Compare commits

...
Author SHA1 Message Date
José Valim 3564c68cdc Release v1.18.0 2024-12-19 18:30:20 +01:00
José Valim 9df42f5a0a Update CHANGELOG 2024-12-19 17:26:59 +01:00
José Valim c6597bcb20 Collapse structs for better pretty printing in different scenarios 2024-12-19 15:13:18 +01:00
José Valim ca6edfd389 Include types in comparison warning 2024-12-19 15:13:14 +01:00
José Valim ed83c407a5 Simplify and optimize NaiveDateTime.utc_now 2024-12-19 15:13:12 +01:00
José Valim a426520b66 Improvements to descr pretty printing 2024-12-18 09:32:08 +01:00
Guillaume Duboc 1c0f585f40 Fuse maps and tuples for printing (#14079) 2024-12-18 09:32:06 +01:00
José Valim 7c86c4bdb3 Update CHANGELOG 2024-12-17 10:11:15 +01:00
José Valim 4931ab40a5 Do not warn for CaseTemplate options, closes #14077 2024-12-17 10:03:53 +01:00
José Valim c7f78e8015 Emit warnings for duplicate patterns instead of errors 2024-12-16 19:30:54 +01:00
José Valim 6a24bafbda Tag exceptions as dynamic, closes #14074 2024-12-16 16:13:04 +01:00
José Valim 8ddef06e81 Also encode Duration to JSON 2024-12-16 12:23:47 +01:00
José Valim 472c4e1a24 Avoid crashes on diagnostics with tabs, closes #14073 2024-12-16 12:09:18 +01:00
José Valim 305d8e603c Do not convert fun() into (... -> any()) and vice-versa in typespecs 2024-12-15 15:18:52 +01:00
Jonatan Kłosko 1c7fc86e9c Make :source_path in docs chunk a charlist (#14071) 2024-12-13 16:57:54 +01:00
José Valim 617dc72759 Fix Range.new deprecation assertion 2024-12-13 10:14:47 +01:00
José Valim 4329a3c2f2 Address bootstrap 2024-12-13 10:01:18 +01:00
José Valim 8e8c9a7334 Wrap Range.new in Function.identity to avoid tail call optimization 2024-12-13 09:55:39 +01:00
Michał Łępicki 36c06c3a56 Fix JSON.decode/3 spec (#14068)
The error clause was missing the :error tuple wrapping
2024-12-13 09:52:42 +01:00
Mathias Polligkeit 048cc2ddb1 Further improve warning message for range with negative step (#14070) 2024-12-13 09:52:42 +01:00
Wojtek Mach 048ae58d6d Do not set docs source_ref to -latest tags (#14065) 2024-12-12 11:51:41 +01:00
José Valim 85e00cddd5 Clarify warnings happen on compiled code 2024-12-12 09:52:30 +01:00
Jean Klingler 6c866e8ab2 Fix type warnings in pop_in/1 (#14064) 2024-12-12 17:47:23 +09:00
José Valim 55eca5dff5 Implement JSON.Encoder for Calendar types (#14061) 2024-12-12 00:38:44 +01:00
Panagiotis Nezis 070a6d51bf Improve warning in Range.new/2 with negative step (#14062) 2024-12-12 00:38:40 +01:00
José Valim 9c28fb7a39 Improve JSON and Float docs 2024-12-11 21:08:20 +01:00
Andrea Leopardi cda1a09c58 Improve error message on empty "mix do" (#14055) 2024-12-11 12:44:47 +01:00
José Valim 3ca0bddd9d Fix encoding specs 2024-12-11 12:16:33 +01:00
José Valim e07a91594b Update deprecations 2024-12-11 12:02:01 +01:00
José Valim dfc659104c Touch the compile path when writing .app file for the first time (#14053)
Closes #14049.
2024-12-11 10:01:00 +01:00
Shenghang Tsai a4fa71ab23 Update CHANGELOG.md (#14052) 2024-12-11 09:43:42 +01:00
EksperimentalandWojtek Mach 2273bafb47 Improve grammar in CHANGELOG.md (#14050)
Co-authored-by: Wojtek Mach <wojtekmach@users.noreply.github.com>
2024-12-11 08:41:11 +01:00
José Valim a78f1715dc Fix error message around single quote keywords 2024-12-10 19:30:39 +01:00
José Valim b709865976 Improvements to markdown rendering in CHANGELOG 2024-12-10 19:18:32 +01:00
José Valim da7e04a540 Improve docs around auto_reloading 2024-12-10 19:16:09 +01:00
José Valim 66c5908619 Release v1.18.0-rc.0 2024-12-10 18:32:26 +01:00
José Valim 31d8da6c77 Avoid mtime races on compile.app_test.exs 2024-12-10 15:41:16 +01:00
Daniel Gomez de Souza 3bdf5a912c Fix typo (#14046) 2024-12-10 15:34:32 +01:00
José Valim 0e9b5ec070 Branch out v1.18 2024-12-10 15:01:43 +01:00
44 changed files with 640 additions and 236 deletions
+28 -16
View File
@@ -4,13 +4,13 @@ Elixir v1.18 is an impressive release with improvements across the two main effo
## Type system improvements ## Type system improvements
The most exciting change in Elixir v1.18 is type checking of function calls, alongside gradual inference of patterns and return types. To understand how this will impact your programs, consider the following code: The most exciting change in Elixir v1.18 is type checking of function calls, alongside gradual inference of patterns and return types. To understand how this will impact your programs, consider the following code in "lib/user.ex":
```elixir ```elixir
defmodule User do defmodule User do
defstruct [:age, :car_choice] defstruct [:age, :car_choice]
def drive(%User{age: age, car_choice: car}, cars_choices) when age >= 18 do def drive(%User{age: age, car_choice: car}, car_choices) when age >= 18 do
if car in car_choices do if car in car_choices do
{:ok, car} {:ok, car}
else else
@@ -24,9 +24,9 @@ defmodule User do
end end
``` ```
Elixir's type system will infer the `drive` function expects a `%User{}` struct as input and returns either `{:ok, dynamic()}`, `{:error, :no_choice}`, or `{:error, :not_allowed}`. Elixir's type system will infer that the `drive/2` function expects a `%User{}` struct and returns either `{:ok, dynamic()}`, `{:error, :no_choice}`, or `{:error, :not_allowed}`.
Therefore, the following code should emit a violation, due to an invalid argument: Therefore, the following code in a separate module (either in a separate or the same file), should emit a violation, due to an invalid argument:
```elixir ```elixir
User.drive({:ok, %User{}}, car_choices) User.drive({:ok, %User{}}, car_choices)
@@ -156,14 +156,11 @@ More migrations may be added in future releases.
This release includes official support for JSON encoding and decoding. This release includes official support for JSON encoding and decoding.
Both encoder and decoder fully conform to [RFC 8259](https://tools.ietf.org/html/rfc8259) and Both encoder and decoder fully conform to [RFC 8259](https://tools.ietf.org/html/rfc8259) and [ECMA 404](https://ecma-international.org/publications-and-standards/standards/ecma-404/) standards.
[ECMA 404](https://ecma-international.org/publications-and-standards/standards/ecma-404/)
standards.
### Encoding ### Encoding
Encoding can be done via `JSON.encode!/1` and `JSON.encode_to_iodata!/1` functions. Encoding can be done via `JSON.encode!/1` and `JSON.encode_to_iodata!/1` functions. The default encoding rules are applied as follows:
The default encoding rules are applied as follows:
| **Elixir** | **JSON** | | **Elixir** | **JSON** |
|------------------------|----------| |------------------------|----------|
@@ -177,19 +174,18 @@ The default encoding rules are applied as follows:
| `%{atom() => _}` | Object | | `%{atom() => _}` | Object |
| `%{integer() => _}` | Object | | `%{integer() => _}` | Object |
You may also implement the `JSON.Encoder` protocol for custom data structures. You may also implement the `JSON.Encoder` protocol for custom data structures. Elixir already implements the protocol for all Calendar types.
If you have a struct, you can derive the implementation of the `JSON.Encoder`
by specifying which fields should be encoded to JSON: If you have a struct, you can derive the implementation of the `JSON.Encoder` by specifying which fields should be encoded to JSON:
```elixir ```elixir
@derive {JSON.Encoder, only: [....]} @derive {JSON.Encoder, only: [...]}
defstruct ... defstruct ...
``` ```
### Decoding ### Decoding
Decoding can be done via `JSON.decode/2` and `JSON.decode!/2` functions. Decoding can be done via `JSON.decode/2` and `JSON.decode!/2` functions. The default decoding rules are applied as follows:
The default decoding rules are applied as follows:
| **JSON** | **Elixir** | | **JSON** | **Elixir** |
|----------|------------------------| |----------|------------------------|
@@ -199,6 +195,16 @@ The default decoding rules are applied as follows:
| String | `binary()` | | String | `binary()` |
| Object | `%{binary() => _}` | | Object | `%{binary() => _}` |
## Language server listeners
4 months ago, we welcomed [the Official Language Server team](https://elixir-lang.org/blog/2024/08/15/welcome-elixir-language-server-team/), with the goal of unifying the efforts behind code intelligence, tools, and editors in Elixir. Elixir v1.18 brings new features on this front by introducing locks and listeners to its compilation. Let's understand what it means.
At the moment, all language server implementations have their own compilation environment. This means that your project and dependencies during development are compiled once, for your own use, and then again for the language server. This duplicate effort could cause the language server experience to lag, when it could be relying on the already compiled artifacts of your project.
This release address by introducing a compiler lock, ensuring that only a single operating system process running Elixir compiles your project at a given moment, and by providing the ability for one operating system process to listen to the compilation results of others. In other words, different Elixir instances can now communicate over the same compilation build, instead of racing each other.
These enhancements do not only improve editor tooling, but they also directly benefit projects like IEx and Phoenix. For example, you can invoke `IEx.configure(auto_reload: true)` and IEx will automatically reload modules changed elsewhere, either by a separate terminal or your IDE.
## Potential incompatibilities ## Potential incompatibilities
This release no longer supports WERL (a graphical user interface on Windows used by Erlang 25 and earlier). For a better user experience on Windows terminals, use Erlang/OTP 26+ (this is also the last Elixir release to support Erlang/OTP 25). This release no longer supports WERL (a graphical user interface on Windows used by Erlang 25 and earlier). For a better user experience on Windows terminals, use Erlang/OTP 26+ (this is also the last Elixir release to support Erlang/OTP 25).
