f lib/*/ebin/*.beam ./ebin
-	bin/elixir ../ex_doc/bin/ex_doc "Elixir" "$(VERSION)" -m Kernel -u "https://github.com/elixir-lang/elixir" --source-ref "$(call SOURCE_REF)"
-	rm -rf ebin
-
-../ex_doc/bin/ex_doc:
-	@ echo "ex_doc is not found in ../ex_doc as expected. See README for more information."
-	@ false
-
-release_zip: compile
-	rm -rf v$(VERSION).zip
-	zip -9 -r v$(VERSION).zip bin CHANGELOG.md LEGAL lib/*/ebin LICENSE README.md rel VERSION
-
-release_docs: docs
-	cd ../elixir-lang.github.com
-	rm -rf ../elixir-lang.github.com/docs/master
-	mv docs ../elixir-lang.github.com/docs/master
-
-release_erl: compile
-	$(Q) rm -rf rel/elixir
-	$(Q) cd rel && ../rebar generate
-
-#==> Tests tasks
-
-test: test_erlang test_elixir
-
-test_erlang: compile
-	@ echo "==> elixir (eunit)"
-	$(Q) mkdir -p lib/elixir/test/ebin
-	$(Q) $(ERLC) -pa lib/elixir/ebin -o lib/elixir/test/ebin lib/elixir/test/erlang/*.erl
-	$(Q) $(ERL) -pa lib/elixir/test/ebin -s test_helper test -s erlang halt;
-	@ echo ""
-
-test_elixir: test_kernel test_ex_unit test_doc_test test_mix test_eex test_iex
-
-test_doc_test: compile
-	@ echo "==> doctest (exunit)"
-	$(Q) cd lib/elixir && ../../bin/elixir -r "test/doc_test.exs";
-
-test_kernel: compile
-	@ echo "==> kernel (exunit)"
-	$(Q) cd lib/elixir && ../../bin/elixir -r "test/elixir/test_helper.exs" -pr "test/elixir/**/*_test.exs";
-
-.dialyzer.base_plt:
-	@ echo "==> Adding Erlang/OTP basic applications to a new base PLT"
-	$(Q) dialyzer --output_plt .dialyzer.base_plt --build_plt --apps erts kernel stdlib compiler tools syntax_tools parsetools
-
-dialyze: .dialyzer.base_plt
-	$(Q) rm -f .dialyzer_plt
-	$(Q) cp .dialyzer.base_plt .dialyzer_plt
-	@ echo "==> Adding Elixir to PLT..."
-	$(Q) dialyzer --plt .dialyzer_plt --add_to_plt -r lib/elixir/ebin lib/ex_unit/ebin lib/eex/ebin lib/iex/ebin lib/mix/ebin
-	@ echo "==> Dialyzing Elixir..."
-	$(Q) dialyzer --plt .dialyzer_plt -r lib/elixir/ebin lib/ex_unit/ebin lib/eex/ebin lib/iex/ebin lib/mix/ebin
@@ -1,57 +1,94 @@
-![Elixir](https://github.com/elixir-lang/elixir-lang.github.com/raw/master/images/logo/logo.png)
-=========
-[![Build Status](https://secure.travis-ci.org/elixir-lang/elixir.png?branch=master "Build Status")](http://travis-ci.org/elixir-lang/elixir)
+# n8n-openai-adapter

-For more about Elixir, installation and documentation, [check Elixir's website](http://elixir-lang.org/).
+An OpenAI-compatible HTTP adapter that exposes self-hosted **n8n chat agents**
+behind a standard `/v1/chat/completions` API, so any OpenAI client (Cursor,
+LibreChat, the `openai` SDK, a custom app) can talk to your n8n agents as if
+they were OpenAI models.

-## Usage
+n8n itself does **not** ship an inbound OpenAI-compatible endpoint (its "AI
+Gateway" is an outbound proxy to n8n Cloud). This small Elixir service is the
+bridge: one `/v1/chat/completions` endpoint, routed to whichever n8n agent you
+name in the `model` field.

-If you want to contribute to Elixir or run it from source, clone this repository to your machine, compile and test it:
+## How it works

-    $ git clone https://github.com/elixir-lang/elixir.git
-    $ cd elixir
-    $ make clean test
+```
+Your OpenAI client
+   POST /v1/chat/completions {"model":"scholar-agent","thread_id":"abc","messages":[...]}
+        |
+        v
+   n8n-openai-adapter (Plug + Bandit)
+     - authorize (Bearer <ADAPTER_API_KEY>)
+     - look up "scholar-agent" -> n8n chat webhook URL (AgentRegistry GenServer)
+     - take the last user message
+     - forward to the n8n webhook {sessionId: thread_id, action: sendMessage, chatInput}
+        |
+        v
+   n8n agent (its MCP tools, memory, etc. run as usual)
+        |
+        v
+   returns OpenAI-shaped {"choices":[{"message":{"role":"assistant","content":...}}]}
+```

-If Elixir fails to build (specifically when pulling in a new version via git), be sure to remove any previous build artifacts by running `make clean`, then `make test`.
+Multiple agents = multiple `model` names, each mapped to a different n8n webhook
+in the `AGENTS` env var.

-If tests pass, you are ready to move on to the [Getting Started guide][1] or to try Interactive Elixir by running: `bin/iex` in your terminal.
+## Configuration (env vars)

-However, if tests fail, it is likely you have an outdated Erlang version (Elixir requires Erlang R16B or later). You can check your Erlang version by calling `erl` in the command line. You will see some information as follows:
+| Var              | Required | Purpose                                                            |
+|------------------|----------|---------------------------------------------------------------------|
+| `ADAPTER_API_KEY`| yes      | Bearer key that OpenAI clients send.                                |
+| `ADMIN_API_KEY`  | yes      | Bearer key for the admin API / web admin page.                      |
+| `AGENTS_FILE`    | no       | Path to the JSON store (default `/var/lib/n8n-openai/agents.json`). |
+| `PORT`           | no       | HTTP port (default `8000`).                                         |
+| `CHAT_WEBHOOK_BASIC` | no   | `"user:password"` if your n8n Chat Trigger is Basic-auth protected. |

-    Erlang R16B (erts-5.10.1) [source] [64-bit] [smp:2:2] [rq:2] [async-threads:0] [hipe] [kernel-poll:false]
+Agents are **not** configured via env — they're managed at runtime through the
+web admin page / admin API and persisted to `AGENTS_FILE`. The store starts
+empty; add agents after boot.

-If you have the correct version and tests still fail, feel free to [open an issue][2].
+## Admin API (manage agents at runtime)

-## Building documentation
+Agents are persisted to `AGENTS_FILE` and can be added/removed without a
+redeploy, using the `ADMIN_API_KEY`:

-Building the documentation requires [ex_doc](https://github.com/elixir-lang/ex_doc) to be installed and built in the same containing folder as elixir.
+```bash
+# list
+curl -H "Authorization: Bearer $ADMIN_API_KEY" https://openai.bueso.eu/admin/agents

-    # After cloning and compiling Elixir
-    $ git clone git://github.com/elixir-lang/ex_doc.git
-    $ cd ex_doc && ../elixir/bin/mix compile
-    $ cd ../elixir && make docs
+# add / update an agent
+curl -X POST -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" \
+  -d '{"model":"media-agent","webhook":"https://n8n.bueso.eu/webhook/<id>/chat"}' \
+  https://openai.bueso.eu/admin/agents

-## Contributing
+# remove
+curl -X DELETE -H "Authorization: Bearer $ADMIN_API_KEY" \
+  https://openai.bueso.eu/admin/agents/media-agent
+```

-We appreciate any contribution to Elixir, so check out our [CONTRIBUTING.md](CONTRIBUTING.md) guide for more information. We usually keep a list of features and bugs [in the issue tracker][2].
+The store is authoritative and persists across restarts; no env config needed.

-## Important links
+## Building & running

-* #elixir-lang on freenode IRC
-* [Website][1]
-* [Issue tracker][2]
-* [elixir-talk Mailing list (questions)][3]
-* [elixir-core Mailing list (development)][4]
+```bash
+mix deps.get
+mix compile
+ADAPTER_API_KEY=secret AGENTS='{"scholar-agent":"https://n8n.bueso.eu/webhook/<id>/chat"}' \
+  PORT=8000 mix run --no-halt
+```

-  [1]: http://elixir-lang.org
-  [2]: https://github.com/elixir-lang/elixir/issues
-  [3]: http://groups.google.com/group/elixir-lang-talk
-  [4]: http://groups.google.com/group/elixir-lang-core
+## Testing

-## License
+```bash
+MIX_ENV=test mix test
+```

-"Elixir" and the Elixir logo are copyright (c) 2012 Plataformatec.
+## Nix

-Elixir source code is released under Apache 2 License with some parts under Erlang's license (EPL).
+The repo ships a `flake.nix` exporting `overlays.default` and a `packages.default`
+(the packaged BEAM release), so it can be consumed as a flake input from your
+NixOS config just like any other flake — e.g.:

