Activity
From 08/07/2022 to 08/13/2022
08/13/2022
-
03:31 AM Bug #18435: Calling `protected` on ancestor method changes result of `instance_methods(false)`
- nagachika (Tomoyuki Chikanaga) wrote in #note-14:
> I agree to backport the revert to ruby_3_1, but I think removing the existing methods in stable releases could be a severe damage to the depending applications/libraries.
> ...
I don... -
03:27 AM Bug #18435: Calling `protected` on ancestor method changes result of `instance_methods(false)`
- I agree to backport the revert to ruby_3_1, but I think removing the existing methods in stable releases could be a severe damage to the depending applications/libraries.
I'd like to add dummy methods {Method,UnboundMethod}#{public?,pr... -
02:24 AM Feature #18809 (Closed): Add Numeric#ceildiv
- Applied in changeset commit:git|0617cba197cdff626ee9c74cece480df31d384ef.
----------
[DOC] Add the link to [Feature #18809] -
-
12:29 AM Revision d1d1c9ba (git): Add a NEWS entry about Integer#ceildiv [ci skip]
- 12:00 AM Revision f9382a3b (git): * 2022-08-13 [ci skip]
-
12:00 AM Revision fa65800e (git): [ruby/rdoc] [DOC] Enhances text about escapes (https://github.com/ruby/rdoc/pull/917)
- https://github.com/ruby/rdoc/commit/c40bac829c
08/12/2022
-
10:26 PM Bug #18963 (Closed): Ripper.tokenize(code).join != code when heredoc and multiline string is on the same line
- Expect Ripper.tokenize(code).join to match the original code but it doesn't.
~~~ruby
code = %[
<<A + "hello
A
world"
]
eval code # => "ruby\nhello\nworld"
code == Ripper.tokenize(code).join # => false
eval Ripper.tokenize(code... -
05:24 PM Misc #18954: DevMeeting-2022-08-18
- * [Feature #18408] Allow pattern match to set non-local variables
* That would make, in particular, `42 => @a` possible
* This addition seems natural after we added non-local vars support for pinning
* It also "restores" the... -
-
-
-
07:59 AM Revision 77fdb3a4 (git): Introduce with_warn_vsprintf macro
-
- Applied in changeset commit:git|844a0edbae6e74293e3d0cb1ceeeb66a4371d06d.
----------
[Bug #18962] Do not read again once reached EOF
`Ripper::Lexer#parse` re-parses the source code with syntax errors
when `raise_errors: false`.
Co-Aut... -
-
- `Ripper::Lexer#parse` re-parses the source code with syntax errors
when `raise_errors: false`.
Co-Authored-By: tompng <tomoyapenguin@gmail.com> -
- This patch is suggested by nobu.
Benchmark result:
```
require 'benchmark'
n = 10 ** 7
Benchmark.bm do |x|
x.report("Fixnum/Fixnum") { a, b = 5, 2; n.times { a.ceildiv(b) } }
x.report("Bignum/Bignum") { a, b = 10**100, 10**99 - 1... -
-
-
- As commented in include/ruby/internal/abi.h, since teeny versions of
Ruby should guarantee ABI compatibility, `RUBY_ABI_VERSION` has no role
in released versions of Ruby. -
-
-
-
- `RbInstall::GemInstaller#build_extensions` has nothing to do.
-
05:04 AM Revision d9f1b8ba (git): [ruby/error_highlight] Add a note about the current limitation of ErrorHighlight.spot
- https://github.com/ruby/error_highlight/commit/489ce80a62
08/11/2022
-
- https://github.com/ruby/rdoc/commit/5506d4d67e
-
- https://github.com/ruby/rdoc/commit/d00ddfe57c
-
03:48 PM Bug #16842 (Closed): `inspect` prints the UTF-8 character U+0085 (NEXT LINE) verbatim even though it is not printable
- Applied in changeset commit:git|49517b3bb436456407e0ee099c7442f3ab5ac53d.
----------
Fix inspect for unicode codepoint 0x85
This is an inelegant hack, by manually checking for this specific
code point in rb_str_inspect. Some testing i... - 03:47 PM Revision a677aa0f (git): * 2022-08-12 [ci skip]
-
- This is an inelegant hack, by manually checking for this specific
code point in rb_str_inspect. Some testing indicates that this is
the only code point affected.
