Activity
From 11/01/2025 to 11/07/2025
Today
-
05:34 PM Feature #21653: Unify Hash methods and preserving default/default_proc
- I just confirmed the behavior described on version 3.5.0dev, but for Hash#select. I'll check other methods to see which of them are not preserving `default` and `default_proc`
-
- I confirmed the behavior you described on version 3.2.4 while testing Hash#compact, but looks like it is already fixed on version 3.5.0dev (Although I did not checked the other methods you mentioned).
- 04:54 PM Revision 9bbe4b60 (git): ZJIT: Carve out IseqPayload into a separate module (#15098)
-
03:29 PM Revision ae60b0bf (git): Use DOT_WAIT for old GNU make [ci skip]
-
-
-
02:35 PM Revision a2201570 (git): Remove `rb_path_check` declaration
- Implementation was removed in https://github.com/ruby/ruby/commit/a4c051b870ac7f7b3c5482baf05600e1f6751b47
-
-
-
-
-
-
-
09:31 AM Bug #21671: Rails CI raises `Assertion Failed: rbimpl_rstring_getmem:RB_TYPE_P(str, RUBY_T_STRING): actual type: 26` with "-DENABLE_PATH_CHECK=0 -DRUBY_DEBUG=1" enabled
- The failure comes from `rb_intern3 -> sym_find_or_insert_static_symbol` (during Prism parsing).
cc @luke-gru and @jhawthorn, maybe this ring a bell? -
07:14 AM Bug #21671: Rails CI raises `Assertion Failed: rbimpl_rstring_getmem:RB_TYPE_P(str, RUBY_T_STRING): actual type: 26` with "-DENABLE_PATH_CHECK=0 -DRUBY_DEBUG=1" enabled
- I tried to identify which commit introduces this assertion failure but not found.
-
- Recently Rails CI using "-DENABLE_PATH_CHECK=0 -DRUBY_DEBUG=1" enabled ruby raises `Assertion Failed: rbimpl_rstring_getmem:RB_TYPE_P(str, RUBY_T_STRING): actual type: 26` assertion failures.
https://buildkite.com/rails/rails-nightly/... - 09:07 AM Revision a00f425e (git): Update default gems list at a881f2a0f441bf6d06a68bf711ca81 [ci skip]
-
- https://github.com/ruby/json/commit/5a12067f88
-
- This prevent from freezing and sharing state instances.
If you needs some sort of arguments or extra state to the generator
methods, consider using `JSON::Coder` instead.
https://github.com/ruby/json/commit/e9fbc8937f -
- https://github.com/ruby/json/commit/826cb2a4f4
-
08:32 AM Misc #21647: DevMeeting-2025-11-13
- * [Feature #21637] tracepoint: add support for global variables read/write events (viralpraxis)
* There seems to be no straightforward way to trace gvar reads/writes in runtime
* `Kernel.trace_var` only works when gvar name is know... -
07:41 AM Revision 6d81969b (git): Development of 4.0.0 started.
-
07:41 AM Bug #21498: Windows - Ruby Overrides C Library APIs thus breaking them
- I have attached a screenshot showing the problem. Notice in the stack trace on the bottom of the page that the method ` bmp::Bitmap::save(const std::filesystem::path & filename)` calls fclose but ends up calling the Ruby version! Not the...
-
- In my case I created a Ruby extension based on https://github.com/baderouaich/BitmapPlusPlus. Every time I tried to save a bitmap though it was corrupted. I submitted a bug to the project - https://github.com/baderouaich/BitmapPlusPlus/i...
-
- I can't the exact cause how the problem occurs, maybe passing `FILE*` opened in other than Ruby to `rb_w32_fclose`?
usa (Usaku NAKAMURA) wrote in #note-7:
> I'm still a bit concerned about compatibility with existing extension libr... -
- The universal crt was introduced by Microsoft in 2015 (https://learn.microsoft.com/en-us/cpp/porting/upgrade-your-code-to-the-universal-crt?view=msvc-170) and Ruby for Windows has used it since Ruby 3.1 in 2021 (https://rubyinstaller.org...
- 04:14 AM Revision aaf1f53d (git): Follow renaming from Namespace to Ruby::Box
- 04:14 AM Revision 8ebeb751 (git): Update Ruby Box document, delete problems/discussions already resolved
- 04:14 AM Revision 49c0502f (git): Apply renaming to Ruby::Box in test codes
- 04:14 AM Revision 19d4663d (git): Use RUBY_BOX environment variable instead of RUBY_NAMESPACE
- 04:14 AM Revision ccad753c (git): Re-rename files wrongly renamed
- 04:14 AM Revision c4691ef0 (git): Rename Namespace to Ruby::Box
- 04:14 AM Revision 95a110a9 (git): Define ::Ruby module earlier to define classes under it
- 04:14 AM Revision d2a587c7 (git): renaming internal data structures and functions from namespace to box
- 04:14 AM Revision aaa12347 (git): update referenced filenames from namespace to box
- 04:14 AM Revision 50b9d9d3 (git): rename namespace.c (and others) to box.c
- 04:14 AM Revision f19b29d1 (git): Add namespace tests about global variables, toplevel methods
11/06/2025
-
11:49 PM Bug #21498: Windows - Ruby Overrides C Library APIs thus breaking them
- @alanwu
I'm still a bit concerned about compatibility with existing extension libraries, but with Ruby4 about to be released, it might be a good time to make the change. So, I'm not opposed to it this time. -
- Thanks @alanwu. I added comments to the commit in Github, but copying here:
This MR will result in the Ruby dll exporting the symbol _fclose instead of fclose. The exporting happens here:
https://github.com/ruby/ruby/blob/c6c92bdce3ad2... -
03:51 PM Bug #21498: Windows - Ruby Overrides C Library APIs thus breaking them
- Recent C runtimes on Windows have C99 compliant fclose(), and I think overriding it and breaks otherwise standard-compliant extensions. Removing the fclose() override passes CI, at least. <https://github.com/ruby/ruby/pull/15073>
@usa W... -
11:46 PM Revision cf4a034d (git): Use rb_set_memsize for constant cache tables
- These were converted to a set in
c0417bd094abcc68be913ce49a430df7cefbcd44 - 10:14 PM Revision 844132ae (git): ZJIT: Remove obsolete register spill counters (#15089)
- 10:14 PM Revision 93430176 (git): ZJIT: Fix an incomplete comment (#15088)
- 09:32 PM Revision 2998c8d6 (git): ns_subclasses refcount accesses need to be atomic (#15083)
- We were seeing errors like:
```
* thread #8, stop reason = EXC_BAD_ACCESS (code=1, address=0x803)
* frame #0: 0x00000001001fe944 ruby`rb_st_lookup(tab=0x00000000000007fb, key=1, value=0x00000001305b7490) at st.c:1066:22
frame #1: ... -
08:07 PM Revision 38d31dc4 (git): ZJIT: Untag block handler (#15085)
- Storing the tagged block handler in profiles is not GC-safe (nice catch,
Kokubun). Store the untagged block handler instead.