-Check LEGAL and LICENSE files for more information.
+```nix
+inputs.n8n-openai-adapter.url = "git+https://gitea.bueso.eu/<owner>/n8n-openai-adapter";
+```
@@ -1,31 +0,0 @@
-## Release process
-
-This document simply outlines the release process:
-
-1) Remove `-dev` extension from VERSION
-
-2) Run `make clean test` to ensure all tests pass from scratch and the CI is green
-
-3) Ensure CHANGELOG is updated and tag release version with timestamp in it
-
-4) Commit changes above with title "Release vVERSION"
-
-5) Push master and create tag from master branch
-
-6) Release new docs with `make release_docs`, move docs to `docs/stable`
-
-7) Release new zip with `make release_zip`, push new zip to Github Releases
-
-8) Push package to expm with `expm publish package.exs`
-
-9) Merge master into stable branch and push it
-
-10) After release, bump versions and add `-dev` back
-
-11) `make release_docs` once again and push `elixir-lang.github.com`
-
-## Places where version is mentioned
-
-* VERSION
-* CHANGELOG
-* src/elixir.app.src
@@ -1 +0,0 @@
-0.11.0
@@ -1,83 +0,0 @@
-#!/bin/sh
-if [ $# -eq 0 ] || [ "$1" = "--help" ] || [ "$1" = "-h" ]; then
-  echo "Usage: `basename $0` [options] [.exs file] [data]
-
-  -v                Prints version and exit
-  -e \"command\"      Evaluates the given command (*)
-  -r \"file\"         Requires the given files/patterns (*)
-  -S \"script\"       Finds and executes the given script (*)
-  -pr \"file\"        Requires the given files/patterns in parallel (*)
-  -pa \"path\"        Prepends the given path to Erlang code path (*)
-  -pz \"path\"        Appends the given path to Erlang code path (*)
-  --app \"app\"       Start the given app and its dependencies (*)
-  --erl \"switches\"  Switches to be passed down to erlang (*)
-  --name \"name\"     Makes and assigns a name to the distributed node
-  --sname \"name\"    Makes and assigns a short name to the distributed node
-  --cookie \"cookie\" Sets a cookie for this distributed node
-  --hidden          Makes a hidden node
-  --detached        Starts the Erlang VM detached from console
-  --no-halt         Does not halt the Erlang VM after execution
-
-** Options marked with (*) can be given more than once
-** Options given after the .exs file or -- are passed down to the executed code
-** Options can be passed to the erlang runtime using ELIXIR_ERL_OPTS or --erl" >&2
-  exit 1
-fi
-
-readlink_f () {
-  cd "$(dirname "$1")" > /dev/null
-  filename="$(basename "$1")"
-  if [ -h "$filename" ]; then
-    readlink_f "$(readlink "$filename")"
-  else
-    echo "`pwd -P`/$filename"
-  fi
-}
-
-ERL=""
-I=1
-
-while [ $I -le $# ]; do
-  S=1
-  eval "PEEK=\${$I}"
-  case "$PEEK" in
-    -v|--compile|--no-halt|+iex|+compile)
-        ;;
-    -e|-r|-pr|-pa|-pz|--remsh|--app)
-        S=2
-        ;;
-    --detached|--hidden)
-        ERL="$ERL `echo $PEEK | cut -c 2-`"
-        ;;
-    --cookie)
-        I=$(expr $I + 1)
-        eval "VAL=\${$I}"
-        ERL="$ERL -setcookie "$VAL""
-        ;;
-    --sname|--name)
-        I=$(expr $I + 1)
-        eval "VAL=\${$I}"
-        ERL="$ERL `echo $PEEK | cut -c 2-` "$VAL""
-        ;;
-    --erl)
-        I=$(expr $I + 1)
-        eval "VAL=\${$I}"
-        ERL="$ERL "$VAL""
-        ;;
-    *)
-        break
-        ;;
-  esac
-  I=$(expr $I + $S)
-done
-
-SELF=$(readlink_f "$0")
-SCRIPT_PATH=$(dirname "$SELF")
-if [ "$ELIXIR_NO_CLI" != "1" ]; then ERL="$ERL -s elixir start_cli"; fi
-
-if [ -f "$SCRIPT_PATH/../releases/RELEASES" ] && [ -f "$SCRIPT_PATH/erl" ]
-then
-  exec "$SCRIPT_PATH"/erl -pa "$SCRIPT_PATH"/../lib/*/ebin -noshell $ELIXIR_ERL_OPTS $ERL -boot start -extra "$@"
-else
-  exec erl -pa "$SCRIPT_PATH"/../lib/*/ebin -noshell $ELIXIR_ERL_OPTS $ERL -extra "$@"
-fi
@@ -1,88 +0,0 @@
-@echo off
-set argc=0
-for %%x in (%*) do set /A argc+=1
-if  %argc%== 0 (
-  goto documentation
-) else (
-  goto parseopts
-)
-:documentation
-echo Usage: %~nx0 [options] [.exs file] [data]
-echo.
-echo   -v                Prints version and exit
-echo   -e command        Evaluates the given command (*)
-echo   -r file           Requires the given files/patterns (*)
-echo   -S script         Finds and executes the given script (*)
-echo   -pr file          Requires the given files/patterns in parallel (*)
-echo   -pa path          Prepends the given path to Erlang code path (*)
-echo   -pz path          Appends the given path to Erlang code path (*)
-echo   --app app         Start the given app and its dependencies (*)
-echo   --erl switches    Switches to be passed down to erlang (*)
-echo   --name name       Makes and assigns a name to the distributed node
-echo   --sname name      Makes and assigns a short name to the distributed node
-echo   --cookie cookie   Sets a cookie for this distributed node
-echo   --hidden          Makes a hidden node
-echo   --detached        Starts the Erlang VM detached from console
-echo   --no-halt         Does not halt the Erlang VM after execution
-echo.
-echo ** Options marked with (*) can be given more than once
-echo ** Options given after the .exs file or -- are passed down to the executed code
-echo ** Options can be passed to the erlang runtime using ELIXIR_ERL_OPTS or --erl
-goto :EOF
-
-:parseopts
-
-rem Parameters for Erlang
-set parsErlang=
-
-rem Make sure we keep a copy of all parameters
-set allPars=%*
-
-rem Get the original path name from the batch file
-set originPath=%~dp0
-
-rem Optional parameters before the "-extra" parameter
-set beforeExtra=
-
-rem Recursive loop called for each parameter that parses the cmd line parameters
-:startloop
-set par="%1"
-shift
-if "%par%"=="" (
-  rem if no parameters defined
-  goto :expand_erl_libs
-)
-if "%par%"=="""" (
-  rem if no parameters defined - special case for parameter that is already quoted
-  goto :expand_erl_libs
-)
-rem ******* ERLANG PARAMETERS **********************
-IF NOT "%par%"=="%par:--detached=%" (Set parsErlang=%parsErlang% -detached) 
-IF NOT "%par%"=="%par:--hidden=%"   (Set parsErlang=%parsErlang% -hidden)
-IF NOT "%par%"=="%par:--cookie=%"   (Set parsErlang=%parsErlang% -setcookie %1 && shift)
-IF NOT "%par%"=="%par:--sname=%"    (Set parsErlang=%parsErlang% -sname %1 && shift) 
-IF NOT "%par%"=="%par:--name=%"     (Set parsErlang=%parsErlang% -name %1 && shift) 
-IF NOT "%par%"=="%par:--erl=%"      (Set beforeExtra=%beforeExtra% %~1 && shift) 
-rem ******* elixir parameters **********************
-rem Note: we don't have to do anything with options that don't take an argument
-IF NOT "%par%"=="%par:-e=%"      (shift) 
-IF NOT "%par%"=="%par:-r=%"      (shift) 
-IF NOT "%par%"=="%par:-pr=%"     (shift) 
-IF NOT "%par%"=="%par:-pa=%"     (shift) 
-IF NOT "%par%"=="%par:-pz=%"     (shift) 
-IF NOT "%par%"=="%par:--app=%"   (shift) 
-IF NOT "%par%"=="%par:--remsh=%" (shift) 
-goto:startloop
-
-rem ******* assume all pre-params are parsed ********************
-:expand_erl_libs
-rem ******* expand all ebin paths as Windows does not support the ..\*\ebin wildcard ********************
-SETLOCAL enabledelayedexpansion
-set ext_libs=
-for  /d %%d in ("%originPath%..\lib\*.") do (
-  set ext_libs=!ext_libs! -pa "%%~fd\ebin"
-)
-SETLOCAL disabledelayedexpansion
-:run
-erl %ext_libs% -noshell %ELIXIR_ERL_OPTS% %parsErlang% -s elixir start_cli %beforeExtra% -extra %*
-
@@ -1,29 +0,0 @@
-#!/bin/sh
-if [ $# -eq 0 ] || [ "$1" = "--help" ] || [ "$1" = "-h" ]; then
-  echo "Usage: `basename $0` [elixir switches] [compiler switches] [.ex files]
-
-  -o               The directory to output compiled files
-  --no-docs        Do not attach documentation to compiled modules
-  --no-debug-info  Do not attach debug info to compiled modules
-  --ignore-module-conflict
-  --warnings-as-errors Treat warnings as errors and return non-zero exit code
-  --verbose        Print informational messages.
-
-** Options given after -- are passed down to the executed code
-** Options can be passed to the erlang runtime using ELIXIR_ERL_OPTS" >&2
-  exit 1
-fi
-
-readlink_f () {
-  cd "$(dirname "$1")" > /dev/null
-  filename="$(basename "$1")"
-  if [ -h "$filename" ]; then
-    readlink_f "$(readlink "$filename")"
-  else
-    echo "`pwd -P`/$filename"
-  fi
-}
-
-SELF=$(readlink_f "$0")
-SCRIPT_PATH=$(dirname "$SELF")
-exec "$SCRIPT_PATH"/elixir +compile "$@"
@@ -1,23 +0,0 @@
-@echo off
-set argc=0
-for %%x in (%*) do set /A argc+=1
-if %argc% == 0 (
-  goto documentation
-) else (
-  goto run
-)
-:documentation
-echo Usage: %~nx0 [elixir switches] [compiler switches] [.ex files]
-echo.
-echo  -o               The directory to output compiled files
-echo  --no-docs        Do not attach documentation to compiled modules
-echo  --no-debug-info  Do not attach debug info to compiled modules
-echo  --ignore-module-conflict
-echo  --warnings-as-errors Treat warnings as errors and return non-zero exit code
-echo  --verbose        Print informational messages.
-echo.
-echo ** Options marked with (*) can be given more than once
-echo ** Options given after -- are passed down to the executed code
-echo ** Options can be passed to the erlang runtime using ELIXIR_ERL_OPTS" >&2
-:run
-call "%~dp0\elixir.bat" +compile %*
@@ -1,41 +0,0 @@
-#!/bin/sh
-if [ $# -gt 0 ] && ([ "$1" = "--help" ] || [ "$1" = "-h" ]); then
-  echo "Usage: `basename $0` [options] [.exs file] [data]
-
-  -v                Prints version
-  -e \"command\"      Evaluates the given command (*)
-  -r \"file\"         Requires the given files/patterns (*)
-  -S \"script\"       Finds and executes the given script (*)
-  -pr \"file\"        Requires the given files/patterns in parallel (*)
-  -pa \"path\"        Prepends the given path to Erlang code path (*)
-  -pz \"path\"        Appends the given path to Erlang code path (*)
-  --app \"app\"       Start the given app and its dependencies (*)
-  --erl \"switches\"  Switches to be passed down to erlang (*)
-  --name \"name\"     Makes and assigns a name to the distributed node
-  --sname \"name\"    Makes and assigns a short name to the distributed node
-  --cookie \"cookie\" Sets a cookie for this distributed node
-  --hidden          Makes a hidden node
-  --detached        Starts the Erlang VM detached from console
-  --remsh \"name\"    Connects to a node using a remote shell
-  --dot-iex \"path\"  Overrides default .iex file and uses path instead;
-                    path can be empty, then no file will be loaded
-
-** Options marked with (*) can be given more than once
-** Options given after the .exs file or -- are passed down to the executed code
-** Options can be passed to the erlang runtime using ELIXIR_ERL_OPTS or --erl" >&2
-  exit 1
-fi
-
-readlink_f () {
-  cd "$(dirname "$1")" > /dev/null
-  filename="$(basename "$1")"
-  if [ -h "$filename" ]; then
-    readlink_f "$(readlink "$filename")"
-  else
-    echo "`pwd -P`/$filename"
-  fi
-}
-
-SELF=$(readlink_f "$0")
-SCRIPT_PATH=$(dirname "$SELF")
-ELIXIR_NO_CLI=1 exec "$SCRIPT_PATH"/elixir --no-halt --erl "-user Elixir.IEx.CLI" +iex "$@"
@@ -1,2 +0,0 @@
-@echo off
-call "%~dp0\elixir.bat" +iex --no-halt -e "IEx.start" %*
@@ -1,3 +0,0 @@
-#!/usr/bin/env elixir
-Mix.start
-Mix.CLI.main
@@ -1,2 +0,0 @@
-@echo off
-call "%~dp0\elixir.bat" "%~dp0\mix" %*
@@ -0,0 +1,7 @@
+import Config
+
+import_config "#{config_env()}.exs"
+
+if config_env() == :test do
+  config :logger, level: :warning
+end
@@ -0,0 +1,3 @@
+import Config
+
+# Dev: no special config — all runtime settings come from env vars.
@@ -0,0 +1,5 @@
+import Config
+
+# Production: no hardcoded values here. All runtime config (PORT, AGENTS,
+# ADAPTER_API_KEY, CHAT_WEBHOOK_BASIC) comes from the systemd EnvironmentFile
+# in the NixOS service module.
@@ -0,0 +1,7 @@
+import Config
+
+# Test environment: the app starts with an empty agent store (no AGENTS env
+# seeding — agents are managed via the admin API). ADAPTER_API_KEY /
+# ADMIN_API_KEY are set in test/test_helper.exs. AGENTS_FILE must be set HERE
+# (config loads before the app boots) to a writable tmp path.
+System.put_env("AGENTS_FILE", Path.join(System.tmp_dir!(), "n8n-openai-test-agents.json"))
@@ -0,0 +1,27 @@
+{
+  "nodes": {
+    "nixpkgs": {
+      "locked": {
+        "lastModified": 1788881743,
+        "narHash": "sha256-2V9GZGvPfrNzxFozhI9dcqV+c3QdA8YZrvAAzqEB+dI=",
+        "owner": "NixOS",
+        "repo": "nixpkgs",
+        "rev": "d6524aaca2ff07876657ae2b323f24be4874944b",
+        "type": "github"
+      },
+      "original": {
+        "owner": "NixOS",
+        "ref": "nixos-unstable",
+        "repo": "nixpkgs",
+        "type": "github"
+      }
+    },
+    "root": {
+      "inputs": {
+        "nixpkgs": "nixpkgs"
+      }
+    }
+  },
+  "root": "root",
+  "version": 7
+}
@@ -0,0 +1,48 @@
+{
+  description = "OpenAI-compatible adapter exposing n8n chat agents behind /v1/chat/completions";
+
+  inputs = {
+    nixpkgs.url = "github:NixOS/nixpkgs/nixos-unstable";
+  };
+
+  outputs =
+    { self, nixpkgs, ... }:
+    let
+      supportedSystems = [
+        "x86_64-linux"
+        "aarch64-linux"
+      ];
+      forAllSystems = nixpkgs.lib.genAttrs supportedSystems;
+    in
+    {
+      packages = forAllSystems (
+        system:
+        let
+          pkgs = import nixpkgs { inherit system; };
+          beamPackages = pkgs.beamPackages;
+        in
+        {
+          default = beamPackages.mixRelease {
+            pname = "n8n-openai-adapter";
+            version = "0.1.0";
+            src = self;
+            mixFodDeps = beamPackages.fetchMixDeps {
+              pname = "n8n-openai-adapter";
+              version = "0.1.0";
+              src = self;
+              hash = "sha256-sdAhpZUeF33V9xjEa/z/aTmCllfetMjO/1XyfJfUNao=";
+            };
+          };
+        }
+      );
+
+      overlays.default = final: prev: {
+        n8n-openai-adapter = self.packages.${final.stdenv.system}.default;
+      };
+
+      # Proper NixOS module: consume with
+      #   imports = [ inputs.n8n-openai-adapter.nixosModules.default ];
+      #   services.n8n-openai-adapter = { enable = true; domain = "..."; port = 8134; };
+      nixosModules.default = import ./nixos-module.nix;
+    };
+}
@@ -1,201 +0,0 @@
-defexception EEx.SyntaxError, message: nil
-
-defmodule EEx do
-  @moduledoc %S"""
-  EEx stands for Embedded Elixir. It allows you to embed
-  Elixir code inside a string in a robust way:
-
-      EEx.eval_string "foo <%= bar %>", [bar: "baz"]
-      #=> "foo baz"
-
-  ## API
-
-  This module provides 3 main APIs for you to use:
-
-  1) Evaluate a string (`eval_string`) or a file (`eval_file`)
-     directly. This is the simplest API to use but also the
-     slowest, since the code is evaluated and not compiled before;
-
-  2) Define a function from a string (`function_from_string`)
-     or a file (`function_from_file`). This allows you to embed
-     the template as a function inside a module which will then
-     be compiled. This is the preferred API if you have access
-     to the template at compilation time;
-
-  3) Compile a string (`compile_string`) or a file (`compile_file`)
-     into Elixir syntax tree. This is the API used by both functions
-     above and is available to you if you want to provide your own
-     ways of handling the compiled template.
-
-  ## Engine
-
-  EEx has the concept of engines which allows you to modify or
-  transform the code extracted from the given string or file.
-
-  By default, `EEx` uses the `EEx.SmartEngine` that provides some
-  conveniences on top of the simple `EEx.Engine`.
-
-  ### Tags
-
-  `EEx.SmartEngine` supports the following tags:
-
-      <% Elixir expression - inline with output %>
-      <%= Elixir expression - replace with result %>
-      <%% EEx quotation - returns the contents inside %>
-      <%# Comments - they are discarded from source %>
-
-  All expressions that output something to the template
-  **must** use the equals sign (`=`). Since everything in
-  Elixir is a macro, there are no exceptions for this rule.
-  For example, while some template languages would special-
-  case `if` clauses, they are treated the same in EEx and
-  also require `=` in order to have their result printed:
-
-      <%= if true do %>
-        It is obviously true
-      <% else %>
-        This will never appear
-      <% end %>
-
-  Notice that different engines may have different rules
-  for each tag. Other tags may be added in future versions.
-
-  ### Macros
-
-  `EEx.SmartEngine` also adds some macros to your template.
-  An example is the `@` macro which allows easy data access
-  in a template:
-
-      EEx.eval_string "<%= @foo %>", assigns: [foo: 1]
-      #=> 1
-
-  In other words, <%= @foo %> is simply translated to:
-
-      <%= Keyword.get assigns, :foo %>
-
-  The assigns extension is useful when the number of variables
-  required by the template is not specified at compilation time.
-  """
-
-  @doc """
-  Generates a function definition from the string.
-  The kind (`:def` or `:defp`) must be given, the
-  function name, its arguments and the compilation options.
-
-  ## Examples
-
-      defmodule Sample do
-        require EEx
-        EEx.function_from_string :def, :sample, "<%= a + b %>", [:a, :b]
-      end
-
-      Sample.sample(1, 2) #=> "3"
-
-  """
-  defmacro function_from_string(kind, name, source, args // [], options // []) do
-    quote bind_quoted: binding do
-      info = Keyword.merge [file: __ENV__.file, line: __ENV__.line], options
-      args = Enum.map args, fn arg -> { arg, [line: info[:line]], nil } end
-      compiled = EEx.compile_string(source, info)
-
-      case kind do
-        :def  -> def(unquote(name)(unquote_splicing(args)), do: unquote(compiled))
-        :defp -> defp(unquote(name)(unquote_splicing(args)), do: unquote(compiled))
-      end
-    end
-  end
-
-  @doc """
-  Generates a function definition from the file contents.
-  The kind (`:def` or `:defp`) must be given, the
-  function name, its arguments and the compilation options.
-
-  This function is useful in case you have templates but
-  you want to precompile inside a module for speed.
-
-  ## Examples
-
-      # sample.eex
-      <%= a + b %>
-
-      # sample.ex
-      defmodule Sample do
-        require EEx
-        EEx.function_from_file :def, :sample, "sample.eex", [:a, :b]
-      end
-
-      # iex
-      Sample.sample(1, 2) #=> "3"
-
-  """
-  defmacro function_from_file(kind, name, file, args // [], options // []) do
-    quote bind_quoted: binding do
-      info = Keyword.merge options, [file: file, line: 1]
-      args = Enum.map args, fn arg -> { arg, [line: 1], nil } end
-      compiled = EEx.compile_file(file, info)
-
-      @file file
-      case kind do
-        :def  -> def(unquote(name)(unquote_splicing(args)), do: unquote(compiled))
-        :defp -> defp(unquote(name)(unquote_splicing(args)), do: unquote(compiled))
-      end
-    end
-  end
-
-  @doc """
-  Get a string `source` and generate a quoted expression
-  that can be evaluated by Elixir or compiled to a function.
-  """
-  def compile_string(source, options // []) do
-    EEx.Compiler.compile(source, options)
-  end
-
-  @doc """
-  Get a `filename` and generate a quoted expression
-  that can be evaluated by Elixir or compiled to a function.
-  """
-  def compile_file(filename, options // []) do
-    options = Keyword.merge options, [file: filename, line: 1]
-    compile_string(File.read!(filename), options)
-  end
-
-  @doc """
-  Get a string `source` and evaluate the values using the `bindings`.
-
-  ## Examples
-
-      EEx.eval_string "foo <%= bar %>", [bar: "baz"]
-      #=> "foo baz"
-
-  """
-  def eval_string(source, bindings // [], options // []) do
-    compiled = compile_string(source, options)
-    do_eval(compiled, bindings, options)
-  end
-
-  @doc """
-  Get a `filename` and evaluate the values using the `bindings`.
-
-  ## Examples
-
-      # sample.ex
-      foo <%= bar %>
-
-      # iex
-      EEx.eval_file "sample.ex", [bar: "baz"]
-      #=> "foo baz"
-
-  """
-  def eval_file(filename, bindings // [], options // []) do
-    options  = Keyword.put options, :file, filename
-    compiled = compile_file(filename, options)
-    do_eval(compiled, bindings, options)
-  end
-
-  ### Helpers
-
-  defp do_eval(compiled, bindings, options) do
-    { result, _ } = Code.eval_quoted(compiled, bindings, options)
-    result
-  end
-end
@@ -1,117 +0,0 @@
-defrecord EEx.State, engine: EEx.SmartEngine, dict: [], file: 'nofile', line: 1, start_line: 1
-
-defmodule EEx.Compiler do
-  @moduledoc false
-
-  @doc """
-  This is the compilation entry point. It glues the tokenizer
-  and the engine together by handling the tokens and invoking
-  the engine every time a full expression or text is received.
-  """
-  def compile(source, options) do
-    line   = Keyword.get(options, :line, 1)
-    tokens = EEx.Tokenizer.tokenize(source, line)
-    state  = EEx.State.new(options)
-    generate_buffer(tokens, "", [], state)
-  end
-
-  # Generates the buffers by handling each expression from the tokenizer
-
-  defp generate_buffer([{ :text, _line, chars }|t], buffer, scope, state) do
-    buffer = state.engine.handle_text(buffer, chars)
-    generate_buffer(t, buffer, scope, state)
-  end
-
-  defp generate_buffer([{ :expr, line, mark, chars }|t], buffer, scope, state) do
-    expr = maybe_block :elixir_translator.forms!(chars, line, state.file, [])
-    buffer = state.engine.handle_expr(buffer, mark, expr)
-    generate_buffer(t, buffer, scope, state)
-  end
-
-  defp generate_buffer([{ :start_expr, line, mark, chars }|t], buffer, scope, state) do
-    { contents, t } = generate_buffer(t, "", [chars|scope], state.dict([]).line(line).start_line(line))
-    buffer = state.engine.handle_expr(buffer, mark, contents)
-    generate_buffer(t, buffer, scope, state)
-  end
-
-  defp generate_buffer([{ :middle_expr, line, _, chars }|t], buffer, [current|scope], state) do
-    { wrapped, state } = wrap_expr(current, line, buffer, chars, state)
-    generate_buffer(t, "", [wrapped|scope], state.line(line))
-  end
-
-  defp generate_buffer([{ :end_expr, line, _, chars }|t], buffer, [current|_], state) do
-    { wrapped, state } = wrap_expr(current, line, buffer, chars, state)
-    tuples = maybe_block :elixir_translator.forms!(wrapped, state.start_line, state.file, [])
-    buffer = insert_quotes(tuples, state.dict)
-    { buffer, t }
-  end
-
-  defp generate_buffer([{ :end_expr, line, _, chars }|_], _buffer, [], _state) do
-    raise EEx.SyntaxError, message: "unexpected token: #{inspect chars} at line #{inspect line}"
-  end
-
-  defp generate_buffer([], buffer, [], _state) do
-    buffer
-  end
-
-  defp generate_buffer([], _buffer, _scope, _state) do
-    raise EEx.SyntaxError, message: "unexpected end of string. expecting a closing <% end %>."
-  end
-
-  # Creates a placeholder and wrap it inside the expression block
-
-  defp wrap_expr(current, line, buffer, chars, state) do
-    new_lines = List.duplicate(?\n, line - state.line)
-
-    if state.dict == [] and is_empty?(buffer) do
-      { current ++ new_lines ++ chars, state }
-    else
-      key = length(state.dict)
-      placeholder = '__EEX__(' ++ integer_to_list(key) ++ ');'
-      { current ++ placeholder ++ new_lines ++ chars, state.update_dict(&[{key, buffer}|&1]) }
-    end
-  end
-
-  # Check if the syntax node represents an empty string
-
-  defp is_empty?(bin) when is_binary(bin) do
-    bc(<<c>> inbits bin, not c in [?\s, ?\t, ?\r, ?\n], do: <<c>>) == ""
-  end
-
-  defp is_empty?({ :<>, _, [left, right] }) do
-    is_empty?(left) and is_empty?(right)
-  end
-
-  defp is_empty?(_) do
-    false
-  end
-
-  # Block wrapping
-
-  defp maybe_block([]),    do: nil
-  defp maybe_block([h]),   do: h
-  defp maybe_block(other), do: { :__block__, [], other }
-
-  # Changes placeholder to real expression
-
-  defp insert_quotes({ :__EEX__, _, [key] }, dict) do
-    { ^key, value } = List.keyfind dict, key, 0
-    value
-  end
-
-  defp insert_quotes({ left, line, right }, dict) do
-    { insert_quotes(left, dict), line, insert_quotes(right, dict) }
-  end
-
-  defp insert_quotes({ left, right }, dict) do
-    { insert_quotes(left, dict), insert_quotes(right, dict) }
-  end
-
-  defp insert_quotes(list, dict) when is_list(list) do
-    Enum.map list, &insert_quotes(&1, dict)
-  end
-
-  defp insert_quotes(other, _dict) do
-    other
-  end
-end
@@ -1,57 +0,0 @@
-defmodule EEx.Engine do
-  @moduledoc %S"""
-  This is the basic EEx engine that ships with Elixir.
-  An engine needs to implement two functions:
-
-  * `handle_text(buffer, text)` - it receives the buffer,
-    the text and must return a new quoted expression;
-
-  * `handle_expr(buffer, marker, expr)` - it receives the buffer,
-    the marker, the expr and must return a new quoted expression;
-
-    The marker is what follows exactly after `<%`. For example,
-    `<% foo %>` has an empty marker, but `<%= foo %>` has `'='`
-    as marker. The allowed markers so far are:
-
-    * `''`
-    * `'='`
-
-    Read `handle_expr/3` below for more information about the markers
-    implemented by default by this engine.
-  """
-
-  use Behaviour
-
-  defcallback handle_text(Macro.t, binary) :: Macro.t
-  defcallback handle_expr(Macro.t, binary, Macro.t) :: Macro.t
-
-  @doc """
-  The default implementation simply concatenates text to the buffer.
-  """
-  def handle_text(buffer, text) do
-    quote do: unquote(buffer) <> unquote(text)
-  end
-
-  @doc """
-  Implements expressions according to the markers.
-
-      <% Elixir expression - inline with output %>
-      <%= Elixir expression - replace with result %>
-
-  All other markers are not implemented by this engine.
-  """
-  def handle_expr(buffer, "=", expr) do
-    quote do
-      tmp = unquote(buffer)
-      tmp <> to_string(unquote(expr))
-    end
-  end
-
-  def handle_expr(buffer, "", expr) do
-    quote do
-      tmp = unquote(buffer)
-      unquote(expr)
-      tmp
-    end
-  end
-end
@@ -1,112 +0,0 @@
-defmodule EEx.TransformerEngine do
-  @moduledoc """
-  An abstract engine that is meant to be used and
-  built upon in other modules. This engine implements
-  the `EEx.Engine` behavior and provides a `transform`
-  overridable directive that allows a developer to
-  customize the expression returned by the engine.
-
-  Check `EEx.AssignsEngine` and `EEx.SmartEngine` for
-  examples of using this module.
-  """
-
-  @doc false
-  defmacro __using__(_) do
-    quote do
-      @behavior EEx.Engine
-
-      def handle_text(buffer, text) do
-        EEx.Engine.handle_text(buffer, text)
-      end
-
-      def handle_expr(buffer, mark, expr) do
-        EEx.Engine.handle_expr(buffer, mark, transform(expr))
-      end
-
-      defp transform({ a, b, c }) do
-        { transform(a), b, transform(c) }
-      end
-
-      defp transform({ a, b }) do
-        { transform(a), transform(b) }
-      end
-
-      defp transform(list) when is_list(list) do
-        lc i inlist list, do: transform(i)
-      end
-
-      defp transform(other) do
-        other
-      end
-
-      defoverridable [transform: 1, handle_expr: 3, handle_text: 2]
-    end
-  end
-end
-
-defmodule EEx.AssignsEngine do
-  @moduledoc """
-  An abstract engine that, when used with the
-  `TransformerEngine`, allows a developer to access
-  assigns using `@` as syntax.
-
-  This engine is included by default on the SmartEngine.
-
-  ## Examples
-
-      defmodule MyEngine do
-        use EEx.TransformerEngine
-        use EEx.AssignsEngine
-      end
-
-      EEx.eval_string("<%= @foo %>", assigns: [foo: 1])
-      #=> 1
-
-  In the example above, we can access the value `foo` under
-  the binding `assigns` using `@foo`. This is useful when
-  a template, after compiled, may receive different assigns
-  and the developer don't want to recompile it for each
-  variable set.
-
-  Assigns can also be used when compiled to a function:
-
-      # sample.eex
-      <%= @a + @b %>
-
-      # sample.ex
-      defmodule Sample do
-        require EEx
-        EEx.function_from_file :def, :sample, "sample.eex", [:assigns]
-      end
-
-      # iex
-      Sample.sample(a: 1, b: 2) #=> "3"
-
-  """
-
-  @doc false
-  defmacro __using__(_) do
-    quote unquote: false do
-      defp transform({ :@, line, [{ name, _, atom }] }) when is_atom(name) and is_atom(atom) do
-        quote do: Keyword.get(var!(assigns), unquote(name))
-      end
-
-      defp transform(arg) do
-        super(arg)
-      end
-
-      defoverridable [transform: 1]
-    end
-  end
-end
-
-defmodule EEx.SmartEngine do
-  use EEx.TransformerEngine
-  use EEx.AssignsEngine
-
-  @moduledoc """
-  An engine meant for end-user usage that includes
-  `EEx.AssignsEngine` and other conveniences. Read
-  `EEx.AssignsEngine` for examples.
-  """
-end
@@ -1,159 +0,0 @@
-defmodule EEx.Tokenizer do
-  @moduledoc false
-
-  @doc """
-  Tokenizes the given char list or binary.
-  It returns 4 different types of tokens as result:
-
-  * { :text, line, contents }
-  * { :expr, line, marker, contents }
-  * { :start_expr, line, marker, contents }
-  * { :end_expr, line, marker, contents }
-
-  """
-  def tokenize(bin, line) when is_binary(bin) do
-    tokenize(String.to_char_list!(bin), line)
-  end
-
-  def tokenize(list, line) do
-    Enum.reverse(tokenize(list, line, line, [], []))
-  end
-
-  defp tokenize('<%%' ++ t, current_line, line, buffer, acc) do
-    { buffer, new_line, rest } = tokenize_expr t, line, [?%, ?<|buffer]
-    tokenize rest, current_line, new_line, [?>, ?%|buffer], acc
-  end
-
-  defp tokenize('<%#' ++ t, current_line, line, buffer, acc) do
-    { _, new_line, rest } = tokenize_expr t, line, []
-    tokenize rest, current_line, new_line, buffer, acc
-  end
-
-  defp tokenize('<%' ++ t, current_line, line, buffer, acc) do
-    { marker, t } = retrieve_marker(t)
-    { expr, new_line, rest } = tokenize_expr t, line, []
-
-    token = token_name(expr)
-    acc   = tokenize_text(current_line, buffer, acc)
-    final = { token, line, marker, Enum.reverse(expr) }
-    tokenize rest, new_line, new_line, [], [final | acc]
-  end
-
-  defp tokenize('\n' ++ t, current_line, line, buffer, acc) do
-    tokenize t, current_line, line + 1, [?\n|buffer], acc
-  end
-
-  defp tokenize([h|t], current_line, line, buffer, acc) do
-    tokenize t, current_line, line, [h|buffer], acc
-  end
-
-  defp tokenize([], current_line, _line, buffer, acc) do
-    tokenize_text(current_line, buffer, acc)
-  end
-
-  # Retrieve marker for <%
-
-  defp retrieve_marker('=' ++ t) do
-    { "=", t }
-  end
-
-  defp retrieve_marker(t) do
-    { "", t }
-  end
-
-  # Tokenize an expression until we find %>
-
-  defp tokenize_expr([?%, ?>|t], line, buffer) do
-    { buffer, line, t }
-  end
-
-  defp tokenize_expr('\n' ++ t, line, buffer) do
-    tokenize_expr t, line + 1, [?\n|buffer]
-  end
-
-  defp tokenize_expr([h|t], line, buffer) do
-    tokenize_expr t, line, [h|buffer]
-  end
-
-  defp tokenize_expr([], _line, _buffer) do
-    raise EEx.SyntaxError, message: "missing token: %>"
-  end
-
-  # Receive an expression content and check
-  # if it is a start, middle or an end token.
-  #
-  # Start tokens finish with `do` and `fn ->`
-  # Middle tokens are marked with `->` or keywords
-  # End tokens contain only the end word
-
-  defp token_name([h|t]) when h in [?\s, ?\t] do
-    token_name(t)
-  end
-
-  defp token_name('od' ++ [h|_]) when h in [?\s, ?\t, ?)] do
-    :start_expr
-  end
-
-  defp token_name('>-' ++ rest) do
-    rest = Enum.reverse(rest)
-
-    # Tokenize the remaining passing check_terminators as
-    # false, which relax the tokenizer to not error on
-    # unmatched pairs. Then, we check if there is a "fn"
-    # token and, if so, it is not followed by an "end"
-    # token. If this is the case, we are on a start expr.
-    case :elixir_tokenizer.tokenize(rest, 1, file: "eex", check_terminators: false) do
-      { :ok, _line, tokens } ->
-        tokens   = Enum.reverse(tokens)
-        fn_index = fn_index(tokens)
-
-        if fn_index && end_index(tokens) > fn_index do
-          :start_expr
-        else
-          :middle_expr
-        end
-      _error ->
-        :middle_expr
-    end
-  end
-
-  defp token_name('esle' ++ t),   do: check_spaces(t, :middle_expr)
-  defp token_name('retfa' ++ t),  do: check_spaces(t, :middle_expr)
-  defp token_name('hctac' ++ t),  do: check_spaces(t, :middle_expr)
-  defp token_name('eucser' ++ t), do: check_spaces(t, :middle_expr)
-  defp token_name('dne' ++ t),    do: check_spaces(t, :end_expr)
-
-  defp token_name(_) do
-    :expr
-  end
-
-  defp fn_index(tokens) do
-    Enum.find_index tokens, fn
-      { :fn_paren, _ } -> true
-      { :fn, _ }       -> true
-      _                -> false
-    end
-  end
-
-  defp end_index(tokens) do
-    Enum.find_index(tokens, &match?({ :end, _ }, &1)) || :infinity
-  end
-
-  defp check_spaces(string, token) do
-    if only_spaces?(string), do: token, else: :expr
-  end
-
-  defp only_spaces?([h|t]) when h in [?\s, ?\t], do: only_spaces?(t)
-  defp only_spaces?(other), do: other == []
-
-  # Tokenize the buffered text by appending
-  # it to the given accumulator.
-
-  defp tokenize_text(_line, [], acc) do
-    acc
-  end
-
-  defp tokenize_text(line, buffer, acc) do
-    [{ :text, line, String.from_char_list!(Enum.reverse(buffer)) } | acc]
-  end
-end
@@ -1,7 +0,0 @@
-defmodule EEx.Mixfile do
-  use Mix.Project
-
-  def project do
-    [app: :eex, version: System.version]
-  end
-end
@@ -1,22 +0,0 @@
-Code.require_file "../test_helper.exs", __DIR__
-
-defmodule EEx.SmartEngineTest do
-  use ExUnit.Case, async: true
-
-  test "evaluates simple string" do
-    assert_eval "foo bar", "foo bar"
-  end
-
-  test "evaluates with assigns" do
-    assert_eval "1", "<%= @foo %>", assigns: [foo: 1]
-  end
-
-  test "evaluates with loops" do
-    assert_eval "1\n2\n3\n", "<%= lc x inlist [1, 2, 3] do %><%= x %>\n<% end %>"
-  end
-
-  defp assert_eval(expected, actual, binding // []) do
-    result = EEx.eval_string(actual, binding, file: __FILE__)
-    assert result == expected
-  end
-end
@@ -1,105 +0,0 @@
-Code.require_file "../test_helper.exs", __DIR__
-
-defmodule EEx.TokenizerTest do
-  use ExUnit.Case, async: true
-  require EEx.Tokenizer, as: T
-
-  test "simple chars lists" do
-    assert T.tokenize('foo', 1) == [ { :text, 1, "foo" } ]
-  end
-
-  test "simple strings" do
-    assert T.tokenize("foo", 1) == [ { :text, 1, "foo" } ]
-  end
-
-  test "strings with embedded code" do
-    assert T.tokenize('foo <% bar %>', 1) == [ { :text, 1, "foo " }, { :expr, 1, "", ' bar ' } ]
-  end
-
-  test "strings with embedded equals code" do
-    assert T.tokenize('foo <%= bar %>', 1) == [ { :text, 1, "foo " }, { :expr, 1, "=", ' bar ' } ]
-  end
-
-  test "strings with more than one line" do
-    assert T.tokenize('foo\n<%= bar %>', 1) == [ { :text, 1, "foo\n" }, { :expr, 2, "=", ' bar ' } ]
-  end
-
-  test "strings with more than one line and expression with more than one line" do
-    string = '''
-foo <%= bar
-
-baz %>
-<% foo %>
-'''
-
-    assert T.tokenize(string, 1) == [
-      {:text, 1, "foo "},
-      {:expr, 1, "=", ' bar\n\nbaz '},
-      {:text, 3, "\n"},
-      {:expr, 4, "", ' foo '},
-      {:text, 4, "\n"}
-    ] 
-  end
-
-  test "quotation" do
-    assert T.tokenize('foo <%% true %>', 1) == [
-      { :text, 1, "foo <% true %>" }
-    ]
-  end
-
-  test "quotation with do/end" do
-    assert T.tokenize('foo <%% true do %>bar<%% end %>', 1) == [
-      { :text, 1, "foo <% true do %>bar<% end %>" }
-    ]
-  end
-
-  test "comments" do
-    assert T.tokenize('foo <%# true %>', 1) == [
-      { :text, 1, "foo " }
-    ]
-  end
-
-  test "comments with do/end" do
-    assert T.tokenize('foo <%# true do %>bar<%# end %>', 1) == [
-      { :text, 1, "foo bar" }
-    ]
-  end
-
-  test "strings with embedded do end" do
-    assert T.tokenize('foo <% if true do %>bar<% end %>', 1) == [
-      { :text, 1, "foo " },
-      { :start_expr, 1, "", ' if true do ' },
-      { :text, 1, "bar" },
-      { :end_expr, 1, "", ' end ' }
-    ]
-  end
-
-  test "strings with embedded -> end" do
-    assert T.tokenize('foo <% cond do %><% false -> %>bar<% true -> %>baz<% end %>', 1) == [
-      { :text, 1, "foo " },
-      { :start_expr, 1, "", ' cond do ' },
-      { :middle_expr, 1, "", ' false -> ' },
-      { :text, 1, "bar" },
-      { :middle_expr, 1, "", ' true -> ' },
-      { :text, 1, "baz" },
-      { :end_expr, 1, "", ' end ' }
-    ]
-  end
-
-  test "strings with embedded keywords blocks" do
-    assert T.tokenize('foo <% if true do %>bar<% else %>baz<% end %>', 1) == [
-      { :text, 1, "foo " },
-      { :start_expr, 1, "", ' if true do ' },
-      { :text, 1, "bar" },
-      { :middle_expr, 1, "", ' else ' },
-      { :text, 1, "baz" },
-      { :end_expr, 1, "", ' end ' }
-    ]
-  end
-
-  test "raise syntax error when there is start mark and no end mark" do
-    assert_raise EEx.SyntaxError, "missing token: %>", fn ->
-      T.tokenize('foo <% :bar', 1)
-    end
-  end
-end
@@ -1,351 +0,0 @@
-Code.require_file "test_helper.exs", __DIR__
-
-require EEx
-
-defmodule EExText.Compiled do
-  def before_compile do
-    fill_in_stacktrace
-    { __ENV__.line, hd(tl(System.stacktrace)) }
-  end
-  { :erlang, 1, 2 }.tuple_to_list
-  EEx.function_from_string :def, :string_sample, "<%= a + b %>", [:a, :b]
-
-  filename = Path.join(__DIR__, "fixtures/eex_template_with_bindings.eex")
-  EEx.function_from_file :defp, :private_file_sample, filename, [:bar]
-  def file_sample(arg), do: private_file_sample(arg)
-
-  def after_compile do
-    fill_in_stacktrace
-    { __ENV__.line, hd(tl(System.stacktrace)) }
-  end
-
-  @file "unknown"
-  def unknown do
-    fill_in_stacktrace
-    { __ENV__.line, hd(tl(System.stacktrace)) }
-  end
-
-  defp fill_in_stacktrace do
-    try do
-      :erlang.error "failed"
-    catch
-      :error, _, stack -> stack
-    end
-  end
-end
-
-defmodule Clause do
-  defmacro defclause(expr, block) do
-    quote do
-      def unquote(expr), unquote(block)
-    end
-  end
-end
-
-defmodule EExTest do
-  use ExUnit.Case, async: true
-
-  test "evaluates simple string" do
-    assert_eval "foo bar", "foo bar"
-  end
-
-  test "evaluates with embedded" do
-    assert_eval "foo bar", "foo <%= :bar %>"
-  end
-
-  test "evaluates with embedded and the binding" do
-    assert EEx.eval_string("foo <%= bar %>", [bar: 1]) == "foo 1"
-  end
-
-  test "evaluates with embedded do end" do
-    assert_eval "foo bar", "foo <%= if true do %>bar<% end %>"
-  end
-
-  test "evaluates with embedded do end and eval the expression" do
-    assert_eval "foo ", "foo <%= if false do %>bar<% end %>"
-  end
-
-  test "evaluates with embedded do end and nested print expression" do
-    assert_eval "foo bar", "foo <%= if true do %><%= :bar %><% end %>"
-  end
-
-  test "evaluates with embedded do end and nested expressions" do
-    assert_eval "foo bar baz", "foo <%= if true do %>bar <% Process.put(:eex_text, 1) %><%= :baz %><% end %>"
-    assert Process.get(:eex_text) == 1
-  end
-
-  test "evaluates with embedded middle expression" do
-    assert_eval "foo bar", "foo <%= if true do %>bar<% else %>baz<% end %>"
-  end
-
-  test "evaluates with embedded middle expression and eval the expression" do
-    assert_eval "foo baz", "foo <%= if false do %>bar<% else %>baz<% end %>"
-  end
-
-  test "evaluates with nested start expression" do
-    assert_eval "foo bar", "foo <%= if true do %><%= if true do %>bar<% end %><% end %>"
-  end
-
-  test "evaluates with nested middle expression" do
-    assert_eval "foo baz", "foo <%= if true do %><%= if false do %>bar<% else %>baz<% end %><% end %>"
-  end
-
-  test "evaluates with defined variable" do
-    assert_eval "foo 1", "foo <% bar = 1 %><%= bar %>"
-  end
-
-  test "evaluates with require code" do
-    assert_eval "foo 1,2,3", "foo <% require Enum, as: E %><%= E.join [1, 2, 3], \",\" %>"
-  end
-
-  test "evaluates with end of token" do
-    assert_eval "foo bar %>", "foo bar %>"
-  end
-
-  test "raises a syntax error when the token is invalid" do
-    assert_raise EEx.SyntaxError, "missing token: %>", fn ->
-      EEx.compile_string "foo <%= bar"
-    end
-  end
-
-  test "raises a syntax error when end expression is found without a start expression" do
-    assert_raise EEx.SyntaxError, "unexpected token: ' end ' at line 1",  fn ->
-      EEx.compile_string "foo <% end %>"
-    end
-  end
-
-  test "raises a syntax error when start expression is found without an end expression" do
-    assert_raise EEx.SyntaxError, "unexpected end of string. expecting a closing <% end %>.", fn ->
-      EEx.compile_string "foo <% if true do %>"
-    end
-  end
-
-  test "raises a syntax error when nested end expression is found without an start expression" do
-    assert_raise EEx.SyntaxError, "unexpected token: ' end ' at line 1", fn ->
-      EEx.compile_string "foo <% if true do %><% end %><% end %>"
-    end
-  end
-
-  test "respects line numbers" do
-    expected = """
-foo
-2
-"""
-
-    string = """
-foo
-<%= __ENV__.line %>
-"""
-
-    assert_eval expected, string
-  end
-
-  test "respects line numbers inside nested expressions" do
-    expected = """
-foo
-
-3
-
-5
-"""
-
-    string = """
-foo
-<%= if true do %>
-<%= __ENV__.line %>
-<% end %>
-<%= __ENV__.line %>
-"""
-
-    assert_eval expected, string
-  end
-
-  test "respects line numbers inside start expression" do
-    expected = """
-foo
-
-true
-
-5
-"""
-
-    string = """
-foo
-<%= if __ENV__.line == 2 do %>
-<%= true %>
-<% end %>
-<%= __ENV__.line %>
-"""
-
-    assert_eval expected, string
-  end
-
-  test "respects line numbers inside middle expression with ->" do
-    expected = """
-foo
-
-true
-
-7
-"""
-
-    string = """
-foo
-<%= cond do %>
-<% false -> %> false
-<% __ENV__.line == 4 -> %>
-<%= true %>
-<% end %>
-<%= __ENV__.line %>
-"""
-
-    assert_eval expected, string
-  end
-
-  test "respects line number inside middle expressions with keywords" do
-    expected = """
-foo
-
-5
-
-7
-"""
-
-    string = """
-foo
-<%= if false do %>
-<%= __ENV__.line %>
-<% else %>
-<%= __ENV__.line %>
-<% end %>
-<%= __ENV__.line %>
-"""
-
-    assert_eval expected, string
-  end
-
-  test "properly handle functions" do
-    expected = """
-
-Number 1
-
-Number 2
-
-Number 3
-
-"""
-
-    string = """
-<%= Enum.map [1, 2, 3], fn x -> %>
-Number <%= x %>
-<% end %>
-"""
-
-    assert_eval expected, string
-  end
-
-  test "do not consider already finished functions" do
-    expected = """
-foo
-
-true
-
-"""
-
-    string = """
-foo
-<%= cond do %>
-<% false -> %> false
-<% fn -> 1 end -> %>
-<%= true %>
-<% end %>
-"""
-
-    assert_eval expected, string
-  end
-
-  test "evaluates nested do expressions" do
-    string = """
-    <% y = ["a", "b", "c"] %>
-    <%= cond do %>
-     <% "a" in y -> %>
-      Good
-     <% true -> %>
-      <% if true do %>true<% else %>false<% end %>
-      Bad
-    <% end %>
-    """
-
-    assert_eval "\n\n  Good\n \n", string
-  end
-
-  test "unicode" do
-    template = """
-      • <%= "•" %> •
-      <%= "Jößé Vâlìm" %> Jößé Vâlìm
-    """
-    result = EEx.eval_string(template)
-    assert result == "  • • •\n  Jößé Vâlìm Jößé Vâlìm\n"
-  end
-
-  test "evaluates the source from a given file" do
-    filename = Path.join(__DIR__, "fixtures/eex_template.eex")
-    result = EEx.eval_file(filename)
-    assert result == "foo bar.\n"
-  end
-
-  test "evaluates the source from a given file with bindings" do
-    filename = Path.join(__DIR__, "fixtures/eex_template_with_bindings.eex")
-    result = EEx.eval_file(filename, [bar: 1])
-    assert result == "foo 1\n"
-  end
-
-  test "raises an Exception when there's an error with the given file" do
-    assert_raise File.Error, "could not read file non-existent.eex: no such file or directory", fn ->
-      filename = "non-existent.eex"
-      EEx.compile_file(filename)
-    end
-  end
-
-  test "defined from string" do
-    assert EExText.Compiled.string_sample(1, 2) == "3"
-  end
-
-  test "defined from file" do
-    assert EExText.Compiled.file_sample(1) == "foo 1\n"
-  end
-
-  test "defined from file do not affect backtrace" do
-    assert EExText.Compiled.before_compile ==
-      { 8,
-        { EExText.Compiled,
-          :before_compile,
-          0,
-          [file: to_char_list(__FILE__), line: 7]
-        }
-      }
-
-    assert EExText.Compiled.after_compile ==
-      { 19,
-        { EExText.Compiled,
-          :after_compile,
-          0,
-          [file: to_char_list(__FILE__), line: 18]
-        }
-      }
-
-    assert EExText.Compiled.unknown ==
-      { 25,
-        { EExText.Compiled,
-          :unknown,
-          0,
-          [file: 'unknown', line: 24]
-        }
-      }
-  end
-
-  defp assert_eval(expected, actual) do
-    result = EEx.eval_string(actual, [], file: __FILE__, engine: EEx.Engine)
-    assert result == expected
-  end
-end
@@ -1 +0,0 @@
-foo <%= if true do %>bar.<% end %>
@@ -1 +0,0 @@
-foo <%= bar %>
@@ -1 +0,0 @@
-ExUnit.start [trace: "--trace" in System.argv]
@@ -1,86 +0,0 @@
-define(wrap_call(Line, Module, Function, Args),
-  { call, Line,
-    { remote, Line, { atom, Line, Module }, { atom, Line, Function } },
-    Args
-  }).
-
-define(atom_concat(Atoms), list_to_atom(lists:concat(Atoms))).
-define(elixir_macro(Macro), list_to_atom(lists:concat(['MACRO-',Macro]))).
-define(line(Opts), elixir_utils:get_line(Opts)).
-
-record(elixir_scope, {
-  context=nil,             %% can be assign, guards or nil
-  extra=nil,               %% extra information about the context, like fn_match for fns
-  noname=false,            %% when true, don't add new names (used by try)
-  super=false,             %% when true, it means super was invoked
-  caller=false,            %% when true, it means caller was invoked
-  module=nil,              %% the current module
-  function=nil,            %% the current function
-  vars=[],                 %% a dict of defined variables and their alias
-  backup_vars=nil,         %% a copy of vars to be used on ^var
-  temp_vars=nil,           %% a set of all variables defined in a particular assign
-  clause_vars=nil,         %% a dict of all variables defined in a particular clause
-  extra_guards=nil,        %% extra guards from args expansion
-  counter=[],              %% a counter for the variables defined
-  local=nil,               %% the scope to evaluate local functions against
-  context_modules=[],      %% modules defined in the current context
-  macro_aliases=[],        %% keep aliases defined inside a macro
-  macro_counter=0,         %% macros expansions counter
-  lexical_tracker=nil,     %% holds the lexical tracker pid
-  aliases,                 %% an orddict with aliases by new -> old names
-  file,                    %% the current scope filename
-  requires,                %% a set with modules required
-  macro_macros=[],         %% a list with macros imported from module inside a macro
-  macros,                  %% a list with macros imported from module
-  macro_functions=[],      %% a list with functions imported from module inside a macro
-  functions                %% a list with functions imported from module
-}).
-
-record(elixir_quote, {
-  line=nil,
-  context=nil,
-  vars_hygiene=true,
-  aliases_hygiene=true,
-  imports_hygiene=true,
-  unquote=true,
-  unquoted=false,
-  escape=false
-}).
-
-record(elixir_tokenizer, {
-  file,
-  terminators=[],
-  check_terminators=true,
-  existing_atoms_only=false
-}).
-
-%% Introspection
-define(defs(Kind), Kind == def; Kind == defp; Kind == defmacro; Kind == defmacrop).
-
-%% Used in tokenization and interpolation
-
-%% Numbers
-define(is_hex(S), ?is_digit(S) orelse (S >= $A andalso S =< $F) orelse (S >= $a andalso S =< $f)).
-define(is_bin(S), S >= $0 andalso S =< $1).
-define(is_octal(S), S >= $0 andalso S =< $7).
-define(is_leading_octal(S), S >= $0 andalso S =< $3).
-
-%% Digits and letters
-define(is_digit(S), S >= $0 andalso S =< $9).
-define(is_upcase(S), S >= $A andalso S =< $Z).
-define(is_downcase(S), S >= $a andalso S =< $z).
-
-%% Atoms
-define(is_atom_start(S), ?is_quote(S) orelse ?is_upcase(S) orelse ?is_downcase(S) orelse (S == $_)).
-define(is_atom(S), ?is_identifier(S) orelse (S == $@)).
-
-define(is_identifier(S), ?is_digit(S) orelse ?is_upcase(S) orelse ?is_downcase(S) orelse (S == $_)).
-define(is_terminator(S), (S == $?) orelse (S == $!) orelse (S == $:)).
-
-%% Quotes
-define(is_quote(S), S == $" orelse S == $').
-
-%% Spaces
-define(is_horizontal_space(S), (S == $\s) orelse (S == $\t)).
-define(is_vertical_space(S), (S == $\r) orelse (S == $\n)).
-define(is_space(S), ?is_horizontal_space(S) orelse ?is_vertical_space(S)).
@@ -1,56 +0,0 @@
-import Kernel, except: [access: 2]
-
-defprotocol Access do
-  @moduledoc """
-  The Access protocol is the underlying protocol invoked
-  when the brackets syntax is used. For instance, `foo[bar]`
-  is translated to `access foo, bar` which, by default,
-  invokes the `Access.access` protocol.
-
-  This protocol is limited and is implemented only for the
-  following built-in types: keywords, records and functions.
-  """
-
-  @doc """
-  Receives the element being accessed and the access item.
-  """
-  def access(container, key)
-end
-
-defimpl Access, for: List do
-  @doc """
-  Access the given key in a tuple list.
-  The key is found via the `===` operator.
-
-  ## Examples
-
-      iex> keywords = [a: 1, b: 2]
-      ...> keywords[:a]
-      1
-
-      iex> star_ratings = [{1.0, "★"}, {1.5, "★☆"}, {2.0, "★★"}]
-      ...> star_ratings[1.5]
-      "★☆"
-
-  """
-  def access(dict, key)
-  def access([{ key, value }|_], key), do: value
-  def access([{ _, _ }|t], key), do: access(t, key)
-  def access([], _key), do: nil
-end
-
-defimpl Access, for: Atom do
-  @doc """
-  The access protocol can only be accessed by atoms
-  at compilation time. If we reach this, we should raise
-  an exception.
-  """
-  def access(nil, _) do
-    nil
-  end
-
-  def access(atom, _) do
-    raise "The access protocol can only be invoked for atoms at " <>
-      "compilation time, tried to invoke it for #{inspect atom}"
-  end
-end
@@ -1,105 +0,0 @@
-defmodule Application.Behaviour do
-  @moduledoc """
-  Default callbacks for applications.
-
-  In Erlang/OTP, an application is a component that can be started
-  and stopped as a unit, and which can be reused in other systems.
-
-  The first step in creating an application is to define an application specification.
-  For example, if your application is named `:my_app`, an app specification
-  should exist at `ebin/my_app.app`. This file is usually defined by
-  build tools like Mix.
-
-  With the app specification in hand, we must define
-  application module callbacks that control how to start and stop
-  instances of the application. This module is about defining such callbacks.
-
-  There are two callbacks which must be implemented:
-
-  1. `start(type, args)` - must return `{ :ok, pid }` or
-     `{ :ok, pid, state }`, where `pid` is the process identifier
-     of the supervisor tree root and `state` is application defined 
-     state information;
-
-  2. `stop(state)` - receives the `state` returned by `start` and should
-     do any necessary clean up. Notice that shutting down the supervisor
-     is automatically handled by the VM;
-
-  When using this module, it tags the module behaviour as
-  `:application` and provides a default `stop/1` callback. The `start/2` callback
-  still needs to be implemented by the user.
-
-  You can learn more about the `:application` module, the application
-  specification and the application module callbacks from these sources:
-
-  * http://www.erlang.org/doc/man/application.html
-  * http://www.erlang.org/doc/design_principles/applications.html
-  * http://learnyousomeerlang.com/building-otp-applications
-
-  ## Example
-
-      defmodule MyApp do
-        use Application.Behaviour
-
-        def start(_type, args) do
-          MyApp.Sup.start_link(args)
-        end
-      end
-
-  """
-
-  # Starts the given application and all of its dependencies that
-  # have not been started yet recursively.
-  #
-  # ## Supported types
-  #
-  # When starting an application, a type can be given:
-  #
-  # * `:permanent` - If a permanent application terminates, all other
-  #    applications and the runtime system are also terminated;
-  # * `:transient` - If a transient application terminates with reason
-  #    `:normal`, this is reported but no other applications are terminated.
-  #    If a transient application terminates abnormally, all other
-  #    applications and the runtime system are also terminated;
-  # * `:temporary` -  If a temporary application terminates, this is reported
-  #    but no other applications are terminated.
-  #
-  # The type only applies to the application being started. Its dependencies
-  # are all started with default type (which is :temporary).
-  #
-  # Note that transient mode is of little practical use, since when a
-  # supervision tree terminates, the reason is set to shutdown, not normal.
-  #
-  # ## Examples
-  #
-  #     Application.Behaviour.start(:my_app)
-  #
-  @doc false
-  def start(app, type // :temporary) do
-    case :application.start(app, type) do
-      { :error, { :not_started, dep } } ->
-        case start(dep) do
-          :ok   -> start(app, type)
-          other -> other
-        end
-      { :error, { :already_started, _ } } ->
-        :ok
-      other ->
-        other
-    end
-  end
-
-  @doc false
-  defmacro __using__(_) do
-    quote location: :keep do
-      @behavior :application
-
-      @doc false
-      def stop(_state) do
-        :ok
-      end
-
-      defoverridable [stop: 1]
-    end
-  end
-end
@@ -1,142 +0,0 @@
-defmodule Behaviour do
-  @moduledoc """
-  Utilities for defining behaviour intefaces.
-
-  Behaviours can be referenced by other modules
-  to ensure they implement required callbacks.
-
-  For example, you can specify the `URI.Parser`
-  behaviour as follows:
-
-      defmodule URI.Parser do
-        use Behaviour
-
-        @doc "Parses the given URL"
-        defcallback parse(uri_info :: URI.Info.t) :: URI.Info.t
-
-        @doc "Defines a default port"
-        defcallback default_port() :: integer
-      end
-
-  And then a module may use it as:
-
-      defmodule URI.HTTP do
-        @behaviour URI.Parser
-        def default_port(), do: 80
-        def parse(info), do: info
-      end
-
-  If the behaviour changes or `URI.HTTP` does
-  not implement one of the callbacks, a warning
-  will be raised.
-
-  ## Implementation
-
-  Since Erlang R15, behaviours must be defined via
-  `@callback` attributes. `defcallback` is a simple
-  mechanism that defines the `@callback` attribute
-  according to the given type specification. `defcallback` allows
-  documentsion to be created for the callback and defines 
-  a custom function signature.
-
-  The callbacks and their documentation can be retrieved
-  via the `__behaviour__` callback function.
-  """
-
-  @doc """
-  Define a function callback according to the given type specification.
-  """
-  defmacro defcallback({ :::, _, [fun, return] }) do
-    do_defcallback(fun, return, __CALLER__)
-  end
-
-  defmacro defcallback(fun) do
-    do_defcallback(fun, quote(do: term), __CALLER__)
-  end
-
-  @doc """
-  Define a macro callback according to the given type specification.
-  """
-  defmacro defmacrocallback({ :::, _, [fun, return] }) do
-    do_defmacrocallback(fun, return, __CALLER__)
-  end
-
-  defmacro defmacrocallback(fun) do
-    do_defmacrocallback(fun, quote(do: Macro.t), __CALLER__)
-  end
-
-  defp do_defcallback(fun, spec, caller) do
-    case Macro.extract_args(fun) do
-      { name, args } ->
-        do_callback(:def, name, args, name, length(args), spec, caller)
-      :error ->
-        raise ArgumentError, message: "invalid syntax in defcallback #{Macro.to_string(fun)}"
-    end
-  end
-
-  defp do_defmacrocallback(fun, spec, caller) do
-    case Macro.extract_args(fun) do
-      { name, args } ->
-        do_callback(:defmacro, :"MACRO-#{name}", [quote(do: env :: Macro.Env.t)|args], name, length(args), spec, caller)
-      :error ->
-        raise ArgumentError, message: "invalid syntax in defmacrocallback #{Macro.to_string(fun)}"
-    end
-  end
-
-  defp do_callback(kind, name, args, docs_name, docs_arity, return, caller) do
-    Enum.each args, fn
-      { :::, _, [left, right] } ->
-        ensure_not_default(left)
-        ensure_not_default(right)
-        left
-      other ->
-        ensure_not_default(other)
-        other
-    end
-
-    quote do
-      @callback unquote(name)(unquote_splicing(args)) :: unquote(return)
-      Behaviour.store_docs __MODULE__, unquote(caller.line), unquote(kind), unquote(docs_name), unquote(docs_arity)
-    end
-  end
-
-  defp ensure_not_default({ ://, _, [_, _] }) do
-    raise ArgumentError, message: "default arguments // not supported in defcallback"
-  end
-
-  defp ensure_not_default(_), do: :ok
-
-  @doc false
-  def store_docs(module, line, kind, name, arity) do
-    doc = Module.get_attribute module, :doc
-    Module.delete_attribute module, :doc
-    Module.put_attribute module, :behaviour_docs, { { name, arity }, line, kind, doc }
-  end
-
-  @doc false
-  defmacro __using__(_) do
-    quote do
-      Module.register_attribute(__MODULE__, :behaviour_docs, accumulate: true)
-      @before_compile unquote(__MODULE__)
-      import unquote(__MODULE__)
-    end
-  end
-
-  @doc false
-  defmacro __before_compile__(env) do
-    docs = if Code.compiler_options[:docs] do
-      Enum.reverse Module.get_attribute(env.module, :behaviour_docs)
-    end
-
-    quote do
-      @doc false
-      def __behaviour__(:callbacks) do
-        __MODULE__.behaviour_info(:callbacks)
-      end
-
-      def __behaviour__(:docs) do
-        unquote(Macro.escape(docs))
-      end
-    end
-  end
-end
@@ -1,128 +0,0 @@
-defmodule Bitwise do
-  @moduledoc """
-  This module provide macros and operators for bitwise operators.
-  These macros can be used in guards.
-
-  The easiest way to use is to simply import them into
-  your module:
-
-      iex> use Bitwise
-      iex> bnot 1
-      -2
-      iex> 1 &&& 1
-      1
-
-  You can select to include only or skip operators by passing options:
-
-      iex> use Bitwise, only_operators: true
-      ...> 1 &&& 1
-      1
-
-  """
-
-  @doc """
-  Allow a developer to use this module in their programs with
-  the following options:
-
-  * `:only_operators` - Include only operators;
-  * `:skip_operators` - Skip operators;
-
-  """
-  defmacro __using__(options) do
-    except = cond do
-      Keyword.get(options, :only_operators) ->
-        [bnot: 1, band: 2, bor: 2, bxor: 2, bsl: 2, bsr: 2]
-      Keyword.get(options, :skip_operators) ->
-        [~~~: 1, &&&: 2, |||: 2, ^^^: 2, <<<: 2, >>>: 2]
-      true -> []
-    end
-
-    quote do
-      import Bitwise, except: unquote(except)
-    end
-  end
-
-  @doc """
-  Bitwise not.
-  """
-  defmacro bnot(expr) do
-    quote do: __op__(:bnot, unquote(expr))
-  end
-
-  @doc """
-  Bitwise not as operator.
-  """
-  defmacro ~~~expr do
-    quote do: __op__(:bnot, unquote(expr))
-  end
-
-  @doc """
-  Bitwise and.
-  """
-  defmacro band(left, right) do
-    quote do: __op__(:band, unquote(left), unquote(right))
-  end
-
-  @doc """
-  Bitwise and as operator.
-  """
-  defmacro left &&& right do
-    quote do: __op__(:band, unquote(left), unquote(right))
-  end
-
-  @doc """
-  Bitwise or.
-  """
-  defmacro bor(left, right) do
-    quote do: __op__(:bor, unquote(left), unquote(right))
-  end
-
-  @doc """
-  Bitwise or as operator.
-  """
-  defmacro left ||| right do
-    quote do: __op__(:bor, unquote(left), unquote(right))
-  end
-
-  @doc """
-  Bitwise xor.
-  """
-  defmacro bxor(left, right) do
-    quote do: __op__(:bxor, unquote(left), unquote(right))
-  end
-
-  @doc """
-  Bitwise xor as operator.
-  """
-  defmacro left ^^^ right do
-    quote do: __op__(:bxor, unquote(left), unquote(right))
-  end
-
-  @doc """
-  Arithmetic bitshift left.
-  """
-  defmacro bsl(left, right) do
-    quote do: __op__(:bsl, unquote(left), unquote(right))
-  end
-
-  @doc """
-  Arithmetic bitshift left as operator.
-  """
-  defmacro left <<< right do
-    quote do: __op__(:bsl, unquote(left), unquote(right))
-  end
-
-  @doc """
-  Arithmetic bitshift right.
-  """
-  defmacro bsr(left, right) do
-    quote do: __op__(:bsr, unquote(left), unquote(right))
-  end
-
-  @doc """
-  Arithmetic bitshift right as operator.
-  """
-  defmacro left >>> right do
-    quote do: __op__(:bsr, unquote(left), unquote(right))
-  end
-end
@@ -1,484 +0,0 @@
-defmodule Code do
-  defexception LoadError, file: nil do
-    def message(exception) do
-      "could not load #{exception.file}"
-    end
-  end
-
-  @moduledoc """
-  Utilities for managing code compilation, code evaluation and code loading.
-
-  This module complements [Erlang's code module](http://www.erlang.org/doc/man/code.html)
-  to add behavior which is specific to Elixir.
-
-  """
-
-  @doc """
-  List all loaded files.
-  """
-  def loaded_files do
-    :elixir_code_server.call :loaded
-  end
-
-  @doc """
-  Remove files from the loaded files list.
-
-  The modules defined in the file are not removed;
-  calling this function only removes them from the list,
-  allowing them to be required again.
-  """
-  def unload_files(files) do
-    :elixir_code_server.cast { :unload_files, files }
-  end
-
-  @doc """
-  Append a path to the Erlang VM code path.
-
-  The path is expanded with `Path.expand/1` before being appended.
-  """
-  def append_path(path) do
-    :code.add_pathz(Path.expand to_char_list(path))
-  end
-
-  @doc """
-  Prepend a path to the Erlang VM code path.
-
-  The path is expanded with `Path.expand/1` before being prepended.
-  """
-  def prepend_path(path) do
-    :code.add_patha(Path.expand to_char_list(path))
-  end
-
-  @doc """
-  Delete a path from the Erlang VM code path.
-
-  The path is expanded with `Path.expand/1` before being deleted.
-  """
-  def delete_path(path) do
-    :code.del_path(Path.expand to_char_list(path))
-  end
-
-  @doc """
-  Evaluate the contents given by `string`. 
-
-  The `binding` argument is a keyword list of variable bindings. 
-  The `opts` argument is a keyword list of environment options. 
-  
-  Those options can be:
-
-  * `:file` - the file to be considered in the evaluation
-  * `:line` - the line on which the script starts
-  * `:delegate_locals_to` - delegate local calls to the given module,
-    the default is to not delegate
-
-  Additionally, the following scope values can be configured:
-
-  * `:aliases` - a list of tuples with the alias and its target
-  * `:requires` - a list of modules required
-  * `:functions` - a list of tuples where the first element is a module
-    and the second a list of imported function names and arity. The list
-    of function names and arity must be sorted
-  * `:macros` - a list of tuples where the first element is a module
-    and the second a list of imported macro names and arity. The list
-    of function names and arity must be sorted
-
-  Notice that setting any of the values above overrides Elixir's default
-  values. For example, setting `:requires` to `[]`, will no longer
-  automatically require the `Kernel` module; in the same way setting
-  `:macros` will no longer auto-import `Kernel` macros like `if`, `case`,
-  etc.
-
-  Returns a tuple of the form `{ value, binding }`,
-  where `value` is the the value returned from evaluating `string`.
-  If an error occurs while evaluating `string` an exception will be raised.
-
-  `binding` is a keyword list with the value of all variable bindings
-  after evaluating `string`. The binding key is usually an atom, but it
-  may be a tuple for variables defined in a different context.
-
-  ## Examples
-
-      iex> Code.eval_string("a + b", [a: 1, b: 2], file: __ENV__.file, line: __ENV__.line)
-      {3, [a: 1, b: 2]}
-
-      iex> Code.eval_string("c = a + b", [a: 1, b: 2], __ENV__)
-      {3, [a: 1, b: 2, c: 3]}
-
-      iex> Code.eval_string("a = a + b", [a: 1, b: 2])
-      {3, [a: 3, b: 2]}
-
-  For convenience, you can pass `__ENV__` as the `opts` argument and
-  all imports, requires and aliases defined in the current environment
-  will be automatically carried over:
-
-      iex> Code.eval_string("a + b", [a: 1, b: 2], __ENV__)
-      {3, [a: 1, b: 2]}
-
-  """
-  def eval_string(string, binding // [], opts // [])
-
-  def eval_string(string, binding, Macro.Env[] = env) do
-    do_eval_string(string, binding, env.to_keywords)
-  end
-
-  def eval_string(string, binding, opts) when is_list(opts) do
-    validate_eval_opts(opts)
-    do_eval_string(string, binding, opts)
-  end
-
-  defp do_eval_string(string, binding, opts) when is_list(binding) do
-    { value, binding, _scope } = :elixir.eval to_char_list(string), binding, opts
-    { value, binding }
-  end
-
-  @doc """
-  Evaluate the quoted contents.
-
-  See `eval_string/3` for a description of arguments and return values.
-
-  ## Examples
-
-      iex> contents = quote(hygiene: [vars: false], do: a + b)
-      ...> Code.eval_quoted(contents, [a: 1, b: 2], file: __ENV__.file, line: __ENV__.line)
-      {3, [a: 1, b: 2]}
-
-  For convenience, you can pass `__ENV__` as the `opts` argument and
-  all options will be automatically extracted from the current environment:
-
-      iex> contents = quote(hygiene: [vars: false], do: a + b)
-      ...> Code.eval_quoted(contents, [a: 1, b: 2], __ENV__)
-      {3, [a: 1, b: 2]}
-
-  """
-  def eval_quoted(quoted, binding // [], opts // [])
-
-  def eval_quoted(quoted, binding, Macro.Env[] = env) do
-    do_eval_quoted(quoted, binding, env.to_keywords)
-  end
-
-  def eval_quoted(quoted, binding, opts) when is_list(opts) do
-    validate_eval_opts(opts)
-    do_eval_quoted(quoted, binding, opts)
-  end
-
-  defp do_eval_quoted(quoted, binding, opts) when is_list(binding) do
-    { value, binding, _scope } =
-      :elixir.eval_quoted [quoted], binding, opts
-    { value, binding }
-  end
-
-  defp validate_eval_opts(opts) do
-    if f = opts[:functions], do: validate_imports(:functions, f)
-    if m = opts[:macros],    do: validate_imports(:macros, m)
-    if a = opts[:aliases],   do: validate_aliases(:aliases, a)
-    if r = opts[:requires],  do: validate_requires(:requires, r)
-  end
-
-  defp validate_requires(kind, requires) do
-    valid = is_list(requires) and Enum.all?(requires, &is_atom(&1))
-
-    unless valid do
-      raise ArgumentError, message: "expected :#{kind} option given to eval in the format: [module]"
-    end
-  end
-
-  defp validate_aliases(kind, aliases) do
-    valid = is_list(aliases) and Enum.all?(aliases, fn { k, v } ->
-      is_atom(k) and is_atom(v)
-    end)
-
-    unless valid do
-      raise ArgumentError, message: "expected :#{kind} option given to eval in the format: [{ module, module }]"
-    end
-  end
-
-  defp validate_imports(kind, imports) do
-    valid = is_list(imports) and Enum.all?(imports, fn { k, v } ->
-      is_atom(k) and is_list(v) and Enum.all?(v, fn { name, arity } ->
-        is_atom(name) and is_integer(arity)
-      end)
-    end)
-
-    unless valid do
-      raise ArgumentError, message: "expected :#{kind} option given to eval in the format: [{ module, [{ name, arity }] }]"
-    end
-  end
-
-  @doc """
-  Convert the given string to its quoted form. 
-  
-  Returns `{ :ok, quoted_form }`
-  if it succeeds, `{ :error, { line, error, token } }` otherwise.
-
-  ## Options
-
-  * `:file` - The filename to be used in stacktraces
-    and the file reported in the `__ENV__` variable.
-
-  * `:line` - The line reported in the `__ENV__` variable.
-
-  * `:existing_atoms_only` - When `true`, raises an error
-    when non-existing atoms are found by the tokenizer.
-
-  ## Macro.to_string/2
-
-  The opposite of converting a string to its quoted form is
-  `Macro.to_string/2`, which converts a quoted form to a string/binary
-  representation.
-  """
-  def string_to_quoted(string, opts // []) when is_list(opts) do
-    file = Keyword.get opts, :file, "nofile"
-    line = Keyword.get opts, :line, 1
-    res  = :elixir_translator.forms(to_char_list(string), line, file, opts)
-
-    case res do
-      { :ok, forms } -> { :ok, unpack_quote(line, forms) }
-      _ -> res
-    end
-  end
-
-  @doc """
-  Convert the given string to its quoted form. 
-  
-  It returns the ast if it succeeds,
-  raises an exception otherwise. The exception is a `TokenMissingError`
-  in case a token is missing (usually because the expression is incomplete),
-  `SyntaxError` otherwise.
-
-  Check `string_to_quoted/2` for options information.
-  """
-  def string_to_quoted!(string, opts // []) when is_list(opts) do
-    file = Keyword.get opts, :file, "nofile"
-    line = Keyword.get opts, :line, 1
-    res  = :elixir_translator.forms!(to_char_list(string), line, file, opts)
-    unpack_quote(line, res)
-  end
-
-  defp unpack_quote(_line, []),                              do: nil
-  defp unpack_quote(_line, [forms]) when not is_list(forms), do: forms
-  defp unpack_quote(line, forms),                            do: { :__block__, [line: line], forms }
-
-  @doc """
-  Load the given file. 
-  
-  Accepts `relative_to` as an argument to tell where
-  the file is located. If the file was already required/loaded, loads it again.
-  It returns a list of tuples `{ ModuleName, <<byte_code>> }`, one tuple for each
-  module defined in the file.
-
-  Notice that if `load_file` is invoked by different processes
-  concurrently, the target file will be invoked concurrently
-  many times. I.e. if `load_file` is called N times with
-  a given file, the given file will be loaded N times. Check
-  `require_file/2` if you don't want a file to be loaded concurrently.
-  """
-  def load_file(file, relative_to // nil) when is_binary(file) do
-    file = find_file(file, relative_to)
-    :elixir_code_server.call { :acquire, file }
-    loaded = :elixir_compiler.file file
-    :elixir_code_server.cast { :loaded, file }
-    loaded
-  end
-
-  @doc """
-  Require the given `file`. 
-  
-  Accepts `relative_to` as an argument to tell where
-  the file is located. The return value is the same as that of `load_file/2`. If
-  the file was already required/loaded, doesn't do anything and returns `nil`.
-
-  Notice that if `require_file` is invoked by different processes concurrently,
-  the first process to invoke `require_file` acquires a lock and the remaining
-  ones will block until the file is available. I.e. if `require_file` is called
-  N times with a given file, it will be loaded only once. The first process to
-  call `require_file` will get the list of loaded modules, others will get `nil`.
-
-  Check `load_file/2` if you want a file to be loaded concurrently.
-  """
-  def require_file(file, relative_to // nil) when is_binary(file) do
-    file = find_file(file, relative_to)
-
-    case :elixir_code_server.call({ :acquire, file }) do
-      :loaded  ->
-        nil
-      { :queued, ref }  ->
-        receive do { :elixir_code_server, ^ref, :loaded } -> nil end
-      :proceed ->
-        loaded = :elixir_compiler.file file
-        :elixir_code_server.cast { :loaded, file }
-        loaded
-    end
-  end
-
-  @doc """
-  Load the compilation options from the code server.
-
-  Check `compiler_options/1` for more information.
-  """
-  def compiler_options do
-    :elixir_code_server.call :compiler_options
-  end
-
-  @doc """
-  Set compilation options. 
-
-  These options are global since they are stored by Elixir's Code Server.
-
-  Available options are:
-
-  * `:docs` - when `true`, retain documentation in the compiled module,
-    `true` by default;
-
-  * `:debug_info` - when `true`, retain debug information in the compiled module.
-    This allows a developer to reconstruct the original source
-    code, `false` by default;
-
-  * `:ignore_module_conflict` - when `true`, override modules that were already defined
-    without raising errors, `false` by default;
-
-  * `:warnings_as_errors` - cause compilation to fail when warnings are generated;
-
-  """
-  def compiler_options(opts) do
-    :elixir_code_server.cast { :compiler_options, opts }
-  end
-
-  @doc """
-  Compile the given string. 
-
-  Returns a list of tuples where
-  the first element is the module name and the second one is its
-  binary.
-
-  For compiling many files at once, check `Kernel.ParallelCompiler.files/2`.
-  """
-  def compile_string(string, file // "nofile") when is_binary(file) do
-    :elixir_compiler.string String.to_char_list!(string), file
-  end
-
-  @doc """
-  Compile the quoted expression. 
-  
-  Returns a list of tuples where
-  the first element is the module name and the second one is its
-  binary.
-  """
-  def compile_quoted(quoted, file // "nofile") when is_binary(file) do
-    :elixir_compiler.quoted [quoted], file
-  end
-
-  @doc """
-  Ensure the given module is loaded. 
-  
-  If the module is already
-  loaded, this works as no-op. If the module was not yet loaded,
-  it tries to load it.
-
-  If it succeeds loading the module, it returns
-  `{ :module, module }`. If not, returns `{ :error, reason }` with
-  the error reason.
-
-  ## Code loading on the Erlang VM
-
-  Erlang has two modes to load code: interactive and embedded.
-
-  By default, the Erlang VM runs in interactive mode, where modules
-  are loaded as needed. In embedded mode the opposite happens, as all
-  modules need to be loaded upfront or explicitly.
-
-  Therefore, this function is used to check if a module is loaded
-  before using it and allows one to react accordingly. For example, the `URI`
-  module uses this function to check if a specific parser exists for a given
-  URI scheme.
-
-  ## Code.ensure_compiled
-
-  Elixir also contains an `ensure_compiled/1` function that is a
-  superset of `ensure_loaded/1`.
-
-  Since Elixir's compilation happens in parallel, in some situations
-  you may need to use a module but that was not yet compiled, therefore
-  it can't even be loaded.
-
-  `ensure_compiled/1` halts the current process until the
-  module we are depending on is available.
-
-  In most cases, `ensure_loaded` is enough. `ensure_compiled`
-  must be used in some rare cases, usually involving macros
-  that need to invoke a module for callback information.
-  """
-  def ensure_loaded(module) when is_atom(module) do
-    :code.ensure_loaded(module)
-  end
-
-  @doc """
-  Ensure the given module is loaded.
-
-  Similar to `ensure_loaded/1`, but returns `true` if the module
-  is already loaded or was successfully loaded. Returns `false` otherwise.
-  """
-  def ensure_loaded?(module) do
-    match?({ :module, ^module }, ensure_loaded(module))
-  end
-
-  @doc """
-  Ensure the given module is compiled and loaded. 
-  
-  If the module
-  is already loaded, it works as no-op. If the module was not
-  loaded yet, it checks if it needs to be compiled first and
-  then tries to load it.
-
-  If it succeeds loading the module, it returns
-  `{ :module, module }`. If not, returns `{ :error, reason }` with
-  the error reason.
-
-  Check `ensure_loaded/1` for more information on module loading
-  and when to use `ensure_loaded/1` or `ensure_compiled/1`.
-  """
-  def ensure_compiled(module) when is_atom(module) do
-    case :code.ensure_loaded(module) do
-      { :error, :nofile } = error ->
-        case :erlang.get(:elixir_ensure_compiled) do
-          :undefined -> error
-          _ ->
-            try do
-              module.__info__(:module)
-              { :module, module }