It's possible a better fix would be inside of lower-level encoding
code, ... -
- https://github.com/ruby/rdoc/commit/7e6ef6c855
-
- Ripper.lex does not end with \0, ^D and ^Z, and it contains weird tokens including "\u0000"
~~~ruby
# for syntax ok code
Ripper.lex("hello\0\0world").map{_3} #=> ["hello"]
Ripper.lex("hello\0world").map{_3} #=> ["hello", "\u0000wor... -
- Applied in changeset commit:git|cfb9624460a295e4e1723301486d89058c228e07.
----------
Fix Array#[] with ArithmeticSequence with negative steps (#5739)
* Fix Array#[] with ArithmeticSequence with negative steps
Previously, Array#[] when... -
- * Fix Array#[] with ArithmeticSequence with negative steps
Previously, Array#[] when called with an ArithmeticSequence
with a negative step did not handle all cases correctly,
especially cases involving infinite ranges, inverted ranges,... -
- Clean built directories by `make distclean`, and then clean leftover
makefiles for skipped extensions. -
-
- exts.mk files are one level under the top of extension directories.
-
- Are there any blocking issues?
If exist, I will work to resolve them. -
- Since `RUBY_DEVEL` in cppflags has no effect in the configure script
and makefiles.
08/10/2022
-
09:37 PM Feature #18949: Deprecate and remove replicate and dummy encodings
- Thank you for the response.
It's true synchronization is required for a number of read operations, but I think the most important for performance, `get_encoding`, could become very fast with a max limit. But I digress.
More importantly... -
04:40 PM Feature #18949: Deprecate and remove replicate and dummy encodings
- Dan0042 (Daniel DeLorme) wrote in #note-26:
> Can you clarify this please? With a max limit you need synchronization only when adding encodings. I hope you'll agree such a small level of complexity is not an issue. Apart from that, the t... -
- Eregon (Benoit Daloze) wrote in #note-25:
> But a max limit is not nearly as helpful as a fixed number of encodings in terms of complexity and overhead.
Can you clarify this please? With a max limit you need synchronization only when... -
- duerst (Martin Dürst) wrote in #note-24:
> @Eregon: If you want, you can add this issue to the developers' meeting next week. I will be in Switzerland, but I should be able to participate. Maybe you can also participate?
OK, I will a... -
04:23 AM Feature #18949: Deprecate and remove replicate and dummy encodings
- @Eregon: If you want, you can add this issue to the developers' meeting next week. I will be in Switzerland, but I should be able to participate. Maybe you can also participate?
Eregon (Benoit Daloze) wrote in #note-22:
> duerst (Mar... -
- https://github.com/ruby/rdoc/commit/dc88f1b425
-
- matz (Yukihiro Matsumoto) wrote in #note-11:
> I thought it was OK to accept this behavior, but it caused issues like #18729 and #18751. At the time of the decision, I haven't noticed those corner cases. Although it has already been shi... -
- This reverts commit 27278150685e738f84105d09843d3ba371146c7a and
58dc8bf8f15df9a33d191074e8a5d4946a3d59d5.
Visibility is an attribute of the method entry in a class, not an
attribute of the Method object.
Fixes [#18729]
Fixes [#18751]
... -
- Applied in changeset commit:git|bfa6a8ddc84fffe0aef5a0f91b417167e124dbbf.
----------
Only allow procs created by Symbol#to_proc to call public methods
Fixes [Bug #18826]
Co-authored-by: Nobuyoshi Nakada <nobu@ruby-lang.org> -
- Fixes [Bug #18826]
Co-authored-by: Nobuyoshi Nakada <nobu@ruby-lang.org> - 06:41 PM Revision d115a060 (git): * 2022-08-11 [ci skip]
-
-
06:24 PM Feature #18885: End of boot advisory API for RubyVM
- > Is there any point to precompute inline caches if there is no fork?
Yes, the first "request" (or whatever your unit of work is) won't have to do it. So you are moving some work to boot time, instead of user input processing time.
... -
- I think the terminology used here might cause some confusion in the discussion.
"End of boot" makes it sound like this API would be useful for non-forking servers once they have finished their "boot" sequence. But from what I understa... -
- byroot (Jean Boussier) wrote in #note-13:
> I also think `not` would be a better name for it
Since there is already a `not` operator, I would expect a `not` method to have the same semantics and return either true or false. So since we'... -
- Monkeypatching nil.to_hash seems a veeeery bad idea. For example
```ruby
h = {"y"=>"!"}
"xyz".sub(/y/, h) #=> "x!z"
h = nil
"xyz".sub(/y/, h) #=> TypeError (no implicit conversion of nil into String)
def nil.to_hash; {}; end
... -
04:46 PM Feature #18961 (Open): Introduce support for pattern matching all elements of an array
- When pattern matching arrays, often times we want to assert that a given array's elements all match a pattern.
