Activity
From 04/08/2025 to 04/14/2025
04/14/2025
-
10:57 PM Revision 0606046c (git): Lazily create `objspace->id_to_obj_tbl`
- This inverse table is only useful if `ObjectSpace._id2ref` is used,
which is extremely rare. The only notable excepti... -
06:42 PM Bug #20919: IO#seek and IO#pos= do not clear the character buffer in some cases while transcoding
- Folks, could I ask for a review (and potential merge) on the fix of this issue https://github.com/ruby/ruby/pull/12714?
-
03:39 PM Revision b68172ca (git): Rename `f_optarg` to `f_opt_arg` for consistency in parser rules
-
02:30 PM Misc #21154: Document or change Module#autoload?
- Hi @mame!
> I'm not exactly sure what you meant by "const_get hook", but perhaps you meant "const_added"? I'll assum... -
-
09:28 AM
Feature #20877 (Closed): Introduce (public) debug assertion for holding the GVL.
- Applied in changeset commit:git|c13ac4d615c28edc1580363c13b44eb7a73103f3.
----------
Assert the GVL is held when per... - 09:28 AM Revision c13ac4d6 (git): Assert the GVL is held when performing various `rb_` functions.
- [Feature #20877]
- 09:28 AM Revision 4e970c5d (git): Expose `ruby_thread_has_gvl_p`.
- 09:28 AM Revision 22667fcc (git): Add `RUBY_VERSION_IS_3_5`.
-
08:46 AM Feature #21121 (Closed): Ractor channels
- Superseded by #21266
-
07:33 AM Revision d0b7e5b6 (git): v3.4.3
-
- Reverted the backport since backporting the commit failed CI, e.g. https://github.com/ruby/ruby/actions/runs/14439756...
-
- ruby_3_4 commit:e630a0f7ae909dafe2e1dbc47baf90cd795b4a5f merged revision(s) commit:052794bfe1970e90f4f4f9e37fc362dd27...
- 07:31 AM Revision bef59ede (git): Update default gems list at ad8cb532d52692fe72dea68a0aeaf4 [ci skip]
-
- This reverts commit e630a0f7ae909dafe2e1dbc47baf90cd795b4a5f.
-
- ruby_3_4 commit:7e093fb43a.
-
- ruby_3_4 commit:1b6f9cd6ea5715ad33d8ba4c78382aa8aafdbcd4 merged revision(s) commit:0d6263bd416338a339651fb97fe4d62701...
-
- Fix coverage measurement for negative line numbers
Fixes [Bug #21220]
Co-Authored-By: Mike Bourgeou... -
- We've discussed whether we backport this to Ruby 3.4 or not with @peterzhu2118. Since he's too worried about potentia...
-
- ruby_3_4 commit:9abd48df705c37f69ec9fd954f84cbbbd65db624 merged revision(s) commit:3a7b9ca93b91dcc086b9ac8b9957e59268...
-
- Fix `Integer.sqrt` to never exceed actual value
`Integer.sqrt` uses `sqrt(3)` from libm for small values.
... -
- ruby_3_4 commit:e3fc29a9b7.
-
- ruby_3_4 commit:7b5e4125b9935cd78d80719de018dad446780e40 merged revision(s) commit:f27ed98effff9891d28edc7119d45811bc...
-
- [ruby/prism] Freeze `Prism::VERSION`
Closes https://github.com/ruby/prism/pull/3422
https://github.... - 07:18 AM Revision ad8cb532 (git): [ruby/strscan] Bump version
- https://github.com/ruby/strscan/commit/7b1eb1e4ed
- 07:18 AM Revision 0db87b89 (git): [ruby/strscan] Allow parsing strings larger than 2GiB
- (https://github.com/ruby/strscan/pull/147)
For a reason unknown, even though `pos` is stored as a `long`, the
`#pos`... -
- [ruby/prism] Accept a newline after the defined? keyword
[Bug #21197]
https://github.com/ruby/prism... -
07:17 AM Revision 38024772 (git): [rubygems/rubygems] Refine `bundle update --verbose` logs
- Don't mention "Found changes from the lockfile" because that's not
really true in general.
https://github.com/rubyge... -
- If both a native and a generic version are locked, but the native
version is incompatible with the running Ruby, Bund... -
- `Definition#specs` does the same thing but memoizes the result.
https://github.com/rubygems/rubygems/commit/b62bf9fe41 -
- https://github.com/rubygems/rubygems/commit/458fa5dc4c
- 02:21 AM Revision 8ad5dd79 (git): Update bundled gems list at f65aad61f7c2b5ff15d10be9264d3b [ci skip]
-
02:20 AM Revision f65aad61 (git): Update bundled_gems
-
02:06 AM Bug #21207 (Closed): CSV gem not installing in head
- Thanks.