Fix bug in https://github.com/ruby/ruby/pull/15051 - 06:50 PM Revision c6c92bdc (git): zjit-macos.yml: Unset MAKEFLAGS before ruby-bench (#15084)
-
-
06:25 PM Revision 3150a1d9 (git): [Backport #13671] Fix that "ss" in look-behind causes syntax error
- Fixes k-takata/Onigmo#92.
This fix was ported from oniguruma:
https://github.com/kkos/oniguruma/commit/257082dac8c6019198b56324012f0bd1830ff4ba
https://github.com/k-takata/Onigmo/commit/b1a5445fbeba97b3e94a733c2ce11c033453af73 -
- [PATCH] [Bug #21652] [DOC] Update unmarshalable object list
The `Data` mentioned here was the old `T_DATA` class, not the current
extended `Struct`. -
- https://github.com/sampersand/blog/blob/master/the%20-s%20flag.md#the-segfault
- 06:21 PM Revision ff6c728a (git): ZJIT: Restore dropped_bytes after temporary OOM (#15069)
-
06:18 PM Bug #21667: CVE-2024-12224
- Excellent news, glad to see the patch progressing.
How quickly can we release new versions of Ruby to include this patch? -
- https://rustsec.org/advisories/RUSTSEC-2024-0421.html
This seems to be from MMTk depending on the `idna` crate. MMTk is experimental and requires a separate build step, so ruby-build probably doesn't even build it.
In any case, we ... -
- Wiz reports a servo/rust-url package. Curious if Ruby is using this package strictly at the point in time when the Ruby language is being compiled, possibly even an integration test suite. Or perhaps servo ends up as a portion of the Rub...
-
- What version of ruby were you building? Does Wiz point to some file that this is about?
-
06:01 PM Feature #21665: Revisit Object#deep_freeze to support non-Ractor use cases
- Was it considered keep `make_shareable` name, but port it outside of `Ractor` class to make it independent of the "sharing" implementation and useful for JRuby and others?
-
03:26 PM Feature #21665: Revisit Object#deep_freeze to support non-Ractor use cases
- I really think the naming scheme should clearly distinguish between regular objects from classes/modules.
Freezing modules/classes should be separated into their own ad-hoc methods, since you can break a lot of existing code if you free... -
- We have both `SetIvar` and `SetInstanceVariable`. The former is a purely
dynamic fallback that we can inline `attr_accessor`/`attr_writer` into,
whereas the latter comes straight from the interpreter's
`setinstancevariable` opcode. -
04:19 PM Revision f7e7443a (git): Adjust OpenSSL specs for digest algorithm lookup
- https://github.com/ruby/openssl/pull/958 changed the common logic for
digest algorithm lookup:
- If the argument is neither an OpenSSL::Digest instance nor a String,
it is now implicitly converted to String with #to_str. This is
... -
- * [Bug #21498] Windows - Ruby Overrides C Library APIs thus breaking them (alanwu)
* TL;DR linking with ruby.dll overrides some libc symbols on Windows
* The ticket proposes to stop all overrides, is that acceptable?
* If not, r... -
- * [Bug #20409] Missing reporting some invalid breaks (kddnewton)
* The main point of the issue is already resolved in both parsers
* `END { break }` is not a syntax error in any parser
* Any reason not to treat `END { break }` the ... -
- Fixed with https://github.com/ruby/ruby/commit/d73eba51853fde74624984b3063bd648b664e3e3
-
02:44 PM Bug #21666: Math.lgamma(-1).should == [infinity_value, 1] fails with Fedora glibc-2.42.9000-8.fc44
- Okay, lgamma testsuite now seems good after applying https://github.com/ruby/ruby/pull/15076 with Fedora glibc-2.42.9000-8.fc44
-
- To be sure, I will try building and run test with Fedora glibc-2.42.9000-8.fc44 (after applying the above PR)
-
- > https://github.com/ruby/ruby/compare/master...nobu:ruby:lgamma-unspecified-values?expand=1
Thank you. I've checked the above change and as far as I read the specification (POSIX) this is correct. -
- Is this correct?
https://github.com/ruby/ruby/compare/master...nobu:ruby:lgamma-unspecified-values?expand=1 -
- This is a follow-up to commit https://github.com/ruby/openssl/commit/e74ff3e2722f, which missed the line added
in a different PR.
https://github.com/ruby/openssl/commit/1b01d19456 -
- Remove the following subclasses of OpenSSL::PKey::PKeyError and make
them aliases of it.
- OpenSSL::PKey::DHError
- OpenSSL::PKey::DSAError
- OpenSSL::PKey::ECError
- OpenSSL::PKey::RSAError
Historically, methods defined on OpenSSL... -
- Fixes [#Bug 21660], followup to https://github.com/ruby/prism/pull/3674
https://github.com/ruby/prism/commit/fb445a49e5
Co-Authored-By: tomoya ishida <tomoyapenguin@gmail.com> -
- Likewise, use EVP_MD_fetch() if it is available.
