Activity
From 05/10/2025 to 05/16/2025
05/16/2025
-
10:46 PM Feature #21311: Namespace on read (revised)
- bughit (bug hit) wrote in #note-99:
> Can today a single program have an object that is value unequal to itself.
I mean by the default and intuitive definition of value equality which is comparing instance state. Obviously you can de... -
- fxn (Xavier Noria) wrote in #note-98:
> Think about it, can today a single program have objects with the same ID that are not equal? Yes! Absolutely it can!
Object id in this conversation is a proxy for identity, so the correct question... -
09:53 PM Feature #21311: Namespace on read (revised)
- @bughit I totally see what you mean, and I had myself this very objection up in the thread.
While my post had the aim of spreading the word and make namespaces accessible to a broader audience, I have personally still not reached a de... -
- fxn (Xavier Noria) wrote in #note-96:
>
> ...
This argument would work for other languages where classes are magical templates that can be endowed with all kinds of special traits. But this is ruby where a class is an object, an instanc... -
- bughit (bug hit) wrote in #note-95:
> The example I used to illustrate the absurdity of this is two string references to the "same" string object but having different string values.
I am not sure I follow. The edge case in which th... -
- fxn (Xavier Noria) wrote in #note-90:
>
> ...
There are things that new features should not do, like redefining and corrupting the concept of a object.
> 2. @tagomoris made the good point that in _today_'s Ruby, refinements already... -
07:32 AM Feature #21311: Namespace on read (revised)
- Eregon (Benoit Daloze) wrote in #note-93:
> vo.x (Vit Ondruch) wrote in #note-88:
> ...
Just to be clear, I don't agitate for loading RubyGems, just wonder where the line is. Having clean pristine environment is very unique opportunity... -
10:18 PM Bug #21340: Bump autoconf version to properly handle C23 bool/stdbool defines
- google-protobuf is also failing to compile native extension on windows ruby head and 3.4.4 with msys2 gcc 15.
```
In file included from
D:/rubyinstaller-head-x64/include/ruby-3.5.0+1/ruby/defines.h:74,
from
D:/rubyinstaller-head-x64/inc... -
08:54 PM Feature #14916: Proposal to add Array#===
- This is probably unnecessary now that pattern matching exists.
-
- @matz maybe too obvious, but how about `zip_map`?
-
- If approved, I can add `String#ensure_prefix` and the bang versions of those methods.
-
## Problem
Ensuring a string has a specific suffix or prefix is a common operation in many applications.
Bundler itself uses it:
https://github.com/rubygems/rubygems/blob/d409ec8b5fc647fabe30e37e17cd1ea857634f6b/bundler/lib/bund...-
06:47 PM Bug #21345: crash on evaluating 'a=>a,*,'
- This is most probably a Prism bug, since it seems to work fine with `parse.y` as the parser:
```shell
$ ruby --parser parse.y -ve 'a=>a,*,'
ruby 3.4.4 (2025-05-14 revision a38531fd3f) [arm64-darwin24]
-e:1: syntax error, unexpected end-o... -
04:19 PM Bug #21345 (Closed): crash on evaluating 'a=>a,*,'
- ```
ruby -e 'a=>a,*,'
-e: [BUG] Unexpected node type in pattern matching expression: PM_IMPLICIT_REST_NODE
ruby 3.4.3 (2025-04-14 revision d0b7e5b6a0) +PRISM [arm64-darwin24]
-- Crash Report log information ----------------------... -
06:16 PM Revision 22c09135 (git): rb_copy_generic_ivar: reset shape_id when no ivar are present
-
- The finalizer table can't be read nor modified without the VM lock.
-
- Fix a regression introduced by: https://github.com/ruby/ruby/pull/13155
-
- 05:50 PM Revision d9248856 (git): ZJIT: Create more ergonomic type profiling API (#13339)
-
05:31 PM Revision eead8316 (git): Prevent enabling yjit when zjit enabled (GH-13358)
- `ruby --yjit --zjit` already warns and exits, but it was still possible
to enable both with `ruby --zjit -e 'RubyVM:YJIT.enable`.
This commit prevents that with a warning and an early return. (We could
also exit, but that seems a bit un... -
04:50 PM Revision cc90adb6 (git): [DOC] Tweaks for String#append_as_bytes
-
-
-
03:52 PM Bug #21338: TracePoint Not Triggered for Kernel#block_given?
- I see that you're calling `Numeric#times` in your profiled test code. We used to use a `block_given?` C call inside it to check if a block is given. @nobu changed it to `defined?(yield)` at commit:3dccb716daaee74d2ae00a5766fe1779fe220a81...
-
12:46 PM Bug #21338: TracePoint Not Triggered for Kernel#block_given?
- Yes, we intentionally moved the implementation for `block_given?` to C. Again, a method can move between C and Ruby between versions.
I just tested rb_tracepoint_new() with RUBY_EVENT_C_CALL and `block_given?`, and an event definitely f... -
03:17 PM Revision c5c252c0 (git): Add a test case for `it` in a regex.
