Activity
From 08/17/2025 to 08/23/2025
Today
-
07:04 AM Misc #21458: Test 'make install'?
- > There has also been numerous bugs related to make install and tool/rbinstall.rb.
> Some of those would have likely... -
12:13 AM Revision 5b5b5b3a (git): ZJIT: Spill whole FrameState in `Insn::SendWithoutBlock`
- Previously, we only spilled the arguments necessary for the particular
send. In case the callee raises and a rescue r...
08/22/2025
-
09:47 PM Revision 8c24e668 (git): ZJIT: Allow querying a single ZJIT stat (#14309)
- * ZJIT: Add RubyVM::ZJIT.stats_enabled?
* ZJIT: Allow querying a single ZJIT stat -
07:13 PM Misc #21458: Test 'make install'?
- hsbt (Hiroshi SHIBATA) wrote in #note-9:
> But I'm against to add cli test like `rdoc --version` because we explicit... -
- mame (Yusuke Endoh) wrote in #note-2:
> What do you think?
That looks a lot clearer to me.
I think it's not real... -
02:37 PM Feature #21543: Point ArgumentError to the call site
- I also like the error highlight solution much better.
However I wonder if source is even needed? Presumably we hav... -
01:49 PM Feature #21543: Point ArgumentError to the call site
- mame (Yusuke Endoh) wrote in #note-2:
> At the dev meeting, it was suggested that we try addressing the issue with e... -
03:19 AM Feature #21543: Point ArgumentError to the call site
- At the dev meeting, it was suggested that we try addressing the issue with error_highlight without changing the backt...
-
- mrkn (Kenta Murata) wrote in #note-12:
> In general, adding only `mean` (I prefer `mean` over `average`, see below) ... -
03:15 PM Feature #21518: Statistical helpers to `Enumerable`
- An `average` alias would be nice, though.
-
05:00 AM Feature #21518: Statistical helpers to `Enumerable`
- Hi. I'm a creator of enumerable-statistics gem and the original proposer of `Array#sum` and `Enumerable#sum`.
In gen... -
06:21 PM Revision 1e3fcc28 (git): Fix typo in function name ractor_port_initialize
- 05:49 PM Revision 9db54a1a (git): Fixes to encoding/transcoding for ractors.
- Not all ractor-related encoding issues were fixed by 1afc07e815051e2f73493f055f2130cb642ba12a.
I found more by runnin... - 05:34 PM Revision 48fc41cc (git): ZJIT: Temporarily downgrade assert_compiles to assert_runs
- 05:34 PM Revision b6f0cd51 (git): ZJIT: Update tests
- 05:34 PM Revision bcd9c87d (git): ZJIT: Eagerly place CheckInterrupts
- Ideally we would do this lazily post optimization, but this is easier
for now: we already have the requisite Snapshot... - 05:34 PM Revision d690e832 (git): ZJIT: Add CheckInterrupts HIR instruction
-
05:03 PM Feature #21545: #try_dig, a dig that returns early if it cannot dig deeper
- Maybe I'm reading too much into the examples, but this looks to me like something where pattern matching would be mor...
-
- Applied in changeset commit:git|e3e87258ddadf763d9b0b11678935250f9bcc6c3.
----------
Fix issue where a new GC could ... -
- [Bug #21548]
In lazy sweeping, if we need to allocate an object in a heap where we
weren't able to free any slots, b... -
-
09:40 AM Feature #21550: Ractor.sharable_proc/sharable_lambda to make sharable Proc object
- I understand the concern for future confusion, but it's a trade-off. I'd accept confusion here (option 1) to avoid co...
-
- The CI did pass on Windows e.g. https://github.com/ruby/pathname/actions/runs/16760570256/job/47454475161
Are there ... -
07:55 AM Feature #21532: Define most of Pathname in Ruby code
- https://github.com/ruby/pathname/pull/57 is incomplete.
Some methods are incompatibility with C Version and Ruby ver... -
- matheusrich (Matheus Richard) wrote in #note-7:
> It would be a bummer if the whole proposal was rejected over a min... -
- We discussed about the edge cases 00:00 and 12:00 in developer meeting.
After some discussion we concluded there is ... -
- jemalloc development has been discontinued.
https://jasone.github.io/2025/06/12/jemalloc-postmortem/ -
- We discussed this behavior at https://bugs.ruby-lang.org/issues/21508
I understood this issue. But It's hard to fi...