I close this. -
01:48 AM Bug #21207: CSV gem not installing in head
- @kou,
Thanks. Re ruby-loco, I removed the patch I had Sunday AM US and CSV is now included.
I just checked the ... -
- Could you try master? It includes https://github.com/ruby/ruby/commit/cfa74dda4547c7ef4d3510b729bad8cc6299e523 .
04/13/2025
-
10:03 PM Revision 311167ee (git): [ruby/etc] Suppress unused function warning
- https://github.com/ruby/etc/commit/410c8dd4ae
-
- https://github.com/ruby/etc/commit/15c2696794
-
-
-
-
01:29 PM Bug #21256: `it` value is wrapped in an array when that doesn't happen with numbered parameter
- Thanks for that info, it allowed me to make a potential fix: https://github.com/ruby/ruby/pull/13108
-
01:20 PM Bug #21266 (Assigned): YJIT GC safety crash with proc objects as block argument
- ```ruby
# Run with --yjit-call-threshold=1
def foo(args) = bar(*args, &proc { _1 })
def bar(_, _, _, _, *rest) = y... - 11:46 AM Revision 3700ae2e (git): Fix typos `finised` -> `finished` (#13104)
-
-
- Can we close this because https://github.com/ruby/ruby/pull/9815 was merged?
-
- 07:00 AM Revision cfa74dda (git): Update bundled gems list as of 2025-04-13
-
02:44 AM Revision 29b26fd3 (git): Fix macro for disabled match cache
- The `MEMOIZE_LOOKAROUND_MATCH_CACHE_POINT` macro needs an argument
otherwise we end up with:
```
../regexec.c:3955:2...
04/12/2025
-
05:43 PM Revision 543dd77c (git): [ruby/prism] Fix parsing rescued exception via indexed assignment
- Given this code
```ruby
begin
raise '42'
rescue => A[]
end
```
Prism fails with this backtrace
```
Error: test_u... -
05:39 PM Feature #21262: Proposal: `Ractor::Port`
- ko1 (Koichi Sasada) wrote in #note-8:
> The important point is that we have limited "blocking" synchronization (intr... -
01:13 AM Feature #21262: Proposal: `Ractor::Port`
- The important point is that we have limited "blocking" synchronization (introduced by cross-Ractor communication) to ...
-
- Now I understand that `find_work` takes a time and if no consumers are ready, we can't run `find_work` if there is no...
-
- Change the indent to match the characters used in other generation rules. This makes it easier to find lhs.
04/11/2025
-
11:59 PM Feature #21264: Extract Date library from Ruby repo in the future
- If we are going to remove stdlib `Date`, I would also prefer that `Date` be a core class instead of demoted to bundle...
-
03:51 PM Feature #21264: Extract Date library from Ruby repo in the future
- Is it possible to, on the contrary, make Date a core class? Rails, and many other non-Rails codebases I saw, implicit...
-
- Note: This is not for Ruby 3.5.
`Date` and `DateTime` has no primary maintainer in 10+ years. I would like to depr... -
- ko1 (Koichi Sasada) wrote in #note-5:
> > The problem with the above code is that the consumer Ractor will starve un... -
- > The problem with the above code is that the consumer Ractor will starve until find_work returns.
I think it is s... -
- I like this idea, and I think we should do it, but I think we still need a channel or a queue for the producer / cons...
-
- It looks like stat_insn_usage was introduced with YARV, but as far as I
can tell the field has never been used. I th... -
03:53 PM Feature #21258: Retire CGI library from Ruby 3.5
- `CGI.escape_uri_component` and `URI.encode_uri_component` are almost the same except `*` and `~`.
`CGI.escape` and `... -
03:11 PM Feature #21221: Proposal to upstream ZJIT
- > For context YJIT is essentially a baseline JIT given it does little inlining and as you say only one optimization p...
-
10:16 AM Feature #21221: Proposal to upstream ZJIT
- Interesting to see work starting on a CRuby optimizing method-based JIT.
For context YJIT is essentially a baseline ... -
- ```ruby
class RefinedScope
using(Module.new { refine(Kernel) { def itself = 0 } })
ITSELF = :itself.to_proc
e... -
- I'd like to go through the discussion in the order written in this HackMD (with Matz's approval).
https://hackmd.i... -
-
-
02:30 AM Feature #21263 (Open): eval-after-require hook
- I want to register hook after require.
But I have no idea how to write it.
Expected use cases:
1. Load related... -
-
-
-
04/10/2025
-
10:23 PM Feature #18035: Introduce general model/semantic for immutability.
- I found a good example of why the proposed interface would be a good idea:
https://github.com/ruby/resolv/pull/62
T... -
06:33 PM Revision f13e86a7 (git): Allow gdb's rp to work on classes offline
- It's useful to be able to get information about classes without a
running process, ie. when debugging a coredump. -
04:02 PM Misc #21154: Document or change Module#autoload?