This adds support for AES-GCM-SIV with OpenSSL 3.2 or later.
https://github.com/ruby/openssl/commit/0e565a215e -
- With the introduction of OpenSSL 3 providers, newly implemented
algorithms do not necessarily have a corresponding NID assigned. To use
such an algorithm, it has to be "fetched" from providers using the new
EVP_*_fetch() functions.
For ... -
- Raise OpenSSL::Cipher::CipherError instead of ArgumentError or
RuntimeError for consistency.
https://github.com/ruby/openssl/commit/78601c9c34 -
- We generally raise OpenSSL::OpenSSLError or its subclass for errors
originating from the OpenSSL library, which may include extra details
appended by ossl_raise().
https://github.com/ruby/openssl/commit/9427a05ce5 -
- Use explicit strings instead of relying on OpenSSL::ASN1::ObjectId
methods. It is reduced to just SHA-256 because testing other algorithms
does not improve test coverage for ruby/openssl.
https://github.com/ruby/openssl/commit/dcfd2e7b97 -
01:21 PM Bug #21669: Thoroughly implement void value expression check
- I agree with @mame — I think it is valuable, especially when it will be consistent. Happy to implement this change in prism.
-
- Stop documentation for undocumented private constants and private
class methods. And align private method definition styles.
Also `Socket.tcp_with_fast_fallback` looks private as well as
`Socket.tcp_without_fast_fallback`. -
- `RUBY_PLATFORM` should be invariant within the same process.
-
- I added it to the dev meeting and implemented it in https://github.com/ruby/prism/pull/3707
-
- Backport for 3.4 done in commit:5d2a244bfd844be3a94670ddcb509fb5a6b4e847
-
03:31 AM Revision 6014ed99 (git): Remove dead rb_hash_dump
-
- > introduce language standard versioning, that would make it clear both Ruby 3.4 and Ruby 4.0 are implementing same language standard
Maybe they are but that is not the only important thing. Current news list a bunch of default gems tha... -
01:29 AM Revision 00c9a21c (git): [DOC] Update glossary (#15070)
11/05/2025
-
- I don't believe 21021 is related, "try to mark T_NONE object" is a general symptom of missing a write barrier.
This does show a crash when marking the M_TBL so I would expect this to be fixed by https://github.com/ruby/ruby/commit/502... -
10:28 PM Bug #21670: 3.4.5 crashing with "try to mark T_NONE object"
- nrodriguez (Nicolas Rodriguez) wrote in #note-2:
> @byroot https://bugs.ruby-lang.org/issues/21021
I don't know if #21021 is the same underlying issue. This instance occurs both with and without YJIT enabled on 3.4.5, and it looks li... -
09:10 PM Bug #21670: 3.4.5 crashing with "try to mark T_NONE object"
- @byroot https://bugs.ruby-lang.org/issues/21021
-
- It's in `mark_method_entry_i`, which rings a bell. @jhawthorn didn't you fix that one?
@esaron please try the latest patch release: 3.4.7, your issue looks a lot like a missing write barrier issue and John fixed many of them in 3.4.7:... -
- We're experiencing a large number of crashes while running tests in CI on Ubuntu 22.04 runners. It is inconsistent, and we cannot reproduce reliably, although when the crash happens it appears to always happen in the same place, even if ...
-
- ufuk (Ufuk Kayserilioglu) wrote in #note-18:
> retro (Josef Šimánek) wrote in #note-17:
> ...
@ufuk Sure, I'm aware of and I do fully respect this. Ruby itself is not simple to version, since next to the language (having amazing backwa... -
08:03 PM Misc #21657: Question: Is Ruby 4.0 planned for December 2025 or later?
- I don't personally have a strong opinion about jumping to 4.0 purely for symbolic reasons, but if it's going to happen, it seems clear the community needs to know right now so we can start dealing with these issues.
JRuby made a leap ... -
06:17 PM Misc #21657: Question: Is Ruby 4.0 planned for December 2025 or later?
- retro (Josef Šimánek) wrote in #note-17:
> @mame so why to release 4.0 if not causing major breaking changes?
Quoting [my answer to this](https://bugs.ruby-lang.org/issues/21657#note-3) from above:
> ... that is completely a decisio... -
- @mame so why to release 4.0 if not causing major breaking changes?
-
03:34 PM Misc #21657: Question: Is Ruby 4.0 planned for December 2025 or later?
- Just adding my two cents to the off-topic discussion.
I honestly feel that the `~>` constraint was a terrible idea. It's going to be a huge pain when Ruby 4.0 is released (at the end of this year, perhaps?), and all the gems locked to... -
- I made a special kind of `ProfiledType` that looks at specific objects, not just their classes/shapes (https://github.com/ruby/ruby/pull/15051). Then I profiled some of our benchmarks.
For lobsters:
```
Top-6 invokeblock handler (100.0... -
- 06:44 PM Revision bf0331b9 (git): ZJIT: Add zjit_alloc_bytes and total_mem_bytes stats (#15059)
-
- new ZJIT stats excerpt from liquid-runtime:
```
vm_read_from_parent_iseq_local_count: 10,909,753
guard_type_count: 45,109,441
guard_type_exit_ratio: 4.3%
guard... -
06:28 PM Revision df290e11 (git): Skip an unstable IO test for mswin
- https://github.com/ruby/ruby/actions/runs/19107764906/job/54596244201
-
- https://github.com/ruby/ruby/actions/runs/19111531630/job/54609629054
- 06:01 PM Revision 7334244e (git): [ruby/erb] Fix tag shown in example of ERB expression tag and
- execution tag
(https://github.com/ruby/erb/pull/92)
These were the wrong way around.
https://github.com/ruby/erb/commit/50a5cd76fe - 05:57 PM Revision d3d2357a (git): ZJIT: Run ruby-bench CI for macOS arm64 as well (#15040)
-
05:42 PM
Bug #21664 (Closed): Segfault when running Ractor tests in the same process
- Applied in changeset commit:git|54907db8f3daa6d096e78e7eb78e515842c47789.
