Elixir v1.12.0-rc.0 released

Elixir News & Announcements Elixir News
1

Release: Release v1.12.0-rc.0 · elixir-lang/elixir · GitHub

Note: this is a release candidate. Please try it out and give us feedback! Given this is a pre-release, you may not find it in your package manager. In this case, pick one of the alternative methods for installation from our Install page.

Elixir v1.12 is out with improvements to scripting, tighter Erlang/OTP 24 integration, stepped ranges, and dozen of new functions across the standard library. Overall this is a small release, which continues our tradition of bringing Elixir developers quality of live improvements every 6 months.

This is the last release to support Erlang/OTP 21.

Scripting improvements: Mix.install/2 and System.trap_signal/3

Elixir v1.12 brings new conveniences for those using Elixir for scripting (via .exs files). Elixir has been capable of managing dependencies for a quite long time, but it could only be done within Mix projects. In particular, the Elixir team is wary of global dependencies as any scripts that rely on system packages are brittle and hard to reproduce whenever your system changes.

Mix.install/2 is meant to be a sweetspot between single-file scripts and full-blown Mix projects. With Mix.install/2, you can list your dependencies on top of your scripts. When you execute the script for the first time, Elixir will download, compile, and cache your dependencies before running your script. Future invocations of the script will simply read the compiled artefacts from the cache:

Mix.install([:jason])
IO.puts Jason.encode!(%{hello: :world})

Mix.install/2 also performs protocol consolidation, which gives script developers an option to execute their code in the most performant format possible.

Note: Mix.install/2 is currently experimental and it may change in future releases.

Another improvement to scripting is the ability to trap exit signals via System.trap_signal/3. All you need is the signal name and a callback that will be invoked when the signal triggers. For example, ExUnit leverages this functionality to print all currently running tests when you abort the test suite via SIGQUIT (Ctrl+\\ ):

$ mix test
.......................................................................
.....................^\

Aborting test suite, the following have not completed:

* test query building [test/ecto/query_test.exs:48]
* test placeholders in Repo.insert_all [test/ecto/repo_test.exs:502]

Showing results so far...

78 doctests, 1042 tests, 0 failures

This is particularly useful when your tests get stuck and you want to know which one is the culprit.

Important: Trapping signals may have strong implications on how a system shuts down and behave in production and therefore it is extremely discouraged for libraries to set their own traps. Instead, they should redirect users to configure them themselves. The only cases where it is acceptable for libraries to set their own traps is when using Elixir in script mode, such as in .exs files and via Mix tasks.

Tighter Erlang/OTP 24 integration

Erlang/OTP 24 ships with JIT compilation support and Elixir developers don’t have to do anything to reap its benefits. There are many other features in Erlang/OTP 24 to look forwards to and Elixir v1.12 provides integration with many of them: such as support for 16bit floats in bitstrings as well as performance improvements in the compiler and during code evaluation.

Another excellent feature in Erlang/OTP 24 is the implementation of EEP 54, which provides extended error information for many functions in Erlang’s stdlib. Elixir v1.12 fully leverages this feature to improve reporting for errors coming from Erlang. For example, in earlier OTP versions, inserting an invalid argument into a ETS table that no longer exists would simply error with ArgumentError:

Interactive Elixir (1.11.0)
iex(1)> ets = :ets.new(:example, [])
#Reference<0.3845811859.2669281281.223553>
iex(2)> :ets.delete(ets)
true
iex(3)> :ets.insert(ets, :should_be_a_tuple)
** (ArgumentError) argument error
    (stdlib 3.15) :ets.insert(#Reference<0.3845811859.2669281281.223553>, :should_be_a_tuple)

However, in Elixir v1.12 with Erlang/OTP 24:

Interactive Elixir (1.12.0)
iex(1)> ets = :ets.new(:example, [])
#Reference<0.105641012.1058144260.76455>
iex(2)> :ets.delete(ets)
true
iex(3)> :ets.insert(ets, :should_be_a_tuple)
** (ArgumentError) errors were found at the given arguments:

  * 1st argument: the table identifier does not refer to an existing ETS table
  * 2nd argument: not a tuple

    (stdlib 3.15) :ets.insert(#Reference<0.105641012.1058144260.76455>, :should_be_a_tuple)

Stepped ranges

Elixir has support for ranges from before its v1.0 release. Ranges support only integers and are inclusive, using the mathematic notation a..b. Ranges in Elixir are either increasing 1..10 or decreasing 10..1 and the direction of the range was always inferred from the starting and stop positions. Ranges are always lazy as its values are emitted as they are enumerated rather than being computed upfront.

