Activity
From 11/15/2025 to 11/21/2025
Today
-
02:49 PM Bug #21696: Performance degradation for long running processes in Ruby 4.0.0-preview2
- Thanks for running the tests!
On my machine (x64 Linux/AMD Ryzen 7 5800X) Roda works fine with the Gemfile from the repository. Yes, most gems are (very) old, I have pinned these in order to have a stable set of gems so comparing resu... -
01:33 PM Bug #21498: Windows - Ruby Overrides C Library APIs thus breaking them
- @cfis, thank you for your comments.
The benefits of source code compatibility with other platforms and the shared use of Windows C runtime resources by ruby and C extensions
enforce a single C runtime for the Windows port.
In this ... -
08:43 AM Revision 5f5da2c2 (git): Fix stdatomic case in rbimpl_atomic_u64_fetch_add
- This was failing on crossruby, likely because HAVE_GCC_ATOMIC_BUILTINS
was true, but HAVE_GCC_ATOMIC_BUILTINS_64 was false. We probably should
have feature detection of 64-bit stdatomics like we do for GCC, but for
now let's keep rbimpl_... -
08:34 AM Revision b9a213f7 (git): Fix timezone of `yesterday`
- Set the `TZ environment variable. `git log` does not recognize UTC
offset in `--before` option, unless full datetime is given. -
- `cmd.exe` splits the command line also by equal signs, not only by
space characters. -
08:32 AM Revision eb11b40b (git): [ruby/rubygems] bin/rubocop -a
- https://github.com/ruby/rubygems/commit/fbf6fb667e
-
-
-
06:26 AM Feature #21701 (Open): Enumerator.produce accepts an optional `size` keyword argument
- Enumerator::Producer#size currently always returns `Float::INFINITY`, and it is not specifiable.
However, a produce sequence is known to be at least finite in many cases, and you can even tell or compute the exact size in some cases.
... - 04:34 AM Revision 451c1209 (git): [ruby/rubygems] Check for file existence before deletion from cache
- (https://github.com/ruby/rubygems/pull/9095)
* Rescue when deleting a non-existent cached gem file
When a gem was in the cache, but another process deletes it first, this
delete command fails.
To work around this, I'm rescuing from Er... -
- If we use "system" variable in BUNDLE_VERSION on Bundler configuration,
we can use bundler version provided by system installation.
But the current logic returns the first activated version of bundler
like 2.7.2. It makes to confuse use... -
04:03 AM Revision 917e77be (git): [ruby/rubygems] Deprecate comparing Gem::Version objects with strings
- Comparing version objects is a huge bottleneck in dependency solvers
(like inside Bundler). I would like to make comparing version objects
cheaper. Right now we support comparing version objects with strings by
trying to coerce the str... -
- https://github.com/ruby/rubygems/commit/c1e3d4d63b
-
03:13 AM Revision 9aa09b46 (git): [ruby/rubygems] Fixed RuboCop offense in Rakefile generated by `bundle gem`
- ```
Offenses:
Rakefile:18:1: C: [Correctable] Layout/EmptyLines: Extra blank line detected.
Diff:
@@ -11,4 +11,5 @@
ext.lib_dir = "lib/test_gem"
end
+
task default: :compile
https://github.com/ruby/rubygems/commit/8c414729df -
- https://github.com/ruby/rubygems/commit/64f92d2da0
-
- 01:19 AM Revision 9764306c (git): Accurate GC.stat under multi-Ractor mode
-
12:52 AM Revision 07ddb0ed (git): ZJIT: Read `iseq->body->param` directly instead of through FFI
- Going through a call to a C function just to read a bitfield was a
little extreme. We did it to be super conservative since bitfields
have historically been the trigger of many bugs and surprises. Let's
try directly accessing them with c... -
- This will make reading the parameters nicer for the JITs. Should be
no-op for the C side. - 12:30 AM Revision aa9e15cb (git): Fix multiple bugs in `IO::Buffer.map` and update its documentation. (#15264)
- - Buffer's size did not account for offset when mapping the file, leading to possible crashes.
- Size and offset were not checked properly, leading to many situations raising EINVAL errors with generic messages.
- Documentation was wrong.
11/20/2025
-
11:38 PM Bug #21685: Unnecessary context-switching, especially bad on multi-core machines.
- Thanks for taking a look at this and coming up with an implementation, this is great.
I haven't really played around with it much but I did read the code and I have a few thoughts:
* There's 1 deferred wait thread per ractor, which isn... -
- @jpl-coconut Please do! This seems like a really good demonstration of the issue and a good start on addressing it. The change is a lot smaller than I expected it to be ❤️. We would only make performance improvements like this to the `ma...
-
- https://github.com/ruby/rubygems/commit/f360af8e3b
-
11:10 PM Revision e15b4e1c (git): Bump default compiler to clang-20 in CI
- clang-18 has a bug that causes the latest Ractor btest to crash.
-
- This was a test case for Ractors discovered that causes MMTk to deadlock.