For example:
```ruby
class Robot; end
class User; end
case [User.new, User.new]
in [User, *] => u if u.all? { _1 i... -
02:01 PM Revision a661aac9 (git): Bundle unreleased RBS (#6228)
- 12:52 PM Revision 2ebb428f (git): Update default gems list at 1b32a4c7bb2e1ceb456b191ca88289 [ci skip]
-
- https://github.com/ruby/error_highlight/commit/6edf0a0a5d
-
- We had to keep backtrace_location before opts is overwritten.
https://github.com/ruby/error_highlight/commit/2735e4681a -
- * [Feature #18949] Deprecate and remove replicate and dummy encodings (eregon)
* I will try to attend the dev meeting. Would it be possible to discuss this towards the end of the meeting, 16 JST (= 9am CEST) my preference, or if bette... -
- I still think we should stop hiding ZSUPER method entries, it causes way more confusion than it helps and this bug, and I believe it would cause very little incompatibility.
@jeremyevans0 or @mame Would you be interested to work on that... -
- ... and move things from core_ext.rb to base.rb.
This will confine CRuby-dependent things to ErrorHighlight.spot.
https://github.com/ruby/error_highlight/commit/22d1dd7824 - 07:05 AM Revision 1139bc8c (git): Update bundled gems list at 2022-08-10
-
-
- There's no point in making a copy of an array just to expand it. Saves
an unnecessary array allocation in the multiple assignment case, with
a 35-84% improvement in affected cases in benchmark/masgn.yml. -
- This optimizes unbalanced multiple assignment cases such as:
```ruby
a.b, c.d = e, f, g
a.b, c.d, e.f = g, h
```
Previously, this would use:
```
newarray(3)
expandarray(2, 0)
newarray(2)
expandarray(3, 0)
```
These would both alloca... -
- This allows them to show the effect of the previous newarray/expandarray
to swap/opt_reverse optimization. This shows an 35-83% performance
improvement in the four multiple assignment benchmarks that use this
optimization. -
- This renames the reverse instruction to opt_reverse, since now it
is only added by the optimizer. Then it uses as a more general
form of swap. This optimizes multiple assignment in the popped
case with more than two elements. -
- This reverts commit 5512353d97250e85c13bf10b9b32e750478cf474.
-
- An optimization for multiple assignment in the popped case to avoid
array allocation was lost in my fix to make multiple assignment follow
left-to-right evaluation (50c54d40a81bb2a4794a6be5f1861152900b4fed).
Before, in the two element c... -
- Applied in changeset commit:git|357352af5ee37bf31ba4468ee242ea5a10158461.
----------
Do not enable RUBY_DEVEL by RUBY_PATCHLEVEL
This makes RUBY_DEVEL not enabled automatically. It still can be
enabled manually.
Test manually using R... - 05:13 AM Revision 3569d130 (git): * 2022-08-10 [ci skip]
-
- This makes RUBY_DEVEL not enabled automatically. It still can be
enabled manually.
Test manually using RUBY_DEVEL in CI.
Implements [Feature #17468]
08/09/2022
-
- * [Bug #18955] Kernel#sprintf - %c ignores a non-ASCII character's encoding (jeremyevans0)
* Do we want the %c format specifier to transcode a string argument to the format string's encoding before taking the first character?
* [Bug #1... -
-
-
-
01:44 PM Bug #18960: Module#using raises RuntimeError when called at toplevel from wrapped script
- Pull Request: https://github.com/ruby/ruby/pull/6226
-
- I noticed that this file works when loaded with `load`, but fails if you pass `true` (or a module) as the `wrap` argument.
```ruby
# using.rb
using Module.new
```
This works:
```ruby
load "./using.rb"
```
This doesn't:
... -
- duerst (Martin Dürst) wrote in #note-21:
> It references RFC 2781 (see https://www.rfc-editor.org/rfc/rfc2781). That provides quite a bit of details.
Thanks for the link.