- Aha, when test.rb is required, currently `autoload` does (almost) nothing because test.rb is already being required, ...
-
- @fxn Sorry for the delay. I finally watched your video.
I'm not exactly sure what you meant by "const_get hook", b... -
03:58 PM Revision e3dd766e (git): [ruby/resolv] refactoring class-hash to be ractor-safe
- mutable constants can't be shared across ractors; this changes that design to define the required variables as consta...
-
- https://github.com/ruby/resolv/commit/afb57f40a1
-
- * [Feature #21262] Proposal: `Ractor::Port` (ko1)
* Considering with [`Channel`](https://bugs.ruby-lang.org/issues... -
- * [Feature #21219] `Object#inspect` accept a list of instance variables to display (byroot)
* Redefining `#inspect... -
- # Proposal: `Ractor::Port`
In concurrent Ruby applications using Ractors, safely and efficiently communicating res... -
-
- if irb and that dependency are broken, it caused infinite loop with force_activate("irb").
-
- https://github.com/nobu/ruby/tree/inspect_instance_variables
-
- To show that mingw `sh` expects forward slashes as path separators, not
backslashes, configure in another directory t... -
-
-
-
-
- I see. Alignment with File makes sense. At least my expectation was that, by dup'ing the IO object, I'd receive a sep...
-
-
- This is how `IO#dup` works. What would be the benefit of `StringIO#dup` behaving differently?
```ruby
f = File.op... -
- Originally, if a class was defined with the class keyword, the cref had a
const_added callback, and the superclass an... -
-
- @kou +1, @tompng suggest `URI.escape_query_param`. I prefer these approachs.
@jeremyevans0 I understood your conce...
04/09/2025
-
- tenderlovemaking (Aaron Patterson) wrote in #note-7:
> I made a patch for it [here](https://github.com/ruby/ruby/com... -
- jez (Jake Zimmerman) wrote in #note-6:
> I'm curious: could we add a second check after [this check](https://github.... -
08:57 PM Feature #21254: Inlining Class#new
- @tenderlovemaking Question about an extension to the current implementation.
We have a fair amount of code that lo... -
- 11:02 PM Revision 7e093fb4 (git): Fix crash in TCPSocket.open
- Fix segfault crash observable with TCPSocket.open(nil, nil)
-
- Backport PR: https://github.com/ruby/ruby/pull/13091
-
- https://github.com/ruby/ruby/pull/12934
https://github.com/ruby/ruby/commit/b148dfef5aa208cafa28626fd84c9762bfdfe0ac -
- Applied in changeset commit:git|0d6263bd416338a339651fb97fe4d62701704c4b.
----------
Fix coverage measurement for ne... -
- I created a PR.
https://github.com/ruby/ruby/pull/13089
The PR should prevent `RUBY_EVENT_COVERAGE_LINE` from f... -
03:48 AM Bug #21220: Memory corruption in update_line_coverage() [write at index -1]
- mame (Yusuke Endoh) wrote in #note-8:
> @mbcodeandsound Just FYI, I bet you meant to write `!(1..16).cover?(channel)`... -
- Fixes [Bug #21220]
Co-Authored-By: Mike Bourgeous <[email protected]>
Co-Authored-By: Jean Boussier <jean.bouss... -
- I am in favor of retiring cgi and keeping cgi/escape feature. Of the two options, I prefer option 1 (keep only cgi/es...
-
- `URI.encode` (not `escape`)?
-
08:57 AM Feature #21258: Retire CGI library from Ruby 3.5
- I like the idea of deprecating `CGI` and moving the `escape`/`unescape` methods!
JavaScript calls the features `"h... -
- I would like to retire CGI library from Ruby 3.5.0 release. It means CGI is not promoted bundled gems. The users need...
-
- I've found that, if I `.dup` a stringio, when I read one of the objects, the cursor also moves internally on the othe...
-
- test.rb
```
TracePoint.new(:line) do |tp|
p tp
end.enable
load "./target.rb"
```
target.rb
```
if true... -
- - Compare with the root of the target file
- Fix root path of UNC path -
- It moved in https://github.com/ruby/ruby/pull/13012 but wasn't updated here
-
09:23 AM Revision b200bad6 (git): bump teeny
-
- ruby_3_3 commit:4ca521e91342165ed35cb12c9868f10e2a6aa07c merged revision(s) commit:06919949a60b42a8f30e8bd0cb075e17b0...