08/21/2025
-
11:51 PM Feature #21550 (Open): Ractor.sharable_proc/sharable_lambda to make sharable Proc object
- Let's introduce a way to make a sharable Proc.
* `Ractor.shareable_proc(self: nil, &block)` makes proc.
* `Ractor... -
10:52 PM Revision 6fbe2dd3 (git): [DOC] Tweaks for String#insert
-
10:42 PM Revision 823d55a8 (git): Add lock-free fastpath to callable_method_entry...
-
- Indeed, you're right:
```
Thread 1 "ruby" hit Breakpoint 1, syntax_error_initialize (argc=0, argv=0x0, self=1407366... -
10:45 AM Bug #21540: prism allows `foo && return bar` when parse.y doesn't
- That's just how parse.y shows its syntax errors in general, you can check with any other syntax error. `-c` doesn't i...
-
-
- jhawthorn (John Hawthorn) wrote in #note-27:
> I really do not believe it is a departure from the existing semantics... -
- Log at https://github.com/ruby/dev-meeting-log/blob/master/2025/DevMeeting-2025-08-21.md
matz (Yukihiro Matsumoto) w... -
- We discussed at the developers' meeting, and had the conclusion that the original issue cannot be addressed if we use...
-
- Eregon (Benoit Daloze) wrote in #note-24:
> ```ruby
> counter = 0
> get "/" do # assume the proc gets copied here ... -
- My summary:
* To prohibit local variable assignment, we need to change the logic around local variables.
* as jha... - 09:30 PM Revision 2ee5a892 (git): Make `vm_search_method` return a cme instead of a cc (#14299)
- Make vm_search_method return a cme instead of a cc
Both of its callers ended up calling `vm_cc_cme` on the result an... -
- Thanks to nobu for pointing this out. This is a YJIT file so shouldn't
have ZJIT stuff in it. ZJIT doesn't support bu... -
08:12 PM Feature #21545: #try_dig, a dig that returns early if it cannot dig deeper
- I have never seen an API that can return either a string `"ok"` or a hash `{ code: 200 }`
Swallowing exceptions li... -
- How about adding keyword argument, e.g. `exception: true` to behave `try_dig`.
Matz.
-
- Co-authored-by: Alan Wu <alansi.xingwu@shopify.com>
Co-authored-by: Max Bernstein <ruby@bernsteinbear.com>
Co-authore... -
- I was able to reproduce this issue using the rexml tests as well and I have a fix here: https://github.com/ruby/ruby/...
-
- The following segmentation fault are happened in recent days.
* https://github.com/ruby/ruby/actions/runs/17114960... -
- The single-character syntax itself isn't a problem if limited to instance variables. `foo ?=` is already valid and sh...
-
- Currently, `?=` means "a single character =". We don't want to break current available syntax, unless absolutely nece...
-
- @matz I propose `?=` as the uninitialized assignment operator:
```rb
@foo ?= some_heavy_calculation(...)
``` -
- I understand the motivation, but the proposed "@||=" is unacceptable. Other suggestions were made, but I don't think ...
-
- This no longer deadlocks. Under the new Ractor API the `take` should be replaced with `value`.
``` ruby
require "... -
- I believe the Ractor::Port interface doesn't suffer from this issue
```
ruby -e 'th = 2.times.map { |i| Thread.ne... -
07:26 PM Revision f2715af9 (git): ZJIT: Allocate register for VRegs that begin and end at the same index (#14270)
- If the LiveRange looks like (idx, idx), we will currently not allocate a
register. This change allocates a register a... -
- Applied in changeset commit:git|7ac16eff311f9bc762586bda9540d82e8eb7f135.
----------
Adjust snt < max_cpu calculatio... -
- [Bug #20146]
Previously we dealt with the main Ractor not being enabled for M:N by
incrementing snt_cnt++. This work... -
- Thanks!
-
12:58 PM Feature #21532: Define most of Pathname in Ruby code
- Ok. Merge it.
-
- Fixed by commit:acb29f7fa1497463ed3bdd65549ef20b61beda64
-
03:46 PM Revision acb29f7f (git): Do not respect ruby2_keywords on method/proc with post arguments
- Previously, ruby2_keywords could be used on a method or proc with
post arguments, but I don't think the behavior is d... -
03:41 PM Revision b0c80c2b (git): Remove unused SPECIAL_CONST_SHAPE_ID
- Its usage was removed in 306d50811dd060d876d1eb364a0d5e6106f5e4f1.