----------
Fix ractor move of object with generic ivars (#15056)
This bug was happening only when the `id2ref` table exists. We need
to replace the generic field... - 05:42 PM Revision 54907db8 (git): Fix ractor move of object with generic ivars (#15056)
- This bug was happening only when the `id2ref` table exists. We need
to replace the generic fields before replacing the object id of the
newly moved object.
Fixes [Bug #21664] - 03:36 PM Revision 57040636 (git): Update bundled gems list as of 2025-11-05
-
03:35 PM Revision af6a6a2a (git): Update power_assert to 3.0.1
-
- Fun C topic!
wsfulton (William Fulton) wrote in #note-8:
> A more conventional way for default aggregate initialisation is
> ...
Unfortunately, that empty initializer is a violation of the C99 language specification. You can confir... -
07:53 AM Bug #21659: rstring.h error: missing initializer for field ‘len’ of ‘struct RString’ [-Werror=missing-field-initializers] starting in ruby-3.3.10
- I thought it was UB in c++98, but have just checked the standard and I have this wrong as the missing members are indeed default initialized.
A more conventional way for default aggregate initialisation is
```c++
struct RSt... -
- Eregon (Benoit Daloze) wrote in #note-1:
> This check is getting more and more complicated and I wonder of the value of having it.
IMO, the current is indeed complicated because it is inconsistent. I think the new behavior is consist... -
- In prism, that whole check fits in about 130 lines: https://github.com/ruby/prism/blob/2ecd49dfeea1d0b00c3af609be82dd488d25c7f3/src/prism.c#L1027-L1166. Looking at that function, probably it would not be very difficult for prism to suppo...
-
01:33 PM Bug #21669: Thoroughly implement void value expression check
- The whole concept of "void value expression" is also weird in Ruby, where every expression/statement has a value, so I think it feels like something that doesn't belong to Ruby.
It probably only triggers for pretty broken code, and ev... -
- This check is getting more and more complicated and I wonder of the value of having it.
In my opinion there is little value of have this check, so given the complexity and the cost to check it (needs extra traversals of potentially larg... -
- A void-value-expression check is a syntax check that raises a `SyntaxError` if an expression that cannot grammatically return a value (a "void value expression," such as a `return` expression) appears in a context where a value is expect...
-
- +1 from me.
We see [a lot of](https://github.com/search?q=Ractor.make_shareable+language%3ARuby&type=code&l=Ruby):
```ruby
CONST = [...]
Ractor.make_shareable(CONST) if defined?(Ractor)
```
in gems, but that should really be:
``... -
10:35 AM Revision 946d2d03 (git): Suppressing unused warnings
-
- https://github.com/ruby/strscan/commit/5614095d9c
-
- `$Id$` is for RCS, CVS, and SVN; no information with GIT.
https://github.com/ruby/strscan/commit/9e3db14fa2 -
- https://github.com/ruby/strscan/commit/16ec901356
-
- ```
tool/sync_default_gems.rb:177:in `block in <module:SyncDefaultGems>': undefined local variable or method `it' for SyncDefaultGems:Module (NameError)
it.exclude << "lib/open3/jruby_windows.rb"
^^
from <internal:ke... -
- ruby_3_4 commit:c26a2b12e1f7f0779f2e86f59a405b1913cf72c8 merged revision(s) commit:377aa2a336cc700485c699ac49330f2a58b74906.
-
06:02 AM Bug #21668 (Closed): Backport 6cc66469a007d2fb272397e5e27c20b4ce628f5e
- Improves micro-benchmark of UnicodeNormalize
https://github.com/ruby/ruby/pull/14494
~~~ruby
100000.times{UnicodeNormalize.canonical_ordering_one("s\u0307\u0323")}
# 0.107795s → 0.067496s
~~~
-
- [PATCH] Improve performance of UnicodeNormalize.canonical_ordering_one
Use array_of_integer.sort! instead of buble-sort-like algorithm -
- Find interesting commits by following parents instead of relying on
"git log".
If we encounter a merge commit that may contain a conflict resolution,
fall back to cherry-picking the merge commit as a whole rather than
replaying each ind... -
- Currently, we try to git cherry-pick the upstream commit and then
resolve merge conflicts in the working tree with the help of Git's
rename detection. By the nature of heuristics, it does not work reliably
when the upstream adds or remov... -
- No behavior change is intended by this change.
-
- Use "git commit --amend" instead of "git filter-branch" since we only
need to handle one commit at HEAD. -
- This fixes it for the English gem.
-
- * See https://github.com/ruby/pathname/pull/57#issuecomment-3485646510
-
- https://github.com/ruby/strscan/commit/b4ddc3a2a6
-
- Cannot use C99 syntax, as far as supporting Ruby 2.6 and earlier.
https://github.com/ruby/strscan/commit/f6d178fda5 -
- Instead of an undocumented constant.
-
-
-
-
- After 34b0ac68b31, we use a fallback instead of side exit for splats.
Count splats under `send_fallback_one_or_more_complex_arg_pass`. -
- Kokubun bought up that "complex" is a more fitting name for what these
counters count. Thanks!
Also:
- make the SendFallbackReason enum name consistent with the counter name
- rewrite the printout prompt in zjit.rb - 02:36 AM Revision 033ba3c8 (git): Don't run global variable hook functions with VM lock held (#15053)
- We can't run arbitrary ruby code with the VM lock held.
-
- `eos?` is opposite, cannot be used instead of `rest?`.
https://github.com/ruby/strscan/commit/bee8cc547b -
12:09 AM Revision d24bb1e7 (git): [ruby/stringio] [DOC] Tweaks for StringIO#string=
- (https://github.com/ruby/stringio/pull/172)
https://github.com/ruby/stringio/commit/17ae4daf9a - 12:09 AM Revision bd3b44cb (git): ZJIT: Use a shared trampoline across all ISEQs (#15042)
-
- (https://github.com/ruby/stringio/pull/169)
https://github.com/ruby/stringio/commit/bef6541b55 -
- (https://github.com/ruby/stringio/pull/167)
https://github.com/ruby/stringio/commit/94303ace95 -
- (https://github.com/ruby/stringio/pull/170)
https://github.com/ruby/stringio/commit/da338d7e5d -
- (https://github.com/ruby/stringio/pull/168)
https://github.com/ruby/stringio/commit/9f10c7ae86 -
- (https://github.com/ruby/stringio/pull/166)
https://github.com/ruby/stringio/commit/5eeb61df34
11/04/2025
-
- (https://github.com/ruby/stringio/pull/163)
https://github.com/ruby/stringio/commit/a126fe252f - 09:56 PM Revision a0376eb2 (git): ZJIT: Fix --zjit-mem-size and add --zjit-exec-mem-size (#15041)
- ZJIT: Fix --zjit-mem-size and resurrect --zjit-exec-mem-size
-
- that has not been obsolete.