-            rescue
-              UndefinedFunctionError -> error
-            end
-        end
-      other -> other
-    end
-  end
-
-  @doc """
-  Ensure the given module is compiled and loaded. 
-
-  Similar to `ensure_compiled/1`, but returns `true` if the module
-  is already loaded or was successfully loaded and compiled. 
-  Returns `false` otherwise.
-  """
-  def ensure_compiled?(module) do
-    match?({ :module, ^module }, ensure_compiled(module))
-  end
-
-  ## Helpers
-
-  # Finds the file given the relative_to path.
-  # If the file is found, returns its path in binary, fails otherwise.
-  defp find_file(file, relative_to) do
-    file = if relative_to do
-      Path.expand(file, relative_to)
-    else
-      Path.expand(file)
-    end
-
-    if File.regular?(file) do
-      file
-    else
-      raise LoadError, file: file
-    end
-  end
-end
@@ -1,511 +0,0 @@
-defmodule Dict do
-  @moduledoc %S"""
-  This module specifies the Dict API expected to be
-  implemented by different dictionaries. It also provides
-  functions that redirect to the underlying Dict, allowing
-  a developer to work with different Dict implementations
-  using one API.
-
-  To create a new dict, use the `new` functions defined
-  by each dict type:
-
-      HashDict.new  #=> creates an empty HashDict
-
-  In the examples below, `dict_impl` means a specific
-  `Dict` implementation, for example `HashDict` or `ListDict`.
-
-  ## Protocols
-
-  Besides implementing the functions in this module, all
-  dictionaries are required to implement the `Access`
-  protocol:
-
-      iex> dict = dict_impl.new
-      ...> dict = Dict.put(dict, :hello, :world)
-      ...> dict[:hello]
-      :world
-
-  And the `Enumerable` protocol, allowing one to write:
-
-      Enum.each(dict, fn ({ k, v }) ->
-        IO.puts "#{k}: #{v}"
-      end)
-
-  ## Match
-
-  Dictionaries are required to implement all operations
-  using the match (`===`) operator. Any deviation from
-  this behaviour should be avoided and explicitly documented.
-  """
-
-  use Behaviour
-
-  @type key :: any
-  @type value :: any
-  @type keys :: [ key ]
-  @type t :: tuple | list
-
-  defcallback new :: t
-  defcallback new(Keyword.t) :: t
-  defcallback delete(t, key) :: t
-  defcallback drop(t, keys) :: t
-  defcallback empty(t) :: t
-  defcallback equal?(t, t) :: boolean
-  defcallback get(t, key) :: value
-  defcallback get(t, key, value) :: value
-  defcallback fetch(t, key) :: { :ok, value } | :error
-  defcallback fetch!(t, key) :: value | no_return
-  defcallback has_key?(t, key) :: boolean
-  defcallback keys(t) :: list(key)
-  defcallback merge(t, t) :: t
-  defcallback merge(t, t, (key, value, value -> value)) :: t
-  defcallback pop(t, key) :: {value, t}
-  defcallback pop(t, key, value) :: {value, t}
-  defcallback put(t, key, value) :: t
-  defcallback put_new(t, key, value) :: t
-  defcallback size(t) :: non_neg_integer()
-  defcallback split(t, keys) :: {t, t}
-  defcallback take(t, keys) :: t
-  defcallback to_list(t) :: list()
-  defcallback update(t, key, value, (value -> value)) :: t
-  defcallback update!(t, key, (value -> value)) :: t | no_return
-  defcallback values(t) :: list(value)
-  defcallback reduce(t, any, ({key, value}, any -> any)) :: any
-
-  defmacrop target(dict) do
-    quote do
-      cond do
-        is_tuple(unquote(dict)) ->
-          elem(unquote(dict), 0)
-        is_list(unquote(dict)) ->
-          ListDict
-        true ->
-          unsupported_dict(unquote(dict))
-      end
-    end
-  end
-
-  @doc """
-  Returns a list of all keys in `dict`.
-  The keys are not guaranteed to be in any order.
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1, b: 2])
-      ...> Enum.sort(Dict.keys(d))
-      [:a,:b]
-
-  """
-  @spec keys(t) :: [key]
-  def keys(dict) do
-    target(dict).keys(dict)
-  end
-
-  @doc """
-  Returns a list of all values in `dict`
-  The values are not guaranteed to be in any order.
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1, b: 2])
-      ...> Enum.sort(Dict.values(d))
-      [1,2]
-
-  """
-  @spec values(t) :: [value]
-  def values(dict) do
-    target(dict).values(dict)
-  end
-
-  @doc """
-  Returns the number of elements in `dict`.
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1, b: 2])
-      ...> Dict.size(d)
-      2
-
-  """
-  @spec size(t) :: non_neg_integer
-  def size(dict) do
-    target(dict).size(dict)
-  end
-
-  @doc """
-  Returns whether the given `key` exists in the given `dict`.
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1])
-      iex> Dict.has_key?(d, :a)
-      true
-      iex> Dict.has_key?(d, :b)
-      false
-
-  """
-  @spec has_key?(t, key) :: boolean
-  def has_key?(dict, key) do
-    target(dict).has_key?(dict, key)
-  end
-
-  @doc """
-  Returns the value associated with `key` in `dict`. If `dict` does not
-  contain `key`, returns `default` (or `nil` if not provided).
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1])
-      iex> Dict.get(d, :a)
-      1
-      iex> Dict.get(d, :b)
-      nil
-      iex> Dict.get(d, :b, 3)
-      3
-  """
-  @spec get(t, key, value) :: value
-  def get(dict, key, default // nil) do
-    target(dict).get(dict, key, default)
-  end
-
-  @doc """
-  Returns `{ :ok, value }` associated with `key` in `dict`.
-  If `dict` does not contain `key`, returns `:error`.
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1])
-      iex> Dict.fetch(d, :a)
-      { :ok, 1 }
-      iex> Dict.fetch(d, :b)
-      :error
-
-  """
-  @spec fetch(t, key) :: value
-  def fetch(dict, key) do
-    target(dict).fetch(dict, key)
-  end
-
-  @doc """
-  Returns the value associated with `key` in `dict`. If `dict` does not
-  contain `key`, it raises `KeyError`.
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1])
-      iex> Dict.fetch!(d, :a)
-      1
-      iex> Dict.fetch!(d, :b)
-      ** (KeyError) key not found: :b
-
-  """
-  @spec fetch!(t, key) :: value | no_return
-  def fetch!(dict, key) do
-    target(dict).fetch!(dict, key)
-  end
-
-  @doc """
-  Stores the given `value` under `key` in `dict`.
-  If `dict` already has `key`, the stored value is replaced by the new one.
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1, b: 2])
-      ...> d = Dict.put(d, :a, 3)
-      ...> Dict.get(d, :a)
-      3
-
-  """
-  @spec put(t, key, value) :: t
-  def put(dict, key, val) do
-    target(dict).put(dict, key, val)
-  end
-
-  @doc """
-  Puts the given `value` under `key` in `dict` unless `key` already exists.
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1, b: 2])
-      ...> d = Dict.put_new(d, :a, 3)
-      ...> Dict.get(d, :a)
-      1
-
-  """
-  @spec put_new(t, key, value) :: t
-  def put_new(dict, key, val) do
-    target(dict).put_new(dict, key, val)
-  end
-
-  @doc """
-  Removes the entry stored under the given `key` from `dict`.
-  If `dict` does not contain `key`, returns the dictionary unchanged.
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1, b: 2])
-      ...> d = Dict.delete(d, :a)
-      ...> Dict.get(d, :a)
-      nil
-
-      iex> d = dict_impl.new([b: 2])
-      ...> Dict.delete(d, :a) == d
-      true
-
-  """
-  @spec delete(t, key) :: t
-  def delete(dict, key) do
-    target(dict).delete(dict, key)
-  end
-
-  @doc """
-  Merges the given `enum` into `dict`. If one of the `enum` keys
-  already exists in `dict`, the `dict` value is replaced by the `enum`
-  value.
-
-  The `enum` must yield tuples with two elements on enumeration,
-  where the first element represents the key and the second the value.
-
-  ## Examples
-
-      iex> d1 = dict_impl.new([a: 1, b: 2])
-      ...> d2 = dict_impl.new([a: 3, d: 4])
-      ...> d = Dict.merge(d1, d2)
-      ...> [a: Dict.get(d, :a), b: Dict.get(d, :b), d: Dict.get(d, :d)]
-      [a: 3, b: 2, d: 4]
-
-  """
-  @spec merge(t, t) :: t
-  def merge(dict, enum) do
-    merge(dict, enum, fn(_k, _v1, v2) -> v2 end)
-  end
-
-  @doc """
-  Merges the given `enum` into `dict`. If one of the `enum` entries
-  already exists in `dict`, the given function is invoked to resolve
-  the conflict.
-
-  The `enum` must yield tuples with two elements on enumeration,
-  where the first element represents the key and the second the value.
-
-  ## Examples
-
-      iex> d1 = dict_impl.new([a: 1, b: 2])
-      ...> d2 = dict_impl.new([a: 3, d: 4])
-      ...> d = Dict.merge(d1, d2, fn(_k, v1, v2) ->
-      ...>   v1 + v2
-      ...> end)
-      ...> [a: Dict.get(d, :a), b: Dict.get(d, :b), d: Dict.get(d, :d)]
-      [a: 4, b: 2, d: 4]
-
-  """
-  @spec merge(t, t, (key, value, value -> value)) :: t
-  def merge(dict, enum, fun) do
-    target(dict).merge(dict, enum, fun)
-  end
-
-  @doc """
-  Returns the value associated with `key` in `dict` as
-  well as the `dict` without `key`.
-
-  ## Examples
-
-      iex> dict = dict_impl.new [a: 1]
-      ...> {v, d} = Dict.pop dict, :a
-      ...> {v, Enum.sort(d)}
-      {1,[]}
-
-      iex> dict = dict_impl.new [a: 1]
-      ...> {v, d} = Dict.pop dict, :b
-      ...> {v, Enum.sort(d)}
-      {nil,[a: 1]}
-
-      iex> dict = dict_impl.new [a: 1]
-      ...> {v, d} = Dict.pop dict, :b, 3
-      ...> {v, Enum.sort(d)}
-      {3,[a: 1]}
-
-  """
-  @spec pop(t, key, value) :: {value, t}
-  def pop(dict, key, default // nil) do
-    target(dict).pop(dict, key, default)
-  end
-
-  @doc """
-  Update a value in `dict` by calling `fun` on the value to get a new
-  value. An exception is generated if `key` is not present in the dict.
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1, b: 2])
-      ...> d = Dict.update!(d, :a, fn(val) -> -val end)
-      ...> Dict.get(d, :a)
-      -1
-
-  """
-  @spec update!(t, key, (value -> value)) :: t
-  def update!(dict, key, fun) do
-    target(dict).update!(dict, key, fun)
-  end
-
-  @doc false
-  def update(dict, key, fun) do
-    target(dict).update(dict, key, fun)
-  end
-
-  @doc """
-  Update a value in `dict` by calling `fun` on the value to get a new value. If
-  `key` is not present in `dict` then `initial` will be stored as the first
-  value.
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1, b: 2])
-      ...> d = Dict.update(d, :c, 3, fn(val) -> -val end)
-      ...> Dict.get(d, :c)
-      3
-
-  """
-  @spec update(t, key, value, (value -> value)) :: t
-  def update(dict, key, initial, fun) do
-    target(dict).update(dict, key, initial, fun)
-  end
-
-  @doc """
-  Returns a tuple of two dicts, where the first dict contains only
-  entries from `dict` with keys in `keys`, and the second dict
-  contains only entries from `dict` with keys not in `keys`
-
-  Any non-member keys are ignored.
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1, b: 2, c: 3, d: 4])
-      ...> { d1, d2 } = Dict.split(d, [:a, :c, :e])
-      ...> { Dict.to_list(d1), Dict.to_list(d2) }
-      { [a: 1, c: 3], [b: 2, d: 4] }
-
-      iex> d = dict_impl.new([])
-      ...> { d1, d2 } = Dict.split(d, [:a, :c])
-      ...> { Dict.to_list(d1), Dict.to_list(d2) }
-      { [], [] }
-
-      iex> d = dict_impl.new([a: 1, b: 2])
-      ...> { d1, d2 } = Dict.split(d, [:a, :b, :c])
-      ...> { Dict.to_list(d1), Dict.to_list(d2) }
-      { [a: 1, b: 2], [] }
-
-  """
-  @spec split(t, keys) :: {t, t}
-  def split(dict, keys) do
-    target(dict).split(dict, keys)
-  end
-
-  @doc """
-  Returns a new dict where the given `keys` are removed from `dict`.
-  Any non-member keys are ignored.
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1, b: 2])
-      ...> d = Dict.drop(d, [:a, :c, :d])
-      ...> Dict.to_list(d)
-      [b: 2]
-
-      iex> d = dict_impl.new([a: 1, b: 2])
-      ...> d = Dict.drop(d, [:c, :d])
-      ...> Dict.to_list(d)
-      [a: 1, b: 2]
-
-  """
-  @spec drop(t, keys) :: t
-  def drop(dict, keys) do
-    target(dict).drop(dict, keys)
-  end
-
-  @doc """
-  Returns a new dict where only the keys in `keys` from `dict` are
-  included.
-  Any non-member keys are ignored.
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1, b: 2])
-      ...>
-      ...> d = Dict.take(d, [:a, :c, :d])
-      ...> Dict.to_list(d)
-      [a: 1]
-      ...>
-      ...> d = Dict.take(d, [:c, :d])
-      ...> Dict.to_list(d)
-      []
-
-  """
-  @spec take(t, keys) :: t
-  def take(dict, keys) do
-    target(dict).take(dict, keys)
-  end
-
-  @doc """
-  Returns an empty dict of the same type as `dict`.
-
-  ## Examples
-
-      iex> d = dict_impl.new([a: 1, b: 2])
-      ...> e = Dict.empty(d)
-      ...> Dict.to_list(e)
-      []
-
-  """
-  @spec empty(t) :: t
-  def empty(dict) do
-    target(dict).empty(dict)
-  end
-
-  @doc """
-  Check if two dicts are equal using `===`. If the dicts are
-  of different types, they are first converted to lists.
-
-  ## Examples
-
-      iex> a = dict_impl.new(a: 2, b: 3, f: 5, c: 123)
-      ...> b = ListDict.new(a: 2, b: 3, f: 5, c: 123)
-      ...> Dict.equal?(a, b)
-      true
-
-      iex> a = dict_impl.new(a: 2, b: 3, f: 5, c: 123)
-      ...> b = []
-      ...> Dict.equal?(a, b)
-      false
-
-  """
-  @spec equal?(t, t) :: boolean
-  def equal?(a, b) do
-    a_target = target(a)
-    b_target = target(b)
-
-    cond do
-      a_target == b_target ->
-        a_target.equal?(a, b)
-
-      a_target.size(a) == b_target.size(b) ->
-        ListDict.equal?(a_target.to_list(a), b_target.to_list(b))
-
-      true ->
-        false
-    end
-  end
-
-  @doc """
-  Returns a list of key-value pairs stored in `dict`.
-  No particular order is enforced.
-  """
-  @spec to_list(t) :: list
-  def to_list(dict) do
-    target(dict).to_list(dict)
-  end
-
-  defp unsupported_dict(dict) do
-    raise ArgumentError, message: "unsupported dict: #{inspect dict}"
-  end
-end
@@ -1,125 +0,0 @@
-defmodule Dict.Behaviour do
-  @moduledoc """
-  This module makes it easier to create your own `Dict` compliant
-  module, by providing default implementations for some required functions.
-  
-  Usage:
-
-      defmodule MyDict do
-        use Dict.Behaviour
-
-        # implement required functions (see below)
-
-        # override default implementations if needed
-      end
-
-  The client module must contain following functions: `size/1`, `fetch/2`, 
-  `put/3`, `update/4`, `delete/2` and `reduce/3`. All of them are part of
-  the Dict behaviour, so no extra functions are actually required.
-
-  Based on these functions, `Dict.Behaviour` generates default implementations 
-  for other functions such as `drop`, `take`, etc. All of the functions are
-  defined as overridable, so you can provide your own implementation if
-  needed.
-
-  If you implement `new/0` and `new/1` functions, you can also test your custom 
-  module via `Dict` doctests:
-
-      defmodule MyDict do
-        def new(keywords // []) do
-          ...
-        end
-      end
-
-      defmodule MyTests do
-        use ExUnit.Case
-        doctest Dict
-        defp dict_impl, do: MyDict
-      end
-
-  """
-
-  defmacro __using__(_) do
-    quote do
-      @behaviour Dict
-      
-      def get(dict, key, default // nil) do
-        case fetch(dict, key) do
-          { :ok, value } -> value
-          :error -> default
-        end
-      end
-      defoverridable get: 2, get: 3
-
-      def fetch!(dict, key) do
-        case fetch(dict, key) do
-          { :ok, value } -> value
-          :error -> raise(KeyError, key: key)
-        end
-      end
-      defoverridable fetch!: 2
- 
-      def has_key?(dict, key) do
-        match? { :ok, _ }, fetch(dict, key)
-      end
-      defoverridable has_key?: 2
- 
-      def put_new(dict, key, value) do
-        update(dict, key, value, fn(v) -> v end)
-      end
-      defoverridable put_new: 3
- 
-      def drop(dict, []), do: dict
-
-      def drop(dict, [key|keys]) do
-        drop(delete(dict, key), keys)
-      end
-      defoverridable drop: 2
- 
-      def take(dict, keys) do
-        take(dict, keys, new)
-      end
-      defoverridable take: 2
- 
-      defp take(_dict, [], acc), do: acc
-      defp take(dict, [key|keys], acc) do
-        case fetch(dict, key) do
-          { :ok, value } -> take(dict, keys, put(acc, key, value))
-          :error -> take(dict, keys, acc)
-        end
-      end
- 
-      def to_list(dict), do: reduce(dict, [], &[&1|&2]) |> Enum.reverse
-      defoverridable to_list: 1
- 
-      def keys(dict), do: reduce(dict, [], fn({k, _}, acc) -> [k | acc] end) |> Enum.reverse
-      defoverridable keys: 1
- 
-      def values(dict), do: reduce(dict, [], fn({_, v}, acc) -> [v | acc] end) |> Enum.reverse
-      defoverridable values: 1
- 
-      def equal?(dict1, dict2) do
-        case Dict.size(dict1) == Dict.size(dict2) do
-          false -> false
-          true -> 
-            try do
-              reduce(dict1, nil, fn({ k, v }, _acc) ->
-                unless fetch(dict2, k) == { :ok, v }, do: throw(:error)
-              end)
-              true
-            catch
-              :error -> false
-            end
-        end
-      end
-      defoverridable equal?: 2
- 
-      def merge(dict, enumerable, callback // fn(_k, _v1, v2) -> v2 end) do
-        Enum.reduce(enumerable, dict, fn({key, value}, acc) ->
-          update(acc, key, value, fn(v1) -> callback.(key, v1, value) end)
-        end)
-      end
-      defoverridable merge: 2, merge: 3
-    end
-  end
-end
@@ -1,349 +0,0 @@
-defexception RuntimeError,      message: "runtime error"
-defexception ArgumentError,     message: "argument error"
-defexception ArithmeticError,   message: "bad argument in arithmetic expression"
-defexception SystemLimitError,  message: "a system limit has been reached"
-
-defexception SyntaxError, [file: nil, line: nil, description: "syntax error"] do
-  def message(exception) do
-    "#{Exception.format_file_line(exception.file, exception.line)}#{exception.description}"
-  end
-end
-
-defexception TokenMissingError, [file: nil, line: nil, description: "expression is incomplete"] do
-  def message(exception) do
-    "#{Exception.format_file_line(exception.file, exception.line)}#{exception.description}"
-  end
-end
-
-defexception CompileError, [file: nil, line: nil, description: "compile error"] do
-  def message(exception) do
-    "#{Exception.format_file_line(exception.file, exception.line)}#{exception.description}"
-  end
-end
-
-defexception BadFunctionError, [actual: nil] do
-  def message(exception) do
-    "expected a function, got: #{inspect(exception.actual)}"
-  end
-end
-
-defexception MatchError, [actual: nil] do
-  def message(exception) do
-    "no match of right hand side value: #{inspect(exception.actual)}"
-  end
-end
-
-defexception CaseClauseError, [actual: nil] do
-  def message(exception) do
-    "no case clause matching: #{inspect(exception.actual)}"
-  end
-end
-
-defexception TryClauseError, [actual: nil] do
-  def message(exception) do
-    "no try clause matching: #{inspect(exception.actual)}"
-  end
-end
-
-defexception BadArityError, [function: nil, args: nil] do
-  def message(exception) do
-    "bad arity error: #{inspect(exception.function)} called with #{inspect(exception.args)}"
-  end
-end
-
-defexception UndefinedFunctionError, [module: nil, function: nil, arity: nil] do
-  def message(exception) do
-    if exception.function do
-      formatted = Exception.format_mfa exception.module, exception.function, exception.arity
-      "undefined function: #{formatted}"
-    else
-      "undefined function"
-    end
-  end
-end
-
-defexception FunctionClauseError, [module: nil, function: nil, arity: nil] do
-  def message(exception) do
-    if exception.function do
-      formatted = Exception.format_mfa exception.module, exception.function, exception.arity
-      "no function clause matching in #{formatted}"
-    else
-      "no function clause matches"
-    end
-  end
-end
-
-defexception Protocol.UndefinedError, [protocol: nil, value: nil, description: nil] do
-  def message(exception) do
-    msg = "protocol #{inspect exception.protocol} not implemented for #{inspect exception.value}"
-    if exception.description do
-      msg <> ", " <> exception.description
-    else
-      msg
-    end
-  end
-end
-
-defexception ErlangError, [original: nil] do
-  def message(exception) do
-    "erlang error: #{inspect(exception.original)}"
-  end
-end
-
-defexception KeyError, key: nil do
-  def message(exception) do
-    "key not found: #{inspect exception.key}"
-  end
-end
-
-defexception Enum.OutOfBoundsError, message: "out of bounds error"
-
-defexception Enum.EmptyError, message: "empty error"
-
-defmodule Exception do
-  @moduledoc """
-  Several convenience functions to work with and pretty print
-  exceptions and stacktraces.
-  """
-
-  @doc """
-  Normalizes an exception, converting Erlang exceptions
-  to Elixir exceptions. It takes the kind spilled by
-  `catch` as an argument as a convenience for converting only
-  `:errors`, ignorning the others.
-  """
-  def normalize(:error, exception), do: normalize(exception)
-  def normalize(_kind, other), do: other
-
-  @doc """
-  Normalizes an exception, converting Erlang exceptions
-  to Elixir exceptions. Useful when interfacing Erlang
-  code with Elixir code.
-  """
-  def normalize(exception) when is_exception(exception) do
-    exception
-  end
-
-  def normalize(:badarg) do
-    ArgumentError[]
-  end
-
-  def normalize(:badarith) do
-    ArithmeticError[]
-  end
-
-  def normalize(:system_limit) do
-    SystemLimitError[]
-  end
-
-  def normalize({ :badarity, { fun, args } }) do
-    BadArityError[function: fun, args: args]
-  end
-
-  def normalize({ :badfun, actual }) do
-    BadFunctionError[actual: actual]
-  end
-
-  def normalize({ :badmatch, actual }) do
-    MatchError[actual: actual]
-  end
-
-  def normalize({ :case_clause, actual }) do
-    CaseClauseError[actual: actual]
-  end
-
-  def normalize({ :try_clause, actual }) do
-    TryClauseError[actual: actual]
-  end
-
-  def normalize(:undef) do
-    { mod, fun, arity } = from_stacktrace(:erlang.get_stacktrace)
-    UndefinedFunctionError[module: mod, function: fun, arity: arity]
-  end
-
-  def normalize(:function_clause) do
-    { mod, fun, arity } = from_stacktrace(:erlang.get_stacktrace)
-    FunctionClauseError[module: mod, function: fun, arity: arity]
-  end
-
-  def normalize({ :badarg, payload }) do
-    ArgumentError[message: "argument error: #{inspect(payload)}"]
-  end
-
-  def normalize(other) do
-    ErlangError[original: other]
-  end
-
-  @doc """
-  Receives a tuple representing a stacktrace entry and formats it.
-
-  The current working directory may be given as an argument,
-  otherwise one is automatically retrieved.
-  """
-  def format_stacktrace_entry(entry, cwd // nil)
-
-  # From Macro.Env.stacktrace
-  def format_stacktrace_entry({ module, :__MODULE__, 0, file_line }, cwd) do
-    "#{format_location(file_line, cwd)}#{inspect module} (module)"
-  end
-
-  # From :elixir_compiler_*
-  def format_stacktrace_entry({ _module, :__MODULE__, 2, file_line }, cwd) do
-    "#{format_location(file_line, cwd)}(module)"
-  end
-
-  # From :elixir_compiler_*
-  def format_stacktrace_entry({ _module, :__FILE__, 2, file_line }, cwd) do
-    "#{format_location(file_line, cwd)}(file)"
-  end
-
-  def format_stacktrace_entry({module, fun, arity, file_line}, cwd) do
-    "#{format_location(file_line, cwd)}#{format_mfa(module, fun, arity)}"
-  end
-
-  def format_stacktrace_entry({fun, arity, file_line}, cwd) do
-    "#{format_location(file_line, cwd)}#{format_fa(fun, arity)}"
-  end
-
-  @doc """
-  Formats the stacktrace.
-
-  A stacktrace must be given as an argument. If not, this function
-  calculates the current stacktrace and formats it. As a consequence,
-  the value of `System.stacktrace` is changed.
-  """
-  def format_stacktrace(trace // nil) do
-    trace = trace || try do
-      throw(:stacktrace)
-    catch
-      :stacktrace -> Enum.drop(:erlang.get_stacktrace, 1)
-    end
-
-    cwd = System.cwd
-
-    case trace do
-      [] -> "\n"
-      s  -> "    " <> Enum.map_join(s, "\n    ", &format_stacktrace_entry(&1, cwd)) <> "\n"
-    end
-  end
-
-  @doc """
-  Formats the caller, i.e. the first entry in the stacktrace.
-  Notice that due to tail call optimization, the stacktrace
-  may not report the direct caller of the function.
-  """
-  def format_caller(trace // nil) do
-    trace = trace || try do
-      throw(:stacktrace)
-    catch
-      :stacktrace -> Enum.drop(:erlang.get_stacktrace, 1)
-    end
-
-    if entry = Enum.at(trace, 1) do
-      format_stacktrace_entry(entry)
-    else
-      "nofile:0: "
-    end
-  end
-
-  @doc """
-  Receives an anonymous function and arity and formats it as
-  shown in stacktraces. The arity may also be a list of arguments.
-
-  ## Examples
-
-      Exception.format_fa(fn -> end, 1)
-      #=> "#Function<...>/1"
-
-  """
-  def format_fa(fun, arity) do
-    if is_list(arity) do
-      inspected = lc x inlist arity, do: inspect(x)
-      "#{inspect fun}(#{Enum.join(inspected, ", ")})"
-    else
-      "#{inspect fun}/#{arity}"
-    end
-  end
-
-  @doc """
-  Receives a module, fun and arity and formats it
-  as shown in stacktraces. The arity may also be a list
-  of arguments.
-
-  ## Examples
-
-      iex> Exception.format_mfa Foo, :bar, 1
-      "Foo.bar/1"
-      iex> Exception.format_mfa Foo, :bar, []
-      "Foo.bar()"
-      iex> Exception.format_mfa nil, :bar, []
-      "nil.bar()"
-
-  """
-  def format_mfa(module, fun, arity) do
-    fun =
-      case inspect(fun) do
-        << ?:, erl :: binary >> -> erl
-        elixir -> elixir
-      end
-
-    if is_list(arity) do
-      inspected = lc x inlist arity, do: inspect(x)
-      "#{format_module module}#{fun}(#{Enum.join(inspected, ", ")})"
-    else
-      "#{format_module module}#{fun}/#{arity}"
-    end
-  end
-
-  defp format_module(mod), do: "#{inspect mod}."
-
-  @doc """
-  Formats the given file and line as shown in stacktraces.
-  If any of the values are nil, they are omitted.
-
-  The current working directory may be given as an argument,
-  otherwise one is automatically retrieved.
-
-  ## Examples
-
-      iex> Exception.format_file_line("foo", 1)
-      "foo:1: "
-
-      iex> Exception.format_file_line("foo", nil)
-      "foo: "
-
-      iex> Exception.format_file_line(nil, nil)
-      ""
-
-  """
-  def format_file_line(file, line, cwd // nil) do
-    if file do
-      file = to_string(file)
-      file = if cwd, do: Path.relative_to(file, cwd), else: Path.relative_to_cwd(file)
-
-      if line && line != 0 do
-        "#{file}:#{line}: "
-      else
-        "#{file}: "
-      end
-    else
-      ""
-    end
-  end
-
-  defp format_location(opts, cwd) do
-    format_file_line Keyword.get(opts, :file), Keyword.get(opts, :line), cwd
-  end
-
-  defp from_stacktrace([{ module, function, args, _ }|_]) when is_list(args) do
-    { module, function, length(args) }
-  end
-
-  defp from_stacktrace([{ module, function, arity, _ }|_]) do
-    { module, function, arity }
-  end
-
-  defp from_stacktrace(_) do
-    { nil, nil, nil }
-  end
-end
@@ -1,83 +0,0 @@
-defmodule Float do
-  @moduledoc """
-  Functions for working with floating point numbers.
-  """
-
-  @doc """
-  Parses a binary into a float. 
-  
-  If successful, returns a tuple of the form `{ float, remainder_of_binary }`.
-  Otherwise `:error`.
-
-  ## Examples
-
-      iex> Float.parse("34")
-      {34.0,""}
-      iex> Float.parse("34.25")
-      {34.25,""}
-      iex> Float.parse("56.5xyz")
-      {56.5,"xyz"}
-      iex> Float.parse("pi")
-      :error
-
-  """
-  @spec parse(binary) :: { float, binary } | :error
-  def parse(binary) when is_binary(binary) do
-    case Integer.parse binary do
-      :error -> :error
-      { integer_part, after_integer } -> parse after_integer, integer_part
-    end
-  end
-
-  # Dot followed by digit is required afterwards or we are done
-  defp parse(<< ?., char, rest :: binary >>, int) when char in ?0..?9 do
-    parse(rest, char - ?0, 1, int)
-  end
-
-  defp parse(rest, int) do
-    { :erlang.float(int), rest }
-  end
-
-  # Handle decimal points
-  defp parse(<< char, rest :: binary >>, float, decimal, int) when char in ?0..?9 do
-    parse rest, 10 * float + (char - ?0), decimal + 1, int
-  end
-
-  defp parse(<< ?e, after_e :: binary >>, float, decimal, int) do
-    case Integer.parse after_e do
-      :error ->
-        # Note we rebuild the binary here instead of breaking it apart at
-        # the function clause because the current approach copies a binary
-        # just on this branch. If we broke it apart in the function clause,
-        # the copy would happen when calling Integer.parse/1.
-        { floatify(int, float, decimal), << ?e, after_e :: binary >> }
-      { exponential, after_exponential } ->
-        { floatify(int, float, decimal, exponential), after_exponential }
-    end
-  end
-
-  defp parse(bitstring, float, decimal, int) do
-    { floatify(int, float, decimal), bitstring }
-  end
-
-  defp floatify(int, float, decimal, exponential // 0) do
-    multiplier = if int < 0, do: -1.0, else: 1.0
-
-    # Try to ensure the minimum amount of rounding errors
-    result = multiplier * (abs(int) * :math.pow(10, decimal) + float) * :math.pow(10, exponential - decimal)
-
-    # Try avoiding stuff like this:
-    # iex(1)> 0.0001 * 75
-    # 0.007500000000000001
-    # Due to IEEE 754 floating point standard
-    # http://docs.oracle.com/cd/E19957-01/806-3568/ncg_goldberg.html
-
-    final_decimal_places = decimal - exponential
-    if final_decimal_places > 0 do
-      decimal_power_round = :math.pow(10,  final_decimal_places)
-      trunc(result * decimal_power_round) / decimal_power_round
-    else
-      result
-    end
-  end
-end
@@ -1,101 +0,0 @@
-defmodule GenEvent.Behaviour do
-  @moduledoc """
-  This module is a convenience for defining GenEvent callbacks in Elixir.
-
-  GenEvent is an OTP behaviour that encapsulates event handling functionality.
-
-  ## Example
-
-  Below is an example of a GenEvent that stores notifications
-  until they are fetched:
-
-      defmodule MyEventHandler do
-        use GenEvent.Behaviour
-
-        # Callbacks
-
-        def init(_) do
-          { :ok, [] }
-        end
-
-        def handle_event({:notification, x}, notifications) do
-          { :ok, [x|notifications] }
-        end
-
-        def handle_call(:notifications, notifications) do
-          {:ok, Enum.reverse(notifications), []}
-        end
-
-      end
-
-      { :ok, pid } = :gen_event.start_link
-      #=> {:ok,#PID<0.42.0>}
-
-      :gen_event.add_handler(pid, MyEventHandler, [])
-      #=> :ok
-
-      :gen_event.notify(pid, {:notification, 1})
-      #=> :ok
-
-      :gen_event.notify(pid, {:notification, 2})
-      #=> :ok
-      
-      :gen_event.call(pid, MyEventHandler, :notifications)
-      #=> [1, 2]
-
-      :gen_event.call(pid, MyEventHandler, :notifications)
-      #=> []
-
-  Notice we never call the server callbacks directly, they are called
-  by OTP whenever we interact with the server. 
-
-  Starting and sending messages to the GenEvent is done
-  via Erlang's `:gen_event` module. For more information,
-  please refer to the following:
-
-  * http://www.erlang.org/doc/man/gen_event.html
-  * http://learnyousomeerlang.com/event-handlers
-  """
-
-  @doc false
-  defmacro __using__(_) do
-    quote location: :keep do
-      @behaviour :gen_event
-
-      @doc false
-      def init(args) do
-        { :ok, args }
-      end
-
-      @doc false
-      def handle_event(_event, state) do
-        { :ok, state }
-      end
-
-      @doc false
-      def handle_call(_request, state) do
-        { :ok, :ok, state }
-      end
-
-      @doc false
-      def handle_info(_msg, state) do
-        { :ok, state }
-      end
-
-      @doc false
-      def terminate(reason, state) do
-        :ok
-      end
-
-      @doc false
-      def code_change(_old, state, _extra) do
-        { :ok, state }
-      end
-
-      defoverridable [init: 1, 
-                      handle_event: 2,
-                      handle_call: 2, handle_info: 2,
-                      terminate: 2, code_change: 3]
-    end
-  end
-end
@@ -1,211 +0,0 @@
-defmodule GenFSM.Behaviour do
-  @moduledoc """
-  This module is a convenience for defining GenFSM callbacks in Elixir.
-
-  A finite state machine (FSM) is responsible for reacting to events received;
-  GenFSM is an OTP behaviour that encapsulates common FSM
-  functionalities.
-
-  ## Example
-
-  Below is an example of a GenFSM that runs a very simple minded
-  coffee vending machine (CVM). The CVM treats all coins the same. If
-  you press the request button then the CVM will brew coffee, if you
-  have paid enough coins; if not, it will wait until you have inserted
-  enough coins and then it will instantly brew the coffee, since you
-  had already pressed the request button. As we told you - a very
-  simple minded CVM! And greedy too. If you insert more coins than you
-  need it will gladly eat them until you press the request button.
-
-  We will leave it to the service-minded reader to improve the way the CVM
-  works - we hereby declare a full disclaimer for any lawsuits that the
-  behaviour of the CVM in its original state might encur.
-
-
-      defmodule MyFsm do
-        use GenFSM.Behaviour
-
-          # keeping track of what is going on inside the CVM.
-          # 3 is the target price for a cup of coffee
-          defrecord StateData, coins: 0, price: 3
-
-          #
-          # API functions
-          #
-
-          def start_link() do
-            :gen_fsm.start_link({:local, :cvm}, __MODULE__, [], [])
-          end
-
-          def insert_coin() do
-            :gen_fsm.send_event(:cvm, :coin)
-          end
-
-          def request_coffee() do
-            :gen_fsm.send_event(:cvm, :request_coffee)
-          end
-
-          #
-          # Callbacks
-          #
-
-          def init(_args) do
-            { :ok, :short_paid, StateData.new }
-          end
-
-
-          def short_paid(:coin, state_data = StateData[coins: c, price: p]) 
-           when c + 1 < p do
-            { :next_state, :short_paid, state_data.coins(c + 1) }
-          end
-
-          def short_paid(:coin, state_data) do
-            { :next_state, :paid_in_full, &state_data.update_coins(&1 + 1) }
-          end
-
-          def short_paid(:request_coffee, state_data) do
-            { :next_state, :requested_short_paid, state_data }
-          end
-
-
-          def requested_short_paid(:request_coffee, state_data) do
-            {:next_state, :requested_short_paid, state_data }
-          end
-
-          def requested_short_paid(:coin, state_data=StateData[coins: c, price: p]) 
-           when c+1 < p do
-            { :next_state, :requested_short_paid, state_data.coins(c + 1) }
-          end
-
-          def requested_short_paid(:coin, _state_data) do
-            IO.puts "Here's your coffee!"
-            { :next_state, :short_paid, StateData.new }
-          end
-
-
-          def paid_in_full(:coin, state_data) do
-            { :next_state, :paid_in_full, &state_data.update_coins(&1 + 1) }
-          end
-
-          def paid_in_full(:request_coffee, _state_data) do
-            IO.puts "Here's your coffee!"
-            { :next_state, :short_paid, StateData.new }
-          end
-        end
-
-
-      { :ok, _pid } = MyFsm.start_link()
-
-      MyFsm.insert_coin
-      #=> :ok
-      MyFsm.insert_coin
-      #=> :ok
-
-      MyFsm.request_coffee
-      #=> :ok
-
-      MyFsm.insert_coin
-      #=> :ok
-      #=> Here's your coffee!
-
-  Notice we never call the GenFSM callbacks directly; they are called by
-  OTP whenever we interact with the server throught the API. `send_event` is
-  asynchronous, whereas `sync_send_event` is synchronous. For
-  a GenFSM, the different values a callback can return depend
-  on the type of callback.
-
-  State handling returns for `send_event` callbacks:
-
-      { :next_state, next_state_name, new_state_data }
-      { :next_state, next_state_name, new_state_data, timeout }
-      { :next_state, next_state_name, new_state_data, :hibernate }
-      { :stop, reason, new_state_data }
-
-  State handling returns for `sync_send_event` callbacks:
-
-      { :reply, reply, next_state_name, new_state_data }
-      { :reply, reply, next_state_name, new_state_data, timeout }
-      { :reply, reply, next_state_name, new_state_data, :hibernate }
-      { :next_state, next_state_name, new_state_data }
-      { :next_state, next_state_name, new_state_data, timeout }
-      { :next_state, next_state_name, new_state_data, :hibernate }
-      { :stop, reason, reply, new_state_data }
-      { :stop, reason, new_state_date }
-
-  There are 6 callbacks required to be implemented in a GenFsm plus 1
-  or 2 for each state. The `GenFSM.Behaviour` module defines
-  `handle_sync_event`, `handle_info`, `terminate` and `code_change`
-  for you. The list of callbacks are:
-
-  * `init(args)` - invoked when the FSM is started;
-  * `handle_sync_event(event, from, state_name, state_data)` - invoked to 
-  handle `sync_send_all_state_event` messages;
-  * `handle_event(event, state_name, state_data)` - invoked to handle 
-  `send_all_state_event` messages;
-  * `handle_info(msg, state_name, state_data)` - handle all other 
-  messages which are normally received by processes;
-  * `terminate(reason, state_name, state_data)` - called when the FSM 
-  is about to terminate, useful for cleaning up;
-  * `code_change(old_vsn, state, extra)` - called when the application 
-  code is being upgraded live (hot code swap);
-
-  Unlike `GenServer` and `GenEvent`, the callback `init/1` is not
-  implemented by default, as it requires the next state to be returned.
-
-  For each state you need to define either or both of these:
-
-  * `state_name(event, state_data)` - invoked to handle 
-  `send_event` messages;
-  * `state_name(event, from, state_data)`- invoked to handle 
-  `sync_send_event` messages;
-
-  If you send asynchronous events you only need to implement the
-  `state_name/2` variant and vice-versa for synchronous events and
-  `state_name/3`. Keep in mind that if you mix `send_event` and
-  `sync_send_event` the best thing to do is to implement both
-  callbacks for all states.
-
-  Starting and sending messages to the GenFSM is done via Erlang's
-  `:gen_fsm` module. For more information, please refer to the
-  following:
-
-  * http://www.erlang.org/doc/man/gen_fsm.html
-  * http://www.erlang.org/doc/design_principles/fsm.html
-  * http://learnyousomeerlang.com/finite-state-machines
-  """
-
-  @doc false
-  defmacro __using__(_) do
-    quote location: :keep do
-      @behavior :gen_fsm
-
-      @doc false
-      def handle_event(_event, state_name, state_data) do
-        { :next_state, state_name, state_data }
-      end
-
-      @doc false
-      def handle_sync_event(_event, from, state_name, state_data) do
-        { :reply, :ok, state_name, state_data }
-      end
-
-      @doc false
-      def handle_info(_msg, state_name, state_data) do
-        { :next_state, state_name, state_data }
-      end
-
-      @doc false
-      def terminate(_reason, _state_name, _state_data) do
-        :ok
-      end
-
-      @doc false
-      def code_change(_old, state_name, state_data, _extra) do
-        { :ok, state_name, state_data }
-      end
-
-      defoverridable [handle_event: 3, handle_sync_event: 4,
-                      handle_info: 3, terminate: 3, code_change: 4]
-    end
-  end
-end
@@ -1,125 +0,0 @@
-defmodule GenServer.Behaviour do
-  @moduledoc """
-  This module is a convenience for defining GenServer callbacks in Elixir.
-
-  A server is responsible for reacting to messages received from a client.
-  A GenServer is an OTP behaviour that encapsulates common server
-  functionalities.
-
-  ## Example
-
-  Below is an example of a GenServer that pushes and pops items
-  onto a stack:
-
-      defmodule MyServer do
-        use GenServer.Behaviour
-
-        # Callbacks
-
-        def handle_call(:pop, _from, [h|t]) do
-          { :reply, h, t }
-        end
-
-        def handle_call(request, from, config) do
-          # Call the default implementation from GenServer.Behaviour
-          super(request, from, config)
-        end
-
-        def handle_cast({ :push, item }, config) do
-          { :noreply, [item|config] }
-        end
-
-        def handle_cast(request, config) do
-          super(request, config)
-        end
-      end
-
-      { :ok, pid } = :gen_server.start_link(MyServer, [:hello], [])
-
-      :gen_server.call(pid, :pop)
-      #=> :hello
-
-      :gen_server.cast(pid, { :push, :world })
-      #=> :ok
-
-      :gen_server.call(pid, :pop)
-      #=> :world
-
-  Notice we never call the server callbacks directly, they are called
-  by OTP whenever we interact with the server. **cast** messages are
-  asynchronous while **call** ones are synchronous. For a
-  GenServer, there are 8 different values a callback such as
-  `handle_call` or `handle_cast` can return:
-
-      { :reply, reply, new_state }
-      { :reply, reply, new_state, timeout }
-      { :reply, reply, new_state, :hibernate }
-      { :noreply, new_state }
-      { :noreply, new_state, timeout }
-      { :noreply, new_state, :hibernate }
-      { :stop, reason, new_state }
-      { :stop, reason, reply, new_state }
-
-  There are 6 callbacks required to be implemented in a GenServer. The
-  `GenServer.Behaviour` module defines all of them automatically, but
-  allows us to customize the ones we need. The required callbacks are:
-
-  * `init(args)` - invoked when the server is started;
-  * `handle_call(msg, from, state)` - invoked to handle call messages;
-  * `handle_cast(msg, state)` - invoked to handle cast messages;
-  * `handle_info(msg, state)` - handle all other messages which are
-     normally received by processes;
-  * `terminate(reason, state)` - called when the server is about to
-     terminate, useful for cleaning up;
-  * `code_change(old_vsn, state, extra)` - called when the application
-    code is being upgraded live (hot code swap);
-
-  Starting and sending messages to the GenServer is done
-  via Erlang's `:gen_server` module. For more information,
-  please refer to the following:
-
-  * http://www.erlang.org/doc/man/gen_server.html
-  * http://www.erlang.org/doc/design_principles/gen_server_concepts.html
-  * http://learnyousomeerlang.com/clients-and-servers
-  """
-
-  @doc false
-  defmacro __using__(_) do
-    quote location: :keep do
-      @behavior :gen_server
-
-      @doc false
-      def init(args) do
-        { :ok, args }
-      end
-
-      @doc false
-      def handle_call(_request, _from, state) do
-        { :noreply, state }
-      end
-
-      @doc false
-      def handle_info(_msg, state) do
-        { :noreply, state }
-      end
-
-      @doc false
-      def handle_cast(_msg, state) do
-        { :noreply, state }
-      end
-
-      @doc false
-      def terminate(_reason, _state) do
-        :ok
-      end
-
-      @doc false
-      def code_change(_old, state, _extra) do
-        { :ok, state }
-      end
-
-      defoverridable [init: 1, handle_call: 3, handle_info: 2,
-        handle_cast: 2, terminate: 2, code_change: 3]
-    end
-  end
-end
@@ -1,478 +0,0 @@
-defmodule HashDict do
-  @moduledoc """
-  A key-value store.
-
-  The `HashDict` is meant to work well with both small and
-  large set of keys and it is an implementation of the `Dict`
-  behaviour. For more information about the functions and
-  their APIs, please consult the `Dict` module.
-  """
-
-  use Dict.Behaviour
-
-  # A dictionary (key-value) implementation based on dynamic hashing.
-  #
-  # This implementation is based on hash tries. We first start with
-  # a set of 8 buckets and expand when the density is about 5 entries
-  # per bucket. We use bit shifting to make rehashing faster on
-  # expansion.
-  #
-  # Compared to dict, it provides many enhancements:
-  #
-  # 1. HashDict buckets are ordered sets, this gives us faster access
-  #    and modification times
-  #
-  # 2. It uses phash2 to calculate the hash (instead of phash)
-  #
-  # 3. The dictionary first starts with a single bucket, instead of
-  #    a set of 8 buckets. This allow us to skip hashing altogher
-  #    for small dictionaries, providing faster operations and
-  #    reducing memory consumption
-  #
-  # 4. Once we reach 8 elements, the dictionary is promoted to a
-  #    set of buckets
-
-  # The ordered record contains a single bucket
-  @ordered_threshold 8
-
-  defrecordp :ordered, HashDict,
-    size: 0,
-    bucket: []
-
-  # The bucketed record contains a series of buckets.
-  @expand_load 5
-  @contract_load 2
-  @node_bitmap 0b111
-  @node_shift 3
-  @node_size 8
-  @node_template :erlang.make_tuple(@node_size, [])
-
-  defrecordp :trie, HashDict,
-    size: 0,
-    depth: 0,
-    expand_on: @node_size * @expand_load,
-    contract_on: @contract_load,
-    root: @node_template
-
-  import Bitwise
-
-  # Let's inline common instructions
-  @compile :inline_list_funcs
-  @compile { :inline, bucket_hash: 1, bucket_index: 1, bucket_nth_index: 2, bucket_next: 1 }
-
-  @doc """
-  Creates a new empty dict.
-  """
-  @spec new :: Dict.t
-  def new do
-    ordered()
-  end
-
-  @doc """
-  Creates a new dict from the given enumerable.
-
-  ## Examples
-
-      HashDict.new [{:b, 1}, {:a, 2}]
-      #=> #HashDict<[a: 2, b: 1]>
-
-  """
-  @spec new(list({key :: term, value :: term})) :: Dict.t
-  def new(pairs) do
-    Enum.reduce pairs, ordered(), fn { k, v }, dict ->
-      put(dict, k, v)
-    end
-  end
-
-  @doc """
-  Creates a new dict from the enumerable with the
-  help of the transformation function.
-
-  ## Examples
-
-      HashDict.new ["a", "b"], fn x -> {x, x} end
-      #=> #HashDict<[{"a","a"},{"b","b"}]>
-
-  """
-  @spec new(list, (term -> {key :: term, value ::term})) :: Dict.t
-  def new(list, transform) when is_function(transform) do
-    Enum.reduce list, new(), fn i, dict ->
-      { k, v } = transform.(i)
-      put(dict, k, v)
-    end
-  end
-
-  def put(dict, key, value) do
-    { dict, _ } = dict_put(dict, key, { :put, value })
-    dict
-  end
-
-  def update!(dict, key, fun) when is_function(fun, 1) do
-    case dict_put(dict, key, { :update, nil, fun }) do
-      { dict, 0 } ->
-        dict
-      { _dict, 1 } ->
-        raise KeyError, key: key
-    end
-  end
-
-  def update(dict, key, initial, fun) when is_function(fun, 1) do
-    { dict, _ } = dict_put(dict, key, { :update, initial, fun })
-    dict
-  end
-
-  def fetch(ordered(bucket: bucket), key) do
-    bucket_get(bucket, key)
-  end
-
-  def fetch(trie(root: root, depth: depth), key) do
-    bucket_get(node_bucket(root, depth, bucket_hash(key)), key)
-  end
-
-  def pop(dict, key, default // nil) do
-    case dict_delete(dict, key) do
-      { dict, _, 0 } -> { default, dict }
-      { dict, value, _ } -> { value, dict }
-    end
-  end
-
-  def delete(dict, key) do
-    { dict, _, _ } = dict_delete(dict, key)
-    dict
-  end
-
-  def size(dict) do
-    elem(dict, 1)
-  end
-
-  def empty(_) do
-    ordered()
-  end
-
-  def merge(dict, enum, callback // fn(_k, _v1, v2) -> v2 end)
-
-  def merge(dict1, dict2, callback) when is_record(dict1, HashDict) and is_record(dict2, HashDict) and elem(dict1, 1) < elem(dict2, 1) do
-    dict_fold dict1, dict2, fn [k|v1], acc ->
-      update(acc, k, v1, &callback.(k, v1, &1))
-    end
-  end
-
-  def merge(dict1, dict2, callback) when is_record(dict1, HashDict) and is_record(dict2, HashDict) do
-    dict_fold dict2, dict1, fn [k|v2], acc ->
-      update(acc, k, v2, &callback.(k, &1, v2))
-    end
-  end
-
-  def merge(dict, enumerable, callback) when is_record(dict, HashDict) do
-    super(dict, enumerable, callback)
-  end
-
-  def split(dict, keys) do
-    split(keys, new, dict)
-  end
-
-  defp split([], including, excluding) do
-    { including, excluding }
-  end
-
-  defp split([key|keys], including, excluding) do
-    case dict_delete(excluding, key) do
-      { excluding, _, 0 } -> split(keys, including, excluding)
-      { excluding, value, _ } -> split(keys, put(including, key, value), excluding)
-    end
-  end
-
-  @doc false
-  def reduce(ordered(bucket: bucket), acc, fun) do
-    bucket_reduce(bucket, acc, fun)
-  end
-
-  def reduce(trie(root: root, depth: depth), acc, fun) do
-    node_reduce(root, depth, acc, fun, @node_size)
-  end
-
-  ## Dict-wide functions
-
-  defp dict_fold(ordered(bucket: bucket), acc, fun) do
-    bucket_fold(bucket, acc, fun)
-  end
-
-  defp dict_fold(trie(root: root, depth: depth), acc, fun) do
-    node_fold(root, depth, acc, fun, @node_size)
-  end
-
-  defp dict_put(ordered(size: @ordered_threshold, bucket: bucket), key, value) do
-    root = node_relocate(bucket, 0)
-    dict_put(trie(size: @ordered_threshold, root: root), key, value)
-  end
-
-  defp dict_put(ordered(size: size, bucket: bucket) = dict, key, value) do
-    { new, count } = bucket_put(bucket, key, value)
-    { ordered(dict, size: size + count, bucket: new), count }
-  end
-
-  defp dict_put(trie(root: root, depth: depth, size: size, expand_on: size, contract_on: contract_on) = dict, key, value) do
-    root = node_expand(root, depth, depth + 1)
-    dict = trie(dict, root: root, depth: depth + 1,
-      expand_on: size * @node_size, contract_on: contract_on * @node_size)
-    dict_put(dict, key, value)
-  end
-
-  defp dict_put(trie(root: root, size: size, depth: depth) = dict, key, value) do
-    pos = bucket_hash(key)
-    { root, count } = node_put(root, depth, pos, key, value)
-    { trie(dict, size: size + count, root: root), count }
-  end
-
-  defp dict_delete(ordered(bucket: bucket, size: size) = dict, key) do
-    case bucket_delete(bucket, key) do
-      { _, value, 0 } ->
-        { dict, value, 0 }
-      { new_bucket, value, -1 } ->
-        { ordered(dict, size: size - 1, bucket: new_bucket), value, -1 }
-    end
-  end
-
-  defp dict_delete(trie(root: root, size: size, depth: depth) = dict, key) do
-    pos = bucket_hash(key)
-    case node_delete(root, depth, pos, key) do
-      { _, value, 0 } ->
-        { dict, value, 0 }
-      { root, value, -1 } ->
-        { if depth > 0 and trie(dict, :contract_on) == size do
-          root = node_contract(root, depth)
-          trie(dict,
-            root: root,
-            size: size - 1,
-            depth: depth - 1,
-            contract_on: div(size, @node_size),
-            expand_on: div(trie(dict, :expand_on), @node_size))
-        else
-          trie(dict, size: size - 1, root: root)
-        end, value, -1 }
-    end
-  end
-
-  ## Bucket helpers
-
-  # Get value from the bucket
-  defp bucket_get([[k|_]|_bucket], key) when k > key do
-    :error
-  end
-
-  defp bucket_get([[key|value]|_bucket], key) do
-    { :ok, value }
-  end
-
-  defp bucket_get([_e|bucket], key) do
-    bucket_get(bucket, key)
-  end
-
-  defp bucket_get([], _key) do
-    :error
-  end
-
-  # Puts a value in the bucket
-  defp bucket_put([[k|_]|_]=bucket, key, { :put, value }) when k > key do
-    { [[key|value]|bucket], 1 }
-  end
-
-  defp bucket_put([[k|_]|_]=bucket, key, { :update, initial, _fun }) when k > key do
-    { [[key|initial]|bucket], 1 }
-  end
-
-  defp bucket_put([[key|_]|bucket], key, { :put, value }) do
-    { [[key|value]|bucket], 0 }
-  end
-
-  defp bucket_put([[key|value]|bucket], key, { :update, _initial, fun }) do
-    { [[key|fun.(value)]|bucket], 0 }
-  end
-
-  defp bucket_put([e|bucket], key, value) do
-    { rest, count } = bucket_put(bucket, key, value)
-    { [e|rest], count }
-  end
-
-  defp bucket_put([], key, { :put, value }) do
-    { [[key|value]], 1 }
-  end
-
-  defp bucket_put([], key, { :update, initial, _fun }) do
-    { [[key|initial]], 1 }
-  end
-
-  # Puts a value in the bucket without returning
-  # the operation value
-  defp bucket_put!([[k|_]|_]=bucket, key, value) when k > key, do: [[key|value]|bucket]
-  defp bucket_put!([[key|_]|bucket], key, value), do: [[key|value]|bucket]
-  defp bucket_put!([e|bucket], key, value), do: [e|bucket_put!(bucket, key, value)]
-  defp bucket_put!([], key, value), do: [[key|value]]
-
-  # Deletes a key from the bucket
-  defp bucket_delete([[k|_]|_]=bucket, key) when k > key do
-    { bucket, nil, 0 }
-  end
-
-  defp bucket_delete([[key|value]|bucket], key) do
-    { bucket, value, -1 }
-  end
-
-  defp bucket_delete([e|bucket], key) do
-    { rest, value, count } = bucket_delete(bucket, key)
-    { [e|rest], value, count }
-  end
-
-  defp bucket_delete([], _key) do
-    { [], nil, 0 }
-  end
-
-  # Reduces the bucket
-  defp bucket_reduce([[k|v]|t], acc, fun) do
-    bucket_reduce(t, fun.({ k, v }, acc), fun)
-  end
-
-  defp bucket_reduce([], acc, _fun) do
-    acc
-  end
-
-  defp bucket_fold(bucket, acc, fun) do
-    :lists.foldl(fun, acc, bucket)
-  end
-
-  defp bucket_hash(key) do
-    :erlang.phash2(key)
-  end
-
-  defp bucket_index(hash) do
-    hash &&& @node_bitmap
-  end
-
-  defp bucket_nth_index(hash, n) do
-    (hash >>> (@node_shift * n)) &&& @node_bitmap
-  end
-
-  defp bucket_next(hash) do
-    hash >>> @node_shift
-  end
-
-  ## Node helpers
-
-  # Gets a bucket from the node
-  defp node_bucket(node, 0, hash) do
-    elem(node, bucket_index(hash))
-  end
-
-  defp node_bucket(node, depth, hash) do
-    child = elem(node, bucket_index(hash))
-    node_bucket(child, depth - 1, bucket_next(hash))
-  end
-
-  # Puts a key-value into a node
-  defp node_put(node, 0, hash, key, value) do
-    pos = bucket_index(hash)
-    { new, count } = bucket_put(elem(node, pos), key, value)
-    { set_elem(node, pos, new), count }
-  end
-
-  defp node_put(node, depth, hash, key, value) do
-    pos = bucket_index(hash)
-    { new, count } = node_put(elem(node, pos), depth - 1, bucket_next(hash), key, value)
-    { set_elem(node, pos, new), count }
-  end
-
-  # Deletes a key from the bucket
-  defp node_delete(node, 0, hash, key) do
-    pos = bucket_index(hash)
-    case bucket_delete(elem(node, pos), key) do
-      { _, value, 0 } -> { node, value, 0 }
-      { new, value, -1 } -> { set_elem(node, pos, new), value, -1 }
-    end
-  end
-
-  defp node_delete(node, depth, hash, key) do
-    pos = bucket_index(hash)
-    case node_delete(elem(node, pos), depth - 1, bucket_next(hash), key) do
-      { _, value, 0 } -> { node, value, 0 }
-      { new, value, -1 } -> { set_elem(node, pos, new), value, -1 }
-    end
-  end
-
-  # Reduces a node recursively
-  defp node_reduce(bucket, -1, acc, fun, _) do
-    bucket_reduce(bucket, acc, fun)
-  end
-
-  defp node_reduce(node, depth, acc, fun, count) when count >= 1 do
-    acc = node_reduce(:erlang.element(count, node), depth - 1, acc, fun, @node_size)
-    node_reduce(node, depth, acc, fun, count - 1)
-  end
-
-  defp node_reduce(_node, _, acc, _fun, 0) do
-    acc
-  end
-
-  # Folds a node recursively
-  defp node_fold(bucket, -1, acc, fun, _) do
-    bucket_fold(bucket, acc, fun)
-  end
-
-  defp node_fold(node, depth, acc, fun, count) when count >= 1 do
-    acc = node_fold(:erlang.element(count, node), depth - 1, acc, fun, @node_size)
-    node_fold(node, depth, acc, fun, count - 1)
-  end
-
-  defp node_fold(_node, _, acc, _fun, 0) do
-    acc
-  end
-
-  # Node resizing
-  defp node_expand({ b1, b2, b3, b4, b5, b6, b7, b8 }, 0, n) do
-    { node_relocate(b1, n), node_relocate(b2, n), node_relocate(b3, n),
-      node_relocate(b4, n), node_relocate(b5, n), node_relocate(b6, n),
-      node_relocate(b7, n), node_relocate(b8, n) }
-  end
-
-  defp node_expand({ b1, b2, b3, b4, b5, b6, b7, b8 }, depth, n) do
-    depth = depth - 1
-    { node_expand(b1, depth, n), node_expand(b2, depth, n), node_expand(b3, depth, n),
-      node_expand(b4, depth, n), node_expand(b5, depth, n), node_expand(b6, depth, n),
-      node_expand(b7, depth, n), node_expand(b8, depth, n) }
-  end
-
-  defp node_contract({ b1, b2, b3, b4, b5, b6, b7, b8 }, depth) when depth > 0 do
-    depth = depth - 1
-    { node_contract(b1, depth), node_contract(b2, depth), node_contract(b3, depth),
-      node_contract(b4, depth), node_contract(b5, depth), node_contract(b6, depth),
-      node_contract(b7, depth), node_contract(b8, depth) }
-  end
-
-  defp node_contract({ b1, b2, b3, b4, b5, b6, b7, b8 }, 0) do
-    b1 |> each_contract(b2) |> each_contract(b3) |> each_contract(b4)
-       |> each_contract(b5) |> each_contract(b6) |> each_contract(b7)
-       |> each_contract(b8)
-  end
-
-  defp each_contract([[k|_v]=e|acc], [[key|_value]|_]=bucket) when k < key, do: [e|each_contract(acc, bucket)]
-  defp each_contract(acc, [e|bucket]), do: [e|each_contract(acc, bucket)]
-  defp each_contract([], bucket), do: bucket
-  defp each_contract(acc, []), do: acc
-
-  defp node_relocate(node // @node_template, bucket, n) do
-    :lists.foldl fn [key|value], acc ->
-      pos = key |> bucket_hash() |> bucket_nth_index(n)
-      set_elem(acc, pos, bucket_put!(elem(acc, pos), key, value))
-    end, node, bucket
-  end
-end
-
-defimpl Enumerable, for: HashDict do
-  def reduce(dict, acc, fun),  do: HashDict.reduce(dict, acc, fun)
-  def member?(dict, { k, v }), do: match?({ :ok, ^v }, HashDict.fetch(dict, k))
-  def member?(_dict, _),       do: false
-  def count(dict),             do: HashDict.size(dict)
-end
-
-defimpl Access, for: HashDict do
-  def access(dict, key), do: HashDict.get(dict, key, nil)
-end
@@ -1,485 +0,0 @@
-defmodule HashSet do
-  @moduledoc """
-  A set store.
-
-  The `HashSet` is meant to work well with both small and
-  large sets. It is an implementation of the `Set` behaviour.
-  For more information about the functions and their APIs,
-  please consult the `Set` module.
-  """
-
-  @behaviour Set
-
-  # The ordered record contains a single bucket.
-  @ordered_threshold 8
-
-  defrecordp :ordered, HashSet,
-    size: 0,
-    bucket: []
-
-  # The bucketed record contains a series of buckets.
-  @expand_load 5
-  @contract_load 2
-  @node_bitmap 0b1111
-  @node_shift 4
-  @node_size 16
-  @node_template :erlang.make_tuple(@node_size, [])
-
-  @expand_default (@node_size * @expand_load)
-  @contract_default @contract_load
-
-  defrecordp :trie, HashSet,
-    size: 0,
-    depth: 0,
-    expand_on: @expand_default,
-    contract_on: @contract_default,
-    root: @node_template
-
-  import Bitwise
-
-  @compile :inline_list_funcs
-  @compile { :inline, bucket_hash: 1, bucket_index: 1, bucket_nth_index: 2, bucket_next: 1 }
-
-  @doc """
-  Creates a new empty set.
-  """
-  def new() do
-    ordered()
-  end
-
-  @doc """
-  Creates a new set from the given enumerable.
-
-  ## Examples
-
-      iex> HashSet.new [1, 1, 2, 3, 3] |> HashSet.to_list
-      [1,2,3]
-
-  """
-  def new(members) do
-    Enum.reduce members, ordered(), fn member, set ->
-      put(set, member)
-    end
-  end
-
-  def union(set1, set2) when is_record(set1, HashSet) and is_record(set2, HashSet) and elem(set1, 1) <= elem(set2, 1) do
-    set_fold set1, set2, fn v1, acc ->
-      put(acc, v1)
-    end
-  end
-
-  def union(set1, set2) when is_record(set1, HashSet) and is_record(set2, HashSet) do
-    set_fold set2, set1, fn v, acc ->
-      put(acc, v)
-    end
-  end
-
-  def intersection(set1, set2) when is_record(set1, HashSet) and is_record(set2, HashSet) and elem(set1, 1) <= elem(set2, 1) do
-    set_filter set1, fn e -> set_member?(set2, e) end
-  end
-
-  def intersection(set1, set2) when is_record(set1, HashSet) and is_record(set2, HashSet) do
-    set_filter set2, fn e -> set_member?(set1, e) end
-  end
-
-  def difference(set1, set2) when is_record(set1, HashSet) and is_record(set2, HashSet) do
-    set_filter set1, fn m -> not set_member?(set2, m) end
-  end
-
-  def member?(set, member) when is_record(set, HashSet) do
-    set_member?(set, member)
-  end
-
-  def empty(_) do
-    ordered()
-  end
-
-  def size(set) do
-    elem(set, 1)
-  end
-
-  def to_list(ordered(bucket: bucket)) do
-    bucket
-  end
-
-  def to_list(set) do
-    set_fold(set, [], &[&1|&2]) |> :lists.reverse
-  end
-
-  def put(set, member) do
-    { set, _ } = set_put(set, member)
-    set
-  end
-
-  def delete(set, member) do
-    { set, _ } = set_delete(set, member)
-    set
-  end
-
-  def equal?(set1, set2) do
-    size = elem(set1, 1)
-    case elem(set2, 1) do
-      ^size ->
-        set_equal?(set1, set2)
-      _ ->
-        false
-    end
-  end
-
-  def subset?(set1, set2) do
-    set_equal?(set1, set2)
-  end
-
-  def disjoint?(set1, set2) do
-    set_disjoint?(set1, set2)
-  end
-
-  @doc false
-  def reduce(ordered(bucket: bucket), acc, fun) do
-    :lists.foldl(fun, acc, bucket)
-  end
-
-  def reduce(trie() = set, acc, fun) do
-    set_fold(set, acc, fun)
-  end
-
-  ## HashSet-wide functions
-
-  defp set_filter(ordered(bucket: bucket, size: size), fun) do
-    { new, removed_count } = bucket_filter(bucket, fun, [], 0)
-    ordered(bucket: new, size: size - removed_count)
-  end
-
-  defp set_filter(trie(root: root, depth: depth, size: size) = set, fun) do
-    { new, removed_count } = node_filter(root, depth, fun, @node_size)
-
-    if depth > 0 and trie(set, :contract_on) >= (size - removed_count) do
-      contract_trie(trie(root: new,
-                         size: size - removed_count,
-                         depth: depth,
-                         contract_on: trie(set, :contract_on),
-                         expand_on:   trie(set, :expand_on)))
-    else
-      trie(size: size - removed_count, root: new, depth: depth)
-    end
-  end
-
-  defp set_put(ordered(size: @ordered_threshold, bucket: bucket), member) do
-    root = node_relocate(bucket, 0)
-    set_put(trie(size: @ordered_threshold, root: root), member)
-  end
-
-  defp set_put(ordered(size: size, bucket: bucket) = set, member) do
-    { new, count } = bucket_put(bucket, member)
-    { ordered(set, size: size + count, bucket: new), count }
-  end
-
-  defp set_put(trie(root: root, depth: depth, size: size, expand_on: size, contract_on: contract_on) = set, member) do
-    root = node_expand(root, depth, depth + 1)
-    set = trie(set, root: root, depth: depth + 1,
-      expand_on: size * @node_size, contract_on: contract_on * @node_size)
-    set_put(set, member)
-  end
-
-  defp set_put(trie(root: root, size: size, depth: depth) = set, member) do
-    pos = bucket_hash(member)
-    { root, count } = node_put(root, depth, pos, member)
-    { trie(set, size: size + count, root: root), count }
-  end
-
-  defp set_member?(ordered(bucket: bucket), member) do
-    :lists.member(member, bucket)
-  end
-
-  defp set_member?(trie(root: root, depth: depth), member) do
-    :lists.member(member, node_bucket(root, depth, bucket_hash(member)))
-  end
-
-  defp set_delete(ordered(bucket: bucket, size: size) = set, member) do
-    case bucket_delete(bucket, member) do
-      { _, 0 } ->
-        { set, 0 }
-      { new_bucket, -1 } ->
-        { ordered(set, size: size - 1, bucket: new_bucket), -1 }
-    end
-  end
-
-  defp set_delete(trie(root: root, size: size, depth: depth) = set, member) do
-    pos = bucket_hash(member)
-    case node_delete(root, depth, pos, member) do
-      { _, 0 } ->
-        { set, 0 }
-      { root, -1 } ->
-        { if depth > 0 and trie(set, :contract_on) == size do
-          root = node_contract(root, depth)
-          trie(set,
-            root: root,
-            size: size - 1,
-            depth: depth - 1,
-            contract_on: div(size, @node_size),
-            expand_on: div(trie(set, :expand_on), @node_size))
-        else
-          trie(set, size: size - 1, root: root)
-        end, -1 }
-    end
-  end
-
-  defp set_equal?(set1, set2) do
-    try do
-      reduce(set1, true, fn member, acc ->
-        case member?(set2, member) do
-          true -> acc
-          _    -> throw(:error)
-        end
-      end)
-    catch
-      :error -> false
-    end
-  end
-
-  defp set_disjoint?(set1, set2) do
-    try do
-      reduce(set1, true, fn member, acc ->
-        case member?(set2, member) do
-          false -> acc
-          _     -> throw(:error)
-        end
-      end)
-    catch
-      :error -> false
-    end
-  end
-
-  defp set_fold(ordered(bucket: bucket), acc, fun) do
-    bucket_fold(bucket, acc, fun)
-  end
-
-  defp set_fold(trie(root: root, depth: depth), acc, fun) do
-    node_fold(root, depth, acc, fun, @node_size)
-  end
-
-  ## Bucket helpers
-
-  defp bucket_filter([e|bucket], fun, acc, count) do
-    case fun.(e) do
-      true  -> bucket_filter(bucket, fun, [e|acc], count)
-      false -> bucket_filter(bucket, fun, acc, count + 1)
-    end
-  end
-
-  defp bucket_filter([], _fun, acc, count) do
-    { :lists.reverse(acc), count }
-  end
-
-  defp bucket_put([m|_]=bucket, member) when m > member do
-    { [member|bucket], 1 }
-  end
-
-  defp bucket_put([member|bucket], member) do
-    { [member|bucket], 0 }
-  end
-
-  defp bucket_put([e|bucket], member) do
-    { rest, count } = bucket_put(bucket, member)
-    { [e|rest], count }
-  end
-
-  defp bucket_put([], member) do
-    { [member], 1 }
-  end
-
-  defp bucket_put!([m|_]=bucket, member)     when m > member, do: [member|bucket]
-  defp bucket_put!([member|bucket], member), do: [member|bucket]
-  defp bucket_put!([e|bucket], member),      do: [e|bucket_put!(bucket, member)]
-  defp bucket_put!([], member),              do: [member]
-
-  # Deletes a key from the bucket
-  defp bucket_delete([m,_|_]=bucket, member) when m > member do
-    { bucket, 0 }
-  end
-
-  defp bucket_delete([member|bucket], member) do
-    { bucket, -1 }
-  end
-
-  defp bucket_delete([e|bucket], member) do
-    { rest, count } = bucket_delete(bucket, member)
-    { [e|rest], count }
-  end
-
-  defp bucket_delete([], _member) do
-    { [], 0 }
-  end
-
-  defp bucket_fold(bucket, acc, fun) do
-    :lists.foldl(fun, acc, bucket)
-  end
-
-  defp bucket_hash(key) do
-    :erlang.phash2(key)
-  end
-
-  defp bucket_nth_index(hash, n) do
-    (hash >>> (@node_shift * n)) &&& @node_bitmap
-  end
-
-  defp bucket_index(hash) do
-    hash &&& @node_bitmap
-  end
-
-  defp bucket_next(hash) do
-    hash >>> @node_shift
-  end
-
-  # Trie resizing
-
-  defp contract_trie(trie(depth: 0) = set) do
-    set
-  end
-
-  defp contract_trie(trie(root: root, depth: depth, size: size, contract_on: contract_on, expand_on: expand_on) = set) when size <= contract_on do
-    new_contract_on = div(contract_on, @node_size)
-    new_expand_on   = div(expand_on, @node_size)
-
-    if new_contract_on == 0, do: new_contract_on =  @contract_default
-    if new_expand_on == 0,   do: new_expand_on = @expand_default
-
-    contract_trie(trie(set, root: node_contract(root, depth),
-                             size: size,
-                             depth: depth - 1,
-                             contract_on: new_contract_on,
-                             expand_on: new_expand_on))
-  end
-
-  defp contract_trie(set) do
-    set
-  end
-
-  # Node helpers
-
-  # Gets a bucket from the node
-  defp node_bucket(node, 0, hash) do
-    elem(node, bucket_index(hash))
-  end
-
-  defp node_bucket(node, depth, hash) do
-    child = elem(node, bucket_index(hash))
-    node_bucket(child, depth - 1, bucket_next(hash))
-  end
-
-  defp node_put(node, 0, hash, member) do
-    pos = bucket_index(hash)
-    { new, count } = bucket_put(elem(node, pos), member)
-    { set_elem(node, pos, new), count }
-  end
-
-  defp node_put(node, depth, hash, member) do
-    pos = bucket_index(hash)
-    { new, count } = node_put(elem(node, pos), depth - 1, bucket_next(hash), member)
-    { set_elem(node, pos, new), count }
-  end
-
-  # Deletes a key from the bucket
-  defp node_delete(node, 0, hash, member) do
-    pos = bucket_index(hash)
-    case bucket_delete(elem(node, pos), member) do
-      { _, 0 } -> { node, 0 }
-      { new, -1 } -> { set_elem(node, pos, new), -1 }
-    end
-  end
-
-  defp node_delete(node, depth, hash, member) do
-    pos = bucket_index(hash)
-    case node_delete(elem(node, pos), depth - 1, bucket_next(hash), member) do
-      { _, 0 } -> { node, 0 }
-      { new, -1 } -> { set_elem(node, pos, new), -1 }
-    end
-  end
-
-  defp node_fold(bucket, -1, acc, fun, _) do
-    bucket_fold(bucket, acc, fun)
-  end
-
-  defp node_fold(node, depth, acc, fun, count) when count >= 1 do
-    acc = node_fold(:erlang.element(count, node), depth - 1, acc, fun, @node_size)
-    node_fold(node, depth, acc, fun, count - 1)
-  end
-
-  defp node_fold(_node, _, acc, _fun, 0) do
-    acc
-  end
-
-  defp node_filter(bucket, -1, fun, _) do
-    bucket_filter(bucket, fun, [], 0)
-  end
-
-  defp node_filter(node, depth, fun, count) when count >= 1 do
-    case node_filter(:erlang.element(count, node), depth - 1, fun, @node_size) do
-      { _, 0 } ->
-        node_filter(node, depth, fun, count - 1)
-      { new_element, count1 } ->
-        { new_node, count2 } = node_filter(:erlang.setelement(count, node, new_element), depth, fun, count - 1)
-        { new_node, count1 + count2 }
-    end
-  end
-
-  defp node_filter(node, _,  _fun, 0) do
-   { node, 0 }
-  end
-
-  defp node_relocate(node // @node_template, bucket, n) do
-    :lists.foldl fn member, acc ->
-      pos = member |> bucket_hash() |> bucket_nth_index(n)
-      set_elem(acc, pos, bucket_put!(elem(acc, pos), member))
-    end, node, bucket
-  end
-
-  # Node resizing
-  defp node_expand({ b1, b2, b3, b4, b5, b6, b7, b8, b9, b10, b11, b12, b13, b14, b15, b16 }, 0, n) do
-    { node_relocate(b1, n), node_relocate(b2, n), node_relocate(b3, n),
-      node_relocate(b4, n), node_relocate(b5, n), node_relocate(b6, n),
-      node_relocate(b7, n), node_relocate(b8, n), node_relocate(b9, n),
-      node_relocate(b10, n), node_relocate(b11, n), node_relocate(b12, n),
-      node_relocate(b13, n), node_relocate(b14, n), node_relocate(b15, n),
-      node_relocate(b16, n) }
-  end
-
-  defp node_expand({ b1, b2, b3, b4, b5, b6, b7, b8, b9, b10, b11, b12, b13, b14, b15, b16 }, depth, n) do
-    depth = depth - 1
-    { node_expand(b1, depth, n), node_expand(b2, depth, n), node_expand(b3, depth, n),
-      node_expand(b4, depth, n), node_expand(b5, depth, n), node_expand(b6, depth, n),
-      node_expand(b7, depth, n), node_expand(b8, depth, n), node_expand(b9, depth, n),
-      node_expand(b10, depth, n), node_expand(b11, depth, n), node_expand(b12, depth, n),
-      node_expand(b13, depth, n), node_expand(b14, depth, n), node_expand(b15, depth, n),
-      node_expand(b16, depth, n) }
-  end
-
-  defp node_contract({ b1, b2, b3, b4, b5, b6, b7, b8, b9, b10, b11, b12, b13, b14, b15, b16 }, depth) when depth > 0 do
-    depth = depth - 1
-    { node_contract(b1, depth), node_contract(b2, depth), node_contract(b3, depth),
-      node_contract(b4, depth), node_contract(b5, depth), node_contract(b6, depth),
-      node_contract(b7, depth), node_contract(b8, depth), node_contract(b9, depth),
-      node_contract(b10, depth), node_contract(b11, depth), node_contract(b12, depth),
-      node_contract(b13, depth), node_contract(b14, depth), node_contract(b15, depth),
-      node_contract(b16, depth) }
-  end
-
-  defp node_contract({ b1, b2, b3, b4, b5, b6, b7, b8, b9, b10, b11, b12, b13, b14, b15, b16 }, 0) do
-    b1 |> each_contract(b2) |> each_contract(b3) |> each_contract(b4)
-       |> each_contract(b5) |> each_contract(b6) |> each_contract(b7)
-       |> each_contract(b8) |> each_contract(b9) |> each_contract(b10)
-       |> each_contract(b11) |> each_contract(b12) |> each_contract(b13)
-       |> each_contract(b14) |> each_contract(b15) |> each_contract(b16)
-  end
-
-  defp each_contract([m1|acc], [m2|_]=bucket) when m1 < m2, do: [m1|each_contract(acc, bucket)]
-  defp each_contract(acc, [m|bucket]), do: [m|each_contract(acc, bucket)]
-  defp each_contract([], bucket), do: bucket
-  defp each_contract(acc, []), do: acc
-end
-
-defimpl Enumerable, for: HashSet do
-  def reduce(set, acc, fun), do: HashSet.reduce(set, acc, fun)
-  def member?(set, v),       do: HashSet.member?(set, v)
-  def count(set),            do: HashSet.size(set)
-end
@@ -1,485 +0,0 @@
-import Kernel, except: [inspect: 1]
-import Inspect.Algebra
-
-defrecord Inspect.Opts, raw: false, limit: 50, pretty: false, width: 80
-
-defprotocol Inspect do
-  @moduledoc """
-  The `Inspect` protocol is responsible for converting any Elixir
-  data structure into an algebra document. This document is then
-  formatted, either in pretty printing format or a regular one.
-
-  The `inspect/2` function receives the entity to be inspected
-  followed by the inspecting options, represented by the record
-  `Inspect.Opts`.
-
-  Inspection is done using the functions available in
-  `Inspect.Algebra` and by calling `Kernel.inspect/2` recursively
-  passing the `Inspect.Opts` as an argument. When `Kernel.inspect/2`
-  receives an `Inspect.Opts` record as the second argument, it returns
-  the underlying algebra document instead of the formatted string.
-
-  Many times, inspecting a structure can be implemented in function
-  of existing entities. For example, here is `HashSet`'s `inspect`
-  implementation:
-
-      defimpl Inspect, for: HashSet do
-        import Inspect.Algebra
-
-        def inspect(dict, opts) do
-          concat ["#HashSet<", Kernel.inspect(HashSet.to_list(dict), opts), ">"]
-        end
-      end
-
-  The `concat` function comes from `Inspect.Algebra` and it
-  concatenates algebra documents together. In the example above,
-  it is concatenating the string `"HashSet<"` (all strings are
-  valid algebra documents that keep their formatting when pretty
-  printed), the document returned by `Kernel.inspect/2` and the
-  other string `">"`.
-
-  Since regular strings are valid entities in an algebra document,
-  an implementation of inspect may simply return a string,
-  although that will devoid it of any pretty-printing.
-  """
-
-  def inspect(thing, opts)
-end
-
-defimpl Inspect, for: Atom do
-  require Macro
-
-  @doc """
-  Represents the atom as an Elixir term. The atoms `false`, `true`
-  and `nil` are simply quoted. Modules are properly represented
-  as modules using the dot notation.
-
-  Notice that in Elixir, all operators can be represented using
-  literal atoms (`:+`, `:-`, etc).
-
-  ## Examples
-
-      iex> inspect(:foo)
-      ":foo"
-      iex> inspect(nil)
-      "nil"
-      iex> inspect(Foo.Bar)
-      "Foo.Bar"
-
-  """
-  def inspect(atom, _opts) do
-    inspect(atom)
-  end
-
-  def inspect(false),  do: "false"
-  def inspect(true),   do: "true"
-  def inspect(nil),    do: "nil"
-  def inspect(:""),    do: ":\"\""
-  def inspect(Elixir), do: "Elixir"
-
-  def inspect(atom) do
-    binary = atom_to_binary(atom)
-
-    cond do
-      valid_atom_identifier?(binary) ->
-        ":" <> binary
-      valid_ref_identifier?(binary) ->
-        "Elixir." <> rest = binary
-        rest
-      atom in [:{}, :[], :<<>>] ->
-        ":" <> binary
-      atom in Macro.binary_ops or atom in Macro.unary_ops ->
-        ":" <> binary
-      true ->
-        << ?:, ?", Inspect.BitString.escape(binary, ?") :: binary, ?" >>
-    end
-  end
-
-  # Detect if atom is an atom alias (Elixir.Foo.Bar.Baz)
-
-  defp valid_ref_identifier?("Elixir" <> rest) do
-    valid_ref_piece?(rest)
-  end
-
-  defp valid_ref_identifier?(_), do: false
-
-  defp valid_ref_piece?(<<?., h, t :: binary>>) when h in ?A..?Z do
-    valid_ref_piece? valid_identifier?(t)
-  end
-
-  defp valid_ref_piece?(<<>>), do: true
-  defp valid_ref_piece?(_),    do: false
-
-  # Detect if atom
-
-  defp valid_atom_identifier?(<<h, t :: binary>>) when h in ?a..?z or h in ?A..?Z or h == ?_ do
-    case valid_identifier?(t) do
-      <<>>   -> true
-      <<??>> -> true
-      <<?!>> -> true
-      _      -> false
-    end
-  end
-