-
03:27 PM Misc #21458: Test 'make install'?
- naruse (Yui NARUSE) wrote in #note-8:
> Many years ago, test-all was run for install-ed Ruby, but these days it ru... -
- FYI: We already tested `make install` HEAD version and all of stable versions at https://github.com/ruby/actions.
... -
- Testing `make install` sounds reasonable. But complex multiple testing for installed files seems too complex.
> Ma... - 02:45 PM Revision 18aa7306 (git): ZJIT: Refactor `gen_new_hash` (#14293)
- We can use the `gen_push_opnds` and `gen_pop_opnds` helpers added in #14200 to simplify the code.
-
-
-
- > Formatting times should be done with an appropriately internationalized library.
This is not formatting. This is i... -
01:18 PM Bug #21547: SEGV after 2083fa commit
- Thanks. It works well!
-
- Thanks as always. I can confirm that this fixes this issue.
-
- Applied in changeset commit:git|b6bf44ae0f0196c58a07e13ac1bd7adafd13f8b2.
----------
variable.c: handle cleared fiel... -
- Thanks everyone, I have a fix for it: https://github.com/ruby/ruby/pull/14291
-
- I was able to reproduce with the `cool.io` test suite.
-
- We are running tests using Ruby HEAD on Linux, macOS, and Windows platforms.
https://github.com/fluent/fluentd/actio... -
- Thanks. Perhaps it's a linux only issue, I'll try to investigate further.
-
- I reproduces pretty much all the time for me (with a clean install). A good starting point seems to be the `cool.io` ...
-
- @watson1978 Unfortunately this doesn't seem to reproduce on my machine.
Does your crash report include the C level... -
- [Bug #21547]
Followup: https://github.com/ruby/ruby/pull/14201
When adding an instance variable and the IMEMO/field... -
11:32 AM Revision a837ec09 (git): Drop a workaround for CentOS 6
- `RUBY_CXX_DEPRECATED` is overridden using `RBIMPL_ATTR_DEPRECATED` in
include/ruby/backward/2/attributes.h already. -
- Applied in changeset commit:git|f2b1017b3df5f93aaf1d4ccd758d341298aa8dcc.
----------
Add `Math.log1p` and `Math.expm... -
- Accepted.
Matz.
-
-
- This commit adds two new methods to the `Math` module:
* `Math.log1p(x)`: Computes `Math.log(x + 1)`
* `Math.expm1(x... -
- # The next dev meeting
**Date: 2025/09/11 13:00-17:00** (JST)
Log: *TBD*
- Dev meeting *IS NOT* a decision-mak... -
-
- TIPS: By adding `[DOC]` tag in the commit log, you can download and preview the generated HTML files from the "Misc" ...
-
-
- jpcamara (JP Camara) wrote in #note-3:
> Yea I hear ya. So should I just submit a PR with my suggestions for the doc... -
- Sorry, `/o` is a traditional option. Even though Perl is not popular anymore, we are **not** going to deprecate it. I...
-
- The `o` option for regexp is `once` option originally from Perl regexp.
https://perldoc.perl.org/perlop#Regexp-Quote... -
- @k0kubun `name2ctype.h` is generated by `tool/enc-unicode.rb`. You can run it on ruby_3_4.
-
- `fetch_path` is much better than `dig!` since this bang method is against our naming convention (if we have x and x!,...
-
09:30 AM Feature #21520: Feature Proposal: Enumerator::Lazy#tee
- matz (Yukihiro Matsumoto) wrote in #note-11:
> I don't think `lazy_each` is a good name. We just wanted to peek the ... -
- I don't think `lazy_each` is a good name. We just wanted to peek the element in the stream, being lazy or not. In tha...
-
- nobu (Nobuyoshi Nakada) wrote in #note-9:
> A couple days ago, another name came to me: `tee`.
Thanks, `tee` is a... -
- A couple days ago, another name came to me: `tee`.
-
- I am positive about adding those methods, but I am no expert on Mathematics nor Statistics.
Matz.
-
- I accept `Ruby::Box`. I saw the proposal for `Ruby::Context` but this namespace things are a bit different from the `...
-
-
- OK, now we have updates both for prism and parse.y. Go ahead.
Matz.
- 01:55 AM Revision 60189671 (git): Bump github.com/microsoft/vcpkg
- Bumps [github.com/microsoft/vcpkg](https://github.com/microsoft/vcpkg) from master to dd3097e305afa53f7b4312371f62058...