So that says strings in "UTF-16" (no explicit endian) SHOULD ... -
- duerst (Martin Dürst) wrote in #note-20:
> I think it should be okay to in your implementation assume a maximum number of encodings (currently, the number of actual encodings in CRuby is a bit more than 100, so a max of 127 or whatever ... -
- Eregon (Benoit Daloze) wrote in #note-19:
> duerst (Martin Dürst) wrote in #note-15:
> ...
Having a special case that is used extremely rarely is prone to hidden bugs, and is therefore bad engineering.
It's unclear whether you wan... -
- Eregon (Benoit Daloze) wrote in #note-18:
> Replicate (`Encoding#replicate`) and extra dummy encodings (`rb_define_dummy_encoding`) cause a variable number of encodings, which means extra synchronization, hashtable lookups, indirectio... -
- Poisoned regions cannot be accessed without unpoisoning outside gc.c.
Specifically, debug.gem is terminated by AddressSanitizer.
```
SUMMARY: AddressSanitizer: use-after-poison iseq_collector.c:39 in iseq_i
``` -
03:05 AM Revision 44264b4f (git): Merge rubygems/bundler HEAD.
- Pick from https://github.com/rubygems/rubygems/commit/dfbb5a38114640e0d8d616861607f3de73ee0199
-
02:02 AM Revision f8936b33 (git): Remove documentation about dev
- dev is Shopify's internal tool that doesn't work if you use Intel
Homebrew on M1 (or rbenv, btw). Now that we maintain this outside
Shopify's repository, we should stop talking about it here.
08/08/2022
- 07:35 PM Revision 35b9cd1d (git): [rubygems/rubygems] Bump rb-sys in /test/rubygems/test_gem_ext_cargo_builder/custom_name
- Bumps [rb-sys](https://github.com/oxidize-rb/rb-sys) from 0.9.28 to 0.9.29.
- [Release notes](https://github.com/oxidize-rb/rb-sys/releases)
- [Commits](https://github.com/oxidize-rb/rb-sys/compare/v0.9.28...v0.9.29)
---
updated-depende... -
07:02 PM Feature #18959: Handle gracefully nil kwargs eg. **nil
- Oh wow, I didn't know that 😲 `def nil.to_hash = {}` works perfectly well! Many thanks! 🙏
-
- This was also rejected in #9291. However, since this is a feature request with a use case, I guess it can be considered again.
Note that you can opt in to the behavior you want with one line (`def nil.to_hash = {}`), so this behavior ... -
- The issue:
```ruby
def qwe(a: 1) end
qwe(**nil) #=> fails with `no implicit conversion of nil into Hash (TypeError)` error
{ a:1, **nil } #=> fails with `no implicit conversion of nil into Hash (TypeError)` error
```
Reaso... -
- 04:50 PM Revision 78bc2aae (git): [rubygems/rubygems] Bump rb-sys
- Bumps [rb-sys](https://github.com/oxidize-rb/rb-sys) from 0.9.28 to 0.9.29.
- [Release notes](https://github.com/oxidize-rb/rb-sys/releases)
- [Commits](https://github.com/oxidize-rb/rb-sys/compare/v0.9.28...v0.9.29)
---
updated-depende... - 04:33 PM Revision 9c13a6ce (git): [ruby/psych] Raise specific error when an anchor isn't defined
- https://github.com/ruby/psych/commit/98fbd5247a
- 04:33 PM Revision 4b9cdf4e (git): [ruby/psych] Update to squiggly heredocs in the file
- https://github.com/ruby/psych/commit/42b43de997
- 04:33 PM Revision 38ea6b30 (git): [ruby/psych] Add test for missing anchor
- https://github.com/ruby/psych/commit/5f08137ae6
- 04:33 PM Revision ea1efdf3 (git): [ruby/psych] Add test for anchor reuse
- The spec calls this a "reuse" of an anchor
https://yaml.org/spec/1.2.2/#71-alias-nodes
https://github.com/ruby/psych/commit/57e3b70a56 - 04:32 PM Revision 54219ae8 (git): [ruby/psych] Raise specific error when aliases are not enabled
- https://github.com/ruby/psych/commit/0c11ddcf46
- 04:31 PM Revision 71f89c28 (git): [ruby/psych] Don't hardcode expected alias names
- https://github.com/ruby/psych/commit/b9ab19094f
- 04:31 PM Revision 0b7cfdca (git): [ruby/psych] Test that recursive refs dump as aliases
- https://github.com/ruby/psych/commit/d9f7289190
- 04:31 PM Revision c851bced (git): [ruby/psych] Clarify tests about parsing aliases
- https://github.com/ruby/psych/commit/0bc30cb4cb
- 03:00 PM Revision 1602443c (git): * 2022-08-09 [ci skip]
- 02:59 PM Revision dcf0bb30 (git): Update default gems list at 5c9ce5475736756891238a3e3fe581 [ci skip]
-
- https://github.com/ruby/date/commit/dff37b3dd1
-
- Time is always in the proleptic Gregorian calendar.