@@ -219,7 +225,7 @@ You may also prefer to write using guards:
def foo(x, y, z) when x == y and y == z def foo(x, y, z) when x == y and y == z
## v1.18.0-dev ## v1.18.0 (2024-12-19)
### 1. Enhancements ### 1. Enhancements
@@ -234,6 +240,8 @@ You may also prefer to write using guards:
* [Config] Add `Config.read_config/1` * [Config] Add `Config.read_config/1`
* [Enumerable] Add `Enum.product_by/2` and `Enum.sum_by/2` * [Enumerable] Add `Enum.product_by/2` and `Enum.sum_by/2`
* [Exception] Add `MissingApplicationsError` exception to denote missing applications * [Exception] Add `MissingApplicationsError` exception to denote missing applications
* [JSON] Add a new `JSON` module with encoding and decoding functionality
* [JSON] Implement `JSON.Encoder` for all Calendar types
* [Kernel] Update source code parsing to match [UTS #55](https://www.unicode.org/reports/tr55/) latest recommendations. In particular, mixed script is allowed in identifiers as long as they are separate by underscores (`_`), such as `http_сервер`. Previously allowed highly restrictive identifiers, which mixed Latin and other scripts, such as the japanese word for t-shirt, `Tシャツ`, now require the underscore as well * [Kernel] Update source code parsing to match [UTS #55](https://www.unicode.org/reports/tr55/) latest recommendations. In particular, mixed script is allowed in identifiers as long as they are separate by underscores (`_`), such as `http_сервер`. Previously allowed highly restrictive identifiers, which mixed Latin and other scripts, such as the japanese word for t-shirt, `Tシャツ`, now require the underscore as well
* [Kernel] Warn on bidirectional confusability in identifiers * [Kernel] Warn on bidirectional confusability in identifiers
* [Kernel] Verify the type of the binary generators * [Kernel] Verify the type of the binary generators
@@ -261,6 +269,7 @@ You may also prefer to write using guards:
#### IEx #### IEx
* [IEx] Add `IEx.configure(auto_reload: true)` to automatically pick up modules recompiled from other operating system processes
* [IEx] Add `:dot_iex` support to `IEx.configure/1` * [IEx] Add `:dot_iex` support to `IEx.configure/1`
* [IEx] Add report for normal/shutdown exits in IEx * [IEx] Add report for normal/shutdown exits in IEx
@@ -284,6 +293,7 @@ You may also prefer to write using guards:
* [Code.Fragment] Properly handle keyword keys as their own entry * [Code.Fragment] Properly handle keyword keys as their own entry
* [Inspect.Algebra] Ensure `next_break_fits` respects `line_length` * [Inspect.Algebra] Ensure `next_break_fits` respects `line_length`
* [Kernel] Validate AST on `unquote` and `unquote_splicing` to provide better error reports instead of failing too late inside the compiler * [Kernel] Validate AST on `unquote` and `unquote_splicing` to provide better error reports instead of failing too late inside the compiler
* [Kernel] Avoid crashes when emitting diagnostics on code using \t for indentation
* [Module] Include module attribute line and name when tracing its aliases * [Module] Include module attribute line and name when tracing its aliases
* [Stream] Do not halt streams twice in `Stream.transform/5` * [Stream] Do not halt streams twice in `Stream.transform/5`
* [URI] Fix a bug when a schemaless URI is given to `URI.merge/2` * [URI] Fix a bug when a schemaless URI is given to `URI.merge/2`
@@ -334,6 +344,8 @@ You may also prefer to write using guards:
#### Mix #### Mix
* [mix cmd] Deprecate `mix cmd --app APP` in favor of `mix do --app APP` * [mix cmd] Deprecate `mix cmd --app APP` in favor of `mix do --app APP`
* [mix compile] `:warnings_as_errors` configuration in `:elixirc_options` is deprecated. Instead pass the `--warnings-as-errors` flag to `mix compile`. Alternatively, you might alias the task: `aliases: [compile: "compile --warnings-as-errors"]`
* [mix test] `:warnings_as_errors` configuration in `:test_elixirc_options` is deprecated. Instead pass the `--warnings-as-errors` flag to `mix test`. Alternatively, you might alias the task: `aliases: [test: "test --warnings-as-errors"]`
* [Mix.Tasks.Compile] Deprecate `compilers/0` in favor of `Mix.Task.Compiler.compilers/0` * [Mix.Tasks.Compile] Deprecate `compilers/0` in favor of `Mix.Task.Compiler.compilers/0`
## v1.17 ## v1.17
+2 -2
View File
@@ -2,7 +2,7 @@ PREFIX ?= /usr/local
TEST_FILES ?= "*_test.exs" TEST_FILES ?= "*_test.exs"
SHARE_PREFIX ?= $(PREFIX)/share SHARE_PREFIX ?= $(PREFIX)/share
MAN_PREFIX ?= $(SHARE_PREFIX)/man MAN_PREFIX ?= $(SHARE_PREFIX)/man
CANONICAL := main/ CANONICAL := v1.18/
ELIXIRC := bin/elixirc --ignore-module-conflict $(ELIXIRC_OPTS) ELIXIRC := bin/elixirc --ignore-module-conflict $(ELIXIRC_OPTS)
ERLC := erlc -I lib/elixir/include ERLC := erlc -I lib/elixir/include
ERL_MAKE := erl -make ERL_MAKE := erl -make
@@ -17,7 +17,7 @@ INSTALL_DIR = $(INSTALL) -m755 -d
INSTALL_DATA = $(INSTALL) -m644 INSTALL_DATA = $(INSTALL) -m644
INSTALL_PROGRAM = $(INSTALL) -m755 INSTALL_PROGRAM = $(INSTALL) -m755
GIT_REVISION = $(strip $(shell git rev-parse HEAD 2> /dev/null )) GIT_REVISION = $(strip $(shell git rev-parse HEAD 2> /dev/null ))
GIT_TAG = $(strip $(shell head="$(call GIT_REVISION)"; git tag --points-at $$head 2> /dev/null | tail -1) ) GIT_TAG = $(strip $(shell head="$(call GIT_REVISION)"; git tag --points-at $$head 2> /dev/null | grep -v latest | tail -1))
SOURCE_DATE_EPOCH_PATH = lib/elixir/tmp/ebin_reproducible SOURCE_DATE_EPOCH_PATH = lib/elixir/tmp/ebin_reproducible
SOURCE_DATE_EPOCH_FILE = $(SOURCE_DATE_EPOCH_PATH)/SOURCE_DATE_EPOCH SOURCE_DATE_EPOCH_FILE = $(SOURCE_DATE_EPOCH_PATH)/SOURCE_DATE_EPOCH
+2 -3
View File
@@ -6,12 +6,11 @@ Elixir applies bug fixes only to the latest minor branch. Security patches are a
Elixir version | Support Elixir version | Support
:------------- | :----------------------------- :------------- | :-----------------------------
1.18 | Development 1.18 | Bug fixes and security patches
1.17 | Bug fixes and security patches 1.17 | Security patches only
1.16 | Security patches only 1.16 | Security patches only
1.15 | Security patches only 1.15 | Security patches only
1.14 | Security patches only 1.14 | Security patches only
1.13 | Security patches only
## Announcements ## Announcements
+1 -1
View File
@@ -1 +1 @@
1.18.0-dev 1.18.0
+1 -1
View File
@@ -1,7 +1,7 @@
#!/bin/sh #!/bin/sh
set -e set -e
ELIXIR_VERSION=1.18.0-dev ELIXIR_VERSION=1.18.0
if [ $# -eq 0 ] || { [ $# -eq 1 ] && { [ "$1" = "--help" ] || [ "$1" = "-h" ]; }; }; then if [ $# -eq 0 ] || { [ $# -eq 1 ] && { [ "$1" = "--help" ] || [ "$1" = "-h" ]; }; }; then
cat <<USAGE >&2 cat <<USAGE >&2
+1 -1
View File
@@ -1,6 +1,6 @@
@echo off @echo off
set ELIXIR_VERSION=1.18.0-dev set ELIXIR_VERSION=1.18.0
if ""%1""=="""" if ""%2""=="""" goto documentation if ""%1""=="""" if ""%2""=="""" goto documentation
if /I ""%1""==""--help"" if ""%2""=="""" goto documentation if /I ""%1""==""--help"" if ""%2""=="""" goto documentation
+1 -1
View File
@@ -1,6 +1,6 @@
#!/usr/bin/env pwsh #!/usr/bin/env pwsh
$ELIXIR_VERSION = "1.18.0-dev" $ELIXIR_VERSION = "1.18.0"
$scriptPath = Split-Path -Parent $PSCommandPath $scriptPath = Split-Path -Parent $PSCommandPath
$erlExec = "erl" $erlExec = "erl"
+19 -20
View File
@@ -110,30 +110,12 @@ defmodule NaiveDateTime do
@spec utc_now(Calendar.calendar() | :native | :microsecond | :millisecond | :second) :: t @spec utc_now(Calendar.calendar() | :native | :microsecond | :millisecond | :second) :: t
def utc_now(calendar_or_time_unit \\ Calendar.ISO) def utc_now(calendar_or_time_unit \\ Calendar.ISO)
def utc_now(Calendar.ISO) do
{:ok, {year, month, day}, {hour, minute, second}, microsecond} =
Calendar.ISO.from_unix(:os.system_time(), :native)
%NaiveDateTime{
year: year,
month: month,
day: day,
hour: hour,
minute: minute,
second: second,
microsecond: microsecond,
calendar: Calendar.ISO
}
end
def utc_now(time_unit) when time_unit in [:microsecond, :millisecond, :second, :native] do def utc_now(time_unit) when time_unit in [:microsecond, :millisecond, :second, :native] do
utc_now(time_unit, Calendar.ISO) utc_now(time_unit, Calendar.ISO)
end end
def utc_now(calendar) do def utc_now(calendar) do
calendar utc_now(:native, calendar)
|> DateTime.utc_now()
|> DateTime.to_naive()
end end
@doc """ @doc """
@@ -158,7 +140,20 @@ defmodule NaiveDateTime do
@spec utc_now(:native | :microsecond | :millisecond | :second, Calendar.calendar()) :: t @spec utc_now(:native | :microsecond | :millisecond | :second, Calendar.calendar()) :: t
def utc_now(time_unit, calendar) def utc_now(time_unit, calendar)
when time_unit in [:native, :microsecond, :millisecond, :second] do when time_unit in [:native, :microsecond, :millisecond, :second] do
DateTime.utc_now(time_unit, calendar) |> DateTime.to_naive() {:ok, {year, month, day}, {hour, minute, second}, microsecond} =
Calendar.ISO.from_unix(System.os_time(time_unit), time_unit)
%NaiveDateTime{
year: year,
month: month,
day: day,
hour: hour,
minute: minute,
second: second,
microsecond: microsecond,
calendar: Calendar.ISO
}
|> convert!(calendar)
end end
@doc """ @doc """
@@ -1265,6 +1260,10 @@ defmodule NaiveDateTime do
{:ok, t} | {:error, :incompatible_calendars} {:ok, t} | {:error, :incompatible_calendars}
# Keep it multiline for proper function clause errors. # Keep it multiline for proper function clause errors.
def convert(%NaiveDateTime{calendar: calendar} = ndt, calendar) do
{:ok, ndt}
end
def convert( def convert(
%{ %{
calendar: calendar, calendar: calendar,
-9
View File
@@ -27,11 +27,6 @@ defmodule Code.Typespec do
end end
end end
def spec_to_quoted(name, {:type, anno, :fun, []}) when is_atom(name) do
meta = meta(anno)
{:"::", meta, [{name, meta, []}, quote(do: term)]}
end
def spec_to_quoted(name, {:type, anno, :bounded_fun, [type, constrs]}) when is_atom(name) do def spec_to_quoted(name, {:type, anno, :bounded_fun, [type, constrs]}) when is_atom(name) do
meta = meta(anno) meta = meta(anno)
{:type, _, :fun, [{:type, _, :product, args}, result]} = type {:type, _, :fun, [{:type, _, :product, args}, result]} = type
@@ -317,10 +312,6 @@ defmodule Code.Typespec do
[{:->, meta(anno), [[typespec_to_quoted(args)], typespec_to_quoted(result)]}] [{:->, meta(anno), [[typespec_to_quoted(args)], typespec_to_quoted(result)]}]
end end
defp typespec_to_quoted({:type, anno, :fun, []}) do
typespec_to_quoted({:type, anno, :fun, [{:type, anno, :any}, {:type, anno, :any, []}]})
end
defp typespec_to_quoted({:type, anno, :range, [left, right]}) do defp typespec_to_quoted({:type, anno, :range, [left, right]}) do
{:.., meta(anno), [typespec_to_quoted(left), typespec_to_quoted(right)]} {:.., meta(anno), [typespec_to_quoted(left), typespec_to_quoted(right)]}
end end
+3 -9
View File
@@ -577,15 +577,9 @@ defmodule Float do
Returns a charlist which corresponds to the shortest text representation Returns a charlist which corresponds to the shortest text representation
of the given float. of the given float.