-  defp valid_atom_identifier?(_), do: false
-
-  defp valid_identifier?(<<h, t :: binary>>)
-      when h in ?a..?z
-      when h in ?A..?Z
-      when h in ?0..?9
-      when h == ?_ do
-    valid_identifier? t
-  end
-
-  defp valid_identifier?(other), do: other
-end
-
-defimpl Inspect, for: BitString do
-  @doc %S"""
-  Represents a string as itself escaping all necessary
-  characters. Binaries that contain non-printable characters
-  are printed using the bitstring syntax.
-
-  ## Examples
-
-      iex> inspect("bar")
-      "\"bar\""
-      iex> inspect("f\"oo")
-      "\"f\\\"oo\""
-      iex> inspect(<<0,1,2>>)
-      "<<0, 1, 2>>"
-
-  """
-  def inspect(thing, opts) when is_binary(thing) do
-    if String.printable?(thing) do
-      << ?", escape(thing, ?") :: binary, ?" >>
-    else
-      inspect_bitstring(thing, opts)
-    end
-  end
-
-  def inspect(thing, opts) do
-    inspect_bitstring(thing, opts)
-  end
-
-  ## Escaping
-
-  @doc false
-  def escape(other, char) do
-    escape(other, char, <<>>)
-  end
-
-  defp escape(<< char, t :: binary >>, char, binary) do
-    escape(t, char, << binary :: binary, ?\\, char >>)
-  end
-  defp escape(<<?#, ?{, t :: binary>>, char, binary) do
-    escape(t, char, << binary :: binary, ?\\, ?#, ?{ >>)
-  end
-  defp escape(<<?\a, t :: binary>>, char, binary) do
-    escape(t, char, << binary :: binary, ?\\, ?a >>)
-  end
-  defp escape(<<?\b, t :: binary>>, char, binary) do
-    escape(t, char, << binary :: binary, ?\\, ?b >>)
-  end
-  defp escape(<<?\d, t :: binary>>, char, binary) do
-    escape(t, char, << binary :: binary, ?\\, ?d >>)
-  end
-  defp escape(<<?\e, t :: binary>>, char, binary) do
-    escape(t, char, << binary :: binary, ?\\, ?e >>)
-  end
-  defp escape(<<?\f, t :: binary>>, char, binary) do
-    escape(t, char, << binary :: binary, ?\\, ?f >>)
-  end
-  defp escape(<<?\n, t :: binary>>, char, binary) do
-    escape(t, char, << binary :: binary, ?\\, ?n >>)
-  end
-  defp escape(<<?\r, t :: binary>>, char, binary) do
-    escape(t, char, << binary :: binary, ?\\, ?r >>)
-  end
-  defp escape(<<?\\, t :: binary>>, char, binary) do
-    escape(t, char, << binary :: binary, ?\\, ?\\ >>)
-  end
-  defp escape(<<?\t, t :: binary>>, char, binary) do
-    escape(t, char, << binary :: binary, ?\\, ?t >>)
-  end
-  defp escape(<<?\v, t :: binary>>, char, binary) do
-    escape(t, char, << binary :: binary, ?\\, ?v >>)
-  end
-  defp escape(<<h, t :: binary>>, char, binary) do
-    escape(t, char, << binary :: binary, h >>)
-  end
-  defp escape(<<>>, _char, binary), do: binary
-
-  ## Bitstrings
-
-  defp inspect_bitstring(bitstring, Inspect.Opts[] = opts) do
-    each_bit(bitstring, opts.limit, "<<") <> ">>"
-  end
-
-  defp each_bit(_, 0, acc) do
-    acc <> "..."
-  end
-
-  defp each_bit(<<h, t :: bitstring>>, counter, acc) when t != <<>> do
-    each_bit(t, decrement(counter), acc <> integer_to_binary(h) <> ", ")
-  end
-
-  defp each_bit(<<h :: size(8)>>, _counter, acc) do
-    acc <> integer_to_binary(h)
-  end
-
-  defp each_bit(<<>>, _counter, acc) do
-    acc
-  end
-
-  defp each_bit(bitstring, _counter, acc) do
-    size = bit_size(bitstring)
-    <<h :: size(size)>> = bitstring
-    acc <> integer_to_binary(h) <> "::size(" <> integer_to_binary(size) <> ")"
-  end
-
-  defp decrement(:infinity), do: :infinity
-  defp decrement(counter),   do: counter - 1
-end
-
-defimpl Inspect, for: List do
-  @doc %S"""
-  Represents a list, checking if it can be printed or not.
-  If so, a single-quoted representation is returned,
-  otherwise the brackets syntax is used. Keywords are
-  printed in keywords syntax.
-
-  ## Examples
-
-      iex> inspect('bar')
-      "'bar'"
-      iex> inspect([0|'bar'])
-      "[0, 98, 97, 114]"
-      iex> inspect([:foo,:bar])
-      "[:foo, :bar]"
-
-  """
-
-  def inspect([], _opts), do: "[]"
-
-  def inspect(thing, Inspect.Opts[] = opts) do
-    cond do
-      :io_lib.printable_list(thing) ->
-        << ?', Inspect.BitString.escape(String.from_char_list!(thing), ?') :: binary, ?' >>
-      keyword?(thing) && not opts.raw ->
-        surround_many("[", thing, "]", opts.limit, &keyword(&1, opts))
-      true ->
-        surround_many("[", thing, "]", opts.limit, &Kernel.inspect(&1, opts))
-    end
-  end
-
-  defp keyword({key, value}, opts) do
-    concat(
-      key_to_binary(key) <> ": ",
-      Kernel.inspect(value, opts)
-    )
-  end
-
-  defp key_to_binary(key) do
-    case Inspect.Atom.inspect(key) do
-      ":" <> right -> right
-      other -> other
-    end
-  end
-
-  defp keyword?([{ key, _value } | rest]) when is_atom(key) do
-    case atom_to_list(key) do
-      'Elixir.' ++ _ -> false
-      _ -> keyword?(rest)
-    end
-  end
-
-  defp keyword?([]),     do: true
-  defp keyword?(_other), do: false
-end
-
-defimpl Inspect, for: Tuple do
-  @doc """
-  Represents tuples. If the tuple represents a record,
-  it shows it nicely formatted using the access syntax.
-
-  ## Examples
-
-      iex> inspect({1, 2, 3})
-      "{1, 2, 3}"
-      iex> inspect(ArgumentError.new)
-      "ArgumentError[message: \\\"argument error\\\"]"
-
-  """
-
-  def inspect({}, _opts), do: "{}"
-
-  def inspect(tuple, opts) do
-    unless opts.raw do
-      record_inspect(tuple, opts)
-    end || surround_many("{", tuple_to_list(tuple), "}", opts.limit, &Kernel.inspect(&1, opts))
-  end
-
-  ## Helpers
-
-  defp record_inspect(record, opts) do
-    [name|tail] = tuple_to_list(record)
-
-    if is_atom(name) && (fields = record_fields(name)) && (length(fields) == size(record) - 1) do
-      if Enum.first(tail) == :__exception__ do
-        surround_record(name, tl(fields), tl(tail), opts)
-      else
-        surround_record(name, fields, tail, opts)
-      end
-    end || surround_many("{", [name|tail], "}", opts.limit, &Kernel.inspect(&1, opts))
-  end
-
-  defp record_fields(name) do
-    case atom_to_binary(name) do
-      "Elixir." <> _ ->
-        try do
-          name.__record__(:fields)
-        rescue
-          _ -> nil
-        end
-      _ -> nil
-    end
-  end
-
-  defp surround_record(name, fields, tail, opts) do
-    fields = lc { field, _ } inlist fields, do: field
-
-    concat(
-      Inspect.Atom.inspect(name, opts),
-      surround_many("[", Enum.zip(fields, tail), "]", opts.limit, &keyword(&1, opts))
-    )
-  end
-
-  defp keyword({ k, v }, opts) do
-    concat(
-      atom_to_binary(k) <> ": ",
-      Kernel.inspect(v, opts)
-    )
-  end
-end
-
-defimpl Inspect, for: Integer do
-  @doc """
-  Represents the integer as a string.
-
-  ## Examples
-
-      iex> inspect(1)
-      "1"
-
-  """
-  def inspect(thing, _opts) do
-    integer_to_binary(thing)
-  end
-end
-
-defimpl Inspect, for: Float do
-  @doc """
-  Floats are represented using the shortened, correctly rounded string
-  that converts to float when read back with `binary_to_float/1`. This
-  is done via the Erlang implementation of _Printing Floating-Point
-  Numbers Quickly and Accurately_ in Proceedings of the SIGPLAN '96
-  Conference on Programming Language Design and Implementation.
-
-  ## Examples
-
-      iex> inspect(1.0)
-      "1.0"
-
-  """
-  def inspect(thing, _opts) do
-    iolist_to_binary(:io_lib_format.fwrite_g(thing))
-  end
-end
-
-defimpl Inspect, for: Regex do
-  @doc %S"""
-  Represents the Regex using the `%r""` syntax.
-
-  ## Examples
-
-      iex> inspect(%r/foo/m)
-      "%r\"foo\"m"
-
-  """
-  def inspect(regex, opts) when size(regex) == 5 do
-    concat ["%r", Kernel.inspect(Regex.source(regex), opts), Regex.opts(regex)]
-  end
-
-  def inspect(other, opts) do
-    Kernel.inspect(other, opts.raw(true))
-  end
-end
-
-defimpl Inspect, for: Function do
-  def inspect(function, _opts) do
-    fun_info = :erlang.fun_info(function)
-    mod = fun_info[:module]
-
-    if fun_info[:type] == :external and fun_info[:env] == [] do
-      "&#{Inspect.Atom.inspect(mod)}.#{fun_info[:name]}/#{fun_info[:arity]}"
-    else
-      case atom_to_list(mod) do
-        'elixir_compiler_' ++ _ ->
-          if function_exported?(mod, :__RELATIVE__, 0) do
-            "#Function<#{uniq(fun_info)} in file:#{mod.__RELATIVE__}>"
-          else
-            default_inspect(mod, fun_info)
-          end
-        _ ->
-          default_inspect(mod, fun_info)
-      end
-    end
-  end
-
-  defp default_inspect(mod, fun_info) do
-    "#Function<#{uniq(fun_info)} in #{Inspect.Atom.inspect(mod)}.#{extract_name(fun_info[:name])}>"
-  end
-
-  defp extract_name(name) do
-    case :binary.split(atom_to_binary(name), "-", [:global]) do
-      ["", name | _] -> name
-      _ -> name
-    end
-  end
-
-  defp uniq(fun_info) do
-    integer_to_binary(fun_info[:new_index]) <> "." <> integer_to_binary(fun_info[:uniq])
-  end
-end
-
-defimpl Inspect, for: PID do
-  def inspect(pid, _opts) do
-    "#PID" <> iolist_to_binary(:erlang.pid_to_list(pid))
-  end
-end
-
-defimpl Inspect, for: Port do
-  def inspect(port, _opts) do
-    iolist_to_binary :erlang.port_to_list(port)
-  end
-end
-
-defimpl Inspect, for: Reference do
-  def inspect(ref, _opts) do
-    '#Ref' ++ rest = :erlang.ref_to_list(ref)
-    "#Reference" <> iolist_to_binary(rest)
-  end
-end
-
-defimpl Inspect, for: HashDict do
-  def inspect(dict, opts) do
-    concat ["#HashDict<", Inspect.List.inspect(HashDict.to_list(dict), opts), ">"]
-  end
-end
-
-defimpl Inspect, for: HashSet do
-  def inspect(set, opts) do
-    concat ["#HashSet<", Inspect.List.inspect(HashSet.to_list(set), opts), ">"]
-  end
-end
@@ -1,424 +0,0 @@
-defmodule Inspect.Algebra do
-  @moduledoc %S"""
-  A set of functions for creating and manipulating algebra
-  documents, as described in ["Strictly Pretty" (2000) by Christian Lindig][0].
-
-  An algebra document is represented by an `Inspect.Algebra` node
-  or a regular string.
-
-      iex> Inspect.Algebra.empty
-      :doc_nil
-      iex> "foo"
-      "foo"
-
-  With the functions in this module, we can concatenate different
-  elements together and render them:
-
-      iex> doc = Inspect.Algebra.concat(Inspect.Algebra.empty, "foo")
-      iex> Inspect.Algebra.pretty(doc, 80)
-      "foo"
-
-  The functions `nest/2`, `space/2` and `line/2` help you put the
-  document together into a rigid structure. However, the document
-  algebra gets interesting when using functions like `break/2`, which
-  converts the given string into a line break depending on how much space
-  there is to print. Let's glue two docs together with a break and then
-  render it:
-
-      iex> doc = Inspect.Algebra.glue("a", " ", "b")
-      iex> Inspect.Algebra.pretty(doc, 80)
-      "a b"
-
-  Notice the break was represented as is, because we haven't reached
-  a line limit. Once we do, it is replaced by a newline:
-
-      iex> doc = Inspect.Algebra.glue(String.duplicate("a", 20), " ", "b")
-      iex> Inspect.Algebra.pretty(doc, 10)
-      "aaaaaaaaaaaaaaaaaaaa\nb"
-
-  Finally, this module also contains Elixir related functions, a bit
-  tied to Elixir formatting, namely `surround/3` and `surround_many/5`.
-
-  ## Implementation details
-
-  The original Haskell implementation of the algorithm by [Wadler][1]
-  relies on lazy evaluation to unfold document groups on two alternatives:
-  `:flat` (breaks as spaces) and `:break` (breaks as newlines).
-  Implementing the same logic in a strict language such as Elixir leads
-  to an exponential growth of possible documents, unless document groups
-  are encoded explictly as `:flat` or `:break`. Those groups are then reduced
-  to a simple document, where the layout is already decided, per [Lindig][0].
-
-  This implementation slightly changes the semantic of Lindig's algorithm
-  to allow elements that belong to the same group to be printed together
-  in the same line, even if they do not fit the line fully. This was achieved
-  by changing `:break` to mean a possible break and `:flat` to force a flat
-  structure. Then deciding if a break works as a newline is just a matter
-  of checking if we have enough space until the next break that is not
-  inside a group (which is still flat).
-
-  Custom pretty printers can be implemented using the documents returned
-  by this module and by providing their own rendering functions.
-
-    [0]: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.34.2200
-    [1]: http://homepages.inf.ed.ac.uk/wadler/papers/prettier/prettier.pdf
-
-  """
-
-  @surround_separator ","
-  @tail_separator " |"
-  @newline "\n"
-  @nesting 1
-  @break " "
-
-  defp repeat(_, 0), do: ""
-  defp repeat(s, i), do: :lists.duplicate(i, s)
-
-  # Functional interface to `doc` records
-
-  @type t :: :doc_nil | doc_cons_t | doc_nest_t | doc_break_t | doc_group_t | binary
-  defrecordp :doc_cons, left: :doc_nil :: t, right: :doc_nil :: t
-  defrecordp :doc_nest, indent: 1 :: non_neg_integer, doc: :doc_nil :: t
-  defrecordp :doc_break, str: " " :: binary
-  defrecordp :doc_group, doc: :doc_nil :: t
-
-  defmacrop is_doc(doc) do
-    if __CALLER__.in_guard? do
-      do_is_doc(doc)
-    else
-      var = quote do: doc
-      quote do
-        unquote(var) = unquote(doc)
-        unquote(do_is_doc(var))
-      end
-    end
-  end
-
-  defp do_is_doc(doc) do
-    quote do
-      unquote(doc) |> is_binary or
-      unquote(doc) |> is_integer or
-      unquote(doc) == :doc_nil or
-      (unquote(doc) |> is_tuple and
-       elem(unquote(doc), 0) in [:doc_cons, :doc_nest, :doc_break, :doc_group])
-    end
-  end
-
-  @doc """
-  Returns `:doc_nil` which is a document entity used to represent
-  nothingness. Takes no arguments.
-
-  ## Examples
-
-      iex> Inspect.Algebra.empty
-      :doc_nil
-
-  """
-  @spec empty() :: :doc_nil
-  def empty, do: :doc_nil
-
-  @doc """
-  Concatenates two document entities. Takes two arguments:
-  left doc and right doc. Returns a DocCons doc
-
-  ## Examples
-
-      iex> doc = Inspect.Algebra.concat "Tasteless", "Artosis"
-      iex> Inspect.Algebra.pretty(doc, 80)
-      "TastelessArtosis"
-
-  """
-  @spec concat(t, t) :: doc_cons_t
-  def concat(x, y) when is_doc(x) and is_doc(y) do
-    doc_cons(left: x, right: y)
-  end
-
-  @doc """
-  Concatenates a list of documents.
-  """
-  @spec concat([t]) :: doc_cons_t
-  def concat(docs) do
-    folddoc(docs, &concat(&1, &2))
-  end
-
-  @doc """
-  Nests document entity `x` positions deep. Nesting will be
-  appended to the line breaks.
-
-  ## Examples
-
-      iex> doc = Inspect.Algebra.nest(Inspect.Algebra.concat(Inspect.Algebra.break, "6"), 5)
-      iex> Inspect.Algebra.pretty(doc, 80)
-      " 6"
-
-  """
-  @spec nest(t, non_neg_integer) :: doc_nest_t
-  def nest(x, 0) when is_doc(x) do
-    x
-  end
-
-  def nest(x, i) when is_doc(x) and is_integer(i) do
-    doc_nest(indent: i, doc: x)
-  end
-
-  @doc %S"""
-  Document entity representing a break. This break can
-  be rendered as a linebreak or as spaces, depending on the
-  `mode` of the chosen layout or the provided separator.
-
-  ## Examples
-
-  Let's glue two docs together with a break and then render it:
-
-      iex> doc = Inspect.Algebra.glue("a", " ", "b")
-      iex> Inspect.Algebra.pretty(doc, 80)
-      "a b"
-
-  Notice the break was represented as is, because we haven't reached
-  a line limit. Once we do, it is replaced by a newline:
-
-      iex> doc = Inspect.Algebra.glue(String.duplicate("a", 20), " ", "b")
-      iex> Inspect.Algebra.pretty(doc, 10)
-      "aaaaaaaaaaaaaaaaaaaa\nb"
-
-  """
-  @spec break(binary) :: doc_break_t
-  def break(s) when is_binary(s), do: doc_break(str: s)
-
-  @spec break() :: doc_break_t
-  def break(), do: doc_break(str: @break)
-
-  @doc """
-  Inserts a break between two docs. See `break/1` for more info.
-  """
-  @spec glue(t, t) :: doc_cons_t
-  def glue(x, y), do: concat(x, concat(break, y))
-
-  @doc """
-  Inserts a break, passed as the second argument, between two docs,
-  the first and the third arguments.
-  """
-  @spec glue(t, binary, t) :: doc_cons_t
-  def glue(x, g, y) when is_binary(g), do: concat(x, concat(break(g), y))
-
-  @doc %S"""
-  Returns a group containing the specified document.
-
-  ## Examples
-
-      iex> doc = Inspect.Algebra.group(
-      ...>   Inspect.Algebra.concat(
-      ...>     Inspect.Algebra.group(
-      ...>       Inspect.Algebra.concat(
-      ...>         "Hello,",
-      ...>         Inspect.Algebra.concat(
-      ...>           Inspect.Algebra.break,
-      ...>           "A"
-      ...>         )
-      ...>       )
-      ...>     ),
-      ...>     Inspect.Algebra.concat(
-      ...>       Inspect.Algebra.break,
-      ...>       "B"
-      ...>     )
-      ...> ))
-      iex> Inspect.Algebra.pretty(doc, 80)
-      "Hello, A B"
-      iex> Inspect.Algebra.pretty(doc, 6)
-      "Hello,\nA B"
-
-  """
-  @spec group(t) :: doc_group_t
-  def group(d) when is_doc(d) do
-    doc_group(doc: d)
-  end
-
-  @doc """
-  Inserts a mandatory single space between two document entities.
-
-  ## Examples
-
-      iex> doc = Inspect.Algebra.space "Hughes", "Wadler"
-      iex> Inspect.Algebra.pretty(doc, 80)
-      "Hughes Wadler"
-
-  """
-  @spec space(t, t) :: doc_cons_t
-  def space(x, y), do: concat(x, concat(" ", y))
-
-  @doc %S"""
-  Inserts a mandatory linebreak between two document entities.
-
-  ## Examples
-
-      iex> doc = Inspect.Algebra.line "Hughes", "Wadler"
-      iex> Inspect.Algebra.pretty(doc, 80)
-      "Hughes\nWadler"
-
-  """
-  @spec line(t, t) :: doc_cons_t
-  def line(x, y), do: concat(x, concat(@newline, y))
-
-  @doc """
-  Folds a list of document entities into a document entity
-  using a function that is passed as the first argument.
-
-  ## Examples
-
-      iex> doc = ["A", "B"]
-      iex> doc = Inspect.Algebra.folddoc(doc, fn(x,y) ->
-      ...>   Inspect.Algebra.concat [x, "!", y]
-      ...> end)
-      iex> Inspect.Algebra.pretty(doc, 80)
-      "A!B"
-
-  """
-  @spec folddoc([t], ((t, t) -> t)) :: t
-  def folddoc([], _), do: empty
-  def folddoc([doc], _), do: doc
-  def folddoc([d|ds], f), do: f.(d, folddoc(ds, f))
-
-  # Elixir conveniences
-
-  @doc %S"""
-  Surrounds a document with characters.
-
-  Puts the document between left and right enclosing and nesting it.
-  The document is marked as a group, to show the maximum as possible
-  concisely together.
-
-  ## Examples
-
-      iex> doc = Inspect.Algebra.surround "[", Inspect.Algebra.glue("a", "b"), "]"
-      iex> Inspect.Algebra.pretty(doc, 3)
-      "[a\n b]"
-
-  """
-  @spec surround(binary, t, binary) :: t
-  def surround(left, doc, right) do
-    group concat [left, nest(doc, @nesting), right]
-  end
-
-  @doc %S"""
-  Maps and glues a collection of items together using the given separator
-  and surrounds them. A limit can be passed which, once reached, stops
-  gluing and outputs "..." instead.
-
-  ## Examples
-
-      iex> doc = Inspect.Algebra.surround_many("[", Enum.to_list(1..5), "]", :infinity, &integer_to_binary(&1))
-      iex> Inspect.Algebra.pretty(doc, 5)
-      "[1,\n 2,\n 3,\n 4,\n 5]"
-
-      iex> doc = Inspect.Algebra.surround_many("[", Enum.to_list(1..5), "]", 3, &integer_to_binary(&1))
-      iex> Inspect.Algebra.pretty(doc, 20)
-      "[1, 2, 3, ...]"
-
-  """
-  @spec surround_many(binary, [any], binary, integer | :infinity, (term -> t)) :: t
-  def surround_many(left, [], right, _, _fun) do
-    concat(left, right)
-  end
-
-  def surround_many(left, docs, right, limit, fun) do
-    surround(left, surround_many(docs, limit, fun), right)
-  end
-
-  defp surround_many(_, 0, _fun) do
-    "..."
-  end
-
-  defp surround_many([h], _limit, fun) do
-    fun.(h)
-  end
-
-  defp surround_many([h|t], limit, fun) when is_list(t) do
-    glue(
-      concat(fun.(h), @surround_separator),
-      surround_many(t, decrement(limit), fun)
-    )
-  end
-
-  defp surround_many([h|t], _limit, fun) do
-    glue(
-      concat(fun.(h), @tail_separator),
-      fun.(t)
-    )
-  end
-
-  defp decrement(:infinity), do: :infinity
-  defp decrement(counter),   do: counter - 1
-
-  @doc """
-  The pretty printing function.
-
-  Takes the maximum width and a document to print as its arguments
-  and returns the string representation of the best layout for the
-  document to fit in the given width.
-  """
-  @spec pretty(t, non_neg_integer | :infinity) :: binary
-  def pretty(d, w) do
-    sdoc = format w, 0, [{0, default_mode(w), doc_group(doc: d)}]
-    render(sdoc)
-  end
-
-  defp default_mode(:infinity), do: :flat
-  defp default_mode(_),         do: :break
-
-  # Rendering and internal helpers
-
-  # Records representing __simple__ documents, already on a fixed layout
-  # Those are generalized by `sdoc` type.
-  @type sdoc :: :s_nil | s_text_t | s_line_t
-  defrecordp :s_text, str: "" :: binary, sdoc: :s_nil :: sdoc
-  defrecordp :s_line, indent: 1 :: non_neg_integer, sdoc: :s_nil :: sdoc
-
-  # Record representing the document mode to be rendered: flat or broken
-  @typep mode :: :flat | :break
-
-  @doc false
-  @spec fits?(integer, [{ integer, mode, t }]) :: boolean
-  def fits?(:infinity, _),                                  do: true # no pretty printing
-  def fits?(w, _) when w < 0,                               do: false
-  def fits?(_, []),                                         do: true
-  def fits?(w, [{_, _, :doc_nil} | t]),                     do: fits?(w, t)
-  def fits?(w, [{i, m, doc_cons(left: x, right: y)} | t]),  do: fits?(w, [{i, m, x} | [{i, m, y} | t]])
-  def fits?(w, [{i, m, doc_nest(indent: j, doc: x)} | t]),  do: fits?(w, [{i + j, m, x} | t])
-  def fits?(w, [{_, _, s} | t]) when is_binary(s),          do: fits?((w - byte_size s), t)
-  def fits?(w, [{_, :flat, doc_break(str: s)} | t]),        do: fits?((w - byte_size s), t)
-  def fits?(_, [{_, :break, doc_break(str: _)} | _]),       do: true
-  def fits?(w, [{i, _, doc_group(doc: x)} | t]),            do: fits?(w, [{i, :flat, x} | t])
-
-  @doc false
-  @spec format(integer | :infinity, integer, [{ integer, mode, t }]) :: atom | tuple
-  def format(_, _, []),                                        do: :s_nil
-  def format(w, k, [{_, _, :doc_nil} | t]),                    do: format(w, k, t)
-  def format(w, k, [{i, m, doc_cons(left: x, right: y)} | t]), do: format(w, k, [{i, m, x} | [{i, m, y} | t]])
-  def format(w, k, [{i, m, doc_nest(indent: j, doc: x)} | t]), do: format(w, k, [{i + j, m, x} | t])
-  def format(w, k, [{_, _, s} | t]) when is_binary(s),         do: s_text(str: s, sdoc: format(w, (k + byte_size s), t))
-  def format(w, k, [{i, m, doc_group(doc: x)} | t]),           do: format(w, k, [{i, m, x} | t])
-  def format(w, k, [{_, :flat, doc_break(str: s)} | t]),       do: s_text(str: s, sdoc: format(w, (k + byte_size s), t))
-  def format(w, k, [{i, :break, doc_break(str: s)} | t]) do
-    k = k + byte_size(s)
-
-    if w == :infinity or fits?(w - k, t) do
-      s_text(str: s, sdoc: format(w, k, t))
-    else
-      s_line(indent: i, sdoc: format(w, i, t))
-    end
-  end
-
-  @doc false
-  @spec render(sdoc) :: binary
-  def render(sdoc) do
-    iolist_to_binary do_render sdoc
-  end
-
-  @spec do_render(sdoc) :: [binary]
-  defp do_render(:s_nil), do: [""]
-  defp do_render(s_text(str: s, sdoc: d)), do: [s | do_render(d)]
-  defp do_render(s_line(indent: i, sdoc: d)) do
-    prefix = repeat " ", i
-    [@newline | [prefix | do_render d]]
-  end
-end
@@ -1,70 +0,0 @@
-defmodule Integer do
-  @moduledoc """
-  Functions for working with integers.
-  """
-
-  import Bitwise
-
-  @doc """
-  Determines if an integer is odd.
-
-  Returns `true` if `n` is an odd number, otherwise `false`.
-  Implemented as a macro so it is allowed in guard clauses.
-  """
-  defmacro odd?(n) do
-    quote do: (unquote(n) &&& 1) == 1
-  end
-
-  @doc """
-  Determines if an integer is even.
-
-  Returns `true` if `n` is an even number, otherwise `false`.
-  Implemented as a macro so it is allowed in guard clauses.
-  """
-  defmacro even?(n) do
-    quote do: (unquote(n) &&& 1) == 0
-  end
-
-  @doc """
-  Converts a binary to an integer. 
-  
-  If successful, returns a tuple of the form `{ integer, remainder_of_binary }`.
-  Otherwise `:error`.
-
-  ## Examples
-
-      iex> Integer.parse("34")
-      {34,""}
-      iex> Integer.parse("34.5")
-      {34,".5"}
-      iex> Integer.parse("three")
-      :error
-
-  """
-  @spec parse(binary) :: { integer, binary } | :error
-  def parse(<< ?-, bin :: binary >>) do
-    case do_parse(bin) do
-      :error -> :error
-      { number, remainder } -> { -number, remainder }
-    end
-  end
-
-  def parse(<< ?+, bin :: binary >>) do
-    do_parse(bin)
-  end
-
-  def parse(bin) when is_binary(bin) do
-    do_parse(bin)
-  end
-
-  defp do_parse(<< char, bin :: binary >>) when char in ?0..?9, do: do_parse(bin, char - ?0)
-  defp do_parse(_), do: :error
-
-  defp do_parse(<< char, rest :: binary >>, acc) when char in ?0..?9 do
-    do_parse rest, 10 * acc + (char - ?0)
-  end
-
-  defp do_parse(bitstring, acc) do
-    { acc, bitstring }
-  end
-end
@@ -1,293 +0,0 @@
-defexception IO.StreamError, reason: nil do
-  def message(exception) do
-    formatted = iolist_to_binary(:file.format_error(reason exception))
-    "error during streaming: #{formatted}"
-  end
-end
-
-defmodule IO do
-  @moduledoc """
-  Functions handling IO.
-
-  Many functions in this module expects an IO device as argument.
-  An IO device must be a pid or an atom representing a process.
-  For convenience, Elixir provides `:stdio` and `:stderr` as
-  shortcuts to Erlang's `:standard_io` and `:standard_error`.
-
-  The majority of the functions expect data encoded in UTF-8
-  and will do a conversion to string, via the `String.Chars`
-  protocol (as shown in typespecs).
-
-  The functions starting with `bin*` expects iodata as arguments,
-  i.e. iolists or binaries with no particular encoding.
-
-  """
-
-  @type device   :: atom | pid
-  @type chardata :: char_list | String.Chars.t
-  @type nodata :: { :error, term } | :eof
-
-  import :erlang, only: [group_leader: 0]
-
-  defmacrop is_iolist(data) do
-    quote do
-      is_list(unquote(data)) or is_binary(unquote(data))
-    end
-  end
-
-  @doc """
-  Reads `count` characters from the IO device or until
-  the end of the line if `:line` is given. It returns:
-
-  * `data` - The input characters.
-
-  * `:eof` - End of file was encountered.
-
-  * `{:error, reason}` - Other (rare) error condition,
-    for instance `{:error, :estale}` if reading from an
-    NFS file system.
-  """
-  @spec read(device, :line | non_neg_integer) :: chardata | nodata
-  def read(device // group_leader, chars_or_line)
-
-  def read(device, :line) do
-    :io.get_line(map_dev(device), '')
-  end
-
-  def read(device, count) when count >= 0 do
-    :io.get_chars(map_dev(device), '', count)
-  end
-
-  @doc """
-  Reads `count` bytes from the IO device or until
-  the end of the line if `:line` is given. It returns:
-
-  * `data` - The input characters.
-
-  * `:eof` - End of file was encountered.
-
-  * `{:error, reason}` - Other (rare) error condition,
-    for instance `{:error, :estale}` if reading from an
-    NFS file system.
-  """
-  @spec binread(device, :line | non_neg_integer) :: iodata | nodata
-  def binread(device // group_leader, chars_or_line)
-
-  def binread(device, :line) do
-    case :file.read_line(map_dev(device)) do
-      { :ok, data } -> data
-      other -> other
-    end
-  end
-
-  def binread(device, count) when count >= 0 do
-    case :file.read(map_dev(device), count) do
-      { :ok, data } -> data
-      other -> other
-    end
-  end
-
-  @doc """
-  Writes the given argument to the given device.
-  By default the device is the standard output.
-  The argument is expected to be a chardata (i.e.
-  a char list or an unicode binary).
-
-  It returns `:ok` if it succeeds.
-
-  ## Examples
-
-      IO.write "sample"
-      #=> "sample"
-
-      IO.write :stderr, "error"
-      #=> "error"
-
-  """
-  @spec write(device, chardata) :: :ok
-  def write(device // group_leader(), item) do
-    :io.put_chars map_dev(device), to_chardata(item)
-  end
-
-  @doc """
-  Writes the given argument to the given device
-  as a binary, no unicode conversion happens.
-
-  Check `write/2` for more information.
-  """
-  @spec binwrite(device, iodata) :: :ok | { :error, term }
-  def binwrite(device // group_leader(), item) when is_iolist(item) do
-    :file.write map_dev(device), item
-  end
-
-  @doc """
-  Writes the argument to the device, similar to `write/2`,
-  but adds a newline at the end. The argument is expected
-  to be a chardata.
-  """
-  @spec puts(device, chardata) :: :ok
-  def puts(device // group_leader(), item) do
-    erl_dev = map_dev(device)
-    :io.put_chars erl_dev, [to_chardata(item), ?\n]
-  end
-
-  @doc """
-  Inspects and writes the given argument to the device
-  followed by a newline. A set of options can be given.
-
-  It sets by default pretty printing to true and the
-  width to be the width of the device, with a minimum
-  of 80 characters.
-
-  ## Examples
-
-      IO.inspect Process.list
-
-  """
-  @spec inspect(term, Keyword.t) :: term
-  def inspect(item, opts // []) do
-    inspect group_leader(), item, opts
-  end
-
-  @doc """
-  Inspects the item with options using the given device.
-  """
-  @spec inspect(device, term, Keyword.t) :: term
-  def inspect(device, item, opts) when is_list(opts) do
-    opts = Keyword.put_new(opts, :pretty, true)
-
-    unless Keyword.get(opts, :width) do
-      opts = case :io.columns(device) do
-        { :ok, width } -> Keyword.put(opts, :width, width)
-        { :error, _ }  -> opts
-      end
-    end
-
-    puts device, Kernel.inspect(item, opts)
-    item
-  end
-
-  @doc """
-  Gets a number of bytes from the io device. If the
-  io device is a unicode device, `count` implies
-  the number of unicode codepoints to be retrieved.
-  Otherwise, `count` is the number of raw bytes to be retrieved.
-  It returns:
-
-  * `data` - The input characters.
-
-  * `:eof` - End of file was encountered.
-
-  * `{:error, reason}` - Other (rare) error condition,
-    for instance `{:error, :estale}` if reading from an
-    NFS file system.
-  """
-  @spec getn(chardata, pos_integer) :: chardata | nodata
-  @spec getn(device, chardata) :: chardata | nodata
-  def getn(prompt, count // 1)
-
-  def getn(prompt, count) when is_integer(count) do
-    getn(group_leader, prompt, count)
-  end
-
-  def getn(device, prompt) do
-    getn(device, prompt, 1)
-  end
-
-  @doc """
-  Gets a number of bytes from the io device. If the
-  io device is a unicode device, `count` implies
-  the number of unicode codepoints to be retrieved.
-  Otherwise, `count` is the number of raw bytes to be retrieved.
-  """
-  @spec getn(device, chardata, pos_integer) :: chardata | nodata
-  def getn(device, prompt, count) do
-    :io.get_chars(map_dev(device), to_chardata(prompt), count)
-  end
-
-  @doc """
-  Reads a line from the IO device. It returns:
-
-  * `data` - The characters in the line terminated
-    by a LF (or end of file).
-
-  * `:eof` - End of file was encountered.
-
-  * `{:error, reason}` - Other (rare) error condition,
-    for instance `{:error, :estale}` if reading from an
-    NFS file system.
-  """
-  @spec gets(device, chardata) :: chardata | nodata
-  def gets(device // group_leader(), prompt) do
-    :io.get_line(map_dev(device), to_chardata(prompt))
-  end
-
-  @doc """
-  Converts the io device into a Stream. The device is
-  iterated line by line if `:line` is given or by a given
-  number of codepoints.
-
-  This reads the IO as utf-8. Check out
-  `IO.binstream/2` to handle the IO as a raw binary.
-
-  Note that an IO stream has side effects and every time
-  you go over the stream you may get different results.
-
-  ## Examples
-
-  Here is an example on how we mimic an echo server
-  from the command line:
-
-      Enum.each IO.stream(:stdio, :line), &IO.write(&1)
-
-  """
-  @spec stream(device, :line | pos_integer) :: Enumerable.t
-  def stream(device, line_or_codepoints) do
-    fn(acc, f) -> stream(map_dev(device), line_or_codepoints, acc, f) end
-  end
-
-  @doc """
-  Converts the io device into a Stream. The device is
-  iterated line by line or by a number of bytes. This
-  reads the IO as a raw binary.
-
-  Note that an IO stream has side effects and every time
-  you go over the stream you may get different results.
-  """
-  @spec binstream(device, :line | pos_integer) :: Enumerable.t
-  def binstream(device, line_or_bytes) do
-    fn(acc, f) -> binstream(map_dev(device), line_or_bytes, acc, f) end
-  end
-
-  @doc false
-  def stream(device, what, acc, fun) do
-    case read(device, what) do
-      :eof ->
-        acc
-      { :error, reason } ->
-        raise IO.StreamError, reason: reason
-      data ->
-        stream(device, what, fun.(data, acc), fun)
-    end
-  end
-
-  @doc false
-  def binstream(device, what, acc, fun) do
-    case binread(device, what) do
-      :eof ->
-        acc
-      { :error, reason } ->
-        raise IO.StreamError, reason: reason
-      data ->
-        binstream(device, what, fun.(data, acc), fun)
-    end
-  end
-
-  # Map the Elixir names for standard io and error to Erlang names
-  defp map_dev(:stdio),  do: :standard_io
-  defp map_dev(:stderr), do: :standard_error
-  defp map_dev(other),   do: other
-
-  defp to_chardata(list) when is_list(list), do: list
-  defp to_chardata(other), do: to_string(other)
-end
@@ -1,222 +0,0 @@
-defmodule IO.ANSI.Sequence do
-  @moduledoc false
-
-  defmacro defsequence(name, code // "", terminator // "m") do
-    quote bind_quoted: [name: name, code: code, terminator: terminator] do
-      def unquote(name)() do
-        "\e[#{unquote(code)}#{unquote(terminator)}"
-      end
-
-      defp escape_sequence(<< unquote(atom_to_binary(name)), rest :: binary >>) do
-        { "\e[#{unquote(code)}#{unquote(terminator)}", rest }
-      end
-    end
-  end
-end
-
-defmodule IO.ANSI do
-  @moduledoc """
-  Functionality to render ANSI escape sequences
-  (http://en.wikipedia.org/wiki/ANSI_escape_code) —  characters embedded
-  in text used to control formatting, color, and other output options
-  on video text terminals.
-  """
-
-  import IO.ANSI.Sequence
-
-  @doc """
-  Checks whether the default I/O device is a terminal or a file.
-
-  Used to identify whether printing ANSI escape sequences will likely
-  be displayed as intended.
-  """
-  @spec terminal? :: boolean
-  @spec terminal?(:io.device) :: boolean
-  def terminal?(device // :erlang.group_leader) do
-    match?({:ok, _}, :io.columns(device))
-  end
-
-  @doc "Resets all attributes"
-  defsequence :reset, 0
-
-  @doc "Bright (increased intensity) or Bold"
-  defsequence :bright, 1
-
-  @doc "Faint (decreased intensity), not widely supported"
-  defsequence :faint, 2
-
-  @doc "Italic: on. Not widely supported. Sometimes treated as inverse."
-  defsequence :italic, 3
-
-  @doc "Underline: Single"
-  defsequence :underline, 4
-
-  @doc "Blink: Slow. Less than 150 per minute"
-  defsequence :blink_slow, 5
-
-  @doc "Blink: Rapid. MS-DOS ANSI.SYS; 150 per minute or more; not widely supported"
-  defsequence :blink_rapid, 6
-
-  @doc "Image: Negative. Swap foreground and background"
-  defsequence :inverse, 7
-
-  @doc "Image: Negative. Swap foreground and background"
-  defsequence :reverse, 7
-
-  @doc "Conceal. Not widely supported"
-  defsequence :conceal, 8
-
-  @doc "Crossed-out. Characters legible, but marked for deletion. Not widely supported."
-  defsequence :crossed_out, 9
-
-  @doc "Sets primary (default) font"
-  defsequence :primary_font, 10
-
-  lc font_n inlist [1, 2, 3, 4, 5, 6, 7, 8, 9] do
-    @doc "Sets alternative font #{font_n}"
-    defsequence :"font_#{font_n}", font_n + 10
-  end
-
-  @doc "Normal color or intensity"
-  defsequence :normal, 22
-
-  @doc "Not italic"
-  defsequence :not_italic, 23
-
-  @doc "Underline: None"
-  defsequence :no_underline, 24
-
-  @doc "Blink: off"
-  defsequence :blink_off, 25
-
-  colors = [:black, :red, :green, :yellow, :blue, :magenta, :cyan, :white]
-  colors = Enum.zip(0..(length(colors)-1), colors)
-
-  lc { code, color } inlist colors do
-    @doc "Sets foreground color to #{color}"
-    defsequence color, code + 30
-
-    @doc "Sets background color to #{color}"
-    defsequence :"#{color}_background", code + 40
-  end
-
-  @doc "Default text color"
-  defsequence :default_color, 39
-
-  @doc "Default background color"
-  defsequence :default_background, 49
-
-  @doc "Framed"
-  defsequence :framed, 51
-
-  @doc "Encircled"
-  defsequence :encircled, 52
-
-  @doc "Overlined"
-  defsequence :overlined, 53
-
-  @doc "Not framed or encircled"
-  defsequence :not_framed_encircled, 54
-
-  @doc "Not overlined"
-  defsequence :not_overlined, 55
-
-  @doc "Send cursor home"
-  defsequence :home, "", "H"
-
-  @doc "Clear screen"
-  defsequence :clear, "2", "J"
-
-
-  # Catch spaces between codes
-  defp escape_sequence(<< ?\s, rest :: binary >>) do
-    escape_sequence(rest)
-  end
-
-  defp escape_sequence(other) do
-    [spec|_] = String.split(other, %r/(,|\})/)
-    raise ArgumentError, message: "invalid ANSI sequence specification: #{spec}"
-  end
-
-  @doc %S"""
-  Escapes a string by converting named ANSI sequences into actual ANSI codes.
-
-  The format for referring to sequences is `%{red}` and `%{red,bright}` (for
-  multiple sequences).
-
-  It will also append a `%{reset}` to the string. If you don't want this
-  behaviour, use `escape_fragment/2`.
-
-  An optional boolean parameter can be passed to enable or disable
-  emitting actual ANSI codes. When `false`, no ANSI codes will emitted.
-  By default, standard output will be checked if it is a terminal capable
-  of handling these sequences (using `terminal?/1` function)
-
-  ## Examples
-
-      iex> IO.ANSI.escape("Hello %{red,bright,green}yes", true)
-      "Hello \e[31m\e[1m\e[32myes\e[0m"
-
-  """
-  @spec escape(String.t, emit :: boolean) :: String.t
-  def escape(string, emit // terminal?) do
-    {rendered, emitted} = do_escape(string, false, emit, false, [])
-    if emitted and emit do
-      rendered <> reset
-    else
-      rendered
-    end
-  end
-
-  @doc %S"""
-  Escapes a string by converting named ANSI sequences into actual ANSI codes.
-
-  The format for referring to sequences is `%{red}` and `%{red,bright}` (for
-  multiple sequences).
-
-  An optional boolean parameter can be passed to enable or disable
-  emitting actual ANSI codes. When `false`, no ANSI codes will emitted.
-  By default, standard output will be checked if it is a terminal capable
-  of handling these sequences (using `terminal?/1` function)
-
-  ## Examples
-
-      iex> IO.ANSI.escape_fragment("Hello %{red,bright,green}yes", true)
-      "Hello \e[31m\e[1m\e[32myes"
-      iex> IO.ANSI.escape_fragment("%{reset}bye", true)
-      "\e[0mbye"
-
-  """
-  @spec escape_fragment(String.t, emit :: boolean) :: String.t
-  def escape_fragment(string, emit // terminal?) do
-    {rendered, _emitted} = do_escape(string, false, emit, false, [])
-    rendered
-  end
-
-  defp do_escape(<< ?%, ?{, rest :: binary >>, false, emit, _emitted, acc) do
-    do_escape_sequence(rest, emit, acc)
-  end
-  defp do_escape(<< ?,, rest :: binary >>, true, emit, _emitted, acc) do
-    do_escape_sequence(rest, emit, acc)
-  end
-  defp do_escape(<< ?\s, rest :: binary >>, true, emit, emitted, acc) do
-    do_escape(rest, true, emit, emitted, acc)
-  end
-  defp do_escape(<< ?}, rest :: binary >>, true, emit, emitted, acc) do
-    do_escape(rest, false, emit, emitted, acc)
-  end
-  defp do_escape(<< x :: [binary, size(1)], rest :: binary>>, false, emit, emitted, acc) do
-    do_escape(rest, false, emit, emitted, [x|acc])
-  end
-  defp do_escape("", false, _emit, emitted, acc) do
-    {iolist_to_binary(Enum.reverse(acc)), emitted}
-  end
-
-  defp do_escape_sequence(rest, emit, acc) do
-    {code, rest} = escape_sequence(rest)
-    if emit do
-      acc = [code|acc]
-    end
-    do_escape(rest, true, emit, true, acc)
-  end
-end
@@ -1,376 +0,0 @@
-defmodule Kernel.CLI do
-  @moduledoc false
-
-  defrecord Config, commands: [], output: ".", compile: [],
-                    halt: true, compiler_options: [], errors: [],
-                    verbose_compile: false
-
-  @doc """
-  This is the API invoked by Elixir boot process.
-  """
-  def main(argv) do
-    argv = lc arg inlist argv, do: String.from_char_list!(arg)
-
-    { config, argv } = process_argv(argv, Kernel.CLI.Config.new)
-    System.argv(argv)
-
-    run fn ->
-      command_results = Enum.map(Enum.reverse(config.commands), &process_command(&1, config))
-      command_errors  = lc { :error, msg } inlist command_results, do: msg
-      errors          = Enum.reverse(config.errors) ++ command_errors
-
-      if errors != [] do
-        Enum.each(errors, &IO.puts(:stderr, &1))
-        System.halt(1)
-      end
-    end, config.halt
-  end
-
-  @doc """
-  Runs the given function by catching any failure
-  and printing them to stdout. `at_exit` hooks are
-  also invoked before exiting.
-
-  This function is used by Elixir's CLI and also
-  by escripts generated by Elixir.
-  """
-  def run(fun, halt // true) do
-    try do
-      fun.()
-      if halt do
-        at_exit(0)
-        System.halt(0)
-      end
-    catch
-      :exit, reason when is_integer(reason) ->
-        at_exit(reason)
-        System.halt(reason)
-      :exit, :normal ->
-        at_exit(0)
-        System.halt(0)
-      kind, reason ->
-        at_exit(1)
-        print_error(kind, Exception.normalize(kind, reason), System.stacktrace)
-        System.halt(1)
-    end
-  end
-
-  ## Helpers
-
-  defp at_exit(status) do
-    hooks = :elixir_code_server.call(:flush_at_exit)
-
-    lc hook inlist hooks do
-      try do
-        hook.(status)
-      catch
-        kind, reason ->
-          print_error(kind, Exception.normalize(kind, reason), System.stacktrace)
-      end
-    end
-
-    # If an at_exit callback adds a
-    # new hook we need to invoke it.
-    unless hooks == [], do: at_exit(status)
-  end
-
-  defp shared_option?(list, config, callback) do
-    case process_shared(list, config) do
-      { [h|hs], _ } when h == hd(list) ->
-        new_config = config.update_errors &["#{h} : Unknown option" | &1]
-        callback.(hs, new_config)
-      { new_list, new_config } ->
-        callback.(new_list, new_config)
-    end
-  end
-
-  defp print_error(:error, exception, trace) do
-    IO.puts :stderr, "** (#{inspect exception.__record__(:name)}) #{exception.message}"
-    IO.puts :stderr, format_stacktrace(trace)
-  end
-
-  defp print_error(kind, reason, trace) do
-    IO.puts :stderr, "** #{kind} #{inspect(reason)}"
-    IO.puts :stderr, format_stacktrace(trace)
-  end
-
-  defp format_stacktrace(stack) do
-    Exception.format_stacktrace(prune_stacktrace(stack))
-  end
-
-  @elixir_internals [:elixir_compiler, :elixir_module]
-
-  defp prune_stacktrace([{ mod, _, _, _ }|t]) when mod in @elixir_internals do
-    prune_stacktrace(t)
-  end
-
-  defp prune_stacktrace([{ __MODULE__, :wrapper, 1, _ }|_]) do
-    []
-  end
-
-  defp prune_stacktrace([h|t]) do
-    [h|prune_stacktrace(t)]
-  end
-
-  defp prune_stacktrace([]) do
-    []
-  end
-
-  # Process shared options
-
-  defp process_shared([opt|_t], _config) when opt in ["-v", "--version"] do
-    IO.puts "Elixir #{System.version}"
-    System.halt 0
-  end
-
-  defp process_shared(["-pa", h|t], config) do
-    Enum.each Path.wildcard(Path.expand(h)), &Code.prepend_path(&1)
-    process_shared t, config
-  end
-
-  defp process_shared(["-pz", h|t], config) do
-    Enum.each Path.wildcard(Path.expand(h)), &Code.append_path(&1)
-    process_shared t, config
-  end
-
-  defp process_shared(["--app", h|t], config) do
-    process_shared t, config.update_commands &[{:app, h}|&1]
-  end
-
-  defp process_shared(["--no-halt"|t], config) do
-    process_shared t, config.halt(false)
-  end
-
-  defp process_shared(["-e", h|t], config) do
-    process_shared t, config.update_commands &[{:eval, h}|&1]
-  end
-
-  defp process_shared(["-r", h|t], config) do
-    process_shared t, config.update_commands &[{:require, h}|&1]
-  end
-
-  defp process_shared(["-pr", h|t], config) do
-    process_shared t, config.update_commands &[{:parallel_require, h}|&1]
-  end
-
-  defp process_shared([erl, _|t], config) when erl in ["--erl", "--sname", "--name", "--cookie"] do
-    process_shared t, config
-  end
-
-  defp process_shared([erl|t], config) when erl in ["--detached", "--hidden"] do
-    process_shared t, config
-  end
-
-  defp process_shared(list, config) do
-    { list, config }
-  end
-
-  # Process init options
-
-  defp process_argv(["--"|t], config) do
-    { config, t }
-  end
-
-  defp process_argv(["+compile"|t], config) do
-    process_compiler t, config
-  end
-
-  defp process_argv(["+iex"|t], config) do
-    process_iex t, config
-  end
-
-  defp process_argv(["-S", h|t], config) do
-    { config.update_commands(&[{:script, h}|&1]), t }
-  end
-
-  defp process_argv([h|t] = list, config) do
-    case h do
-      "-" <> _ ->
-        shared_option? list, config, &process_argv(&1, &2)
-      _ ->
-        { config.update_commands(&[{:file, h}|&1]), t }
-    end
-  end
-
-  defp process_argv([], config) do
-    { config, [] }
-  end
-
-  # Process compiler options
-
-  defp process_compiler(["--"|t], config) do
-    { config, t }
-  end
-
-  defp process_compiler(["-o", h|t], config) do
-    process_compiler t, config.output(h)
-  end
-
-  defp process_compiler(["--no-docs"|t], config) do
-    process_compiler t, config.update_compiler_options(&[{:docs, false}|&1])
-  end
-
-  defp process_compiler(["--no-debug-info"|t], config) do
-    process_compiler t, config.update_compiler_options(&[{:debug_info, false}|&1])
-  end
-
-  defp process_compiler(["--ignore-module-conflict"|t], config) do
-    process_compiler t, config.update_compiler_options(&[{:ignore_module_conflict, true}|&1])
-  end
-
-  defp process_compiler(["--warnings-as-errors"|t], config) do
-    process_compiler t, config.update_compiler_options(&[{:warnings_as_errors, true}|&1])
-  end
-  
-  defp process_compiler(["--verbose"|t], config) do
-    process_compiler t, config.verbose_compile(true)
-  end
-
-  defp process_compiler([h|t] = list, config) do
-    case h do
-      "-" <> _ ->
-        shared_option? list, config, &process_compiler(&1, &2)
-      _ ->
-        pattern = if :filelib.is_dir(h), do: "#{h}/**/*.ex", else: h
-        process_compiler t, config.update_compile &[pattern|&1]
-    end
-  end
-
-  defp process_compiler([], config) do
-    { config.update_commands(&[{:compile, config.compile}|&1]), [] }
-  end
-
-  # Process iex options
-
-  defp process_iex(["--"|t], config) do
-    { config, t }
-  end
-
-  # This clause is here so that Kernel.CLI does not error out with "unknown
-  # option"
-  defp process_iex(["--dot-iex", _|t], config) do
-    process_iex t, config
-  end
-
-  defp process_iex([opt, _|t], config) when opt in ["--remsh"] do
-    process_iex t, config
-  end
-
-  defp process_iex(["-S", h|t], config) do
-    { config.update_commands(&[{:script, h}|&1]), t }
-  end
-
-  defp process_iex([h|t] = list, config) do
-    case h do
-      "-" <> _ ->
-        shared_option? list, config, &process_iex(&1, &2)
-      _ ->
-        { config.update_commands(&[{:file, h}|&1]), t }
-    end
-  end
-
-  defp process_iex([], config) do
-    { config, [] }
-  end
-
-  # Process commands
-
-  defp process_command({:cookie, h}, _config) do
-    if Node.alive? do
-      wrapper fn -> Node.set_cookie(binary_to_atom(h)) end
-    else
-      { :error, "--cookie : Cannot set cookie if the node is not alive (set --name or --sname)" }
-    end
-  end
-
-  defp process_command({:eval, expr}, _config) when is_binary(expr) do
-    wrapper fn -> Code.eval_string(expr, []) end
-  end
-
-  defp process_command({:app, app}, _config) when is_binary(app) do
-    case Application.Behaviour.start(binary_to_atom(app)) do
-      { :error, reason } ->
-        { :error, "--app : Could not start application #{app}: #{inspect reason}" }
-      :ok ->
-        :ok
-    end
-  end
-
-  defp process_command({:script, file}, _config) when is_binary(file) do
-    if exec = find_elixir_executable(file) do
-      wrapper fn -> Code.require_file(exec) end
-    else
-      { :error, "-S : Could not find executable #{file}" }
-    end
-  end
-
-  defp process_command({:file, file}, _config) when is_binary(file) do
-    if :filelib.is_regular(file) do
-      wrapper fn -> Code.require_file(file) end
-    else
-      { :error, "No file named #{file}" }
-    end
-  end
-
-  defp process_command({:require, pattern}, _config) when is_binary(pattern) do
-    files = Path.wildcard(pattern)
-    files = Enum.uniq(files)
-    files = Enum.filter files, &:filelib.is_regular(&1)
-
-    if files != [] do
-      wrapper fn -> Enum.map files, &Code.require_file(&1) end
-    else
-      { :error, "-r : No files matched pattern #{pattern}" }
-    end
-  end
-
-  defp process_command({:parallel_require, pattern}, _config) when is_binary(pattern) do
-    files = Path.wildcard(pattern)
-    files = Enum.uniq(files)
-    files = Enum.filter files, &:filelib.is_regular(&1)
-
-    if files != [] do
-      wrapper fn -> Kernel.ParallelRequire.files(files) end
-    else
-      { :error, "-pr : No files matched pattern #{pattern}" }
-    end
-  end
-
-  defp process_command({:compile, patterns}, config) do
-    :filelib.ensure_dir(:filename.join(config.output, "."))
-
-    files = Enum.map patterns, &Path.wildcard(&1)
-    files = Enum.uniq(Enum.concat(files))
-    files = Enum.filter files, &:filelib.is_regular(&1)
-
-    if files != [] do
-      wrapper fn ->
-        Code.compiler_options(config.compiler_options)
-        Kernel.ParallelCompiler.files_to_path(files, config.output,
-          each_file: fn file -> if config.verbose_compile do IO.puts "Compiled #{file}" end end)
-      end
-    else
-      { :error, "--compile : No files matched patterns #{Enum.join(patterns, ",")}" }
-    end
-  end
-
-  defp wrapper(fun) do
-    fun.()
-    :ok
-  end
-
-  defp find_elixir_executable(file) do
-    if exec = System.find_executable(file) do
-      # If we are on Windows, the executable is going to be
-      # a .bat file that must be in the same directory as
-      # the actual Elixir executable.
-      case :os.type() do
-        { :win32, _ } ->
-          exec = Path.rootname(exec)
-          if File.regular?(exec), do: exec
-        _ ->
-          exec
-      end
-    end
-  end
-end
@@ -1,27 +0,0 @@
-# Implement error_handler pattern for Erlang
-# which is integrated with Kernel.ParallelCompiler
-defmodule Kernel.ErrorHandler do
-  @moduledoc false
-
-  def undefined_function(module, fun, args) do
-    ensure_loaded(module)
-    :error_handler.undefined_function(module, fun, args)
-  end
-
-  def undefined_lambda(module, fun, args) do
-    ensure_loaded(module)
-    :error_handler.undefined_lambda(module, fun, args)
-  end
-
-  defp ensure_loaded(module) do
-    case Code.ensure_loaded(module) do
-      { :module, _ } -> []
-      { :error, _ } ->
-        parent = :erlang.get(:elixir_compiler_pid)
-        parent <- { :waiting, self(), module }
-        receive do
-          { :release, ^parent } -> :ok
-        end
-    end
-  end
-end
@@ -1,169 +0,0 @@
-# This is a module Elixir responsible for tracking
-# the usage of aliases, imports and requires in the Elixir scope.
-#
-# The implementation simply stores dispatch information in an
-# ETS table and then consults this table once compilation is done.
-#
-# Note that since this is required for bootstrap, we can't use
-# any of the `GenServer.Behaviour` conveniences.
-defmodule Kernel.LexicalTracker do
-  @timeout  30_000
-  @behavior :gen_server
-
-  @import 2
-  @alias  3
-
-  @doc """
-  Returns all remotes linked to in this lexical scope.
-  """
-  def remotes(pid) do
-    ets = :gen_server.call(to_pid(pid), :ets, @timeout)
-    :ets.match(ets, { :"$1", :_, :_ }) |> List.flatten
-  end
-
-  defp to_pid(pid) when is_pid(pid),  do: pid
-  defp to_pid(mod) when is_atom(mod) do
-    table = :elixir_module.data_table(mod)
-    [{ _, val }] = :ets.lookup(table, :__lexical_tracker)
-    val
-  end
-
-  # Internal API
-
-  # Starts the tracker and returns its pid.
-  @doc false
-  def start_link do
-    { :ok, pid } = :gen_server.start_link(__MODULE__, [], [])
-    pid
-  end
-
-  @doc false
-  def stop(pid) do
-    :gen_server.cast(pid, :stop)
-  end
-
-  @doc false
-  def add_import(pid, module, line, warn) do
-    :gen_server.cast(pid, { :add_import, module, line, warn })
-  end
-
-  @doc false
-  def add_alias(pid, module, line, warn) do
-    :gen_server.cast(pid, { :add_alias, module, line, warn })
-  end
-
-  @doc false
-  def remote_dispatch(pid, module) do
-    :gen_server.cast(pid, { :remote_dispatch, module })
-  end
-
-  @doc false
-  def import_dispatch(pid, module) do
-    :gen_server.cast(pid, { :import_dispatch, module })
-  end
-
-  @doc false
-  def alias_dispatch(pid, module) do
-    :gen_server.cast(pid, { :alias_dispatch, module })
-  end
-
-  @doc false
-  def collect_unused_imports(pid) do
-    unused(pid, @import)
-  end
-
-  @doc false
-  def collect_unused_aliases(pid) do
-    unused(pid, @alias)
-  end
-
-  defp unused(pid, pos) do
-    ets = :gen_server.call(pid, :ets, @timeout)
-    :ets.foldl(fn
-      { module, _, _ } = tuple, acc when is_integer(:erlang.element(pos, tuple)) ->
-        [{ module, :erlang.element(pos, tuple) }|acc]
-      _, acc ->
-        acc
-    end, [], ets) |> Enum.sort
-  end
-
-  # Callbacks
-
-
-  def init([]) do
-    { :ok, :ets.new(:lexical, [:protected]) }
-  end
-
-  def handle_call(:ets, _from, d) do
-    { :reply, d, d }
-  end
-
-  def handle_call(_request, _from, d) do
-    { :noreply, d }
-  end
-
-  def handle_cast({ :remote_dispatch, module }, d) do
-    add_module(d, module)
-    { :noreply, d }
-  end
-
-  def handle_cast({ :import_dispatch, module }, d) do
-    add_dispatch(d, module, @import)
-    { :noreply, d }
-  end
-
-  def handle_cast({ :alias_dispatch, module }, d) do
-    add_dispatch(d, module, @alias)
-    { :noreply, d }
-  end
-
-  def handle_cast({ :add_import, module, line, warn }, d) do
-    add_directive(d, module, line, warn, @import)
-    { :noreply, d }
-  end
-
-  def handle_cast({ :add_alias, module, line, warn }, d) do
-    add_directive(d, module, line, warn, @alias)
-    { :noreply, d }
-  end
-
-  def handle_cast(:stop, d) do
-    { :stop, :normal, d }
-  end
-
-  def handle_cast(_msg, d) do
-    { :noreply, d }
-  end
-
-  def handle_info(_msg, d) do
-    { :noreply, d }
-  end
-
-  def terminate(_reason, _d) do
-    :ok
-  end
-
-  def code_change(_old, d, _extra) do
-    { :ok, d }
-  end
-
-  # Callbacks helpers
-
-  # In the table we keep imports and aliases.
-  # If the value is false, it was not imported/aliased
-  # If the value is true, it was imported/aliased
-  # If the value is a line, it was imported/aliased and has a pending warning
-  defp add_module(d, module) do
-    :ets.insert_new(d, { module, false, false })
-  end
-
-  defp add_dispatch(d, module, pos) do
-    :ets.update_element(d, module, { pos, true })
-  end
-
-  defp add_directive(d, module, line, warn, pos) do
-    add_module(d, module)
-    marker = if warn, do: line, else: true
-    :ets.update_element(d, module, { pos, marker })
-  end
-end
@@ -1,170 +0,0 @@
-defmodule Kernel.ParallelCompiler do
-  @moduledoc """
-  A module responsible for compiling files in parallel.
-  """
-
-  @doc """
-  Compiles the given files.
-
-  Those files are compiled in parallel and can automatically
-  detect dependencies between them. Once a dependency is found,
-  the current file stops being compiled until the dependency is
-  resolved.
-
-  If there is an error during compilation or if `warnings_as_errors`
-  is set to `true` and there is a warning, this function will fail
-  with an exception.
-
-  This function receives a set of callbacks as options:
-
-  * `:each_file` - for each file compiled, invokes the callback passing the file
-  * `:each_module` - for each module compiled, invokes the callback
-                     passing the file, module and the module bytecode
-
-  The compiler doesn't care about the return values of the callbacks.
-  Returns the modules generated by each compiled file.
-  """
-  def files(files, callbacks // [])
-
-  def files(files, callbacks) when is_list(callbacks) do
-    spawn_compilers(files, nil, callbacks)
-  end
-
-  @doc """
-  Compiles the given files to the given path.
-  Read `files/2` for more information.
-  """
-  def files_to_path(files, path, callbacks // [])
-
-  def files_to_path(files, path, callbacks) when is_binary(path) and is_list(callbacks) do
-    spawn_compilers(files, path, callbacks)
-  end
-
-  defp spawn_compilers(files, path, callbacks) do
-    Code.ensure_loaded(Kernel.ErrorHandler)
-    :elixir_code_server.cast(:reset_warnings)
-    schedulers = max(:erlang.system_info(:schedulers_online), 2)
-
-    result = spawn_compilers(files, files, path, callbacks, [], [], schedulers, [])
-
-    # In case --warning-as-errors is enabled and there was a warning,
-    # compilation status will be set to error and we fail with CompileError
-    case :elixir_code_server.call(:compilation_status) do
-      :ok    -> result
-      :error -> raise CompileError, [], []
-    end
-  end
-
-  # We already have 4 currently running, don't spawn new ones
-  defp spawn_compilers(files, original, output, callbacks, waiting, queued, schedulers, result) when
-      length(queued) - length(waiting) >= schedulers do
-    wait_for_messages(files, original, output, callbacks, waiting, queued, schedulers, result)
-  end
-
-  # Spawn a compiler for each file in the list until we reach the limit
-  defp spawn_compilers([h|t], original, output, callbacks, waiting, queued, schedulers, result) do
-    parent = self()
-
-    child  = spawn_link fn ->
-      :erlang.put(:elixir_compiler_pid, parent)
-      :erlang.put(:elixir_ensure_compiled, true)
-      :erlang.process_flag(:error_handler, Kernel.ErrorHandler)
-
-      try do
-        if output do
-          :elixir_compiler.file_to_path(h, output)
-        else
-          :elixir_compiler.file(h)
-        end
-        parent <- { :compiled, self(), h }
-      catch
-        kind, reason ->
-          parent <- { :failure, self(), kind, reason, System.stacktrace }
-      end
-    end
-
-    spawn_compilers(t, original, output, callbacks, waiting, [{child, h}|queued], schedulers, result)
-  end
-
-  # No more files, nothing waiting, queue is empty, we are done
-  defp spawn_compilers([], _original, _output, _callbacks, [], [], _schedulers, result), do: result
-
-  # Queued x, waiting for x: POSSIBLE ERROR! Release processes so we get the failures
-  defp spawn_compilers([], original, output, callbacks, waiting, queued, schedulers, result) when length(waiting) == length(queued) do
-    Enum.each queued, fn { child, _ } -> child <- { :release, self() } end
-    wait_for_messages([], original, output, callbacks, waiting, queued, schedulers, result)
-  end
-
-  # No more files, but queue and waiting are not full or do not match
-  defp spawn_compilers([], original, output, callbacks, waiting, queued, schedulers, result) do
-    wait_for_messages([], original, output, callbacks, waiting, queued, schedulers, result)
-  end
-
-  # Wait for messages from child processes
-  defp wait_for_messages(files, original, output, callbacks, waiting, queued, schedulers, result) do
-    receive do
-      { :compiled, child, file } ->
-        if callback = Keyword.get(callbacks, :each_file) do
-          callback.(file)
-        end
-        new_queued  = List.keydelete(queued, child, 0)
-
-        # Sometimes we may have spurious entries in the waiting
-        # list because someone invoked try/rescue UndefinedFunctionError
-        new_waiting = List.keydelete(waiting, child, 0)
-        spawn_compilers(files, original, output, callbacks, new_waiting, new_queued, schedulers, result)
-      { :module_available, child, ref, file, module, binary } ->
-        if callback = Keyword.get(callbacks, :each_module) do
-          callback.(file, module, binary)
-        end
-
-        # Release the module loader which is waiting for an ack
-        child <- { ref, :ack }
-
-        new_waiting = release_waiting_processes(module, waiting)
-        new_result  = [module|result]
-        wait_for_messages(files, original, output, callbacks, new_waiting, queued, schedulers, new_result)
-      { :waiting, child, on } ->
-        new_waiting = :orddict.store(child, on, waiting)
-        spawn_compilers(files, original, output, callbacks, new_waiting, queued, schedulers, result)
-      { :failure, child, kind, reason, stacktrace } ->
-        if many_missing?(child, files, waiting, queued) do
-          IO.puts "== Compilation failed =="
-          IO.puts "Compilation failed on the following files:\n"
-
-          Enum.each Enum.reverse(queued), fn { pid, file } ->
-            case List.keyfind(waiting, pid, 0) do
-              { _, mod } -> IO.puts "* #{file} is missing module #{inspect mod}"
-              _ -> :ok
-            end
-          end
-
-          IO.puts "\nThe first failure is shown below..."
-        end
-
-        {^child, file} = List.keyfind(queued, child, 0)
-        IO.puts "== Compilation error on file #{file} =="
-        :erlang.raise(kind, reason, stacktrace)
-    end
-  end
-
-  defp many_missing?(child, files, waiting, queued) do
-    waiting_length = length(waiting)
-
-    match?({ ^child, _ }, List.keyfind(waiting, child, 0)) and
-      waiting_length > 1 and files == [] and
-      waiting_length == length(queued)
-  end
-
-  # Release waiting processes that are waiting for the given module
-  defp release_waiting_processes(module, waiting) do
-    Enum.filter waiting, fn { child, waiting_module } ->
-      if waiting_module == module do
-        child <- { :release, self() }
-        false
-      else
-        true
-      end
-    end
-  end
-end
@@ -1,70 +0,0 @@
-defmodule Kernel.ParallelRequire do
-  @moduledoc """
-  A module responsible for requiring files in parallel.
-  """
-
-  defmacrop default_callback, do: quote(do: fn x -> x end)
-
-  @doc """
-  Requires the given files.
-
-  A callback that is invoked every time a file is required
-  can be optionally given as argument.
-
-  Returns the modules generated by each required file.
-  """
-  def files(files, callback // default_callback) do
-    schedulers = max(:erlang.system_info(:schedulers_online), 2)
-    spawn_requires(files, [], callback, schedulers, [])
-  end
-
-  defp spawn_requires([], [], _callback, _schedulers, result), do: result
-
-  defp spawn_requires([], waiting, callback, schedulers, result) do
-    wait_for_messages([], waiting, callback, schedulers, result)
-  end
-
-  defp spawn_requires(files, waiting, callback, schedulers, result) when length(waiting) >= schedulers do
-    wait_for_messages(files, waiting, callback, schedulers, result)
-  end
-
-  defp spawn_requires([h|t], waiting, callback, schedulers, result) do
-    parent = self
-
-    compiler_pid    = :erlang.get(:elixir_compiler_pid)
-    ensure_compiled = :erlang.get(:elixir_ensure_compiled)
-    { :error_handler, handler } = :erlang.process_info(parent, :error_handler)
-
-    child = spawn_link fn ->
-      if compiler_pid != :undefined do
-        :erlang.put(:elixir_compiler_pid, compiler_pid)
-      end
-
-      if ensure_compiled != :undefined do
-        :erlang.put(:elixir_ensure_compiled, ensure_compiled)
-      end
-
-      :erlang.process_flag(:error_handler, handler)
-
-      try do
-        new = Code.require_file(h) || []
-        parent <- { :required, self, Enum.map(new, &elem(&1, 0)), h }
-      catch
-        kind, reason ->
-          parent <- { :failure, self, kind, reason, System.stacktrace }
-      end
-    end
-
-    spawn_requires(t, [child|waiting], callback, schedulers, result)
-  end
-
-  defp wait_for_messages(files, waiting, callback, schedulers, result) do
-    receive do
-      { :required, child, mods, file } ->
-        callback.(file)
-        spawn_requires(files, List.delete(waiting, child), callback, schedulers, mods ++ result)
-      { :failure, _child, kind, reason, stacktrace } ->
-        :erlang.raise(kind, reason, stacktrace)
-    end
-  end
-end
@@ -1,413 +0,0 @@
-# This is an optimization Elixir runs on function clauses.
-# Whenever a variables matches against a record (a tagged
-# tuple), this information is stored in order to optimize
-# record calls.
-defmodule Kernel.RecordRewriter do
-  @moduledoc false
-
-  def optimize_clause(module, clause) do
-    optimize_clause(clause, module, :orddict.new)
-  end
-
-  ## Clause
-
-  defp optimize_clause({ :clause, line, args, guards, body }, module, dict) do
-    { args, dict } = optimize_args(args, module, dict)
-    { body, dict, res } = optimize_body(body, module, dict, [])
-    { { :clause, line, args, guards, body }, dict, res }
-  end
-
-  defp optimize_args(args, module, dict) do
-    Enum.map_reduce args, dict, fn(arg, acc) ->
-      { new_arg, new_acc, _res } = optimize_expr(arg, module, acc)
-      { new_arg, new_acc }
-    end
-  end
-
-  defp optimize_body([], _module, dict, _acc) do
-    { [], dict, nil }
-  end
-
-  defp optimize_body([h], module, dict, acc) do
-    { new_expr, new_dict, new_res } = optimize_expr(h, module, dict)
-    { Enum.reverse([new_expr|acc]), new_dict, new_res }
-  end
-
-  defp optimize_body([h|t], module, dict, acc) do
-    { new_expr, new_dict, _ } = optimize_expr(h, module, dict)
-    optimize_body(t, module, new_dict, [new_expr|acc])
-  end
-
-  ## Record helpers
-
-  defp record_fields(record, record) do
-    fields      = Module.get_attribute(record, :record_fields)
-    optimizable = Module.get_attribute(record, :record_optimizable)
-    unless nil?(fields) or nil?(optimizable) do
-      { (lc { k, _ } inlist fields, do: k), optimizable }
-    end
-  end
-
-  defp record_fields(_module, record) do
-    if Code.ensure_loaded?(record) && function_exported?(record, :__record__, 1) do
-      try do
-        fields      = lc { k, _ } inlist record.__record__(:fields), do: k
-        optimizable = record.__record__(:optimizable)
-        { fields, optimizable }
-      rescue
-        [UndefinedFunctionError, FunctionClauseError] -> nil
-      end
-    end
-  end
-
-  defp record_field_info(function) do
-    case atom_to_list(function) do
-      'update' -> nil
-      'update_' ++ _field -> nil
-      _ -> { :accessor, function }
-    end
-  end
-
-  defp optimize_call(line, module, { record, _ } = res, left, { :atom, _, function }, args) do
-    case record_fields(module, record) do
-      { fields, optimizable } ->
-        opt_call =
-          if :lists.member({ function, length(args) + 1 }, optimizable) do
-            case record_field_info(function) do
-              { kind, field } ->
-                if index = Enum.find_index(fields, &(field == &1)) do
-                  optimize_record_accessor_call(line, res, kind, field, index, left, args)
-                end
-              nil ->
-                optimize_record_other_call(line, record, res, function, left, args)
-            end
-          end
-
-        opt_call || optimize_record_other_call(line, record, nil, function, left, args)
-      nil ->
-        nil
-    end
-  end
-
-  defp optimize_call(_line, _module, _res, _left, _right, _args) do
-    nil
-  end
-
-  defp optimize_record_accessor_call(line, _res, :accessor, _field, index, left, []) do
-    call = { :call, line,
-      { :remote, line, { :atom, 0, :erlang }, { :atom, 0, :element } },
-      [{ :integer, 0, index + 2 }, left]
-    }
-    { call, nil }
-  end
-
-  defp optimize_record_accessor_call(line, res, :accessor, _field, index, left, [arg]) do
-    call = { :call, line,
-      { :remote, line, { :atom, 0, :erlang }, { :atom, 0, :setelement } },
-      [{ :integer, 0, index + 2 }, left, arg]
-    }
-    { call, res }
-  end
-
-  defp optimize_record_other_call(line, record, res, function, left, args) do
-    call = { :call, line,
-      { :remote, line, { :atom, line, record }, { :atom, 0, function } },
-      args ++ [left]
-    }
-    { call, res }
-  end
-
-  ## Expr
-
-  defp optimize_expr({ :call, call_line,
-      { :remote, _, { :atom, _, :erlang }, { :atom, _, :setelement } } = remote,
-      [{ :integer, _, pos } = integer, tuple, value] }, module, dict) when pos > 1 do
-
-    { tuple, dict, res } = optimize_expr(tuple, module, dict)
-    { value, dict, _ }   = optimize_expr(value, module, dict)
-    { { :call, call_line, remote, [integer, tuple, value] }, dict, res }
-  end
-
-  defp optimize_expr({ :call, call_line, { :remote, line, left, right }, args } = call, module, dict) do
-    { left, dict, res } = optimize_expr(left, module, dict)
-    { right, dict, _ } = optimize_expr(right, module, dict)
-    { args, dict } = optimize_args(args, module, dict)
-
-    case optimize_call(call_line, module, res, left, right, args) do
-      { call, call_res } ->
-        { call, dict, call_res }
-      nil ->
-        { { :call, call_line, { :remote, line, left, right }, args }, dict, nil }
-    end
-  end
-
-  defp optimize_expr({ :call, line, expr, args }, module, dict) do
-    { expr, dict, _ } = optimize_expr(expr, module, dict)
-    { args, dict } = optimize_args(args, module, dict)
-    { { :call, line, expr, args }, dict, nil }
-  end
-
-  defp optimize_expr({ :match, line, left, right }, module, dict) do
-    { left,  dict, left_res }  = optimize_expr(left, module, dict)
-    { right, dict, right_res } = optimize_expr(right, module, dict)
-
-    match = { :match, line, left, right }
-
-    if left_res do
-      dict = assign_vars(extract_vars(right, []), dict, left_res)
-    end
-
-    if right_res do
-      dict = assign_vars(extract_vars(left, []), dict, right_res)
-    end
-
-    { match, dict, right_res || left_res }
-  end
-
-  defp optimize_expr({ :op, line, op, left, right }, module, dict) do
-    { left, dict, _ }  = optimize_expr(left, module, dict)
-    { right, dict, _ } = optimize_expr(right, module, dict)
-    { { :op, line, op, left, right }, dict, nil }
-  end
-
-  defp optimize_expr({ :op, line, op, expr }, module, dict) do
-    { expr, dict, _ } = optimize_expr(expr, module, dict)
-    { { :op, line, op, expr }, dict, nil }
-  end
-
-  defp optimize_expr({ :bin, line, elements }, module, dict) do
-    { elements, dict } = optimize_args(elements, module, dict)
-    { { :bin, line, elements }, dict, nil }
-  end
-
-  defp optimize_expr({ :bin_element, line, expr, type1, type2 }, module, dict) do
-    { expr, dict, _ } = optimize_expr(expr, module, dict)
-    { { :bin_element, line, expr, type1, type2 }, dict, nil }
-  end
-
-  defp optimize_expr({ :cons, line, left, right }, module, dict) do
-    { left, dict, _ }  = optimize_expr(left, module, dict)
-    { right, dict, _ } = optimize_expr(right, module, dict)
-    { { :cons, line, left, right }, dict, nil }
-  end
-
-  defp optimize_expr({ :block, line, args }, module, dict) do
-    { args, dict, res } = optimize_body(args, module, dict, [])
-    { { :block, line, args }, dict, res }
-  end
-
-  defp optimize_expr({ :tuple, line, args }, module, dict) do
-    { args, dict, args_res } = optimize_tuple_args(args, module, dict)
-    args_res = if Enum.any?(args_res), do: args_res, else: nil
-
-    res =
-      case args do
-        [{ :atom, _, atom }|_] -> atom
-        _ -> nil
-      end
-
-    { { :tuple, line, args }, dict, { res, args_res } }
-  end
-
-  defp optimize_expr({ :var, _, name } = var, _module, dict) do
-    case :orddict.find(name, dict) do
-      { :ok, res } -> { var, dict, res }
-      :error       -> { var, dict, nil }
-    end
-  end
-
-  defp optimize_expr({ :case, line, expr, clauses }, module, dict) do
-    { expr, dict, _ } = optimize_expr(expr, module, dict)
-    tuples  = lc clause inlist clauses, do: optimize_clause(clause, module, dict)
-    clauses = lc { clause, _, _ } inlist tuples, do: clause
-    dict    = join_dict(tuples)
-    res     = join_result(tuples)
-    { { :case, line, expr, clauses }, dict, res }
-  end
-
-  defp optimize_expr({ :receive, line, clauses }, module, dict) do
-    tuples  = lc clause inlist clauses, do: optimize_clause(clause, module, dict)
-    clauses = lc { clause, _, _ } inlist tuples, do: clause
-    dict    = join_dict(tuples)
-    res     = join_result(tuples)
-    { { :receive, line, clauses }, dict, res }
-  end
-
-  defp optimize_expr({ :receive, line, clauses, after_key, after_value }, module, dict) do
-    tuples  = lc clause inlist clauses, do: optimize_clause(clause, module, dict)
-    clauses = lc { clause, _, _ } inlist tuples, do: clause
-
-    { after_key, dict, _ } = optimize_expr(after_key, module, dict)
-    { after_value, dict, res } = optimize_body(after_value, module, dict, [])
-
-    dict = join_dict(tuples, dict)
-    res  = join_result(tuples, res)
-
-    { { :receive, line, clauses, after_key, after_value }, dict, res }
-  end
-
-  defp optimize_expr({ :try, line, body, elses, catches, try_after }, module, dict) do
-    tuples  = lc clause inlist catches, do: optimize_clause(clause, module, dict)
-    catches = lc { clause, _, _ } inlist tuples, do: clause
-
-    tuples  = lc clause inlist elses, do: optimize_clause(clause, module, dict)
-    elses   = lc { clause, _, _ } inlist tuples, do: clause
-
-    { body, _, res } = optimize_body(body, module, dict, [])
-    res = join_result(tuples, res)
-
-    { try_after, _, _ } = optimize_body(try_after, module, dict, [])
-    { { :try, line, body, elses, catches, try_after }, dict, res }
-  end
-
-  defp optimize_expr({ :fun, line, { :function, receiver, name, arity } }, module, dict) do
-    { receiver, dict, _ } = optimize_expr(receiver, module, dict)
-    { name, dict, _ }     = optimize_expr(name, module, dict)
-    { arity, dict, _ }    = optimize_expr(arity, module, dict)
-    { { :fun, line, { :function, receiver, name, arity } }, dict, nil }
-  end
-
-  defp optimize_expr({ :fun, line, { :clauses, clauses } }, module, dict) do
-    clauses = lc clause inlist clauses do
-      { clause, _, _ } = optimize_clause(clause, module, dict)
-      clause
-    end
-
-    { { :fun, line, { :clauses, clauses } }, dict, nil }
-  end
-
-  defp optimize_expr({ comprehension, line, expr, args }, module, dict) when comprehension in [:lc, :bc] do
-    { args, new_dict } = optimize_args(args, module, dict)
-    { expr, _, _ } = optimize_expr(expr, module, new_dict)
-    { { comprehension, line, expr, args }, dict, nil }
-  end
-
-  defp optimize_expr({ generate, line, left, right }, module, dict) when generate in [:generate, :b_generate] do
-    { left, dict, _ }  = optimize_expr(left, module, dict)
-    { right, dict, _ } = optimize_expr(right, module, dict)
-    { { generate, line, left, right }, dict, nil }
-  end
-
-  defp optimize_expr(other, _module, dict) when elem(other, 0) in [:string, :atom, :integer, :float, :nil, :fun] do
-    { other, dict, nil }
-  end
-
-  ## Helpers
-
-  defp optimize_tuple_args(args, module, dict) do
-    { final_args, { final_dict, final_acc } } =
-      Enum.map_reduce args, { dict, [] }, fn(arg, { acc_dict, acc_res }) ->
-        { new_arg, new_acc, res } = optimize_expr(arg, module, acc_dict)
-        { new_arg, { new_acc, [res|acc_res] } }
-      end
-
-    { final_args, final_dict, Enum.reverse(final_acc) }
-  end
-
-  defp assign_vars([key|t], dict, { _, value } = res) when is_list(key) and is_list(value) and length(key) == length(value) do
-    assign_vars t, assign_nested_vars(key, dict, value), res
-  end
-