Unfortunately, due to this inference, it is not possible to have empty ranges. For example, if you want to create a list of n elements, you can express it with a range from 1..n, as 1..0 is a decreasing range with two elements.

Elixir v1.12 supports stepped ranges via the first..last//step notation. For example: 1..10//2 will emit the numbers 1, 3, 5, 7, and 9. You can consider the // operator to perform “range division”, as it effectively divides and rounds up the number of elements in the range by step. Steps can be either positive (increasing ranges) or negative (decreasing ranges). Stepped ranges bring more expressive power to Elixir ranges and they elegantly solve the empty range problem, as they allow the direction of the steps to be explicitly declared instead of inferred.

As of Elixir v1.12, implicitly decreasing ranges are soft-deprecated and warnings will be emitted in future Elixir versions based on our deprecation policy.

Additional functions

Elixir v1.12 has the additional of many functions across the standard library. The Enum module received additions such as Enum.count_until/2, Enum.product/1, Enum.zip_with/2, and more. The Integer module now includes Integer.pow/2 and Integer.extended_gcd/2. Finally, the Kernel module got two new functions, Kernel.then/2 and Kernel.tap/2, which are specially useful in |> pipelines.

v1.12.0 (2021-04-03)

1. Enhancements

EEx

  • [EEx.Engine] Add c:EEx.Engine.handle_text/3 callback that receives text metadata
  • [EEx.Engine] Emit warnings for unused “do” expression in EEx

Elixir

  • [Code] Do not add newlines around interpolation on code formatting. Note this means formatted code that has interpolation after the line length on Elixir v1.12 won’t be considered as formatted on earlier Elixir versions
  • [Calendar] Support basic datetime format in Calendar.ISO parsing functions
  • [Code] Improve evaluation performance on systems running on Erlang/OTP 24+
  • [Date] Support steps via Date.range/3
  • [DateTime] Add offset to DateTime.to_iso8601/2 (now to_iso8601/3)
  • [Enum] Add Enum.count_until/2 and Enum.count_until/3
  • [Enum] Add Enum.product/1
  • [Enum] Add Enum.zip_with/2, Enum.zip_with/3, Enum.zip_reduce/3, and Enum.zip_reduce/4
  • [Enum] Add support for functions as the second argument of Enum.with_index/2
  • [Exception] Show error_info data for exceptions coming from Erlang
  • [Float] Add Float.pow/2
  • [Integer] Add Integer.pow/2 and Integer.extended_gcd/2
  • [List] Add default value for List.first/1 and List.last/1
  • [Kernel] Add start..stop//step as support for stepped ranges
  • [Kernel] Also warn for literal structs on min/2 and max/2
  • [Kernel] Add Kernel.tap/2 and Kernel.then/2
  • [Kernel] Do not add runtime dependencies to remotes in typespecs
  • [Kernel] When there is an unused variable warning and there is a variable with the same name previously defined, suggest the user may have wanted to use the pin operator
  • [Kernel] Improve error messages on invalid character right after a number
  • [Kernel] Show removal and deprecated tips from Erlang/OTP
  • [Macro] Add export dependencies on Macro.struct!/2
  • [Macro] Support :newline to customize newlines escaping in Macro.unescape_string/2
  • [Module] Raise on invalid @dialyzer attributes
  • [Module] Add Module.get_definition/2 and Module.delete_definition/2
  • [Module] Allow @on_load to be a private function
  • [Module] Validate @dialyzer related module attributes
  • [Range] Add Range.new/3 and Range.size/1
  • [Regex] Add offset option to Regex.scan/3 and Regex.run/3
  • [Registry] Support :compression on Registry tables
  • [Registry] Support Registry.values/3 for reading values under a given key-pid pair
  • [Stream] Add Stream.zip_with/2 and Stream.zip_with/3
  • [String] Add :turkic mode option to String case functions
  • [System] Add System.trap_signal/3 and System.untrap_signal/2
  • [Tuple] Add Tuple.sum/1 and Tuple.product/1
  • [URI] Support RFC3986 compliant encoding and decoding of queries via the :rfc3986 option

ExUnit

  • [ExUnit] Intercept SIGQUIT (via Ctrl+\) and show a list of all aborted tests as well as intermediate test results
  • [ExUnit] Interpolate module attributes in match assertions diffs
  • [ExUnit] Print how much time is spent on async vs sync tests
  • [ExUnit] Improve error messages for doctests
  • [ExUnit] Compile doctests faster (often by two times)

IEx

  • [IEx] Make IEx’ parser configurable to allow special commands
  • [IEx] Show function signature when pressing tab after the opening parens of a function
  • [IEx] If an IEx expression starts with a binary operator, such as |>, automatically pipe in the result of the last expression