The underlying algorithm changes depending on the Erlang/OTP version: It uses the algorithm presented in "Ryū: fast float-to-string conversion"
in Proceedings of the SIGPLAN '2018 Conference on Programming Language
* For OTP >= 24, it uses the algorithm presented in "Ryū: fast Design and Implementation.
float-to-string conversion" in Proceedings of the SIGPLAN '2018
Conference on Programming Language Design and Implementation.
* For OTP < 24, it uses the algorithm presented in "Printing Floating-Point
Numbers Quickly and Accurately" in Proceedings of the SIGPLAN '1996
Conference on Programming Language Design and Implementation.
For a configurable representation, use `:erlang.float_to_list/2`. For a configurable representation, use `:erlang.float_to_list/2`.
+19 -4
View File
@@ -150,6 +150,12 @@ defimpl JSON.Encoder, for: Map do
end end
end end
defimpl JSON.Encoder, for: [Date, Time, NaiveDateTime, DateTime, Duration] do
def encode(value, _encoder) do
[?", @for.to_iso8601(value), ?"]
end
end
defmodule JSON.DecodeError do defmodule JSON.DecodeError do
@moduledoc """ @moduledoc """
The exception raised by `JSON.decode!/1`. The exception raised by `JSON.decode!/1`.
@@ -199,6 +205,8 @@ defmodule JSON do
@moduledoc since: "1.18.0" @moduledoc since: "1.18.0"
@type encoder :: (term(), encoder() -> iodata())
@type decode_error_reason :: @type decode_error_reason ::
{:unexpected_end, non_neg_integer()} {:unexpected_end, non_neg_integer()}
| {:invalid_byte, non_neg_integer(), byte()} | {:invalid_byte, non_neg_integer(), byte()}
@@ -257,7 +265,8 @@ defmodule JSON do
For streaming decoding, see Erlang's `:json` module. For streaming decoding, see Erlang's `:json` module.
""" """
@spec decode(binary(), term(), keyword()) :: {term(), term(), binary()} | decode_error_reason() @spec decode(binary(), term(), keyword()) ::
{term(), term(), binary()} | {:error, decode_error_reason()}
def decode(binary, acc, decoders) when is_binary(binary) and is_list(decoders) do def decode(binary, acc, decoders) when is_binary(binary) and is_list(decoders) do
decoders = Keyword.put_new(decoders, :null, nil) decoders = Keyword.put_new(decoders, :null, nil)
@@ -326,13 +335,19 @@ defmodule JSON do
The second argument is a function that is recursively The second argument is a function that is recursively
invoked to encode a term. invoked to encode a term.
> #### IO and performance {: .tip}
>
> If you need to encode data to be sent over the network
> or written to the filesystem, consider using the more
> efficient `encode_to_iodata!/2`.
## Examples ## Examples
iex> JSON.encode!([123, "string", %{key: "value"}]) iex> JSON.encode!([123, "string", %{key: "value"}])
"[123,\"string\",{\"key\":\"value\"}]" "[123,\"string\",{\"key\":\"value\"}]"
""" """
@spec encode!(a, (a -> iodata())) :: binary() when a: var @spec encode!(term(), encoder()) :: binary()
def encode!(term, encoder \\ &protocol_encode/2) do def encode!(term, encoder \\ &protocol_encode/2) do
IO.iodata_to_binary(encoder.(term, encoder)) IO.iodata_to_binary(encoder.(term, encoder))
end end
@@ -353,7 +368,7 @@ defmodule JSON do
"[123,\"string\",{\"key\":\"value\"}]" "[123,\"string\",{\"key\":\"value\"}]"
""" """
@spec encode_to_iodata!(a, (a -> iodata())) :: iodata() when a: var @spec encode_to_iodata!(term(), encoder()) :: iodata()
def encode_to_iodata!(term, encoder \\ &protocol_encode/2) do def encode_to_iodata!(term, encoder \\ &protocol_encode/2) do
encoder.(term, encoder) encoder.(term, encoder)
end end
@@ -365,7 +380,7 @@ defmodule JSON do
`encode!/2` and `encode_to_iodata!/2`. The default implementation `encode!/2` and `encode_to_iodata!/2`. The default implementation
is an optimized dispatch to the `JSON.Encoder` protocol. is an optimized dispatch to the `JSON.Encoder` protocol.
""" """
@spec protocol_encode(a, (a -> iodata())) :: iodata() when a: var @spec protocol_encode(term(), encoder()) :: iodata()
def protocol_encode(value, encoder) when is_atom(value) do def protocol_encode(value, encoder) when is_atom(value) do
case value do case value do
nil -> "null" nil -> "null"
+3 -3
View File
@@ -3306,7 +3306,7 @@ defmodule Kernel do
end end
defp nest_pop_in(:map, h, [{:access, key}]) do defp nest_pop_in(:map, h, [{:access, key}]) do
quote do quote generated: true do
case unquote(h) do case unquote(h) do
nil -> {nil, nil} nil -> {nil, nil}
h -> Access.pop(h, unquote(key)) h -> Access.pop(h, unquote(key))
@@ -3327,7 +3327,7 @@ defmodule Kernel do
end end
defp nest_pop_in(_, h, [{:access, key}]) do defp nest_pop_in(_, h, [{:access, key}]) do
quote do quote generated: true do
case unquote(h) do case unquote(h) do
nil -> :pop nil -> :pop
h -> Access.pop(h, unquote(key)) h -> Access.pop(h, unquote(key))
@@ -4066,7 +4066,7 @@ defmodule Kernel do
end end
defp stepless_range(nil, first, last, _caller) do defp stepless_range(nil, first, last, _caller) do
quote(do: Elixir.Range.new(unquote(first), unquote(last))) quote(do: Function.identity(Elixir.Range.new(unquote(first), unquote(last))))
end end
defp stepless_range(:guard, first, last, caller) do defp stepless_range(:guard, first, last, caller) do
+1 -8
View File
@@ -668,14 +668,7 @@ defmodule Kernel.Typespec do
when is_list(args) do when is_list(args) do
{args, state} = fn_args(meta, args, vars, caller, state) {args, state} = fn_args(meta, args, vars, caller, state)
{spec, state} = typespec(return, vars, caller, state) {spec, state} = typespec(return, vars, caller, state)
{{:type, location(meta), :fun, [args, spec]}, state}
fun_args =
case [args, spec] do
[{:type, _, :any}, {:type, _, :any, []}] -> []
pair -> pair
end
{{:type, location(meta), :fun, fun_args}, state}
end end
# Handle type operator # Handle type operator
+22 -5
View File
@@ -363,14 +363,14 @@ defmodule Module.Types.Apply do
match?({false, _}, map_fetch(left, :__struct__)) or match?({false, _}, map_fetch(left, :__struct__)) or
match?({false, _}, map_fetch(right, :__struct__)) -> match?({false, _}, map_fetch(right, :__struct__)) ->
warning = {:struct_comparison, expr, context} warning = {:struct_comparison, expr, name, left, right, context}
warn(__MODULE__, warning, elem(expr, 1), stack, context) warn(__MODULE__, warning, elem(expr, 1), stack, context)
number_type?(left) and number_type?(right) -> number_type?(left) and number_type?(right) ->
context context
disjoint?(left, right) -> disjoint?(left, right) ->
warning = {:mismatched_comparison, expr, context} warning = {:mismatched_comparison, expr, name, left, right, context}
warn(__MODULE__, warning, elem(expr, 1), stack, context) warn(__MODULE__, warning, elem(expr, 1), stack, context)
true -> true ->
@@ -395,7 +395,7 @@ defmodule Module.Types.Apply do
context context
disjoint?(left, right) -> disjoint?(left, right) ->
warning = {:mismatched_comparison, expr, context} warning = {:mismatched_comparison, expr, name, left, right, context}
warn(__MODULE__, warning, elem(expr, 1), stack, context) warn(__MODULE__, warning, elem(expr, 1), stack, context)
true -> true ->
@@ -810,7 +810,7 @@ defmodule Module.Types.Apply do
} }
end end
def format_diagnostic({:mismatched_comparison, expr, context}) do def format_diagnostic({:mismatched_comparison, expr, name, left, right, context}) do
traces = collect_traces(expr, context) traces = collect_traces(expr, context)
%{ %{
@@ -821,6 +821,10 @@ defmodule Module.Types.Apply do
comparison between distinct types found: comparison between distinct types found:
#{expr_to_string(expr) |> indent(4)} #{expr_to_string(expr) |> indent(4)}
given types:
#{type_comparison_to_string(name, left, right) |> indent(4)}
""", """,
format_traces(traces), format_traces(traces),
""" """
@@ -833,7 +837,7 @@ defmodule Module.Types.Apply do
} }
end end
def format_diagnostic({:struct_comparison, expr, context}) do def format_diagnostic({:struct_comparison, expr, name, left, right, context}) do
traces = collect_traces(expr, context) traces = collect_traces(expr, context)
%{ %{
@@ -844,6 +848,10 @@ defmodule Module.Types.Apply do
comparison with structs found: comparison with structs found:
#{expr_to_string(expr) |> indent(4)} #{expr_to_string(expr) |> indent(4)}
given types:
#{type_comparison_to_string(name, left, right) |> indent(4)}
""", """,
format_traces(traces), format_traces(traces),
""" """
@@ -933,6 +941,15 @@ defmodule Module.Types.Apply do
alias Inspect.Algebra, as: IA alias Inspect.Algebra, as: IA
defp type_comparison_to_string(fun, left, right) do
{Kernel, fun, [left, right], _} = :elixir_rewrite.erl_to_ex(:erlang, fun, [left, right])
{fun, [], [to_quoted(left, collapse_structs: true), to_quoted(right, collapse_structs: true)]}
|> Code.Formatter.to_algebra()
|> Inspect.Algebra.format(98)
|> IO.iodata_to_binary()
end
defp clauses_args_to_quoted_string([{args, _return}], converter) do defp clauses_args_to_quoted_string([{args, _return}], converter) do
"\n " <> (clause_args_to_quoted_string(args, converter) |> indent(4)) "\n " <> (clause_args_to_quoted_string(args, converter) |> indent(4))
end end
+190 -69
View File
@@ -367,18 +367,30 @@ defmodule Module.Types.Descr do
@doc """ @doc """
Converts a descr to its quoted representation. Converts a descr to its quoted representation.