-
- [Bug #21141] [DOC] Clarify what time is in UTC
[Bug #21141] [DOC] Refine description of `Time#utc?`
... -
-
-
03:33 AM Bug #21257: YJIT can generate infinite loop when OOM
- YJIT compiles the `first` and `second` methods to this (on x86_64-linux):
```
# regenerate_branch
# Block: first@infi... -
- We've found an edge case where YJIT can generate an infinite loop (jump to the same address) when it's out-of-memory....
-
- vs2022 17.14.0 preview 2.0 also has this issue.
-
-
-
-
04/08/2025
-
- https://github.com/rubygems/rubygems/commit/5b312fd040
-
- https://github.com/ruby/json/commit/97f2924d0a
-
- https://github.com/ruby/json/commit/97f47d1cfd
-
- After exchanging impressions with @byroot today, I have created https://github.com/ruby/ruby/pull/13085 to restore th...
- 05:46 PM Revision d5f94941 (git): [ruby/openssl] Fix the tests using SHA-1 Probabilistic Signature Scheme (PSS) parameters.
- Fedora OpenSSL 3.5 on rawhide stopped accepting SHA-1 PSS[1] parameters.
This is different from the SHA-1 signatures ... -
- Here is a repro without rspec.
```ruby
class C
define_method(:foo) do
p it
end
end
class D < C
... -
- I have the following script:
```rb
require "bundler/inline"
gemfile do
source "https://rubygems.org"
g... -
05:34 PM Misc #20968: `Array#fetch_values` unexpected method name in stack trace
- > In any case, it was reaffirmed that matz strongly prefers that `<internal:` not be displayed.
I somewhat agree t... -
- `rake` suppressed these backtrace at https://github.com/ruby/rake/pull/554
I'm +1 to proposal of mame and matz. - 04:52 PM Revision 5aa05f17 (git): Fix lldb debug scripts (#13048)
- In ruby/ruby#13008 `RVALUE` was removed without replacement. This means
the lldb scripts that relied on `RVALUE` stop... -
- Btw, @ko1 came up with this idea, so I want to say thanks to him.
-
- Earlopain (Earlopain _) wrote in #note-3:
> > As you can see in the above output, the Class#new frame is eliminated.... -
- > As you can see in the above output, the Class#new frame is eliminated. I'm not sure if anyone really cares about th...
-
- ko1 (Koichi Sasada) wrote in #note-1:
> `swap` is remained?
I [made a patch to remove `swap`](https://github.com/... -
- `swap` is remained?
-
- Thank you for you trust Matz!
I think it should be helpful to onboard new people if we have an architecture that i... -
- I agree with making ZJIT upstream. And I feel no worry about the migration, since I trust the team with merging proce...
-
- mame (Yusuke Endoh) wrote in #note-9:
> Given the convention of deriving methods such as `key_set` from `keys`, wher... -
- Given the convention of deriving methods such as `key_set` from `keys`, where the former returns a set instead of an ...
-
-
- I tried to build ruby.exe with Windows SDK 10.0.26100. But it failed with:
```
dmyext.obj dmyenc.obj
linking sha... -
- > So I guess I should merge the proposed patch as a precaution?
I think so yes, with that eval code as a test case. -
- byroot (Jean Boussier) wrote in #note-7:
> Can source line ever legitimately be `0`?
With `eval`, it can. It can ... -
- @mbcodeandsound Just FYI, I bet you meant to write `!(1..16).cover?(channel)` in the following line.
https://git... -
- Nice find!
Can source line ever legitimately be `0`? Perhaps we are missing some assertions that would have caught... -
- @byroot Thanks! I think this is an issue of the prism compiler handling the line number around a flip-flop syntax.
t... -
- What I've figured for now it that the corruption is triggered by:
```ruby
# This require line makes sure the or... -
- > The issue reproduces on my machine, but it's very rare.
You can make it happen 100% of the time with the follow... -
- Thanks for the report. The issue reproduces on my machine, but it's very rare. I haven't figured out what happens.
... -
- > I do not know if it's normal for rb_sourceline() to return 0
Yes:
```ruby
const char *
rb_sourcefile(vo... -
- Hi @mame, did you have a chance to watch the video or discuss it in a dev meeting?
-
- ```
2) Error:
TestDir#test_children_long_name:
Test::Unit::ProxyError: No such file or directory @ apply2files - C:... -
- These assertions didn't handle drive letter of Windows
```
1) Failure:
TestFileExhaustive#test_dirname [V:/github.... -
06:00 AM Bug #21104: Net::HTTP connections failing in Ruby >= 3.4.0 on macOS with Happy Eyeballs enabled
>For those experiencing this problem, can you successfully run the following script?
```
➜ ruby-sandbox ruby -...-
03:12 AM Revision d17ab5a4 (git): [ruby/mmtk] Do root scanning in scan_vm_specific_roots
- We rely on scan_vm_specific_roots to reach all stacks via the following
path:
VM -> ractors -> threads -> fibers...