There is a fix for it in https://github.com/ruby/mmtk/pull/49. -
11:01 PM Revision 010b23a7 (git): [ruby/rubygems] Add support for BUNDLE_LOCKFILE environment variable
- This specifies the lockfile location. This allows for easy support
of different lockfiles per Ruby version or platform.
https://github.com/ruby/rubygems/commit/b54d65bc0a
Co-authored-by: Sutou Kouhei <kou@cozmixng.org>
Co-authored-by: ... -
- This allows for the same behavior as including `lockfile false`
in the Gemfile. This allows you to get the behavior without
modifying the Gemfile, which is useful if you do not control the
Gemfile.
This is similar to the --no-lock optio... -
- This allows you to specify the lockfile to use. This is useful if
you want to use different lockfiles for different ruby versions or
platforms. You can also skip writing the lockfile by using a false
value.
https://github.com/ruby/rubyg... -
10:48 PM Feature #21700: `IO::Buffer.map`: offset argument is "broken" and needs to be made more useful
- This makes sense to me. However it might mean we need to return a slice (which IMHO is probably okay). The alternative is to expose the internal implementation and force the user to deal with all that shit. We should also document clearl...
-
08:25 PM Feature #21700: `IO::Buffer.map`: offset argument is "broken" and needs to be made more useful
- > Currently, offset doesn't change the buffer's size if unspecified, leading to crashes. I will be making a PR addressing this but still.
The pull request in question: https://github.com/ruby/ruby/pull/15264 -
- *This was supposed to be a bug report, but it turned into a feature request halfway through.*
`IO::Buffer.map` supports `size` and `offset` arguments to localize buffer to a specific part of mapped file. While `size` works fine, speci... - 10:35 PM Revision fb28d474 (git): ZJIT: Change the output on redefined method tests to verify the new definition is used
- 10:35 PM Revision 447989e5 (git): ZJIT: Update test names to use the same convention as the HIR tests
- 10:35 PM Revision 36f1ab96 (git): ZJIT: Add tests for `opt_newarray_send` with target methods redefined
- 10:35 PM Revision 7d2f9ab9 (git): ZJIT: Handle display formatting for all defined bops
- 10:35 PM Revision b06dd644 (git): ZJIT: Compile the VM_OPT_NEWARRAY_SEND_HASH variant of opt_newarray_send
- 10:35 PM Revision 604fc059 (git): ZJIT: Rename array length reference to make the code easier to follow
-
- before_exec stops the timer thread, which requires locking the Ractor
scheduler lock. This may deadlock if rb_gc_before_fork locks the VM. -
10:08 PM Revision 55938a45 (git): [DOC] Sort some methods in What's Here
-
-
-
- Previously this held a pointer to the Fiber itself, which requires
marking it (which was only implemented recently, prior to that it was
buggy). Using a monotonically increasing integer instead allows us to
avoid having a free function a... -
- 10:05 PM Revision 8b71234a (git): [ruby/rubygems] Change the logger instance for this spec:
- - With the logger change that is now threadsafe, such code no longer
behaves the same:
```ruby
Bundler.ui.silence do
Bundler.ui.level = 'info'
Bundler.ui.info("foo")
# This used to output something. Now it doesn't.
... - 10:05 PM Revision 409c004a (git): [ruby/rubygems] Make the Bundler logger thread safe:
- - The Logger is not thread safe when calling `with_level`.
This now becomes problematic because we are using multiple
threads during the resolution phase in order to fetch git gems.
https://github.com/ruby/rubygems/commit/380653ae74 - 10:05 PM Revision bd60600d (git): [ruby/rubygems] Run git operations in parallel to speed things up:
- - ### Problem
When you have a Gemfile that contains git gems, each repository will
be fetched one by one. This is extremelly slow.
A simple Gemfile with 5 git gems (small repositories) can take up
to 10 seconds just to fetch the... -
- Found by wbcheck
It seems like here the classext was associated with the class, but it
already had Ruby objects attached.
rb_gc_writebarrier_remember works around that issue, but I suspect if we
enabled autocompaction the values copied... -
-
-
10:02 PM Revision cb9c7a6a (git): [ruby/rubygems] Improve error messages and handling in tests
- This is a first pass to improve the way errors are handled and raised in
bundler's tests. The goal is to clean up tests and modernize them -
these were some obvious areas that could be cleaned up.
- Instead of raising "ZOMG" in the load... -
08:09 PM Revision ba47c2f0 (git): [ruby/prism] Add tests to `regexp_encoding_option_mismatch`
- related to #2667
https://github.com/ruby/prism/commit/44f075bae4 - 08:06 PM Revision 0b4420bf (git): [ruby/prism] Use memmove for overlapping memory ranges
- Fixes https://github.com/ruby/prism/pull/3736.
https://github.com/ruby/prism/commit/1f5f192ab7 -
07:47 PM Revision a8f269a2 (git): ZJIT: Deduplicate successor and predecessor sets (#15263)
- Fixes https://github.com/Shopify/ruby/issues/877
I didn't consider the ability to have the successor or predecessor sets having duplicates when originally crafting the Iongraph support PR, but have added this to prevent that happening i... -
05:23 PM Revision 48027256 (git): [ruby/json] Remove unused symbols
- https://github.com/ruby/json/commit/9364d0c761
- 04:32 PM Revision f8cb9f32 (git): ZJIT: Put optional interpreter cache on both GetIvar and SetIvar
- 04:32 PM Revision 0b6daad6 (git): ZJIT: Fix pointer types for SetInstanceVariable
-
- Applied in changeset commit:git|41b8e440e7f2c5d3d1c1a9644de4bdc06a343724.