## Options
* `:collapse_structs` - do not show struct fields that match
their default type
""" """
def to_quoted(descr) do def to_quoted(descr, opts \\ []) do
if term_type?(descr) do if term_type?(descr) do
{:term, [], []} {:term, [], []}
else else
# Dynamic always come first for visibility
{dynamic, descr} =
case :maps.take(:dynamic, descr) do
:error -> {[], descr}
{dynamic, descr} -> {to_quoted(:dynamic, dynamic, opts), descr}
end
# Merge empty list and list together if they both exist
{extra, descr} = {extra, descr} =
case descr do case descr do
# Merge empty list and list together if they both exist
%{list: list, bitmap: bitmap} when (bitmap &&& @bit_empty_list) != 0 -> %{list: list, bitmap: bitmap} when (bitmap &&& @bit_empty_list) != 0 ->
descr = descr |> Map.delete(:list) |> Map.update!(:bitmap, &(&1 - @bit_empty_list)) descr = descr |> Map.delete(:list) |> Map.replace!(:bitmap, bitmap - @bit_empty_list)
case list_to_quoted(list, :list) do case list_to_quoted(list, :list, opts) do
[] -> {[{:empty_list, [], []}], descr} [] -> {[{:empty_list, [], []}], descr}
unions -> {unions, descr} unions -> {unions, descr}
end end
@@ -387,26 +399,32 @@ defmodule Module.Types.Descr do
{[], descr} {[], descr}
end end
case extra ++ Enum.flat_map(descr, fn {key, value} -> to_quoted(key, value) end) do unions =
dynamic ++
Enum.sort(
extra ++ Enum.flat_map(descr, fn {key, value} -> to_quoted(key, value, opts) end)
)
case unions do
[] -> {:none, [], []} [] -> {:none, [], []}
unions -> unions |> Enum.sort() |> Enum.reduce(&{:or, [], [&2, &1]}) unions -> Enum.reduce(unions, &{:or, [], [&2, &1]})
end end
end end
end end
defp to_quoted(:atom, val), do: atom_to_quoted(val) defp to_quoted(:atom, val, _opts), do: atom_to_quoted(val)
defp to_quoted(:bitmap, val), do: bitmap_to_quoted(val) defp to_quoted(:bitmap, val, _opts), do: bitmap_to_quoted(val)
defp to_quoted(:dynamic, descr), do: dynamic_to_quoted(descr) defp to_quoted(:dynamic, descr, opts), do: dynamic_to_quoted(descr, opts)
defp to_quoted(:map, dnf), do: map_to_quoted(dnf) defp to_quoted(:map, dnf, opts), do: map_to_quoted(dnf, opts)
defp to_quoted(:list, dnf), do: list_to_quoted(dnf, :non_empty_list) defp to_quoted(:list, dnf, opts), do: list_to_quoted(dnf, :non_empty_list, opts)
defp to_quoted(:tuple, dnf), do: tuple_to_quoted(dnf) defp to_quoted(:tuple, dnf, opts), do: tuple_to_quoted(dnf, opts)
@doc """ @doc """
Converts a descr to its quoted string representation. Converts a descr to its quoted string representation.
""" """
def to_quoted_string(descr) do def to_quoted_string(descr, opts \\ []) do
descr descr
|> to_quoted() |> to_quoted(opts)
|> Code.Formatter.to_algebra() |> Code.Formatter.to_algebra()
|> Inspect.Algebra.format(98) |> Inspect.Algebra.format(98)
|> IO.iodata_to_binary() |> IO.iodata_to_binary()
@@ -785,17 +803,16 @@ defmodule Module.Types.Descr do
defp atom_to_quoted({:union, a}) do defp atom_to_quoted({:union, a}) do
if :sets.is_subset(@boolset, a) do if :sets.is_subset(@boolset, a) do
:sets.subtract(a, @boolset) entries =
|> :sets.to_list() :sets.subtract(a, @boolset)
|> Enum.sort() |> :sets.to_list()
|> Enum.reduce({:boolean, [], []}, &{:or, [], [&2, literal_to_quoted(&1)]}) |> Enum.map(&literal_to_quoted/1)
[{:boolean, [], []} | entries]
else else
:sets.to_list(a) :sets.to_list(a)
|> Enum.sort()
|> Enum.map(&literal_to_quoted/1) |> Enum.map(&literal_to_quoted/1)
|> Enum.reduce(&{:or, [], [&2, &1]})
end end
|> List.wrap()
end end
defp atom_to_quoted({:negation, a}) do defp atom_to_quoted({:negation, a}) do
@@ -1035,16 +1052,16 @@ defmodule Module.Types.Descr do
end end
end end
defp list_to_quoted(dnf, name) do defp list_to_quoted(dnf, name, opts) do
dnf = list_normalize(dnf) dnf = list_normalize(dnf)
for {list_type, last_type, negs} <- dnf, reduce: [] do for {list_type, last_type, negs} <- dnf, reduce: [] do
acc -> acc ->
arguments = arguments =
if subtype?(last_type, @empty_list) do if subtype?(last_type, @empty_list) do
[to_quoted(list_type)] [to_quoted(list_type, opts)]
else else
[to_quoted(list_type), to_quoted(last_type)] [to_quoted(list_type, opts), to_quoted(last_type, opts)]
end end
if negs == [] do if negs == [] do
@@ -1054,9 +1071,9 @@ defmodule Module.Types.Descr do
|> Enum.map(fn {ty, lst} -> |> Enum.map(fn {ty, lst} ->
args = args =
if subtype?(lst, @empty_list) do if subtype?(lst, @empty_list) do
[to_quoted(ty)] [to_quoted(ty, opts)]
else else
[to_quoted(ty), to_quoted(lst)] [to_quoted(ty, opts), to_quoted(lst, opts)]
end end
{name, [], args} {name, [], args}
@@ -1064,11 +1081,7 @@ defmodule Module.Types.Descr do
|> Enum.reduce(&{:or, [], [&2, &1]}) |> Enum.reduce(&{:or, [], [&2, &1]})
|> Kernel.then( |> Kernel.then(
&[ &[
{:and, [], {:and, [], [{name, [], arguments}, {:not, [], [&1]}]}
[
{name, [], arguments},
{:not, [], [&1]}
]}
| acc | acc
] ]
) )
@@ -1170,7 +1183,7 @@ defmodule Module.Types.Descr do
end end
end end
defp dynamic_to_quoted(descr) do defp dynamic_to_quoted(descr, opts) do
cond do cond do
term_type?(descr) -> term_type?(descr) ->
[{:dynamic, [], []}] [{:dynamic, [], []}]
@@ -1179,7 +1192,7 @@ defmodule Module.Types.Descr do
[single] [single]
true -> true ->
case to_quoted(descr) do case to_quoted(descr, opts) do
{:none, _meta, []} = none -> [none] {:none, _meta, []} = none -> [none]
descr -> [{:dynamic, [], [descr]}] descr -> [{:dynamic, [], [descr]}]
end end
@@ -1691,7 +1704,7 @@ defmodule Module.Types.Descr do
if map_empty_negation?(tag, acc_fields, neg) do if map_empty_negation?(tag, acc_fields, neg) do
{acc_fields, acc_negs} {acc_fields, acc_negs}
else else
case all_but_one?(tag, acc_fields, neg_tag, neg_fields) do case map_all_but_one?(tag, acc_fields, neg_tag, neg_fields) do
{:one, diff_key} -> {:one, diff_key} ->
{Map.update!(acc_fields, diff_key, &difference(&1, neg_fields[diff_key])), {Map.update!(acc_fields, diff_key, &difference(&1, neg_fields[diff_key])),
acc_negs} acc_negs}
@@ -1704,10 +1717,55 @@ defmodule Module.Types.Descr do
{tag, fields, negs} {tag, fields, negs}
end) end)
|> map_fusion()
end
# Given a dnf, fuse maps when possible
# e.g. %{a: integer(), b: atom()} or %{a: float(), b: atom()} into %{a: number(), b: atom()}
defp map_fusion(dnf) do
# Steps:
# 1. Group maps by tags and keys
# 2. Try fusions for each group until no fusion is found
# 3. Merge the groups back into a dnf
{without_negs, with_negs} = Enum.split_with(dnf, fn {_tag, _fields, negs} -> negs == [] end)
without_negs =
without_negs
|> Enum.group_by(fn {tag, fields, _} -> {tag, Map.keys(fields)} end)
|> Enum.flat_map(fn {_, maps} -> map_non_negated_fuse(maps) end)
without_negs ++ with_negs
end
defp map_non_negated_fuse(maps) do
Enum.reduce(maps, [], fn map, acc ->
case Enum.split_while(acc, &non_fusible_maps?(map, &1)) do
{_, []} ->
[map | acc]
{others, [match | rest]} ->
fused = map_non_negated_fuse_pair(map, match)
others ++ [fused | rest]
end
end)
end
# Two maps are fusible if they differ in at most one element.
defp non_fusible_maps?({_, fields1, []}, {_, fields2, []}) do
Enum.count_until(fields1, fn {key, value} -> Map.fetch!(fields2, key) != value end, 2) > 1
end
defp map_non_negated_fuse_pair({tag, fields1, []}, {_, fields2, []}) do
fields =
symmetrical_merge(fields1, fields2, fn _k, v1, v2 ->
if v1 == v2, do: v1, else: union(v1, v2)
end)
{tag, fields, []}
end end
# If all fields are the same except one, we can optimize map difference. # If all fields are the same except one, we can optimize map difference.