-  defp assign_vars([key|t], dict, { value, _ } = res) when is_atom(key) and value != nil do
-    dict =
-      case :orddict.find(key, dict) do
-        { :ok, ^res } ->
-          dict
-        { :ok, { ^value, _ } } ->
-          :orddict.store(key, { value, nil }, dict)
-        { :ok, _ } ->
-          # We are overriding a type of an existing variable,
-          # which means the source code is invalid.
-          :orddict.store(key, nil, dict)
-        :error ->
-          :orddict.store(key, res, dict)
-      end
-
-    assign_vars t, dict, res
-  end
-
-  defp assign_vars([_|t], dict, res) do
-    assign_vars t, dict, res
-  end
-
-  defp assign_vars([], dict, _res) do
-    dict
-  end
-
-  defp assign_nested_vars([vars|vt], dict, [res|rt]) do
-    assign_nested_vars(vt, assign_vars(vars, dict, res), rt)
-  end
-
-  defp assign_nested_vars([], dict, []) do
-    dict
-  end
-
-  defp extract_vars({ :match, _, left, right }, vars) do
-    vars = extract_vars(right, vars)
-    extract_vars(left, vars)
-  end
-
-  defp extract_vars({ :var, _, name }, vars) do
-    [name|vars]
-  end
-
-  defp extract_vars({ :tuple, _, args }, vars) do
-    [Enum.map(args, &extract_vars(&1, []))|vars]
-  end
-
-  defp extract_vars(_, vars) do
-    vars
-  end
-
-  defp join_dict([]) do
-    []
-  end
-
-  defp join_dict([{ _, dict, _ }|t]) do
-    join_dict(t, dict)
-  end
-
-  defp join_dict([{ _, dict, _ }|t], other) do
-    other = Enum.reduce other, other, fn
-      { key, { value, _ } = res }, acc ->
-        case :orddict.find(key, dict) do
-          { :ok, ^res } -> acc
-          { :ok, { ^value, _ } } -> :orddict.store(key, { value, nil }, acc)
-          { :ok, _ } -> :orddict.store(key, nil, acc)
-          :error -> :orddict.erase(key, acc)
-        end
-      { key, nil }, acc ->
-        :orddict.store(key, nil, acc)
-    end
-
-    join_dict(t, other)
-  end
-
-  defp join_dict([], other) do
-    other
-  end
-
-  defp join_result([]) do
-    []
-  end
-
-  defp join_result([{ _, _, res }|t]) do
-    join_result(t, res)
-  end
-
-  defp join_result([{ _, _, res }|t], res) do
-    join_result(t, res)
-  end
-
-  defp join_result([{ _, _, { res, _ } }|t], { res, _ }) do
-    join_result(t, { res, nil })
-  end
-
-  defp join_result([{ _, _, _ }|_], _res) do
-    nil
-  end
-
-  defp join_result([], res) do
-    res
-  end
-end
@@ -1,783 +0,0 @@
-defmodule Kernel.Typespec do
-  @moduledoc """
-  Provides macros and functions for working with typespecs.
-
-  The attributes `@type`, `@opaque`, `@typep`, `@spec` and
-  `@callback` available in modules are handled by the equivalent
-  macros defined by this module.
-
-  ## Defining a type
-
-      @type type_name :: type
-      @typep type_name :: type
-      @opaque type_name :: type
-
-  For more details, see documentation for `deftype`, `deftypep` and `defopaque`
-  below.
-
-  ## Defining a specification
-
-      @spec function_name(type, type) :: type
-      @callback function_name(type, type) :: type
-
-  For more details, see documentation for `defspec` and `defcallback` below.
-
-  ## Types
-
-  The type syntax provided by Elixir is fairly similar to the one
-  in Erlang.
-
-  Most of the built-in types provided in Erlang (for example, `pid()`)
-  are expressed the same way: `pid()` or simply `pid`. Parametrized types
-  are also supported (`list(integer())`) and so are remote types (`Enum.t`).
-
-  Certain data type shortcuts (`[...]`, `<<>>` and `{...}`) are supported as
-  well.
-
-  Main differences lie in how bit strings and functions are defined:
-
-  ### Bit Strings
-
-  Bit string with a base size of 3:
-
-      <<_ :: 3>>
-
-  Bit string with a unit size of 8:
-
-      <<_ :: _ * 8>>
-
-  ### Anonymous functions
-
-  Any anonymous function:
-
-      ((...) -> any)
-      or
-      (... -> any)
-
-  Anonymous function with arity of zero:
-
-      (() -> type)
-
-  Anonymous function with some arity:
-
-      ((type, type) -> type)
-      or
-      (type, type -> type)
-
-  ## Notes
-
-  Elixir discourages the use of type `string()` as it might be confused
-  with binaries which are referred to as "strings" in Elixir (as opposed to
-  character lists). In order to use the type that is called `string()` in Erlang,
-  one has to use the `char_list()` type which is a synonym for `string()`. If you
-  use `string()`, you'll get a warning from the compiler.
-
-  If you want to refer to the "string" type (the one operated by functions in the
-  String module), use `String.t()` type instead.
-
-  See http://www.erlang.org/doc/reference_manual/typespec.html
-  for more information.
-  """
-
-  @doc """
-  Defines a type.
-  This macro is the one responsible for handling the attribute `@type`.
-
-  ## Examples
-
-      @type my_type :: atom
-
-  """
-  defmacro deftype(type) do
-    quote do
-      Kernel.Typespec.deftype(:type, unquote(Macro.escape type), __ENV__)
-    end
-  end
-
-  @doc """
-  Defines an opaque type.
-  This macro is the one responsible for handling the attribute `@opaque`.
-
-  ## Examples
-
-      @opaque my_type :: atom
-
-  """
-  defmacro defopaque(type) do
-    quote do
-      Kernel.Typespec.deftype(:opaque, unquote(Macro.escape type), __ENV__)
-    end
-  end
-
-  @doc """
-  Defines a private type.
-  This macro is the one responsible for handling the attribute `@typep`.
-
-  ## Examples
-
-      @typep my_type :: atom
-
-  """
-  defmacro deftypep(type) do
-    quote do
-      Kernel.Typespec.deftype(:typep, unquote(Macro.escape type), __ENV__)
-    end
-  end
-
-  @doc """
-  Defines a spec.
-  This macro is the one responsible for handling the attribute `@spec`.
-
-  ## Examples
-
-      @spec add(number, number) :: number
-
-  """
-  defmacro defspec(spec) do
-    quote do
-      Kernel.Typespec.defspec(:spec, unquote(Macro.escape spec), __ENV__)
-    end
-  end
-
-  @doc """
-  Defines a callback.
-  This macro is the one responsible for handling the attribute `@callback`.
-
-  ## Examples
-
-      @callback add(number, number) :: number
-
-  """
-  defmacro defcallback(spec) do
-    quote do
-      Kernel.Typespec.defspec(:callback, unquote(Macro.escape spec), __ENV__)
-    end
-  end
-
-  ## Helpers
-
-  @doc """
-  Defines a `type`, `typep` or `opaque` by receiving Erlang's typespec.
-  """
-  def define_type(caller, kind, { name, _, vars } = type) when kind in [:type, :typep, :opaque] do
-    { kind, export } =
-      case kind do
-        :type   -> { :type, true }
-        :typep  -> { :type, false }
-        :opaque -> { :opaque, true }
-      end
-
-    module = caller.module
-    arity  = length(vars)
-
-    Module.compile_typespec module, kind, type
-
-    if export do
-      Module.compile_typespec(module, :export_type, [{ name, arity }])
-    end
-
-    define_doc(caller, kind, name, arity, export)
-    type
-  end
-
-  defp define_doc(caller, kind, name, arity, export) do
-    module = caller.module
-    doc    = Module.get_attribute(module, :typedoc)
-
-    if doc do
-      if export do
-        Module.add_doc(module, caller.line, kind, { name, arity }, doc)
-      else
-        :elixir_errors.warn "#{caller.file}:#{caller.line}: type #{name}/#{arity} is private, " <>
-                            "@typedoc's are always discarded for private types\n"
-      end
-    end
-
-    Module.delete_attribute(module, :typedoc)
-  end
-
-  @doc """
-  Defines a `spec` by receiving Erlang's typespec.
-  """
-  def define_spec(module, tuple, definition) do
-    Module.compile_typespec module, :spec, { tuple, definition }
-  end
-
-  @doc """
-  Defines a `callback` by receiving Erlang's typespec.
-  """
-  def define_callback(module, tuple, definition) do
-    Module.compile_typespec module, :callback, { tuple, definition }
-  end
-
-  @doc """
-  Returns `true` if the current module defines a given type
-  (private, opaque or not). This function is only available
-  for modules being compiled.
-  """
-  def defines_type?(module, name, arity) do
-    finder = &match?({ ^name, _, vars } when length(vars) == arity, &1)
-    :lists.any(finder, Module.get_attribute(module, :type)) or
-      :lists.any(finder, Module.get_attribute(module, :opaque))
-  end
-
-  @doc """
-  Returns `true` if the current module defines a given spec.
-  This function is only available for modules being compiled.
-  """
-  def defines_spec?(module, name, arity) do
-    tuple = { name, arity }
-    :lists.any(&match?(^tuple, &1), Module.get_attribute(module, :spec))
-  end
-
-  @doc """
-  Returns `true` if the current module defines a callback.
-  This function is only available for modules being compiled.
-  """
-  def defines_callback?(module, name, arity) do
-    tuple = { name, arity }
-    :lists.any(&match?(^tuple, &1), Module.get_attribute(module, :callback))
-  end
-
-  @doc """
-  Converts a spec clause back to Elixir AST.
-  """
-  def spec_to_ast(name, { :type, line, :fun, [{:type, _, :product, args}, result] }) do
-    args = lc arg inlist args, do: typespec_to_ast(arg)
-    { :::, [line: line], [{ name, [line: line], args }, typespec_to_ast(result)] }
-  end
-
-  def spec_to_ast(name, { :type, line, :fun, [] }) do
-    { :::, [line: line], [{ name, [line: line], [] }, quote(do: term)] }
-  end
-
-  def spec_to_ast(name, { :type, line, :bounded_fun, [{ :type, _, :fun, [{ :type, _, :product, args }, result] }, constraints] }) do
-    [h|t] =
-      lc {:type, line, :constraint, [{:atom, _, :is_subtype}, [var, type]]} inlist constraints do
-        { :is_subtype, [line: line], [typespec_to_ast(var), typespec_to_ast(type)] }
-      end
-
-    args = lc arg inlist args, do: typespec_to_ast(arg)
-    guards = Enum.reduce t, h, fn(x, acc) -> { :and, line, [acc, x] } end
-
-    { :::, [line: line], [{ :when, [line: line], [{ name, [line: line], args }, guards] }, typespec_to_ast(result)] }
-  end
-
-  @doc """
-  Converts a type clause back to Elixir AST.
-  """
-  def type_to_ast({ { :record, record }, fields, args }) when is_atom(record) do
-    fields = lc field inlist fields, do: typespec_to_ast(field)
-    args = lc arg inlist args, do: typespec_to_ast(arg)
-    type = { :{}, [], [record|fields] }
-    quote do: unquote(record)(unquote_splicing(args)) :: unquote(type)
-  end
-
-  def type_to_ast({ name, type, args }) do
-    args = lc arg inlist args, do: typespec_to_ast(arg)
-    quote do: unquote(name)(unquote_splicing(args)) :: unquote(typespec_to_ast(type))
-  end
-
-  @doc """
-  Returns all type docs available from the module's beam code.
-
-  It is returned as a list of tuples where the first element is the pair of type
-  name and arity and the second element is the documentation.
-
-  The module has to have a corresponding beam file on the disk which can be
-  located by the runtime system.
-  """
-  @spec beam_typedocs(module | binary) :: [tuple] | nil
-  def beam_typedocs(module) when is_atom(module) or is_binary(module) do
-    case abstract_code(module) do
-      { :ok, abstract_code } ->
-        type_docs = lc { :attribute, _, :typedoc, tup } inlist abstract_code, do: tup
-        List.flatten(type_docs)
-      _ ->
-        nil
-    end
-  end
-
-  @doc """
-  Returns all types available from the module's beam code.
-
-  It is returned as a list of tuples where the first
-  element is the type (`:typep`, `:type` and `:opaque`).
-
-  The module has to have a corresponding beam file on the disk which can be
-  located by the runtime system.
-  """
-  @spec beam_types(module | binary) :: [tuple] | nil
-  def beam_types(module) when is_atom(module) or is_binary(module) do
-    case abstract_code(module) do
-      { :ok, abstract_code } ->
-        exported_types = lc { :attribute, _, :export_type, types } inlist abstract_code, do: types
-        exported_types = List.flatten(exported_types)
-
-        lc { :attribute, _, kind, { name, _, args } = type } inlist abstract_code, kind in [:opaque, :type] do
-          cond do
-            kind == :opaque -> { :opaque, type }
-            :lists.member({ name, length(args) }, exported_types) -> { :type, type }
-            true -> { :typep, type }
-          end
-        end
-      _ ->
-        nil
-    end
-  end
-
-  @doc """
-  Returns all specs available from the module's beam code.
-
-  It is returned as a list of tuples where the first
-  element is spec name and arity and the second is the spec.
-
-  The module has to have a corresponding beam file on the disk which can be
-  located by the runtime system.
-  """
-  @spec beam_specs(module | binary) :: [tuple] | nil
-  def beam_specs(module) when is_atom(module) or is_binary(module) do
-    from_abstract_code(module, :spec)
-  end
-
-  @doc """
-  Returns all callbacks available from the module's beam code.
-
-  It is returned as a list of tuples where the first
-  element is spec name and arity and the second is the spec.
-
-  The module has to have a corresponding beam file on the disk
-  which can be located by the runtime system.
-  """
-  @spec beam_callbacks(module | binary) :: [tuple] | nil
-  def beam_callbacks(module) when is_atom(module) or is_binary(module) do
-    from_abstract_code(module, :callback)
-  end
-
-  defp from_abstract_code(module, kind) do
-    case abstract_code(module) do
-      { :ok, abstract_code } ->
-        lc { :attribute, _, abs_kind, value } inlist abstract_code, kind == abs_kind, do: value
-      :error ->
-        nil
-    end
-  end
-
-  defp abstract_code(module) do
-    case :beam_lib.chunks(abstract_code_beam(module), [:abstract_code]) do
-      {:ok, { _, [{ :abstract_code, { _raw_abstract_v1, abstract_code } }] } } ->
-        { :ok, abstract_code }
-      _ ->
-        :error
-    end
-  end
-
-  defp abstract_code_beam(module) when is_atom(module) do
-    case :code.get_object_code(module) do
-      { ^module, beam, _filename } -> beam
-      :error -> module
-    end
-  end
-
-  defp abstract_code_beam(binary) when is_binary(binary) do
-    binary
-  end
-
-  ## Macro callbacks
-
-  @doc false
-  def deftype(kind, { :::, _, [type, definition] }, caller) do
-    do_deftype(kind, type, definition, caller)
-  end
-
-  def deftype(kind, {name, _meta, args} = type, caller)
-      when is_atom(name) and not is_list(args) do
-    do_deftype(kind, type, { :term, [line: caller.line], nil }, caller)
-  end
-
-  def deftype(_kind, other, caller) do
-    type_spec = Macro.to_string(other)
-    compile_error caller, "invalid type specification #{type_spec}"
-  end
-
-  defp do_deftype(kind, { name, _, args }, definition, caller) do
-    args =
-      if is_atom(args) do
-        []
-      else
-        lc(arg inlist args, do: variable(arg))
-      end
-
-    vars = lc { :var, _, var } inlist args, do: var
-    spec = typespec(definition, vars, caller)
-
-    vars = lc { :var, _, _ } = var inlist args, do: var
-    type = { name, spec, vars }
-
-    define_type(caller, kind, type)
-  end
-
-  @doc false
-  def defspec(type, {:::, _, [{ :when, _, [{ name, meta, args }, constraints_guard] }, return] }, caller) do
-    if is_atom(args), do: args = []
-    constraints = guard_to_constraints(constraints_guard, caller)
-    spec = { :type, line(meta), :fun, fn_args(meta, args, return, Keyword.keys(constraints), caller) }
-    spec = { :type, line(meta), :bounded_fun, [spec, Keyword.values(constraints)] }
-    code = { { name, Kernel.length(args) }, spec }
-    Module.compile_typespec(caller.module, type, code)
-    code
-  end
-
-  def defspec(type, {:::, _, [{ name, meta, args }, return]}, caller) do
-    if is_atom(args), do: args = []
-    spec = { :type, line(meta), :fun, fn_args(meta, args, return, [], caller) }
-    code = { { name, Kernel.length(args) }, spec }
-    Module.compile_typespec(caller.module, type, code)
-    code
-  end
-
-  def defspec(_type, other, caller) do
-    spec = Macro.to_string(other)
-    compile_error caller, "invalid function type specification #{spec}"
-  end
-
-  defp guard_to_constraints({ :is_subtype, meta, [{ name, _, _ }, type] }, caller) do
-    line = line(meta)
-    contraints = [{ :atom, line, :is_subtype }, [{:var, line, name}, typespec(type, [], caller)]]
-    [{ name, { :type, line, :constraint, contraints } }]
-  end
-
-  defp guard_to_constraints({ :and, _, [left, right] }, caller) do
-    guard_to_constraints(left, caller) ++ guard_to_constraints(right, caller)
-  end
-
-  ## To AST conversion
-
-  defp typespec_to_ast({ :type, line, :tuple, :any }) do
-    typespec_to_ast({:type, line, :tuple, []})
-  end
-
-  defp typespec_to_ast({ :type, line, :tuple, args }) do
-    args = lc arg inlist args, do: typespec_to_ast(arg)
-    { :{}, [line: line], args }
-  end
-
-  defp typespec_to_ast({ :type, _line, :list, [arg] }) do
-    case unpack_typespec_kw(arg, []) do
-      { :ok, ast } -> ast
-      :error -> [typespec_to_ast(arg)]
-    end
-  end
-
-  defp typespec_to_ast({ :type, _line, :list, args }) do
-    lc arg inlist args, do: typespec_to_ast(arg)
-  end
-
-  defp typespec_to_ast({ :type, line, :binary, [arg1, arg2] }) do
-    [arg1, arg2] = lc arg inlist [arg1, arg2], do: typespec_to_ast(arg)
-    cond do
-      arg2 == 0 ->
-        quote line: line, do: <<_ :: unquote(arg1)>>
-      arg1 == 0 ->
-        quote line: line, do: <<_ :: _ * unquote(arg2)>>
-      true ->
-        quote line: line, do: <<_ :: unquote(arg1) * unquote(arg2)>>
-    end
-  end
-
-  defp typespec_to_ast({ :type, line, :union, args }) do
-    args = lc arg inlist args, do: typespec_to_ast(arg)
-    Enum.reduce tl(args), hd(args),
-      fn(arg, expr) -> { :|, [line: line], [expr, arg] } end
-  end
-
-  defp typespec_to_ast({ :type, line, :fun, [{:type, _, :product, args}, result] }) do
-    args = lc arg inlist args, do: typespec_to_ast(arg)
-    { :->, [line: line], [{args, [line: line], typespec_to_ast(result)}] }
-  end
-
-  defp typespec_to_ast({ :type, line, :fun, [args, result] }) do
-    { :->, [line: line], [{[typespec_to_ast(args)], [line: line], typespec_to_ast(result)}] }
-  end
-
-  defp typespec_to_ast({ :type, line, :fun, [] }) do
-    typespec_to_ast({ :type, line, :fun, [{:type, line, :any}, {:type, line, :any, []} ] })
-  end
-
-  defp typespec_to_ast({ :type, line, :range, [left, right] }) do
-    { :"..", [line: line], [typespec_to_ast(left), typespec_to_ast(right)] }
-  end
-
-  defp typespec_to_ast({ :type, line, name, args }) do
-    args = lc arg inlist args, do: typespec_to_ast(arg)
-    { name, [line: line], args }
-  end
-
-  defp typespec_to_ast({ :var, line, var }) do
-    var =
-      case atom_to_binary(var) do
-        <<"_", c :: [binary, size(1)], rest :: binary>> ->
-          binary_to_atom("_#{String.downcase(c)}#{rest}")
-        <<c :: [binary, size(1)], rest :: binary>> ->
-          binary_to_atom("#{String.downcase(c)}#{rest}")
-      end
-    { var, line, nil }
-  end
-
-  # Special shortcut(s)
-  defp typespec_to_ast({ :remote_type, line, [{:atom, _, :elixir}, {:atom, _, :char_list}, []] }) do
-    typespec_to_ast({:type, line, :char_list, []})
-  end
-
-  defp typespec_to_ast({ :remote_type, line, [{:atom, _, :elixir}, {:atom, _, :as_boolean}, [arg]] }) do
-    typespec_to_ast({:type, line, :as_boolean, [arg]})
-  end
-
-  defp typespec_to_ast({ :remote_type, line, [mod, name, args] }) do
-    args = lc arg inlist args, do: typespec_to_ast(arg)
-    dot  = { :., [line: line], [typespec_to_ast(mod), typespec_to_ast(name)] }
-    { dot, [line: line], args }
-  end
-
-  defp typespec_to_ast({ :ann_type, line, [var, type] }) do
-    { :::, [line: line], [typespec_to_ast(var), typespec_to_ast(type)] }
-  end
-
-  defp typespec_to_ast({ :typed_record_field,
-                         { :record_field, line, { :atom, line1, name }},
-                         type }) do
-    typespec_to_ast({ :ann_type, line, [{ :var, line1, name }, type] })
-  end
-
-  defp typespec_to_ast({:type, _, :any}) do
-    quote do: ...
-  end
-
-  defp typespec_to_ast({:paren_type, _, [type]}) do
-    typespec_to_ast(type)
-  end
-
-  defp typespec_to_ast({ t, _line, atom }) when is_atom(t) do
-    atom
-  end
-
-  defp typespec_to_ast(other), do: other
-
-  ## From AST conversion
-
-  defp line(meta) do
-    case :lists.keyfind(:line, 1, meta) do
-      { :line, line } -> line
-      false -> 0
-    end
-  end
-
-  # Handle unions
-  defp typespec({ :|, meta, [_, _] } = exprs, vars, caller) do
-    exprs = Enum.reverse(collect_union(exprs))
-    union = lc e inlist exprs, do: typespec(e, vars, caller)
-    { :type, line(meta), :union, union }
-  end
-
-  # Handle binaries
-  defp typespec({:<<>>, meta, []}, _, _) do
-    {:type, line(meta), :binary, [{:integer, line(meta), 0}, {:integer, line(meta), 0}]}
-  end
-
-  defp typespec({:<<>>, meta, [{:::, _, [{:_, meta1, atom}, {:*, _, [{:_, meta2, atom}, unit]}]}]}, _, _) when is_atom(atom) do
-    {:type, line(meta), :binary, [{:integer, line(meta1), 0}, {:integer, line(meta2), unit}]}
-  end
-
-  defp typespec({:<<>>, meta, [{:::, meta1, [{:_, meta2, atom}, base]}]}, _, _) when is_atom(atom) do
-    {:type, line(meta), :binary, [{:integer, line(meta1), base}, {:integer, line(meta2), 0}]}
-  end
-
-  # Handle ranges
-  defp typespec({:"..", meta, args}, vars, caller) do
-    typespec({:range, meta, args}, vars, caller)
-  end
-
-  # Handle special forms
-  defp typespec({:__MODULE__, _, atom}, vars, caller) when is_atom(atom) do
-    typespec(caller.module, vars, caller)
-  end
-
-  defp typespec({:__aliases__, _, _} = alias, vars, caller) do
-    atom = Macro.expand alias, caller
-    typespec(atom, vars, caller)
-  end
-
-  # Handle funs
-  defp typespec({:->, meta, [{[{:fun, _, arguments}], cmeta, return}]}, vars, caller) when is_list(arguments) do
-    typespec({:->, meta, [{arguments, cmeta, return}]}, vars, caller)
-  end
-
-  defp typespec({:->, meta, [{arguments, _, return}]}, vars, caller) when is_list(arguments) do
-    args = fn_args(meta, arguments, return, vars, caller)
-    { :type, line(meta), :fun, args }
-  end
-
-  # Handle type operator
-  defp typespec({:"::", meta, [var, expr] }, vars, caller) do
-    left  = typespec(var, [elem(var, 0)|vars], caller)
-    right = typespec(expr, vars, caller)
-    { :ann_type, line(meta), [left, right] }
-  end
-
-  # Handle unary ops
-  defp typespec({op, meta, [integer]}, _, _) when op in [:+, :-] and is_integer(integer) do
-    { :op, line(meta), op, {:integer, line(meta), integer} }
-  end
-
-  # Handle access macro
-  defp typespec({{:., meta, [Kernel, :access]}, meta1, [target, args]}, vars, caller) do
-    access = {{:., meta, [Kernel, :access]}, meta1,
-              [target, args ++ [_: { :any, [], [] }]]}
-    typespec(Macro.expand(access, caller), vars, caller)
-  end
-
-  # Handle remote calls
-  defp typespec({{:., meta, [remote, name]}, _, args} = orig, vars, caller) do
-    remote = Macro.expand remote, caller
-    unless is_atom(remote) do
-      compile_error(caller, "invalid remote in typespec: #{Macro.to_string(orig)}")
-    end
-    remote_type({typespec(remote, vars, caller), meta, typespec(name, vars, caller), args}, vars, caller)
-  end
-
-  # Handle tuples
-  defp typespec({:tuple, meta, atom}, vars, caller) when is_atom(atom) do
-    typespec({:{}, meta, []}, vars, caller)
-  end
-
-  defp typespec({:{}, meta, []}, _, _) do
-    { :type, line(meta), :tuple, :any }
-  end
-
-  defp typespec({:{}, meta, t}, vars, caller) when is_list(t) do
-    args = lc e inlist t, do: typespec(e, vars, caller)
-    { :type, line(meta), :tuple, args }
-  end
-
-  # Handle blocks
-  defp typespec({:__block__, _meta, [arg]}, vars, caller) do
-    typespec(arg, vars, caller)
-  end
-
-  # Handle variables or local calls
-  defp typespec({name, meta, atom}, vars, caller) when is_atom(atom) do
-    if :lists.member(name, vars) do
-      { :var, line(meta), name }
-    else
-      typespec({name, meta, []}, vars, caller)
-    end
-  end
-
-  # Handle local calls
-  defp typespec({:string, meta, arguments}, vars, caller) do
-    IO.write "warning: string() type use is discouraged. For character lists, use " <>
-      "char_list() type, for strings, String.t()\n#{Exception.format_stacktrace(caller.stacktrace)}"
-    arguments = lc arg inlist arguments, do: typespec(arg, vars, caller)
-    { :type, line(meta), :string, arguments }
-  end
-
-  defp typespec({:char_list, _meta, arguments}, vars, caller) do
-    typespec((quote do: :elixir.char_list(unquote_splicing(arguments))), vars, caller)
-  end
-
-  defp typespec({:as_boolean, _meta, arguments}, vars, caller) do
-    typespec((quote do: :elixir.as_boolean(unquote_splicing(arguments))), vars, caller)
-  end
-
-  defp typespec({name, meta, arguments}, vars, caller) do
-    arguments = lc arg inlist arguments, do: typespec(arg, vars, caller)
-    { :type, line(meta), name, arguments }
-  end
-
-  # Handle literals
-  defp typespec(atom, _, _) when is_atom(atom) do
-    { :atom, 0, atom }
-  end
-
-  defp typespec(integer, _, _) when is_integer(integer) do
-    { :integer, 0, integer }
-  end
-
-  defp typespec([], vars, caller) do
-    typespec({ nil, [], [] }, vars, caller)
-  end
-
-  defp typespec([spec], vars, caller) do
-    typespec({ :list, [], [spec] }, vars, caller)
-  end
-
-  defp typespec([spec, {:"...", _, quoted}], vars, caller) when is_atom(quoted) do
-    typespec({ :nonempty_list, [], [spec] }, vars, caller)
-  end
-
-  defp typespec([h|t] = l, vars, caller) do
-    union = Enum.reduce(t, validate_kw(h, l, caller), fn(x, acc) ->
-      { :|, [], [acc, validate_kw(x, l, caller)] }
-    end)
-    typespec({ :list, [], [union] }, vars, caller)
-  end
-
-  defp typespec(t, vars, caller) when is_tuple(t) do
-    args = lc e inlist tuple_to_list(t), do: typespec(e, vars, caller)
-    { :type, 0, :tuple, args }
-  end
-
-  ## Helpers
-
-  defp compile_error(caller, desc) do
-    raise CompileError, file: caller.file, line: caller.line, description: desc
-  end
-
-  defp remote_type({remote, meta, name, arguments}, vars, caller) do
-    arguments = lc arg inlist arguments, do: typespec(arg, vars, caller)
-    { :remote_type, line(meta), [ remote, name, arguments ] }
-  end
-
-  defp collect_union({ :|, _, [a, b] }), do: [b|collect_union(a)]
-  defp collect_union(v), do: [v]
-
-  defp validate_kw({ key, _ } = t, _, _caller) when is_atom(key), do: t
-  defp validate_kw(_, original, caller) do
-    compile_error(caller, "unexpected list #{Macro.to_string original} in typespec")
-  end
-
-  defp fn_args(meta, args, return, vars, caller) do
-    case [fn_args(meta, args, vars, caller), typespec(return, vars, caller)] do
-      [{:type, _, :any}, {:type, _, :any, []}] -> []
-      x -> x
-    end
-  end
-
-  defp fn_args(meta, [{:"...", _, _}], _vars, _caller) do
-    { :type, line(meta), :any }
-  end
-
-  defp fn_args(meta, args, vars, caller) do
-    args = lc arg inlist args, do: typespec(arg, vars, caller)
-    { :type, line(meta), :product, args }
-  end
-
-  defp variable({name, meta, _}) do
-    {:var, line(meta), name}
-  end
-
-  defp unpack_typespec_kw({ :type, _, :union, [
-         next,
-         { :type, _, :tuple, [{ :atom, _, atom }, type] }
-       ] }, acc) do
-    unpack_typespec_kw(next, [{atom, typespec_to_ast(type)}|acc])
-  end
-
-  defp unpack_typespec_kw({ :type, _, :tuple, [{ :atom, _, atom }, type] }, acc) do
-    { :ok, [{atom, typespec_to_ast(type)}|acc] }
-  end
-
-  defp unpack_typespec_kw(_, _acc) do
-    :error
-  end
-end
@@ -1,540 +0,0 @@
-defmodule Keyword do
-  @moduledoc """
-  A keyword is a list of tuples where the first element
-  of the tuple is an atom and the second element can be
-  any value.
-
-  A keyword may have duplicated keys so it is not strictly
-  a dictionary. However most of the functions in this module
-  behaves exactly as a dictionary and mimic the API defined
-  by the `Dict` behaviour.
-
-  For example, `Keyword.get` will get the first entry matching
-  the given key, regardless if duplicated entries exist.
-  Similarly, `Keyword.put` and `Keyword.delete` ensure all
-  duplicated entries for a given key are removed when invoked.
-  A handful of functions exist to handle duplicated keys, in
-  particular, `from_enum` allows creating a new keywords without
-  removing duplicated keys, `get_values` returns all values for
-  a given key and `delete_first` deletes just one of the existing
-  entries.
-
-  Since a keyword list is simply a list, all the operations defined
-  in `Enum` and `List` can also be applied.
-  """
-
-  @type key :: atom
-  @type value :: any
-  @type t :: [{key, value}]
-
-  @doc """
-  Creates a Keyword from an enum. Unlike `Keyword.new`
-  which behaves as a dict, `Keyword.from_enum` does not remove
-  duplicated entries.
-  """
-  @spec from_enum(Enum.t) :: t
-  def from_enum(enum) do
-    Enum.to_list(enum)
-  end
-
-  @doc """
-  Checks if the given argument is a keywords list or not.
-  """
-  @spec keyword?(term) :: boolean
-  def keyword?([{ key, _value } | rest]) when is_atom(key) do
-    keyword?(rest)
-  end
-
-  def keyword?([]),     do: true
-  def keyword?(_other), do: false
-
-  @doc """
-  Returns an empty keyword list, i.e. an empty list.
-  """
-  @spec new :: t
-  def new do
-    []
-  end
-
-  @doc """
-  Creates a Keyword from an enumerable. Similar to dicts,
-  duplicated entries are removed, the latest one prevails.
-
-  ## Examples
-
-      iex> Keyword.new([{:b, 1}, {:a, 2}])
-      [a: 2, b: 1]
-
-  """
-  @spec new(Enum.t) :: t
-  def new(pairs) do
-    Enum.reduce pairs, [], fn { k, v }, keywords ->
-      put(keywords, k, v)
-    end
-  end
-
-  @doc """
-  Creates a Keyword from an enumerable with the
-  help of the transformation function. Duplicated
-  entries are removed, the latest one prevails.
-
-  ## Examples
-
-      iex> Keyword.new([:a, :b], fn (x) -> {x, x} end) |> Enum.sort
-      [a: :a, b: :b]
-
-  """
-  @spec new(Enum.t, ({key, value} -> {key, value})) :: t
-  def new(pairs, transform) do
-    Enum.reduce pairs, [], fn i, keywords ->
-      { k, v } = transform.(i)
-      put(keywords, k, v)
-    end
-  end
-
-  @doc """
-  Gets the value for a specific `key`.
-
-  If `key` does not exist, return default value (`nil` if no default value).
-
-  If duplicated entries exist, the first one is returned.
-  Use `get_values/2` to retrieve all entries.
-
-  ## Examples
-
-      iex> Keyword.get([a: 1], :a)
-      1
-
-      iex> Keyword.get([a: 1], :b)
-      nil
-
-      iex> Keyword.get([a: 1], :b, 3)
-      3
-
-  """
-  @spec get(t, key) :: value
-  @spec get(t, key, value) :: value
-  def get(keywords, key, default // nil) when is_atom(key) do
-    case :lists.keyfind(key, 1, keywords) do
-      { ^key, value } -> value
-      false -> default
-    end
-  end
-
-  @doc """
-  Fetches the value for a specific `key` and returns it in a tuple.
-  If the `key` does not exist, returns `:error`.
-
-  ## Examples
-
-      iex> Keyword.fetch([a: 1], :a)
-      { :ok, 1 }
-
-      iex> Keyword.fetch([a: 1], :b)
-      :error
-
-  """
-  @spec fetch(t, key) :: value
-  def fetch(keywords, key) when is_atom(key) do
-    case :lists.keyfind(key, 1, keywords) do
-      { ^key, value } -> { :ok, value }
-      false -> :error
-    end
-  end
-
-  @doc """
-  Fetches the value for specific `key`. If `key` does not exist,
-  a `KeyError` is raised.
-
-  ## Examples
-
-      iex> Keyword.fetch!([a: 1], :a)
-      1
-
-      iex> Keyword.fetch!([a: 1], :b)
-      ** (KeyError) key not found: :b
-
-  """
-  @spec fetch!(t, key) :: value | no_return
-  def fetch!(keywords, key) when is_atom(key) do
-    case :lists.keyfind(key, 1, keywords) do
-      { ^key, value } -> value
-      false -> raise(KeyError, key: key)
-    end
-  end
-
-  @doc """
-  Gets all values for a specific `key`.
-
-  ## Examples
-
-      iex> Keyword.get_values([a: 1, a: 2], :a)
-      [1,2]
-
-  """
-  @spec get_values(t, key) :: [value]
-  def get_values(keywords, key) when is_atom(key) do
-    lc { k, v } inlist keywords, key == k, do: v
-  end
-
-  @doc """
-  Returns all keys from the keyword list. Duplicated
-  keys appear duplicated in the final list of keys.
-
-  ## Examples
-
-      iex> Keyword.keys([a: 1, b: 2])
-      [:a,:b]
-
-      iex> Keyword.keys([a: 1, b: 2, a: 3])
-      [:a,:b,:a]
-
-  """
-  @spec keys(t) :: [key]
-  def keys(keywords) do
-    lc { key, _ } inlist keywords, do: key
-  end
-
-  @doc """
-  Returns all values from the keyword list.
-
-  ## Examples
-
-      iex> Keyword.values([a: 1, b: 2])
-      [1,2]
-
-  """
-  @spec values(t) :: [value]
-  def values(keywords) do
-    lc { _, value } inlist keywords, do: value
-  end
-
-  @doc """
-  Deletes all entries in the keyword list for a specific `key`.
-  If the `key` does not exist, returns the keyword list unchanged.
-  Use `delete_first` to delete just the first entry in case of
-  duplicated keys.
-
-  ## Examples
-
-      iex> Keyword.delete([a: 1, b: 2], :a)
-      [b: 2]
-
-      iex> Keyword.delete([a: 1, b: 2, a: 3], :a)
-      [b: 2]
-
-      iex> Keyword.delete([b: 2], :a)
-      [b: 2]
-
-  """
-  @spec delete(t, key) :: t
-  def delete(keywords, key) when is_atom(key) do
-    lc { k, _ } = tuple inlist keywords, key != k, do: tuple
-  end
-
-  @doc """
-  Deletes the first entry in the keyword list for a specific `key`.
-  If the `key` does not exist, returns the keyword list unchanged.
-
-  ## Examples
-
-      iex> Keyword.delete_first([a: 1, b: 2, a: 3], :a)
-      [b: 2, a: 3]
-
-      iex> Keyword.delete_first([b: 2], :a)
-      [b: 2]
-
-  """
-  @spec delete_first(t, key) :: t
-  def delete_first(keywords, key) when is_atom(key) do
-    :lists.keydelete(key, 1, keywords)
-  end
-
-  @doc """
-  Puts the given `value` under `key`.
-
-  If a previous value is already stored, all entries are
-  removed and the value is overriden.
-
-  ## Examples
-
-      iex> Keyword.put([a: 1, b: 2], :a, 3)
-      [a: 3, b: 2]
-
-      iex> Keyword.put([a: 1, b: 2, a: 4], :a, 3)
-      [a: 3, b: 2]
-
-  """
-  @spec put(t, key, value) :: t
-  def put(keywords, key, value) when is_atom(key) do
-    [{key, value}|delete(keywords, key)]
-  end
-
-  @doc """
-  Puts the given `value` under `key` unless the entry `key`
-  already exists.
-
-  ## Examples
-      iex> Keyword.put_new([a: 1], :b, 2)
-      [b: 2, a: 1]
-
-      iex> Keyword.put_new([a: 1, b: 2], :a, 3)
-      [a: 1, b: 2]
-
-  """
-  @spec put_new(t, key, value) :: t
-  def put_new(keywords, key, value) when is_atom(key) do
-    case :lists.keyfind(key, 1, keywords) do
-      { ^key, _ } -> keywords
-      false -> [{key, value}|keywords]
-    end
-  end
-
-  @doc """
-  Checks if two keywords are equal. I.e. they contain
-  the same keys and those keys contain the same values.
-
-  ## Examples
-
-      iex> Keyword.equal?([a: 1, b: 2], [b: 2, a: 1])
-      true
-
-  """
-  @spec equal?(t, t) :: boolean
-  def equal?(left, right) do
-    :lists.sort(left) == :lists.sort(right)
-  end
-
-  @doc """
-  Merges two keyword lists into one. If they have duplicated
-  entries, the one given as second argument wins.
-
-  ## Examples
-
-      iex> Keyword.merge([a: 1, b: 2], [a: 3, d: 4]) |> Enum.sort
-      [a: 3, b: 2, d: 4]
-
-  """
-  @spec merge(t, t) :: t
-  def merge(d1, d2) do
-    d2 ++ lc({ k, _ } = tuple inlist d1, not has_key?(d2, k), do: tuple)
-  end
-
-  @doc """
-  Merges two keyword lists into one. If they have duplicated
-  entries, the given function is invoked to solve conflicts.
-
-  ## Examples
-
-      iex> Keyword.merge([a: 1, b: 2], [a: 3, d: 4], fn (_k, v1, v2) ->
-      ...>  v1 + v2
-      iex> end)
-      [a: 4, b: 2, d: 4]
-
-  """
-  @spec merge(t, t, (key, value, value -> value)) :: t
-  def merge(d1, d2, fun) do
-    do_merge(d2, d1, fun)
-  end
-
-  defp do_merge([{ k, v2 }|t], acc, fun) do
-    do_merge t, update(acc, k, v2, fn(v1) -> fun.(k, v1, v2) end), fun
-  end
-
-  defp do_merge([], acc, _fun) do
-    acc
-  end
-
-  @doc """
-  Returns whether a given `key` exists in the given `keywords`.
-
-  ## Examples
-
-      iex> Keyword.has_key?([a: 1], :a)
-      true
-
-      iex> Keyword.has_key?([a: 1], :b)
-      false
-
-  """
-  @spec has_key?(t, key) :: boolean
-  def has_key?(keywords, key) when is_atom(key) do
-    :lists.keymember(key, 1, keywords)
-  end
-
-  @doc """
-  Updates the `key` with the given function. If the `key` does
-  not exist, raises `KeyError`.
-
-  ## Examples
-
-      iex> Keyword.update!([a: 1], :a, &(&1 * 2))
-      [a: 2]
-
-      iex> Keyword.update!([a: 1], :b, &(&1 * 2))
-      ** (KeyError) key not found: :b
-
-  """
-  @spec update!(t, key, (value -> value)) :: t | no_return
-  def update!([{key, value}|keywords], key, fun) do
-    [{key, fun.(value)}|delete(keywords, key)]
-  end
-
-  def update!([{_, _} = e|keywords], key, fun) do
-    [e|update!(keywords, key, fun)]
-  end
-
-  def update!([], key, _fun) when is_atom(key) do
-    raise(KeyError, key: key)
-  end
-
-  @doc """
-  Updates the `key` with the given function. If the `key` does
-  not exist, inserts the given `initial` value.
-
-  ## Examples
-
-      iex> Keyword.update([a: 1], :a, 13, &(&1 * 2))
-      [a: 2]
-
-      iex> Keyword.update([a: 1], :b, 11, &(&1 * 2))
-      [a: 1, b: 11]
-
-  """
-  @spec update(t, key, value, (value -> value)) :: t
-  def update([{key, value}|keywords], key, _initial, fun) do
-    [{key, fun.(value)}|delete(keywords, key)]
-  end
-
-  def update([{_, _} = e|keywords], key, initial, fun) do
-    [e|update(keywords, key, initial, fun)]
-  end
-
-  def update([], key, initial, _fun) when is_atom(key) do
-    [{key, initial}]
-  end
-
-  @doc """
-  Splits the given keywords in two given the given keys.
-  Duplicated keys are preserved in the split keyword list.
-
-  ## Examples
-
-      iex> d = [a: 1, b: 2, c: 3, d: 4]
-      iex> Keyword.split(d, [:a, :c, :e])
-      { [a: 1, c: 3], [b: 2, d: 4] }
-
-      iex> d = [a: 1, b: 2, c: 3, d: 4, a: 5]
-      iex> Keyword.split(d, [:a, :c, :e])
-      { [a: 1, c: 3, a: 5], [b: 2, d: 4] }
-
-  """
-  def split(dict, keys) do
-    acc = { [], [] }
-
-    { take, drop } = Enum.reduce dict, acc, fn({ k, v }, { take, drop }) ->
-      case :lists.member(k, keys) do
-        true  -> { [{k, v}|take], drop }
-        false -> { take, [{k, v}|drop] }
-      end
-    end
-
-    { Enum.reverse(take), Enum.reverse(drop) }
-  end
-
-  @doc """
-  Takes the given keys from the dict.
-  Duplicated keys are preserved in the new keyword list.
-
-  ## Examples
-
-      iex> d = [a: 1, b: 2, c: 3, d: 4]
-      iex> Keyword.take(d, [:a, :c, :e])
-      [a: 1, c: 3]
-
-      iex> d = [a: 1, b: 2, c: 3, d: 4, a: 5]
-      iex> Keyword.take(d, [:a, :c, :e])
-      [a: 1, c: 3, a: 5]
-
-  """
-  def take(dict, keys) do
-    lc { k, _ } = tuple inlist dict, :lists.member(k, keys), do: tuple
-  end
-
-  @doc """
-  Drops the given keys from the dict.
-  Duplicated keys are preserved in the new keyword list.
-
-  ## Examples
-
-      iex> d = [a: 1, b: 2, c: 3, d: 4]
-      iex> Keyword.drop(d, [:b, :d])
-      [a: 1, c: 3]
-
-      iex> d = [a: 1, b: 2, c: 3, d: 4, a: 5]
-      iex> Keyword.drop(d, [:b, :d])
-      [a: 1, c: 3, a: 5]
-
-  """
-  def drop(dict, keys) do
-    lc { k, _ } = tuple inlist dict, not :lists.member(k, keys), do: tuple
-  end
-
-  @doc """
-  Returns the first value associated with `key` in the keyword
-  list as well as the keyword list without `key`.
-
-  All duplicated entries are removed. See `pop_first/3` for
-  removing only the first entry.
-
-  ## Examples
-
-      iex> Keyword.pop [a: 1], :a
-      {1,[]}
-
-      iex> Keyword.pop [a: 1], :b
-      {nil,[a: 1]}
-
-      iex> Keyword.pop [a: 1], :b, 3
-      {3,[a: 1]}
-
-      iex> Keyword.pop [a: 1], :b, 3
-      {3,[a: 1]}
-
-      iex> Keyword.pop [a: 1, a: 2], :a
-      {1,[]}
-
-  """
-  def pop(dict, key, default // nil) do
-    { get(dict, key, default), delete(dict, key) }
-  end
-
-  @doc """
-  Returns the first value associated with `key` in the keyword
-  list as well as the keyword list without that particular ocurrence
-  of `key`.
-
-  Duplicated entries are not removed.
-
-  ## Examples
-
-      iex> Keyword.pop_first [a: 1], :a
-      {1,[]}
-
-      iex> Keyword.pop_first [a: 1], :b
-      {nil,[a: 1]}
-
-      iex> Keyword.pop_first [a: 1], :b, 3
-      {3,[a: 1]}
-
-      iex> Keyword.pop_first [a: 1], :b, 3
-      {3,[a: 1]}
-
-      iex> Keyword.pop_first [a: 1, a: 2], :a
-      {1,[a: 2]}
-
-  """
-  def pop_first(dict, key, default // nil) do
-    { get(dict, key, default), delete_first(dict, key) }
-  end
-end
@@ -1,417 +0,0 @@
-defmodule List do
-  @moduledoc """
-  Implements functions that only make sense for lists
-  and cannot be part of the Enum protocol. In general,
-  favor using the Enum API instead of List.
-
-  A decision was taken to delegate most functions to
-  Erlang's standard library but follow Elixir's convention
-  of receiving the target (in this case, a list) as the
-  first argument.
-  """
-
-  @compile :inline_list_funcs
-
-  @doc """
-  Deletes the given item from the list. Returns a list without the item.
-  If the item occurs more than once in the list, just the first occurrence
-  is removed.
-
-  ## Examples
-
-      iex> List.delete([1, 2, 3], 1)
-      [2,3]
-
-      iex> List.delete([1, 2, 2, 3], 2)
-      [1, 2, 3]
-
-  """
-  def delete(list, item) do
-    :lists.delete(item, list)
-  end
-
-  @doc """
-  Duplicates the given element `n` times in a list.
-
-  ## Examples
-
-      iex> List.duplicate("hello", 3)
-      ["hello","hello","hello"]
-
-      iex> List.duplicate([1, 2], 2)
-      [[1,2],[1,2]]
-  """
-  def duplicate(elem, n) do
-    :lists.duplicate(n, elem)
-  end
-
-  @doc """
-  Flattens the given `list` of nested lists.
-
-  ## Examples
-
-      iex> List.flatten([1, [[2], 3]])
-      [1,2,3]
-
-  """
-  def flatten(list) do
-    :lists.flatten(list)
-  end
-
-  @doc """
-  Flattens the given `list` of nested lists. 
-  The list `tail` will be added at the end of
-  the flattened list.
-
-  ## Examples
-
-      iex> List.flatten([1, [[2], 3]], [4, 5])
-      [1,2,3,4,5]
-
-  """
-  def flatten(list, tail) do
-    :lists.flatten(list, tail)
-  end
-
-  @doc """
-  Folds (reduces) the given list to the left with
-  a function. Requires an accumulator.
-
-  ## Examples
-
-      iex> List.foldl([5, 5], 10, fn (x, acc) -> x + acc end)
-      20
-
-      iex> List.foldl([1, 2, 3, 4], 0, fn (x, acc) -> x - acc end)
-      2
-
-  """
-  def foldl(list, acc, function) when is_list(list) and is_function(function) do
-    :lists.foldl(function, acc, list)
-  end
-
-  @doc """
-  Folds (reduces) the given list to the right with
-  a function. Requires an accumulator.
-
-  ## Examples
-
-      iex> List.foldr([1, 2, 3, 4], 0, fn (x, acc) -> x - acc end)
-      -2
-
-  """
-  def foldr(list, acc, function) when is_list(list) and is_function(function) do
-    :lists.foldr(function, acc, list)
-  end
-
-  @doc """
-  Returns the last element in `list` or `nil` if `list` is empty.
-
-  ## Examples
-
-      iex> List.last([])
-      nil
-      iex> List.last([1])
-      1
-      iex> List.last([1, 2, 3])
-      3
-
-  """
-  def last([]), do: nil
-
-  def last(list) do
-    :lists.last(list)
-  end
-
-  @doc """
-  Receives a list of tuples and returns the first tuple
-  where the item at `position` in the tuple matches the
-  given `item`.
-
-  ## Examples
-
-      iex> List.keyfind([a: 1, b: 2], :a, 0)
-      { :a, 1 }
-
-      iex> List.keyfind([a: 1, b: 2], 2, 1)
-      { :b, 2 }
-
-      iex> List.keyfind([a: 1, b: 2], :c, 0)
-      nil
-
-  """
-  def keyfind(list, key, position, default // nil) do
-    :lists.keyfind(key, position + 1, list) || default
-  end
-
-  @doc """
-  Receives a list of tuples and returns `true` if there is
-  a tuple where the item at `position` in the tuple matches
-  the given `item`.
-
-  ## Examples
-
-      iex> List.keymember?([a: 1, b: 2], :a, 0)
-      true
-
-      iex> List.keymember?([a: 1, b: 2], 2, 1)
-      true
-
-      iex> List.keymember?([a: 1, b: 2], :c, 0)
-      false
-
-  """
-  def keymember?(list, key, position) do
-    :lists.keymember(key, position + 1, list)
-  end
-
-  @doc """
-  Receives a list of tuples and replaces the item
-  identified by `key` at `position` if it exists.
-
-  ## Examples
-
-      iex> List.keyreplace([a: 1, b: 2], :a, 0, { :a, 3 })
-      [a: 3, b: 2]
-
-  """
-  def keyreplace(list, key, position, new_tuple) do
-    :lists.keyreplace(key, position + 1, list, new_tuple)
-  end
-
-  @doc """
-  Receives a list of tuples and sorts the items
-  at `position` of the tuples. The sort is stable.
-
-  ## Examples
-
-      iex> List.keysort([a: 5, b: 1, c: 3], 1)
-      [b: 1, c: 3, a: 5]
-
-      iex> List.keysort([a: 5, c: 1, b: 3], 0)
-      [a: 5, b: 3, c: 1]
-
-  """
-  def keysort(list, position) do
-    :lists.keysort(position + 1, list)
-  end
-
-  @doc """
-  Receives a list of tuples and replaces the item
-  identified by `key` at `position`. If the item
-  does not exist, it is added to the end of the list.
-
-  ## Examples
-
-      iex> List.keystore([a: 1, b: 2], :a, 0, { :a, 3 })
-      [a: 3, b: 2]
-
-      iex> List.keystore([a: 1, b: 2], :c, 0, { :c, 3 })
-      [a: 1, b: 2, c: 3]
-
-  """
-  def keystore(list, key, position, new_tuple) do
-    :lists.keystore(key, position + 1, list, new_tuple)
-  end
-
-  @doc """
-  Receives a list of tuples and deletes the first tuple
-  where the item at `position` matches the
-  given `item`. Returns the new list.
-
-  ## Examples
-
-      iex> List.keydelete([a: 1, b: 2], :a, 0)
-      [{ :b, 2 }]
-
-      iex> List.keydelete([a: 1, b: 2], 2, 1)
-      [{ :a, 1 }]
-
-      iex> List.keydelete([a: 1, b: 2], :c, 0)
-      [{ :a, 1 }, { :b, 2 }]
-
-  """
-  def keydelete(list, key, position) do
-    :lists.keydelete(key, position + 1, list)
-  end
-
-  @doc """
-  Wraps the argument in a list.
-  If the argument is already a list, returns the list.
-  If the argument is `nil`, returns an empty list.
-
-  ## Examples
-
-      iex> List.wrap("hello")
-      ["hello"]
-
-      iex> List.wrap([1, 2, 3])
-      [1,2,3]
-
-      iex> List.wrap(nil)
-      []
-
-  """
-  def wrap(list) when is_list(list) do
-    list
-  end
-
-  def wrap(nil) do
-    []
-  end
-
-  def wrap(other) do
-    [other]
-  end
-
-  @doc """
-  Zips corresponding elements from each list in `list_of_lists`.
-
-  ## Examples
-
-      iex> List.zip([[1, 2], [3, 4], [5, 6]])
-      [{1, 3, 5}, {2, 4, 6}]
-
-      iex> List.zip([[1, 2], [3], [5, 6]])
-      [{1, 3, 5}]
-
-  """
-  def zip([]), do: []
-  def zip(list_of_lists) when is_list(list_of_lists) do
-    do_zip(list_of_lists, [])
-  end
-
-  @doc """
-  Unzips the given list of lists or tuples into separate lists and returns a
-  list of lists.
-
-  ## Examples
-
-      iex> List.unzip([{1, 2}, {3, 4}])
-      [[1, 3], [2, 4]]
-
-      iex> List.unzip([{1, :a, "apple"}, {2, :b, "banana"}, {3, :c}])
-      [[1, 2, 3], [:a, :b, :c]]
-
-  """
-  def unzip(list) when is_list(list) do
-    :lists.map &tuple_to_list/1, zip(list)
-  end
-
-  @doc """
-  Returns a list with `value` inserted at the specified `index`. Note that `index`
-  is capped at the list length. Negative indices indicate an offset from the
-  end of the list.
-
-  ## Examples
-
-      iex> List.insert_at([1, 2, 3, 4], 2, 0)
-      [1, 2, 0, 3, 4]
-
-      iex> List.insert_at([1, 2, 3], 10, 0)
-      [1, 2, 3, 0]
-
-      iex> List.insert_at([1, 2, 3], -1, 0)
-      [1, 2, 3, 0]
-
-      iex> List.insert_at([1, 2, 3], -10, 0)
-      [0, 1, 2, 3]
-
-  """
-  def insert_at(list, index, value) do
-    if index < 0 do
-      do_insert_at(list, length(list) + index + 1, value)
-    else
-      do_insert_at(list, index, value)
-    end
-  end
-
-  @doc """
-  Returns a list with a replaced value at the specified `index`. Negative indices
-  indicate an offset from the end of the list. If `index` is out of bounds, the
-  original `list` is returned.
-
-  ## Examples
-
-      iex> List.replace_at([1, 2, 3], 0, 0)
-      [0, 2, 3]
-
-      iex> List.replace_at([1, 2, 3], 10, 0)
-      [1, 2, 3]
-
-      iex> List.replace_at([1, 2, 3], -1, 0)
-      [1, 2, 0]
-
-      iex> List.replace_at([1, 2, 3], -10, 0)
-      [1, 2, 3]
-
-  """
-  def replace_at(list, index, value) do
-    if index < 0 do
-      do_replace_at(list, length(list) + index, value)
-    else
-      do_replace_at(list, index, value)
-    end
-  end
-
-  ## Helpers
-
-  # replace_at
-
-  defp do_replace_at([], _index, _value) do
-    []
-  end
-
-  defp do_replace_at(list, index, _value) when index < 0 do
-    list
-  end
-
-  defp do_replace_at([_old|rest], 0, value) do
-    [ value | rest ]
-  end
-
-  defp do_replace_at([h|t], index, value) do
-    [ h | do_replace_at(t, index - 1, value) ]
-  end
-
-  # insert_at
-
-  defp do_insert_at([], _index, value) do
-    [ value ]
-  end
-
-  defp do_insert_at(list, index, value) when index <= 0 do
-    [ value | list ]
-  end
-
-  defp do_insert_at([h|t], index, value) do
-    [ h | do_insert_at(t, index - 1, value) ]
-  end
-
-  # zip
-
-  defp do_zip(list, acc) do
-    converter = fn x, acc -> do_zip_each(to_list(x), acc) end
-    {mlist, heads} = :lists.mapfoldl converter, [], list
-
-    case heads do
-      nil -> :lists.reverse acc
-      _   -> do_zip mlist, [list_to_tuple(:lists.reverse(heads))|acc]
-    end
-  end
-
-  defp do_zip_each(_, nil) do
-    { nil, nil }
-  end
-
-  defp do_zip_each([h|t], acc) do
-    { t, [h|acc] }
-  end
-
-  defp do_zip_each([], _) do
-    { nil, nil }
-  end
-
-  defp to_list(tuple) when is_tuple(tuple), do: tuple_to_list(tuple)
-  defp to_list(list)  when is_list(list),   do: list
-end
@@ -1,56 +0,0 @@
-defprotocol List.Chars do
-  @moduledoc %S"""
-  The List.Chars protocol is responsible for
-  converting a structure to a list (only if applicable).
-  The only function required to be implemented is
-  `to_char_list` which does the conversion.
-
-  The `to_char_list` function automatically imported
-  by Kernel invokes this protocol.
-  """
-
-  def to_char_list(thing)
-end
-
-defimpl List.Chars, for: Atom do
-  def to_char_list(atom), do: atom_to_list(atom)
-end
-
-defimpl List.Chars, for: BitString do
-  @doc """
-  Returns the given binary converted to a char list.
-  """
-  def to_char_list(thing) when is_binary(thing) do
-    String.to_char_list!(thing)
-  end
-
-  def to_char_list(thing) do
-    raise Protocol.UndefinedError,
-             protocol: @protocol,
-                value: thing,
-          description: "cannot convert a bitstring to a char list"
-  end
-end
-
-defimpl List.Chars, for: List do
-  def to_char_list(list), do: list
-end
-
-defimpl List.Chars, for: Integer do
-  def to_char_list(thing) do
-    integer_to_list(thing)
-  end
-end
-
-defimpl List.Chars, for: Float do
-  @digits 20
-  @limit  :math.pow(10, @digits)
-
-  def to_char_list(thing) when thing > @limit do
-    float_to_list(thing, scientific: @digits)
-  end
-
-  def to_char_list(thing) do
-    float_to_list(thing, compact: true, decimals: @digits)
-  end
-end
@@ -1,146 +0,0 @@
-defmodule ListDict do
-  @moduledoc """
-  A Dict implementation that works on lists of two-items tuples.
-
-  This dictionary is only recommended for keeping a small amount
-  of values. Other dict alternatives are more viable for keeping
-  any other amount than a handful.
-
-  For more information about the functions and their APIs, please
-  consult the `Dict` module.
-  """
-
-  @doc """
-  Returns a new `ListDict`, i.e. an empty list.
-  """
-  def new, do: []
-
-  @doc """
-  Creates a new `ListDict` from the given pairs.
-  """
-  def new(pairs) do
-    Enum.to_list pairs
-  end
-
-  @doc """
-  Creates a new `ListDict` from the given pairs
-  via the given transformation function.
-  """
-  def new(list, transform) when is_function(transform) do
-    Enum.map list, transform
-  end
-
-  def keys(dict) do
-    lc { key, _ } inlist dict, do: key
-  end
-
-  def values(dict) do
-    lc { _, value } inlist dict, do: value
-  end
-
-  def size(dict) do
-    length(dict)
-  end
-
-  def has_key?(dict, key)
-  def has_key?([{ key, _ }|_], key), do: true
-  def has_key?([{ _, _ }|t], key), do: has_key?(t, key)
-  def has_key?([], _key), do: false
-
-  def get(dict, key, default // nil)
-  def get([{ key, value }|_], key, _default), do: value
-  def get([{ _, _ }|t], key, default), do: get(t, key, default)
-  def get([], _key, default), do: default
-
-  def fetch(dict, key)
-  def fetch([{ key, value }|_], key), do: { :ok, value }
-  def fetch([{ _, _ }|t], key), do: fetch(t, key)
-  def fetch([], _key), do: :error
-
-  def fetch!(dict, key) do
-    case fetch(dict, key) do
-      { :ok, value } -> value
-      :error -> raise(KeyError, key: key)
-    end
-  end
-
-  def pop(dict, key, default // nil) do
-    { get(dict, key, default), delete(dict, key) }
-  end
-
-  def put(dict, key, val) do
-    [{key, val}|delete(dict, key)]
-  end
-
-  def put_new(dict, key, val) do
-    case has_key?(dict, key) do
-      true  -> dict
-      false -> [{key, val}|dict]
-    end
-  end
-
-  def delete(dict, key)
-  def delete([{ key, _ }|t], key), do: t
-  def delete([{ _, _ } = h|t], key), do: [h|delete(t, key)]
-  def delete([], _key), do: []
-
-  def merge(dict, enum, callback // fn(_k, _v1, v2) -> v2 end) do
-    Enum.reduce enum, dict, fn { k, v2 }, acc ->
-      update(acc, k, v2, fn(v1) -> callback.(k, v1, v2) end)
-    end
-  end
-
-  def split(dict, keys) do
-    acc = { [], [] }
-
-    {take, drop} = Enum.reduce dict, acc, fn({ k, v }, { take, drop }) ->
-      if :lists.member(k, keys) do
-        { [{k, v}|take], drop }
-      else
-        { take, [{k, v}|drop] }
-      end
-    end
-
-    {Enum.reverse(take), Enum.reverse(drop)}
-  end
-
-  def take(dict, keys) do
-    lc { k, _ } = tuple inlist dict, :lists.member(k, keys), do: tuple
-  end
-
-  def drop(dict, keys) do
-    lc { k, _ } = tuple inlist dict, not :lists.member(k, keys), do: tuple
-  end
-
-  def update!([{key, value}|dict], key, fun) do
-    [{key, fun.(value)}|delete(dict, key)]
-  end
-
-  def update!([{_, _} = e|dict], key, fun) do
-    [e|update!(dict, key, fun)]
-  end
-
-  def update!([], key, _fun) do
-    raise(KeyError, key: key)
-  end
-
-  def update([{key, value}|dict], key, _initial, fun) do
-    [{key, fun.(value)}|delete(dict, key)]
-  end
-
-  def update([{_, _} = e|dict], key, initial, fun) do
-    [e|update(dict, key, initial, fun)]
-  end
-
-  def update([], key, initial, _fun) do
-    [{key, initial}]
-  end
-
-  def empty(_dict), do: []
-
-  def equal?(dict, other) do
-    :lists.keysort(1, dict) === :lists.keysort(1, other)
-  end
-
-  def to_list(dict), do: dict
-end
@@ -1,704 +0,0 @@
-import Kernel, except: [to_string: 1]
-
-defmodule Macro do
-  @moduledoc """
-  This module provides conveniences for working with macros.
-  """
-
-  @typedoc "Abstract Syntax Tree (AST) node"
-  @type t :: { t, t } | { t, Keyword.t, t } | atom | number | binary | list
-
-  @doc false
-  defmacro binary_ops do
-    [ :===, :!==,
-      :==, :!=, :<=, :>=,
-      :&&, :||, :<>, :++, :--, :**, ://, :::, :<-, :.., :|>, :=~,
-      :<, :>, :->,
-      :+, :-, :*, :/, :=, :|, :.,
-      :and, :or, :xor, :when, :in, :inlist, :inbits,
-      :<<<, :>>>, :|||, :&&&, :^^^, :~~~ ]
-  end
-
-  @doc false
-  defmacro unary_ops do
-    [:!, :@, :^, :not, :+, :-, :~~~, :&]
-  end
-
-  @spec binary_op_props(atom) :: { :left | :right, precedence :: integer }
-  defp binary_op_props(o) do
-    case o do
-      :::                                                       -> {:right, 30}
-      :when                                                     -> {:right, 40}
-      o when o in [:inlist, :inbits]                            -> {:left, 50}
-      ://                                                       -> {:right, 60}
-      :|                                                        -> {:left, 70}
-      :=                                                        -> {:right, 80}
-      o when o in [:||, :|||, :or, :xor]                        -> {:left, 130}
-      o when o in [:&&, :&&&, :and]                             -> {:left, 140}
-      o when o in [:==, :!=, :<, :<=, :>=, :>, :=~, :===, :!==] -> {:left, 150}
-      o when o in [:<-, :|>, :<<<, :>>>]                        -> {:right, 160}
-      :in                                                       -> {:left, 170}
-      :..                                                       -> {:left, 200}
-      o when o in [:+, :-]                                      -> {:left, 210}
-      o when o in [:*, :/]                                      -> {:left, 220}
-      o when o in [:<>]                                         -> {:right, 230}
-      :^^^                                                      -> {:left, 250}
-      :.                                                        -> {:left, 310}
-    end
-  end
-
-  @doc """
-  Breaks a pipeline expression into a list. Raises if
-  the pipeline is ill-formed.
-  """
-  @spec unpipe(Macro.t) :: [Macro.t]
-  def unpipe({ :|> , _, [left, right] }) do
-    [left|unpipe(right)]
-  end
-
-  def unpipe(other) do
-    [other]
-  end
-
-  @doc """
-  Pipes the given `expr` in to the `call_expr` as the
-  argument in the given `position`.
-  """
-  @spec pipe(Macro.t, Macro.t, integer) :: Macro.t | no_return
-  def pipe(expr, call_args, integer // 0)
-
-  def pipe(expr, { call, line, atom }, integer) when is_atom(atom) do
-    { call, line, List.insert_at([], integer, expr) }
-  end
-
-  def pipe(expr, { call, line, args }, integer) when is_list(args) do
-    { call, line, List.insert_at(args, integer, expr) }
-  end
-
-  def pipe(expr, call_args, _integer) do
-    raise ArgumentError,
-      message: "cannot pipe #{to_string expr} into #{to_string call_args}"
-  end
-
-  @doc """
-  Receives an expression representing a possible definition
-  and extracts its arguments. It returns a tuple with the
-  function name and the arguments list or `:error` if not
-  a valid call syntax.
-
-  This is useful for macros that want to provide the same
-  argument syntax available in def/defp/defmacro and friends.
-
-  ## Examples
-
-      extract_args(quote do: foo)        == { :foo, [] }
-      extract_args(quote do: foo())      == { :foo, [] }
-      extract_args(quote do: foo(1, 2, 3)) == { :foo, [1, 2, 3] }
-      extract_args(quote do: 1.(1, 2, 3))  == :error
-
-  """
-  @spec extract_args(Macro.t) :: { atom, [Macro.t] } | :error
-  def extract_args(expr) do
-    :elixir_clauses.extract_args(expr)
-  end
-
-  @doc """
-  Recursively escapes a value so it can be inserted
-  into a syntax tree.
-
-  One may pass `unquote: true` to `escape/2`
-  which leaves unquote statements unescaped, effectively
-  unquoting the contents on escape.
-
-  ## Examples
-
-      iex> Macro.escape(:foo)
-      :foo
-
-      iex> Macro.escape({ :a, :b, :c })
-      { :{}, [], [:a, :b, :c] }
-
-      iex> Macro.escape({ :unquote, [], [1] }, unquote: true)
-      1
-
-  """
-  @spec escape(term) :: Macro.t
-  @spec escape(term, Keyword.t) :: Macro.t
-  def escape(expr, opts // []) do
-    elem(:elixir_quote.escape(expr, Keyword.get(opts, :unquote, false)), 0)
-  end
-
-  @doc %S"""
-  Unescape the given chars. This is the unescaping behavior
-  used by default in Elixir single- and double-quoted strings.
-  Check `unescape_string/2` for information on how to customize
-  the escaping map.
-
-  In this setup, Elixir will escape the following: `\a`, `\b`,
-  `\d`, `\e`, `\f`, `\n`, `\r`, `\s`, `\t` and `\v`. Octals are
-  also escaped according to the latin1 set they represent.
-
-  This function is commonly used on sigil implementations
-  (like `%r`, `%b` and others) which receive a raw, unescaped
-  string.
-
-  ## Examples
-
-      iex> Macro.unescape_string("example\\n")
-      "example\n"
-
-  In the example above, we pass a string with `\n` escaped
-  and we return a version with it unescaped.
-  """
-  @spec unescape_string(String.t) :: String.t
-  def unescape_string(chars) do
-    :elixir_interpolation.unescape_chars(chars)
-  end
-
-  @doc %S"""
-  Unescape the given chars according to the map given.
-  Check `unescape_string/1` if you want to use the same map
-  as Elixir single- and double-quoted strings.
-
-  ## Map
-
-  The map must be a function. The function receives an integer
-  representing the number of the characters it wants to unescape.
-  Here is the default mapping function implemented by Elixir:
-
-      def unescape_map(?a), do: ?\a
-      def unescape_map(?b), do: ?\b
-      def unescape_map(?d), do: ?\d
-      def unescape_map(?e), do: ?\e
-      def unescape_map(?f), do: ?\f
-      def unescape_map(?n), do: ?\n
-      def unescape_map(?r), do: ?\r
-      def unescape_map(?s), do: ?\s
-      def unescape_map(?t), do: ?\t
-      def unescape_map(?v), do: ?\v
-      def unescape_map(e),  do: e
-
-  If the `unescape_map` function returns `false`. The char is
-  not escaped and `\` is kept in the char list.
-
-  ## Octals
-
-  Octals will by default be escaped unless the map function
-  returns false for ?0.
-
-  ## Hex
-
-  Hexadecimals will by default be escaped unless the map function
-  returns false for ?x.
-
-  ## Examples
-
-  Using the unescape_map defined above is easy:
-
-      Macro.unescape_string "example\\n", &unescape_map(&1)
-
-  """
-  @spec unescape_string(String.t, (non_neg_integer -> non_neg_integer | false)) :: String.t
-  def unescape_string(chars, map) do
-    :elixir_interpolation.unescape_chars(chars, map)
-  end
-
-  @doc """
-  Unescape the given tokens according to the default map.
-  Check `unescape_string/1` and `unescape_string/2` for more
-  information about unescaping.
-
-  Only tokens that are binaries are unescaped, all others are
-  ignored. This function is useful when implementing your own
-  sigils. Check the implementation of `Kernel.sigil_b`
-  for examples.
-  """
-  @spec unescape_tokens([Macro.t]) :: [Macro.t]
-  def unescape_tokens(tokens) do
-    :elixir_interpolation.unescape_tokens(tokens)
-  end
-
-  @doc """
-  Unescape the given tokens according to the given map.
-  Check `unescape_tokens/1` and `unescape_string/2` for more information.
-  """
-  @spec unescape_tokens([Macro.t], (non_neg_integer -> non_neg_integer | false)) :: [Macro.t]
-  def unescape_tokens(tokens, map) do
-    :elixir_interpolation.unescape_tokens(tokens, map)
-  end
-
-  @doc """
-  Converts the given expression to a binary.
-
-  ## Examples
-
-      iex> Macro.to_string(quote do: foo.bar(1, 2, 3))
-      "foo.bar(1, 2, 3)"
-
-  """
-  @spec to_string(Macro.t) :: String.t
-  @spec to_string(Macro.t, (Macro.t, String.t -> String.t)) :: String.t
-  def to_string(tree, fun // fn(_ast, string) -> string end)
-
-  # Variables
-  def to_string({ var, _, atom } = ast, fun) when is_atom(atom) do
-    fun.(ast, atom_to_binary(var))
-  end
-
-  # Aliases
-  def to_string({ :__aliases__, _, refs } = ast, fun) do
-    fun.(ast, Enum.map_join(refs, ".", &call_to_string(&1, fun)))
-  end
-
-  # Blocks
-  def to_string({ :__block__, _, [expr] } = ast, fun) do
-    fun.(ast, to_string(expr, fun))
-  end
-
-  def to_string({ :__block__, _, _ } = ast, fun) do
-    block = adjust_new_lines block_to_string(ast, fun), "\n  "
-    fun.(ast, "(\n  " <> block <> "\n)")
-  end
-
-  # Bits containers
-  def to_string({ :<<>>, _, args } = ast, fun) do
-    fun.(ast, case Enum.map_join(args, ", ", &to_string(&1, fun)) do
-      "<" <> rest -> "<< <" <> rest  <> " >>"
-      rest -> "<<" <> rest <> ">>"
-    end)
-  end
-
-  # Tuple containers
-  def to_string({ :{}, _, args } = ast, fun) do
-    fun.(ast, "{" <> Enum.map_join(args, ", ", &to_string(&1, fun)) <> "}")
-  end
-
-  # Fn keyword
-  def to_string({ :fn, _, [{ :->, _, [{_, _, tuple}] } = arrow] } = ast, fun)
-      when not is_tuple(tuple) or elem(tuple, 0) != :__block__ do
-    fun.(ast, "fn " <> arrow_to_string(arrow, fun) <> " end")
-  end
-
-  def to_string({ :fn, _, [{ :->, _, [_] } = block] } = ast, fun) do
-    fun.(ast, "fn " <> block_to_string(block, fun) <> "\nend")
-  end
-
-  def to_string({ :fn, _, [block] } = ast, fun) do
-    block = adjust_new_lines block_to_string(block, fun), "\n  "
-    fun.(ast, "fn\n  " <> block <> "\nend")
-  end
-
-  # left -> right
-  def to_string({ :->, _, _ } = ast, fun) do
-    fun.(ast, "(" <> arrow_to_string(ast, fun, true) <> ")")
-  end
-
-  # Binary ops
-  def to_string({ op, _, [left, right] } = ast, fun) when op in binary_ops do
-    fun.(ast, op_to_string(left, fun, op, :left) <> " #{op} " <> op_to_string(right, fun, op, :right))
-  end
-
-  # Splat when
-  def to_string({ :when, _, args } = ast, fun) do
-    { left, right } = :elixir_utils.split_last(args)
-    fun.(ast, "(" <> Enum.map_join(left, ", ", &to_string(&1, fun)) <> ") when " <> to_string(right, fun))
-  end
-
-  # Unary ops
-  def to_string({ :not, _, [arg] } = ast, fun)  do
-    fun.(ast, "not " <> to_string(arg, fun))
-  end
-
-  def to_string({ op, _, [arg] } = ast, fun) when op in unary_ops do
-    fun.(ast, atom_to_binary(op) <> to_string(arg, fun))
-  end
-
-  # Access
-  def to_string({ { :., _, [Kernel, :access] }, _, [left, right] } = ast, fun) do
-    fun.(ast, to_string(left, fun) <> to_string(right, fun))
-  end
-
-  # All other calls
-  def to_string({ target, _, args } = ast, fun) when is_list(args) do
-    { list, last } = :elixir_utils.split_last(args)
-    fun.(ast, case is_kw_blocks?(last) do
-      true  -> call_to_string_with_args(target, list, fun) <> kw_blocks_to_string(last, fun)
-      false -> call_to_string_with_args(target, args, fun)
-    end)
-  end
-
-  # Two-item tuples
-  def to_string({ left, right }, fun) do
-    to_string({ :{}, [], [left, right] }, fun)
-  end
-
-  # Lists
-  def to_string(list, fun) when is_list(list) do
-    if Keyword.keyword?(list) do
-      fun.(list, "[" <> kw_list_to_string(list, fun) <> "]")
-    else
-      fun.(list, "[" <> Enum.map_join(list, ", ", &to_string(&1, fun)) <> "]")
-    end
-  end
-
-  # All other structures
-  def to_string(other, fun), do: fun.(other, inspect(other, raw: true))
-
-  # Block keywords
-  defmacrop kw_keywords, do: [:do, :catch, :rescue, :after, :else]
-
-  defp is_kw_blocks?([_|_] = kw) do
-    Enum.all?(kw, &match?({x, _} when x in kw_keywords, &1))
-  end
-  defp is_kw_blocks?(_), do: false
-
-  defp module_to_string(atom, _fun) when is_atom(atom), do: inspect(atom, raw: true)
-  defp module_to_string(other, fun), do: call_to_string(other, fun)
-
-  defp call_to_string(atom, _fun) when is_atom(atom), do: atom_to_binary(atom)
-  defp call_to_string({ :., _, [arg] }, fun),         do: module_to_string(arg, fun) <> "."
-  defp call_to_string({ :., _, [left, right] }, fun), do: module_to_string(left, fun) <> "." <> call_to_string(right, fun)
-  defp call_to_string(other, fun),                    do: to_string(other, fun)
-
-  defp call_to_string_with_args(target, args, fun) do
-    { list, last } = :elixir_utils.split_last(args)
-    target = call_to_string(target, fun)
-
-    case last != [] and Keyword.keyword?(last) do
-      true  ->
-        args = Enum.map_join(list, ", ", &to_string(&1, fun))
-        if list != [], do: args = args <> ", "
-        args = args <> kw_list_to_string(last, fun)
-        target <> "(" <> args <> ")"
-      false ->
-        args = Enum.map_join(args, ", ", &to_string(&1, fun))
-        target <> "(" <> args <> ")"
-    end
-  end
-
-  defp kw_blocks_to_string(kw, fun) do
-    Enum.reduce(kw_keywords, " ", fn(x, acc) ->
-      case Keyword.has_key?(kw, x) do
-        true  -> acc <> kw_block_to_string(x, Keyword.get(kw, x), fun)
-        false -> acc
-      end
-    end) <> "end"
-  end
-
-  defp kw_block_to_string(key, value, fun) do
-    block = adjust_new_lines block_to_string(value, fun), "\n  "
-    atom_to_binary(key) <> "\n  " <> block <> "\n"
-  end
-
-  defp block_to_string({ :->, _, exprs }, fun) do
-    Enum.map_join(exprs, "\n", fn({ left, _, right }) ->
-      left = comma_join_or_empty_paren(left, fun, false)
-      left <> "->\n  " <> adjust_new_lines block_to_string(right, fun), "\n  "
-    end)
-  end
-
-  defp block_to_string({ :__block__, _, exprs }, fun) do
-    Enum.map_join(exprs, "\n", &to_string(&1, fun))
-  end
-
-  defp block_to_string(other, fun), do: to_string(other, fun)
-
-  defp kw_list_to_string(list, fun) do
-    Enum.map_join(list, ", ", fn { key, value } ->
-      atom_to_binary(key) <> ": " <> to_string(value, fun)
-    end)
-  end
-
-  defp parenthise(expr, fun) do
-    "(" <> to_string(expr, fun) <> ")"
-  end
-
-  defp op_to_string({ op, _, [_, _] } = expr, fun, parent_op, side) when op in binary_ops do
-    { parent_assoc, parent_prec } = binary_op_props(parent_op)
-    { _, prec }                   = binary_op_props(op)
-    cond do
-      parent_prec < prec -> to_string(expr, fun)
-      parent_prec > prec -> parenthise(expr, fun)
-      true ->
-        # parent_prec == prec, so look at associativity.
-        if parent_assoc == side do
-          to_string(expr, fun)
-        else
-          parenthise(expr, fun)
-        end
-    end
-  end
-
-  defp op_to_string(expr, fun, _, _), do: to_string(expr, fun)
-
-  defp arrow_to_string({ :->, _, pairs }, fun, paren // false) do
-    Enum.map_join(pairs, "; ", fn({ left, _, right }) ->
-      left = comma_join_or_empty_paren(left, fun, paren)
-      left <> "-> " <> to_string(right, fun)
-    end)
-  end
-
-  defp comma_join_or_empty_paren([], _fun, true),  do: "() "
-  defp comma_join_or_empty_paren([], _fun, false), do: ""
-
-  defp comma_join_or_empty_paren(left, fun, _) do
-    Enum.map_join(left, ", ", &to_string(&1, fun)) <> " "
-  end
-
-  defp adjust_new_lines(block, replacement) do
-    bc <<x>> inbits block do
-      << case x == ?\n do
-        true  -> replacement
-        false -> <<x>>
-      end :: binary >>
-    end
-  end
-
-  @doc """
-  Receives a AST node and expands it once. The following contents are expanded:
-
-  * Macros (local or remote);
-  * Aliases are expanded (if possible) and return atoms;
-  * All pseudo-variables (`__FILE__`, `__MODULE__`, etc);
-  * Module attributes reader (`@foo`);
-
-  In case the expression cannot be expanded, it returns the expression
-  itself. Notice that `expand_once/2` performs the expansion just
-  once and it is not recursive. Check `expand/2` for expansion
-  until the node no longer represents a macro and `expand_all/2`
-  for recursive expansion.
-
-  ## Examples
-
-  In the example below, we have a macro that generates a module
-  with a function named `name_length` that returns the length
-  of the module name. The value of this function will be calculated
-  at compilation time and not at runtime.
-
-  Consider the implementation below:
-
-      defmacro defmodule_with_length(name, do: block) do
-        length = length(atom_to_list(name))
-
-        quote do
-          defmodule unquote(name) do
-            def name_length, do: unquote(length)
-            unquote(block)
-          end
-        end
-      end
-
-  When invoked like this:
-
-      defmodule_with_length My.Module do
-        def other_function, do: ...
-      end
-
-  The compilation will fail because `My.Module` when quoted
-  is not an atom, but a syntax tree as follow:
-
-      {:__aliases__, [], [:My, :Module] }
-
-  That said, we need to expand the aliases node above to an
-  atom, so we can retrieve its length. Expanding the node is
-  not straight-forward because we also need to expand the
-  caller aliases. For example:
-
-      alias MyHelpers, as: My
-
-      defmodule_with_length My.Module do
-        def other_function, do: ...
-      end
-
-  The final module name will be `MyHelpers.Module` and not
-  `My.Module`. With `Macro.expand`, such aliases are taken
-  into consideration. Local and remote macros are also
-  expanded. We could rewrite our macro above to use this
-  function as:
-
-      defmacro defmodule_with_length(name, do: block) do
-        expanded = Macro.expand(name, __CALLER__)
-        length   = length(atom_to_list(expanded))
-
-        quote do
-          defmodule unquote(name) do
-            def name_length, do: unquote(length)
-            unquote(block)
-          end
-        end
-      end
-
-  """
-  def expand_once(ast, env) do
-    elem(expand_once(ast, env, nil), 0)
-  end
-
-  defp expand_once({ :__aliases__, _, _ } = original, env, cache) do
-    case :elixir_aliases.expand(original, env.aliases, env.macro_aliases, env.lexical_tracker) do
-      receiver when is_atom(receiver) ->
-        :elixir_lexical.record_remote(receiver, env.lexical_tracker)
-        { receiver, true, cache }
-      aliases ->
-        aliases = lc alias inlist aliases, do: elem(expand_once(alias, env, cache), 0)
-
-        case :lists.all(&is_atom/1, aliases) do
-          true ->
-            receiver = :elixir_aliases.concat(aliases)
-            :elixir_lexical.record_remote(receiver, env.lexical_tracker)
-            { receiver, true, cache }
-          false -> { original, false, cache }
-        end
-    end
-  end
-
-  # Expand @ calls
-  defp expand_once({ :@, _, [{ name, _, args }] } = original, env, cache) when is_atom(args) or args == [] do
-    case (module = env.module) && Module.open?(module) do
-      true  -> { Module.get_attribute(module, name), true, cache }
-      false -> { original, false, cache }
-    end
-  end
-
-  # Expand pseudo-variables
-  defp expand_once({ :__MODULE__, _, atom }, env, cache) when is_atom(atom),
-    do: { env.module, true, cache }
-  defp expand_once({ :__FILE__, _, atom }, env, cache)   when is_atom(atom),
-    do: { env.file, true, cache }
-  defp expand_once({ :__DIR__, _, atom }, env, cache)    when is_atom(atom),
-    do: { :filename.dirname(env.file), true, cache }
-  defp expand_once({ :__ENV__, _, atom }, env, cache)    when is_atom(atom),
-    do: { env, true, cache }
-
-  # Expand possible macro import invocation
-  defp expand_once({ atom, line, args } = original, env, cache) when is_atom(atom) do
-    args = case is_atom(args) do
-      true  -> []
-      false -> args
-    end
-
-    case not is_partial?(args) do
-      false -> { original, false, cache }
-      true  ->
-        module = env.module
-
-        extra  = if function_exported?(module, :__info__, 1) do
-          [{ module, module.__info__(:macros) }]
-        else
-          []
-        end
-
-        cache  = to_erl_env(env, cache)
-        expand = :elixir_dispatch.expand_import(line, { atom, length(args) }, args,
-          env.module, extra, cache)
-
-        case expand do
-          { :ok, _, expanded } -> { expanded, true, cache }
-          { :error, _ }        -> { original, false, cache }
-        end
-    end
-  end
-
-  # Expand possible macro require invocation
-  defp expand_once({ { :., _, [left, right] }, line, args } = original, env, cache) when is_atom(right) do
-    { receiver, _, _ } = expand_once(left, env, cache)
-
-    case is_atom(receiver) and not is_partial?(args) do
-      false -> { original, false, cache }
-      true  ->
-        cache  = to_erl_env(env, cache)
-        expand = :elixir_dispatch.expand_require(line, receiver, { right, length(args) },
-          args, env.module, cache)
-
-        case expand do
-          { :ok, _receiver, expanded } -> { expanded, true, cache }
-          { :error, _ }                -> { original, false, cache }
-        end
-    end
-  end
-
-  # Anything else is just returned
-  defp expand_once(other, _env, cache), do: { other, false, cache }
-
-  defp to_erl_env(env, nil),    do: :elixir_scope.to_erl_env(env)
-  defp to_erl_env(_env, cache), do: cache
-
-  defp is_partial?(args) do
-    :lists.any(&match?({ :&, _, [_] }, &1), args)
-  end
-
-  @doc """
-  Receives a AST node and expands it until it no longer represents
-  a macro. Check `expand_once/2` for more information on how