----------
Support backwards compatibility for Set subclasses
For subclasses from Set, require `set/subclass_compatible`, and
extend the subclass and include a m... -
- Applied in changeset commit:git|41b8e440e7f2c5d3d1c1a9644de4bdc06a343724.
----------
Support backwards compatibility for Set subclasses
For subclasses from Set, require `set/subclass_compatible`, and
extend the subclass and include a m... -
- For subclasses from Set, require `set/subclass_compatible`, and
extend the subclass and include a module in it that makes it more
backwards compatible with the pure Ruby Set implementation used
before Ruby 4.
The module included in the ... -
- `rb_zjit_option_enabled_p` seems no longer used/defined since
ruby/ruby#f84bbb423836d9d0d018b8ab71ecceb5868fd5be. -
- `realloc` is not guaranteed to return the same address when shrinking.
-
- Now win32/registry depends on fiddle, and its conversion is complex
and too generic for the purpose of resolv.
https://github.com/ruby/resolv/commit/bd24870d2d -
04:54 AM Feature #21675: Advent of Pattern Matching
- We briefly discussed this at the monthly dev meeting. While we haven't reached a consensus on whether the scope of pattern matching should be expanded in general, I would like to share one specific concern that came up during the discuss...
- 02:42 AM Revision 4c752508 (git): Update default gems list at 0653a01ada3c3c286de128074b58b3 [ci skip]
-
- https://github.com/ruby/rubygems/commit/9be811c01a
-
02:07 AM Feature #19979: Allow methods to declare that they don't accept a block via `&nil`
- Given that the next release will be Ruby 4.0, and that the Syntax moratorium should be over.
Do you think it's possible you could reconsider adding this feature? -
01:07 AM Feature #21695: Optimizing Ruby performance with Ruby itself instead of Rust
- Currently, my goal is to rewrite the C++ code I have into C. To achieve this, I am creating a set of macros in Ruby on top of C that run at compile time, and the output of these macros is C code. Here is an example of what I am doing, ...
-
- This test has been reliably failing on recent trunk versions.
See: <https://github.com/ruby/ruby/actions/runs/19519712433/job/55880266450#step:14:120>
11/19/2025
-
11:28 PM Bug #21685: Unnecessary context-switching, especially bad on multi-core machines.
- How can I help move the discussion forward? Is it better if I create a PR and we discuss there?
-
- Applied in changeset commit:git|fa02d7a01f5e7516de8eb3c7f92ec75c50c06e3f.
----------
Implement heap_live_slots in GC.stat_heap
[Feature #20408] -
- [Feature #20408]
-
- [Feature #20408]
-
- [Feature #20408]
-
- As can be seen in vm_block_handler_verify(), VM_BLOCK_HANDLER_NONE is
not a valid argument for vm_block_handler(). Store nil in the profiler
when seen instead of crashing. -
-
- ## Components
This PR adds functionality to visualize HIR using the [Iongraph](https://spidermonkey.dev/blog/2025/10/28/iongraph-web.html) tool first created for use with Spidermonkey.
## Justification
Iongraph's viewer is (as mention... - 10:54 PM Revision 63a6290c (git): [DOC] Update yjit.md to use a different email
-
-
10:37 PM Revision ba2b97a9 (git): Update to ruby/spec@6e62695
-
- I ran the benchmarks, but unfortunately some of the gems are out of date for anyone looking into this. The Gemfile needs:
```ruby
gem "sqlite3" # instead of "1.4.2"
gem "nio4r", "2.7.5" # instead of "2.5.8"
```
I also had to create a c... -
- Previously calling a singleton method with invalid arguments would give:
RuntimeError: Incompatible locations
This is because `join` wants the operator to come before the location
https://github.com/ruby/error_highlight/commit/449... -
- kbrock (Keenan Brock) wrote:
> I'm hoping we can make `ruby_thread_has_gvl_p` a public method and no longer experimental.
This has been done as part of #20877 in https://github.com/ruby/ruby/pull/11975/commits/979c05185270d088b4531a6... -
08:58 PM Revision 4a1af72a (git): ZJIT: Count all calls to C functions from generated code (#15240)
- lobsters:
```
Top-20 calls to C functions from JIT code (79.9% of total 97,004,883):
rb_vm_opt_send_without_block: 19,874,212 (20.5%)
rb_vm_setinstancevariable: 9,774,841 (10... -
-
08:03 PM Misc #21699 (Open): Updating CAPI docs
- Hello.
How often are the CAPI docs hosted at https://docs.ruby-lang.org/capi/en/master/ refreshed?
It looks like it's been more than 2 weeks since they were last refreshed.
Last revision 4a3d8346a6d0e068508631541f6bc43e8b154ea1 is... -
05:31 PM Bug #21498: Windows - Ruby Overrides C Library APIs thus breaking them
- Hi YO4 - thank you for your last comment - I appreciate it. I too need to strive for better communication, because clearly I am not communicating very well. I hope we can all assume the best intentions in others, and realize communicatio...
-
- Looking back now, I realize my recent comments were not good communication.
My comments were overly aggressive.