defp all_but_one?(tag1, fields1, tag2, fields2) do defp map_all_but_one?(tag1, fields1, tag2, fields2) do
keys1 = Map.keys(fields1) keys1 = Map.keys(fields1)
keys2 = Map.keys(fields2) keys2 = Map.keys(fields2)
@@ -1735,55 +1793,77 @@ defmodule Module.Types.Descr do
end)) end))
end end
defp map_to_quoted(dnf) do defp map_to_quoted(dnf, opts) do
dnf dnf
|> map_normalize() |> map_normalize()
|> Enum.map(&map_each_to_quoted/1) |> Enum.map(&map_each_to_quoted(&1, opts))
|> case do
[] -> []
dnf -> Enum.reduce(dnf, &{:or, [], [&2, &1]}) |> List.wrap()
end
end end
defp map_each_to_quoted({tag, positive_map, negative_maps}) do defp map_each_to_quoted({tag, positive_map, negative_maps}, opts) do
case negative_maps do case negative_maps do
[] -> [] ->
map_literal_to_quoted({tag, positive_map}) map_literal_to_quoted({tag, positive_map}, opts)
_ -> _ ->
negative_maps negative_maps
|> Enum.map(&map_literal_to_quoted/1) |> Enum.map(&map_literal_to_quoted(&1, opts))
|> Enum.reduce(&{:or, [], [&2, &1]}) |> Enum.reduce(&{:or, [], [&2, &1]})
|> Kernel.then( |> Kernel.then(
&{:and, [], [map_literal_to_quoted({tag, positive_map}), {:not, [], [&1]}]} &{:and, [], [map_literal_to_quoted({tag, positive_map}, opts), {:not, [], [&1]}]}
) )
end end
end end
def map_literal_to_quoted({:closed, fields}) when map_size(fields) == 0 do def map_literal_to_quoted({:closed, fields}, _opts) when map_size(fields) == 0 do
{:empty_map, [], []} {:empty_map, [], []}
end end
def map_literal_to_quoted({tag, fields}) do def map_literal_to_quoted({tag, fields}, opts) do
case tag do case tag do
:closed -> :closed ->
with %{__struct__: struct_descr} <- fields, with %{__struct__: struct_descr} <- fields,
{_, [struct]} <- atom_fetch(struct_descr) do {_, [struct]} <- atom_fetch(struct_descr) do
fields = Map.delete(fields, :__struct__)
fields =
with true <- Keyword.get(opts, :collapse_structs, false),
[_ | _] = info <- maybe_struct(struct),
true <- Enum.all?(info, &is_map_key(fields, &1.field)) do
Enum.reduce(info, fields, fn %{field: field}, acc ->
# TODO: This should consider the struct default value
if Map.fetch!(acc, field) == term() do
Map.delete(acc, field)
else
acc
end
end)
else
_ -> fields
end
{:%, [], {:%, [],
[ [
literal_to_quoted(struct), literal_to_quoted(struct),
{:%{}, [], map_fields_to_quoted(tag, Map.delete(fields, :__struct__))} {:%{}, [], map_fields_to_quoted(tag, fields, opts)}
]} ]}
else else
_ -> {:%{}, [], map_fields_to_quoted(tag, fields)} _ -> {:%{}, [], map_fields_to_quoted(tag, fields, opts)}
end end
:open -> :open ->
{:%{}, [], [{:..., [], nil} | map_fields_to_quoted(tag, fields)]} {:%{}, [], [{:..., [], nil} | map_fields_to_quoted(tag, fields, opts)]}
end end
end end
defp map_fields_to_quoted(tag, map) do defp maybe_struct(struct) do
try do
struct.__info__(:struct)
rescue
_ -> nil
end
end
defp map_fields_to_quoted(tag, map, opts) do
sorted = Enum.sort(Map.to_list(map)) sorted = Enum.sort(Map.to_list(map))
keyword? = Inspect.List.keyword?(sorted) keyword? = Inspect.List.keyword?(sorted)
@@ -1799,9 +1879,9 @@ defmodule Module.Types.Descr do
{optional?, type} = pop_optional_static(type) {optional?, type} = pop_optional_static(type)
cond do cond do
not optional? -> {key, to_quoted(type)} not optional? -> {key, to_quoted(type, opts)}
empty?(type) -> {key, {:not_set, [], []}} empty?(type) -> {key, {:not_set, [], []}}
true -> {key, {:if_set, [], [to_quoted(type)]}} true -> {key, {:if_set, [], [to_quoted(type, opts)]}}
end end
end end
end end
@@ -1922,37 +2002,78 @@ defmodule Module.Types.Descr do
# This is a cheap optimization that relies on structural equality. # This is a cheap optimization that relies on structural equality.
defp tuple_union(left, right), do: left ++ (right -- left) defp tuple_union(left, right), do: left ++ (right -- left)
defp tuple_to_quoted(dnf) do defp tuple_to_quoted(dnf, opts) do
dnf dnf
|> tuple_simplify() |> tuple_simplify()
|> Enum.map(&tuple_each_to_quoted/1) |> tuple_fusion()
|> case do |> Enum.map(&tuple_each_to_quoted(&1, opts))
[] -> []
dnf -> Enum.reduce(dnf, &{:or, [], [&2, &1]}) |> List.wrap()
end
end end
defp tuple_each_to_quoted({tag, positive_map, negative_maps}) do # Given a dnf of tuples, fuses the tuple unions when possible,
case negative_maps do # e.g. {integer(), atom()} or {float(), atom()} into {number(), atom()}
# The negations of two fused tuples are just concatenated.
defp tuple_fusion(dnf) do
# Steps:
# 1. Consider tuples without negations apart from those with
# 2. Group tuples by size and tag
# 3. Try fusions for each group until no fusion is found
# 4. Merge the groups back into a dnf
{without_negs, with_negs} = Enum.split_with(dnf, fn {_tag, _elems, negs} -> negs == [] end)
without_negs =
without_negs
|> Enum.group_by(fn {tag, elems, _} -> {tag, length(elems)} end)
|> Enum.flat_map(fn {_, tuples} -> tuple_non_negated_fuse(tuples) end)
without_negs ++ with_negs
end
defp tuple_non_negated_fuse(tuples) do
Enum.reduce(tuples, [], fn tuple, acc ->
case Enum.split_while(acc, &non_fusible_tuples?(tuple, &1)) do
{_, []} ->
[tuple | acc]
{others, [match | rest]} ->
fused = tuple_non_negated_fuse_pair(tuple, match)
others ++ [fused | rest]
end
end)
end
# Two tuples are fusible if they have no negations and differ in at most one element.
defp non_fusible_tuples?({_, elems1, []}, {_, elems2, []}) do
Enum.zip(elems1, elems2) |> Enum.count_until(fn {a, b} -> a != b end, 2) > 1
end
defp tuple_non_negated_fuse_pair({tag, elems1, []}, {_, elems2, []}) do
fused_elements =
Enum.zip_with(elems1, elems2, fn a, b -> if a == b, do: a, else: union(a, b) end)
{tag, fused_elements, []}
end
defp tuple_each_to_quoted({tag, positive_tuple, negative_tuples}, opts) do
case negative_tuples do
[] -> [] ->
tuple_literal_to_quoted({tag, positive_map}) tuple_literal_to_quoted({tag, positive_tuple}, opts)
_ -> _ ->
negative_maps negative_tuples
|> Enum.map(&tuple_literal_to_quoted/1) |> Enum.map(&tuple_literal_to_quoted(&1, opts))
|> Enum.reduce(&{:or, [], [&2, &1]}) |> Enum.reduce(&{:or, [], [&2, &1]})
|> Kernel.then( |> Kernel.then(
&{:and, [], [tuple_literal_to_quoted({tag, positive_map}), {:not, [], [&1]}]} &{:and, [], [tuple_literal_to_quoted({tag, positive_tuple}, opts), {:not, [], [&1]}]}
) )
end end
end end
defp tuple_literal_to_quoted({:closed, []}), do: {:{}, [], []} defp tuple_literal_to_quoted({:closed, []}, _opts), do: {:{}, [], []}
defp tuple_literal_to_quoted({tag, elements}) do defp tuple_literal_to_quoted({tag, elements}, opts) do
case tag do case tag do
:closed -> {:{}, [], Enum.map(elements, &to_quoted/1)} :closed -> {:{}, [], Enum.map(elements, &to_quoted(&1, opts))}
:open -> {:{}, [], Enum.map(elements, &to_quoted/1) ++ [{:..., [], nil}]} :open -> {:{}, [], Enum.map(elements, &to_quoted(&1, opts)) ++ [{:..., [], nil}]}
end end
end end
+2 -1
View File
@@ -444,7 +444,8 @@ defmodule Module.Types.Expr do
# to avoid export dependencies. So we do it here. # to avoid export dependencies. So we do it here.
if Code.ensure_loaded?(exception) and function_exported?(exception, :__struct__, 0) do if Code.ensure_loaded?(exception) and function_exported?(exception, :__struct__, 0) do
{info, context} = Of.struct_info(exception, meta, stack, context) {info, context} = Of.struct_info(exception, meta, stack, context)
{Of.struct_type(exception, info, args), context} # TODO: For properly defined structs, this should not be dynamic
{dynamic(Of.struct_type(exception, info, args)), context}
else else
# If the exception cannot be found or is invalid, fetch the signature to emit warnings. # If the exception cannot be found or is invalid, fetch the signature to emit warnings.
{_, context} = Apply.signature(exception, :__struct__, 0, meta, stack, context) {_, context} = Apply.signature(exception, :__struct__, 0, meta, stack, context)
+1 -1
View File
@@ -141,7 +141,7 @@ defmodule Module.Types.Helpers do
column: meta[:column], column: meta[:column],
hints: formatter_hints ++ expr_hints(expr), hints: formatter_hints ++ expr_hints(expr),
formatted_expr: formatted_expr, formatted_expr: formatted_expr,
formatted_type: Module.Types.Descr.to_quoted_string(type) formatted_type: Module.Types.Descr.to_quoted_string(type, collapse_structs: true)
} }
end) end)
|> Enum.sort_by(&{&1.line, &1.column}) |> Enum.sort_by(&{&1.line, &1.column})
+1 -3
View File
@@ -495,13 +495,11 @@ defmodule Module.Types.Of do
unknown key .#{key} in expression: unknown key .#{key} in expression:
#{expr_to_string(expr) |> indent(4)} #{expr_to_string(expr) |> indent(4)}
""",
empty_if(dot_var?(expr), """
the given type does not have the given key: the given type does not have the given key:
#{to_quoted_string(type) |> indent(4)} #{to_quoted_string(type) |> indent(4)}
"""), """,
format_traces(traces) format_traces(traces)
]) ])
} }
+1 -1
View File
@@ -192,7 +192,7 @@ defmodule Range do
IO.warn_once( IO.warn_once(
{__MODULE__, :new}, {__MODULE__, :new},
fn -> fn ->
"Range.new/2 has a default step of -1, please call Range.new/3 explicitly passing the step of -1 instead" "Range.new/2 and first..last default to a step of -1 when last < first. Use Range.new(first, last, -1) or first..last//-1, or pass 1 if that was your intention"
end, end,
3 3
) )
@@ -8,12 +8,11 @@ Elixir applies bug fixes only to the latest minor branch. Security patches are a
Elixir version | Support Elixir version | Support
:------------- | :----------------------------- :------------- | :-----------------------------
1.18 | Development 1.18 | Bug fixes and security patches
1.17 | Bug fixes and security patches 1.17 | Security patches only
1.16 | Security patches only 1.16 | Security patches only
1.15 | Security patches only 1.15 | Security patches only
1.14 | Security patches only 1.14 | Security patches only
1.13 | Security patches only
New releases are announced in the read-only [announcements mailing list](https://groups.google.com/group/elixir-lang-ann). All security releases [will be tagged with `[security]`](https://groups.google.com/forum/#!searchin/elixir-lang-ann/%5Bsecurity%5D%7Csort:date). New releases are announced in the read-only [announcements mailing list](https://groups.google.com/group/elixir-lang-ann). All security releases [will be tagged with `[security]`](https://groups.google.com/forum/#!searchin/elixir-lang-ann/%5Bsecurity%5D%7Csort:date).