-  expansion works and `expand_all/2` for recursive expansion.
-  """
-  def expand(tree, env) do
-    elem(expand(tree, env, nil), 0)
-  end
-
-  @doc false # Used internally by Elixir
-  def expand(tree, env, cache) do
-    expand_until({ tree, true, cache }, env)
-  end
-
-  defp expand_until({ tree, true, cache }, env) do
-    expand_until(expand_once(tree, env, cache), env)
-  end
-
-  defp expand_until({ tree, false, cache }, _env) do
-    { tree, cache }
-  end
-
-  @doc false
-  def expand_all(tree, env) do
-    IO.write "Macro.expand_all/2 is deprecated, please avoid recursive code expansion\n#{Exception.format_stacktrace}"
-    elem(expand_all(tree, env, nil), 0)
-  end
-
-  @doc false # Used internally by Elixir
-  def expand_all(tree, env, cache) do
-    expand_all_until(expand(tree, env, cache), env)
-  end
-
-  defp expand_all_until({ { left, meta, right }, cache }, env) do
-    { left, cache }  = expand_all(left, env, cache)
-    { right, cache } = expand_all(right, env, cache)
-    { { left, meta, right }, cache }
-  end
-
-  defp expand_all_until({ { left, right }, cache }, env) do
-    { left, cache }  = expand_all(left, env, cache)
-    { right, cache } = expand_all(right, env, cache)
-    { { left, right }, cache }
-  end
-
-  defp expand_all_until({ list, cache }, env) when is_list(list) do
-    :lists.mapfoldl(&expand_all(&1, env, &2), cache, list)
-  end
-
-  defp expand_all_until({ other, cache }, _env) do
-    { other, cache }
-  end
-
-  @doc """
-  Recurs the quoted expression checking if all sub-terms are
-  safe (i.e. they represent data structures and don't actually
-  evaluate code) and returns `:ok` unless a given term is unsafe,
-  which is returned as `{ :unsafe, term }`.
-  """
-  def safe_term(terms) do
-    do_safe_term(terms) || :ok
-  end
-
-  defp do_safe_term({ local, _, terms }) when local in [:{}, :__aliases__] do
-    do_safe_term(terms)
-  end
-
-  defp do_safe_term({ unary, _, [term] }) when unary in [:+, :-] do
-    do_safe_term(term)
-  end
-
-  defp do_safe_term({ left, right }), do: do_safe_term(left) || do_safe_term(right)
-  defp do_safe_term(terms) when is_list(terms),  do: Enum.find_value(terms, &do_safe_term(&1))
-  defp do_safe_term(terms) when is_tuple(terms), do: { :unsafe, terms }
-  defp do_safe_term(_), do: nil
-end
@@ -1,84 +0,0 @@
-defmodule Macro.Env do
-  @moduledoc """
-  A record that holds compile time environment information.
-
-  The current environment can be accessed at any time as
-  `__ENV__`. Inside macros, the caller environment can be
-  accessed as `__CALLER__`. It contains the following fields:
-
-  * `module` - the current module name.
-  * `file` - the current file name as a binary
-  * `line` - the current line as an integer
-  * `function` - a tuple as `{ atom, integer` }, where the first
-    element is the function name and the seconds its arity. Returns
-    `nil` if not inside a function
-  * `aliases` -  a list of two item tuples, where the first
-    item is the aliased name and the second the actual name
-  * `context` - the context of the environment. It can be nil
-    (default context), inside a guard or inside an assign
-  * `requires` - the list of required modules
-  * `functions` - a list of functions imported from each module
-  * `macros` - a list of macros imported from each module
-  * `context_modules` - a list of modules defined in the current context
-  * `macro_aliases` - a list of aliases defined inside the current macro
-  * `vars` - a list keeping all defined varaibles as { var, context }
-  """
-
-  @type name_arity :: { atom, non_neg_integer }
-  @type file :: binary
-  @type line :: non_neg_integer
-  @type aliases :: [{ module, module }]
-  @type context :: :match | :guard | nil
-  @type requires :: [module]
-  @type functions :: [{ module, [name_arity] }]
-  @type macros :: [{ module, [name_arity] }]
-  @type context_modules :: [module]
-  @type vars :: [{ atom, atom }]
-  @type lexical_tracker :: pid
-
-  fields = [:module, :file, :line, :function, :aliases, :context, :requires, :functions,
-            :macros, :context_modules, :macro_aliases, :vars, :lexical_tracker]
-
-  types  = quote do: [module: module, file: file, line: line,
-    function: name_arity, aliases: aliases, requires: requires,
-    functions: functions, macros: macros, context_modules: context_modules,
-    macro_aliases: aliases, vars: vars, lexical_tracker: lexical_tracker]
-
-  Record.deffunctions(fields, __MODULE__)
-  Record.deftypes(fields, types, __MODULE__)
-
-  @doc """
-  Returns a keyword list containing the file and line
-  information as keys.
-  """
-  def location(record) do
-    [file: file(record), line: line(record)]
-  end
-
-  @doc """
-  Returns wether the compilation environment is currently
-  inside a guard.
-  """
-  def in_guard?(record), do: context(record) == :guard
-
-  @doc """
-  Returns wether the compilation environment is currently
-  inside a match clause.
-  """
-  def in_match?(record), do: context(record) == :match
-
-  @doc """
-  Returns the environment stacktrace.
-  """
-  def stacktrace(record) do
-    cond do
-      nil?(record.module) ->
-        [{ :elixir_compiler, :__FILE__, 2, location(record) }]
-      nil?(record.function) ->
-        [{ module(record), :__MODULE__, 0, location(record) }]
-      true ->
-        { name, arity } = record.function
-        [{ module(record), name, arity, location(record) }]
-    end
-  end
-end
@@ -1,930 +0,0 @@
-defmodule Module do
-  defmacrop is_env(env) do
-    quote do
-      is_tuple(unquote(env)) and size(unquote(env)) > 1 and elem(unquote(env), 0) == Macro.Env
-    end
-  end
-
-  @moduledoc %S'''
-  This module provides many functions to deal with modules during
-  compilation time. It allows a developer to dynamically attach
-  documentation, add, delete and register attributes and so forth.
-
-  After a module is compiled, using many of the functions in
-  this module will raise errors, since it is out of their purpose
-  to inspect runtime data. Most of the runtime data can be inspected
-  via the `__info__(attr)` function attached to each compiled module.
-
-  ## Module attributes
-
-  Each module can be decorated with one or more attributes. The following ones
-  are currently defined by Elixir:
-
-  * `@after_compile`
-
-      A hook that will be invoked right after the current module is compiled.
-
-      Accepts a module or a tuple `{ <module>, <function atom> }`. The function
-      must take two arguments: the module environment and its bytecode.
-      When just a module is provided, the function is assumed to be
-      `__after_compile__/2`.
-
-      ### Example
-
-          defmodule M do
-            @after_compile __MODULE__
-
-            def __after_compile__(env, _bytecode) do
-              IO.inspect env
-            end
-          end
-
-  * `@before_compile`
-
-      A hook that will be invoked before the module is compiled.
-
-      Accepts a module or a tuple `{ <module>, <function/macro atom> }`. The
-      function/macro must take one argument: the module environment. If it's a
-      macro, its returned value will be injected at the end of the module definition
-      before the compilation starts.
-
-      When just a module is provided, the function/macro is assumed to be
-      `__before_compile__/1`.
-
-      ### Example
-
-          defmodule M do
-            @before_compile __MODULE__
-
-            defmacro __before_compile__(_env) do
-              quote do
-                def hello, do: "world"
-              end
-            end
-          end
-
-  * `@behaviour`   (notice the british spelling)
-
-      Specify an OTP or user-defined behaviour.
-
-      ### Example
-
-          defmodule M do
-            @behaviour gen_event
-
-            # ...
-          end
-
-  * `@compile`
-
-      Define options for module compilation that are passed to the Erlang
-      compiler.
-
-      Accepts an atom, a tuple, or a list of atoms and tuples.
-
-      See http://www.erlang.org/doc/man/compile.html for the list of supported
-      options.
-
-      ### Example
-
-            defmodule M do
-              @compile { :inline, myfun: 1 }
-
-              def myfun(arg) do
-                to_string(arg)
-              end
-            end
-
-  * `@doc`
-
-      Provide documentation for the function or macro that follows the
-      attribute.
-
-      Accepts a string (often a heredoc) or `false` where `@doc false` will
-      make the function/macro invisible to the documentation extraction tools
-      like ExDoc.
-
-      Can be invoked more than once.
-
-      ### Example
-
-            defmodule M do
-              @doc "Hello world"
-              def hello do
-                "world"
-              end
-
-              @doc """
-              Sum.
-              """
-              def sum(a, b) do
-                a + b
-              end
-            end
-
-  * `@file`
-
-      Change the filename used in stacktraces for the function or macro that
-      follows the attribute.
-
-      Accepts a string. Can be used more than once.
-
-      ### Example
-
-            defmodule M do
-              @doc "Hello world"
-              @file "hello.ex"
-              def hello do
-                "world"
-              end
-            end
-
-  * `@moduledoc`
-
-      Provide documentation for the current module.
-
-      Accepts a string (which is often a heredoc) or `false` where
-      `@moduledoc false` will make the module invisible to the
-      documentation extraction tools like ExDoc.
-
-      ### Example
-
-            defmodule M do
-              @moduledoc """
-              A very useful module
-              """
-            end
-
-
-  * `@on_definition`
-
-      A hook that will be invoked after each function or macro in the current
-      module is defined. Useful when annotating functions.
-
-      Accepts a module or a tuple `{ <module>, <function atom> }`. The function
-      must take 6 arguments:
-
-        - the module environment
-        - kind: `:def`, `:defp`, `:defmacro`, or `:defmacrop`
-        - function/macro name
-        - list of quoted arguments
-        - list of quoted guards
-        - quoted function body
-
-      If the function/macro being defined has multiple clauses, the hook will
-      be called for each clause.
-
-      When just a module is provided, the function is assumed to be
-      `__on_definition__/6`.
-
-      Note that you can\'t provide the current module to `@on_definition`
-      because the hook function will not be defined in time. Finally, since
-      the `on_definition` hook is executed inside the context of the defined
-      function (i.e. `env.function` returns the current function), the hook
-      can only be a function, not a macro.
-
-      ### Example
-
-            defmodule H do
-              def on_def(_env, kind, name, args, guards, body) do
-                IO.puts "Defining #{kind} named #{name} with args:"
-                IO.inspect args
-                IO.puts "and guards"
-                IO.inspect guards
-                IO.puts "and body"
-                IO.puts Macro.to_string(body)
-              end
-            end
-
-            defmodule M do
-              @on_definition { H, :on_def }
-
-              def hello(arg) when is_binary(arg) or is_list(arg) do
-                "Hello" <> to_string(arg)
-              end
-
-              def hello(_) do
-                :ok
-              end
-            end
-
-  * `@on_load`
-
-      A hook that will be invoked whenever the module is loaded.
-
-      Accepts a function atom of a function in the current module. The function
-      must have arity 0 (no arguments) and has to return `:ok`, otherwise the
-      loading of the module will be aborted.
-
-      ### Example
-
-            defmodule M do
-              @on_load :load_check
-
-              def load_check do
-                if some_condition() do
-                  :ok
-                else
-                  nil
-                end
-              end
-
-              def some_condition do
-                false
-              end
-            end
-
-  * `@vsn`
-
-      Specify the module version. Accepts any valid Elixir value.
-
-      ### Example
-
-            defmodule M do
-              @vsn "1.0"
-            end
-
-  The following attributes are part of typespecs and are also reserved by
-  Elixir (see `Kernel.Typespec` for more information about typespecs):
-
-  * `@type`        - defines a type to be used in `@spec`
-  * `@typep`       - defines a private type to be used in `@spec`
-  * `@opaque`      - defines an opaque type to be used in `@spec`
-  * `@spec`        - provides a specification for a function
-  * `@callback`    - provides a specification for the behavior callback
-
-  In addition to the built-in attributes outlined above, custom attributes may
-  also be added. A custom attribute is any valid identifier prefixed with an
-  `@` and followed by a valid Elixir value:
-
-        defmodule M do
-          @custom_attr [some: "stuff"]
-        end
-
-  For more advanced options available when defining custom attributes, see
-  `register_attribute/3`.
-
-  ## Runtime information about a module
-
-  It is possible to query a module at runtime to find out which functions and
-  macros it defines, extract its docstrings, etc. See `__info__/1`.
-  '''
-
-  @doc """
-  Provides runtime information about functions and macros defined by the
-  module, enables docstring extraction, etc.
-
-  Each module gets an `__info__/1` function when it's compiled. The function
-  takes one of the following atoms:
-
-  * `:functions`  - keyword list of public functions along with their arities
-
-  * `:macros`     - keyword list of public macros along with their arities
-
-  * `:docs`       - list of all docstrings attached to functions and macros
-                    using the `@doc` attribute
-
-  * `:moduledoc`  - tuple `{ <line>, <doc> }` where `line` is the line on
-                    which module definition starts and `doc` is the string
-                    attached to the module using the `@moduledoc` attribute
-
-  * `:module`     - module name (`Module == Module.__info__(:module)`)
-
-  In addition to the above, you may also pass to `__info__/1` any atom supported
-  by Erlang's `module_info` function which also gets defined for each compiled
-  module. See http://erlang.org/doc/reference_manual/modules.html#id74571 for
-  more information.
-  """
-  def __info__(kind)
-
-  @doc """
-  Check if a module is open, i.e. it is currently being defined
-  and its attributes and functions can be modified.
-  """
-  def open?(module) do
-    table = data_table_for(module)
-    table == :ets.info(table, :name)
-  end
-
-  @doc """
-  Evaluates the quoted contents in the given module's context.
-
-  A list of environment options can also be given as argument.
-  See `Code.eval_string` for more information.
-
-  Raises an error if the module was already compiled.
-
-  ## Examples
-
-      defmodule Foo do
-        contents = quote do: (def sum(a, b), do: a + b)
-        Module.eval_quoted __MODULE__, contents
-      end
-
-      Foo.sum(1, 2) #=> 3
-
-  For convenience, you can my pass `__ENV__` as argument and
-  all options will be automatically extracted from the environment:
-
-      defmodule Foo do
-        contents = quote do: (def sum(a, b), do: a + b)
-        Module.eval_quoted __MODULE__, contents, [], __ENV__
-      end
-
-      Foo.sum(1, 2) #=> 3
-
-  """
-  def eval_quoted(module, quoted, binding // [], opts // [])
-
-  def eval_quoted(env, quoted, binding, opts) when is_env(env) do
-    eval_quoted(env.module, quoted, binding, Keyword.merge(env.to_keywords, opts))
-  end
-
-  def eval_quoted(module, quoted, binding, env) when is_env(env) do
-    eval_quoted(module, quoted, binding, env.to_keywords)
-  end
-
-  def eval_quoted(module, quoted, binding, opts) do
-    assert_not_compiled!(:eval_quoted, module)
-    :elixir_module.eval_quoted(module, quoted, binding, opts)
-  end
-
-  @doc """
-  Creates a module with the given name and given by
-  the given quoted expressions. The line where the module
-  is defined and its file can be given as options.
-
-  ## Examples
-
-      contents =
-        quote do
-          def world, do: true
-        end
-
-      Module.create(Hello, contents, __ENV__.location)
-
-      Hello.world #=> true
-
-  ## Differences with `defmodule`
-
-  `Module.create` works similarly to `defmodule` and
-  return the same results. While one could also use
-  `defmodule` to define modules dynamically, this
-  function is preferred when the module body is given
-  by a quoted expression.
-
-  Another important distinction is that `Module.create`
-  allows you to control the environment variables used
-  when defining the module, while `defmodule` automatically
-  shares the same environment.
-  """
-  def create(module, quoted, opts // [])
-
-  def create(module, quoted, env) when is_env(env) do
-    create(module, quoted, env.to_keywords)
-  end
-
-  def create(module, quoted, opts) when is_atom(module) do
-    line = Keyword.get(opts, :line, 1)
-    :elixir_module.compile(line, module, quoted, [], :elixir.scope_for_eval(opts))
-  end
-
-  @doc """
-  Concatenates the list of aliases and returns a new alias.
-  It handles char lists, binaries and atoms.
-
-  ## Examples
-
-      iex> Module.concat([Foo, Bar])
-      Foo.Bar
-      iex> Module.concat([Foo, "Bar"])
-      Foo.Bar
-      iex> Module.concat([Foo, 'Bar'])
-      Foo.Bar
-
-  """
-  def concat(list) when is_list(list) do
-    :elixir_aliases.concat(list)
-  end
-
-  @doc """
-  Concatenates the two given aliases and returns a new alias.
-  It handles char lists, binaries and atoms.
-
-  ## Examples
-
-      iex> Module.concat(Foo, Bar)
-      Foo.Bar
-      iex> Module.concat(Foo, "Bar")
-      Foo.Bar
-      iex> Module.concat(Foo, 'Bar')
-      Foo.Bar
-
-  """
-  def concat(left, right) do
-    :elixir_aliases.concat([left, right])
-  end
-
-  @doc """
-  Concatenates the list aliases and returns a new alias only
-  if the alias was already referenced. If the alias was not
-  referenced yet, fails with ArgumentError.
-  It handles char lists, binaries and atoms.
-
-  ## Examples
-
-      iex> Module.safe_concat([Unknown, Module])
-      ** (ArgumentError) argument error
-
-      iex> Module.safe_concat([List, Chars])
-      List.Chars
-
-  """
-  def safe_concat(list) when is_list(list) do
-    :elixir_aliases.safe_concat(list)
-  end
-
-  @doc """
-  Concatenates the two aliases and returns a new alias only
-  if the alias was already referenced. If the alias was not
-  referenced yet, fails with ArgumentError.
-  It handles char lists, binaries and atoms.
-
-  ## Examples
-
-      iex> Module.safe_concat(Unknown, Module)
-      ** (ArgumentError) argument error
-
-      iex> Module.safe_concat(List, Chars)
-      List.Chars
-
-  """
-  def safe_concat(left, right) do
-    :elixir_aliases.safe_concat([left, right])
-  end
-
-  @doc """
-  Gets an anonymous function from the given module, function
-  and arity. The module and function are not verified to exist.
-
-      iex> fun = Module.function(Kernel, :is_atom, 1)
-      iex> fun.(:hello)
-      true
-
-  """
-  def function(mod, fun, arity) do
-    :erlang.make_fun(mod, fun, arity)
-  end
-
-  @doc """
-  Attaches documentation to a given function or type. It expects
-  the module the function/type belongs to, the line (a non negative
-  integer), the kind (def or defmacro), a tuple representing
-  the function and its arity, the function signature (the signature
-  should be omitted for types) and the documentation, which should
-  be either a binary or a boolean.
-
-  ## Examples
-
-      defmodule MyModule do
-        Module.add_doc(__MODULE__, __ENV__.line + 1, :def, { :version, 0 }, [], "Manually added docs")
-        def version, do: 1
-      end
-
-  """
-  def add_doc(module, line, kind, tuple, signature // [], doc)
-
-  def add_doc(_module, _line, kind, _tuple, _signature, doc) when kind in [:defp, :defmacrop, :typep] do
-    if doc, do: { :error, :private_doc }, else: :ok
-  end
-
-  def add_doc(module, line, kind, tuple, signature, doc) when
-      kind in [:def, :defmacro, :type, :opaque] and (is_binary(doc) or is_boolean(doc) or doc == nil) do
-    assert_not_compiled!(:add_doc, module)
-    table = docs_table_for(module)
-
-    { signature, _ } = Enum.map_reduce signature, 1, fn(x, acc) ->
-      { simplify_signature(x, line, acc), acc + 1 }
-    end
-
-    case :ets.lookup(table, tuple) do
-      [] ->
-        :ets.insert(table, { tuple, line, kind, signature, doc })
-        :ok
-      [{ tuple, line, _old_kind, old_sign, old_doc }] ->
-        :ets.insert(table, {
-          tuple,
-          line,
-          kind,
-          merge_signatures(old_sign, signature, 1),
-          if(nil?(doc), do: old_doc, else: doc)
-        })
-        :ok
-    end
-  end
-
-  # Simplify signatures to be stored in docs
-
-  defp simplify_signature({ ://, defline, [left, right ] }, line, i) do
-    { ://, defline, [simplify_signature(left, line, i), right] }
-  end
-
-  defp simplify_signature({ var, line, atom }, _, _i) when is_atom(atom) do
-    case atom_to_list(var) do
-      [?_|_]    -> { var, line, :guess }
-      _         -> { var, line, nil }
-    end
-  end
-
-  defp simplify_signature({ :=, _, [_, right] }, line, i) do
-    simplify_signature(right, line, i)
-  end
-
-  defp simplify_signature(other, line, i) when is_integer(other), do: { :"int#{i}", line, :guess }
-  defp simplify_signature(other, line, i) when is_boolean(other), do: { :"bool#{i}", line, :guess }
-  defp simplify_signature(other, line, i) when is_atom(other),    do: { :"atom#{i}", line, :guess }
-  defp simplify_signature(other, line, i) when is_list(other),    do: { :"list#{i}", line, :guess }
-  defp simplify_signature(other, line, i) when is_float(other),   do: { :"float#{i}", line, :guess }
-  defp simplify_signature(other, line, i) when is_binary(other),  do: { :"binary#{i}", line, :guess }
-  defp simplify_signature(_, line, i), do: { :"arg#{i}", line, :guess }
-
-  # Merge
-
-  defp merge_signatures([h1|t1], [h2|t2], i) do
-    [merge_signature(h1, h2, i)|merge_signatures(t1, t2, i + 1)]
-  end
-
-  defp merge_signatures([], [], _) do
-    []
-  end
-
-  defp merge_signature({ ://, line, [left, right] }, newer, i) do
-    { ://, line, [merge_signature(left, newer, i), right] }
-  end
-
-  defp merge_signature(older, { ://, _, [left, _] }, i) do
-    merge_signature(older, left, i)
-  end
-
-  # The older signature, when given, always have higher precedence
-  defp merge_signature({ _, _, nil } = older, _newer, _),        do: older
-  defp merge_signature(_older, { _, _, nil } = newer, _),        do: newer
-
-  # Both are a guess, so check if they are the same guess
-  defp merge_signature({ var, _, _ } = older, { var, _, _ }, _), do: older
-
-  # Otherwise, returns a generic guess
-  defp merge_signature({ _, line, _ }, _newer, i), do: { :"arg#{i}", line, :guess }
-
-  @doc """
-  Checks if the module defines the given function or macro.
-  Use `defines?/3` to assert for an specific type.
-
-  ## Examples
-
-      defmodule Example do
-        Module.defines? __MODULE__, { :version, 0 } #=> false
-        def version, do: 1
-        Module.defines? __MODULE__, { :version, 0 } #=> true
-      end
-
-  """
-  def defines?(module, tuple) when is_tuple(tuple) do
-    assert_not_compiled!(:defines?, module)
-    table = function_table_for(module)
-    :ets.lookup(table, tuple) != []
-  end
-
-  @doc """
-  Checks if the module defines a function or macro with the
-  given `kind`. `kind` can be either `:def`, `:defp`,
-  `:defmacro` or `:defmacrop`.
-
-  ## Examples
-
-      defmodule Example do
-        Module.defines? __MODULE__, { :version, 0 }, :defp #=> false
-        def version, do: 1
-        Module.defines? __MODULE__, { :version, 0 }, :defp #=> false
-      end
-
-  """
-  def defines?(module, tuple, kind) do
-    assert_not_compiled!(:defines?, module)
-    table = function_table_for(module)
-    case :ets.lookup(table, tuple) do
-      [{ _, ^kind, _, _, _, _, _ }] -> true
-      _ -> false
-    end
-  end
-
-  @doc """
-  Return all functions defined in the given module.
-
-  ## Examples
-
-      defmodule Example do
-        def version, do: 1
-        Module.definitions_in __MODULE__ #=> [{:version,1}]
-      end
-
-  """
-  def definitions_in(module) do
-    assert_not_compiled!(:definitions_in, module)
-    table = function_table_for(module)
-    lc { tuple, _, _, _, _, _, _ } inlist :ets.tab2list(table), do: tuple
-  end
-
-  @doc """
-  Returns all functions defined in te given module according
-  to its kind.
-
-  ## Examples
-
-      defmodule Example do
-        def version, do: 1
-        Module.definitions_in __MODULE__, :def  #=> [{:version,1}]
-        Module.definitions_in __MODULE__, :defp #=> []
-      end
-
-  """
-  def definitions_in(module, kind) do
-    assert_not_compiled!(:definitions_in, module)
-    table = function_table_for(module)
-    lc { tuple, stored_kind, _, _, _, _, _ } inlist :ets.tab2list(table), stored_kind == kind, do: tuple
-  end
-
-  @doc """
-  Makes the given functions in the given module overridable.
-  An overridable function is lazily defined, allowing a
-  developer to customize it. See `Kernel.defoverridable` for
-  more information and documentation.
-  """
-  def make_overridable(module, tuples) do
-    assert_not_compiled!(:make_overridable, module)
-
-    lc tuple inlist tuples do
-      case :elixir_def.lookup_definition(module, tuple) do
-        false ->
-          { name, arity } = tuple
-          raise "Cannot make function #{name}/#{arity} overridable because it was not defined"
-        clause ->
-          :elixir_def.delete_definition(module, tuple)
-          neighbours = Module.DispatchTracker.yank(module, tuple)
-
-          old    = get_attribute(module, :__overridable)
-          merged = :orddict.update(tuple, fn({ count, _, _, _ }) ->
-            { count + 1, clause, neighbours, false }
-          end, { 1, clause, neighbours, false }, old)
-
-          put_attribute(module, :__overridable, merged)
-      end
-    end
-  end
-
-  @doc """
-  Returns true if the given tuple in module is marked as overridable.
-  """
-  def overridable?(module, tuple) do
-    !! List.keyfind(get_attribute(module, :__overridable), tuple, 0)
-  end
-
-  @doc """
-  Puts an Erlang attribute to the given module with the given
-  key and value. The semantics of putting the attribute depends
-  if the attribute was registered or not via `register_attribute/2`.
-
-  ## Examples
-
-      defmodule MyModule do
-        Module.put_attribute __MODULE__, :custom_threshold_for_lib, 10
-      end
-
-  """
-  def put_attribute(module, key, value) when is_atom(key) do
-    assert_not_compiled!(:put_attribute, module)
-    table = data_table_for(module)
-    value = normalize_attribute(key, value)
-    acc   = :ets.lookup_element(table, :__acc_attributes, 2)
-
-    new =
-      if :lists.member(key, acc) do
-        case :ets.lookup(table, key) do
-          [{^key, old}] -> [value|old]
-          [] -> [value]
-        end
-      else
-        value
-      end
-
-    :ets.insert(table, { key, new })
-  end
-
-  @doc """
-  Gets the given attribute from a module. If the attribute
-  was marked with `accumulate` with `Module.register_attribute`,
-  a list is always returned.
-
-  The `@` macro compiles to a call to this function. For example,
-  the following code:
-
-      @foo
-
-  Expands to:
-
-      Module.get_attribute(__MODULE__, :foo, true)
-
-  Notice the third argument is used to indicate if a warning
-  should be emitted when the attribute was not previously defined.
-  This is true for `@foo` attributes but false for direct calls.
-
-  ## Examples
-
-      defmodule Foo do
-        Module.put_attribute __MODULE__, :value, 1
-        Module.get_attribute __MODULE__, :value #=> 1
-
-        Module.register_attribute __MODULE__, :value, accumulate: true
-        Module.put_attribute __MODULE__, :value, 1
-        Module.get_attribute __MODULE__, :value #=> [1]
-      end
-
-  """
-  def get_attribute(module, key, warn // false) when is_atom(key) do
-    assert_not_compiled!(:get_attribute, module)
-    table = data_table_for(module)
-
-    case :ets.lookup(table, key) do
-      [{^key, val}] -> val
-      [] ->
-        acc = :ets.lookup_element(table, :__acc_attributes, 2)
-
-        if :lists.member(key, acc) do
-          []
-        else
-          warn && :elixir_errors.warn "#{Exception.format_caller} undefined module attribute @#{key}, " <>
-            "please remove access to @#{key} or explicitly set it to nil before access\n"
-          nil
-        end
-    end
-  end
-
-  @doc """
-  Deletes all attributes that matches the given key.
-
-  ## Examples
-
-      defmodule MyModule do
-        Module.put_attribute __MODULE__, :custom_threshold_for_lib, 10
-        Module.delete_attribute __MODULE__, :custom_threshold_for_lib
-      end
-
-  """
-  def delete_attribute(module, key) when is_atom(key) do
-    assert_not_compiled!(:delete_attribute, module)
-    table = data_table_for(module)
-    :ets.delete(table, key)
-  end
-
-  @doc """
-  Registers an attribute. By registering an attribute, a developer
-  is able to customize how Elixir will store and accumulate the
-  attribute values.
-
-  ## Options
-
-  When registering an attribute, two options can be given:
-
-  * `:accumulate` - Several calls to the same attribute will
-    accumulate instead of override the previous one. New attributes
-    are always added to the top of the accumulated list.
-
-  * `:persist` - The attribute will be persisted in the Erlang
-    Abstract Format. Useful when interfacing with Erlang libraries.
-
-  By default, both options are false.
-
-  ## Examples
-
-      defmodule MyModule do
-        Module.register_attribute __MODULE__,
-          :custom_threshold_for_lib,
-          accumulate: true, persist: false
-
-        @custom_threshold_for_lib 10
-        @custom_threshold_for_lib 20
-        @custom_threshold_for_lib #=> [20, 10]
-      end
-
-  """
-  def register_attribute(module, new, opts) when is_atom(new) do
-    assert_not_compiled!(:register_attribute, module)
-    table = data_table_for(module)
-
-    if Keyword.get(opts, :persist) do
-      old = :ets.lookup_element(table, :__persisted_attributes, 2)
-      :ets.insert(table, { :__persisted_attributes,  [new|old] })
-    end
-
-    if Keyword.get(opts, :accumulate) do
-      old = :ets.lookup_element(table, :__acc_attributes, 2)
-      :ets.insert(table, { :__acc_attributes,  [new|old] })
-    end
-  end
-
-  @doc """
-  Split the given module name into binary parts.
-
-  ## Examples
-
-      Module.split Very.Long.Module.Name.And.Even.Longer
-      #=> ["Very", "Long", "Module", "Name", "And", "Even", "Longer"]
-
-  """
-  def split(module) do
-    tl(String.split(String.Chars.to_string(module), "."))
-  end
-
-  @doc false
-  # Used internally to compile documentation. This function
-  # is private and must be used only internally.
-  def compile_doc(env, kind, name, args, _guards, _body) do
-    module = env.module
-    line   = env.line
-    arity  = length(args)
-    pair   = { name, arity }
-    doc    = get_attribute(module, :doc)
-
-    case add_doc(module, line, kind, pair, args, doc) do
-      :ok ->
-        :ok
-      { :error, :private_doc } ->
-        :elixir_errors.warn "#{env.file}:#{line} function #{name}/#{arity} is private, @doc's are always discarded for private functions\n"
-    end
-
-    delete_attribute(module, :doc)
-  end
-
-  @doc false
-  # Used internally to compile types. This function
-  # is private and must be used only internally.
-  def compile_typespec(module, key, value) when is_atom(key) do
-    assert_not_compiled!(:put_attribute, module)
-    table = data_table_for(module)
-
-    new =
-      case :ets.lookup(table, key) do
-        [{^key, old}] -> [value|old]
-        [] -> [value]
-      end
-
-    :ets.insert(table, { key, new })
-  end
-
-  ## Helpers
-
-  defp normalize_attribute(:on_load, atom) when is_atom(atom) do
-    { atom, 0 }
-  end
-
-  defp normalize_attribute(kind, atom) when kind in [:behavior, :behaviour] and is_atom(atom) do
-    Code.ensure_compiled(atom)
-    atom
-  end
-
-  defp normalize_attribute(:file, file) when is_binary(file) do
-    file
-  end
-
-  defp normalize_attribute(key, atom) when is_atom(atom) and
-      key in [:before_compile, :after_compile, :on_definition] do
-    { atom, :"__#{key}__" }
-  end
-
-  defp normalize_attribute(key, _value) when key in [:type, :typep, :export_type, :opaque, :callback] do
-    raise ArgumentError, message: "attributes type, typep, export_type, opaque and callback " <>
-      "must be set via Kernel.Typespec"
-  end
-
-  defp normalize_attribute(_key, value) do
-    value
-  end
-
-  defp data_table_for(module) do
-    module
-  end
-
-  defp function_table_for(module) do
-    :elixir_def.table(module)
-  end
-
-  defp docs_table_for(module) do
-    :elixir_module.docs_table(module)
-  end
-
-  defp assert_not_compiled!(fun, module) do
-    open?(module) ||
-      raise ArgumentError,
-        message: "could not call #{fun} on module #{inspect module} because it was already compiled"
-  end
-end
@@ -1,364 +0,0 @@
-# This is a module Elixir responsible for tracking
-# calls in order to extract Elixir modules' behaviour
-# during compilation time.
-#
-# ## Implementation
-#
-# The implementation uses the digraph module to track
-# all dependencies. The graph starts with one main vertice:
-#
-# * `:local` - points to local functions
-#
-# We also have can the following vertices:
-#
-# * `Module` - a module that was invoked via an import or remotely
-# * `{ name, arity }` - a local function/arity pair
-# * `{ :import, name, arity }` - an invoked function/arity import
-# * `{ :remote, name, arity }` - an remotely invoked function/arity
-#
-# Each of those vertices can associate to other vertices
-# as described below:
-#
-# * `Module`
-#   * in neighbours:  `{ :import, name, arity }` and `{ :remote, name arity }`
-#
-# * `{ name, arity }`
-#   * in neighbours: `:local`, `{ name, arity }`
-#   * out neighbours: `{ :import, name, arity }` and `{ :remote, name arity }`
-#
-# * `{ :import, name, arity }`
-#   * in neighbours: `{ name, arity }`
-#   * out neighbours: `Module`
-#
-# * `{ :remote, name, arity }`
-#   * in neighbours: `{ name, arity }`
-#   * out neighbours: `Module`
-#
-# Note that since this is required for bootstrap, we can't use
-# any of the `GenServer.Behaviour` conveniences.
-defmodule Module.DispatchTracker do
-  @moduledoc false
-
-  @timeout  30_000
-  @behavior :gen_server
-
-  @type ref :: pid | module
-  @type name :: atom
-  @type name_arity :: { name, arity }
-
-  @type local :: { name, arity }
-  @type import :: { :import, name, arity }
-  @type remote :: { :remote, name, arity }
-
-  # Public API
-
-  @doc """
-  Receives a dispatch or a module and returns all dispatches
-  that calls it.
-
-  In case the argument is a module, the response will be
-  made by import and remote dispatches.
-
-  In case the argument is another dispatch, the response
-  will be made by local dispatches.
-
-  This function is not recursive, so if A dispatches to
-  B which dispatches to C, A does not appear in the result,
-  only B.
-  """
-  @spec dispatches_to(ref, module) :: [import | remote]
-  @spec dispatches_to(ref, local | import | remote) :: [local]
-  def dispatches_to(ref, dispatch) do
-    d = :gen_server.call(to_pid(ref), :digraph, @timeout)
-    :digraph.in_neighbours(d, dispatch) |> only_tuples
-  end
-
-  @doc """
-  Receives a local and returns all dispatches from that local.
-
-  This function is not recursive, so if A dispatches to
-  B which dispatches to C, C does not appear in the result,
-  only B.
-  """
-  @spec dispatches_from(ref, local) :: [local | import | remote]
-  def dispatches_from(ref, { name, arity }) do
-    d = :gen_server.call(to_pid(ref), :digraph, @timeout)
-    :digraph.out_neighbours(d, { name, arity }) |> only_tuples
-  end
-
-  @doc """
-  Returns all imported modules that had the given
-  `{ name, arity }` invoked.
-  """
-  @spec imports_with_dispatch(ref, name_arity) :: [module]
-  def imports_with_dispatch(ref, { name, arity }) do
-    d = :gen_server.call(to_pid(ref), :digraph, @timeout)
-    :digraph.out_neighbours(d, { :import, name, arity })
-  end
-
-  @doc """
-  Returns all modules that had the given `{ name, arity }`
-  invoked remotely.
-  """
-  @spec remotes_with_dispatch(ref, name_arity) :: [module]
-  def remotes_with_dispatch(ref, { name, arity }) do
-    d = :gen_server.call(to_pid(ref), :digraph, @timeout)
-    :digraph.out_neighbours(d, { :remote, name, arity })
-  end
-
-  @doc """
-  Returns all locals that are reachable.
-
-  By default, all public functions are reachable.
-  A private function is only reachable if it has
-  a public function that it invokes directly.
-  """
-  @spec reachable(ref) :: [local]
-  def reachable(ref) do
-    d = :gen_server.call(to_pid(ref), :digraph, @timeout)
-    reduce_reachable(d, :local, [])
-  end
-
-  defp reduce_reachable(d, vertex, vertices) do
-    neighbours = :digraph.out_neighbours(d, vertex)
-    neighbours = (lc { _, _ } = t inlist neighbours, do: t) |> :ordsets.from_list
-    remaining  = :ordsets.subtract(neighbours, vertices)
-    vertices   = :ordsets.union(neighbours, vertices)
-    :lists.foldl(&reduce_reachable(d, &1, &2), vertices, remaining)
-  end
-
-  defp to_pid(pid) when is_pid(pid),  do: pid
-  defp to_pid(mod) when is_atom(mod) do
-    table = :elixir_module.data_table(mod)
-    [{ _, val }] = :ets.lookup(table, :__dispatch_tracker)
-    val
-  end
-
-  defp only_tuples(list) do
-    lc x inlist list, is_tuple(x), do: x
-  end
-
-  # Internal API
-
-  # Starts the tracker and returns its pid.
-  @doc false
-  def start_link do
-    { :ok, pid } = :gen_server.start_link(__MODULE__, [], [])
-    pid
-  end
-
-  # Adds a definition into the tracker. A public
-  # definition is connected with the :local node
-  # while a private one is left unreachable until
-  # a call is made to.
-  @doc false
-  def add_definition(pid, kind, tuple) when kind in [:def, :defp, :defmacro, :defmacrop] do
-    :gen_server.cast(pid, { :add_definition, kind, tuple })
-  end
-
-  # Adds and tracks defaults for a definition into the tracker.
-  @doc false
-  def add_defaults(pid, kind, tuple, defaults) when kind in [:def, :defp, :defmacro, :defmacrop] do
-    :gen_server.cast(pid, { :add_defaults, kind, tuple, defaults })
-  end
-
-  # Adds a local dispatch to the given target.
-  def add_local(pid, to) when is_tuple(to) do
-    :gen_server.cast(pid, { :add_local, :local, to })
-  end
-
-  # Adds a local dispatch from-to the given target.
-  @doc false
-  def add_local(pid, from, to) when is_tuple(from) and is_tuple(to) do
-    :gen_server.cast(pid, { :add_local, from, to })
-  end
-
-  # Adds a remote dispatch to the given target.
-  @doc false
-  def add_remote(pid, function, module, target) when is_atom(module) and is_tuple(target) do
-    :gen_server.cast(pid, { :add_external, :remote, function, module, target })
-  end
-
-  # Adds a import dispatch to the given target.
-  @doc false
-  def add_import(pid, function, module, target) when is_atom(module) and is_tuple(target) do
-    :gen_server.cast(pid, { :add_external, :import, function, module, target })
-  end
-
-  # Yanks a local node. Returns its in and out vertices in a tuple.
-  @doc false
-  def yank(pid, local) do
-    :gen_server.call(to_pid(pid), { :yank, local }, @timeout)
-  end
-
-  # Reattach a previously yanked node
-  @doc false
-  def reattach(pid, kind, tuple, neighbours) do
-    pid = to_pid(pid)
-    add_definition(pid, kind, tuple)
-    :gen_server.cast(pid, { :reattach, tuple, neighbours })
-  end
-
-  # Collecting all conflicting imports with the given functions
-  @doc false
-  def collect_imports_conflicts(pid, all_defined) do
-    d = :gen_server.call(pid, :digraph, @timeout)
-
-    lc { name, arity } inlist all_defined,
-       n = :digraph.out_neighbours(d, { :import, name, arity }),
-       n != [] do
-      { n, name, arity }
-    end
-  end
-
-  # Collect all unused definitions based on the private
-  # given also accounting the expected amount of default
-  # clauses a private function have.
-  @doc false
-  def collect_unused_locals(pid, private) do
-    reachable = reachable(pid)
-    :lists.foldl(&collect_unused_locals(&1, &2, reachable), [], private)
-  end
-
-  defp collect_unused_locals({ tuple, kind, 0 }, acc, reachable) do
-    if :lists.member(tuple, reachable) do
-      acc
-    else
-      [{ :unused_def, tuple, kind }|acc]
-    end
-  end
-
-  defp collect_unused_locals({ tuple, kind, default }, acc, reachable) when default > 0 do
-    { name, arity } = tuple
-    min = arity - default
-    max = arity
-
-    invoked = lc { n, a } inlist reachable, n == name, a in min..max, do: a
-
-    if invoked == [] do
-      [{ :unused_def, tuple, kind }|acc]
-    else
-      case :lists.min(invoked) - min do
-        0 -> acc
-        ^default -> [{ :unused_args, tuple }|acc]
-        unused_args -> [{ :unused_args, tuple, unused_args }|acc]
-      end
-    end
-  end
-
-  # Stops the gen server
-  @doc false
-  def stop(pid) do
-    :gen_server.cast(pid, :stop)
-  end
-
-  # Callbacks
-
-  def init([]) do
-    d = :digraph.new([:protected])
-    :digraph.add_vertex(d, :local)
-    { :ok, d }
-  end
-
-  def handle_call({ :yank, local }, _from, d) do
-    in_vertices  = :digraph.in_neighbours(d, local)
-    out_vertices = :digraph.out_neighbours(d, local)
-    :digraph.del_vertex(d, local)
-    { :reply, { in_vertices, out_vertices }, d }
-  end
-
-  def handle_call(:digraph, _from, d) do
-    { :reply, d, d }
-  end
-
-  def handle_call(_request, _from, d) do
-    { :noreply, d }
-  end
-
-  def handle_info(_msg, d) do
-    { :noreply, d }
-  end
-
-  def handle_cast({ :add_local, from, to }, d) do
-    handle_add_local(d, from, to)
-    { :noreply, d }
-  end
-
-  def handle_cast({ :add_external, kind, function, module, { name, arity } }, d) do
-    handle_import_or_remote(d, kind, function, module, name, arity)
-    { :noreply, d }
-  end
-
-  def handle_cast({ :add_definition, kind, tuple }, d) do
-    handle_add_definition(d, kind, tuple)
-    { :noreply, d }
-  end
-
-  def handle_cast({ :add_defaults, kind, { name, arity }, defaults }, d) do
-    lc i inlist :lists.seq(arity - defaults, arity - 1) do
-      handle_add_definition(d, kind, { name, i })
-      handle_add_local(d, { name, i }, { name, i + 1 })
-    end
-    { :noreply, d }
-  end
-
-  def handle_cast({ :reattach, tuple, { in_neigh, out_neigh } }, d) do
-    lc from inlist in_neigh, do: replace_edge(d, from, tuple)
-    lc to inlist out_neigh,  do: replace_edge(d, tuple, to)
-    { :noreply, d }
-  end
-
-  def handle_cast(:stop, d) do
-    { :stop, :normal, d }
-  end
-
-  def handle_cast(_msg, d) do
-    { :noreply, d }
-  end
-
-  def terminate(_reason, _d) do
-    :ok
-  end
-
-  def code_change(_old, d, _extra) do
-    { :ok, d }
-  end
-
-  defp handle_import_or_remote(d, kind, function, module, name, arity) do
-    :digraph.add_vertex(d, module)
-
-    tuple = { kind, name, arity }
-    :digraph.add_vertex(d, tuple)
-    replace_edge!(d, tuple, module)
-
-    if function != nil do
-      replace_edge!(d, function, tuple)
-    end
-  end
-
-  defp handle_add_local(d, from, to) do
-    :digraph.add_vertex(d, to)
-    replace_edge!(d, from, to)
-  end
-
-  defp handle_add_definition(d, public, tuple) when public in [:def, :defmacro] do
-    :digraph.add_vertex(d, tuple)
-    replace_edge!(d, :local, tuple)
-  end
-
-  defp handle_add_definition(d, private, tuple) when private in [:defp, :defmacrop] do
-    :digraph.add_vertex(d, tuple)
-  end
-
-  defp replace_edge!(d, from, to) do
-    unless :lists.member(to, :digraph.out_neighbours(d, from)) do
-      [:"$e"|_] = :digraph.add_edge(d, from, to)
-    end
-  end
-
-  defp replace_edge(d, from, to) do
-    unless :lists.member(to, :digraph.out_neighbours(d, from)) do
-      :digraph.add_edge(d, from, to)
-    end
-  end
-end
@@ -1,186 +0,0 @@
-defmodule Node do
-  @moduledoc """
-  Functions related to Erlang nodes.
-  """
-
-  @type t :: atom
-
-  @doc """
-  Returns the current node. It returns the same as the built-in `node()`.
-  """
-  @spec self :: t
-  def self do
-    :erlang.node()
-  end
-
-  @doc """
-  Returns `true` if the local node is alive; that is, if the node can be
-  part of a distributed system. Otherwise, it returns `false`.
-  """
-  @spec alive? :: boolean
-  def alive? do
-    :erlang.is_alive()
-  end
-
-  @doc """
-  Returns a list of all visible nodes in the system, excluding
-  the local node. Same as `list(:visible)`.
-  """
-  @spec list :: [t]
-  def list do
-    :erlang.nodes()
-  end
-
-  @doc """
-  Returns a list of nodes according to argument given. The result
-  returned when the argument is a list, is the list of nodes
-  satisfying the disjunction(s) of the list elements.
-
-  See http://www.erlang.org/doc/man/erlang.html#nodes-1 for more info.
-  """
-  @typep list_arg :: :visible | :hidden | :connected | :this | :known
-  @spec list(list_arg | [list_arg]) :: [t]
-  def list(args) do
-    :erlang.nodes(args)
-  end
-
-  @doc """
-  Monitors the status of the node. If `flag` is `true`, monitoring is
-  turned on. If `flag` is `false`, monitoring is turned off.
-
-  See http://www.erlang.org/doc/man/erlang.html#monitor_node-2 for more info.
-  """
-  @spec monitor(t, boolean) :: true
-  def monitor(node, flag) do
-    :erlang.monitor_node(node, flag)
-  end
-
-  @doc """
-  Behaves as `monitor/2` except that it allows an extra
-  option to be given, namely `:allow_passive_connect`.
-
-  See http://www.erlang.org/doc/man/erlang.html#monitor_node-3 for more info.
-  """
-  @spec monitor(t, boolean, [:allow_passive_connect]) :: true
-  def monitor(node, flag, options) do
-    :erlang.monitor_node(node, flag, options)
-  end
-
-  @doc """
-  Forces the disconnection of a node. This will appear to the `node` as if
-  the local node has crashed. This BIF is mainly used in the Erlang network
-  authentication protocols. Returns `true` if disconnection succeeds, otherwise
-  `false`. If the local node is not alive, the function returns `:ignored`.
-
-  See http://www.erlang.org/doc/man/erlang.html#disconnect_node-1 for more info.
-  """
-  @spec disconnect(t) :: boolean | :ignored
-  def disconnect(node) do
-    :erlang.disconnect_node(node)
-  end
-
-  @doc """
-  Establishes a connection to `node`. Returns `true` if successful,
-  `false` if not, and the atom `:ignored` if the local node is not
-  alive.
-
-  See http://erlang.org/doc/man/net_kernel.html#connect_node-1 for more info.
-  """
-  @spec connect(t) :: boolean | :ignored
-  def connect(node) do
-    :net_kernel.connect_node(node)
-  end
-
-  @doc """
-  Returns the pid of a new process started by the application of `fun`
-  on `node`. If `node` does not exist, a useless pid is returned.
-
-  Check http://www.erlang.org/doc/man/erlang.html#spawn-2 for
-  the list of available options.
-  """
-  @spec spawn(t, (() -> any)) :: pid
-  def spawn(node, fun) do
-    :erlang.spawn(node, fun)
-  end
-
-  @doc """
-  Returns the pid of a new process started by the application of `fun`
-  on `node`. If `node` does not exist, a useless pid is returned.
-
-  Check http://www.erlang.org/doc/man/erlang.html#spawn_opt-3 for
-  the list of available options.
-  """
-  @spec spawn(t, (() -> any), Process.spawn_opts) :: pid | {pid, reference}
-  def spawn(node, fun, opts) do
-    :erlang.spawn_opt(node, fun, opts)
-  end
-
-  @doc """
-  Returns the pid of a new process started by the application of
-  `module.function(args)` on `node`. If `node` does not exists,
-  a useless pid is returned.
-
-  Check http://www.erlang.org/doc/man/erlang.html#spawn-4 for
-  the list of available options.
-  """
-  @spec spawn(t, module, atom, [any]) :: pid
-  def spawn(node, module, fun, args) do
-    :erlang.spawn(node, module, fun, args)
-  end
-
-  @doc """
-  Returns the pid of a new process started by the application of
-  `module.function(args)` on `node`. If `node` does not exists,
-  a useless pid is returned.
-
-  Check http://www.erlang.org/doc/man/erlang.html#spawn_opt-5 for
-  the list of available options.
-  """
-  @spec spawn(t, module, atom, [any], Process.spawn_opts) :: pid | {pid, reference}
-  def spawn(node, module, fun, args, opts) do
-    :erlang.spawn_opt(node, module, fun, args, opts)
-  end
-
-  @doc """
-  Returns the pid of a new process started by the application of `fun`
-  on `node`. A link is created between the calling process and the
-  new process, atomically. If `node` does not exist, a useless pid is returned
-  (and due to the link, an exit signal with exit reason `:noconnection` will be
-  received).
-  """
-  @spec spawn_link(t, (() -> any)) :: pid
-  def spawn_link(node, fun) do
-    :erlang.spawn_link(node, fun)
-  end
-
-  @doc """
-  Returns the pid of a new process started by the application of
-  `module.function(args)` on `node`. A link is created between the calling
-  process and the new process, atomically. If `node` does not exist, a useless
-  pid is returned (and due to the link, an exit signal with exit reason
-  `:noconnection` will be received).
-  """
-  @spec spawn_link(t, module, atom, [any]) :: pid
-  def spawn_link(node, module, fun, args) do
-    :erlang.spawn_link(node, module, fun, args)
-  end
-
-  @doc """
-  Sets the magic cookie of `node` to the atom `cookie`. The default node
-  is `Node.self`, the local node. If `node` is the local node, the function also
-  sets the cookie of all other unknown nodes to `cookie`.
-
-  This function will raise `FunctionClauseError` if the given `node` is not alive.
-  """
-  def set_cookie(node // Node.self, cookie) when is_atom(cookie) do
-    :erlang.set_cookie(node, cookie)
-  end
-
-  @doc """
-  Returns the magic cookie of the local node, if the node is alive;
-  otherwise `:nocookie`.
-  """
-  def get_cookie() do
-    :erlang.get_cookie()
-  end
-end
@@ -1,227 +0,0 @@
-defmodule OptionParser do
-  @moduledoc """
-  This module contains functions to parse command line arguments.
-  """
-
-  @doc """
-  Parses `argv` and returns a tuple with the parsed options, its
-  arguments, and a list options that couldn't be parsed.
-
-  ## Examples
-
-      iex> OptionParser.parse(["--debug"])
-      { [debug: true], [], [] }
-
-      iex> OptionParser.parse(["--source", "lib"])
-      { [source: "lib"], [], [] }
-
-      iex> OptionParser.parse(["--source-path", "lib", "test/enum_test.exs", "--verbose"])
-      { [source_path: "lib", verbose: true], ["test/enum_test.exs"], [] }
-
-  Notice how Elixir automatically translates the "--source-path"
-  switch to the underscored atom `:source_path`, which better follows
-  Elixir conventions.
-
-  ## Aliases
-
-  A set of aliases can be given as the second argument:
-
-      iex> OptionParser.parse(["-d"], aliases: [d: :debug])
-      { [debug: true], [], [] }
-
-  ## Switches
-
-  Extra information about switches can be given as arguments, too.
-  This is useful when a switch must behave as a boolean
-  or if duplicated switches should be kept, overriden or accumulated.
-
-  The following types are supported:
-
-  * `:boolean` - Marks the given switch as a boolean. Boolean switches
-                 never consume the following value unless it is
-                 `true` or `false`;
-  * `:integer` - Parses the switch as an integer;
-  * `:float`   - Parses the switch as a float;
-
-  If a switch can't be parsed, the option is returned in the invalid
-  options list (third element of the returned tuple).
-
-  The following extra options are supported:
-
-  * `:keep` - Keeps duplicated items in the list instead of overriding;
-
-  Examples:
-
-      iex> OptionParser.parse(["--unlock", "path/to/file"], switches: [unlock: :boolean])
-      { [unlock: true], ["path/to/file"], [] }
-
-      iex> OptionParser.parse(["--unlock", "--limit", "0", "path/to/file"],
-      ...>                    switches: [unlock: :boolean, limit: :integer])
-      { [unlock: true, limit: 0], ["path/to/file"], [] }
-
-      iex> OptionParser.parse(["-limit", "3"], switches: [limit: :integer])
-      { [limit: 3], [], [] }
-
-      iex> OptionParser.parse(["-limit", "yyz"], switches: [limit: :integer])
-      { [], [], [limit: "yyz"] }
-
-  ## Negation switches
-
-  Any switches starting with `--no-` are always considered to be
-  booleans and never parse the next value:
-
-      iex> OptionParser.parse(["--no-op", "path/to/file"])
-      { [no_op: true], ["path/to/file"], [] }
-
-  In case the negated switch exists as a boolean, it sets the boolean to false:
-
-      iex> OptionParser.parse(["--no-op", "path/to/file"], switches: [op: :boolean])
-      { [op: false], ["path/to/file"], [] }
-
-  """
-  def parse(argv, opts // []) when is_list(argv) and is_list(opts) do
-    parse(argv, opts, true)
-  end
-
-  @doc """
-  Similar to `parse/2` but only parses the head of `argv`;
-  as soon as it finds a non-switch, it stops parsing.
-
-  See `parse/2` for more information.
-
-  ## Example
-
-      iex> OptionParser.parse_head(["--source", "lib", "test/enum_test.exs", "--verbose"])
-      { [source: "lib"], ["test/enum_test.exs", "--verbose"], [] }
-
-      iex> OptionParser.parse_head(["--verbose", "--source", "lib", "test/enum_test.exs", "--unlock"])
-      { [verbose: true, source: "lib"], ["test/enum_test.exs", "--unlock"], [] }
-  """
-  def parse_head(argv, opts // []) when is_list(argv) and is_list(opts) do
-    parse(argv, opts, false)
-  end
-
-  ## Helpers
-
-  defp parse(argv, opts, bool) do
-    aliases  = opts[:aliases]  || []
-    switches = opts[:switches] || []
-    parse(argv, aliases, switches, bool)
-  end
-
-  defp parse(argv, aliases, switches, all) do
-    parse(argv, aliases, switches, [], [], [], all)
-  end
-
-  defp parse(["--"|_] = value, _aliases, _switches, dict, _args, invalid, _all) do
-    { Enum.reverse(dict), value, Enum.reverse(invalid) }
-  end
-
-  defp parse(["-" <> option|t], aliases, switches, dict, args, invalid, all) do
-    { option, kinds, value } = normalize_option(option, switches, aliases)
-
-    if nil?(value) do
-      { value, t } =
-        if :boolean in kinds do
-          { true, t }
-        else
-          value_from_tail(t)
-        end
-    end
-
-    { dict, invalid } = store_option(dict, invalid, option, value, kinds)
-
-    parse(t, aliases, switches, dict, args, invalid, all)
-  end
-
-  defp parse([h|t], aliases, switches, dict, args, invalid, true) do
-    parse(t, aliases, switches, dict, [h|args], invalid, true)
-  end
-
-  defp parse([], _, _switches, dict, args, invalid, true) do
-    { Enum.reverse(dict), Enum.reverse(args), Enum.reverse(invalid) }
-  end
-
-  defp parse(value, _, _switches, dict, _args, invalid, false) do
-    { Enum.reverse(dict), value, Enum.reverse(invalid) }
-  end
-
-  defp value_from_tail(["-" <> _|_] = t), do: { true, t }
-  defp value_from_tail([h|t]),            do: { h, t }
-  defp value_from_tail([]),               do: { true, [] }
-
-  defp store_option(dict, invalid, option, value, kinds) do
-    { invalid_option, value } =
-      cond do
-        :boolean in kinds ->
-          { nil, value in [true, "true"] }
-        :integer in kinds ->
-          case Integer.parse(value) do
-            { value, "" } -> { nil, value }
-            _ -> { option, value }
-          end
-        :float in kinds ->
-          case Float.parse(value) do
-            { value, "" } -> { nil, value }
-            _ -> { option, value }
-          end
-        true ->
-          { nil, value }
-      end
-
-    if invalid_option do
-      { dict, [{ option, value }|invalid] }
-    else
-      { do_store_option(dict, option, value, kinds), invalid }
-    end
-  end
-
-  defp do_store_option(dict, option, value, kinds) do
-    cond do
-      :keep in kinds ->
-        [{ option, value }|dict]
-      true ->
-        [{ option, value }|Keyword.delete(dict, option)]
-    end
-  end
-
-  defp normalize_option(<<?-, option :: binary>>, switches, aliases) do
-    normalize_option(option, switches, aliases)
-  end
-
-  defp normalize_option(option, switches, aliases) do
-    { option, value } = split_option(option)
-
-    if non_neg = get_non_negated(option, aliases) do
-      kinds = List.wrap(switches[non_neg])
-
-      if :boolean in kinds do
-        { non_neg, kinds, false }
-      else
-        { get_aliased(option, aliases), [:boolean], true }
-      end
-    else
-      atom = get_aliased(option, aliases)
-      { atom, List.wrap(switches[atom]), value }
-    end
-  end
-
-  defp split_option(option) do
-    case :binary.split(option, "=") do
-      [h]    -> { h, nil }
-      [h, t] -> { h, t }
-    end
-  end
-
-  defp to_underscore(option) do
-    bc <<c>> inbits option, do: << if(c == ?-, do: ?_, else: c) >>
-  end
-
-  defp get_aliased(option, aliases) do
-    atom = option |> to_underscore |> binary_to_atom
-    aliases[atom] || atom
-  end
-
-  defp get_non_negated("no-" <> rest, aliases), do: get_aliased(rest, aliases)
-  defp get_non_negated(_, _),                   do: nil
-end
@@ -1,551 +0,0 @@
-defmodule Path do
-  @moduledoc """
-  This module provides conveniences for manipulating or
-  retrieving file system paths.
-
-  The functions in this module may receive a char list or
-  a binary as an argument and will return a value of the same
-  type.
-
-  The majority of the functions in this module do not
-  interact with the file system, except for a few functions
-  that require it (like `wildcard/1` and `expand/1`).
-  """
-
-  alias :filename, as: FN
-
-  @type t :: char_list | atom | binary
-  @type r :: char_list | binary
-
-  @doc """
-  Converts the given path to an absolute one. Unlike
-  `expand/1`, no attempt is made to resolve `..`, `.` or `~`.
-
-  ## Unix examples
-
-      Path.absname("foo")
-      #=> "/usr/local/foo"
-
-      Path.absname("../x")
-      #=> "/usr/local/../x"
-
-  ## Windows
-
-      Path.absname("foo").
-      "D:/usr/local/foo"
-      Path.absname("../x").
-      "D:/usr/local/../x"
-
-  """
-  def absname(path) do
-    FN.absname(path, get_cwd(path))
-  end
-
-  @doc """
-  Builds a path from `relative_to` to `path`. If `path` is already
-  an absolute path, `relative_to` is ignored. See also `relative_to/2`.
-
-  Unlike `expand/2`, no attempt is made to
-  resolve `..`, `.` or `~`.
-
-  ## Examples
-
-      iex> Path.absname("foo", "bar")
-      "bar/foo"
-
-      iex> Path.absname("../x", "bar")
-      "bar/../x"
-
-  """
-  def absname(path, relative_to) do
-    FN.absname(path, relative_to)
-  end
-
-  @doc """
-  Converts the path to an absolute one and expands
-  any `.` and `..` characters and a leading `~`.
-
-  ## Examples
-
-      Path.expand("/foo/bar/../bar")
-      "/foo/bar"
-
-  """
-  def expand(path) do
-    normalize FN.absname(expand_home(path), get_cwd(path))
-  end
-
-  @doc """
-  Expands the path relative to the path given as the second argument
-  expanding any `.` and `..` characters. If the path is already an
-  absolute path, `relative_to` is ignored.
-
-  Note, that this function treats `path` with a leading `~` as
-  an absolute one.
-
-  The second argument is first expanded to an absolute path.
-
-  ## Examples
-
-      # Assuming that the absolute path to baz is /quux/baz
-      Path.expand("foo/bar/../bar", "baz")
-      #=> "/quux/baz/foo/bar"
-
-      Path.expand("foo/bar/../bar", "/baz")
-      "/baz/foo/bar"
-      Path.expand("/foo/bar/../bar", "/baz")
-      "/foo/bar"
-
-  """
-  def expand(path, relative_to) do
-    normalize FN.absname(FN.absname(expand_home(path), expand_home(relative_to)), get_cwd(path))
-  end
-
-  @doc """
-  Returns the path type.
-
-  ## Unix examples
-
-      Path.type("/usr/local/bin")   #=> :absolute
-      Path.type("usr/local/bin")    #=> :relative
-      Path.type("../usr/local/bin") #=> :relative
-      Path.type("~/file")           #=> :relative
-
-  ## Windows examples
-
-      Path.type("D:/usr/local/bin") #=> :absolute
-      Path.type("usr/local/bin")    #=> :relative
-      Path.type("D:bar.ex")         #=> :volumerelative
-      Path.type("/bar/foo.ex")      #=> :volumerelative
-
-  """
-  def type(name) when is_list(name) or is_binary(name) do
-    case :os.type() do
-      { :win32, _ } -> win32_pathtype(name)
-      _             -> unix_pathtype(name)
-    end |> elem(0)
-  end
-
-  @doc """
-  Forces the path to be a relative path.
-
-  ## Unix examples
-
-      Path.relative("/usr/local/bin")   #=> "usr/local/bin"
-      Path.relative("usr/local/bin")    #=> "usr/local/bin"
-      Path.relative("../usr/local/bin") #=> "../usr/local/bin"
-
-  ## Windows examples
-
-      Path.relative("D:/usr/local/bin") #=> "usr/local/bin"
-      Path.relative("usr/local/bin")    #=> "usr/local/bin"
-      Path.relative("D:bar.ex")         #=> "bar.ex"
-      Path.relative("/bar/foo.ex")      #=> "bar/foo.ex"
-
-  """
-  def relative(name) do
-    case :os.type() do
-      { :win32, _ } -> win32_pathtype(name)
-      _             -> unix_pathtype(name)
-    end |> elem(1)
-  end
-
-  defp unix_pathtype(<<?/, relative :: binary>>), do:
-    { :absolute, relative }
-  defp unix_pathtype([?/|relative]), do:
-    { :absolute, relative }
-  defp unix_pathtype([list|rest]) when is_list(list), do:
-    unix_pathtype(list ++ rest)
-  defp unix_pathtype([atom|rest]) when is_atom(atom), do:
-    unix_pathtype(atom_to_list(atom) ++ rest)
-  defp unix_pathtype(relative), do:
-    { :relative, relative }
-
-  @slash [?/, ?\\]
-
-  defp win32_pathtype([list|rest]) when is_list(list), do:
-    win32_pathtype(list++rest)
-  defp win32_pathtype([atom|rest]) when is_atom(atom), do:
-    win32_pathtype(atom_to_list(atom)++rest)
-  defp win32_pathtype([char, list|rest]) when is_list(list), do:
-    win32_pathtype([char|list++rest])
-  defp win32_pathtype(<<c1, c2, relative :: binary>>) when c1 in @slash and c2 in @slash, do:
-    { :absolute, relative }
-  defp win32_pathtype(<<c, relative :: binary>>) when c in @slash, do:
-    { :volumerelative, relative }
-  defp win32_pathtype(<<_letter, ?:, c, relative :: binary>>) when c in @slash, do:
-    { :absolute, relative }
-  defp win32_pathtype(<<_letter, ?:, relative :: binary>>), do:
-    { :volumerelative, relative }
-
-  defp win32_pathtype([c1, c2 | relative]) when c1 in @slash and c2 in @slash, do:
-    { :absolute, relative }
-  defp win32_pathtype([c | relative]) when c in @slash, do:
-    { :volumerelative, relative }
-  defp win32_pathtype([c1, c2, list|rest]) when is_list(list), do:
-    win32_pathtype([c1, c2|list++rest])
-  defp win32_pathtype([_letter, ?:, c | relative]) when c in @slash, do:
-    { :absolute, relative }
-  defp win32_pathtype([_letter, ?: | relative]), do:
-    { :volumerelative, relative }
-  defp win32_pathtype(relative), do:
-    { :relative, relative }
-
-  @doc """
-  Returns the given `path` relative to the given `from` path.
-  In other words, it tries to strip the `from` prefix from `path`.
-
-  This function does not query the file system, so it assumes
-  no symlinks in between the paths.
-
-  In case a direct relative path cannot be found, it returns
-  the original path.
-
-  ## Examples
-
-      iex> Path.relative_to("/usr/local/foo", "/usr/local")
-      "foo"
-      iex> Path.relative_to("/usr/local/foo", "/")
-      "usr/local/foo"
-      iex> Path.relative_to("/usr/local/foo", "/etc")
-      "/usr/local/foo"
-
-  """
-  def relative_to(path, from) when is_list(path) and is_binary(from) do
-    path = filename_string_to_binary(path)
-    relative_to(FN.split(path), FN.split(from), path)
-  end
-
-  def relative_to(path, from) when is_binary(path) and is_list(from) do
-    relative_to(FN.split(path), FN.split(filename_string_to_binary(from)), path)
-  end
-
-  def relative_to(path, from) do
-    relative_to(FN.split(path), FN.split(from), path)
-  end
-
-  defp relative_to([h|t1], [h|t2], original) do
-    relative_to(t1, t2, original)
-  end
-
-  defp relative_to([_|_] = l1, [], _original) do
-    FN.join(l1)
-  end
-
-  defp relative_to(_, _, original) do
-    original
-  end
-
-  @doc """
-  Convenience to get the path relative to the current working
-  directory. If, for some reason, the current working directory
-  cannot be retrieved, returns the full path.
-  """
-  def relative_to_cwd(path) do
-    case :file.get_cwd do
-      { :ok, base } -> relative_to(path, base)
-      _ -> path
-    end
-  end
-
-  @doc """
-  Returns the last component of the path or the path
-  itself if it does not contain any directory separators.
-
-  ## Examples
-
-      iex> Path.basename("foo")
-      "foo"
-
-      iex> Path.basename("foo/bar")
-      "bar"
-
-      iex> Path.basename("/")
-      ""
-
-  """
-  def basename(path) do
-    FN.basename(path)
-  end
-
-  @doc """
-  Returns the last component of `path` with the `extension`
-  stripped. This function should be used to remove a specific
-  extension which may, or may not, be there.
-
-  ## Examples
-
-      iex> Path.basename("~/foo/bar.ex", ".ex")
-      "bar"
-      iex> Path.basename("~/foo/bar.exs", ".ex")
-      "bar.exs"
-      iex> Path.basename("~/foo/bar.old.ex", ".ex")
-      "bar.old"
-
-  """
-  def basename(path, extension) do
-    FN.basename(path, extension)
-  end
-
-  @doc """
-  Returns the directory component of `path`.
-
-  ## Examples
-
-      Path.dirname("/foo/bar.ex")
-      #=> "/foo"
-      Path.dirname("/foo/bar/baz.ex")
-      #=> "/foo/bar"
-
-  """
-  def dirname(path) do
-    FN.dirname(path)
-  end
-
-  @doc """
-  Returns the extension of the last component of `path`.
-
-  ## Examples
-
-      iex> Path.extname("foo.erl")
-      ".erl"
-      iex> Path.extname("~/foo/bar")
-      ""
-
-  """
-  def extname(path) do
-    FN.extension(path)
-  end
-
-  @doc """
-  Returns the `path` with the `extension` stripped.
-
-  ## Examples
-
-      iex> Path.rootname("/foo/bar")
-      "/foo/bar"
-      iex> Path.rootname("/foo/bar.ex")
-      "/foo/bar"
-
-  """
-  def rootname(path) do
-    FN.rootname(path)
-  end
-
-  @doc """
-  Returns the `path` with the `extension` stripped. This function should be used to
-  remove a specific extension which might, or might not, be there.
-
-  ## Examples
-
-      iex> Path.rootname("/foo/bar.erl", ".erl")
-      "/foo/bar"
-      iex> Path.rootname("/foo/bar.erl", ".ex")
-      "/foo/bar.erl"
-
-  """
-  def rootname(path, extension) do
-    FN.rootname(path, extension)
-  end
-
-  @doc """
-  Returns a string with one or more path components joined by the path separator.
-  This function should be used to convert a list of strings to a path.
-
-  ## Examples
-
-      iex> Path.join(["~", "foo"])
-      "~/foo"
-      iex> Path.join(["foo"])
-      "foo"
-      iex> Path.join(["/", "foo", "bar"])
-      "/foo/bar"
-
-  """
-  def join([name1, name2|rest]), do:
-    join([join(name1, name2)|rest])
-  def join([name]) when is_list(name), do:
-    binary_to_filename_string(do_join(filename_string_to_binary(name), <<>>, [], major_os_type()))
-  def join([name]) when is_binary(name), do:
-    do_join(name, <<>>, [], major_os_type())
-
-  @doc """
-  Joins two paths.
-
-  ## Examples
-
-      iex> Path.join("foo", "bar")
-      "foo/bar"
-
-  """
-  def join(left, right) when is_binary(left) and is_binary(right), do:
-    do_join(left, Path.relative(right), [], major_os_type())
-
-  def join(left, right) when is_binary(left) and is_list(right), do:
-    join(left, filename_string_to_binary(right))
-
-  def join(left, right) when is_list(left) and is_binary(right), do:
-    join(filename_string_to_binary(left), right)
-
-  def join(left, right) when is_list(left) and is_list(right), do:
-    binary_to_filename_string join(filename_string_to_binary(left), filename_string_to_binary(right))
-
-  def join(left, right) when is_atom(left), do:
-    join(atom_to_binary(left), right)
-
-  def join(left, right) when is_atom(right), do:
-    join(left, atom_to_binary(right))
-
-  defp major_os_type do
-    :os.type |> elem(0)
-  end
-
-  defp do_join(<<uc_letter, ?:, rest :: binary>>, relativename, [], :win32) when uc_letter in ?A..?Z, do:
-    do_join(rest, relativename, [?:, uc_letter+?a-?A], :win32)
-  defp do_join(<<?\\, rest :: binary>>, relativename, result, :win32), do:
-    do_join(<<?/, rest :: binary>>, relativename, result, :win32)
-  defp do_join(<<?/, rest :: binary>>, relativename, [?., ?/|result], os_type), do:
-    do_join(rest, relativename, [?/|result], os_type)
-  defp do_join(<<?/, rest :: binary>>, relativename, [?/|result], os_type), do:
-    do_join(rest, relativename, [?/|result], os_type)
-  defp do_join(<<>>, <<>>, result, os_type), do:
-    iolist_to_binary(maybe_remove_dirsep(result, os_type))
-  defp do_join(<<>>, relativename, [?:|rest], :win32), do:
-    do_join(relativename, <<>>, [?:|rest], :win32)
-  defp do_join(<<>>, relativename, [?/|result], os_type), do:
-    do_join(relativename, <<>>, [?/|result], os_type)
-  defp do_join(<<>>, relativename, result, os_type), do:
-    do_join(relativename, <<>>, [?/|result], os_type)
-  defp do_join(<<char, rest :: binary>>, relativename, result, os_type) when is_integer(char), do:
-    do_join(rest, relativename, [char|result], os_type)
-
-  defp maybe_remove_dirsep([?/, ?:, letter], :win32), do:
-    [letter, ?:, ?/]
-  defp maybe_remove_dirsep([?/], _), do:
-    [?/]
-  defp maybe_remove_dirsep([?/|name], _), do:
-    :lists.reverse(name)
-  defp maybe_remove_dirsep(name, _), do:
-    :lists.reverse(name)
-
-  @doc """
-  Returns a list with the path split by the path separator.
-  If an empty string is given, returns the root path.
-
-  ## Examples
-
-       iex> Path.split("")
-       []
-       iex> Path.split("foo")
-       ["foo"]
-       iex> Path.split("/foo/bar")
-       ["/", "foo", "bar"]
-
-  """
-  # Work around a bug in Erlang on UNIX
-  def split(""), do: []
-
-  def split(path) do
-    FN.split(path)
-  end
-
-  @doc """
-  Traverses paths according to the given `glob` expression.
-
-  The wildcard looks like an ordinary path, except that certain
-  "wildcard characters" are interpreted in a special way. The
-  following characters are special:
-
-  * `?` - Matches one character.
-  * `*` - Matches any number of characters up to the end of
-          the filename, the next dot, or the next slash.
-  * `**` - Two adjacent <c>*</c>'s used as a single pattern will
-           match all files and zero or more directories and subdirectories.
-  * `[char1,char2,...]` - Matches any of the characters listed. Two characters
-                          separated by a hyphen will match a range of characters.
-  * `{item1,item2,...}` - Matches one of the alternatives.
-
-  Other characters represent themselves. Only paths that have
-  exactly the same character in the same position will match. Note
-  that matching is case-sensitive; i.e. "a" will not match "A".
-
-  ## Examples
-
-  Imagine you have a directory called `projects` with three Elixir projects
-  inside of it: `elixir`, `ex_doc` and `dynamo`. You can find all `.beam` files
-  inside the `ebin` directory of each project as follows:
-
-      Path.wildcard("projects/*/ebin/**/*.beam")
-
-  If you want to search for both `.beam` and `.app` files, you could do:
-
-      Path.wildcard("projects/*/ebin/**/*.{beam,app}")
-
-  """
-  def wildcard(glob) when is_binary(glob) do
-    paths = :filelib.wildcard binary_to_filename_string(glob)
-    encoding = :file.native_name_encoding()
-    Enum.map paths, &flatten_filename_to_binary(&1, encoding)
-  end
-
-  def wildcard(glob) when is_list(glob) do
-    :filelib.wildcard glob
-  end
-
-  ## Helpers
-
-  defp get_cwd(path) when is_list(path), do: System.cwd! |> binary_to_filename_string
-  defp get_cwd(_), do: System.cwd!
-
-  defp binary_to_filename_string(binary) do
-    case :unicode.characters_to_list(binary) do
-      { :error, _, _ } ->
-        :erlang.error(:badarg)
-      list when is_list(list) ->
-        list
-    end
-  end
-
-  defp filename_string_to_binary(list) do
-    flatten_filename_to_binary(:filename.flatten(list), :file.native_name_encoding())
-  end
-
-  defp flatten_filename_to_binary(list, encoding) do
-    case :unicode.characters_to_binary(list, :unicode, encoding) do
-      { :error, _, _ } ->
-        :erlang.error(:badarg)
-      bin when is_binary(bin) ->
-        bin
-    end
-  end
-
-  # Normalize the given path by expanding "..", "." and "~".
-
-  defp expand_home(<<?~, rest :: binary>>) do
-    System.user_home! <> rest
-  end
-
-  defp expand_home('~' ++ rest) do
-    (System.user_home! |> binary_to_filename_string) ++ rest
-  end
-
-  defp expand_home(other), do: other
-
-  defp normalize(path), do: normalize(FN.split(path), [])
-
-  defp normalize([top|t], [_|acc]) when top in ["..", '..'] do
-    normalize t, acc
-  end
-
-  defp normalize([top|t], acc) when top in [".", '.'] do
-    normalize t, acc
-  end
-
-  defp normalize([h|t], acc) do
-    normalize t, [h|acc]
-  end
-
-  defp normalize([], acc) do
-    join Enum.reverse(acc)
-  end
-end
@@ -1,68 +0,0 @@
-defmodule Port do
-  @moduledoc """
-  Functions related to Erlang ports.
-  """
-
-  @doc """
-  See http://www.erlang.org/doc/man/erlang.html#open_port-2.
-  """
-  def open(name, settings) do
-    :erlang.open_port(name, settings)
-  end
-
-  @doc """
-  See http://www.erlang.org/doc/man/erlang.html#port_close-1.
-  """
-  def close(port) do
-    :erlang.port_close(port)
-  end
-
-  @doc """
-  See http://www.erlang.org/doc/man/erlang.html#port_command-2.
-  """
-  def command(port, data, options // []) do
-    :erlang.port_command(port, data, options)
-  end
-
-  @doc """
-  See http://www.erlang.org/doc/man/erlang.html#port_connect-2.
-  """
-  def connect(port, pid) do
-    :erlang.port_connect(port, pid)
-  end
-
-  @doc """
-  See http://www.erlang.org/doc/man/erlang.html#port_control-3.
-  """
-  def control(port, operation, data) do
-    :erlang.port_control(port, operation, data)
-  end
-
-  @doc """
-  See http://www.erlang.org/doc/man/erlang.html#port_call-3.
-  """
-  def call(port, operation, data) do
-    :erlang.port_call(port, operation, data)
-  end
-
-  @doc """
-  See http://www.erlang.org/doc/man/erlang.html#port_info-1.
-  """
-  def info(port) do
-    :erlang.port_info(port)
-  end
-
-  @doc """
-  See http://www.erlang.org/doc/man/erlang.html#port_info-2.
-  """
-  def info(port, item) do
-    :erlang.port_info(port, item)
-  end
-
-  @doc """
-  See http://www.erlang.org/doc/man/erlang.html#ports-0.
-  """
-  def list do
-    :erlang.ports
-  end
-end
@@ -1,375 +0,0 @@
-defmodule Process do
-  @moduledoc """
-  This module provides convenience functions around processes and
-  the process dictionary. In Erlang, most of these functions are
-  auto-imported, but in Elixir they are grouped in a module for
-  convenience. Notice that these functions, different from Erlang's,
-  always return nil instead of undefined. You can use their Erlang
-  version if you want the undefined value.
-  """
-
-  @doc """
-  Returns true if the process exists and is alive, that is,
-  is not exiting and has not exited. Otherwise, returns false.
-
-  `pid` must refer to a process at the local node.
-  """
-  @spec alive?(pid) :: boolean
-  def alive?(pid) do
-    :erlang.is_process_alive(pid)