Particularly noteworthy the final statement was made in a manner that seemed to represent the community, cutting off @cif... -
03:40 PM Bug #21698: Should simple ArgumentError refer to Object?
- Thanks for clarification 👍
-
- Displaying the method owner class was decided in #19117. The fact that it was not displayed for `"wrong number of arguments"` errors was simply a bug due to an oversight.
We were aware at the time of #19117 that this change would be i... -
- #19117 was closed more then a year ago, therefore I'd assume it was part of Ruby 3.4. But I am pointing out change between Ruby 3.4 and 4.0discussing. But there are also exceptions such as #20275, so maybe this is the case for another on...
-
- It's just #19117 and it's intentional.
That feature is missing a NEWS entry though, @mame could you add one? -
- Having this simple test file:
~~~
$ cat test.rb
def foo(x, y)
end
foo(1)
~~~
This is the Ruby 3.4 behavior:
~~~
$ ruby --disable-gems test.rb
test.rb:1:in 'foo': wrong number of arguments (given 1, expected 2) (Argumen... -
-
-
11:36 AM Revision 85cd08e4 (git): Update to ruby/spec@2e11d2a
-
-
-
- Applied in changeset commit:git|8986115e0a2a989f2b2ea5945f02c7a13989d640.
----------
[Bug #21697] Keep revision.h outside VCS -
- In 4.0.0-preview2 package, `nmake up` generate the broken `revision.h`.
I could reproduce this:
```
PS C:\Users\hsbt\Downloads> Invoke-WebRequest https://cache.ruby-lang.org/pub/ruby/4.0/ruby-4.0.0-preview2.zip -OutFile ruby-4.0.0... -
07:48 AM Feature #21311: Namespace on read (revised)
- Is there any reason this feature couldn't be released as a gem? It seem like it has evolved far beyond it's original intention (including name change to "Box"), and I worry that it's inclusion in core will lead to inconsistent usage and ...
-
- https://github.com/ruby/rubygems/commit/6cc7d71dac
-
- This reverts commit 8986115e0a2a989f2b2ea5945f02c7a13989d640.
`RELEASE_DATE` including `YEAR`, `MONTH`, `DAY` are mandatory, while
`REVISION` is not. -
-
-
-
- 02:52 AM Revision 3ee08c8d (git): Bump actions/checkout in /.github/actions/setup/directories
- Bumps [actions/checkout](https://github.com/actions/checkout) from 5.0.0 to 5.0.1.
- [Release notes](https://github.com/actions/checkout/releases)
- [Changelog](https://github.com/actions/checkout/blob/main/CHANGELOG.md)
- [Commits](http... -
- Ruby 2.7 supports MSVC 12.0/_MSC_VER 1800 or later.
https://github.com/ruby/etc/commit/6f4404ec88 -
- https://github.com/ruby/etc/commit/d047bb6856
-
- In rb_gc_impl_before_fork, it locks the VM and barriers all the Ractors
before calling mmtk_before_fork. However, since rb_mmtk_block_for_gc is
a barrier point, one or more Ractors could be paused there. However,
mmtk_before_fork is not ... -
- We need the VM barrier in rb_gc_impl_before_fork to stop the other Ractors
because otherwise they could be allocating objects in the fast path which
could be calling mmtk_add_obj_free_candidate. Since mmtk_add_obj_free_candidate
acquires... -
- Visual C++ 2015 (14.0):
- _MSC_VER: 1900
- MSVCRT_VERSION: 140 -
- Visual C++ 2013 (12.0):
- _MSC_VER: 1800
- MSVCRT_VERSION: 120 -
- Visual C++ 2010 (10.0):
- _MSC_VER: 1600
- MSVCRT_VERSION: 100 -
- Visual C++ 2008 (9.0):
- _MSC_VER: 1500
- MSVCRT_VERSION: 90 -
- Visual C++ 2005 (8.0):
- _MSC_VER: 1400
- MSVCRT_VERSION: 80 - 02:01 AM Revision 1979f8c0 (git): [ruby/prism] Add docs for super nodes
- https://github.com/ruby/prism/commit/69abcdbb18
- 01:37 AM Revision 3b953917 (git): [ruby/rubygems] Warn users that `bundle` now display the help:
- - In https://github.com/ruby/rubygems/commit/31d67ecc056fb5a9193bc66a6e69e21576a87702
we enforced the new behaviour where running `bundle` no longer
installs gems but displays the help.
Users now have a way to configure their prefe... - 01:27 AM Revision 83b0cfe1 (git): [ruby/rubygems] Handle BUNDLER_VERSION being set to an empty string
- This is useful, in case you're using Docker, and an upstream
Dockerfile sets BUNDLER_VERSION to something you don't want.
It's impossible to unset it... only override to be the empty
string.
https://github.com/ruby/rubygems/commit/ffa3e... -
- Previously any T_OBJECT with >= 94 IVARs would crash during compaction
attempting to make an object too large to embed. - 12:47 AM Revision 4423facb (git): [ruby/rubygems] Add `--ext=go` to `bundle gem`
- (https://github.com/ruby/rubygems/pull/8183)
* Add new gem templates
* Add `--ext=go` in `bundle gem`
* Add setup-go to .github/workflows/main.yml
* Embed go version in go.mod
* Use go in bundler CI
* Add example method to template... -
11/18/2025
-
- Name contradictory now, and we have other tests testing the same thing.