Also DateTime#to_time should convert to the Gregorian calendar first,
before extracting its components.
https://bugs.ruby-lang.org/issues/18946#change-98527
https://github.com/ruby/da... -
- Applied in changeset commit:git|e0dfa5967e7063da8b65dc3c062ef4652e246e34.
----------
[Bug #18946] Use Gregorian dates to test -
03:02 AM Bug #18946: Time#to_date returns incorrect date
- I checked commit history and found out that to_date has been use GREGORIAN calendar since 2011-05-31 and to_datetime preserved the old DEFAULT_SG (ITALY). I assume this is a mistake and both should use GREGORIAN (ext/date/date_core.c:864...
-
-
-
01:26 PM Revision a24c607e (git): [DOC] Fix formatting issue in Enumerable
-
- - The method was renamed from `get` to `get_value`
- Comparing to `String#unpack` isn't quite equivalent, `unpack1` is closer.
- Use frozen_string_literal to avoid allocating a format string every time.
- Use `N` format which is equivale... -
- https://github.com/ruby/date/commit/0ae93e26aa
-
- duerst (Martin Dürst) wrote in #note-15:
> Yes, exactly. I already said so in https://bugs.ruby-lang.org/issues/18949#note-8.
Right, I wasn't sure this was what you meant there, thank you for confirming.
> ...
Only for extremely-r... -
- nobu (Nobuyoshi Nakada) wrote in #note-17:
> Eregon (Benoit Daloze) wrote:
> ...
It very much is, as explained in the description (https://bugs.ruby-lang.org/issues/18949#Summary).
Replicate (`Encoding#replicate`) and extra dummy enco... -
- Eregon (Benoit Daloze) wrote:
> Replicate and dummy encodings seem to clearly fall into this category, almost nobody uses them but they add a significant complexity and also add a significant performance overhead.
> ...
These are not r... -
- Eregon (Benoit Daloze) wrote in #note-14:
> Could it be replaced by the BINARY encoding in programs using it?
Yes, it could, and was exactly the way in 1.8 or earlier.
I don't want it again. -
- I also think `not` would be a better name for it:
```ruby
class Object
def not
self unless yield self
end
end
```
Something that comes up frequently if to check wether an argument is either nil or empty (e.g. https://... -
07:36 AM Feature #18822: Ruby lack a proper method to percent-encode strings for URIs (RFC 3986)
- +1, for context a similar issue we recently hit:
https://github.com/sporkmonger/addressable/issues/472
Turns out tons of projects these days rely on addressable for a more complete API, it would be nice only to need to lean on it in e... -
- - [Feature #18944] Add `SizedQueue#push(timeout:)` (byroot)
- Matz said this should be required separately.
- When implementing `Queue#pop(timeout:)` it was very evident this was needed.
- Spec is return `self` on success, `nil`... -
08/07/2022
- 04:13 PM Revision 40f2078d (git): * 2022-08-08 [ci skip]
-
- https://github.com/ruby/ruby/commit/e4e054e3ce40 used four footnotes
without blank lines. And the ChangeLog generated from that commit
resulted in ``undefined method `parts' for nil`` error.
For now, let a footnote terminated by the ne... -
- https://github.com/ruby/rdoc/commit/4e44c9c6cf
-
-
- Since the change at f310ac1cb2964f635f582862763b2155aacf2c12 to show
the backtraces by default, this test started to show the backtraces.
As the backtraces are not the subject of this test, silence them by
using Gem::SilentUI. -
- Collected `files` lacked `lib` prefix.
-
- When LOAD_RELATIVE, as `sopath` is truncated to the prefix path, make
the duplicate before it. Also make `rb_libruby_selfpath` frozen and
hidden. -
-
- Eregon (Benoit Daloze) wrote in #note-14:
> znz (Kazuhiro NISHIYAMA) wrote in #note-10:
> ...
Yes, exactly. I already said so in https://bugs.ruby-lang.org/issues/18949#note-8.
> Could it be replaced by the BINARY encoding in progra...