Mix

  • [Mix] Add Mix.install/2 for dynamically installing a list of dependencies
  • [Mix] Support :exit_code option in Mix.raise/2
  • [Mix] Discard MIX_ENV and MIX_TARGET values if they are empty strings
  • [Mix] Print the time taken to execute a task with on MIX_DEBUG=1
  • [mix compile.erlang] Compile multiple files in parallel
  • [mix escript.build] Deep merge configuration and ensure argv is set when executing config/runtime.exs
  • [mix release] Add RELEASE_PROG to releases with the name of the executable starting the release
  • [mix release] Support remote.vm.args to customize how the connecting VM boots
  • [mix test] Run all available tests if there are no pending --failed tests. This provides a better workflow as you no longer need to toggle the --failed flag between runs

2. Bug fixes

Elixir

  • [CLI] Ensure -e "" (with an empty string) parses correctly on Windows
  • [Inspect] Do not override user supplied :limit option for derived implementations
  • [Kernel] Allow heredoc inside a heredoc interpolation
  • [Kernel] Preserve CRLF on heredocs
  • [Kernel] Public functions without documentation now appear as an empty map on Code.fetch_docs/1, unless they start with underscore, where they remain as :none. This aligns Elixir’s implementation with EEP48
  • [Kernel] Do not crash when complex literals (binaries and maps) are used in guards
  • [Kernel] Properly parse keywords (such as end) followed by the :: operator
  • [Macro] Macro.decompose_call/1 now also consider tuples with more than 2 elements to not be valid calls
  • [Macro] Fix Macro.to_string/1 double-escaping of escape characters in sigils
  • [Macro] Fix Macro.underscore/1 on digits preceded by capitals: “FOO10” now becomes “foo10” instead of “fo_o10”
  • [Macro] Preserve underscores between digits on Macro.underscore/1
  • [OptionParser] Properly parse when numbers follow-up aliases, for example, -ab3 is now parsed as -a -b 3
  • [Path] Fix Path.relative_to/2 when referencing self
  • [Task] Ensure Task.async_stream/2 with ordered: false discard results as they are emitted, instead of needlessly accumulating inside the stream manager
  • [URI] Do not discard empty paths on URI.merge/2

ExUnit

  • [ExUnit.Case] Make @tag tmp_dir an absolute directory, avoiding inconsistencies if the test changes the current working directory
  • [ExUnit.Diff] Fix cases where the diffing algorithm would fail to print a pattern correct

IEx

  • [IEx] Fix auto-completion inside remote shells

Mix

  • [mix app.config] Do not emit false positive warnings when configured dependencies that have runtime: false set
  • [mix compile.elixir] Ensure that a manifest is generated even with no source code
  • [mix compile.elixir] Make sure export dependencies trigger recompilation when the dependency is removed as well as when the whole file is removed
  • [mix compile.elixir] Do not emit false positive warnings when a path dependency adds a module that is then used by the current application in the same mix compile cycle
  • [mix test] Ensure protocols within the current project are consolidated when --cover is given
  • [mix release] Improve compliance of release scripts with stripped down Linux installations
  • [mix release] Preserve file mode when copying non-beam ebin files
  • [mix xref] Ensure args are passed to the underlying mix compile call

3. Soft-deprecations (no warnings emitted)

Elixir

  • [Kernel] Using start..stop to match on ranges is soft-deprecated and will warn on future Elixir versions. Use start..stop//step instead
  • [Kernel] Using start..stop to create decreasing ranges is soft-deprecated and will warn on future versions. Use start..stop//-1 instead

4. Hard-deprecations

EEx

  • [EEx.Engine] use EEx.Engine is deprecated in favor of explicit delegation

Elixir

  • [Kernel] The binary operator ^^^ is deprecated. If you are using Bitwise.^^^/2, use Bitwise.bxor/2 instead
  • [Kernel] Deprecate @foo() in favor of @foo
  • [System] Deprecate System.stacktrace/0 (it was already deprecated outside of catch/rescue and now it is deprecated everywhere)

Mix

  • [mix compile] The :xref compiler is deprecated and it has no effect. Please remove it from your mix.exs file.

Checksums

  • Precompiled.zip SHA1: e6db54961faac708946fa218f2c771dadd9f8bd9
  • Precompiled.zip SHA512: d9741ba49fe3a45d9418b20ac329d03d96bb1ad32cbb2dca919b424c0051d9edd504490e54e697cacec62750b886f5e9767cd8c17f67134c80eddebdffb3132c
  • Docs.zip SHA1: 9dea071f1a530e3cd14c1da5b8408c6008c2c035
  • Docs.zip SHA512: fd5cf68d8730b3c531b24d84ad6608d1cbef7b070eb0f187c5daa4f9f3d00d85534c9d4f8bd736207d4c7c63214f9243e51d2fe14fece2f93b2061fa442a16d8

Have fun!