-
- `Gem::Installer.instance_methods(false).include?(:generate_plugins)` is
63x slower than `Gem::Installer.method_defined?(:generate_plugins)` in a
microbenchmark. The latter is a direct lookup, whereas the former will
create an array, whic... -
10:45 PM Revision d5d12efd (git): [ruby/json] parser.c: Remove unued JSON_ParserStruct.parsing_name
- https://github.com/ruby/json/commit/ab5efca015
-
10:32 PM Revision f3f3e768 (git): Extract `KW_SPECIFIED_BITS_MAX` for JITs (GH-15039)
- Rename to `VM_KW_SPECIFIED_BITS_MAX` now that it's in `vm_core.h`.
-
09:57 PM Revision cbe65ebb (git): ZJIT: Skip empty counter sections in stats
-
-
- jeremyevans0 (Jeremy Evans) wrote in #note-10:
> I've submitted a PR that implements backwards compatibility for Set subclasses, without affecting Set itself: https://github.com/ruby/ruby/pull/15228
Thank you, I think it's a good sol... -
03:24 AM Bug #21375: Set[] does not call #initialize
- I've submitted a PR that implements backwards compatibility for Set subclasses, without affecting Set itself: https://github.com/ruby/ruby/pull/15228
@Eregon can you try it with concurrent-ruby and see if it resolves your issue? - 07:24 PM Revision ff2d2fc1 (git): YJIT: omit single ractor mode assumption for `proc#call` (#15092)
- The comptime receiver, which is a proc, is either shareable or from this
ractor so we don't need to assume single-ractor mode. We should never get
the "defined with an un-shareable Proc in a different ractor" error. -
06:58 PM Feature #21693: Allow calling any callable object as a method
- You're very close to valid Ruby with
```
class Greeter
def call(name)
puts "Hello, #{name}!"
end
end
hello = Greeter.new
hello.("World")
```
(Note the 1-character difference here.) Doing `hello.("World")` will a... -
- It would be a gigantic incompatibility so there is no chance for that specific syntax, as your example would call method `hello` before and no longer with your proposal.
Just
```ruby
p = 42
p(43)
```
would break for example with th... -
-
- * When writing to an object, the receiver should be checked if it's frozen,
not the value, so this avoids an error-prone autocomplete. -
- * Add Insn::StoreField and Insn::WriteBarrier
-
- What are you proposing exactly? A "tool" is very vague.
Do you mean a JIT compiler for Ruby written in Ruby? Something else?
Are you interested in speeding up ERB specifically, or Ruby in general?
Why do you think it would achieve bet... -
02:09 AM Feature #21695: Optimizing Ruby performance with Ruby itself instead of Rust
- I am interested. What kind of API do you need? Since bare memory allocation could break the whole app (leads to DoS), it's not that easy to disclose general purpose memory allocation (that's the reason YJIT, RJIT etc. has their own alloc...
-
04:35 PM Revision f84bbb42 (git): ZJIT: add support for lazy `RubyVM::ZJIT.enable`
- This implements Shopify#854:
- Splits boot-time and enable-time initialization,
tracks progress with `InitializationState` enum
- Introduces `RubyVM::ZJIT.enable` Ruby method for
enabling the JIT lazily, if not already enabled
- I... -
- * This can catch subtle errors early, so avoid a fallback case and
handle every instruction explicitly. -
- ufuk (Ufuk Kayserilioglu) wrote in #note-1:
> Did 3.5-preview1 have the same behaviour? If not, could you bisect between the two releases to find the commit that changed the behaviour?
Just ran the test, only for the Roda app. Result... -
12:06 PM Bug #21696: Performance degradation for long running processes in Ruby 4.0.0-preview2
- Did 3.5-preview1 have the same behaviour? If not, could you bisect between the two releases to find the commit that changed the behaviour?
-
- When running my RubyMeasureResponsetime tool (https://github.com/easydatawarehousing/ruby_measure_responsetime) on Ruby 4.0.0-preview2, a slow but steady performance degradation is measurable.
Both the Rails and the Roda based test appl... -
12:49 PM Revision 522b7d82 (git): [ruby/openssl] ssl: fix test_pqc_sigalg on RHEL 9.7
- RHEL 9.7 ships OpenSSL 3.5.1 with ML-DSA support, but it is disabled
for TLS by default, according to the system configuration file:
/etc/crypto-policies/back-ends/opensslcnf.config
Specify SSLContext#sigalgs to override the default lis... -
12:46 PM Bug #21669: Thoroughly implement void value expression check
- I have an implementation in https://github.com/ruby/prism/pull/3728 (maybe needs more tests)
How about waiting on Ruby 4.1 for this? We can then introduce it early in the development cycle to find potential compatibility concerns. pre... -
- The discussion you cited seemed to me to be ignoring the actual problem by pinning the cause solely on ruby's behavior.
You used it to solidify your argument within the ruby community, and even tried to borrow Microsoft's authority to d... -
- Hi @YO4 - thank you for your reply. I absolutely agree a cooperative approach is best. I am saddened / quite surprised to hear that you think I have criticized someone. I certainly had no intention to do that. If I did then I apologize.