@@ -43,6 +42,7 @@ Erlang/OTP versioning is independent from the versioning of Elixir. Erlang relea
Elixir version | Supported Erlang/OTP versions Elixir version | Supported Erlang/OTP versions
:------------- | :------------------------------- :------------- | :-------------------------------
1.18 | 25 - 27
1.17 | 25 - 27 1.17 | 25 - 27
1.16 | 24 - 26 1.16 | 24 - 26
1.15 | 24 - 26 1.15 | 24 - 26
@@ -82,6 +82,14 @@ The first column is the version the feature was hard deprecated. The second colu
Version | Deprecated feature | Replaced by (available since) Version | Deprecated feature | Replaced by (available since)
:-------| :-------------------------------------------------- | :--------------------------------------------------------------- :-------| :-------------------------------------------------- | :---------------------------------------------------------------
[v1.18] | `<%#` in EEx | `<%!--` (v1.14) or `<% #` (v1.0)
[v1.18] | `handle_text/2` callback in EEx | `handle_text/3` (v1.14)
[v1.18] | Returning 2-arity fun from `Enumerable.slice/1` | Returning 3-arity (v1.14)
[v1.18] | Ranges with negative steps in `Range.new/2` | Explicit steps in ranges (v1.11)
[v1.18] | `Tuple.append/2` | `Tuple.insert_at/3` (v1.0)
[v1.18] | `mix cmd --app APP` | `mix do --app APP` (v1.14)
[v1.18] | `List.zip/1` | `Enum.zip/1` (v1.0)
[v1.18] | `Module.eval_quoted/3` | `Code.eval_quoted/3` (v1.0)
[v1.17] | Single-quoted charlists (`'foo'`) | `~c"foo"` (v1.0) [v1.17] | Single-quoted charlists (`'foo'`) | `~c"foo"` (v1.0)
[v1.17] | `left..right` in patterns and guards | `left..right//step` (v1.11) [v1.17] | `left..right` in patterns and guards | `left..right//step` (v1.11)
[v1.17] | `ExUnit.Case.register_test/4` | `register_test/6` (v1.10) [v1.17] | `ExUnit.Case.register_test/4` | `register_test/6` (v1.10)
@@ -214,4 +222,5 @@ Version | Deprecated feature | Replaced by (ava
[v1.14]: https://github.com/elixir-lang/elixir/blob/v1.14/CHANGELOG.md#4-hard-deprecations [v1.14]: https://github.com/elixir-lang/elixir/blob/v1.14/CHANGELOG.md#4-hard-deprecations
[v1.15]: https://github.com/elixir-lang/elixir/blob/v1.15/CHANGELOG.md#4-hard-deprecations [v1.15]: https://github.com/elixir-lang/elixir/blob/v1.15/CHANGELOG.md#4-hard-deprecations
[v1.16]: https://github.com/elixir-lang/elixir/blob/v1.16/CHANGELOG.md#4-hard-deprecations [v1.16]: https://github.com/elixir-lang/elixir/blob/v1.16/CHANGELOG.md#4-hard-deprecations
[v1.17]: https://github.com/elixir-lang/elixir/blob/main/CHANGELOG.md#4-hard-deprecations [v1.17]: https://github.com/elixir-lang/elixir/blob/v1.17/CHANGELOG.md#4-hard-deprecations
[v1.18]: https://github.com/elixir-lang/elixir/blob/v1.18/CHANGELOG.md#4-hard-deprecations
+15 -4
View File
@@ -11,7 +11,7 @@
parallel_match(Meta, Expr, S, #{context := match} = E) -> parallel_match(Meta, Expr, S, #{context := match} = E) ->
#elixir_ex{vars={_Read, Write}} = S, #elixir_ex{vars={_Read, Write}} = S,
Matches = unpack_match(Expr, Meta, []), Matches = unpack_match(Expr, Meta, [], E),
{[{_, EHead} | ETail], EWrites, SM, EM} = {[{_, EHead} | ETail], EWrites, SM, EM} =
lists:foldl(fun({EMeta, Match}, {AccMatches, AccWrites, SI, EI}) -> lists:foldl(fun({EMeta, Match}, {AccMatches, AccWrites, SI, EI}) ->
@@ -31,9 +31,13 @@ parallel_match(Meta, Expr, S, #{context := match} = E) ->
VWrite = (Write /= false) andalso elixir_env:merge_vars(Write, PWrites), VWrite = (Write /= false) andalso elixir_env:merge_vars(Write, PWrites),
{EMatch, SM#elixir_ex{vars={VRead, VWrite}, prematch={PRead, PCycles, PInfo}}, EM}. {EMatch, SM#elixir_ex{vars={VRead, VWrite}, prematch={PRead, PCycles, PInfo}}, EM}.
unpack_match({'=', Meta, [Left, Right]}, _Meta, Acc) -> unpack_match({'=', Meta, [{_, VarMeta, _} = Node, Node]}, _Meta, Acc, E) ->
unpack_match(Left, Meta, unpack_match(Right, Meta, Acc)); %% TODO: remove this clause on Elixir v1.23
unpack_match(Node, Meta, Acc) -> file_warn(VarMeta, ?key(E, file), ?MODULE, {duplicate_match, Node}),
unpack_match(Node, Meta, Acc, E);
unpack_match({'=', Meta, [Left, Right]}, _Meta, Acc, E) ->
unpack_match(Left, Meta, unpack_match(Right, Meta, Acc, E), E);
unpack_match(Node, Meta, Acc, _E) ->
[{Meta, Node} | Acc]. [{Meta, Node} | Acc].
store_cycles([Write | Writes], {Cycles, SkipList}, Acc) -> store_cycles([Write | Writes], {Cycles, SkipList}, Acc) ->
@@ -532,6 +536,13 @@ origin(Meta, Default) ->
false -> Default false -> Default
end. end.
format_error({duplicate_match, Expr}) ->
String = 'Elixir.Macro':to_string(Expr),
io_lib:format(
"this pattern is matched against itself inside a match: ~ts = ~ts",
[String, String]
);
format_error({recursive, Vars, TypeExpr}) -> format_error({recursive, Vars, TypeExpr}) ->
Code = Code =
case TypeExpr of case TypeExpr of
+1
View File
@@ -203,6 +203,7 @@ bootstrap_files() ->
<<"path.ex">>, <<"path.ex">>,
<<"file.ex">>, <<"file.ex">>,
<<"map.ex">>, <<"map.ex">>,
<<"function.ex">>,
<<"range.ex">>, <<"range.ex">>,
<<"access.ex">>, <<"access.ex">>,
<<"io.ex">>, <<"io.ex">>,
+1 -1
View File
@@ -526,7 +526,7 @@ docs_chunk(Map, Set, Module, Anno, Def, Defmacro, Types, Callbacks, ChunkOpts) -
TypeDocs = get_type_docs(Set, Types), TypeDocs = get_type_docs(Set, Types),
ModuleMeta = ModuleDocMeta#{ ModuleMeta = ModuleDocMeta#{
source_path => File, source_path => elixir_utils:characters_to_list(File),
source_annos => [Anno], source_annos => [Anno],
behaviours => [Mod || {behaviour, Mod} <- Attributes] behaviours => [Mod || {behaviour, Mod} <- Attributes]
}, },
+4 -4
View File
@@ -216,12 +216,12 @@ highlight_at_position(Column, Severity, Length) ->
highlight_below_line(Line, Severity) -> highlight_below_line(Line, Severity) ->
% Don't highlight leading whitespaces in line % Don't highlight leading whitespaces in line
{_, SpacesMatched} = trim_line(Line, 0), {Rest, SpacesMatched} = trim_line(Line, 0),
Length = string:length(Line), Length = string:length(Rest),
Highlight = case Severity of Highlight = case Severity of
warning -> highlight(lists:duplicate(Length - SpacesMatched, $~), warning); warning -> highlight(lists:duplicate(Length, $~), warning);
error -> highlight(lists:duplicate(Length - SpacesMatched, $^), error) error -> highlight(lists:duplicate(Length, $^), error)
end, end,
[n_spaces(SpacesMatched), Highlight]. [n_spaces(SpacesMatched), Highlight].
+19 -7
View File
@@ -266,11 +266,7 @@ tokenize([$" | T], Line, Column, Scope, Tokens) ->
%% TODO: Remove me in Elixir v2.0 %% TODO: Remove me in Elixir v2.0
tokenize([$' | T], Line, Column, Scope, Tokens) -> tokenize([$' | T], Line, Column, Scope, Tokens) ->
Message = "single-quoted strings represent charlists. " handle_strings(T, Line, Column + 1, $', Scope, Tokens);
"Use ~c\"\" if you indeed want a charlist or use \"\" instead.\n"
"You may run \"mix format --migrate\" to fix this warning automatically.",
NewScope = prepend_warning(Line, Column, Message, Scope),
handle_strings(T, Line, Column + 1, $', NewScope, Tokens);
% Operator atoms % Operator atoms
@@ -786,7 +782,11 @@ handle_strings(T, Line, Column, H, Scope, Tokens) ->
"number do not require quotes", "number do not require quotes",
[hd(Parts)] [hd(Parts)]
), ),
prepend_warning(Line, Column, WarnMsg, InterScope); prepend_warning(Line, Column-1, WarnMsg, InterScope);
false when H =:= $' ->
WarnMsg = "single quotes around keywords are deprecated. Use double quotes instead",
prepend_warning(Line, Column-1, WarnMsg, InterScope);
false -> false ->
InterScope InterScope
@@ -814,7 +814,19 @@ handle_strings(T, Line, Column, H, Scope, Tokens) ->
error(Reason, Rest, NewScope, Tokens) error(Reason, Rest, NewScope, Tokens)
end; end;
{NewLine, NewColumn, Parts, Rest, NewScope} -> {NewLine, NewColumn, Parts, Rest, InterScope} ->
NewScope =
case H of
$' ->
Message = "single-quoted strings represent charlists. "
"Use ~c\"\" if you indeed want a charlist or use \"\" instead.\n"
"You may run \"mix format --migrate\" to fix this warning automatically.",
prepend_warning(Line, Column-1, Message, InterScope);
_ ->
InterScope
end,
case unescape_tokens(Parts, Line, Column, NewScope) of case unescape_tokens(Parts, Line, Column, NewScope) of
{ok, Unescaped} -> {ok, Unescaped} ->
Token = {string_type(H), {Line, Column - 1, nil}, Unescaped}, Token = {string_type(H), {Line, Column - 1, nil}, Unescaped},
+8
View File
@@ -46,6 +46,14 @@ defmodule JSONTest do
assert JSON.encode!(%Token{value: :example}) == "[\"example\"]" assert JSON.encode!(%Token{value: :example}) == "[\"example\"]"
assert JSON.encode!(%Token{value: "hello\0world"}) == "[\"hello\\u0000world\"]" assert JSON.encode!(%Token{value: "hello\0world"}) == "[\"hello\\u0000world\"]"
end end
test "calendar" do
assert JSON.encode!(~D[2010-04-17]) == "\"2010-04-17\""
assert JSON.encode!(~T[14:00:00.123]) == "\"14:00:00.123\""
assert JSON.encode!(~N[2010-04-17 14:00:00.123]) == "\"2010-04-17T14:00:00.123\""
assert JSON.encode!(~U[2010-04-17 14:00:00.123Z]) == "\"2010-04-17T14:00:00.123Z\""
assert JSON.encode!(Duration.new!(month: 2, hour: 3)) == "\"P2MT3H\""
end
end end
describe "JSON.Encoder" do describe "JSON.Encoder" do
@@ -859,6 +859,35 @@ defmodule Kernel.DiagnosticsTest do
purge(Sample) purge(Sample)
end end
@tag :tmp_dir
test "simple warning with tabs (line + file)", %{tmp_dir: tmp_dir} do
path = make_relative_tmp(tmp_dir, "long-warning.ex")
source = """
defmodule Sample do
\t@file "#{path}"
\tdefp a do
\t\tUnknown.b()
\tend
end
"""
File.write!(path, source)
expected = """
warning: Unknown.b/0 is undefined (module Unknown is not available or is yet to be defined). Make sure the module name is correct and has been specified in full (or that an alias has been defined)
│
4 │ \t\tUnknown.b()
│ ~~~~~~~~~~~
│
└─ #{path}:4: Sample.a/0
"""
assert capture_eval(source, columns: false) =~ expected
after
purge(Sample)
end
test "simple warning (no file)" do test "simple warning (no file)" do
source = """ source = """
defmodule Sample do defmodule Sample do
+1 -1
View File
@@ -255,7 +255,7 @@ defmodule Kernel.DocsTest do