Partially reverting https://github.com/ruby/ruby/pull/15049. -
- (https://github.com/ruby/strscan/pull/169)
Partially revert https://github.com/ruby/strscan/pull/168 because
strscan_rest_p did not have `rb_warning("StringScanner#rest? is
obsolete")`.
It is actively used by the latest tzinfo.gem, and... -
-
- > the core idea of freezing something is to make it immutable, to hold it in place. It could be immutable or another analogy
The `immutable` name is an interesting concept but maybe more in the domain of #18035 than `deep_freeze`. One b... -
06:44 PM Feature #21665: Revisit Object#deep_freeze to support non-Ractor use cases
- I think this is generally useful. I hit a bug in syntax_suggest that took me about 4 hours to track down that boiled down to contents of an array being shared in an unintuitive way so their mutation wasn't cleanly isolated.
On the nam... -
- ## Proposal: Introduce `Object#deep_freeze` (or similar name) to freeze an entire object graph
I would like to re-propose the addition of Object#deep_freeze as a way to explicitly freeze an entire object graph. This proposal was rejec... - 07:46 PM Revision f1f2dfeb (git): Release VM lock before running finalizers (#15050)
- We shouldn't run any ruby code with the VM lock held.
-
- https://github.com/ruby/strscan/commit/1387def685
-
- ruby/strscan#168
-
06:54 PM Revision 480080b5 (git): [ruby/stringio] fix: (jruby) failing to clean buffer's code-range
- same bug as: https://github.com/jruby/jruby/issues/9035
https://github.com/ruby/stringio/commit/65b144b175 -
- Since https://github.com/ruby/strscan/commit/92961cde2b42.
https://github.com/ruby/strscan/commit/911f9c682a -
- vm_method.c already mentions it.
-
- This implements it the same as the other modular GC functions
-
- ruby-build triggers Wiz finding CVE-2024-12224 for the leftover build files, when compiling Ruby from source.
- 04:16 PM Revision ee74b97e (git): Update default gems list at 29847070f00184d7c0a97f8e1f18f5 [ci skip]
-
- https://github.com/ruby/io-wait/commit/57bc0b752b
-
- With Fedora glibc-2.42.9000-8.fc44 the following test:
https://github.com/ruby/ruby/blob/cdcb490d2bceb4ff8d6fce349047a2b722e6ae87/spec/ruby/core/math/lgamma_spec.rb#L9
fails like
```
1)
Math.lgamma returns [Infinity, 1] when pas... -
- I think `END { break }` should still be a syntax error, as it always results in an error.
-
- I think `END` should be syntax error. The same is already true for `BEGIN`, seems inconsistent.
-
- It corresponds to TRANSSODIR, that contains `$(arch)`, so should
contain it as well. -
- ```
parser.c:87:77: error: missing binary operator before token "("
#if JSON_CPU_LITTLE_ENDIAN_64BITS && defined(__has_builtin) && __has_builtin(__builtin_bswap64)
```
https://github.com/ruby/json/commit/fce1c7e84a -
- Closes: https://github.com/ruby/json/pull/888
- Mark it as `inline`.
- Use `RSTRING_GETMEM`, instead of `RSTRING_LEN` and `RSTRING_PTR`.
- Use an inlinable version of `memcmp`.
```
== Parsing activitypub.json (58160 bytes)
ruby 3.4.6 (... -
- Only apply these definitions on 64 bits archs, as it's unclear
if they have performance benefits or compatibility issues on
32bit archs.
https://github.com/ruby/json/commit/ddad00b746 - 06:54 AM Revision 55257b50 (git): Update default gems list at 83c2e3b92e25b766615600913cdca8 [ci skip]
-
06:53 AM Revision 83c2e3b9 (git): [ruby/uri] v1.1.1
- https://github.com/ruby/uri/commit/f1b05c89ab
- 06:45 AM Revision 397bb127 (git): [ruby/uri] Re-allow consecutive, leading and trailing dots in EMAIL_REGEXP
- Effectively reverts commit https://github.com/ruby/uri/commit/788274b180d6 and
https://github.com/ruby/uri/commit/0abac721d8fe.
EMAIL_REGEXP was mostly drawn from WHATWG HTML LS. This spec states that
it intentionally violates RFC 5322 ... -
- https://github.com/ruby/io-wait/commit/c66a90f5b1
- 02:39 AM Revision 0eac75df (git): Bump gacts/run-and-post-run in /.github/actions/setup/directories
- Bumps [gacts/run-and-post-run](https://github.com/gacts/run-and-post-run) from 1.4.2 to 1.4.3.
- [Release notes](https://github.com/gacts/run-and-post-run/releases)
- [Commits](https://github.com/gacts/run-and-post-run/compare/d803f6920a... -
- (https://github.com/ruby/stringio/pull/164)
https://github.com/ruby/stringio/commit/10e991e31d -
- This reverts commit 2c2eaa3103e5cf1cbfc2b16d9db975a9b8a0399a.
-
- (https://github.com/ruby/stringio/pull/162)
https://github.com/ruby/stringio/commit/95a7dd592c - 12:49 AM Revision 9ca94075 (git): ZJIT: Implement register spill (#14936)
-
- (https://github.com/ruby/stringio/pull/161)
https://github.com/ruby/stringio/commit/92656f5c66 -
- (https://github.com/ruby/stringio/pull/160)
https://github.com/ruby/stringio/commit/5034156245 -
12:22 AM Revision 2c2eaa31 (git): [ruby/rubygems] Fix constants in TAR to be frozen
- I would like to use the tar implementation inside a Ractor, but two of
the constants are not frozen. This patch freezes the constants so we
can use it in a Ractor.
https://github.com/ruby/rubygems/commit/0ff4790f4c -
12:17 AM Revision bac6a25a (git): [ruby/rubygems] Remove open-ended and prerelease dependency warnings when building gems
- In general, rubygems should provide mechanism and not policy.