... -
- Ruby has made significant efforts to achieve cross-platform support at the C level.
This includes compensating for the lack of C runtime libraries.
The statement you quoted is taken out of context and that seems unfair.
The override... -
- https://github.com/ruby/rubygems/actions/runs/19458155903/job/55676075439?pr=3857
```
-Did you mean 'methods' or 'method'?
+Could not find gem 'methosd'.
+Did you mean 'method' or 'methods'?
```
https://github.com/... -
- https://github.com/ruby/rubygems/commit/d604c1d1cb
-
- https://github.com/ruby/rubygems/commit/fee8dd2f08
-
- https://github.com/ruby/rubygems/commit/40ace48651
- 10:38 AM Revision 2c169e15 (git): [ruby/rubygems] More tests
- https://github.com/ruby/rubygems/commit/210fa87f65
- 10:38 AM Revision 55afec32 (git): [ruby/rubygems] fix tests
- https://github.com/ruby/rubygems/commit/1dc669a0ab
- 10:38 AM Revision dd6ccb44 (git): [ruby/rubygems] Progressively enhance if DidYouMean is available
- https://github.com/ruby/rubygems/commit/a02353fb96
- 10:38 AM Revision 6aa16246 (git): [ruby/rubygems] Rubocop
- https://github.com/ruby/rubygems/commit/a6bc30a827
- 10:38 AM Revision f3e8bc87 (git): [ruby/rubygems] use DidYouMean::SpellChecker for gem suggestions
- replaces Bundler::SimilarityDetector with DidYouMean::SpellChecker
https://github.com/ruby/rubygems/commit/959bea1506 -
-
- This might be related to the `--prefer-local` issue:
https://github.com/ruby/rubygems/issues/8750
I suspect there is some problem with the arches of the gems. But I might be wrong. -
- Ah, it seems this happens on Ubuntu 24.04, without a `Gemfile.lock`, and when the `Gemfile` does not list the indirect dependencies.
I found three workarounds.
### 1\. Add a `Gemfile.lock`
If a proper `Gemfile.lock` is present, ... - 08:43 AM Revision 85abc59c (git): [DOC] Add documentation about Ruby's VM stack
- 08:20 AM Revision f272aabb (git): [ruby/json] Use #if instead of #ifdef when checking for JSON_DEBUG so debugging code is not generated when JSON_DEBUG=0.
- https://github.com/ruby/json/commit/4f1adb10d3
-
- https://github.com/ruby/resolv/commit/599f78c451
- 06:52 AM Revision ccdf83f1 (git): Update bundled gems list as of 2025-11-18
-
-
-
- https://github.com/ruby/rubygems/commit/778426fb73
-
- https://github.com/ruby/rubygems/commit/4e02243f66
-
- https://github.com/ruby/rubygems/commit/f8fe7a5208
-
- https://github.com/ruby/rubygems/commit/b6deff99c9
-
- https://github.com/ruby/rubygems/commit/07f2daf51e
-
- https://github.com/ruby/rubygems/commit/b237f759b0
-
- It seems like we were trying to deprecate passing `nil` to
Gem::Version.new. This breaks existing code, and I don't think there is
a good reason to deprecate this usage.
I believe what we want to prevent is the following code:
```ruby... -
- Eregon (Benoit Daloze) wrote in #note-13:
> jeremyevans0 (Jeremy Evans) wrote in #note-11:
> ...
I submitted a pull request that implements what @knu suggested: https://github.com/ruby/ruby/pull/15228
It allows subclasses to use the... - 02:34 AM Revision b5d0aac9 (git): Bump actions/checkout from 5.0.0 to 5.0.1
- Bumps [actions/checkout](https://github.com/actions/checkout) from 5.0.0 to 5.0.1.
- [Release notes](https://github.com/actions/checkout/releases)
- [Commits](https://github.com/actions/checkout/compare/v5...v5.0.1)
---
updated-dependen... -
-
- The regexp used in EnvUtil.current_parser did not allow square brackets
for feature names. Modular GC uses square brackets for the GC name (such
as +GC[mmtk]). -
- https://github.com/ruby/rubygems/commit/3471646d43
11/17/2025
-
- https://github.com/ruby/rubygems/commit/a49d315ecd
-
- Hi @mame - Thank you for your response and I am sorry that you feel that way, that was not my intention. I very much appreciate your help and our discussion. Maybe we can try again?
I see the problem is that Ruby exports overridden C ru... -
- I feel your communication on this ticket suffers from what is known as the [XY problem](https://en.wikipedia.org/wiki/XY_problem). You are focusing on your proposed solution ("remove the override"), without first detailing the original p...
-
- I agree having one UCRT is best. But the very first sentence in the paragraph I linked to says:
> Every executable image (EXE or DLL) can have its own statically linked CRT, or can dynamically link to a CRT. The version of the CRT stati... -
- I commend YO4-san for their insight into the debug/release build discrepancy.
I'm now convinced the root cause is loading multiple runtimes. Therefore, I agree with @nobu's fix: we should prohibit loading multiple runtime versions, no... -
- @nobu - When developing an extension it can be very helpful to use a debug build. A runtime check would prohibit that, unless you also make a debug version of ruby. But that seems like a lot of unnecessary overhead.