-  end
-
-  @doc """
-  Returns all key-values in the dictionary.
-  """
-  @spec get :: [{term, term}]
-  def get do
-    :erlang.get()
-  end
-
-  @doc """
-  Returns the value for the given key.
-  """
-  @spec get(term) :: term
-  @spec get(term, default :: term) :: term
-  def get(key, default // nil) do
-    case :erlang.get(key) do
-      :undefined ->
-        default
-      value ->
-        value
-    end
-  end
-
-  @doc """
-  Returns all keys that have the given `value`.
-  """
-  @spec get_keys(term) :: [term]
-  def get_keys(value) do
-    :erlang.get_keys(value)
-  end
-
-  @doc """
-  Stores the given key-value in the process dictionary.
-  """
-  @spec put(term, term) :: term | nil
-  def put(key, value) do
-    nillify :erlang.put(key, value)
-  end
-
-  @doc """
-  Deletes all items in the dictionary.
-  """
-  @spec delete :: [{term, term}]
-  def delete() do
-    :erlang.erase()
-  end
-
-  @doc """
-  Deletes the given key from the dictionary.
-  """
-  @spec delete(term) :: term | nil
-  def delete(key) do
-    nillify :erlang.erase(key)
-  end
-
-  @doc """
-  Sends an exit signal with the given reason to the pid.
-
-  The following behavior applies if reason is any term except `:normal` or `:kill`:
-
-  1) If pid is not trapping exits, pid will exit with the given reason;
-
-  2) If pid is trapping exits, the exit signal is transformed into a message
-     {'EXIT', from, reason} and delivered to the message queue of pid;
-
-  3) If reason is the atom `:normal`, pid will not exit. If it is trapping exits,
-     the exit signal is transformed into a message {'EXIT', from, :normal} and
-     delivered to its message queue;
-
-  4) If reason is the atom `:kill`, that is if `exit(pid, :kill)` is called, an
-     untrappable exit signal is sent to pid which will unconditionally exit with
-     exit reason `:killed`.
-
-  ## Examples
-
-      Process.exit(pid, :kill)
-
-  """
-  @spec exit(pid, term) :: true
-  def exit(pid, reason) do
-    :erlang.exit(pid, reason)
-  end
-
-  @doc """
-  Returns the pid (process identifier) of the calling process.
-  """
-  @spec self() :: pid
-  def self() do
-    :erlang.self()
-  end
-
-  @doc """
-  Returns the pid of a new process started by the application of `fun`.
-  It behaves exactly the same as `Kernel.spawn/1`.
-  """
-  @spec spawn((() -> any)) :: pid
-  def spawn(fun) do
-    :erlang.spawn(fun)
-  end
-
-  @type spawn_opt  :: :link | :monitor | {:priority, :low | :normal | :high} |
-                      {:fullsweep_after, non_neg_integer} |
-                      {:min_heap_size, non_neg_integer} |
-                      {:min_bin_vheap_size, non_neg_integer}
-  @type spawn_opts :: [spawn_opt]
-
-  @doc """
-  Returns the pid of a new process started by the application of `fun`.
-
-  It also accepts extra options, for the list of available options
-  check http://www.erlang.org/doc/man/erlang.html#spawn_opt-2
-  """
-  @spec spawn((() -> any), spawn_opts) :: pid | {pid, reference}
-  def spawn(fun, opts) do
-    :erlang.spawn_opt(fun, opts)
-  end
-
-  @doc """
-  Returns the pid of a new process started by the application of
-  `module.function(args)`. The new process created will be placed in the system
-  scheduler queue and be run some time later.
-
-  It behaves exactly the same as the `Kernel.spawn/3` function.
-  """
-  @spec spawn(module, atom, [any]) :: pid
-  def spawn(mod, fun, args) do
-    :erlang.spawn(mod, fun, args)
-  end
-
-  @doc """
-  Returns the pid of a new process started by the application of
-  `module.function(args)`. The new process created will be placed in the system
-  scheduler queue and be run some time later.
-
-  It also accepts extra options, for the list of available options
-  check http://www.erlang.org/doc/man/erlang.html#spawn_opt-4
-
-  """
-  @spec spawn(module, atom, [any], spawn_opts) :: pid | {pid, reference}
-  def spawn(mod, fun, args, opts) do
-    :erlang.spawn_opt(mod, fun, args, opts)
-  end
-
-  @doc """
-  Returns the pid of a new process started by the application of `fun`.
-  A link is created between the calling process and the new
-  process, atomically.
-  """
-  @spec spawn_link((() -> any)) :: pid
-  def spawn_link(fun) do
-    :erlang.spawn_link(fun)
-  end
-
-  @doc """
-  Returns the pid of a new process started by the application of
-  `module.function(args)`. A link is created between the calling process
-  and the new process, atomically. Otherwise works like spawn/3.
-  """
-  @spec spawn_link(module, atom, [any]) :: pid
-  def spawn_link(mod, fun, args) do
-    :erlang.spawn_link(mod, fun, args)
-  end
-
-  @doc """
-  Returns the pid of a new process started by the application of `fun`
-  and reference for a monitor created to the new process.
-  """
-  @spec spawn_monitor((() -> any)) :: {pid, reference}
-  def spawn_monitor(fun) do
-    :erlang.spawn_monitor(fun)
-  end
-
-  @doc """
-  A new process is started by the application of `module.function(args)`
-  and the process is monitored at the same time. Returns the pid and a
-  reference for the monitor. Otherwise works like spawn/3.
-  """
-  @spec spawn_monitor(module, atom, [any]) :: {pid, reference}
-  def spawn_monitor(mod, fun, args) do
-    :erlang.spawn_monitor(mod, fun, args)
-  end
-
-  @doc """
-  The calling process starts monitoring the item given.
-  It returns the monitor reference.
-
-  See http://www.erlang.org/doc/man/erlang.html#monitor-2 for more info.
-  """
-  @spec monitor(pid | {reg_name :: atom, node :: atom} | reg_name :: atom) :: reference
-  def monitor(item) do
-    :erlang.monitor(:process, item)
-  end
-
-  @doc """
-  If monitor_ref is a reference which the calling process
-  obtained by calling monitor/1, this monitoring is turned off.
-  If the monitoring is already turned off, nothing happens.
-
-  See http://www.erlang.org/doc/man/erlang.html#demonitor-2 for more info.
-  """
-  @spec demonitor(reference) :: true
-  @spec demonitor(reference, options :: [:flush | :info]) :: boolean
-  def demonitor(monitor_ref, options // []) do
-    :erlang.demonitor(monitor_ref, options)
-  end
-
-  @doc """
-  Returns a list of process identifiers corresponding to all the
-  processes currently existing on the local node.
-
-  Note that a process that is exiting, exists but is not alive, i.e.,
-  alive?/1 will return false for a process that is exiting,
-  but its process identifier will be part of the result returned.
-
-  See http://www.erlang.org/doc/man/erlang.html#processes-0 for more info.
-  """
-  @spec list :: [pid]
-  def list do
-    :erlang.processes()
-  end
-
-  @doc """
-  Creates a link between the calling process and another process
-  (or port) `pid`, if there is not such a link already.
-
-  See http://www.erlang.org/doc/man/erlang.html#link-1 for more info.
-  """
-  @spec link(pid | port) :: true
-  def link(pid) do
-    :erlang.link(pid)
-  end
-
-  @doc """
-  Removes the link, if there is one, between the calling process and
-  the process or port referred to by `pid`. Returns true and does not
-  fail, even if there is no link or `id` does not exist
-
-  See http://www.erlang.org/doc/man/erlang.html#unlink-1 for more info.
-  """
-  @spec unlink(pid | port) :: true
-  def unlink(pid) do
-    :erlang.unlink(pid)
-  end
-
-  @doc """
-  Associates the name with a pid or a port identifier. name, which must
-  be an atom, can be used instead of the pid / port identifier in the
-  send operator (name <- message).
-
-  See http://www.erlang.org/doc/man/erlang.html#register-2 for more info.
-  """
-  @spec register(pid | port, atom) :: true
-  def register(pid, name) do
-    :erlang.register(name, pid)
-  end
-
-  @doc """
-  Removes the registered name, associated with a pid or a port identifier.
-
-  See http://www.erlang.org/doc/man/erlang.html#unregister-1 for more info.
-  """
-  @spec unregister(atom) :: true
-  def unregister(name) do
-    :erlang.unregister(name)
-  end
-
-  @doc """
-  Returns the pid or port identifier with the registered name.
-  Returns nil if the name is not registered.
-
-  See http://www.erlang.org/doc/man/erlang.html#whereis-1 for more info.
-  """
-  @spec whereis(atom) :: pid | port | nil
-  def whereis(name) do
-    nillify :erlang.whereis(name)
-  end
-
-  @doc """
-  Returns the pid of the group leader for the process which evaluates the function.
-  """
-  @spec group_leader :: pid
-  def group_leader do
-    :erlang.group_leader
-  end
-
-  @doc """
-  Sets the group leader of `pid` to `leader`. Typically, this is used when a processes
-  started from a certain shell should have another group leader than `:init`.
-  """
-  @spec group_leader(pid, leader :: pid) :: true
-  def group_leader(pid, leader) do
-    :erlang.group_leader(leader, pid)
-  end
-
-  @doc """
-  Returns a list of names which have been registered using register/2.
-  """
-  @spec registered :: [atom]
-  def registered do
-    :erlang.registered()
-  end
-
-  @typep process_flag :: :trap_exit | :error_handler | :min_heap_size |
-                         :min_bin_vheap_size | :priority | :save_calls |
-                         :sensitive
-  @doc """
-  Sets certain flags for the process which calls this function.
-  Returns the old value of the flag.
-
-  See http://www.erlang.org/doc/man/erlang.html#process_flag-2 for more info.
-  """
-  @spec flag(process_flag, term) :: term
-  def flag(flag, value) do
-    :erlang.process_flag(flag, value)
-  end
-
-  @doc """
-  Sets certain flags for the process Pid, in the same manner as flag/2.
-  Returns the old value of the flag. The allowed values for Flag are
-  only a subset of those allowed in flag/2, namely: save_calls.
-
-  See http://www.erlang.org/doc/man/erlang.html#process_flag-3 for more info.
-  """
-  @spec flag(pid, process_flag, term) :: term
-  def flag(pid, flag, value) do
-    :erlang.process_flag(pid, flag, value)
-  end
-
-  @doc """
-  Returns information about the process identified by pid or nil if the process
-  is not alive.
-  Use this only for debugging information.
-
-  See http://www.erlang.org/doc/man/erlang.html#process_info-1 for more info.
-  """
-  @spec info(pid) :: Keyword.t
-  def info(pid) do
-    nillify :erlang.process_info(pid)
-  end
-
-  @doc """
-  Returns information about the process identified by pid
-  or nil if the process is not alive.
-
-  See http://www.erlang.org/doc/man/erlang.html#process_info-2 for more info.
-  """
-  @spec info(pid, atom) :: {atom, term}
-  def info(pid, spec) do
-    nillify :erlang.process_info(pid, spec)
-  end
-
-  @compile { :inline, nillify: 1 }
-  defp nillify(:undefined), do: nil
-  defp nillify(other),      do: other
-end
@@ -1,298 +0,0 @@
-defmodule Protocol do
-  @moduledoc false
-
-  # Callback for defprotocol.
-  @doc false
-  def defprotocol(name, [do: block]) do
-    quote do
-      defmodule unquote(name) do
-        # We don't allow function definition inside protocols
-        import Kernel, except: [
-          defmacrop: 1, defmacrop: 2, defmacro: 1, defmacro: 2,
-          defp: 1, defp: 2, def: 1, def: 2
-        ]
-
-        # Import the new dsl that holds the new def
-        import Protocol.DSL, only: :macros
-
-        # Compile with debug info for consolidation
-        @compile :debug_info
-
-        # Set up a clear slate to store defined functions
-        @functions []
-        @fallback_to_any false
-
-        # Deprecated
-        @only nil
-        @except nil
-
-        # Invoke the user given block
-        unquote(block)
-
-        # Finalize expansion
-        unquote(after_defprotocol)
-      end
-    end
-  end
-
-  defp after_defprotocol do
-    quote unquote: false do
-      msg = "is deprecated and should be removed. " <>
-            "Note if you want to fallback to Any, you have to set @fallback_to_any true"
-
-      if @only do
-        IO.puts "warning: @only in protocol #{inspect __MODULE__} " <> msg
-        @fallback_to_any @fallback_to_any || Any in @only
-      end
-
-      if @except do
-        IO.puts "warning: @except in protocol #{inspect __MODULE__} " <> msg
-        @fallback_to_any @fallback_to_any || not(Any in @except)
-      end
-
-      { arg, bodies, rec } = Protocol.impl_for(__MODULE__)
-
-      @spec impl_for(term) :: module | nil
-      Kernel.def impl_for(data)
-
-      lc { guard, body } inlist bodies do
-        Kernel.def impl_for(unquote(arg)) when unquote(guard), do: unquote(body)
-      end
-
-      @spec impl_for!(term) :: module | no_return
-      Kernel.def impl_for!(data) do
-        impl_for(data) || raise(Protocol.UndefinedError, protocol: __MODULE__, value: data)
-      end
-
-      # Handle special Record type
-      Kernel.defp rec_impl_for(unquote(arg)), do: unquote(rec)
-
-      # Handle special Any type
-      if @fallback_to_any do
-        Kernel.defp any_impl_for do
-          try do
-            __MODULE__.Any.__impl__(:name)
-          catch
-            :error, :undef, [[{ __MODULE__.Any, :__impl__, [:name], _ }|_]|_] ->
-              nil
-          end
-        end
-      else
-        Kernel.defp any_impl_for, do: nil
-      end
-
-      # Inline both helpers
-      @compile { :inline, any_impl_for: 0, rec_impl_for: 1 }
-
-      unless Kernel.Typespec.defines_type?(__MODULE__, :t, 0) do
-        @type t :: term
-      end
-
-      # Store information as an attribute so it
-      # can be read without loading the module.
-      Module.register_attribute(__MODULE__, :protocol, persist: true)
-      @protocol [fallback_to_any: !!@fallback_to_any, consolidated: false]
-
-      @doc false
-      Kernel.def __protocol__(:name),      do: __MODULE__
-      Kernel.def __protocol__(:functions), do: unquote(:lists.sort(@functions))
-    end
-  end
-
-  # Callback for defimpl.
-  @doc false
-  def defimpl(protocol, opts) do
-    do_defimpl(protocol, :lists.keysort(1, opts))
-  end
-
-  defp do_defimpl(protocol, [do: block, for: for]) when is_list(for) do
-    lc f inlist for, do: do_defimpl(protocol, [do: block, for: f])
-  end
-
-  defp do_defimpl(protocol, [do: block, for: for]) do
-    quote do
-      protocol = unquote(protocol)
-      for      = unquote(for)
-      name     = Module.concat(protocol, for)
-
-      Protocol.assert_protocol(protocol)
-
-      defmodule name do
-        @behaviour unquote(protocol)
-        @protocol  unquote(protocol)
-        @for       unquote(for)
-
-        unquote(block)
-
-        Module.register_attribute(__MODULE__, :impl, persist: true)
-        @impl [protocol: @protocol, for: @for]
-
-        @doc false
-        def __impl__(:name),     do: __MODULE__
-        def __impl__(:protocol), do: @protocol
-        def __impl__(:for),      do: @for
-      end
-    end
-  end
-
-  # Check if the given module is a protocol. Raises an error
-  # if not loaded or not a protocol.
-  @doc false
-  def assert_protocol(module) do
-    case Code.ensure_compiled(module) do
-      { :module, ^module } -> nil
-      _ -> raise ArgumentError, message: "#{inspect module} is not loaded"
-    end
-
-    try do
-      module.__protocol__(:name)
-    rescue
-      UndefinedFunctionError ->
-        raise ArgumentError, message: "#{inspect module} is not a protocol"
-    end
-  end
-
-  # Builtin types.
-  @doc false
-  def builtin do
-    [ Tuple, Atom, List, BitString, Integer, Float,
-      Function, PID, Port, Reference, Any ]
-  end
-
-  # Implements the function that detects the protocol and
-  # returns the module to dispatch to.
-  @doc false
-  def impl_for(current) do
-    arg = quote(do: arg)
-    all = [Record|builtin]
-
-    { arg,
-      lc(mod inlist all, do: impl_for(current, mod, arg)),
-      rec_impl_for(current, arg) }
-  end
-
-  defp rec_impl_for(current, arg) do
-    fallback = impl_for(current, Tuple, arg) |> elem(1)
-
-    quote do
-      target = Module.concat(unquote(current), unquote(arg))
-      try do
-        target.__impl__(:name)
-      catch
-        :error, :undef, [[{ ^target, :__impl__, [:name], _ }|_]|_] ->
-          unquote(fallback)
-      end
-    end
-  end
-
-  defp impl_for(current, Record, arg) do
-    fallback = impl_for(current, Tuple, arg) |> elem(1)
-
-    dispatch = quote do
-      atom = :erlang.element(1, unquote(arg))
-
-      case not(atom in unquote(builtin)) and match?('Elixir.' ++ _, atom_to_list(atom)) do
-        true  -> rec_impl_for(atom)
-        false -> unquote(fallback)
-      end
-    end
-
-    quote do
-      { is_record(unquote(arg)), unquote(dispatch) }
-    end
-  end
-
-  defp impl_for(current, Tuple, arg),     do: impl_with_fallback(Tuple, :is_tuple, current, Any, arg)
-  defp impl_for(current, Atom, arg),      do: impl_with_fallback(Atom, :is_atom, current, Any, arg)
-  defp impl_for(current, List, arg),      do: impl_with_fallback(List, :is_list, current, Any, arg)
-  defp impl_for(current, BitString, arg), do: impl_with_fallback(BitString, :is_bitstring, current, Any, arg)
-  defp impl_for(current, Integer, arg),   do: impl_with_fallback(Integer, :is_integer, current, Any, arg)
-  defp impl_for(current, Float, arg),     do: impl_with_fallback(Float, :is_float, current, Any, arg)
-  defp impl_for(current, Function, arg),  do: impl_with_fallback(Function, :is_function, current, Any, arg)
-  defp impl_for(current, PID, arg),       do: impl_with_fallback(PID, :is_pid, current, Any, arg)
-  defp impl_for(current, Port, arg),      do: impl_with_fallback(Port, :is_port, current, Any, arg)
-  defp impl_for(current, Reference, arg), do: impl_with_fallback(Reference, :is_reference, current, Any, arg)
-
-  defp impl_for(_current, Any, _arg) do
-    { true, quote(do: any_impl_for) }
-  end
-
-  # Defines an implementation with fallback to the given module.
-  defp impl_with_fallback(mod, guard, current, fallback, arg) do
-    quote do
-      { unquote(guard)(unquote(arg)),
-        unquote(with_fallback(Module.concat(current, mod), current, fallback, arg)) }
-    end
-  end
-
-  # Tries to dispatch to a given target, fallbacks to the
-  # given `fallback` implementation if the target does not exist.
-  defp with_fallback(target, current, fallback, arg) when is_atom(target) do
-    quote do
-      try do
-        unquote(target).__impl__(:name)
-      catch
-        :error, :undef, [[{ unquote(target), :__impl__, [:name], _ }|_]|_] ->
-          unquote(impl_for(current, fallback, arg) |> elem(1))
-      end
-    end
-  end
-end
-
-defmodule Protocol.DSL do
-  @moduledoc false
-
-  @doc false
-  defmacro def({ _, _, args }) when args == [] or is_atom(args) do
-    raise ArgumentError, message: "protocol functions expect at least one argument"
-  end
-
-  defmacro def({ name, _, args }) when is_atom(name) and is_list(args) do
-    arity = length(args)
-
-    type_args = lc _ inlist :lists.seq(2, arity), do: quote(do: term)
-    type_args = [quote(do: t) | type_args]
-
-    call_args = lc i inlist :lists.seq(2, arity),
-                  do: { binary_to_atom(<<?x, i + 64>>), [], __MODULE__ }
-    call_args = [quote(do: t) | call_args]
-
-    quote do
-      name  = unquote(name)
-      arity = unquote(arity)
-
-      @functions [{name, arity}|@functions]
-
-      # Generate a fake definition with the user
-      # signature that will be used by docs
-      Kernel.def unquote(name)(unquote_splicing(args))
-
-      # Generate the actual implementation
-      Kernel.def unquote(name)(unquote_splicing(call_args)) do
-        impl_for!(t).unquote(name)(unquote_splicing(call_args))
-      end
-
-      # Convert the spec to callback if possible,
-      # otherwise generate a dummy callback
-      Protocol.DSL.callback_from_spec(__MODULE__, name, arity) ||
-        @callback unquote(name)(unquote_splicing(type_args)) :: term
-    end
-  end
-
-  defmacro def(_) do
-    raise ArgumentError, message: "invalid args for def inside defprotocol"
-  end
-
-  @doc false
-  def callback_from_spec(module, name, arity) do
-    tuple = { name, arity }
-    specs = Module.get_attribute(module, :spec)
-
-    found = lc { k, v } inlist specs, k == tuple do
-      Kernel.Typespec.define_callback(module, tuple, v)
-      true
-    end
-
-    found != []
-  end
-end
@@ -1,266 +0,0 @@
-defmodule Protocol.Consolidation do
-  @moduledoc """
-  Module responsible for consolidating protocols and helpers for
-  extracting protocols and implementations from code paths for
-  consolidation.
-  """
-
-  @doc """
-  Extract all protocols from the given paths.
-
-  The paths can be either a char list or a string. Internally
-  they are worked on as char lists, so passing them as lists
-  avoid extra conversion.
-
-  ## Examples
-
-      # Get Elixir's ebin and retrieve all protocols
-      iex> path = :code.lib_dir(:elixir, :ebin)
-      iex> mods = Protocol.Consolidation.extract_protocols([path])
-      iex> Enumerable in mods
-      true
-
-  """
-  @spec extract_protocols([char_list | String.t]) :: [atom]
-  def extract_protocols(paths) do
-    extract_matching_by_attribute paths, 'Elixir.',
-      fn module, attributes ->
-        case attributes[:protocol] do
-          [fallback_to_any: _, consolidated: _] -> module
-          _ -> nil
-        end
-      end
-  end
-
-  @doc """
-  Extract all types implemented for the given protocol from
-  the given paths.
-
-  The paths can be either a char list or a string. Internally
-  they are worked on as char lists, so passing them as lists
-  avoid extra conversion.
-
-  ## Examples
-
-      # Get Elixir's ebin and retrieve all protocols
-      iex> path = :code.lib_dir(:elixir, :ebin)
-      iex> mods = Protocol.Consolidation.extract_impls(Enumerable, [path])
-      iex> List in mods
-      true
-
-  """
-  @spec extract_impls(module, [char_list | String.t]) :: [atom]
-  def extract_impls(protocol, paths) when is_atom(protocol) do
-    prefix = atom_to_list(protocol) ++ '.'
-    extract_matching_by_attribute paths, prefix, fn
-      _mod, attributes ->
-        case attributes[:impl] do
-          [protocol: ^protocol, for: for] -> for
-          _ -> nil
-        end
-    end
-  end
-
-  defp extract_matching_by_attribute(paths, prefix, callback) do
-    lc path inlist paths,
-       file inlist list_dir(path),
-       mod = extract_from_file(path, file, prefix, callback),
-       do: mod
-  end
-
-  defp list_dir(path) when is_list(path) do
-    case :file.list_dir(path) do
-      { :ok, files } -> files
-      _ -> []
-    end
-  end
-
-  defp list_dir(path), do: list_dir(to_char_list(path))
-
-  defp extract_from_file(path, file, prefix, callback) do
-    if :lists.prefix(prefix, file) and Path.extname(file) == '.beam' do
-      extract_from_beam(Path.join(path, file), callback)
-    end
-  end
-
-  defp extract_from_beam(file, callback) do
-    case :beam_lib.chunks(file, [:attributes]) do
-      {:ok, { module, [attributes: attributes] } } ->
-        callback.(module, attributes)
-       _ ->
-         nil
-    end
-  end
-
-  defmacrop if_ok(expr, call) do
-    quote do
-      case unquote(expr) do
-        { :ok, var } -> unquote(Macro.pipe(quote(do: var), call))
-        other -> other
-      end
-    end
-  end
-
-  @doc """
-  Receives a protocol and a list of implementations and
-  consolidates the given protocol. Consolidation happens
-  by changing the protocol `impl_for` in the abstract
-  format to have fast lookup rules.
-
-  It returns the updated version of the protocol bytecode.
-  A given bytecode or protocol implementation can be checked
-  to be consolidated or not by analyzing the protocol
-  attribute:
-
-      Enumerable.__info__(:attributes)[:protocol]
-
-  If the first element of the tuple is true, it means
-  the protocol was consolidated.
-
-  This function does not load the protocol at any point
-  nor loads the new bytecode for the compiled module.
-  """
-  @spec apply_to(module, [module]) ::
-    { :ok, binary } |
-    { :error, :not_a_protocol } |
-    { :error, :no_beam_info }
-  def apply_to(protocol, types) when is_atom(protocol) do
-    ensure_protocol(protocol)
-    |> if_ok(change_debug_info types)
-    |> if_ok(compile)
-  end
-
-  # Ensure the given module is loaded and is a protocol.
-  defp ensure_protocol(protocol) do
-    case :beam_lib.chunks(beam_file(protocol), [:abstract_code, :attributes]) do
-      { :ok, { ^protocol, [abstract_code: { _raw, abstract_code },
-                           attributes: attributes] } } ->
-        case attributes[:protocol] do
-          [fallback_to_any: any, consolidated: _] ->
-            { :ok, { protocol, any, abstract_code } }
-          _ ->
-            { :error, :not_a_protocol }
-        end
-      _ ->
-        { :error, :no_beam_info }
-    end
-  end
-
-  defp beam_file(module) when is_atom(module) do
-    case :code.which(module) do
-      :non_existing -> module
-      file -> file
-    end
-  end
-
-  # Change the debug information to the optimized
-  # impl_for/1 dispatch version.
-  defp change_debug_info({ protocol, any, code }, types) do
-    types   = if any, do: types, else: List.delete(types, Any)
-    records = types -- Protocol.builtin
-    builtin = Protocol.builtin -- (Protocol.builtin -- types)
-    builtin = if records != [], do: [Record|builtin], else: builtin
-    change_impl_for(code, protocol, builtin, records, false, [])
-  end
-
-  defp change_impl_for([{ :attribute, line, :protocol, _ }|t], protocol, builtin, records, _, acc) do
-    attr = [fallback_to_any: Any in builtin, consolidated: true]
-    change_impl_for(t, protocol, builtin, records, true,
-                    [{ :attribute, line, :protocol, attr }|acc])
-  end
-
-  defp change_impl_for([{ :function, line, :impl_for, 1, _ }|t], protocol, builtin, records, is_protocol, acc) do
-    clauses = lc type inlist builtin, do: clause_for(type, protocol, line)
-
-    unless Any in builtin do
-      clauses = clauses ++ [fallback_clause_for(nil, protocol, line)]
-    end
-
-    change_impl_for(t, protocol, builtin, records, is_protocol,
-                    [{ :function, line, :impl_for, 1, clauses }|acc])
-  end
-
-  defp change_impl_for([{ :function, line, :rec_impl_for, 1, _ }|t], protocol, builtin, records, is_protocol, acc) do
-    fallback = if Tuple in builtin, do: Module.concat(protocol, Tuple)
-    clauses  = lc type inlist records, do: record_clause_for(type, protocol, line)
-    clauses  = clauses ++ [fallback_clause_for(fallback, protocol, line)]
-
-    change_impl_for(t, protocol, builtin, records, is_protocol,
-                    [{ :function, line, :rec_impl_for, 1, clauses }|acc])
-  end
-
-  defp change_impl_for([h|t], protocol, info, types, is_protocol, acc) do
-    change_impl_for(t, protocol, info, types, is_protocol, [h|acc])
-  end
-
-  defp change_impl_for([], protocol, _info, _types, is_protocol, acc) do
-    if is_protocol do
-      { :ok, { protocol, Enum.reverse(acc) } }
-    else
-      { :error, :not_a_protocol }
-    end
-  end
-
-  defp clause_for(Tuple, protocol, line),     do: builtin_clause_for(Tuple, :is_tuple, protocol, line)
-  defp clause_for(Atom, protocol, line),      do: builtin_clause_for(Atom, :is_atom, protocol, line)
-  defp clause_for(List, protocol, line),      do: builtin_clause_for(List, :is_list, protocol, line)
-  defp clause_for(BitString, protocol, line), do: builtin_clause_for(BitString, :is_bitstring, protocol, line)
-  defp clause_for(Integer, protocol, line),   do: builtin_clause_for(Integer, :is_integer, protocol, line)
-  defp clause_for(Float, protocol, line),     do: builtin_clause_for(Float, :is_float, protocol, line)
-  defp clause_for(Function, protocol, line),  do: builtin_clause_for(Function, :is_function, protocol, line)
-  defp clause_for(PID, protocol, line),       do: builtin_clause_for(PID, :is_pid, protocol, line)
-  defp clause_for(Port, protocol, line),      do: builtin_clause_for(Port, :is_port, protocol, line)
-  defp clause_for(Reference, protocol, line), do: builtin_clause_for(Reference, :is_reference, protocol, line)
-
-  defp clause_for(Any, protocol, line) do
-    {:clause, line, [{:var, line, :_}], [],
-      [{ :atom, line, Module.concat(protocol, Any) }]}
-  end
-
-  defp clause_for(Record, _protocol, line) do
-    {:clause, line, [{:var, line, :x}],
-      [[{:op, line, :andalso,
-          {:call, line,
-            {:remote, line, {:atom, line, :erlang}, {:atom, line, :is_tuple}},
-            [{:var, line, :x}]},
-          {:call, line,
-            {:remote, line, {:atom, line, :erlang}, {:atom, line, :is_atom}},
-            [{:call, line,
-              {:remote, line, {:atom, line, :erlang}, {:atom, line, :element}},
-              [{:integer, line, 1}, {:var, line, :x}]
-            }]},
-      }]],
-      [{:call, line,
-          {:atom, line, :rec_impl_for},
-          [{:call, line,
-            {:remote, line, {:atom, line, :erlang}, {:atom, line, :element}},
-            [{:integer, line, 1}, {:var, line, :x}]}]}]}
-  end
-
-  defp builtin_clause_for(mod, guard, protocol, line) do
-    {:clause, line,
-      [{:var, line, :x}],
-      [[{:call, line,
-          {:remote, line, {:atom, line, :erlang}, {:atom, line, guard}},
-          [{:var, line, :x}],
-      }]],
-      [{:atom, line, Module.concat(protocol, mod)}]}
-  end
-
-  defp record_clause_for(other, protocol, line) do
-    {:clause, line, [{:atom, line, other}], [],
-      [{:atom, line, Module.concat(protocol, other)}]}
-  end
-
-  defp fallback_clause_for(value, _protocol, line) do
-    {:clause, line, [{:var, line, :_}], [],
-      [{ :atom, line, value }]}
-  end
-
-  # Finally compile the module and emit its bytecode.
-  defp compile({ protocol, code }) do
-    opts = if Code.compiler_options[:debug_info], do: [:debug_info], else: []
-    { :ok, ^protocol, binary, _warnings } = :compile.forms(code, [:return|opts])
-    { :ok, binary }
-  end
-end
@@ -1,72 +0,0 @@
-defrecord Range, [:first, :last] do
-  @moduledoc """
-  Defines a Range.
-  """
-end
-
-defprotocol Range.Iterator do
-  @doc """
-  Reduces the range based on the type of the first argument.
-  """
-  def reduce(first, range, acc, fun)
-
-  @doc """
-  Count how many items are in the range.
-  """
-  def count(first, range)
-end
-
-defimpl Enumerable, for: Range do
-  def reduce(first .. _ = range, acc, fun) do
-    Range.Iterator.reduce(first, range, acc, fun)
-  end
-
-  def member?(first .. last, value) do
-    first <= value and value <= last
-  end
-
-  def count(first .. _ = range) do
-    Range.Iterator.count(first, range)
-  end
-end
-
-defimpl Range.Iterator, for: Integer do
-  def reduce(first, Range[last: last], acc, fun) when is_integer(last) do
-    reducer = if last >= first do
-      fn(acc, fun) -> do_reducer_up(first, last, acc, fun) end
-    else
-      fn(acc, fun) -> do_reducer_down(first, last, acc, fun) end
-    end
-    Enumerable.Function.reduce(reducer, acc, fun)
-  end
-
-  defp do_reducer_up(counter, last, acc, _fun) when counter > last do
-    acc
-  end
-
-  defp do_reducer_up(counter, last, acc, fun) do
-    do_reducer_up(counter + 1, last, fun.(counter, acc), fun)
-  end
-
-  defp do_reducer_down(counter, last, acc, _fun) when counter < last do
-    acc
-  end
-
-  defp do_reducer_down(counter, last, acc, fun) do
-    do_reducer_down(counter - 1, last, fun.(counter, acc), fun)
-  end
-
-  def count(first, Range[last: last]) when is_integer(last) and last >= first do
-    last - first + 1
-  end
-
-  def count(first, Range[last: last]) when is_integer(last) do
-    first - last + 1
-  end
-end
-
-defimpl Inspect, for: Range do
-  def inspect(Range[first: first, last: last], opts) do
-    Inspect.Algebra.concat [Kernel.inspect(first, opts), "..", Kernel.inspect(last, opts)]
-  end
-end
@@ -1,929 +0,0 @@
-defmodule Record do
-  @moduledoc %S"""
-  Functions for working with Records.
-
-  A record is a tagged tuple which contains one or more elements
-  where the first element is an atom. We can manually create a record
-  by simply defining such a tuple:
-
-      iex> record = { User, "José", 25 }
-      iex> is_record(record, User)
-      true
-
-  However, manually constructing tuples can be quite error prone.
-  If we need to add a new field to our User, it would require
-  us to carefully change all places where the tuple is used.
-  Furthermore, as more items are added to the tuple, they
-  lose semantic value.
-
-  This module solves these problems by allowing us to name each
-  element and encapsulate the generation and manipulation of
-  such tuples. Much of the functionality provided by this module happens
-  at compilation time, meaning they don't add any runtime overhead while
-  considerably improving the quality of our code.
-
-  For these reasons, Records are frequently used in Elixir and
-  are also very useful when combined with Protocols. This module
-  provides different mechanisms for working with records and we
-  are going to explore them in the following sections.
-
-  ## defrecordp
-
-  The simplest way of working with records is via `defrecordp`:
-
-      defmodule User do
-        defrecordp :user, name: "José", age: 25
-      end
-
-  In the example above, `defrecordp` is going to generate a set of
-  macros named `user` that allows us to create, update and match
-  on a record. Our record is going to have two fields, a `name` with
-  default value of "José" and `age` with default value 25.
-
-  Let's see some examples:
-
-      # To create records
-      user()        #=> { :user, "José", 25 }
-      user(age: 26) #=> { :user, "José", 26 }
-
-  By using the `user` macro, we no longer need to explicitly create
-  a tuple with all elements. It also allows us to create and modify
-  values by name:
-
-      # Create a new record
-      sample_user = user()
-
-      # And now change its age to 26
-      user(sample_user, age: 26)
-
-  Since `user` is a macro, all the work happens at compilation time.
-  This means all operations, like changing the `age` above, work
-  as a simple tuple operation at runtime:
-
-      # This update operation...
-      user(sample_user, age: 26)
-
-      # Literally translates to this one:
-      set_elem(sample_user, 2, 26)
-
-  For this reason, the following operation is not allowed as all
-  values need to be explicit:
-
-      new_values = [age: 26]
-      user(sample_user, new_values)
-
-  As the name says, `defrecordp` is useful when you don't want to
-  expose the record definition. The `user` macro used above, for
-  example, is only available inside the `User` module and nowhere else.
-  You can find more information in `Kernel.defrecordp/3` docs.
-
-  ## defrecord
-
-  By using `defrecord`, a developer can make a Record definition
-  available everywhere within Elixir code. Let's see an example:
-
-      defrecord User, name: "José", age: 25
-
-  Notice that, unlike `defrecordp`, `defrecord` expects an
-  alias as the first argument. This is because `defrecord` is going
-  to create a module named `User` with all the relevant metadata.
-  This module can then be imported and we can manipulate the user
-  as with `defrecordp`:
-
-      Record.import User, as: :user
-      user()        #=> User[name: "José", age: 25]
-      user(age: 26) #=> User[name: "José", age: 26]
-
-  Notice that now since the record definition is accessible, Elixir
-  shows the record nicely formatted, no longer as a simple tuple. We
-  can get the raw formatting by passing `raw: true` to `inspect`:
-
-      inspect user(), raw: true
-      { User, "José", 25 }
-
-  Since working with external records is common, Elixir allows
-  developers to skip importing the record altogether in favor
-  of a `Module[args]` syntax:
-
-      # Skipping a field uses its default value
-      User[]        #=> User[name: "José", age: 25]
-      User[age: 26] #=> User[name: "José", age: 26]
-
-  The macro name is replaced by the module name and the parentheses
-  are replaced by brackets. When this syntax is used, there is no
-  need to import the record.
-
-  Before we sum up the differences between `defrecord` and
-  `defrecordp`, there is one last functionality introduced by
-  `defrecord` that we need to discuss.
-
-  ## Runtime access
-
-  All the functionality discussed above happens at compilation time.
-  This means that both `user(age: 26)` and `User[age: 26]` are expanded
-  into a tuple at compile time.
-
-  However, there are some situations where we want to set or update
-  fields dynamically. `defrecord` (and not `defrecordp` ) supports
-  this behaviour out of the box:
-
-      defrecord User, name: "José", age: 25
-
-      opts = [name: "Hello"]
-      user = User.new(opts)
-      #=> User[name: "Hello", age: 25]
-      user.update(age: 26)
-      #=> User[name: "Hello", age: 26]
-
-  All the calls above happen at runtime. It gives Elixir records
-  flexibility at the cost of performance since there is more work
-  happening at runtime.
-
-  The above calls (`new` and `update`) can accept both atom and string
-  keys for field names, however not both at the same time. This feature
-  allows to "sanitize" untrusted dictionaries and initialize/update
-  records without using `Kernel.binary_to_existing_atom/1`.
-
-  To sum up, `defrecordp` should be used when you don't want to expose
-  the record information while `defrecord` should be used whenever you
-  want to share a record within your code or with other libraries or
-  whenever you need to dynamically set or update fields.
-
-  The standard library contains excellent examples of both use cases,
-  with [`HashDict`](HashDict.html) being implemented with `defrecordp` 
-  and [`Range`](Range.html) with `defrecord`.
-
-  You can learn more about records in the `Kernel.defrecord/3` docs. Now
-  let's discuss the usefulness of combining records with protocols.
-
-  ## Protocols
-
-  Developers can extend existing protocols by creating their own
-  records and implementing the desired protocols. For instance,
-  imagine that you have created a new representation for storing
-  date and time, represented by the year, the week of the year
-  and the week day:
-
-      defrecord WeekDate, year: nil, week: nil, week_day: nil
-
-  Now we want this date to be represented as a string and this
-  can be done by implementing the `String.Chars` protocol for
-  our record:
-
-      defimpl String.Chars, for: WeekDate do
-        def to_string(WeekDate[year: year, week: week, week_day: day]) do
-          "#{year}-W#{week}-#{day}"
-        end
-      end
-
-  Now we can explicitly convert our `WeekDate`:
-
-      to_string WeekDate[year: 2013, week: 26, week_day: 4]
-      "2013-W26-4"
-
-  A protocol can be implemented for any record, whether
-  generated with `defrecordp` or `defrecord`.
-  """
-
-  @type t :: tuple
-
-  @doc """
-  Extract record information from an Erlang file. 
-  
-  Returns the fields as a list of tuples.
-
-  ## Examples
-
-      > Record.extract(:file_info, from_lib: "kernel/include/file.hrl")
-      [size: :undefined, type: :undefined, access: :undefined, atime: :undefined,
-         mtime: :undefined, ctime: :undefined, mode: :undefined, links: :undefined,
-          major_device: :undefined, minor_device: :undefined, inode: :undefined,
-           uid: :undefined, gid: :undefined]
-
-      defrecord FileInfo, Record.extract(:file_info, from_lib: "kernel/include/file.hrl")
-
-  """
-  def extract(name, opts) do
-    Record.Extractor.retrieve(name, opts)
-  end
-
-  @doc """
-  Import a public record definition as a set of private macros.
-
-  Macros defined as in `Kernel.defrecordp/3`. This is useful when one
-  desires to manipulate a record via a set of macros instead
-  of the regular access syntax.
-
-  ## Example
-
-      defmodule Test do
-        Record.import File.Stat, as: :file_stat
-
-        def size(file_stat(size: size)), do: size
-      end
-
-  """
-  defmacro import(module, as: name) do
-    quote do
-      module = unquote(module)
-
-      fields = if module == __MODULE__ do
-        @record_fields
-      else
-        module.__record__(:fields)
-      end
-
-      Record.defmacros(unquote(name), fields, __ENV__, module)
-    end
-  end
-
-  @doc false
-  def defrecord(name, fields, opts) do
-    block = Keyword.get(opts, :do, nil)
-    record_check!(fields)
-
-    quote do
-      unquoted_fields = unquote(fields)
-
-      defmodule unquote(name) do
-        import Record.DSL
-
-        @record_fields []
-        @record_types  []
-
-        Record.deffunctions(unquoted_fields, __ENV__)
-        value = unquote(block)
-        Record.deftypes(@record_fields, @record_types, __ENV__)
-        value
-      end
-    end
-  end
-
-  defp record_check!([{ field, { :::, _, [_, _] }}|_]) when is_atom(field) do
-    raise ArgumentError, message: "typespecs are not supported inlined with defrecord, " <>
-                                  "please use record_type instead"
-  end
-
-  defp record_check!([_|t]), do: record_check!(t)
-  defp record_check!(_), do: :ok
-
-  @doc false
-  def defrecordp(name, tag, fields) when is_atom(name) and is_atom(tag) and is_list(fields) do
-    { fields, types, def_type } = recordp_split(fields, [], [], false)
-    type = binary_to_atom(atom_to_binary(name) <> "_t")
-    tag  = tag || name
-
-    quote do
-      Record.defmacros(unquote(name), unquote(fields), __ENV__, unquote(tag))
-
-      if unquote(def_type) do
-        @typep unquote(type)() :: { unquote(tag), unquote_splicing(types) }
-      end
-    end
-  end
-
-  defp recordp_split([{ field, { :::, _, [default, type] }}|t], defaults, types, _) do
-    recordp_split t, [{ field, default }|defaults], [type|types], true
-  end
-
-  defp recordp_split([other|t], defaults, types, def_type) do
-    recordp_split t, [other|defaults], [quote(do: term)|types], def_type
-  end
-
-  defp recordp_split([], defaults, types, def_type) do
-    { :lists.reverse(defaults), :lists.reverse(types), def_type }
-  end
-
-  @doc """
-  Define record functions skipping the module definition.
-
-  This is called directly by `Kernel.defrecord/3`. It expects the record
-  values, a set of options and the module environment.
-
-  ## Examples
-
-      defmodule CustomRecord do
-        Record.deffunctions [:name, :age], __ENV__
-        Record.deftypes [:name, :age], [name: :binary, age: :integer], __ENV__
-      end
-
-  """
-  def deffunctions(values, env) do
-    values  = lc value inlist values, do: convert_value(value)
-    escaped = Macro.escape(values)
-
-    contents = [
-      reflection(escaped),
-      initializer(escaped),
-      conversions(values),
-      record_optimizable(),
-      updater(values),
-      accessors(values, 1),
-      switch_recorder()
-    ]
-
-    contents = [quote(do: @record_fields unquote(escaped))|contents]
-
-    # Special case for bootstrapping purposes
-    if env == Macro.Env do
-      Module.eval_quoted(env, contents, [], [])
-    else
-      Module.eval_quoted(env.module, contents, [], env.location)
-    end
-  end
-
-  @doc """
-  Define types and specs for the record.
-  """
-  def deftypes(values, types, env) do
-    types  = types || []
-    values = lc value inlist values do
-      { name, default } = convert_value(value)
-      { name, default, find_spec(types, name) }
-    end
-
-    contents = [
-      core_specs(values),
-      accessor_specs(values, 1, [])
-    ]
-
-    if env == Macro.Env do
-      Module.eval_quoted(env, contents, [], [])
-    else
-      Module.eval_quoted(env.module, contents, [], env.location)
-    end
-  end
-
-  @doc """
-  Define macros for manipulating records. 
-  
-  This is called
-  directly by `Kernel.defrecordp/3`. It expects the macro name, the
-  record values and the environment.
-
-  ## Examples
-
-      defmodule CustomRecord do
-        Record.defmacros :user, [:name, :age], __ENV__
-      end
-
-  """
-  def defmacros(name, values, env, tag // nil) do
-    escaped = lc value inlist values do
-      { key, value } = convert_value(value)
-      { key, Macro.escape(value) }
-    end
-
-    tag = tag || name
-
-    contents = quote do
-      defmacrop unquote(name)() do
-        Record.access(unquote(tag), unquote(escaped), [], __CALLER__)
-      end
-
-      defmacrop unquote(name)(record) when is_tuple(record) do
-        Record.to_keywords(unquote(tag), unquote(escaped), record)
-      end
-
-      defmacrop unquote(name)(args) do
-        Record.access(unquote(tag), unquote(escaped), args, __CALLER__)
-      end
-
-      defmacrop unquote(name)(record, args) do
-        Record.dispatch(unquote(tag), unquote(escaped), record, args, __CALLER__)
-      end
-    end
-
-    Module.eval_quoted(env.module, contents, [], env.location)
-  end
-
-  ## Callbacks
-
-  # Store all optimizable fields in the record as well
-  @doc false
-  defmacro __before_compile__(_) do
-    quote do
-      def __record__(:optimizable), do: @record_optimizable
-    end
-  end
-
-  # Store fields that can be optimized and that cannot be
-  # optimized as they are overriden
-  @doc false
-  def __on_definition__(env, kind, name, args, _guards, _body) do
-    tuple     = { name, length(args) }
-    module    = env.module
-    functions = Module.get_attribute(module, :record_optimizable)
-
-    functions =
-      if kind in [:def] and Module.get_attribute(module, :record_optimized) do
-        [tuple|functions]
-      else
-        List.delete(functions, tuple)
-      end
-
-    Module.put_attribute(module, :record_optimizable, functions)
-  end
-
-  # Implements the access macro used by records.
-  # It returns a quoted expression that defines
-  # a record or a match in case the record is
-  # inside a match.
-  @doc false
-  def access(atom, fields, keyword, caller) do
-    unless is_keyword(keyword) do
-      raise ArgumentError, message: "expected contents inside brackets to be a keyword list, got: #{inspect keyword}"
-    end
-
-    in_match = caller.in_match?
-
-    has_underscore_value = Keyword.has_key?(keyword, :_)
-    underscore_value     = Keyword.get(keyword, :_, { :_, [], nil })
-    keyword              = Keyword.delete keyword, :_
-
-    iterator = fn({field, default}, each_keyword) ->
-      new_fields =
-        case Keyword.has_key?(each_keyword, field) do
-          true  -> Keyword.get(each_keyword, field)
-          false ->
-            case in_match or has_underscore_value do
-              true  -> underscore_value
-              false -> Macro.escape(default)
-            end
-        end
-
-      { new_fields, Keyword.delete(each_keyword, field) }
-    end
-
-    { match, remaining } = :lists.mapfoldl(iterator, keyword, fields)
-
-    case remaining do
-      [] ->
-        quote do: { unquote_splicing([atom|match]) }
-      _  ->
-        keys = lc { key, _ } inlist remaining, do: key
-        raise ArgumentError, message: "record #{inspect atom} does not have the keys: #{inspect keys}"
-    end
-  end
-
-  # Implements to_keywords macro defined by defmacros.
-  # It returns a quoted expression that represents
-  # converting record to keywords list.
-  @doc false
-  def to_keywords(_atom, fields, record) do
-    { var, extra } = cache_var(record)
-
-    keywords = Enum.map fields,
-      fn { key, _default } ->
-        index = find_index(fields, key, 0)
-        quote do
-          { unquote(key), :erlang.element(unquote(index + 2), unquote(var)) }
-        end
-      end
-
-    quote do
-      unquote_splicing(extra)
-      unquote(keywords)
-    end
-  end
-
-  # Dispatch the call to either get, update or to_list depending on the args given.
-  @doc false
-  def dispatch(atom, fields, record, args, caller) do
-    cond do
-      is_atom(args) ->
-        get(atom, fields, record, args)
-      is_keyword(args) ->
-        update(atom, fields, record, args, caller)
-      is_list(args) ->
-        to_list(atom, fields, record, args)
-      true ->
-        raise ArgumentError, message: "expected arguments to be a compile time atom, list or keywords"
-    end
-  end
-
-  # Implements the update macro defined by defmacros.
-  # It returns a quoted expression that represents
-  # the access given by the keywords.
-  defp update(atom, fields, var, keyword, caller) do
-    unless is_keyword(keyword) do
-      raise ArgumentError, message: "expected arguments to be compile time keywords"
-    end
-
-    if caller.in_match? do
-      raise ArgumentError, message: "cannot invoke update style macro inside match context"
-    end
-
-    Enum.reduce keyword, var, fn({ key, value }, acc) ->
-      index = find_index(fields, key, 0)
-      if index do
-        quote do
-          :erlang.setelement(unquote(index + 2), unquote(acc), unquote(value))
-        end
-      else
-        raise ArgumentError, message: "record #{inspect atom} does not have the key: #{inspect key}"
-      end
-    end
-  end
-
-  # Implements the get macro defined by defmacros.
-  # It returns a quoted expression that represents
-  # getting the value of a given field.
-  defp get(atom, fields, var, key) do
-    index = find_index(fields, key, 0)
-    if index do
-      quote do
-        :erlang.element(unquote(index + 2), unquote(var))
-      end
-    else
-      raise ArgumentError, message: "record #{inspect atom} does not have the key: #{inspect key}"
-    end
-  end
-
-  # Implements to_list macro defined by defmacros.
-  # It returns a quoted expression that represents
-  # extracting given fields from record.
-  @doc false
-  defp to_list(atom, fields, record, keys) do
-    unless is_list(fields) do
-      raise ArgumentError, message: "expected arguments to be a compile time list"
-    end
-
-    { var, extra } = cache_var(record)
-
-    list = Enum.map keys,
-      fn(key) ->
-        index = find_index(fields, key, 0)
-        if index do
-          quote do: :erlang.element(unquote(index + 2), unquote(var))
-        else
-          raise ArgumentError, message: "record #{inspect atom} does not have the key: #{inspect key}"
-        end
-      end
-
-    quote do
-      unquote_splicing(extra)
-      unquote(list)
-    end
-  end
-
-  defp cache_var({ var, _, kind } = tuple) when is_atom(var) and is_atom(kind) do
-    { tuple, [] }
-  end
-
-  defp cache_var(other) do
-    quote do
-      { x, [x = unquote(other)] }
-    end
-  end
-
-  ## Function generation
-
-  # Define __record__/1, __record__/2, __record__/3 as reflection
-  # functions that return the record names and fields.
-  #
-  # Note that fields are *not* keywords. They are in the same
-  # order as given as parameter and reflects the order of the
-  # fields in the tuple.
-  #
-  # ## Examples
-  #
-  #     defrecord FileInfo, atime: nil, mtime: nil
-  #
-  #     FileInfo.__record__(:name)   #=> FileInfo
-  #     FileInfo.__record__(:fields) #=> [atime: nil, mtime: nil]
-  #     FileInfo.__record__(:index, :atime) #=> 1
-  #     FileInfo.__record__(:index, :mtime) #=> 2
-  #
-  defp reflection(values) do
-    quoted = lc { k, _ } inlist values do
-      index = find_index(values, k, 0)
-      quote do
-        def __record__(:index, unquote(k)), do: unquote(index + 1)
-      end
-    end
-
-    quote do
-      unquote(quoted)
-
-      @doc false
-      def __record__(:index, _), do: nil
-      @doc false
-      def __record__(:index, arg, _), do: __record__(:index, arg)
-
-      @doc false
-      def __record__(kind, _), do: __record__(kind)
-
-      @doc false
-      def __record__(:name),   do: __MODULE__
-      def __record__(:fields), do: unquote(values)
-    end
-  end
-
-  # Define initializers methods. For a declaration like:
-  #
-  #     defrecord FileInfo, atime: nil, mtime: nil
-  #
-  # It will define three methods:
-  #
-  #     def new() do
-  #       new([])
-  #     end
-  #
-  #     def new([]) do
-  #       { FileInfo, nil, nil }
-  #     end
-  #
-  #     def new(opts) do
-  #       { FileInfo, Keyword.get(opts, :atime), Keyword.get(opts, :mtime) }
-  #     end
-  #
-  defp initializer(values) do
-    defaults = lc { _, value } inlist values, do: value
-
-    # For each value, define a piece of code that will receive
-    # an ordered dict of options (opts) and it will try to fetch
-    # the given key from the ordered dict, falling back to the
-    # default value if one does not exist.
-    atom_selective   = lc { k, v } inlist values, do: initialize_lookup(k, v)
-    string_selective = lc { k, v } inlist values, do: initialize_lookup(atom_to_binary(k), v)
-
-    quote do
-      @doc false
-      def new(), do: new([])
-
-      @doc false
-      def new([]), do: { __MODULE__, unquote_splicing(defaults) }
-      def new([{key, _}|_] = opts) when is_atom(key), do: { __MODULE__, unquote_splicing(atom_selective) }
-      def new([{key, _}|_] = opts) when is_binary(key), do: { __MODULE__, unquote_splicing(string_selective) }
-    end
-  end
-
-  defp initialize_lookup(k, v) do
-    quote do
-      case :lists.keyfind(unquote(k), 1, opts) do
-        false -> unquote(v)
-        {_, v} -> v
-      end
-    end
-  end
-
-  # Define converters method(s). For a declaration like:
-  #
-  #     defrecord FileInfo, atime: nil, mtime: nil
-  #
-  # It will define one method, to_keywords, which will return a Keyword
-  #
-  #    [atime: nil, mtime: nil]
-  #
-  defp conversions(values) do
-    sorted = lc { k, _ } inlist values do
-      index = find_index(values, k, 0)
-      { k, quote(do: :erlang.element(unquote(index + 2), record)) }
-    end
-
-    quote do
-      @doc false
-      def to_keywords(record) do
-        unquote(:orddict.from_list(sorted))
-      end
-    end
-  end
-
-  # Implement accessors. For a declaration like:
-  #
-  #     defrecord FileInfo, atime: nil, mtime: nil
-  #
-  # It will define four methods:
-  #
-  #     def atime(record) do
-  #       elem(record, 1)
-  #     end
-  #
-  #     def mtime(record) do
-  #       elem(record, 2)
-  #     end
-  #
-  #     def atime(value, record) do
-  #       set_elem(record, 1, value)
-  #     end
-  #
-  #     def mtime(record) do
-  #       set_elem(record, 2, value)
-  #     end
-  #
-  #     def atime(callback, record) do
-  #       set_elem(record, 1, callback.(elem(record, 1)))
-  #     end
-  #
-  #     def mtime(callback, record) do
-  #       set_elem(record, 2, callback.(elem(record, 2)))
-  #     end
-  #
-  defp accessors([{ :__exception__, _ }|t], 1) do
-    accessors(t, 2)
-  end
-
-  defp accessors([{ key, _default }|t], i) do
-    update = binary_to_atom "update_" <> atom_to_binary(key)
-
-    contents = quote do
-      @doc false
-      def unquote(key)(record) do
-        :erlang.element(unquote(i + 1), record)
-      end
-
-      @doc false
-      def unquote(key)(value, record) do
-        :erlang.setelement(unquote(i + 1), record, value)
-      end
-
-      @doc false
-      def unquote(update)(function, record) do
-        :erlang.setelement(unquote(i + 1), record,
-          function.(:erlang.element(unquote(i + 1), record)))
-      end
-    end
-
-    [contents|accessors(t, i + 1)]
-  end
-
-  defp accessors([], _i) do
-    []
-  end
-
-  # Define an updater method that receives a
-  # keyword list and updates the record.
-  defp updater(values) do
-    atom_fields =
-      lc {key, _default} inlist values, do: updater_lookup(key, key, values)
-
-    string_fields =
-      lc {key, _default} inlist values, do: updater_lookup(atom_to_binary(key), key, values)
-
-    atom_contents = quote do: { __MODULE__, unquote_splicing(atom_fields) }
-    string_contents = quote do: { __MODULE__, unquote_splicing(string_fields) }
-
-    quote do
-      @doc false
-      def update([], record) do
-        record
-      end
-
-      def update([{key, _}|_] = keywords, record) when is_atom(key) do
-        unquote(atom_contents)
-      end
-      def update([{key, _}|_] = keywords, record) when is_binary(key) do
-        unquote(string_contents)
-      end
-    end
-  end
-
-  defp updater_lookup(k, key, values) do
-    v = find_index(values, key, 1)
-
-    quote do
-      case :lists.keyfind(unquote(k), 1, keywords) do
-        false -> elem(record, unquote(v))
-        {_, value} -> value
-      end
-    end
-  end
-
-  defp record_optimizable do
-    quote do
-      @record_optimized true
-      @record_optimizable []
-      @before_compile { unquote(__MODULE__), :__before_compile__ }
-      @on_definition { unquote(__MODULE__), :__on_definition__ }
-    end
-  end
-
-  defp switch_recorder do
-    quote do: @record_optimized false
-  end
-
-  ## Types/specs generation
-
-  defp core_specs(values) do
-    types   = lc { _, _, spec } inlist values, do: spec
-    options = if values == [], do: [], else: [options_specs(values)]
-    values_specs = if values == [], do: [], else: values_specs(values)
-
-    quote do
-      unless Kernel.Typespec.defines_type?(__MODULE__, :t, 0) do
-        @type t :: { __MODULE__, unquote_splicing(types) }
-      end
-
-      unless Kernel.Typespec.defines_type?(__MODULE__, :options, 0) do
-        @type options :: unquote(options) | [{String.t, unquote(values_specs)}]
-      end
-
-      @spec new :: t
-      @spec new(options) :: t
-      @spec to_keywords(t) :: options
-      @spec update(options, t) :: t
-      @spec __record__(:name) :: atom
-      @spec __record__(:fields) :: [{atom, any}]
-      @spec __record__(:index, atom) :: non_neg_integer | nil
-    end
-  end
-
-  defp options_specs([{ k, _, v }|t]) do
-    :lists.foldl fn { k, _, v }, acc ->
-      { :|, [], [{ k, v }, acc] }
-    end, { k, v }, t
-  end
-  defp values_specs([{ _, _, v }|t]) do
-    :lists.foldl fn { _, _, v }, acc ->
-      { :|, [], [v, acc] }
-    end, v, t
-  end
-
-  defp accessor_specs([{ :__exception__, _, _ }|t], 1, acc) do
-    accessor_specs(t, 2, acc)
-  end
-
-  defp accessor_specs([{ key, _default, spec }|t], i, acc) do
-    update = binary_to_atom "update_" <> atom_to_binary(key)
-
-    contents = quote do
-      @spec unquote(key)(t) :: unquote(spec)
-      @spec unquote(key)(unquote(spec), t) :: t
-      @spec unquote(update)((unquote(spec) -> unquote(spec)), t) :: t
-    end
-
-    accessor_specs(t, i + 1, [contents | acc])
-  end
-
-  defp accessor_specs([], _i, acc), do: acc
-
-  ## Helpers
-
-  defp is_keyword(list) when is_list(list), do: :lists.all(&is_keyword_tuple/1, list)
-  defp is_keyword(_), do: false
-
-  defp is_keyword_tuple({ x, _ }) when is_atom(x), do: true
-  defp is_keyword_tuple(_), do: false
-
-  defp convert_value(atom) when is_atom(atom), do: { atom, nil }
-
-  defp convert_value({ atom, other }) when is_atom(atom), do:
-    { atom, check_value(atom, other) }
-
-  defp convert_value({ field, _ }), do:
-    raise(ArgumentError, message: "record field name has to be an atom, got #{inspect field}")
-
-  defp check_value(atom, other) when is_list(other) do
-    lc(i inlist other, do: check_value(atom, i))
-    other
-  end
-
-  defp check_value(atom, other) when is_tuple(other) do
-    lc(i inlist tuple_to_list(other), do: check_value(atom, i))
-    other
-  end
-
-  defp check_value(atom, other) when is_function(other) do
-    unless :erlang.fun_info(other, :env) == { :env, [] } and
-           :erlang.fun_info(other, :type) == { :type, :external } do
-      raise ArgumentError, message: "record field default value #{inspect atom} can only contain " <>
-                                    "functions that point to an existing &Mod.fun/arity"
-    end
-  end
-
-  defp check_value(atom, other) when is_reference(other) or is_pid(other) or is_port(other) do
-    raise(ArgumentError, message: "record field default value #{inspect atom} cannot contain a reference, pid or port")
-  end
-
-  defp check_value(_atom, other), do: other
-
-  defp find_index([{ k, _ }|_], k, i), do: i
-  defp find_index([{ _, _ }|t], k, i), do: find_index(t, k, i + 1)
-  defp find_index([], _k, _i), do: nil
-
-  defp find_spec(types, name) do
-    matches = lc { k, v } inlist types, name == k, do: v
-    case matches do
-      [h|_] -> h
-      _     -> quote do: term
-    end
-  end
-end
-
-defmodule Record.DSL do
-  @moduledoc false
-
-  @doc """
-  Defines the type for each field in the record.
-
-  Expects a keyword list.
-  """
-  defmacro record_type(opts) when is_list(opts) do
-    escaped = lc { k, v } inlist opts, do: { k, Macro.escape(v) }
-
-    quote do
-      @record_types Keyword.merge(@record_types || [], unquote(escaped))
-    end
-  end
-end
@@ -1,91 +0,0 @@
-defmodule Record.Extractor do
-  @moduledoc false
-
-  # Retrieve a record definition from an Erlang file using
-  # the same lookup as the *include* attribute from Erlang modules.
-  def retrieve(name, from: file) when is_binary(file) do
-    file = String.to_char_list!(file)
-
-    case :code.where_is_file(file) do
-      :non_existing -> realfile = file
-      realfile -> :ok
-    end
-
-    retrieve_record(name, realfile)
-  end
-
-  # Retrieve a record definition from an Erlang file using
-  # the same lookup as the *include_lib* attribute from Erlang modules.
-  def retrieve(name, from_lib: file) when is_binary(file) do
-    [app|path] = :filename.split(String.to_char_list!(file))
-
-    case :code.lib_dir(list_to_atom(app)) do
-      { :error, _ } ->
-        raise ArgumentError, message: "lib file #{file} could not be found"
-      libpath ->
-        retrieve_record name, :filename.join([libpath|path])
-    end
-  end
-
-  # Retrieve the record with the given name from the given file
-  defp retrieve_record(name, file) do
-    form = read_file(file)
-    records = retrieve_records(form)
-    if record = List.keyfind(records, name, 0) do
-      parse_record(record, form)
-    else
-      raise ArgumentError, message: "no record #{name} found at #{file}"
-    end
-  end
-
-  # Parse the given file and retrieve all existent records.
-  defp retrieve_records(form) do
-    lc { :attribute, _, :record, record } inlist form, do: record
-  end
-
-  # Read a file and return its abstract syntax form that also
-  # includes record and other preprocessor modules. This is done
-  # by using Erlang's epp_dodger.
-  defp read_file(file) do
-    case :epp_dodger.quick_parse_file(file) do
-      { :ok, form } ->
-        form
-      other ->
-        raise "error parsing file #{file}, got: #{inspect(other)}"
-    end
-  end
-
-  # Parse a tuple with name and fields and returns a
-  # list of tuples where the first element is the field
-  # and the second is its default value.
-  defp parse_record({ _name, fields }, form) do
-    cons = List.foldr fields, { nil, 0 }, fn f, acc ->
-      { :cons, 0, parse_field(f), acc }
-    end
-    eval_record(cons, form)
-  end
-
-  defp parse_field({ :typed_record_field, record_field, _type }) do
-    parse_field(record_field)
-  end
-
-  defp parse_field({ :record_field, _, key }) do
-    { :tuple, 0, [key, {:atom, 0, :undefined}] }
-  end
-
-  defp parse_field({ :record_field, _, key, value }) do
-    { :tuple, 0, [key, value] }
-  end
-
-  defp eval_record(cons, form) do
-    form = form ++
-      [ { :function, 0, :hello, 0, [
-          { :clause, 0, [], [], [ cons ] } ] } ]
-
-    { :function, 0, :hello, 0, [
-      { :clause, 0, [], [], [ record_ast ] } ] } = :erl_expand_records.module(form, []) |> List.last
-
-    { :value, record, _ } = :erl_eval.expr(record_ast, [])
-    record
-  end
-end
@@ -1,396 +0,0 @@
-defmodule Regex do
-  @moduledoc %S"""
-  Regular expressions for Elixir built on top of the `re` module
-  in the Erlang Standard Library. More information can be found
-  in the [`re` documentation](http://www.erlang.org/doc/man/re.html).
-
-  Regular expressions in Elixir can be created using `Regex.compile!`
-  or using the special form with `%r`:
-
-      # A simple regular expressions that matches foo anywhere in the string
-      %r/foo/
-
-      # A regular expression with case insensitive options
-      %r/foo/i
-
-  The `re` module provides several options, the ones available in Elixir, followed by
-  their shortcut in parenthesis, are:
-
-  * `caseless` (i) - add case insensitivity
-  * `dotall` (s) - causes dot to match newlines and also set newline to anycrlf.
-    The new line setting can be overridden by setting `(*CR)` or `(*LF)` or
-    `(*CRLF)` or `(*ANY)` according to re documentation
-  * `multiline` (m) - causes `^` and `$` to mark the beginning and end of each line.
-    Use `\A` and `\z` to match the end or beginning of the string
-  * `extended` (x) - whitespace characters are ignored except when escaped and
-    allow `#` to delimit comments
-  * `firstline` (f) - forces the unanchored pattern to match before or at the first
-    newline, though the matched text may continue over the newline
-  * `ungreedy` (r) - inverts the "greediness" of the regexp
-  * `groups` (g) - compiles with info about groups available
-
-  The options not available are:
-
-  * `anchored` - not available, use `^` or `\A` instead
-  * `dollar_endonly` - not available, use `\z` instead
-  * `no_auto_capture` - not available, use `?:` instead
-  * `newline` - not available, use `(*CR)` or `(*LF)` or `(*CRLF)` or `(*ANYCRLF)`
-    or `(*ANY)` at the beginning of the regexp according to the re documentation
-
-  Most of the functions in this module accept either a binary or a char list
-  as subject. The result is based on the argument (a binary will return
-  a binary, a char list will return a char list).
-  """
-
-  defrecordp :regex, Regex, [:re_pattern, :source, :options, :groups]
-  @type t :: { Regex, term, binary, binary, [atom] | nil }
-
-  defexception CompileError, message: "regex could not be compiled"
-
-  @doc """
-  Compiles the regular expression.
-
-  The given options can either be a binary with the characters
-  representing the same regex options given to the `%r` sigil,
-  or a list of options, as expected by the [Erlang `re` docs](http://www.erlang.org/doc/man/re.html).
-
-  It returns `{ :ok, regex }` in case of success,
-  `{ :error, reason }` otherwise.
-
-  ## Examples
-
-      iex> Regex.compile("foo")
-      {:ok, %r"foo"}
-      iex> Regex.compile("*foo")
-      {:error, {'nothing to repeat', 0}}
-
-  """
-  @spec compile(binary, binary | [term]) :: { :ok, t } | { :error, any }
-  def compile(source, options // "")
-
-  def compile(source, options) when is_binary(options) do
-    case translate_options(options) do
-      { :error, rest } ->
-        { :error, { :invalid_option, rest } }
-
-      translated_options ->
-        # Always use the unicode option, we don't have a latin1 legacy like
-        # Erlang.
-        compile(source, [:unicode|translated_options], options)
-    end
-  end
-
-  def compile(source, options) when is_list(options) do
-    compile(source, options, "")
-  end
-
-  defp compile(source, opts, doc_opts) when is_binary(source) do
-    re_opts = opts -- [:groups]
-    groups  = if opts != re_opts, do: parse_groups(source)
-
-    case :re.compile(source, re_opts) do
-      { :ok, re_pattern } ->
-        { :ok, regex(re_pattern: re_pattern, source: source, options: doc_opts, groups: groups) }
-      error ->
-        error
-    end
-  end
-
-  @doc """
-  Compiles the regular expression according to the given options.
-  Fails with `Regex.CompileError` if the regex cannot be compiled.
-  """
-  def compile!(source, options // "") do
-    case compile(source, options) do
-      { :ok, regex } -> regex
-      { :error, { reason, at } } -> raise Regex.CompileError, message: "#{reason} at position #{at}"
-    end
-  end
-
-  @doc """
-  Returns a boolean indicating whether there was a match or not.
-
-  ## Examples
-
-      iex> Regex.match?(%r/foo/, "foo")
-      true
-      iex> Regex.match?(%r/foo/, "bar")
-      false
-
-  """
-  def match?(regex(re_pattern: compiled), string) do
-    :re.run(string, compiled, [{ :capture, :none }]) == :match
-  end
-
-  @doc """
-  Runs the regular expression against the given string until the first match.
-  It returns a list with all captures or `nil` if no match occurred.
-
-  When the option `:capture` is set to `:groups`, it will capture all
-  the groups in the regex.
-
-  ## Examples
-
-      iex> Regex.run(%r/c(d)/, "abcd")
-      ["cd", "d"]
-      iex> Regex.run(%r/e/, "abcd")
-      nil
-      iex> Regex.run(%r/c(d)/, "abcd", return: :index)
-      [{2,2},{3,1}]
-
-  """
-  def run(regex, string, options // [])
-
-  def run(regex(re_pattern: compiled, groups: groups), string, options) do
-    return = Keyword.get(options, :return, return_for(string))
-
-    captures =
-      case Keyword.get(options, :capture, :all) do
-        :groups -> groups || raise ArgumentError, message: "regex was not compiled with g"
-        others  -> others
-      end
-
-    case :re.run(string, compiled, [{ :capture, captures, return }]) do
-      :nomatch -> nil
-      :match   -> []
-      { :match, results } -> results
-    end
-  end
-
-  @doc """
-  Returns the given captures as a keyword list or `nil` if no captures
-  are found. Requires the regex to be compiled with the groups option.
-
-  ## Examples
-
-      iex> Regex.named_captures(%r/c(?<foo>d)/g, "abcd")
-      [foo: "d"]
-      iex> Regex.named_captures(%r/a(?<foo>b)c(?<bar>d)/g, "abcd")
-      [foo: "b", bar: "d"]
-      iex> Regex.named_captures(%r/a(?<foo>b)c(?<bar>d)/g, "efgh")
-      nil
-
-  """
-  def named_captures(regex(groups: groups) = regex, string, options // []) do
-    options = Keyword.put_new(options, :capture, :groups)
-    results = run(regex, string, options)
-    if results, do: Enum.zip(groups, results)
-  end
-
-  @doc """
-  Returns the underlying `re_pattern` in the regular expression.
-  """
-  def re_pattern(regex(re_pattern: compiled)) do
-    compiled
-  end
-
-  @doc """
-  Returns the regex source as a binary.
-
-  ## Examples
-
-      iex> Regex.source(%r(foo))
-      "foo"
-
-  """
-  def source(regex(source: source)) do
-    source
-  end
-
-  @doc """
-  Returns the regex options as a string.
-
-  ## Examples
-
-      iex> Regex.opts(%r(foo)m)
-      "m"
-
-  """
-  def opts(regex(options: options)) do
-    options
-  end
-
-  @doc """
-  Returns a list of named groups in the regex.
-
-  ## Examples
-
-      iex> Regex.groups(%r/(?<foo>bar)/g)
-      [:foo]
-
-  """
-  def groups(regex(groups: groups)) do
-    groups
-  end
-
-  @doc """
-  Same as `run/3`, but scans the target several times collecting all
-  matches of the regular expression. A list of lists is returned,
-  where each entry in the primary list represents a match and each
-  entry in the secondary list represents the captured contents.
-
-  The captured contents defaults to `:all`, which includes the whole
-  regex match and each capture.
-
-  When the option `:capture` is set to `:groups`, it will capture all
-  the groups in the regex.
-
-  ## Examples
-
-      iex> Regex.scan(%r/c(d|e)/, "abcd abce")
-      [["cd", "d"], ["ce", "e"]]
-      iex> Regex.scan(%r/c(?:d|e)/, "abcd abce")
-      [["cd"], ["ce"]]
-      iex> Regex.scan(%r/e/, "abcd")
-      []
-
-  """
-  def scan(regex, string, options // [])
-
-  def scan(regex(re_pattern: compiled, groups: groups), string, options) do
-    return  = Keyword.get(options, :return, return_for(string))
-
-    captures =
-      case Keyword.get(options, :capture, :all) do
-        :groups -> groups || raise ArgumentError, message: "regex was not compiled with g"
-        others  -> others
-      end
-
-    options = [{ :capture, captures, return }, :global]
-    case :re.run(string, compiled, options) do
-      :match -> []
-      :nomatch -> []
-      { :match, results } -> results
-    end
-  end
-
-  @doc """
-  Splits the given target into the number of parts specified.
-  If no number of parts is given, it defaults to `:infinity`.
-
-  ## Examples
-      iex> Regex.split(%r/-/, "a-b-c")
-      ["a","b","c"]
-      iex> Regex.split(%r/-/, "a-b-c", [parts: 2])
-      ["a","b-c"]
-      iex> Regex.split(%r/-/, "abc")
-      ["abc"]
-      iex> Regex.split(%r//, "abc")
-      ["a", "b", "c", ""]
-      iex> Regex.split(%r//, "abc", trim: true)
-      ["a", "b", "c"]
-  """
-
-  def split(regex, string, options // [])
-
-  def split(regex(re_pattern: compiled), string, options) do
-    parts =
-      cond do
-        Keyword.get(options, :global) == false -> 2
-        p = Keyword.get(options, :parts)       -> p
-        true                                   -> :infinity
-      end
-
-    return = Keyword.get(options, :return, return_for(string))
-    opts   = [return: return, parts: parts]
-    splits = :re.split(string, compiled, opts)
-
-    if Keyword.get(options, :trim, false) do
-      lc split inlist splits, split != "", do: split
-    else
-      splits
-    end
-  end
-
-  @doc %S"""
-  Receives a regex, a binary and a replacement, returns a new
-  binary where the all matches are replaced by replacement.
-
-  Inside the replacement, you can either give `&` to access the
-  whole regular expression or `\N`, where `N` is in integer to access
-  a specific matching parens. You can also set `:global` to `false`
-  if you want to replace just the first occurrence.
-
-  ## Examples
-
-      iex> Regex.replace(%r/d/, "abc", "d")
-      "abc"
-      iex> Regex.replace(%r/b/, "abc", "d")
-      "adc"
-      iex> Regex.replace(%r/b/, "abc", "[&]")
-      "a[b]c"
-      iex> Regex.replace(%r/b/, "abc", "[\\&]")
-      "a[&]c"
-      iex> Regex.replace(%r/(b)/, "abc", "[\\1]")
-      "a[b]c"
-
-  """
-  def replace(regex(re_pattern: compiled), string, replacement, options // []) do
-    opts   = if Keyword.get(options, :global) != false, do: [:global], else: []
-    return = Keyword.get(options, :return, return_for(string))
-    opts   = [{ :return, return }|opts]
-    :re.replace(string, compiled, replacement, opts)
-  end
-
-  { :ok, pattern } = :re.compile(%S"[.^$*+?()[{\\\|\s#]", [:unicode])
-  @escape_pattern pattern
-
-  @doc %S"""
-  Escapes a string to be literally matched in a regex.
-
-  ## Examples
-
-      iex> Regex.escape(".")
-      "\\."
-      iex> Regex.escape("\\what if")
-      "\\\\what\\ if"
-
-  """
-  @spec escape(String.t | char_list) :: String.t | char_list
-  def escape(string) do
-    :re.replace(string, @escape_pattern, "\\\\&", [:global, { :return, return_for(string) }])
-  end
-
-  # Helpers
-
-  @doc false
-  # Unescape map function used by Macro.unescape_string.
-  def unescape_map(?f), do: ?\f
-  def unescape_map(?n), do: ?\n
-  def unescape_map(?r), do: ?\r
-  def unescape_map(?t), do: ?\t
-  def unescape_map(?v), do: ?\v
-  def unescape_map(?a), do: ?\a
-  def unescape_map(_),  do: false
-
-  # Private Helpers
-
-  defp return_for(element) when is_binary(element), do: :binary
-  defp return_for(element) when is_list(element),   do: :list
-
-  defp translate_options(<<?u, t :: binary>>) do
-    IO.write "The /u flag for regular expressions is no longer needed\n#{Exception.format_stacktrace}"
-    translate_options(t)
-  end
-  defp translate_options(<<?i, t :: binary>>), do: [:caseless|translate_options(t)]
-  defp translate_options(<<?x, t :: binary>>), do: [:extended|translate_options(t)]
-  defp translate_options(<<?f, t :: binary>>), do: [:firstline|translate_options(t)]
-  defp translate_options(<<?r, t :: binary>>), do: [:ungreedy|translate_options(t)]
-  defp translate_options(<<?s, t :: binary>>), do: [:dotall, {:newline, :anycrlf}|translate_options(t)]
-  defp translate_options(<<?m, t :: binary>>), do: [:multiline|translate_options(t)]
-  defp translate_options(<<?g, t :: binary>>), do: [:groups|translate_options(t)]
-  defp translate_options(<<>>), do: []
-  defp translate_options(rest), do: { :error, rest }
-
-  { :ok, pattern } = :re.compile(%S"\(\?<(?<G>[^>]*)>")
-  @groups_pattern pattern
-
-  defp parse_groups(source) do
-    options = [:global, {:capture, ['G'], :binary}]
-    case :re.run(source, @groups_pattern, options) do
-      :nomatch -> []
-      { :match, results } ->
-        lc [group] inlist results, do: binary_to_atom(group)
-    end
-  end
-end
@@ -1,248 +0,0 @@
-defmodule Set do