Pessimistic versioning is not universally better, and in many
cases, it can cause more problems than it solves. Rubygems should
not be warning against open-ended versioning w... -
11/03/2025
-
- > * There might be situations when such package passes it test suite, but later it is not usable just due to the metadata.
Yup, this is frustrating and I do understand for some it can justify open ended Ruby constraint. Let's find a s... -
10:43 PM Misc #21657: Question: Is Ruby 4.0 planned for December 2025 or later?
- I agree with @byroot that the upper bound restriction is from my experience very impractical.
As a maintainer of Ruby in Fedora, once we update Ruby in the distribution, we also make sure that all other packages are compatible with it... -
- > This arrogant behavior
Arrogant? I'm just sharing my experience and the same metrics Brooke shared to put my experience in perspective.
> ...
My intent was never to be unfriendly, but if that's how you feel I promise to never int... -
- > Well no. This all stems from you following a very inefficient and user-unfriendly workflow (which you are entirely free to).
@byroot Well yes, I think I have clearly stated that your workflow could be also very inefficient and user-... -
- > this all stems from Ruby being chaotic with the version releases
Well no. This all stems from you following a very inefficient and user-unfriendly workflow (which you are entirely free to).
I maintains 2x as many gems gems as you... -
02:32 PM Misc #21657: Question: Is Ruby 4.0 planned for December 2025 or later?
- **Jean**:
Sorry, I didn't provide enough context to my original statement. Each year -- about ~2 months before Christmas when a new Ruby version drops -- I release a *patch* version of all my gems that accepts the current version and ... -
- @byroot just to make it clear, I'm not saying I'm big fan of the current situation, but I just described how the whole ecosystem is designed right now. It is quite old rusty design, not being revisited for years. I agree with you to disa...
-
- @retro, hard disagree, because if everyone did what you suggest, it means that every year:
- You need to wait on every single one of you dependency to cut a new release with an updated gemspec. When you get hundreds of dependencies th... -
- @byroot I would not suggest this in general. It can work both ways. Having strict Ruby constraints help to revisit gems after new Ruby release and provides confidence to resolver to install only tested versions for given Ruby. What you s...
-
- @bkuhlmann please don't restrict the upper versions unless you know for a fact it's incompatible. This prevent the community from testing early release and fixing the compatibility issues in advance. This practice is slowing the communit...
-
- In any case, I think we should remove the code from `Set.new`. The original ticket was about the behavior of `Range#to_set` vs `Range#to_a`, _not_ about `Set.new`. At least `Set.new`'s behavior should be restored.
- 08:46 PM Revision 4001e81a (git): ZJIT: Inline String#bytesize (#15033)
- Inline the `String#bytesize` function and remove the C call.
-
- These just call to the C functions that do the optimized test but this avoids the side exit.
See https://github.com/ruby/ruby/pull/12123 for the original CRuby/YJIT implementation. - 07:30 PM Revision 16af7279 (git): Avoid taking vm barrier in heap_prepare() (#14425)
- We can avoid taking this barrier if we're not incremental marking or lazy sweeping.
I found this was taking a significant amount of samples when profiling `Psych.load`
in multiple ractors due to the vm barrier. With this change, we get s... -
- https://github.com/ruby/json/commit/3bc1787bd4
-
- Followup: https://github.com/ruby/json/pull/889
https://github.com/ruby/json/commit/591510392a -
-
- Today I get a syntax error for all the examples in both parsers except for `END { break }`. Can this be closed or should `END { break }` still be considered?
-
- ```
== Parsing activitypub.json (58160 bytes)
ruby 3.4.6 (2025-09-16 revision https://github.com/ruby/json/commit/dbd83256b1) +YJIT +PRISM [arm64-darwin24]
Warming up --------------------------------------
after 1.174k... -
- I appreciate the explanation. I openend https://github.com/ruby/prism/pull/3702, it would be nice if you can check that the added tests match your expectation. Basically it is no endless method definition only between `|...|`, which is m...
-
- @hsbt not sure if you got notification for the last message, could you reopen this?
-
- https://github.com/ruby/json/commit/21284ea649
-
- We can share that logic between the two functions.
https://github.com/ruby/json/commit/ac580458e0 -
- https://github.com/ruby/json/commit/31453b8e95
Co-Authored-By: Scott Myron <samyron@gmail.com> -
- The caller already know if the string contains escape sequences
so this check is redundant.
Also stop calling `rstring_cache_fetch` from `json_string_unescape`
as we know it won't match anyways.
```
== Parsing twitter.json (567916 byte... -
- https://github.com/ruby/json/commit/1576ea7d47
-
- `-Wmissing-field-initializers` is in `-Wextra`, not even in `-Wall`. From the [manual](https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html#index-Wmissing-field-initializers) (emphasis mine)
> Warn if a structure’s initializer has ... -
01:10 AM Revision 2da2d33c (git): merge revision(s) e500222de1a8d5e7a844209a7e912b03db8cdf76:
- [PATCH] fix last commit
`th` is gone. -
12:44 AM Bug #21559: Unicode normalization nfd -> nfc -> nfd is not reversible
- Note to potential backporters: https://github.com/ruby/ruby/commit/bd51b20c50 should also be backported.
-
12:03 AM Bug #21662: Variables other than those in the conditional score are replaced.
- nobu (Nobuyoshi Nakada) wrote in #note-1:
> I try to guesstimate your intension by telepathy.
> ...
It looks like a trap.
11/02/2025
-
- I would also like to know if the next Ruby version is 3.5.0 or 4.0.0 precisely because Ruby doesn't follow Semantic Versioning. Having this knowledge sooner rather than later means I can pin my gemspecs to the correct version (and/or ver...