In addition, it wo... -
- cfis (Charlie Savage) wrote in #note-21:
> Summary - don't pass memory/crt objects across boundaries. By overriding `fclose` and other UCRT functions Ruby causes this to happen.
Yes for the former half.
It is the reason we embed the CRT... -
- Ah yes, you are correct @YO4.
I used MSVC 2026 to build both Ruby and the extension (a dll). By default Ruby builds with /Md. This is the generated `Makefile.sub` in the win32 directory:
```
!if !defined(RUNTIMEFLAG)
RUNTIMEFLAG = -M... -
- When calling getnameinfo we spawn a thread because it may do a slow,
blocking reverse-DNS lookup. Spawning a thread is relatively fast (~20µs
on my Linux machine) but still an order of magnitude slower than when
getnameinfo is simply tra... -
- I am using ERB on top of C for a game I am making now. We can use the tool to improve the productivity, memory safety, and runtime speed for Ruby. It can also replace a Rust JIT compiler for better results. The only problem is that it...
-
- Initially reported on Reddit by easydwh https://old.reddit.com/r/ruby/comments/1ozhupr/is_a_ruby_segmentation_fault_a_bug_if_you_are/
Repro:
```ruby
class BasicObject
def basic_call_no_super
super
end
end
"test".bas... -
06:53 PM Bug #21692: Basic gems like `date` require a compiler
- mame (Yusuke Endoh) wrote in #note-1:
> I think what you want is `bundle install --prefer-local`. (I wonder why this option is not the default.)
Unfortunately this does not seem to work for me.
```
$ gem list | grep date
date (default:... -
- I think what you want is `bundle install --prefer-local`. (I wonder why this option is not the default.)
-
- I'm trying to use Ruby in a high security environment which does not allow a compiler to be installed. This was easy years ago when all of standard library was included with Ruby itself and I could use `apt install ruby`. But as the stan...
-
- I noticed some of the keys have been gone or renamed for a while.
-
06:29 PM Feature #21693 (Open): Allow calling any callable object as a method
- Callable objects are popular in Ruby. A very common pattern of Service objects with a single public `call` method can be found in many Rails apps, too.
I propose to extend syntax so that adding `()` to any object that has a `call` met... - 06:01 PM Revision f9e31f43 (git): [ruby/time] Defer default year for zone_offset until needed
- While parsing ISO 8601 timestamps like `2025-11-13T19:34:23.681726000Z`
I noticed that a lot of time was spent to compute a default `year`
argument to `zone_offset`. For ISO 8601, that year is never used, as
all valid time zone designato... - 03:53 PM Revision 371a295e (git): Fix improper termlen fill in `str_duplicate_setup_embed`.
- When term len != 1 (for example: Encoding::UTF32BE), term fill is wrong size.
-
- https://github.com/ruby/prism/commit/b00d098f9a
-
- Since https://github.com/ruby/ruby/pull/15212 these are proper syntax
errors, so no need to handle this explicitly anymore.
Also updated the example in the docs for this -
01:30 PM Revision a0ef2366 (git): remove old code
-
01:27 PM Bug #21168: Prism doesn't require argument parentheses (in some cases) when a block is present but parse.y does
- PR: https://github.com/ruby/ruby/pull/15217
-
- When using the binary read method of IO on a file containing “\r\n”, it should read “\r”, but some methods do not function correctly on Windows.
```ruby
File.binwrite("crlf.txt", "\r\n")
methods = {
readn: proc { |f| f.read(1) },... -
- I made PR [#15216](https://github.com/ruby/ruby/pull/15216) for main issue.
After some investigation, it appears the case with #note-1 is mainly related to the behavior of eof(), so it seems appropriate to address it in #21634. -
-
-
11:15 AM Misc #21690 (Open): Inconsistent `rb_popcount64()` definition
- The `rb_popcount64()` function is defined both in `internal/bits.h` and `parser_bits.h`, but the definition of these functions is different on the `#else` branch.
```c
// internal/bits.h
static inline unsigned int
rb_popcount64(uin... -
- I should also mention for the concurrent-ruby case that using `::Set` on Ruby 4+ is not an option because it's not thread-safe enough:
https://github.com/ruby-concurrency/concurrent-ruby/issues/1093#issuecomment-3540742547 -
- This change also breaks concurrent-ruby: https://github.com/ruby-concurrency/concurrent-ruby/actions/runs/19414490796/job/55540724035
https://github.com/ruby-concurrency/concurrent-ruby/issues/1093
The code relying on it seems pretty re... -
- jeremyevans0 (Jeremy Evans) wrote in #note-11:
> knu (Akinori MUSHA) wrote in #note-10:
> ...
Yes, this sounds like a good way, it avoids breaking code and keeps performance optimal for Set.
The only downside is some extra complexity ... -
- * [Bug #21375] Set[] does not call #initialize (eregon)
* This broke concurrent-ruby and there is no good fix.
* I believe Set[] should call #initialize, it's simple and avoids a major breaking change. The performance impact seems in... -
- # The next dev meeting
**Date: 2025/12/11 13:00-17:00** (JST)
Log: *TBD*
- Dev meeting *IS NOT* a decision-making place. All decisions should be done at the bug tracker.