08/20/2025
-
- Given it's `ruby: compile error (SyntaxError)` from parse.y I guess it's a check done in `compile.c` and missing in `...
-
- @jhawthorn I apologize if my reply sounded disrespectful or so, it was not my intention, I am/was genuinely surprised...
-
- > It makes very little sense to me that example 4 is fine but example 2 isn't.
Ideally example 4 would also be forb... -
- 08:17 PM Revision 19ad72d2 (git): ZJIT: Remove unnecessary option return type on gen_branch_params (#14286)
- 07:53 PM Revision a7a026ae (git): YJIT: Improve locals names (#14285)
-
- We lock the VM to invalidate method entries. However, we do not lock the
VM to call methods, so it's possible that du... -
- When calling getnameinfo we spawn a thread because it may do a slow,
blocking reverse-DNS lookup. Spawning a thread i... -
-
10:28 AM Bug #21534: ppc64le bootstraptest/test_ractor.rb aborted (core dumped) on configure cppflags=-DRUBY_DEBUG
- I found the following text in my assumption. we should use `RUBY_DEBUG` macro without optimization?
https://github... -
- 06:48 AM Revision 609fe7e4 (git): Bump lewagon/wait-on-check-action
- Bumps [lewagon/wait-on-check-action](https://github.com/lewagon/wait-on-check-action) from 31f07a800aa1ba8518509dc856...
-
- 05:46 AM Revision 3443a0bc (git): Bump actions/cache from 4.2.3 to 4.2.4
- Bumps [actions/cache](https://github.com/actions/cache) from 4.2.3 to 4.2.4.
- [Release notes](https://github.com/act... - 04:10 AM Revision f7687c4a (git): Enable Dependabot updates for vcpkg
- See https://github.blog/changelog/2025-08-12-dependabot-version-updates-now-support-vcpkg/
-
- I am maintaining [Fluentd](https://github.com/fluent/fluentd).
When I run Fluentd's unit tests after [2083fa](https:... -
08/19/2025
-
-
-
- gen_prepare_call_with_gc() was not enough because of the rb_funcall()
usage in range_init().
Co-authored-by: Takashi... -
- 32-bit platforms do not have flonum and something about the static symbol test was flaky.
-
- tenderlovemaking (Aaron Patterson) wrote in #note-21:
>
> ```ruby
> foo = 123
> Ractor.shareable_proc { foo }
>... -
-
- Both the X509 store and X509 store context were missing write barriers.
To the callback object being stored in the ex... -
-
02:02 PM Revision fc5ee247 (git): ZJIT: Compile toregexp (#14200)
- `toregexp` is fairly similar to `concatstrings`, so this commit extracts
a helper for pushing and popping operands on... -
-
-
-
-
-
- stanhu (Stan Hu) wrote:
> 1. Should CXX just be set to `nil`? Or should all C++ extensions be expected to check for... -
- I think the issue itself is already fixed. Let's discuss adding a test for `make install` in #21458
-
- `logger` is now bundled gems. We should discuss this at https://github.com/ruby/logger/pull/132
-
-
- Because TypedData_Make_Struct both allocates an object and after
xmallocs memory, all added references must always be... -
- Update included file list automatically.
-
- https://github.com/ruby/syntax_suggest/commit/54bb8ab330
08/18/2025
-
- It only makes sense for heap objects.
-
- I think the second line starting with `for ` looks very strange and breaks the flow, also the order is then incorrect...
-
-
- We recently converted this to be WB_PROTECTED, but missed some locations
the seed was set.
This is a little bit conf... -
- When we dup a complex object we need to issue a writebarrier_remember
on the new object. This was caught by wbcheck i... -
05:49 PM Bug #21342: Segfault: invalid keeping_mutexes when using Mutex in Thread then Fiber after GC
- I created a PR here: https://github.com/ruby/ruby/pull/14262
-
-
-
- * ZJIT: Handle ISEQ moves in IseqCall
* ZJIT: Handle ISEQ moves in Invariants
* Let gen_iseq_call take a reference
... - 04:03 PM Revision 68c7f10b (git): ZJIT: Temporarily disable buggy defined? codegen
- 04:03 PM Revision 33224236 (git): ZJIT: Refine type of Defined
- 04:03 PM Revision c663fe1e (git): ZJIT: Make Defined retain its Snapshot
- 04:03 PM Revision f3dbf7de (git): ZJIT: Make a fix a bunch of other cases using Option
- 04:03 PM Revision d2a53c7c (git): ZJIT: Don't return Option from counter increment
- 04:03 PM Revision 3c2cff62 (git): ZJIT: Don't return Option from side exits
- 04:03 PM Revision 1e42f9ed (git): ZJIT: Remove empty returns from control instructions in codegen
- 04:03 PM Revision ad9fda7a (git): ZJIT: Don't return Option from codegen helper functions
- They don't need it anymore; jit.get_opnd doesn't return Option anymore.