assert module_doc == "Module doc" assert module_doc == "Module doc"
file = __ENV__.file file = String.to_charlist(__ENV__.file)
source_annos = [:erl_anno.new({line + 3, 19})] source_annos = [:erl_anno.new({line + 3, 19})]
@@ -190,17 +190,6 @@ defmodule Kernel.ExpansionTest do
end end
test "errors on directly recursive definitions" do test "errors on directly recursive definitions" do
assert_compile_error(
~r"""
recursive variable definition in patterns:
x = x
the variable "x" \(context Kernel.ExpansionTest\) is defined in function of itself
""",
fn -> expand(quote(do: (x = x) = :ok)) end
)
assert_compile_error( assert_compile_error(
~r""" ~r"""
recursive variable definition in patterns: recursive variable definition in patterns:
+15 -1
View File
@@ -190,7 +190,7 @@ defmodule Kernel.WarningTest do
) )
assert_warn_eval( assert_warn_eval(
["nofile:1:3", "found quoted keyword \"foo\" but the quotes are not required"], ["nofile:1:2", "found quoted keyword \"foo\" but the quotes are not required"],
~s/["foo": :bar]/ ~s/["foo": :bar]/
) )
@@ -264,6 +264,20 @@ defmodule Kernel.WarningTest do
purge(Sample) purge(Sample)
end end
test "duplicate pattern" do
output =
capture_eval("""
defmodule Sample do
var = quote(do: x)
def hello(unquote(var) = unquote(var)), do: unquote(var)
end
""")
assert output =~ "this pattern is matched against itself inside a match: x = x"
after
purge(Sample)
end
test "unused compiler variable" do test "unused compiler variable" do
output = output =
capture_eval(""" capture_eval("""
+2
View File
@@ -1136,6 +1136,8 @@ defmodule KernelTest do
test "pop_in/1" do test "pop_in/1" do
users = %{"john" => %{age: 27}, "meg" => %{age: 23}} users = %{"john" => %{age: 27}, "meg" => %{age: 23}}
assert pop_in(users["john"]) == {%{age: 27}, %{"meg" => %{age: 23}}}
assert pop_in(users["john"][:age]) == {27, %{"john" => %{}, "meg" => %{age: 23}}} assert pop_in(users["john"][:age]) == {27, %{"john" => %{}, "meg" => %{age: 23}}}
assert pop_in(users["john"][:name]) == {nil, %{"john" => %{age: 27}, "meg" => %{age: 23}}} assert pop_in(users["john"][:name]) == {nil, %{"john" => %{age: 27}, "meg" => %{age: 23}}}
assert pop_in(users["bob"][:age]) == {nil, %{"john" => %{age: 27}, "meg" => %{age: 23}}} assert pop_in(users["bob"][:age]) == {nil, %{"john" => %{age: 27}, "meg" => %{age: 23}}}
@@ -1184,7 +1184,7 @@ defmodule Module.Types.DescrTest do
test "boolean" do test "boolean" do
assert boolean() |> to_quoted_string() == "boolean()" assert boolean() |> to_quoted_string() == "boolean()"
assert atom([true, false, :a]) |> to_quoted_string() == "boolean() or :a" assert atom([true, false, :a]) |> to_quoted_string() == ":a or boolean()"
assert atom([true, :a]) |> to_quoted_string() == ":a or true" assert atom([true, :a]) |> to_quoted_string() == ":a or true"
assert difference(atom(), boolean()) |> to_quoted_string() == "atom() and not boolean()" assert difference(atom(), boolean()) |> to_quoted_string() == "atom() and not boolean()"
end end
@@ -1199,7 +1199,7 @@ defmodule Module.Types.DescrTest do
assert intersection(atom(), dynamic()) |> to_quoted_string() == "dynamic(atom())" assert intersection(atom(), dynamic()) |> to_quoted_string() == "dynamic(atom())"
assert union(atom([:foo, :bar]), dynamic()) |> to_quoted_string() == assert union(atom([:foo, :bar]), dynamic()) |> to_quoted_string() ==
"dynamic() or (:bar or :foo)" "dynamic() or :bar or :foo"
assert intersection(dynamic(), closed_map(a: integer())) |> to_quoted_string() == assert intersection(dynamic(), closed_map(a: integer())) |> to_quoted_string() ==
"dynamic(%{a: integer()})" "dynamic(%{a: integer()})"
@@ -1256,7 +1256,7 @@ defmodule Module.Types.DescrTest do
assert open_tuple([integer(), atom()]) |> to_quoted_string() == "{integer(), atom(), ...}" assert open_tuple([integer(), atom()]) |> to_quoted_string() == "{integer(), atom(), ...}"
assert union(tuple([integer(), atom()]), open_tuple([atom()])) |> to_quoted_string() == assert union(tuple([integer(), atom()]), open_tuple([atom()])) |> to_quoted_string() ==
"{integer(), atom()} or {atom(), ...}" "{atom(), ...} or {integer(), atom()}"
assert difference(tuple([integer(), atom()]), open_tuple([atom()])) |> to_quoted_string() == assert difference(tuple([integer(), atom()]), open_tuple([atom()])) |> to_quoted_string() ==
"{integer(), atom()}" "{integer(), atom()}"
@@ -1268,6 +1268,96 @@ defmodule Module.Types.DescrTest do
# assert difference(tuple([number(), term()]), tuple([integer(), atom()])) # assert difference(tuple([number(), term()]), tuple([integer(), atom()]))
# |> to_quoted_string() == # |> to_quoted_string() ==
# "{float(), term()} or {number(), term() and not atom()}" # "{float(), term()} or {number(), term() and not atom()}"
assert union(tuple([integer(), atom()]), tuple([integer(), atom()])) |> to_quoted_string() ==
"{integer(), atom()}"
assert union(tuple([integer(), atom()]), tuple([float(), atom()])) |> to_quoted_string() ==
"{float() or integer(), atom()}"
assert union(tuple([integer(), atom()]), tuple([float(), atom()]))
|> union(tuple([pid(), pid(), port()]))
|> union(tuple([pid(), pid(), atom()]))
|> to_quoted_string() ==
"{float() or integer(), atom()} or {pid(), pid(), atom() or port()}"
assert union(open_tuple([integer()]), open_tuple([float()])) |> to_quoted_string() ==
"{float() or integer(), ...}"
# {:ok, {term(), integer()}} or {:ok, {term(), float()}} or {:exit, :kill} or {:exit, :timeout}
assert tuple([atom([:ok]), tuple([term(), empty_list()])])
|> union(tuple([atom([:ok]), tuple([term(), open_map()])]))
|> union(tuple([atom([:exit]), atom([:kill])]))
|> union(tuple([atom([:exit]), atom([:timeout])]))
|> to_quoted_string() ==
"{:exit, :kill or :timeout} or {:ok, {term(), %{...} or empty_list()}}"
# Detection of duplicates
assert tuple([atom([:ok]), term()])
|> union(tuple([atom([:ok]), term()]))
|> to_quoted_string() == "{:ok, term()}"
assert tuple([closed_map(a: integer(), b: atom()), open_map()])
|> union(tuple([closed_map(a: integer(), b: atom()), open_map()]))
|> to_quoted_string() ==
"{%{a: integer(), b: atom()}, %{...}}"
# Nested fusion
assert tuple([closed_map(a: integer(), b: atom()), open_map()])
|> union(tuple([closed_map(a: float(), b: atom()), open_map()]))
|> to_quoted_string() ==
"{%{a: float() or integer(), b: atom()}, %{...}}"
# Complex simplification of map/tuple combinations. Initial type is:
# ```
# dynamic(
# :error or
# ({%Decimal{coef: :inf, exp: integer(), sign: integer()}, binary()} or
# {%Decimal{coef: :NaN, exp: integer(), sign: integer()}, binary()} or
# {%Decimal{coef: integer(), exp: integer(), sign: integer()}, term()} or
# {%Decimal{coef: :inf, exp: integer(), sign: integer()} or
# %Decimal{coef: :NaN, exp: integer(), sign: integer()} or
# %Decimal{coef: integer(), exp: integer(), sign: integer()}, term()})
# )
# ```
decimal_inf =
closed_map(
__struct__: atom([Decimal]),
coef: atom([:inf]),
exp: integer(),
sign: integer()
)
decimal_nan =
closed_map(
__struct__: atom([Decimal]),
coef: atom([:NaN]),
exp: integer(),
sign: integer()
)
decimal_int =
closed_map(__struct__: atom([Decimal]), coef: integer(), exp: integer(), sign: integer())
assert atom([:error])
|> union(
tuple([decimal_inf, binary()])
|> union(
tuple([decimal_nan, binary()])
|> union(
tuple([decimal_int, term()])
|> union(tuple([union(decimal_inf, union(decimal_nan, decimal_int)), term()]))
)
)
)
|> dynamic()
|> to_quoted_string() ==
"""
dynamic(
:error or {%Decimal{coef: :NaN or :inf, exp: integer(), sign: integer()}, binary()} or
{%Decimal{coef: :NaN or :inf or integer(), exp: integer(), sign: integer()}, term()}
)\
"""
end end
test "map" do test "map" do
@@ -1311,6 +1401,50 @@ defmodule Module.Types.DescrTest do
assert difference(open_map(a: number(), b: atom()), open_map(a: integer())) assert difference(open_map(a: number(), b: atom()), open_map(a: integer()))
|> to_quoted_string() == "%{..., a: float(), b: atom()}" |> to_quoted_string() == "%{..., a: float(), b: atom()}"
# Basic map fusion
assert union(closed_map(a: integer()), closed_map(a: integer())) |> to_quoted_string() ==
"%{a: integer()}"
assert union(closed_map(a: integer()), closed_map(a: float())) |> to_quoted_string() ==
"%{a: float() or integer()}"
# Nested fusion
assert union(closed_map(a: integer(), b: atom()), closed_map(a: float(), b: atom()))
|> union(closed_map(x: pid(), y: pid(), z: port()))
|> union(closed_map(x: pid(), y: pid(), z: atom()))
|> to_quoted_string() ==
"%{a: float() or integer(), b: atom()} or %{x: pid(), y: pid(), z: atom() or port()}"
# Open map fusion
assert union(open_map(a: integer()), open_map(a: float())) |> to_quoted_string() ==
"%{..., a: float() or integer()}"
# Fusing complex nested maps with unions
assert closed_map(status: atom([:ok]), data: closed_map(value: term(), count: empty_list()))
|> union(
closed_map(status: atom([:ok]), data: closed_map(value: term(), count: open_map()))
)
|> union(closed_map(status: atom([:error]), reason: atom([:timeout])))
|> union(closed_map(status: atom([:error]), reason: atom([:crash])))
|> to_quoted_string() ==
"%{data: %{count: %{...} or empty_list(), value: term()}, status: :ok} or\n %{reason: :crash or :timeout, status: :error}"
# Difference and union tests
assert closed_map(status: atom([:ok]), value: term())
|> difference(closed_map(status: atom([:ok]), value: float()))
|> union(
closed_map(status: atom([:ok]), value: term())
|> difference(closed_map(status: atom([:ok]), value: integer()))
)
|> to_quoted_string() ==
"%{status: :ok, value: term()}"
# Nested map fusion
assert closed_map(data: closed_map(x: integer(), y: atom()), meta: open_map())
|> union(closed_map(data: closed_map(x: float(), y: atom()), meta: open_map()))
|> to_quoted_string() ==
"%{data: %{x: float() or integer(), y: atom()}, meta: %{...}}"
# Test complex combinations # Test complex combinations
assert intersection(open_map(a: number(), b: atom()), open_map(a: integer(), c: boolean())) assert intersection(open_map(a: number(), b: atom()), open_map(a: integer(), c: boolean()))