- Dev meeting is a place we can ask Matz, nobu, nurse and o... -
-
- https://github.com/ruby/actions/actions/runs/19420968008/job/55558315338
```
1) Error:
TestCommitEmail#test_sendmail_encoding:
Test::Unit::ProxyError: No such file or directory - git
/home/runner/work/actions/actions/ruby-... -
- When there are nested capture variables inside of a pattern match
that has an alternation pattern, it is a syntax error. Currently it
only adds a syntax error when it is at the top level of the pattern. -
- Cross referencing: https://github.com/socketry/async/issues/424
-
11/16/2025
- 11:49 PM Revision 4870fbd0 (git): vm_dump.c: unbreak unwind on powerpc (#14650)
-
-
-
-
-
-
- @nobu, I had not correctly understood the 'rt' case in #note-2.
In the case where 'universal_newline: true' attribute, Ctrl-Z is not interpreted as an EOF, so I believe the behavior you pointed out is correct.
```
> ruby -ve "open('eo... -
02:31 PM Misc #21688 (Open): Ruby::Box maturing path
- A quick couple of questions in the light of upcoming Ruby 4.0 release and `Ruby::Box` being available there (even if as an experimental feature):
1. Should the documentation of the class be available as RDoc? I mean, as standard class... -
- Here is additional issue.
When an intermediate EOF character met, reading operation reports EOF but sometimes IO#eof? does not.
```ruby
require 'tempfile'
Tempfile.open do |f|
str = "0123456789\x1A"
f.write(str + "x"*(1024_0 ... -
- If `rb_win32_fclose` calls ucrt's `fflush`/`fclose`, it's questionable why it doesn't work. Is it truly calling the same ucrts?
The screenshot suggests the C++ code is built in Debug build. From the C++ code, the the ucrt debug DLL migh... -
- As proposed in this issue, resolving this problem required determining a solution strategy across multiple methods with considering compatibility.
While I'm only an external contributor and not a core team member, I'll boldly say that...
11/15/2025
-
- Thank you @mame an @nobu for looking into this.
@mame I don't know why the override doesn't work. The calling C++ code is from `Microsoft Visual Studio\18\Insiders\VC\Tools\MSVC\14.50.35717\include\__msvc_filebuf.hpp`:
``` cpp
_EXPO... -
- cfis (Charlie Savage) wrote in #note-15:
> When I step through with a debugger, this line forces an early return from the function. So fclose is never called.
> ...
Thank you for debugging. The early return (not calling `fclose` when `... -
- I gave a try to @nobu's code above (https://bugs.ruby-lang.org/issues/21498#note-9).
That results in a linker error:
``` bash
linking shared-object -test-/gvl/call_without_gvl.so
Creating library call_without_gvl-x64-mswin64_140.l... -
- I have attached the "good" file and the "bad" file. The good file is generated by rewriting the extension to not use fclose, and thus not have the call go through `rb_w32_fclose`. The bad file is generated by using flcose which does end ...
-
- When I step through with a debugger, this line forces an early return from the function. So fclose is never called.
https://github.com/ruby/ruby/blob/master/win32/win32.c#L6627
``` cpp
if (fflush(fp)) return -1;
```
-
-
06:39 PM Bug #21686: In combination with IO#ungetbyte, the write position may become unpredictable.
- #20919 may be related. If the PR for this issue is accepted, it likely determines how that issue should be addressed. There's already a PR that is several months old associated with it.
-
- I made [PR #15204](https://github.com/ruby/ruby/pull/15204)
## Changed and Unchanged
The behavior of this PR is listed below.
Since IO#ungetc has a different file position behavior when combined with encoding conversion, IO#getbyt... -
- In the current implementation, using `IO#ungetbyte` can cause `IO#pos` to become negative.
Writing to the same file descriptor in this state will result in unexpected write positions.
```ruby
require 'tempfile'
Tempfile.open do |... -
- In Windows environment, when opening a file with the "r", encountering an EOF character (Ctrl-Z, "\x1A") during reading causes the IO to report END-OF-FILE.
```ruby
require 'tempfile'
Tempfile.open do |f|
str = "0123456789\x1A"... -
- (https://github.com/ruby/stringio/pull/174)
Method #seek deserves (and will get) documentation independent of that
in class IO.
Meanwhile, the link should go someplace sensible and useful.
https://github.com/ruby/stringio/commit/d0265... -
- (https://github.com/ruby/stringio/pull/171)
https://github.com/ruby/stringio/commit/95a111017a -
- (https://github.com/ruby/stringio/pull/165)
Adds to "Position": pos inside a character.
Makes a couple of minor corrections.
---------
https://github.com/ruby/stringio/commit/ff332abafa
Co-authored-by: Sutou Kouhei <kou@cozmixng.org> -
- It has been merged into `doc/ruby/options.md` with
`field_processing.md` at ruby/ruby#10138. -
-
01:18 AM Revision f1765cd4 (git): [Doc] Remove leftover references to namespace from box.md
- And fix the indentation a little bit, since `box` is one character
longer than `ns`. -
12:34 AM Revision d7369f02 (git): ZJIT: Add individual tests for complex arg pass counters
- Make it easier to see what happens when one is changed.