49 Likes
2

So with 1..n//1 we get an empty range?

1 Like
3

So many nice things :+1:

Where can I check these 2?

1 Like
4

Nope. That’ s equivalent to 1..n

FRom the docs:

Definition

An increasing range first..last//step is a range from first to last increasing by step where step must be a positive integer and all values v must be first <= v and v <= last. Therefore, a range 10..0//1 is an empty range because there is no value v that is 10 <= v and v <= 0.

Similarly, a decreasing range first..last//step is a range from first to last decreasing by step where step must be a negative integer and values v must be first >= v and v >= last. Therefore, a range 0..10//-1 is an empty range because there is no value v that is 0 >= v and v >= 10.

5

Well, what you have cited, it does solve the problem, as the example has 10..0//1 as an example for the empty range, so 1..n//1 has to be empty as well for any n less than 1.

7

Sorry, for your example I was thinking n to be >= 1

8

Correct.

2 Likes
9

Love the addition of tap and then to streamline pipelines.

18 Likes
10

v1.12 91344a24a8610c87acc34a471bd53d0d9f48a40b 2021-04-01T20:48:58Z a2b3a1cd90b4eb300abb17315556ac27964b9e183003ecd2f016c2aca360afeb
v1.12-otp-21 91344a24a8610c87acc34a471bd53d0d9f48a40b 2021-04-01T20:48:53Z a2b3a1cd90b4eb300abb17315556ac27964b9e183003ecd2f016c2aca360afeb
v1.12-otp-22 91344a24a8610c87acc34a471bd53d0d9f48a40b 2021-04-01T21:02:50Z f61999360118bc070cf7c762402a1148dd3a907f1f8a0002c6aa68796e498844
v1.12-otp-23 91344a24a8610c87acc34a471bd53d0d9f48a40b 2021-04-01T21:01:38Z fccabc4a893558368fc0dfac10a45506d040d459a16fcb16ebea4888d4050997

Source: https://repo.hex.pm/builds/elixir/builds.txt

Well … v1.12 without rc - looks like a 1st April joke. :smiling_imp:

Also Bob got into it and have builds from future:

obraz
obraz1680×1050 76.8 KB

but seriously … after what time those lists gets updated?

11

The builds.txt does get updated each time a build by hex‘s bob finishes. The name afaik is directly pulled from the branch/tag names on github. So maybe this was an april fools joke or just mistake. Reverting it afaik would need manual intervention here. Not sure why the proper rc version would be missing still.

1 Like
12

v1.12 is the branch. The released version is v1.12.0-rc.0.

1 Like
13

The tag for the release was created 6 hours ago, but the datetime in bob‘s builds.txt points to evening of april 1st, so it seems somethings off here for those builds. Also the 1.11. rcs are correctly labeled in the file, while for 1.12 there should be a 1.12.0-rc.0 as part of the 1.12 segment of versions.

1 Like
14

There was a hiccup when building docs and releases but it should be fine now (thanks @ericmj for fixing it!)

3 Likes
15

Speaking of docs, you can also check out the rendered v1.12 changelog on hexdocs.pm where all the mentions of modules and functions will have nice links :slight_smile:

4 Likes
17

Here’s a note of thanks for Mix.install, which I’ve been testing the past few weeks. The Mix.install method of managing script dependencies is IMO superior to NPM or RubyGems. It’s a small feature but can greatly expand the utility of Elixir - bravo!

18 Likes
18

:tada: tap and then are superb! Thanks a lot.

I love how the documentation on then reads

This is most commonly used in pipelines, allowing you to pipe a value to a function outside of its first argument.

… as if people were already commonly using it anywhere :slight_smile:

6 Likes
19

Do you really think that literally nobody were having a fun using the Elixir’s master branch? :smiling_imp:

5 Likes
20

I wonder why is tap a macro ?

  defmacro tap(value, fun) do
    quote bind_quoted: [fun: fun, value: value] do
      fun.(value)
      value
    end
  end

I would believe that the following would work too. (Well I know it works since like many I have a bunch of those in different projects).

def tap(value, fun) do
  fun.(value)
  value
end

Edit: oh, I guess it is an optimization to remove a useless call, since a pipeline AST will be unflattened?

1 Like
21

Here is the PR which also has a link to the mailing list discussion:

5 Likes