|> union(difference(open_map(x: atom()), open_map(x: boolean()))) |> union(difference(open_map(x: atom()), open_map(x: boolean())))
@@ -798,9 +798,9 @@ defmodule Module.Types.ExprTest do
x.foo_bar x.foo_bar
where "x" was given the type: the given type does not have the given key:
# type: dynamic(%URI{ dynamic(%URI{
authority: term(), authority: term(),
fragment: term(), fragment: term(),
host: term(), host: term(),
@@ -810,6 +810,10 @@ defmodule Module.Types.ExprTest do
scheme: term(), scheme: term(),
userinfo: term() userinfo: term()
}) })
where "x" was given the type:
# type: dynamic(%URI{})
# from: types_test.ex:LINE-4:43 # from: types_test.ex:LINE-4:43
x = %URI{} x = %URI{}
""" """
@@ -845,6 +849,10 @@ defmodule Module.Types.ExprTest do
min(x, y) min(x, y)
given types:
min(dynamic(:foo), integer())
where "x" was given the type: where "x" was given the type:
# type: dynamic(:foo) # type: dynamic(:foo)
@@ -868,6 +876,10 @@ defmodule Module.Types.ExprTest do
x === y x === y
given types:
integer() === float()
where "x" was given the type: where "x" was given the type:
# type: integer() # type: integer()
@@ -893,6 +905,10 @@ defmodule Module.Types.ExprTest do
mod.<=(x, y) mod.<=(x, y)
given types:
dynamic(:foo) <= dynamic(%Point{})
where "mod" was given the type: where "mod" was given the type:
# type: dynamic(Kernel) # type: dynamic(Kernel)
@@ -907,7 +923,7 @@ defmodule Module.Types.ExprTest do
where "y" was given the type: where "y" was given the type:
# type: dynamic(%Point{x: term(), y: term(), z: term()}) # type: dynamic(%Point{})
# from: types_test.ex:LINE-2 # from: types_test.ex:LINE-2
y = %Point{} y = %Point{}
@@ -1224,16 +1240,18 @@ defmodule Module.Types.ExprTest do
e e
end end
) == ) ==
union( dynamic(
closed_map( union(
__struct__: atom([ArgumentError]), closed_map(
__exception__: atom([true]), __struct__: atom([ArgumentError]),
message: term() __exception__: atom([true]),
), message: term()
closed_map( ),
__struct__: atom([RuntimeError]), closed_map(
__exception__: atom([true]), __struct__: atom([RuntimeError]),
message: term() __exception__: atom([true]),
message: term()
)
) )
) )
end end
@@ -166,9 +166,13 @@ defmodule Module.Types.PatternTest do
x.foo_bar x.foo_bar
the given type does not have the given key:
dynamic(%Point{x: term(), y: term(), z: term()})
where "x" was given the type: where "x" was given the type:
# type: dynamic(%Point{x: term(), y: term(), z: term()}) # type: dynamic(%Point{})
# from: types_test.ex:LINE-1 # from: types_test.ex:LINE-1
x = %Point{} x = %Point{}
""" """
+1 -1
View File
@@ -35,7 +35,7 @@ defmodule RangeTest do
assert ExUnit.CaptureIO.capture_io(:stderr, fn -> assert ExUnit.CaptureIO.capture_io(:stderr, fn ->
assert Range.new(3, 1) == 3..1//-1 assert Range.new(3, 1) == 3..1//-1
end) =~ "has a default step of -1" end) =~ "default to a step of -1"
end end
test "fields" do test "fields" do
+2 -4
View File
@@ -710,7 +710,8 @@ defmodule TypespecTest do
@type my_type :: (... -> any) @type my_type :: (... -> any)
end end
assert [type: {:my_type, {:type, _, :fun, []}, []}] = types(bytecode) assert [type: {:my_type, {:type, _, :fun, [{:type, _, :any}, {:type, _, :any, []}]}, []}] =
types(bytecode)
end end
test "@type with a fun with multiple arguments and return type" do test "@type with a fun with multiple arguments and return type" do
@@ -1520,9 +1521,6 @@ defmodule TypespecTest do
assert ast_string == assert ast_string ==
"@type literal_struct_all_fields_key_type() :: %TypespecTest.SomeStruct{key: integer()}" "@type literal_struct_all_fields_key_type() :: %TypespecTest.SomeStruct{key: integer()}"
{:built_in_fun, _, _} ->
assert ast_string == "@type built_in_fun() :: (... -> any())"
{:built_in_nonempty_list, _, _} -> {:built_in_nonempty_list, _, _} ->
assert ast_string == "@type built_in_nonempty_list() :: [...]" assert ast_string == "@type built_in_nonempty_list() :: [...]"
+6 -1
View File
@@ -315,6 +315,11 @@ defmodule ExUnit.Case do
end end
end end
@keys [:async, :group, :parameterize, :register]
@doc false
def __keys__(opts), do: Keyword.take(opts, @keys)
@doc false @doc false
def __register__(module, opts) do def __register__(module, opts) do
if not Keyword.keyword?(opts) do if not Keyword.keyword?(opts) do
@@ -324,7 +329,7 @@ defmodule ExUnit.Case do
end end
{register?, opts} = Keyword.pop(opts, :register, true) {register?, opts} = Keyword.pop(opts, :register, true)
{next_opts, opts} = Keyword.split(opts, [:async, :group, :parameterize]) {next_opts, opts} = Keyword.split(opts, @keys)
if opts != [] do if opts != [] do
IO.warn("unknown options given to ExUnit.Case: #{inspect(opts)}") IO.warn("unknown options given to ExUnit.Case: #{inspect(opts)}")
+1 -1
View File
@@ -83,7 +83,7 @@ defmodule ExUnit.CaseTemplate do
# We inject this code in the module that calls "use MyTemplate". # We inject this code in the module that calls "use MyTemplate".
def __proxy__(module, opts) do def __proxy__(module, opts) do
quote do quote do
use ExUnit.Case, unquote(opts) use ExUnit.Case, ExUnit.Case.__keys__(unquote(opts))
setup_all context do setup_all context do
unquote(module).__ex_unit__(:setup_all, context) unquote(module).__ex_unit__(:setup_all, context)
@@ -36,7 +36,7 @@ defmodule ExUnit.NestedCase do
end end
defmodule ExUnit.CaseTemplateTest do defmodule ExUnit.CaseTemplateTest do
use ExUnit.SampleCase, async: true use ExUnit.SampleCase, async: true, another_option: 123
use ExUnit.NestedCase use ExUnit.NestedCase
two = 2 two = 2
+16 -12
View File
@@ -391,16 +391,17 @@ defmodule IEx do
The supported options are: The supported options are:
* `:colors` * `:auto_reload`
* `:inspect`
* `:width`
* `:history_size`
* `:default_prompt`
* `:continuation_prompt`
* `:alive_prompt`
* `:alive_continuation_prompt` * `:alive_continuation_prompt`
* `:parser` * `:alive_prompt`
* `:colors`
* `:continuation_prompt`
* `:default_prompt`
* `:dot_iex` * `:dot_iex`
* `:history_size`
* `:inspect`
* `:parser`
* `:width`
They are discussed individually in the sections below. They are discussed individually in the sections below.
@@ -488,9 +489,6 @@ defmodule IEx do
* `:alive_continuation_prompt` - used when `Node.alive?/0` returns * `:alive_continuation_prompt` - used when `Node.alive?/0` returns
`true` and more input is expected `true` and more input is expected
* `:auto_reload` - when set to `true`, automatically purges in-memory
modules when they get invalidated by a concurrent compilation
The following values in the prompt string will be replaced appropriately: The following values in the prompt string will be replaced appropriately:
* `%counter` - the index of the history * `%counter` - the index of the history
@@ -510,10 +508,16 @@ defmodule IEx do
If the parser raises, the buffer is reset to an empty string. If the parser raises, the buffer is reset to an empty string.
## dot_iex ## `.iex`
Configure the file loaded into your IEx session when it starts. Configure the file loaded into your IEx session when it starts.
See more information [in the `.iex.exs` documentation](`m:IEx#module-the-iex-exs-file`). See more information [in the `.iex.exs` documentation](`m:IEx#module-the-iex-exs-file`).
## Auto reloading
When set to `true`, the `:auto_reload` option automatically purges
in-memory modules when they get invalidated by a concurrent compilation
happening in the Operating System.
""" """
@spec configure(keyword()) :: :ok @spec configure(keyword()) :: :ok
def configure(options) do def configure(options) do
+7
View File
@@ -189,6 +189,13 @@ defmodule Mix.Tasks.Compile.App do
Mix.Project.ensure_structure() Mix.Project.ensure_structure()
File.write!(target, IO.chardata_to_string(contents)) File.write!(target, IO.chardata_to_string(contents))
File.touch!(target, new_mtime) File.touch!(target, new_mtime)
# If we just created the .app file, it will have touched
# the directory mtime, so we need to reset it.
if current_properties == [] do
File.touch!(compile_path, new_mtime)
end
Mix.shell().info("Generated #{app} app") Mix.shell().info("Generated #{app} app")
{:ok, []} {:ok, []}
else else
+19 -6
View File
@@ -58,12 +58,25 @@ defmodule Mix.Tasks.Do do
{apps, args} = extract_apps_from_args(args) {apps, args} = extract_apps_from_args(args)
show_forgotten_apps_warning(apps) show_forgotten_apps_warning(apps)
Enum.each(gather_commands(args), fn [task | args] -> Enum.each(gather_commands(args), fn
if apps == [] do [task | args] ->
Mix.Task.run(task, args) if apps == [] do
else Mix.Task.run(task, args)
Mix.Task.run_in_apps(task, apps, args) else
end Mix.Task.run_in_apps(task, apps, args)
end
[] ->
Mix.raise("""
One of the commands passed to "mix do" is empty. Each command passed to "mix do" must \
have at least the task name. These are all invalid:
mix do
mix do my_task +
mix do + my_task
Run "mix help do" for more information.
""")
end) end)
end end
@@ -66,8 +66,6 @@ defmodule Mix.Tasks.Compile.AppTest do
test "generates .app file when changes happen" do test "generates .app file when changes happen" do
in_fixture("no_mixfile", fn -> in_fixture("no_mixfile", fn ->
Mix.Project.push(MixTest.Case.Sample) Mix.Project.push(MixTest.Case.Sample)
# Pre-create the compilation path to avoid mtime races
File.mkdir_p(Mix.Project.compile_path())
Mix.Tasks.Compile.Elixir.run([]) Mix.Tasks.Compile.Elixir.run([])
assert Mix.Tasks.Compile.App.run([]) == {:ok, []} assert Mix.Tasks.Compile.App.run([]) == {:ok, []}
+8
View File
@@ -13,6 +13,14 @@ defmodule Mix.Tasks.DoTest do
end) end)
end end
test "raises if a task is empty" do
for args <- [~w(), ~w(+), ~w(help +), ~w(+ help)] do
assert_raise Mix.Error, ~r"^One of the commands passed to \"mix do\" is empty", fn ->
Mix.Tasks.Do.run(args)
end
end
end
test "gather_command returns a list of commands" do test "gather_command returns a list of commands" do
assert gather_commands(["help", "+", "compile"]) == assert gather_commands(["help", "+", "compile"]) ==
[["help"], ["compile"]] [["help"], ["compile"]]