- 04:03 PM Revision 5b3c87a0 (git): ZJIT: Make jit.get_opnd noisily fail
- We have a verifier that runs in debug mode that should prevent this.
Simplify a bunch of call sites. -
- Make LIR SSA.
-
- I tested at the latest master branch commit <7ddc53bc0c57b1c079e40a74d233f537111b3e92> below.
Interestingly, the a... - 02:59 PM Revision cb31be5d (git): ZJIT: Codegen for NewHash (#14059)
- This is my first contribution to ZJIT.
Co-authored-by: Takashi Kokubun <takashikkbn@gmail.com>
Co-authored-by: Max B... -
- Just want to point out that Ruby itself doesn't have any concept of `try`. This *might* be best as a proposal for Act...
-
02:51 PM Feature #12282: Hash#dig! for repeated applications of Hash#fetch
- Agreed, this feature would be useful. I would like to essentially assert that all of the nodes along the path are fou...
-
-
-
- Since we handle embedded arrays in the if statement above, we don't need
to handle it here. -
- 03:31 AM Revision e3ce56c9 (git): Fix typos
-
- https://github.com/ruby/json/commit/72e231f929
-
- https://github.com/ruby/json/commit/2d2e0d403d
-
03:31 AM Revision f0ee7630 (git): Merge specs checking CLI flags and subcommands are documented
- We had them duplicated, but with slightly different features:
* The ones in `other/cli_man_pages.rb` enforced a spec... -
- https://github.com/rubygems/rubygems/commit/b58829a868
-
- https://github.com/rubygems/rubygems/commit/05199ae0c1
-
- https://github.com/rubygems/rubygems/commit/d41b8d303c
-
- To save some unnecessary `bundle install` commands.
https://github.com/rubygems/rubygems/commit/61e7d9d09a -
- Make synopsis, subcommands, and CLI flags use a format consistent with
the other docs, and also reword some sentences... -
- Name default value placeholders in a more standard way. That's what our
specs check, but they don't yet work for subc... -
- For synopsis, subcommand and flag documentation like other commands do.
https://github.com/rubygems/rubygems/commit/... -
- https://github.com/rubygems/rubygems/commit/2cbe7ea0a5
-
- We had actually cancelled this change. After a second look though, it
does seem like a reasonable plan since `bundle ... -
- https://github.com/rubygems/rubygems/commit/7cad1e4947
-
- If upgrading `foo` needs an indirect dependency to be downgraded,
Bundler would not be able to upgrade foo.
This is ... -
03:31 AM Revision 81360399 (git): [rubygems/rubygems] Introduce `bundle list --format=json`
- The `bundle list` command is a convenient way for human to know what gems and versions are available. By introducing ...
-
- This reverts commit https://github.com/rubygems/rubygems/commit/bea87eab0b17 and adds a
regression spec for it.
http... -
- Like others, it's a remembered option which we are deprecating in favor
of configuration.
https://github.com/rubygem... -
- For consistency.
https://github.com/rubygems/rubygems/commit/3e3364e19f -
- https://github.com/rubygems/rubygems/commit/4d0c058e6a
-
- Since the lowest supported version is now 3.4.1.
https://github.com/rubygems/rubygems/commit/d00e03c52e - 03:31 AM Revision d4bf58b5 (git): [rubygems/rubygems] Don't worry about missing Makefile.
- https://github.com/rubygems/rubygems/commit/0e92346d88
-
- The current oldest support Ruby version is 3.2. And Ruby 3.2 bundled
Bundler 2.5. It means RG 4.0 can drop to support... -
-
-
-
08/17/2025
-
-
- The VPATH rule of NMake is different from others. Abandon using
them in the rules for the generated source, locate t... -
- Applied in changeset commit:git|7c28fb2fb2544e5fed75ef216c4dd08084b38671.
----------
[Bug #21546] Make the generated... -
- It is relocatable and should reflect the installed location.
```console
$ export PKG_CONFIG_PATH=/opt/local/lib/p... -