-
- Dan0042 (Daniel DeLorme) wrote in #note-28:
> But to be honest, as @hsbt said in #17473, "I'm happy to use Pathname without require it." and that's really all I am looking for as well.
That's a trap though because without requiring i... -
- Once this bug is fixed I think it would be valuable to run these tests together in the same process, as that's a much stronger test than running tiny pieces of code each in their own process (what the bootstrap test harness does).
Notab... -
- When running the tests from https://github.com/ruby/ruby/blob/master/bootstraptest/test_ractor.rb in a single process, I see segfaults occurring every time on the following tests:
```
# move object with generic ivar
# move object with... -
- 05:06 PM Revision c83a249f (git): pend on Windows for timeouts
- 05:06 PM Revision 89fa15b6 (git): Fix incorrect RUBY_DEBUG range
- 05:06 PM Revision 0116dd5e (git): Make Namespace.root visible not only for debugging
- There are many APIs that expects application codes overwrite global methods.
For example, warn() expects Warning.warn() is overwritten to hook warning
messages.
If we enable namespace, Warning.warn defined in the app code is visible
only... - 05:06 PM Revision fa4c04a7 (git): Use CFUNC namespace only for IFUNC frames, its behavior should be unchanged
- 05:06 PM Revision 12303e2e (git): Fix use of inappropriate debug flag
- 05:06 PM Revision bb62a1cf (git): Add flag to ignore EXPERIMENTAL warnings
- 05:06 PM Revision e89eecce (git): No need to call rb_define_class/module_under_id
- Classes/modules defined in a namespace are defined under ::Object
as usual (as without namespaces), and it'll be set into the const_tbl
of ::Object.
In namespaces, namespace objects' const_tbl is equal to the one of ::Object.
So constant... - 05:06 PM Revision bc77a11b (git): Add basic namespace tests
-
- Isn't this related to this problem with GCC? https://bugs.ruby-lang.org/issues/21655#change-114967
-
04:18 PM Bug #21659: rstring.h error: missing initializer for field ‘len’ of ‘struct RString’ [-Werror=missing-field-initializers] starting in ruby-3.3.10
- wsfulton (William Fulton) wrote in #note-4:
> Fixing this warning, in particular, removes undefined behaviour.
Can you explain what the undefined behavior is? My understanding is that in both C and C++, the behavior is defined to ze... -
- Keep code warning free improves the code quality. Fixing this warning, in particular, removes undefined behaviour. Are the core Ruby developers supportive of these goals and consider a patch to fix this warning and possibly even making i...
-
- Inside `| |` style block parameter, binary operators (example: `1+2`, `1|2`) are rejected because we need to reject ambiguous `|`.
On the right hand side of assignment and endless def, binary operators are allowed, unless we introduce a... -
- It isn't specifically about block locals, `p do |a = def f = 1, b| end` is the same.
`p do |a = def f = 1| 2; b|c end` becomes `p do |a = (def f = 1| 2); b|c end`, so one block local `b` and method body of `c`. The same issue as the f... -
- Fix: https://github.com/ruby/json/issues/873
This allow users to encode binary strings if they so wish.
e.g. they can use Base64 or similar, or chose to replace invalid
characters with something else.
https://github.com/ruby/json/commi... - 02:39 PM Revision 11583f53 (git): [DOC] Update bundled gems list at ee7ce9e6d71c7707d13541e9baa3a5
-
-
-
- We need to allocate at least sizeof(struct RArray) when the array is
embedded on garbage collectors that support slot sizes less than 40 bytes. -
- We need to allocate at least sizeof(struct RString) when the string is
embedded on garbage collectors that support slot sizes less than 40 bytes. -
11:38 AM Bug #21663 (Open): IO#pos will corrupt the file position in specific situation.
- `IO#pos` and `IO#tell` will corrupt the file position in the following situation.
### Situation:
- Running on Windows. (`RUBY_CRLF_ENVIRONMENT` is defined.)
- The filesize is over 0x100000000 bytes.
- The file is opened in text mod... -
-
- I try to guesstimate your intension by telepathy.
binh (binh chau) wrote:
> ```ruby
> ...
You might had expected the third `bar` is a `Bar`.
In fact, the first `bar` in `!bar.is_a?(Bar)` is the attribute.
But the assignment in t... -
- ```ruby
class Bar
attr_accessor :name
end
class Foo
attr_accessor :bar
def initialize(bar=nil)
@bar = bar || Bar.new
end
def bug!
if !bar.is_a?(Bar)
bar = ''
elsif bar.name == 'bar'
end
... - 08:21 AM Revision f827b116 (git): [ruby/erb] Fix typo in documentation
- (https://github.com/ruby/erb/pull/91)
https://github.com/ruby/erb/commit/6bceee7d6e -
- ruby_3_3 commit:609f957ebede1a062b1f34515382e4c306f77444 merged revision(s) commit:ffc69eec0a5746d48ef3cf649639c67631a6a609, commit:0531fa4d6fea100f69f0bac9e03973fe49ecd570.
-
- [PATCH] `struct rb_thread_sched_waiting`
Introduce `struct rb_thread_sched_waiting` and `timer_th.waiting`
can contain other than `rb_thread_t`.
[PATCH] mn timer thread: force wakeups for timeouts -
- ruby_3_3 commit:f0feca1a8495eba2706a7914f0c4f8128c281366.
-
- Remove the code for Boyer-Moore-Horspool search.
Now we are using Sunday's quick search.
https://github.com/k-takata/Onigmo/commit/3d9072419a1578b742a422412d004fd8a54209fd -
- This fixes k-takata/Onigmo#120.
The commit k-takata/Onigmo@9c13de8d0684ebde97e3709d7693997c81ca374b was insufficient.
https://github.com/k-takata/Onigmo/commit/1de602ddff140d91419e3f86dd35c81d7bd2d8e7 -
- Better fix for k-takata/Onigmo#107.
https://github.com/k-takata/Onigmo/commit/85393e4a63223b538529e7095255ce1153c09cff -
- * Multiplication result may overflow 'int' before it is converted to
'OnigDistance'.