-  @moduledoc %S"""
-  This module specifies the Set API expected to be
-  implemented by different representations.
-
-  It also provides functions that redirect to the
-  underlying Set, allowing a developer to work with
-  different Set implementations using one API.
-
-  To create a new set, use the `new` functions defined
-  by each set type:
-
-      HashSet.new  #=> creates an empty HashSet
-
-  For simplicity's sake, in the examples below every time
-  `new` is used, it implies one of the module-specific
-  calls like above.
-
-  ## Protocols
-
-  Sets are required to implement the `Enumerable` protocol,
-  allowing one to write:
-
-      Enum.each(set, fn k ->
-        IO.inspect k
-      end)
-
-  ## Match
-
-  Sets are required to implement all operations
-  using the match (`===`) operator. Any deviation from
-  this behaviour should be avoided and explicitly documented.
-  """
-
-  use Behaviour
-
-  @type value :: any
-  @type values :: [ value ]
-  @type t :: tuple
-
-  defcallback delete(t, value) :: t
-  defcallback difference(t, t) :: t
-  defcallback disjoint?(t, t) :: boolean
-  defcallback empty(t) :: t
-  defcallback equal?(t, t) :: boolean
-  defcallback intersection(t, t) :: t
-  defcallback member?(t, value) :: boolean
-  defcallback put(t, value) :: t
-  defcallback size(t) :: non_neg_integer
-  defcallback subset?(t, t) :: boolean
-  defcallback to_list(t) :: list()
-  defcallback union(t, t) :: t
-
-  defmacrop target(set) do
-    quote do
-      if is_tuple(unquote(set)) do
-        elem(unquote(set), 0)
-      else
-        unsupported_set(unquote(set))
-      end
-    end
-  end
-
-  @doc """
-  Deletes `value` from `set`.
-
-  ## Examples
-
-      iex> s = HashSet.new([1, 2, 3])
-      ...> Set.delete(s, 4) |> HashSet.to_list
-      [1, 2, 3]
-
-      iex> s = HashSet.new([1, 2, 3])
-      ...> Set.delete(s, 2) |> HashSet.to_list
-      [1, 3]
-  """
-  @spec delete(t, value) :: t
-  def delete(set, value) do
-    target(set).delete(set, value)
-  end
-
-  @doc """
-  Returns a set that is `set1` without the members of `set2`.
-
-  ## Examples
-
-      iex> Set.difference(HashSet.new([1,2]), HashSet.new([2,3,4])) |> HashSet.to_list
-      [1]
-
-  """
-  @spec difference(t, t) :: t
-  def difference(set1, set2) do
-    target(set1).difference(set1, set2)
-  end
-
-  @doc """
-  Checks if `set1` and `set2` have no members in common.
-
-  ## Examples
-
-      iex> Set.disjoint?(HashSet.new([1, 2]), HashSet.new([3, 4]))
-      true
-      iex> Set.disjoint?(HashSet.new([1, 2]), HashSet.new([2, 3]))
-      false
-
-  """
-  @spec disjoint?(t, t) :: boolean
-  def disjoint?(set1, set2) do
-    target(set1).disjoint?(set1, set2)
-  end
-
-  @doc """
-  Returns an empty set of the same type as `set`.
-  """
-  @spec empty(t) :: t
-  def empty(set) do
-    target(set).empty(set)
-  end
-
-  @doc """
-  Checks if `set1` and `set2` are equal.
-
-  ## Examples
-
-      iex> Set.equal?(HashSet.new([1, 2]), HashSet.new([2, 1, 1]))
-      true
-      
-      iex> Set.equal?(HashSet.new([1, 2]), HashSet.new([3, 4]))
-      false
-
-  """
-  @spec equal?(t, t) :: boolean
-  def equal?(set1, set2) do
-    target(set1).equal?(set1, set2)
-  end
-
-  @doc """
-  Returns a set containing only members in common between `set1` and `set2`.
-
-  ## Examples
-
-      iex> Set.intersection(HashSet.new([1,2]), HashSet.new([2,3,4])) |> HashSet.to_list
-      [2]
-
-      iex> Set.intersection(HashSet.new([1,2]), HashSet.new([3,4])) |> HashSet.to_list  
-      []
-
-  """
-  @spec intersection(t, t) :: t
-  def intersection(set1, set2) do
-    target(set1).intersection(set1, set2)
-  end
-
-  @doc """
-  Checks if `set` contains `value`.
-
-  ## Examples
-
-      iex> Set.member?(HashSet.new([1, 2, 3]), 2)
-      true
-
-      iex> Set.member?(HashSet.new([1, 2, 3]), 4) 
-      false
-
-  """
-  @spec member?(t, value) :: boolean
-  def member?(set, value) do
-    target(set).member?(set, value)
-  end
-
-  @doc """
-  Inserts `value` into `set` if it does not already contain it.
-
-  ## Examples
-
-      iex> Set.put(HashSet.new([1, 2, 3]), 3) |> Set.to_list
-      [1, 2, 3]
-
-      iex> Set.put(HashSet.new([1, 2, 3]), 4) |> Set.to_list
-      [1, 2, 3, 4]
-
-  """
-  @spec put(t, value) :: t
-  def put(set, value) do
-    target(set).put(set, value)
-  end
-
-  @doc """
-  Returns the number of elements in `set`.
-
-  ## Examples
-
-      iex> Set.size(HashSet.new([1, 2, 3]))
-      3
-
-  """
-  @spec size(t) :: non_neg_integer
-  def size(set) do
-    target(set).size(set)
-  end
-
-  @doc """
-  Checks if `set1`'s members are all contained in `set2`.
-
-  ## Examples
-
-      iex> Set.subset?(HashSet.new([1, 2]), HashSet.new([1, 2, 3]))
-      true
-      iex> Set.subset?(HashSet.new([1, 2, 3]), HashSet.new([1, 2]))
-      false
-  """
-  @spec subset?(t, t) :: boolean
-  def subset?(set1, set2) do
-    target(set1).subset?(set1, set2)
-  end
-
-  @doc """
-  Converts `set` to a list.
-
-  ## Examples
-
-      iex> HashSet.to_list(HashSet.new([1, 2, 3]))
-      [1,2,3]
-
-  """
-  @spec to_list(t) :: list
-  def to_list(set) do
-    target(set).to_list(set)
-  end
-
-  @doc """
-  Returns a set containing all members of `set1` and `set2`.
-
-  ## Examples
-
-      iex> Set.union(HashSet.new([1,2]), HashSet.new([2,3,4])) |> HashSet.to_list
-      [1,2,3,4]
-
-  """
-  @spec union(t, t) :: t
-  def union(set1, set2) do
-    target(set1).union(set1, set2)
-  end
-
-  defp unsupported_set(set) do
-    raise ArgumentError, message: "unsupported set: #{inspect set}"
-  end
-end
@@ -1,494 +0,0 @@
-defmodule Stream do
-  @moduledoc """
-  Module for creating and composing streams.
-
-  Streams are composable, lazy enumerables. Any enumerable that generates
-  items one by one during enumeration is called a stream. For example,
-  Elixir's `Range` is a stream:
-
-      iex> range = 1..5
-      1..5
-      iex> Enum.map range, &(&1 * 2)
-      [2,4,6,8,10]
-
-  In the example above, as we mapped over the range, the elements being
-  enumerated were created one by one, during enumeration. The `Stream`
-  module allows us to map the range, without triggering its enumeration:
-
-      iex> range = 1..3
-      iex> stream = Stream.map(range, &(&1 * 2))
-      iex> Enum.map(stream, &(&1 + 1))
-      [3,5,7]
-
-  Notice we started with a range and then we created a stream that is
-  meant to multiply each item in the range by 2. At this point, no
-  computation was done yet. Just when `Enum.map/2` is called we
-  enumerate over each item in the range, multiplying it by 2 and adding 1.
-  We say the functions in `Stream` are *lazy* and the functions in `Enum`
-  are *eager*.
-
-  Due to their laziness, streams are useful when working with large
-  (or even infinite) collections. When chaining many operations with `Enum`,
-  intermediate lists are created, while `Stream` creates a recipe of
-  computations that are executed at a later moment. Let's see another
-  example:
-
-      1..3 |>
-        Enum.map(&IO.inspect(&1)) |>
-        Enum.map(&(&1 * 2)) |>
-        Enum.map(&IO.inspect(&1))
-      1
-      2
-      3
-      2
-      4
-      6
-      #=> [2,4,6]
-
-  Notice that we first printed each item in the list, then multiplied each
-  element by 2 and finally printed each new value. In this example, the list
-  was iterated three times. Let's see an example with streams:
-
-      stream = 1..3 |>
-        Stream.map(&IO.inspect(&1)) |>
-        Stream.map(&(&1 * 2)) |>
-        Stream.map(&IO.inspect(&1))
-      Enum.to_list(stream)
-      1
-      2
-      2
-      4
-      3
-      6
-      #=> [2,4,6]
-
-  Although the end result is the same, the order in which the items were
-  printed changed! With streams, we print the first item and then print
-  its double. In this example, the list was iterated just once!
-
-  That's what we meant when we first said that streams are composable,
-  lazy enumerables. Notice we could call `Stream.map/2` multiple times,
-  effectively composing the streams and they are lazy. The computations
-  are performed only when you call a function from the `Enum` module.
-
-  ## Creating Streams
-
-  There are many functions in Elixir's standard library that return
-  streams, some examples are:
-
-  * `IO.stream/1` - Streams input lines, one by one;
-  * `URI.query_decoder/1` - Decodes a query string, pair by pair;
-
-  This module also allows us to create streams from any enumerable:
-
-      iex> stream = Stream.map([1,2,3], &(&1 * 2))
-      iex> Enum.map(stream, &(&1 + 1))
-      [3,5,7]
-
-  By simply passing a list (which is an enumerable) as the first argument
-  to `Stream.map/2`, we have automatically created a stream that will
-  multiply the items in the list by 2 on enumeration.
-
-  This module also provides other functions for creating streams, such as
-  `Stream.cycle/1`.
-  """
-
-  defrecord Lazy, [:enumerable, :fun, :acc]
-
-  defimpl Enumerable, for: Lazy do
-    def reduce(Lazy[] = lazy, acc, fun) do
-      do_reduce(lazy, acc, fun, 0)
-    end
-
-    def count(Lazy[] = lazy) do
-      do_reduce(lazy, 0, fn _, acc -> acc + 1 end, 0)
-    end
-
-    def member?(Lazy[] = lazy, value) do
-      do_reduce(lazy, false, fn(entry, _) ->
-        if entry === value, do: throw({ :stream_lazy, 0, true }), else: false
-      end, 0)
-    end
-
-    defp do_reduce(Lazy[enumerable: enumerable, fun: f1, acc: nil], acc, fun, nesting) do
-      do_reduce(enumerable, acc, f1.(fun), nesting)
-    end
-
-    defp do_reduce(Lazy[enumerable: enumerable, fun: f1, acc: side], acc, fun, nesting) do
-      do_reduce(enumerable, { acc, side }, f1.(fun, nesting), nesting + 1)
-    end
-
-    defp do_reduce(enumerable, acc, fun, nesting) do
-      Enumerable.reduce(enumerable, acc, fun) |> remove_nesting(nesting)
-    catch
-      { :stream_lazy, nesting, res } -> remove_nesting(res, nesting)
-    end
-
-    defp remove_nesting(acc, 0),       do: acc
-    defp remove_nesting(acc, nesting), do: remove_nesting(elem(acc, 0), nesting - 1)
-  end
-
-  @type t :: Lazy.t | (acc, (element, acc -> acc) -> acc)
-  @type acc :: any
-  @type element :: any
-  @type index :: non_neg_integer
-  @type default :: any
-
-  @doc """
-  Creates a stream that enumerates each enumerable in an enumerable.
-
-  ## Examples
-
-      iex> stream = Stream.concat([1..3, 4..6, 7..9])
-      iex> Enum.to_list(stream)
-      [1,2,3,4,5,6,7,8,9]
-
-  """
-  @spec concat(Enumerable.t) :: t
-  def concat(enumerables) do
-    &do_concat(enumerables, &1, &2)
-  end
-
-  @doc """
-  Creates a stream that enumerates the first argument, followed by the second.
-
-  ## Examples
-
-      iex> stream = Stream.concat(1..3, 4..6)
-      iex> Enum.to_list(stream)
-      [1,2,3,4,5,6]
-
-      iex> stream1 = Stream.cycle([1, 2, 3])
-      iex> stream2 = Stream.cycle([4, 5, 6])
-      iex> stream = Stream.concat(stream1, stream2)
-      iex> Enum.take(stream, 6)
-      [1,2,3,1,2,3]
-
-  """
-  @spec concat(Enumerable.t, Enumerable.t) :: t
-  def concat(first, second) do
-    &do_concat([first, second], &1, &2)
-  end
-
-  defp do_concat(enumerables, acc, fun) do
-    Enumerable.reduce(enumerables, acc, &Enumerable.reduce(&1, &2, fun))
-  end
-
-  @doc """
-  Creates a stream that cycles through the given enumerable,
-  infinitely.
-
-  ## Examples
-
-      iex> stream = Stream.cycle([1,2,3])
-      iex> Enum.take(stream, 5)
-      [1,2,3,1,2]
-
-  """
-  @spec cycle(Enumerable.t) :: t
-  def cycle(enumerable) do
-    &do_cycle(enumerable, &1, &2)
-  end
-
-  defp do_cycle(enumerable, acc, fun) do
-    acc = Enumerable.reduce(enumerable, acc, fun)
-    do_cycle(enumerable, acc, fun)
-  end
-
-  @doc """
-  Lazily drops the next `n` items from the enumerable.
-
-  ## Examples
-
-      iex> stream = Stream.drop(1..10, 5)
-      iex> Enum.to_list(stream)
-      [6,7,8,9,10]
-
-  """
-  @spec drop(Enumerable.t, non_neg_integer) :: t
-  def drop(enumerable, n) when n >= 0 do
-    Lazy[enumerable: enumerable,
-         fun: fn(f1, _) ->
-           fn
-             _entry, { acc, n } when n > 0 ->
-               { acc, n - 1 }
-             entry, { acc, n } ->
-               { f1.(entry, acc), n }
-           end
-         end,
-         acc: n]
-  end
-
-  @doc """
-  Lazily drops elements of the enumerable while the given
-  function returns true.
-
-  ## Examples
-
-      iex> stream = Stream.drop_while(1..10, &(&1 <= 5))
-      iex> Enum.to_list(stream)
-      [6,7,8,9,10]
-
-  """
-  @spec drop_while(Enumerable.t, (element -> as_boolean(term))) :: t
-  def drop_while(enumerable, f) do
-    Lazy[enumerable: enumerable,
-         fun: fn(f1, _) ->
-           fn
-             entry, { acc, true } ->
-               if f.(entry), do: { acc, true }, else: { f1.(entry, acc), false }
-             entry, { acc, false } ->
-               { f1.(entry, acc), false }
-           end
-         end,
-         acc: true]
-  end
-
-  @doc """
-  Creates a stream that will filter elements according to
-  the given function on enumeration.
-
-  ## Examples
-
-      iex> stream = Stream.filter([1, 2, 3], fn(x) -> rem(x, 2) == 0 end)
-      iex> Enum.to_list(stream)
-      [2]
-
-  """
-  @spec filter(Enumerable.t, (element -> as_boolean(term))) :: t
-  def filter(enumerable, f) do
-    Lazy[enumerable: enumerable,
-         fun: fn(f1) ->
-           fn(entry, acc) ->
-             if f.(entry), do: f1.(entry, acc), else: acc
-           end
-         end]
-  end
-
-  @doc """
-  Emit a sequence of values, starting with `start_value`. Successive
-  values are generated by calling `next_fun` on the previous value.
-
-  ## Examples
-
-      iex> Stream.iterate(0, &(&1+1)) |> Enum.take(5)
-      [0,1,2,3,4]
-
-  """
-
-  @spec iterate(element, (element -> element)) :: t
-  def iterate(start_value, next_fun) do
-    fn acc, fun ->
-      do_iterate(start_value, next_fun, fun.(start_value, acc), fun)
-    end
-  end
-
-  defp do_iterate(value, next_fun, acc, fun) do
-    next = next_fun.(value)
-    do_iterate(next, next_fun, fun.(next, acc), fun)
-  end
-
-  @doc """
-  Creates a stream that will apply the given function on
-  enumeration.
-
-  ## Examples
-
-      iex> stream = Stream.map([1, 2, 3], fn(x) -> x * 2 end)
-      iex> Enum.to_list(stream)
-      [2,4,6]
-
-  """
-  @spec map(Enumerable.t, (element -> any)) :: t
-  def map(enumerable, f) do
-    Lazy[enumerable: enumerable,
-         fun: fn(f1) ->
-           fn(entry, acc) ->
-             f1.(f.(entry), acc)
-           end
-         end]
-  end
-
-  @doc """
-  Creates a stream that will apply the given function on enumeration and
-  flatten the result.
-
-  ## Examples
-
-      iex> stream = Stream.flat_map([1, 2, 3], fn(x) -> [x, x * 2] end)
-      iex> Enum.to_list(stream)
-      [1, 2, 2, 4, 3, 6]
-
-  """
-
-  @spec flat_map(Enumerable.t, (element -> any)) :: t
-  def flat_map(enumerable, f) do
-    Lazy[enumerable: enumerable,
-         fun: fn(f1) ->
-           fn(entry, acc) -> do_flat_map(f.(entry), acc, f1) end
-         end]
-  end
-
-  defp do_flat_map(Lazy[] = lazy, acc, f1) do
-    try do
-      Enumerable.reduce(lazy, acc, fn x, y ->
-        try do
-          f1.(x, y)
-        catch
-          { :stream_lazy, nesting, rest } ->
-            throw({ :stream_flat_map, nesting, rest })
-        end
-      end)
-    catch
-      { :stream_flat_map, nesting, rest } ->
-        throw({ :stream_lazy, nesting, rest })
-    end
-  end
-
-  defp do_flat_map(enum, acc, f1) do
-    Enumerable.reduce(enum, acc, f1)
-  end
-
-  @doc """
-  Creates a stream that will reject elements according to
-  the given function on enumeration.
-
-  ## Examples
-
-      iex> stream = Stream.reject([1, 2, 3], fn(x) -> rem(x, 2) == 0 end)
-      iex> Enum.to_list(stream)
-      [1,3]
-
-  """
-  @spec reject(Enumerable.t, (element -> as_boolean(term))) :: t
-  def reject(enumerable, f) do
-    Lazy[enumerable: enumerable,
-         fun: fn(f1) ->
-           fn(entry, acc) ->
-             unless f.(entry), do: f1.(entry, acc), else: acc
-           end
-         end]
-  end
-
-  @doc """
-  Returns a stream generated by calling `generator_fun` repeatedly.
-
-  ## Examples
-
-      iex> Stream.repeatedly(&:random.uniform/0) |> Enum.take(3)
-      [0.4435846174457203, 0.7230402056221108, 0.94581636451987]
-
-  """
-  @spec repeatedly((() -> element)) :: t
-  def repeatedly(generator_fun)
-  when is_function(generator_fun, 0) do
-    &do_repeatedly(generator_fun, &1, &2)
-  end
-
-  defp do_repeatedly(generator_fun, acc, fun) do
-    do_repeatedly(generator_fun, fun.(generator_fun.(), acc), fun)
-  end
-
-  @doc """
-  Lazily takes the next `n` items from the enumerable and stops
-  enumeration.
-
-  ## Examples
-
-      iex> stream = Stream.take(1..100, 5)
-      iex> Enum.to_list(stream)
-      [1,2,3,4,5]
-
-      iex> stream = Stream.cycle([1, 2, 3]) |> Stream.take(5)
-      iex> Enum.to_list(stream)
-      [1,2,3,1,2]
-
-  """
-  @spec take(Enumerable.t, non_neg_integer) :: t
-  def take(_enumerable, 0), do: Lazy[enumerable: [], fun: &(&1)]
-
-  def take(enumerable, n) when n > 0 do
-    Lazy[enumerable: enumerable,
-         fun: fn(f1, nesting) ->
-           fn(entry, { acc, n }) ->
-             res = f1.(entry, acc)
-             if n > 1, do: { res, n-1 }, else: throw { :stream_lazy, nesting, res }
-           end
-         end,
-         acc: n]
-  end
-
-  @doc """
-  Lazily takes elements of the enumerable while the given
-  function returns true.
-
-  ## Examples
-
-      iex> stream = Stream.take_while(1..100, &(&1 <= 5))
-      iex> Enum.to_list(stream)
-      [1,2,3,4,5]
-
-  """
-  @spec take_while(Enumerable.t, (element -> as_boolean(term))) :: t
-  def take_while(enumerable, f) do
-    Lazy[enumerable: enumerable,
-         fun: fn(f1, nesting) ->
-           fn(entry, { acc, true }) ->
-             if f.(entry) do
-               { f1.(entry, acc), true }
-             else
-               throw { :stream_lazy, nesting, acc }
-             end
-           end
-         end,
-         acc: true]
-  end
-
-  @doc """
-  Emit a sequence of values and accumulators. Successive values are generated by
-  calling `next_fun` with the previous accumulator.
-
-  If the return value is nil iteration ends.
-
-  ## Examples
-
-      iex> Stream.unfold(5, fn 0 -> nil; n -> {n, n-1} end) |> Enum.to_list()
-      [5, 4, 3, 2, 1]
-  """
-  @spec unfold(acc, (acc -> { element, acc } | nil)) :: t
-  def unfold(acc, f) do
-    fn acc1, f1 ->
-      do_unfold(acc, f, acc1, f1)
-    end
-  end
-
-  defp do_unfold(gen_acc, gen_fun, acc, fun) do
-    case gen_fun.(gen_acc) do
-      nil                -> acc
-      { v, new_gen_acc } -> do_unfold(new_gen_acc, gen_fun, fun.(v, acc), fun)
-    end
-  end
-
-  @doc """
-  Creates a stream where each item in the enumerable will
-  be accompanied by its index.
-
-  ## Examples
-
-      iex> stream = Stream.with_index([1, 2, 3])
-      iex> Enum.to_list(stream)
-      [{1,0},{2,1},{3,2}]
-
-  """
-  @spec with_index(Enumerable.t) :: t
-  def with_index(enumerable) do
-    Lazy[enumerable: enumerable,
-         fun: fn(f1, _) ->
-           fn(entry, { acc, counter }) ->
-             acc = f1.({ entry, counter }, acc)
-             { acc, counter + 1 }
-           end
-         end,
-         acc: 0]
-  end
-end
@@ -1,84 +0,0 @@
-import Kernel, except: [to_string: 1]
-
-defprotocol String.Chars do
-  @moduledoc %S"""
-  The String.Chars protocol is responsible for
-  converting a structure to a Binary (only if applicable).
-  The only function required to be implemented is
-  `to_string` which does the conversion.
-
-  The `to_string` function automatically imported
-  by Kernel invokes this protocol. String
-  interpolation also invokes to_string in its
-  arguments. For example, `"foo#{bar}"` is the same
-  as `"foo" <> to_string(bar)`.
-  """
-
-  def to_string(thing)
-end
-
-defimpl String.Chars, for: Atom do
-  @doc """
-  Convert the atom literally to a binary, except
-  `nil` which is converted to an empty string.
-  """
-  def to_string(nil) do
-    ""
-  end
-
-  def to_string(atom) do
-    atom_to_binary(atom)
-  end
-end
-
-defimpl String.Chars, for: BitString do
-  @doc """
-  Returns the given binary or raises an error for bitstrings.
-  """
-  def to_string(thing) when is_binary(thing) do
-    thing
-  end
-
-  def to_string(thing) do
-    raise Protocol.UndefinedError,
-             protocol: @protocol,
-                value: thing,
-          description: "cannot convert a bitstring to a string"
-  end
-end
-
-defimpl String.Chars, for: List do
-  @doc """
-  Consider the list is an iolist and converts it
-  to a binary. This allows a list of binaries, or
-  a charlist, or a mix of both, to be converted
-  successfully.
-
-  ## Examples
-
-      iex> to_string('foo')
-      "foo"
-      iex> to_string(["foo", 'bar'])
-      "foobar"
-
-  """
-  def to_string(char_list), do: String.from_char_list!(char_list)
-end
-
-defimpl String.Chars, for: Integer do
-  @doc """
-  Simply converts the integer to a string.
-  """
-  def to_string(thing) do
-    integer_to_binary(thing)
-  end
-end
-
-defimpl String.Chars, for: Float do
-  @doc """
-  Simply converts the float to a string.
-  """
-  def to_string(thing) do
-    iolist_to_binary(:io_lib_format.fwrite_g(thing))
-  end
-end
@@ -1,184 +0,0 @@
-defmodule Supervisor.Behaviour do
-  @moduledoc """
-  This module is a convenience to define Supervisor
-  callbacks in Elixir. By using this module, you get
-  the module behaviour automatically tagged as
-  `:supervisor` and some helper functions are imported
-  to make defining supervisors easier.
-
-  For more information on supervisors, please check the
-  remaining functions defined in this module or refer to
-  the following:
-
-  http://www.erlang.org/doc/man/supervisor.html
-  http://www.erlang.org/doc/design_principles/sup_princ.html
-  http://learnyousomeerlang.com/supervisors
-
-  ## Example
-
-      defmodule ExUnit.Sup do
-        use Supervisor.Behaviour
-
-        def init(user_options) do
-          tree = [ worker(ExUnit.Runner, [user_options]) ]
-          supervise(tree, strategy: :one_for_one)
-        end
-      end
-
-      { :ok, pid } = :supervisor.start_link(MyServer, [])
-
-  """
-
-  @doc false
-  defmacro __using__(_) do
-    quote location: :keep do
-      @behaviour :supervisor
-      import unquote(__MODULE__)
-    end
-  end
-
-  @doc """
-  Receives a list of children (worker or supervisors) to
-  supervise and a set of options. Returns a tuple containing
-  the supervisor specification.
-
-  ## Examples
-
-      supervise children, strategy: :one_for_one
-
-  ## Options
-
-  * `:strategy` - the restart strategy option It can be either
-    `:one_for_one`, `:rest_for_one`, `:one_for_all` and
-    `:simple_one_for_one`;
-
-  * `:max_restarts` - the maximum amount of restarts allowed in
-    a time frame. Defaults to 5;
-
-  * `:max_seconds` - the time frame in which max_restarts applies.
-    Defaults to 5;
-
-  The `:strategy` option is required and by default maximum 5 restarts
-  are allowed in 5 seconds.
-
-  ## Strategies
-
-  * `:one_for_one` - If a child process terminates, only that
-    process is restarted;
-
-  * `:one_for_all` - If a child process terminates, all other child
-    processes are terminated and then all child processes, including
-    the terminated one, are restarted;
-
-  * `:rest_for_one` - If a child process terminates, the "rest" of
-    the child processes, i.e. the child processes after the terminated
-    process in start order, are terminated. Then the terminated child
-    process and the rest of the child processes are restarted;
-
-  * `:simple_one_for_one` - Similar to `:one_for_one` but suits better
-    when dynamically attaching children;
-
-  """
-  def supervise(children, options) do
-    unless strategy = options[:strategy] do
-      raise ArgumentError, message: "expected :strategy option to be given to supervise"
-    end
-
-    maxR = Keyword.get(options, :max_restarts, 5)
-    maxS = Keyword.get(options, :max_seconds, 5)
-
-    { :ok, { { strategy, maxR, maxS }, children } }
-  end
-
-  @child_doc """
-  ## Options
-
-  * `:id` - a name used to identify the child specification
-    internally by the supervisor. Defaults to the module name;
-
-  * `:function` - the function to invoke on the child to start it.
-    Defaults to `:start_link`;
-
-  * `:restart` - defines when the child process should restart.
-    Defaults to `:permanent`;
-
-  * `:shutdown` - defines how a child process should be terminated.
-    Defaults to `5000` for a worker and `:infinity` for a supervisor;
-
-  * `:modules` - it should be a list with one element `[module]`,
-    where module is the name of the callback module only if the
-    child process is a supervisor, `gen_server` or `gen_fsm`. If the
-    child process is a gen_event, modules should be `:dynamic`.
-    Defaults to a list with the given module;
-
-  ## Restart values
-
-  The following restart values are supported:
-
-  * `:permanent` - the child process is always restarted;
-
-  * `:temporary` - the child process is never restarted (not even
-    when the supervisor's strategy is `:rest_for_one` or `:one_for_all`);
-
-  * `:transient` - the child process is restarted only if it
-    terminates abnormally, i.e. with another exit reason than
-    `:normal`, `:shutdown` or `{ :shutdown, term }`;
-
-  ## Shutdown values
-
-  The following shutdown values are supported:
-
-  * `:brutal_kill` - the child process is unconditionally terminated
-    using `exit(child, :kill)`;
-
-  * `:infinity` - if the child process is a supervisor, it is a mechanism
-    to give the subtree enough time to shutdown. It can also be used with
-    workers with care;
-
-  * Finally, it can also be any integer meaning that the supervisor tells
-    the child process to terminate by calling `exit(child, :shutdown)` and
-    then waits for an exit signal back. If no exit signal is received within
-    the specified time (in miliseconds), the child process is unconditionally
-    terminated using `exit(child, :kill)`;
-  """
-
-  @doc """
-  Defines the given `module` as a worker which will be started
-  with the given arguments.
-
-      worker ExUnit.Runner, [], restart: :permanent
-
-  By default, the function `:start_link` is invoked on the given module.
-
-  #{@child_doc}
-  """
-  def worker(module, args, options // []) do
-    child(:worker, module, args, options)
-  end
-
-  @doc """
-  Defines the given `module` as a supervisor which will be started
-  with the given arguments.
-
-      supervisor ExUnit.Runner, [], restart: :permanent
-
-  By default, the function `:start_link` is invoked on the given module.
-
-  #{@child_doc}
-  """
-  def supervisor(module, args, options // []) do
-    options = Keyword.update(options, :shutdown, :infinity, fn(x) -> x end)
-    child(:supervisor, module, args, options)
-  end
-
-  defp child(type, module, args, options) do
-    id       = Keyword.get(options, :id, module)
-    modules  = Keyword.get(options, :modules, [module])
-    function = Keyword.get(options, :function, :start_link)
-    restart  = Keyword.get(options, :restart, :permanent)
-    shutdown = Keyword.get(options, :shutdown, 5000)
-
-    { id, { module, function, args },
-      restart, shutdown, type, modules }
-  end
-end
@@ -1,388 +0,0 @@
-defmodule System do
-  defexception NoHomeError,
-    message: "could not find the user home, please set the HOME environment variable"
-
-  defexception NoTmpDirError,
-    message: "could not get a writable temporary directory, please set the TMPDIR environment variable"
-
-  defexception NoAccessCwdError,
-    message: "could not get a current working directory, the current location is not accessible"
-
-  @moduledoc """
-  The System module provides access to variables used or
-  maintained by the VM and to functions that interact directly
-  with the VM or the host system.
-  """
-
-  defp strip_re(iodata, pattern) do
-    :re.replace(iodata, pattern, "", [return: :binary])
-  end
-
-  defp read_stripped(path) do
-    case :file.read_file(path) do
-      { :ok, binary } ->
-        strip_re(binary, "^\s+|\s+$")
-      _ -> ""
-    end
-  end
-
-  # Read and strip the version from the `VERSION` file.
-  defmacrop get_version do
-    case read_stripped("VERSION") do
-      ""   -> raise CompileError, message: "could not read the version number from VERSION"
-      data -> data
-    end
-  end
-
-  # Tries to run `git describe --always --tags`. In the case of success returns
-  # the most recent tag. If that is not available, tries to read the commit hash
-  # from .git/HEAD. If that fails, returns an empty string.
-  defmacrop get_describe do
-    dirpath = ".git"
-    case :file.read_file_info(dirpath) do
-      { :ok, _ } ->
-        if :os.find_executable('git') do
-          data = :os.cmd('git describe --always --tags')
-          strip_re(data, "\n")
-        else
-          read_stripped(:filename.join(".git", "HEAD"))
-        end
-      _ -> ""
-    end
-  end
-
-  # Get the date at compilation time.
-  defmacrop get_date do
-    iolist_to_binary :httpd_util.rfc1123_date
-  end
-
-  @doc """
-  Elixir version information.
-
-  Returns Elixir's version as binary.
-  """
-  @spec version() :: String.t
-  def version, do: get_version
-
-  @doc """
-  Elixir build information.
-
-  Returns a keyword list with Elixir version, git tag info and compilation date.
-  """
-  @spec build_info() :: Keyword.t
-  def build_info do
-    [version: version, tag: get_describe, date: get_date]
-  end
-
-  @doc """
-  List command line arguments.
-
-  Returns the list of command line arguments passed to the program.
-  """
-  @spec argv() :: [String.t]
-  def argv do
-    :elixir_code_server.call :argv
-  end
-
-  @doc """
-  Modify command line arguments.
-
-  Changes the list of command line arguments. Use it with caution,
-  as it destroys any previous argv information.
-  """
-  @spec argv([String.t]) :: :ok
-  def argv(args) do
-    :elixir_code_server.cast({ :argv, args })
-  end
-
-  @doc """
-  Current working directory.
-
-  Returns the current working directory or `nil` if one
-  is not available.
-  """
-  def cwd do
-    case :file.get_cwd do
-      { :ok, base } -> String.from_char_list!(base)
-      _ -> nil
-    end
-  end
-
-  @doc """
-  Current working directory, exception on error.
-
-  Returns the current working directory or raises `System.NoAccessCwdError`.
-  """
-  def cwd! do
-    cwd || raise NoAccessCwdError
-  end
-
-  @doc """
-  User home directory.
-
-  Returns the user home directory (platform independent).
-  Returns `nil` if no user home is set.
-  """
-  def user_home do
-    case :os.type() do
-      { :win32, _ } -> get_windows_home
-      _             -> get_unix_home
-    end
-  end
-
-  @doc """
-  User home directory, exception on error.
-
-  Same as `user_home/0` but raises `System.NoHomeError`
-  instead of returning `nil` if no user home is set.
-  """
-  def user_home! do
-    user_home || raise NoHomeError
-  end
-
-  defp get_unix_home do
-    get_env("HOME")
-  end
-
-  defp get_windows_home do
-    :filename.absname(
-      get_env("USERPROFILE") || (
-        hd = get_env("HOMEDRIVE")
-        hp = get_env("HOMEPATH")
-        hd && hp && hd <> hp
-      )
-    )
-  end
-
-  @doc %S"""
-  Writable temporary directory.
-
-  Returns a writable temporary directory.
-  Searches for directories in the following order:
-
-  1. The directory named by the TMPDIR environment variable
-  2. The directory named by the TEMP environment variable
-  3. The directory named by the TMP environment variable
-  4. `C:\TMP` on Windows or `/tmp` on Unix
-  5.  As a last resort, the current working directory
-
-  Returns `nil` if none of the above are writable.
-  """
-  def tmp_dir do
-    write_env_tmp_dir('TMPDIR') ||
-      write_env_tmp_dir('TEMP') ||
-      write_env_tmp_dir('TMP')  ||
-      write_tmp_dir('/tmp')     ||
-      ((cwd = cwd()) && write_tmp_dir(cwd))
-  end
-
-  @doc """
-  Writable temporary directory, exception on error.
-
-  Same as `tmp_dir/0` but raises `System.NoTmpDirError`
-  instead of returning `nil` if no temp dir is set.
-  """
-  def tmp_dir! do
-    tmp_dir || raise NoTmpDirError
-  end
-
-  defp write_env_tmp_dir(env) do
-    case :os.getenv(env) do
-      false -> nil
-      tmp   -> write_tmp_dir(tmp)
-    end
-  end
-
-  defp write_tmp_dir(dir) do
-    case :file.read_file_info(dir) do
-      {:ok, info} ->
-        type_index = File.Stat.__record__(:index, :type)
-        access_index = File.Stat.__record__(:index, :access)
-        case { elem(info, type_index), elem(info, access_index) } do
-          { :directory, access } when access in [:read_write, :write] ->
-            String.from_char_list!(dir)
-          _ ->
-            nil
-        end
-      { :error, _ } -> nil
-    end
-  end
-
-  @doc """
-  Register a program exit handler function.
-
-  Registers a function that will be invoked
-  at the end of program execution. Useful for
-  invoking a hook in "script" mode.
-
-  The function must receive the exit status code
-  as an argument.
-  """
-  def at_exit(fun) when is_function(fun, 1) do
-    :elixir_code_server.cast { :at_exit, fun }
-  end
-
-  @doc """
-  Execute a system command.
-
-  Executes `command` in a command shell of the target OS,
-  captures the standard output of the command and returns
-  the result as a binary.
-
-  If `command` is a char list, a char list is returned.
-  Returns a binary otherwise.
-  """
-  @spec cmd(binary) :: binary
-  @spec cmd(char_list) :: char_list
-
-  def cmd(command) when is_list(command) do
-    :os.cmd(command)
-  end
-
-  def cmd(command) when is_binary(command) do
-    # Notice we don't use unicode for conversion
-    # because the OS is expecting and returning raw bytes
-    :binary.list_to_bin :os.cmd(:binary.bin_to_list(command))
-  end
-
-  @doc """
-  Locate an executable on the system.
-
-  This function looks up an executable program given
-  its name using the environment variable PATH on Unix
-  and Windows. It also considers the proper executable
-  extension for each OS, so for Windows it will try to
-  lookup files with `.com`, `.cmd` or similar extensions.
-
-  If `program` is a char list, a char list is returned.
-  Returns a binary otherwise.
-  """
-  @spec find_executable(binary) :: binary | nil
-  @spec find_executable(char_list) :: char_list | nil
-
-  def find_executable(program) when is_list(program) do
-    :os.find_executable(program) || nil
-  end
-
-  def find_executable(program) when is_binary(program) do
-    # Notice we don't use unicode for conversion
-    # because the OS is expecting and returning raw bytes
-    case :os.find_executable(:binary.bin_to_list(program)) do
-      false -> nil
-      other -> :binary.list_to_bin(other)
-    end
-  end
-
-  @doc """
-  System environment variables.
-
-  Returns a list of all environment variables. Each variable is given as a
-  `{name, value}` tuple where both `name` and `value` are strings.
-  """
-  @spec get_env() :: [{String.t, String.t}]
-  def get_env do
-    Enum.map(:os.getenv, fn var ->
-        var = String.from_char_list! var
-        [k, v] = String.split var, "=", global: false
-        {k, v}
-    end)
-  end
-
-  @doc """
-  Environment variable value.
-
-  Returns the value of the environment variable
-  `varname` as a binary, or `nil` if the environment
-  variable is undefined.
-  """
-  @spec get_env(binary) :: binary | nil
-  def get_env(varname) when is_binary(varname) do
-    case :os.getenv(String.to_char_list!(varname)) do
-      false -> nil
-      other -> String.from_char_list!(other)
-    end
-  end
-
-  @doc """
-  Erlang VM process identifier.
-
-  Returns the process identifier of the current Erlang emulator
-  in the format most commonly used by the operating system environment.
-
-  See http://www.erlang.org/doc/man/os.html#getpid-0 for more info.
-  """
-  @spec get_pid() :: binary
-  def get_pid, do: iolist_to_binary(:os.getpid)
-
-  @doc """
-  Set an environment variable value.
-
-  Sets a new `value` for the environment variable `varname`.
-  """
-  @spec put_env(binary, binary) :: :ok
-  def put_env(varname, value) when is_binary(varname) and is_binary(value) do
-   :os.putenv :binary.bin_to_list(varname), String.to_char_list!(value)
-   :ok
-  end
-
-  @doc """
-  Set multiple environment variables.
-
-  Sets a new value for each environment variable corresponding
-  to each key in `dict`.
-  """
-  @spec put_env(Dict.t) :: :ok
-  def put_env(dict) do
-    Enum.each dict, fn {key, val} -> put_env key, val end
-  end
-
-  @doc """
-  Last exception stacktrace.
-
-  Note that the Erlang VM (and therefore this function) does not
-  return the current stacktrace but rather the stacktrace of the
-  latest exception.
-  """
-  def stacktrace do
-    :erlang.get_stacktrace
-  end
-
-  @doc """
-  Halt the Erlang runtime system.
-
-  Halts the Erlang runtime system where the argument `status` must be a
-  non-negative integer, the atom `:abort` or a binary.
-
-  * If an integer, the runtime system exits with the integer value which
-    is returned to the operating system;
-
-  * If `:abort`, the runtime system aborts producing a core dump, if that is
-    enabled in the operating system;
-
-  * If a char list, an erlang crash dump is produced with status as slogan,
-    and then the runtime system exits with status code 1;
-
-  Note that on many platforms, only the status codes 0-255 are supported
-  by the operating system.
-
-  For more information, check: http://www.erlang.org/doc/man/erlang.html#halt-1
-
-  ## Examples
-
-      System.halt(0)
-      System.halt(1)
-      System.halt(:abort)
-
-  """
-  @spec halt() :: no_return
-  @spec halt(non_neg_integer | binary | :abort) :: no_return
-  def halt(status // 0)
-
-  def halt(status) when is_integer(status) or status == :abort do
-    :erlang.halt(status)
-  end
-
-  def halt(status) when is_binary(status) do
-    :erlang.halt(:binary.bin_to_list(status))
-  end
-end
@@ -1,60 +0,0 @@
-defmodule Tuple do
-  @moduledoc """
-  Functions for working with tuples.
-  """
-
-  @doc """
-  Create a new tuple.
-
-  Creates a tuple of size `size` containing the
-  given `data` at every position.
-
-  ## Examples
-
-      iex> Tuple.duplicate(:hello, 3)
-      { :hello, :hello, :hello }
-
-  """
-  @spec duplicate(term, non_neg_integer) :: tuple
-  def duplicate(data, size) do
-    :erlang.make_tuple(size, data)
-  end
-
-  @doc """
-  Insert an element into a tuple.
-
-  Inserts `value` into `tuple` at the given zero-based `index`.
-  Raises an `ArgumentError` if `index` is greater than the
-  length of `tuple`.
-
-  ## Examples
-
-      iex> tuple = { :bar, :baz }
-      ...> Tuple.insert_at(tuple, 0, :foo)
-      { :foo, :bar, :baz }
-
-  """
-  @spec insert_at(tuple, non_neg_integer, term) :: tuple
-  def insert_at(tuple, index, term) do
-    :erlang.insert_element(index + 1, tuple, term)
-  end
-
-  @doc """
-  Remove an element from a tuple.
-
-  Deletes the element at the zero-based `index` from `tuple`.
-  Raises an `ArgumentError` if `index` is greater than 
-  or equal to the length of `tuple`.
-
-  ## Examples
-
-      iex> tuple = { :foo, :bar, :baz }
-      ...> Tuple.delete_at(tuple, 0)
-      { :bar, :baz }
-
-  """
-  @spec delete_at(tuple, non_neg_integer) :: tuple
-  def delete_at(tuple, index) do
-    :erlang.delete_element(index + 1, tuple)
-  end
-end
@@ -1,233 +0,0 @@
-defmodule URI do
-  @moduledoc """
-  Utilities for working with and creating URIs.
-  """
-
-  defrecord Info, [scheme: nil, path: nil, query: nil,
-                   fragment: nil, authority: nil,
-                   userinfo: nil, host: nil, port: nil]
-
-  import Bitwise
-
-  @ports [
-    { "ftp", 21 },
-    { "http", 80 },
-    { "https", 443 },
-    { "ldap", 389 },
-    { "sftp", 22 },
-    { "tftp", 69 },
-  ]
-
-  Enum.each @ports, fn { scheme, port } ->
-    def normalize_scheme(unquote(scheme)), do: unquote(scheme)
-    def default_port(unquote(scheme)),     do: unquote(port)
-  end
-
-  @doc """
-  Normalizes the scheme according to the spec by downcasing it.
-  """
-  def normalize_scheme(nil),     do: nil
-  def normalize_scheme(scheme),  do: String.downcase(scheme)
-
-  @doc """
-  Returns the default port for a given scheme.
-  If the scheme is unknown to URI, returns `nil`.
-  Any scheme may be registered via `default_port/2`.
-  """
-  def default_port(scheme) when is_binary(scheme) do
-    { :ok, dict } = :application.get_env(:elixir, :uri)
-    Dict.get(dict, scheme)
-  end
-
-  @doc """
-  Registers a scheme with a default port.
-  """
-  def default_port(scheme, port) when is_binary(scheme) and port > 0 do
-    { :ok, dict } = :application.get_env(:elixir, :uri)
-    :application.set_env(:elixir, :uri, Dict.put(dict, scheme, port))
-  end
-
-  @doc """
-  Takes an enumerable (containing a sequence of two-item tuples)
-  and returns a string of the form "k=v&k2=v2..." where keys and values are
-  URL encoded as per `encode/1`. Keys and values can be any term
-  that implements the `String.Chars` protocol (i.e. can be converted
-  to a binary).
-  """
-  def encode_query(l), do: Enum.map_join(l, "&", &pair/1)
-
-  @doc """
-  Given a query string of the form "key1=value1&key=value2...", produces an
-  orddict with one entry for each key-value pair. Each key and value will be a
-  binary. It also does percent-unescaping of both keys and values.
-
-  Use `query_decoder/1` if you want to iterate over each value manually.
-  """
-  def decode_query(q, dict // HashDict.new) when is_binary(q) do
-    Enum.reduce query_decoder(q), dict, fn({ k, v }, acc) -> Dict.put(acc, k, v) end
-  end
-
-  @doc """
-  Returns an iterator function over the query string that decodes
-  the query string in steps.
-  """
-  def query_decoder(q) when is_binary(q) do
-    fn(acc, fun) ->
-      do_decoder(q, acc, fun)
-    end
-  end
-
-  defp do_decoder("", acc, _fun) do
-    acc
-  end
-
-  defp do_decoder(q, acc, fun) do
-    next =
-      case :binary.split(q, "&") do
-        [first, rest] -> rest
-        [first]       -> ""
-      end
-
-    current =
-      case :binary.split(first, "=") do
-        [ key, value ] -> { decode(key), decode(value) }
-        [ key ]        -> { decode(key), nil }
-      end
-
-    do_decoder(next, fun.(current, acc), fun)
-  end
-
-  defp pair({k, v}) do
-    encode(to_string(k)) <> "=" <> encode(to_string(v))
-  end
-
-  @doc """
-  Percent (URL) encodes a URI.
-  """
-  def encode(s), do: bc(<<c>> inbits s, do: <<percent(c) :: binary>>)
-
-  defp percent(32), do: <<?+>>
-  defp percent(?-), do: <<?->>
-  defp percent(?_), do: <<?_>>
-  defp percent(?.), do: <<?.>>
-
-  defp percent(c)
-      when c >= ?0 and c <= ?9
-      when c >= ?a and c <= ?z
-      when c >= ?A and c <= ?Z do
-    <<c>>
-  end
-
-  defp percent(c), do: "%" <> hex(bsr(c, 4)) <> hex(band(c, 15))
-
-  defp hex(n) when n <= 9, do: <<n + ?0>>
-  defp hex(n), do: <<n + ?A - 10>>
-
-  @doc """
-  Unpercent (URL) decodes a URI.
-  """
-  def decode(<<?%, hex1, hex2, tail :: binary >>) do
-    << bsl(hex2dec(hex1), 4) + hex2dec(hex2) >> <> decode(tail)
-  end
-
-  def decode(<<head, tail :: binary >>) do
-    <<check_plus(head)>> <> decode(tail)
-  end
-
-  def decode(<<>>), do: <<>>
-
-  defp hex2dec(n) when n in ?A..?F, do: n - ?A + 10
-  defp hex2dec(n) when n in ?0..?9, do: n - ?0
-
-  defp check_plus(?+), do: 32
-  defp check_plus(c),  do: c
-
-  @doc """
-  Parses a URI into components.
-
-  URIs have portions that are handled specially for the
-  particular scheme of the URI. For example, http and https
-  have different default ports. Sometimes the parsing
-  of portions themselves are different. This parser
-  is extensible via behavior modules. If you have a
-  module named `URI.MYSCHEME` with a function called
-  `parse` that takes a single argument, the generically
-  parsed URI, that function will be called when this
-  parse function is passed a URI of that scheme. This
-  allows you to build on top of what the URI library
-  currently offers. You also need to define `default_port`
-  which takes no arguments and returns the default port
-  for that particular scheme. Take a look at `URI.HTTPS` for an
-  example of one of these extension modules.
-  """
-  def parse(s) when is_binary(s) do
-    # From http://tools.ietf.org/html/rfc3986#appendix-B
-    regex = %r/^(([^:\/?#]+):)?(\/\/([^\/?#]*))?([^?#]*)(\?([^#]*))?(#(.*))?/
-    parts = nillify(Regex.run(regex, s))
-
-    destructure [_, _, scheme, _, authority, path, _, query, _, fragment], parts
-    { userinfo, host, port } = split_authority(authority)
-
-    if authority do
-      authority = ""
-
-      if userinfo, do: authority = authority <> userinfo <> "@"
-      if host, do: authority = authority <> host
-      if port, do: authority = authority <> ":" <> integer_to_binary(port)
-    end
-
-    scheme = normalize_scheme(scheme)
-
-    if nil?(port) and not nil?(scheme) do
-      port = default_port(scheme)
-    end
-
-    URI.Info[
-      scheme: scheme, path: path, query: query,
-      fragment: fragment, authority: authority,
-      userinfo: userinfo, host: host, port: port
-    ]
-  end
-
-  # Split an authority into its userinfo, host and port parts.
-  defp split_authority(s) do
-    s = s || ""
-    components = Regex.run %r/(^(.*)@)?(\[[a-zA-Z0-9:.]*\]|[^:]*)(:(\d*))?/, s
-
-    destructure [_, _, userinfo, host, _, port], nillify(components)
-    port = if port, do: binary_to_integer(port)
-    host = if host, do: host |> String.lstrip(?[) |> String.rstrip(?])
-
-    { userinfo, host, port }
-  end
-
-  # Regex.run returns empty strings sometimes. We want
-  # to replace those with nil for consistency.
-  defp nillify(l) do
-    lc s inlist l do
-      if size(s) > 0, do: s, else: nil
-    end
-  end
-end
-
-defimpl String.Chars, for: URI.Info do
-  def to_string(URI.Info[] = uri) do
-    scheme = uri.scheme
-
-    if scheme && (port = URI.default_port(scheme)) do
-      if uri.port == port, do: uri = uri.port(nil)
-    end
-
-    result = ""
-
-    if uri.scheme,   do: result = result <> uri.scheme <> "://"
-    if uri.userinfo, do: result = result <> uri.userinfo <> "@"
-    if uri.host,     do: result = result <> uri.host
-    if uri.port,     do: result = result <> ":" <> integer_to_binary(uri.port)
-    if uri.path,     do: result = result <> uri.path
-    if uri.query,    do: result = result <> "?" <> uri.query
-    if uri.fragment, do: result = result <> "#" <> uri.fragment
-
-    result
-  end
-end
@@ -1,528 +0,0 @@
-defmodule Version do
-  @moduledoc %S"""
-  Functions for parsing and matching versions against requirements.
-
-  A version is a string in a specific format or a `Version.Schema` 
-  generated after parsing via `Version.parse/1`.
-
-  `Version` parsing and requirements follow
-  [SemVer 2.0 schema](http://semver.org/) and you will get
-  the most of Mix's version system by following it. In order
-  to support integration with projects that may
-  follow different versioning schemas, Elixir won't choke
-  on unknown versions, however you won't be able to use
-  Mix requirements with such unformatted versions.
-
-  ## Versions
-
-  In a nutshell, a version is given by three numbers:
-
-      MAJOR.MINOR.PATCH
-
-  Pre-releases are supported by appending `-[0-9A-Za-z-\.]`:
-
-      "1.0.0-alpha.3"
-
-  Build information can be added by appending `+[0-9A-Za-z-\.]`:
-
-      "1.0.0-alpha.3+20130417140000"
-
-  ## Requirements
-
-  Requirements allow you to specify which versions of a given
-  dependency you are willing to work against. It supports common
-  operators like `>=`, `<=`, `>`, `==` and friends that 
-  work as one would expect:
-
-      # Only version 2.0.0
-      "== 2.0.0"
-
-      # Anything later than 2.0.0
-      "> 2.0.0"
-
-  Requirements also support `and` and `or` for complex conditions:
-
-      # 2.0.0 and later until 2.1.0
-      ">= 2.0.0 and < 2.1.0"
-
-  Since the example above is such a common requirement, it can
-  be expressed as:
-
-      "~> 2.0.0"
-
-  """
-
-  @type t :: String.t | Version.Schema.t
-  @type requirement :: String.t | Version.Requirement.t
-
-  @type matchable :: { major :: String.t | non_neg_integer,
-                       minor :: non_neg_integer | nil,
-                       patch :: non_neg_integer | nil,
-                       pre   :: [String.t] }
-
-  import Kernel, except: [match?: 2]
-
-  defrecord Schema, major: 0, minor: 0, patch: 0, pre: nil, build: nil, source: nil
-  defrecord Requirement, source: nil, matchspec: nil
-
-  defexception InvalidRequirement, reason: :invalid_requirement do
-    def message(InvalidRequirement[reason: reason]) when is_binary(reason) do
-      { first, rest } = String.next_grapheme(reason)
-      String.downcase(first) <> rest
-    end
-
-    def message(InvalidRequirement[]) do
-      "invalid version specification"
-    end
-  end
-
-  @doc """
-  Check if the given version matches the specification.
-
-  Returns `true` if `version` satisfies `requirement`, `false` otherwise.
-  Raises a `Version.InvalidRequirement` exception if `requirement` is not parseable. 
-
-  ## Examples
-
-      iex> Version.match?("2.0", ">1.0")
-      true
-      iex> Version.match?("2.0", "==1.0")
-      false
-
-  """
-  @spec match?(t, requirement) :: boolean
-  def match?(version, requirement) when is_binary(requirement) do
-    case Version.Parser.parse_requirement(requirement) do
-      { :ok, req } ->
-        match?(version, req)
-      { :error, reason } ->
-        raise InvalidRequirement, reason: reason
-    end
-  end
-
-  def match?(version, requirement) when is_binary(version) do
-    match?(parse(version), requirement)
-  end
-
-  def match?(Schema[] = version, Requirement[matchspec: spec]) do
-    case :ets.test_ms(to_matchable(version), spec) do
-      { :ok, result } ->
-        result != false
-      { :error, reason } ->
-        raise InvalidRequirement, reason: reason
-    end
-  end
-
-  @doc """
-  Check if a version string is compatible with [semver](http://semver.org/).
-
-  ## Examples
-
-      iex> Version.valid?("2.0")
-      true
-      iex> Version.valid?("invalid")
-      false
-
-  """
-  @spec valid?(String.t | Schema.t) :: boolean
-  def valid?(string) when is_binary(string) do
-    Version.Parser.valid_version?(string)
-  end
-
-  def valid?(Version.Schema[major: nil]), do: false
-  def valid?(Version.Schema[]),           do: true
-
-  @doc """
-  Parse a version string into a `Version.Schema`.
-
-  ## Examples
-
-      > Version.parse("2.0.1-alpha1")
-      #Version.Schema<2.0.1-alpha1>
-
-  """
-  @spec parse(String.t) :: Schema.t
-  def parse(string) when is_binary(string) do
-    case Version.Parser.parse_version(string) do
-      { :ok, matchable } -> from_matchable(matchable).source(string).build(get_build(string))
-      { :error, _ } -> Version.Schema[source: string]
-    end
-  end
-
-  defp get_build(string) do
-    case Regex.run(%r/\+([^\s]+)$/, string) do
-      nil ->
-        nil
-
-      [_, build] ->
-        build
-    end
-  end
-
-  @doc """
-  Convert a version to a `Version.matchable`
-
-  ## Examples
-
-      iex> Version.to_matchable("2.0.1-alpha.1")
-      {2, 0, 1, ["alpha", 1]}
-
-  """
-  @spec to_matchable(String.t | Schema.t) :: Version.matchable
-  def to_matchable(Schema[major: nil, source: source]) do
-    { source, nil, nil, [] }
-  end
-
-  def to_matchable(Version.Schema[major: major, minor: minor, patch: patch, pre: nil]) do
-    { major, minor, patch, [] }
-  end
-
-  def to_matchable(Version.Schema[major: major, minor: minor, patch: patch, pre: pre]) do
-    { major, minor, patch, Version.Parser.parse_pre(pre) }
-  end
-
-  def to_matchable(string) do
-    to_matchable(parse(string))
-  end
-
-  @doc """
-  Convert a matchable to a `Version.Schema`.
-
-  ## Examples
-
-      > Version.from_matchable({2, 0, 1, ["alpha", 1]})
-      #Version.Schema<2.0.1-alpha.1>
-
-  """
-  @spec from_matchable(Version.matchable) :: Schema.t
-  def from_matchable({ source, nil, nil, nil }) when is_binary(source) do
-    Version.Schema[source: source]
-  end
-
-  def from_matchable({ major, minor, patch, pre }) do
-    source = "#{major}"
-
-    if minor do
-      source = "#{source}.#{minor}"
-
-      if patch do
-        source = "#{source}.#{patch}"
-
-        case pre do
-          [] ->
-            pre = nil
-
-          list ->
-            pre    = Enum.join(list, ".")
-            source = "#{source}-#{pre}"
-        end
-      end
-    end
-
-    Version.Schema[major: major, minor: minor, patch: patch, pre: pre, source: source]
-  end
-
-  defmodule Parser.DSL do
-    @moduledoc false
-
-    defmacro deflexer(match, do: body) when is_binary(match) do
-      quote do
-        def lexer(unquote(match) <> rest, acc) do
-          lexer(rest, [unquote(body) | acc])
-        end
-      end
-    end
-
-    defmacro deflexer(acc, do: body) do
-      quote do
-        def lexer("", unquote(acc)) do
-          unquote(body)
-        end
-      end
-    end
-
-    defmacro deflexer(char, acc, do: body) do
-      quote do
-        def lexer(<< unquote(char) :: utf8, rest :: binary >>, unquote(acc)) do
-          unquote(char) = << unquote(char) :: utf8 >>
-
-          lexer(rest, unquote(body))
-        end
-      end
-    end
-  end
-
-  defmodule Parser do
-    @moduledoc false
-    import Parser.DSL
-
-    deflexer ">=",    do: :'>='
-    deflexer "<=",    do: :'<='
-    deflexer "~>",    do: :'~>'
-    deflexer ">",     do: :'>'
-    deflexer "<",     do: :'<'
-    deflexer "==",    do: :'=='
-    deflexer "!=",    do: :'!='
-    deflexer "!",     do: :'!='
-    deflexer " or ",  do: :'||'
-    deflexer " and ", do: :'&&'
-    deflexer " ",     do: :' '
-
-    deflexer x, [] do
-      [x, :'==']
-    end
-
-    deflexer x, [h | acc] do
-      cond do
-        is_binary h ->
-          [h <> x | acc]
-
-        h in [:'||', :'&&'] ->
-          [x, :'==', h | acc]
-
-        true ->
-          [x, h | acc]
-      end
-    end
-
-    deflexer acc do
-      Enum.filter(Enum.reverse(acc), &(&1 != :' '))
-    end
-
-    @version_regex %r/^(\d+)(?:\.(\d+)(?:\.(\d+))?)?(?:\-([^\s]+))?(?:\+[^\d]+)?$/
-
-    @spec parse_requirement(String.t) :: { :ok, Version.Requirement.t } | { :error, binary | atom }
-    def parse_requirement(source) do
-      lexed = lexer(source, [])
-
-      if valid_requirement?(lexed) do
-        spec = to_matchspec(lexed)
-
-        case :ets.test_ms({}, spec) do
-          { :ok, _ } ->
-            { :ok, Requirement[source: source, matchspec: spec] }
-
-          { :error, errors } ->
-            { :error, Enum.map(errors, fn { :error, reason } ->
-              to_string(reason)
-            end) }
-        end
-      else
-        { :error, :invalid_requirement }
-      end
-    end
-
-    defp nillify(""), do: nil
-    defp nillify(o),  do: o
-
-    @spec parse_version(String.t) :: { :ok, Version.matchable } | { :error, :invalid_version }
-    def parse_version(string) when is_binary(string) do
-      if valid_version?(string) do
-        destructure [_, major, minor, patch, pre], Regex.run(@version_regex, string)
-
-        major = binary_to_integer(major)
-        minor = binary_to_integer(minor |> nillify || "0")
-        patch = binary_to_integer(patch |> nillify || "0")
-        pre   = pre && parse_pre(pre) || []
-
-        { :ok, { major, minor, patch, pre } }
-      else
-        { :error, :invalid_version }
-      end
-    end
-
-    @doc false
-    def parse_pre(pre) do
-      String.split(pre, ".") |> Enum.map fn piece ->
-        if piece =~ %r/^(0|[1-9][0-9]*)$/ do
-          binary_to_integer(piece)
-        else
-          piece
-        end
-      end
-    end
-
-    @spec valid_requirement?(list) :: boolean
-    def valid_requirement?([]) do
-      false
-    end
-
-    def valid_requirement?([a | next]) do
-      valid_requirement?(a, next)
-    end
-
-    # it must finish with a version
-    defp valid_requirement?(a, []) when is_binary(a) do
-      true
-    end
-
-    # version version
-    defp valid_requirement?(a, [b | _]) when is_binary(a) and is_binary(b) do
-      false
-    end
-
-    # or <op> | and <op>
-    defp valid_requirement?(a, [b | next]) when is_atom(a) and is_atom(b) and a in [:'||', :'&&'] do
-      valid_requirement?(b, next)
-    end
-
-    # <version> or | <version> and
-    defp valid_requirement?(a, [b | next]) when is_binary(a) and is_atom(b) and b in [:'||', :'&&'] do
-      valid_requirement?(b, next)
-    end
-
-    # or <version> | and <version>
-    defp valid_requirement?(a, [b | next]) when is_atom(a) and is_binary(b) and a in [:'||', :'&&'] do
-      valid_requirement?(b, next)
-    end
-
-    # <op> <version>; also checks operators work on valid versions
-    defp valid_requirement?(a, [b | next]) when is_atom(a) and is_binary(b) do
-      if valid_version?(b) do
-        valid_requirement?(b, next)
-      else
-        if a in [:'==', :'!='] and Regex.match? %r/^\w/, b do
-          valid_requirement?(b, next)
-        else
-          false
-        end
-      end
-    end
-
-    defp valid_requirement?(_, _) do
-      false
-    end
-
-    @spec valid_version?(String.t) :: boolean
-    def valid_version?(string) do
-      Regex.match? %r/^\d+(\.\d+(\.\d+)?)?(\-[^\s]+)?(?:\+[^\s]+)?$/, string
-    end
-
-    defp approximate(version) do
-      Version.from_matchable(case Regex.run(@version_regex, version) do
-        [_, major] ->
-          { binary_to_integer(major) + 1, 0, 0, [] }
-
-        [_, major, _] ->
-          { binary_to_integer(major) + 1, 0, 0, [] }
-
-        [_, major, minor, _] ->
-          { binary_to_integer(major), binary_to_integer(minor) + 1, 0, [] }
-
-        [_, major, minor, _, _] ->
-          { binary_to_integer(major), binary_to_integer(minor) + 1, 0, [] }
-      end)
-    end
-
-    defp to_matchspec(lexed) do
-      first = to_condition(lexed)
-      rest  = Enum.drop(lexed, 2)
-
-      [{{ :'$1', :'$2', :'$3', :'$4' }, [to_condition(first, rest)], [:'$_'] }]
-    end
-
-    defp to_condition([:'==', version | _]) do
-      version = Version.to_matchable(version)
-
-      { :'==', :'$_', { :const, version } }
-    end
-
-    defp to_condition([:'!=', version | _]) do
-      version = Version.to_matchable(version)
-
-      { :'/=', :'$_', { :const, version } }
-    end
-
-    defp to_condition([:'~>', version | _]) do
-      from = Version.parse(version)
-      to   = approximate(version)
-
-      { :andalso, to_condition([:'>=', to_string(from)]),
-                  to_condition([:'<', to_string(to)]) }
-    end
-
-    defp to_condition([:'>', version | _]) do
-      { major, minor, patch, pre } = Version.to_matchable(version)
-
-      { :andalso, { :not, { :is_binary, :'$1' } },
-                  { :orelse, { :'>', {{ :'$1', :'$2', :'$3' }},
-                                     { :const, { major, minor, patch } } },
-                             { :andalso, { :'==', {{ :'$1', :'$2', :'$3' }},
-                                                  { :const, { major, minor, patch } } },
-                             { :orelse, { :andalso, { :'==', { :length, :'$4' }, 0 },
-                                                    { :'/=', length(pre), 0 } },
-                                        { :andalso, { :'/=', length(pre), 0 },
-                                                    { :orelse, { :'>', { :length, :'$4' }, length(pre) },
-                                                               { :andalso, { :'==', { :length, :'$4' }, length(pre) },
-                                                                           { :'>', :'$4', { :const, pre } } } } } } } } }
-    end
-
-    defp to_condition([:'>=', version | _]) do
-      matchable = Version.to_matchable(version)
-
-      { :orelse, { :andalso, { :not, { :is_binary, :'$1' } },
-                             { :'==', :'$_', { :const, matchable } } },
-                 to_condition([:'>', version]) }
-    end
-
-    defp to_condition([:'<', version | _]) do
-      { major, minor, patch, pre } = Version.to_matchable(version)
-
-      { :andalso, { :not, { :is_binary, :'$1' } },
-                  { :orelse, { :'<', {{ :'$1', :'$2', :'$3' }},
-                                     { :const, { major, minor, patch } } },
-                             { :andalso, { :'==', {{ :'$1', :'$2', :'$3' }},
-                                                  { :const, { major, minor, patch } } },
-                             { :orelse, { :andalso, { :'/=', { :length, :'$4' }, 0 },
-                                                    { :'==', length(pre), 0 } },
-                                        { :andalso, { :'/=', { :length, :'$4' }, 0 },
-                                                    { :orelse, { :'<', { :length, :'$4' }, length(pre) },
-                                                               { :andalso, { :'==', { :length, :'$4' }, length(pre) },
-                                                                           { :'<', :'$4', { :const, pre } } } } } } } } }
-    end
-
-    defp to_condition([:'<=', version | _]) do
-      matchable = Version.to_matchable(version)
-
-      { :orelse, { :andalso, { :not, { :is_binary, :'$1' } },
-                             { :'==', :'$_', { :const, matchable } } },
-                 to_condition([:'<', version]) }
-    end
-
-    defp to_condition(current, []) do
-      current
-    end
-
-    defp to_condition(current, [:'&&', operator, version | rest]) do
-      to_condition({ :andalso, current, to_condition([operator, version]) }, rest)
-    end
-
-    defp to_condition(current, [:'||', operator, version | rest]) do
-      to_condition({ :orelse, current, to_condition([operator, version]) }, rest)
-    end
-  end
-end
-
-defimpl String.Chars, for: Version.Schema do
-  def to_string(Version.Schema[source: source]) do
-    source
-  end
-end
-
-defimpl Inspect, for: Version.Schema do
-  def inspect(self, _opts) do
-    "#Version.Schema<" <> to_string(self) <> ">"
-  end
-end
-
-defimpl String.Chars, for: Version.Requirement do
-  def to_string({ _, source, _ }) do
-    source
-  end
-end
-
-defimpl Inspect, for: Version.Requirement do
-  def inspect({ _, source, _ }, _opts) do
-    "#Version.Requirement<" <> source <> ">"
-  end
-end
@@ -1,11 +0,0 @@
-defmodule Elixir.Mixfile do
-  use Mix.Project
-
-  def project do
-    [ app: :elixir,
-      version: System.version,
-      escript_embed_elixir: false,
-      escript_main_module: :elixir,
-      escript_emu_args: "%%! -noshell\n" ]
-  end
-end
@@ -1,103 +0,0 @@
-00DF; 00DF; 0053 0073; 0053 0053; # LATIN SMALL LETTER SHARP S
-0130; 0069 0307; 0130; 0130; # LATIN CAPITAL LETTER I WITH DOT ABOVE
-FB00; FB00; 0046 0066; 0046 0046; # LATIN SMALL LIGATURE FF
-FB01; FB01; 0046 0069; 0046 0049; # LATIN SMALL LIGATURE FI
-FB02; FB02; 0046 006C; 0046 004C; # LATIN SMALL LIGATURE FL
-FB03; FB03; 0046 0066 0069; 0046 0046 0049; # LATIN SMALL LIGATURE FFI
-FB04; FB04; 0046 0066 006C; 0046 0046 004C; # LATIN SMALL LIGATURE FFL
-FB05; FB05; 0053 0074; 0053 0054; # LATIN SMALL LIGATURE LONG S T
-FB06; FB06; 0053 0074; 0053 0054; # LATIN SMALL LIGATURE ST
-0587; 0587; 0535 0582; 0535 0552; # ARMENIAN SMALL LIGATURE ECH YIWN
-FB13; FB13; 0544 0576; 0544 0546; # ARMENIAN SMALL LIGATURE MEN NOW
-FB14; FB14; 0544 0565; 0544 0535; # ARMENIAN SMALL LIGATURE MEN ECH
-FB15; FB15; 0544 056B; 0544 053B; # ARMENIAN SMALL LIGATURE MEN INI
-FB16; FB16; 054E 0576; 054E 0546; # ARMENIAN SMALL LIGATURE VEW NOW
-FB17; FB17; 0544 056D; 0544 053D; # ARMENIAN SMALL LIGATURE MEN XEH
-0149; 0149; 02BC 004E; 02BC 004E; # LATIN SMALL LETTER N PRECEDED BY APOSTROPHE
-0390; 0390; 0399 0308 0301; 0399 0308 0301; # GREEK SMALL LETTER IOTA WITH DIALYTIKA AND TONOS
-03B0; 03B0; 03A5 0308 0301; 03A5 0308 0301; # GREEK SMALL LETTER UPSILON WITH DIALYTIKA AND TONOS
-01F0; 01F0; 004A 030C; 004A 030C; # LATIN SMALL LETTER J WITH CARON
-1E96; 1E96; 0048 0331; 0048 0331; # LATIN SMALL LETTER H WITH LINE BELOW
-1E97; 1E97; 0054 0308; 0054 0308; # LATIN SMALL LETTER T WITH DIAERESIS
-1E98; 1E98; 0057 030A; 0057 030A; # LATIN SMALL LETTER W WITH RING ABOVE
-1E99; 1E99; 0059 030A; 0059 030A; # LATIN SMALL LETTER Y WITH RING ABOVE
-1E9A; 1E9A; 0041 02BE; 0041 02BE; # LATIN SMALL LETTER A WITH RIGHT HALF RING
-1F50; 1F50; 03A5 0313; 03A5 0313; # GREEK SMALL LETTER UPSILON WITH PSILI
-1F52; 1F52; 03A5 0313 0300; 03A5 0313 0300; # GREEK SMALL LETTER UPSILON WITH PSILI AND VARIA
-1F54; 1F54; 03A5 0313 0301; 03A5 0313 0301; # GREEK SMALL LETTER UPSILON WITH PSILI AND OXIA
-1F56; 1F56; 03A5 0313 0342; 03A5 0313 0342; # GREEK SMALL LETTER UPSILON WITH PSILI AND PERISPOMENI
-1FB6; 1FB6; 0391 0342; 0391 0342; # GREEK SMALL LETTER ALPHA WITH PERISPOMENI
-1FC6; 1FC6; 0397 0342; 0397 0342; # GREEK SMALL LETTER ETA WITH PERISPOMENI
-1FD2; 1FD2; 0399 0308 0300; 0399 0308 0300; # GREEK SMALL LETTER IOTA WITH DIALYTIKA AND VARIA
-1FD3; 1FD3; 0399 0308 0301; 0399 0308 0301; # GREEK SMALL LETTER IOTA WITH DIALYTIKA AND OXIA
-1FD6; 1FD6; 0399 0342; 0399 0342; # GREEK SMALL LETTER IOTA WITH PERISPOMENI
-1FD7; 1FD7; 0399 0308 0342; 0399 0308 0342; # GREEK SMALL LETTER IOTA WITH DIALYTIKA AND PERISPOMENI
-1FE2; 1FE2; 03A5 0308 0300; 03A5 0308 0300; # GREEK SMALL LETTER UPSILON WITH DIALYTIKA AND VARIA
-1FE3; 1FE3; 03A5 0308 0301; 03A5 0308 0301; # GREEK SMALL LETTER UPSILON WITH DIALYTIKA AND OXIA
-1FE4; 1FE4; 03A1 0313; 03A1 0313; # GREEK SMALL LETTER RHO WITH PSILI
-1FE6; 1FE6; 03A5 0342; 03A5 0342; # GREEK SMALL LETTER UPSILON WITH PERISPOMENI
-1FE7; 1FE7; 03A5 0308 0342; 03A5 0308 0342; # GREEK SMALL LETTER UPSILON WITH DIALYTIKA AND PERISPOMENI
-1FF6; 1FF6; 03A9 0342; 03A9 0342; # GREEK SMALL LETTER OMEGA WITH PERISPOMENI
-1F80; 1F80; 1F88; 1F08 0399; # GREEK SMALL LETTER ALPHA WITH PSILI AND YPOGEGRAMMENI
-1F81; 1F81; 1F89; 1F09 0399; # GREEK SMALL LETTER ALPHA WITH DASIA AND YPOGEGRAMMENI
-1F82; 1F82; 1F8A; 1F0A 0399; # GREEK SMALL LETTER ALPHA WITH PSILI AND VARIA AND YPOGEGRAMMENI
-1F83; 1F83; 1F8B; 1F0B 0399; # GREEK SMALL LETTER ALPHA WITH DASIA AND VARIA AND YPOGEGRAMMENI
-1F84; 1F84; 1F8C; 1F0C 0399; # GREEK SMALL LETTER ALPHA WITH PSILI AND OXIA AND YPOGEGRAMMENI
-1F85; 1F85; 1F8D; 1F0D 0399; # GREEK SMALL LETTER ALPHA WITH DASIA AND OXIA AND YPOGEGRAMMENI
-1F86; 1F86; 1F8E; 1F0E 0399; # GREEK SMALL LETTER ALPHA WITH PSILI AND PERISPOMENI AND YPOGEGRAMMENI
-1F87; 1F87; 1F8F; 1F0F 0399; # GREEK SMALL LETTER ALPHA WITH DASIA AND PERISPOMENI AND YPOGEGRAMMENI
-1F88; 1F80; 1F88; 1F08 0399; # GREEK CAPITAL LETTER ALPHA WITH PSILI AND PROSGEGRAMMENI
-1F89; 1F81; 1F89; 1F09 0399; # GREEK CAPITAL LETTER ALPHA WITH DASIA AND PROSGEGRAMMENI
-1F8A; 1F82; 1F8A; 1F0A 0399; # GREEK CAPITAL LETTER ALPHA WITH PSILI AND VARIA AND PROSGEGRAMMENI
-1F8B; 1F83; 1F8B; 1F0B 0399; # GREEK CAPITAL LETTER ALPHA WITH DASIA AND VARIA AND PROSGEGRAMMENI
-1F8C; 1F84; 1F8C; 1F0C 0399; # GREEK CAPITAL LETTER ALPHA WITH PSILI AND OXIA AND PROSGEGRAMMENI
-1F8D; 1F85; 1F8D; 1F0D 0399; # GREEK CAPITAL LETTER ALPHA WITH DASIA AND OXIA AND PROSGEGRAMMENI
-1F8E; 1F86; 1F8E; 1F0E 0399; # GREEK CAPITAL LETTER ALPHA WITH PSILI AND PERISPOMENI AND PROSGEGRAMMENI
-1F8F; 1F87; 1F8F; 1F0F 0399; # GREEK CAPITAL LETTER ALPHA WITH DASIA AND PERISPOMENI AND PROSGEGRAMMENI
-1F90; 1F90; 1F98; 1F28 0399; # GREEK SMALL LETTER ETA WITH PSILI AND YPOGEGRAMMENI
-1F91; 1F91; 1F99; 1F29 0399; # GREEK SMALL LETTER ETA WITH DASIA AND YPOGEGRAMMENI
-1F92; 1F92; 1F9A; 1F2A 0399; # GREEK SMALL LETTER ETA WITH PSILI AND VARIA AND YPOGEGRAMMENI
-1F93; 1F93; 1F9B; 1F2B 0399; # GREEK SMALL LETTER ETA WITH DASIA AND VARIA AND YPOGEGRAMMENI
-1F94; 1F94; 1F9C; 1F2C 0399; # GREEK SMALL LETTER ETA WITH PSILI AND OXIA AND YPOGEGRAMMENI
-1F95; 1F95; 1F9D; 1F2D 0399; # GREEK SMALL LETTER ETA WITH DASIA AND OXIA AND YPOGEGRAMMENI
-1F96; 1F96; 1F9E; 1F2E 0399; # GREEK SMALL LETTER ETA WITH PSILI AND PERISPOMENI AND YPOGEGRAMMENI
-1F97; 1F97; 1F9F; 1F2F 0399; # GREEK SMALL LETTER ETA WITH DASIA AND PERISPOMENI AND YPOGEGRAMMENI
-1F98; 1F90; 1F98; 1F28 0399; # GREEK CAPITAL LETTER ETA WITH PSILI AND PROSGEGRAMMENI
-1F99; 1F91; 1F99; 1F29 0399; # GREEK CAPITAL LETTER ETA WITH DASIA AND PROSGEGRAMMENI
-1F9A; 1F92; 1F9A; 1F2A 0399; # GREEK CAPITAL LETTER ETA WITH PSILI AND VARIA AND PROSGEGRAMMENI
-1F9B; 1F93; 1F9B; 1F2B 0399; # GREEK CAPITAL LETTER ETA WITH DASIA AND VARIA AND PROSGEGRAMMENI
-1F9C; 1F94; 1F9C; 1F2C 0399; # GREEK CAPITAL LETTER ETA WITH PSILI AND OXIA AND PROSGEGRAMMENI
-1F9D; 1F95; 1F9D; 1F2D 0399; # GREEK CAPITAL LETTER ETA WITH DASIA AND OXIA AND PROSGEGRAMMENI
-1F9E; 1F96; 1F9E; 1F2E 0399; # GREEK CAPITAL LETTER ETA WITH PSILI AND PERISPOMENI AND PROSGEGRAMMENI
-1F9F; 1F97; 1F9F; 1F2F 0399; # GREEK CAPITAL LETTER ETA WITH DASIA AND PERISPOMENI AND PROSGEGRAMMENI
-1FA0; 1FA0; 1FA8; 1F68 0399; # GREEK SMALL LETTER OMEGA WITH PSILI AND YPOGEGRAMMENI
-1FA1; 1FA1; 1FA9; 1F69 0399; # GREEK SMALL LETTER OMEGA WITH DASIA AND YPOGEGRAMMENI
-1FA2; 1FA2; 1FAA; 1F6A 0399; # GREEK SMALL LETTER OMEGA WITH PSILI AND VARIA AND YPOGEGRAMMENI
-1FA3; 1FA3; 1FAB; 1F6B 0399; # GREEK SMALL LETTER OMEGA WITH DASIA AND VARIA AND YPOGEGRAMMENI
-1FA4; 1FA4; 1FAC; 1F6C 0399; # GREEK SMALL LETTER OMEGA WITH PSILI AND OXIA AND YPOGEGRAMMENI
-1FA5; 1FA5; 1FAD; 1F6D 0399; # GREEK SMALL LETTER OMEGA WITH DASIA AND OXIA AND YPOGEGRAMMENI
-1FA6; 1FA6; 1FAE; 1F6E 0399; # GREEK SMALL LETTER OMEGA WITH PSILI AND PERISPOMENI AND YPOGEGRAMMENI
-1FA7; 1FA7; 1FAF; 1F6F 0399; # GREEK SMALL LETTER OMEGA WITH DASIA AND PERISPOMENI AND YPOGEGRAMMENI
-1FA8; 1FA0; 1FA8; 1F68 0399; # GREEK CAPITAL LETTER OMEGA WITH PSILI AND PROSGEGRAMMENI
-1FA9; 1FA1; 1FA9; 1F69 0399; # GREEK CAPITAL LETTER OMEGA WITH DASIA AND PROSGEGRAMMENI
-1FAA; 1FA2; 1FAA; 1F6A 0399; # GREEK CAPITAL LETTER OMEGA WITH PSILI AND VARIA AND PROSGEGRAMMENI
-1FAB; 1FA3; 1FAB; 1F6B 0399; # GREEK CAPITAL LETTER OMEGA WITH DASIA AND VARIA AND PROSGEGRAMMENI
-1FAC; 1FA4; 1FAC; 1F6C 0399; # GREEK CAPITAL LETTER OMEGA WITH PSILI AND OXIA AND PROSGEGRAMMENI
-1FAD; 1FA5; 1FAD; 1F6D 0399; # GREEK CAPITAL LETTER OMEGA WITH DASIA AND OXIA AND PROSGEGRAMMENI
-1FAE; 1FA6; 1FAE; 1F6E 0399; # GREEK CAPITAL LETTER OMEGA WITH PSILI AND PERISPOMENI AND PROSGEGRAMMENI
-1FAF; 1FA7; 1FAF; 1F6F 0399; # GREEK CAPITAL LETTER OMEGA WITH DASIA AND PERISPOMENI AND PROSGEGRAMMENI
-1FB3; 1FB3; 1FBC; 0391 0399; # GREEK SMALL LETTER ALPHA WITH YPOGEGRAMMENI
-1FBC; 1FB3; 1FBC; 0391 0399; # GREEK CAPITAL LETTER ALPHA WITH PROSGEGRAMMENI
-1FC3; 1FC3; 1FCC; 0397 0399; # GREEK SMALL LETTER ETA WITH YPOGEGRAMMENI
-1FCC; 1FC3; 1FCC; 0397 0399; # GREEK CAPITAL LETTER ETA WITH PROSGEGRAMMENI
-1FF3; 1FF3; 1FFC; 03A9 0399; # GREEK SMALL LETTER OMEGA WITH YPOGEGRAMMENI
-1FFC; 1FF3; 1FFC; 03A9 0399; # GREEK CAPITAL LETTER OMEGA WITH PROSGEGRAMMENI
-1FB2; 1FB2; 1FBA 0345; 1FBA 0399; # GREEK SMALL LETTER ALPHA WITH VARIA AND YPOGEGRAMMENI
-1FB4; 1FB4; 0386 0345; 0386 0399; # GREEK SMALL LETTER ALPHA WITH OXIA AND YPOGEGRAMMENI
-1FC2; 1FC2; 1FCA 0345; 1FCA 0399; # GREEK SMALL LETTER ETA WITH VARIA AND YPOGEGRAMMENI
-1FC4; 1FC4; 0389 0345; 0389 0399; # GREEK SMALL LETTER ETA WITH OXIA AND YPOGEGRAMMENI
-1FF2; 1FF2; 1FFA 0345; 1FFA 0399; # GREEK SMALL LETTER OMEGA WITH VARIA AND YPOGEGRAMMENI
-1FF4; 1FF4; 038F 0345; 038F 0399; # GREEK SMALL LETTER OMEGA WITH OXIA AND YPOGEGRAMMENI
-1FB7; 1FB7; 0391 0342 0345; 0391 0342 0399; # GREEK SMALL LETTER ALPHA WITH PERISPOMENI AND YPOGEGRAMMENI
-1FC7; 1FC7; 0397 0342 0345; 0397 0342 0399; # GREEK SMALL LETTER ETA WITH PERISPOMENI AND YPOGEGRAMMENI
-1FF7; 1FF7; 03A9 0342 0345; 03A9 0342 0399; # GREEK SMALL LETTER OMEGA WITH PERISPOMENI AND YPOGEGRAMMENI
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
hermes-agent	c345dab9c1	Merge pull request 'adapter: support OpenAI streaming (SSE) in /v1/chat/completions' (#1 ) from feat/adapter-streaming into main	2026-09-12 12:51:34 +02:00
hermes-agent	2d46eda619	adapter: support OpenAI streaming (SSE) in /v1/chat/completions OpenCode's @ai-sdk/openai-compatible sends stream:true and would render an empty response because the adapter always returned a single non-streaming chat.completion JSON body. Now when stream:true, emit OpenAI-compatible SSE chat.completion.chunk events (role, content, [DONE]) so streaming clients render text. Non-streaming path unchanged. Adds a local EchoServer test that exercises the streaming route end-to-end.	2026-09-12 10:50:47 +00:00
hermes-agent	beb9b1b3ef	Add temporary chat-request logging (diagnose SwiftChat error)	2026-09-11 18:44:45 +00:00
hermes-agent	cb512a7f17	admin: serve the page without requiring the Bearer header A browser opening /admin can't send an Authorization header, so the admin page was unreachable (401 blank). Serve the HTML form openly — it exposes no data — and let the in-page ADMIN_API_KEY field drive the auth'd /admin/agents CRUD calls.	2026-09-11 15:36:15 +00:00
hermes-agent	e4fdeb6b79	nixos-module: set RELEASE_COOKIE so the release starts The Elixir release's start script reads releases/COOKIE which isn't baked in, so the service crashed on boot (cat: releases/COOKIE: No such file). Set RELEASE_COOKIE in the systemd Environment to fix startup.	2026-09-10 14:35:28 +00:00
hermes-agent	6b22171018	flake: fill mixFodDeps hash	2026-09-10 07:21:42 +00:00
hermes-agent	f0112289e6	flake: add mixFodDeps (fetchMixDeps) for Hex deps	2026-09-10 07:17:56 +00:00
hermes-agent	72e18a0a0c	Remove AGENTS env seeding; agents managed only via admin API The store now starts empty and agents are added/removed exclusively through the web admin page / admin API, persisted to AGENTS_FILE. No AGENTS env var needed in the sops secret.	2026-09-10 07:06:26 +00:00
hermes-agent	e2be3f652e	Add web admin page to manage agents GET /admin serves a self-contained HTML page (ADMIN_API_KEY protected) that lists agents and lets you add/update/remove them via the admin API — no redeploy needed to add an agent.	2026-09-10 06:59:08 +00:00
hermes-agent	490bd32322	Add admin API to manage agents at runtime - AgentRegistry is now file-backed (AGENTS_FILE, default /var/lib/n8n-openai/agents.json): agents persist across restarts and can be added/removed without a redeploy. - New admin endpoints (separate ADMIN_API_KEY): GET /admin/agents POST /admin/agents {model, webhook} DELETE /admin/agents/:model - AGENTS env only seeds the store on first boot; the file is authoritative. - NixOS module sets AGENTS_FILE under the writable StateDirectory.	2026-09-10 06:46:05 +00:00
hermes-agent	1bfcba133a	Add NixOS module for the adapter service Export nixosModules.default so the service (systemd unit, service user, sops secret) is defined in the flake, not re-declared in each host config. Consume with imports = [ inputs.n8n-openai-adapter.nixosModules.default ] + services.n8n-openai-adapter = { enable = true; domain = ...; port = ...; }.	2026-09-10 06:36:44 +00:00
hermes-agent	c3d024c16c	Apply mix format	2026-09-09 21:46:16 +00:00
hermes-agent	1530a761d5	Add flake.lock pinning nixpkgs	2026-09-09 21:44:55 +00:00
hermes-agent	5197b6ece6	OpenAI-compatible adapter for n8n chat agents (Elixir) Exposes self-hosted n8n chat agents behind /v1/chat/completions and /v1/models. Model -> n8n webhook routing via a GenServer registry, so multiple agents map to multiple models. Plug + Bandit, req for the n8n webhook call, Bearer auth (ADAPTER_API_KEY). Ships a flake.nix (beamPackages.mixRelease) so it can be consumed as a NixOS flake input.	2026-09-09 21:43:21 +00:00
				`@@ -1 +0,0 @@`
				`lib/elixir/test/elixir/fixtures/*.txt text eol=lf`
				`@@ -1 +0,0 @@`
				`ExUnit.start [trace: "--trace" in System.argv]`