* Comparison is always true because code <= 122.
* This statement makes ExprStmt unreachable.
* Empty block without comment
https://github.com/k-taka... -
- Fixes k-takata/Onigmo#92.
This fix was ported from oniguruma:
https://github.com/kkos/oniguruma/commit/257082dac8c6019198b56324012f0bd1830ff4ba
https://github.com/k-takata/Onigmo/commit/b1a5445fbeba97b3e94a733c2ce11c033453af73 -
- This reverts commit 282338f88a8bf0807a7a1d21b06f78abe9de8fac.
It seems that the commit didn't improve the performance.
Revert it to fix k-takata/Onigmo#100.
https://github.com/k-takata/Onigmo/commit/cef834cb3a6e278fa252f52b704c65175a970ac0 -
- E.g.
For the pattern `/----k/i`, optimization was totally turned off.
Make it possible to use the characters before `k` (i.e. `----`) for
optimization.
https://github.com/k-takata/Onigmo/commit/9c13de8d0684ebde97e3709d7693997c81ca374b -
05:05 AM Revision e02c892a (git): Fix indents in Onigmo files to use spaces instead of tabs (#14047) [no ci]
-
- ruby_3_3 commit:30f661951bb19f69cf55388c76ecaf7b0be41079 merged revision(s) commit:12350eb9e0d3317da57b5a37c0c2810946b48850.
- 04:34 AM Revision 2c7fd073 (git): [DOC] Update bundled gems list at 22496e2fe6ea4d354d039927a33458
-
-
- [PATCH] [Bug #21625] Allow io/wait methods with `IO#ungetc` in text mode
-
- Backport would only be needed if the upgrade to Unicode 16.0.0 (see https://bugs.ruby-lang.org/issues/20724) is backported.
-
- Added regression test at https://github.com/ruby/ruby/commit/a122d7a58e91ed6cd531e906cb398688d7cc8b17
and fix at https://github.com/ruby/ruby/commit/e4c8e3544237b8c0efba6b945173dc66552d641c.
Many thanks to Tomoya Ishida for finding thi... -
-
-
11/01/2025
-
-
- The 3.4 backport was merged, but the 3.3 backport doesn't apply cleanly.
-
- I forgot it was introduced in 3.3. It is eligible for backport, but ultimately that decision is up to each branch maintainers.
-
03:51 PM Revision 7ec94afa (git): mn timer thread: force wakeups for timeouts
-
- Sorry, I forgot about your comment on https://github.com/ruby/prism/pull/3674. https://github.com/ruby/prism/pull/3700 will fix this
-
- This is syntax error in parse.y but not in Prism.
~~~ruby
def f a, (b) = 1
~~~
Related to #21623 -
- The following script leaks memory:
10.times do
100_000.times do
"\ufffd".encode(Encoding::US_ASCII, fallback: proc { Object.new })
rescue
end
puts `ps -o rss= -p #{$$}`
end
Before:
450244
... -
- https://github.com/ruby/json/commit/82b030f294
-
- https://github.com/ruby/json/commit/3ea744ad67
-
- There's a new windows-head build that I tried now. Unfortunately I still get this error. The difference seems to be that `bundler-cache: true` specifies a vendor directory for bundle install, which I skipped when applying the change. So ...
-
- ```
../../../../../../ext/json/ext/parser/parser.c:1142:40: warning: suggest parentheses around ‘&&’ within ‘||’ [-Wparentheses]
1142 | if (RB_UNLIKELY(first_digit == '0' && mantissa_digits > 1 || negative && mantissa_digits == 0))... -
- Closes: https://github.com/ruby/json/pull/881
If we encounter a newline, it is likely that the document is pretty printed,
hence that the newline is followed by multiple spaces.
In such case we can use SWAR to count up to eight consecu... -
-
-
- Closes: https://github.com/ruby/json/pull/878
```
== Parsing float parsing (2251051 bytes)
ruby 3.4.6 (2025-09-16 revision https://github.com/ruby/json/commit/dbd83256b1) +YJIT +PRISM [arm64-darwin24]
Warming up ------------------------... -
-
- Split flushing NEWS and bumping up versions
-
- https://github.com/ruby/json/commit/11f4e7b7be
-
- https://github.com/ruby/json/commit/f0150e2944
-
- Encapsulate pointer arithmetic to reduce possibility of mistakes.
https://github.com/ruby/json/commit/8b39407225 -
- https://github.com/ruby/json/commit/1bf405ecc6
-
- https://github.com/ruby/json/commit/2681b23b87
-
- wsfulton (William Fulton) wrote in #note-2:
> This happens when compiling an EXTENSION. We have always used -Wmissing-field-initializers in our extensions and now they are broken. New Ruby releases really MUST NOT introduce new warnings... -
- This happens when compiling an EXTENSION. We have always used -Wmissing-field-initializers in our extensions and now they are broken. New Ruby releases really MUST NOT introduce new warnings for extensions which have to include Ruby.h.
-
- In configure.ac, we add `-Wno-missing-field-initializers` explicitly if available.
Are you adding that `-Werror -Wmissing-field-initializer` by yourself? -
- These are syntax error in parse.y but not in Prism.
~~~ruby
p do |a = def f = 1; b| end
p do |a = def f = 1| 2; b|c end # `|` inside block parameter
~~~
Normal assignment as a default value `p do |a = b = 1| end` is already syntax e... - 06:50 AM Revision 8c190052 (git): Update bundled gems list as of 2025-10-31
-
- That is not an instance variable of an `Encoding` instance, but the instance variable for the encoding of the name.
-
01:23 AM Bug #21654: Set#new calls extra methods compared to previous versions
- mame (Yusuke Endoh) wrote in #note-13:
> My concern is that the expectation to "if it's at all possible to raise early" is a slippery slope with no clear boundary.
I don't think it's slippery at all. There's a very clear difference betw... -
12:18 AM Revision 17a7b4e0 (git): [DOC] Fix typo in `Hash#compare_by_identity` docs