- Co-authored-by: Alan Wu <[email protected]>
-
- The following is crashing for me:
```shell
ruby --yjit --yjit-call-threshold=1 -e '1.tap { raise rescue p it }'
ruby: YJIT has panicked. More info to follow...
thread '<unnamed>' panicked at ./yjit/src/codegen.rs:2402:14:
...
```
It s... -
01:11 PM Feature #21335: Namespaces should be present in the backtrace
- mame (Yusuke Endoh) wrote in #note-4:
> ```
> ...
I took me quite a while to notice that 0x00007ff4d61f9f20 and 0x00007ff4d61f93e0 were different.
Just my personal preference, but if `#<Namespace:0x00007ff4d61f93e0>` is assigned to ... -
12:22 PM Revision 097d742a (git): [Bug #20009] Support marshaling non-ASCII name class/module
-
12:15 PM Bug #20009: Marshal.load raises exception when load dumped class include non-ASCII
- @nobu What happens when using your patch to Marshal.dump a class and then trying to load it on an older Ruby? Will it create an instance variable `E` or error?
I guess if there is no matching BINARY constant it would error anyway, but i... -
- So my suggestion would be:
* Interpret the serialized module/class name as UTF-8, not as BINARY. And of course if it's only 7-bit as US-ASCII (already the case).
* If the module/class name uses another encoding, we could either transcode... -
- Mmh, but Marshal.dump+load of such non-7-bit modules/classes works on TruffleRuby, although it needs a tiny fix:
```patch
diff --git a/src/main/ruby/truffleruby/core/marshal.rb b/src/main/ruby/truffleruby/core/marshal.rb
index 102468e... -
- I made a patch which works well except for tests added by [ruby/spec@907cb35](https://github.com/ruby/spec/commit/907cb35e21).
Since the dumped strings in these tests have not been loadable, I think they are useless actually.
https://gi... -
- Currently instance variables of class/module are prohibited.
It may be possible to put the encoding information there. - 12:08 PM Revision 5e01c0e4 (git): ZJIT: Remove unnecessary cloning
-
- mame (Yusuke Endoh) wrote in #note-13:
> Unless there is a valid use case for that, it would be good to raise an Exception, instead of no-op.
FYI TruffleRuby implemented FrozenError when `rb_str_locktmp()` is used on a frozen String ... -
-
10:49 AM Bug #21344: Segment Fault Caused by no Backported Patches
- Ah, I have misunderstood that the fix was on rexml.gem.
-
- rexml is already bundled gems in ruby 3.2/3.3.
The users can use any versions of rexml specified with lock file.
-
- Close to mark for backport.
-
08:04 AM Bug #21344 (Closed): Segment Fault Caused by no Backported Patches
In my local machine(3.2.3 (2024-01-18 revision 52bb2ac0a6) [x86_64-linux-gnu]), which is the default version in ubuntu2404, a segmentation fault can be stably triggered by the poc including malicious xml data.
The poc here:
```ruby...-
- Seems adjacent to https://bugs.ruby-lang.org/issues/21330.
The `klass` field like the `flags` one isn't part of the `classext_t`, so it's not effectively namespaced.
- 05:14 AM Revision a0fe0095 (git): Don't enumerate `io->blocking_operations` if fork generation is different. (#13359)
-
- I encountered this issue today and attempted a fix: https://github.com/ruby/ruby/pull/13360
I hope that was appropriate. -
01:14 AM Revision 35000ac2 (git): Prevent double free for too big repetition quantifiers (#13332)
- Prevent double free for too big repetition quantifiers
The previous implementation calls `free(node)` twice (on parsing and compiling a
regexp) when it has an error, so it leads to a double-free issue. This
commit enforces `free(node)` ...
05/15/2025
-
11:35 PM Bug #21343 (Open): Namespace: singleton classes of root namespace leak between namespaces
- Ruby classes all have singleton classes, those singleton classes have singleton classes, and so on with infinite nesting. Since we don't have infinite resources, we create these lazily.
Because of this, only a certain number of single... - 10:31 PM Revision 6b10d401 (git): ZJIT: Bail out of recursive compilation if we can't compile callee
- Right now we just crash if we can't compile an ISEQ for any reason
(unimplemented in HIR, unimplemented in codegen, ...) and this fixes
that by bailing out. -
- Although it almost certainly works in this case, volatile is best not
used for multi-threaded code. Using atomics instead avoids warnings from
TSan.
This also simplifies some logic, as system_working was previously only
ever assigned to... -
- Co-authored-by: Ivo Anjo <[email protected]>
Co-authored-by: Luke Gruber <[email protected]> -
09:19 PM Revision 55c9c75b (git): Maintain same behavior regardless of tracepoint state
- Always use opt_new behavior regardless of tracepoint state.
-
- vo.x (Vit Ondruch) wrote in #note-88:
> can you also elaborate how does it look with RubyGems? Do I need to `require "rubygems"` to be able to load e.g. bundled gems libraries?
Each user namespace will automatically load its own copy... -
- > But, since namespaced code starts with a blank slate
I was more wondering where is the cut line, because it might not be obvious. E.g. in the case of RubyGems, the overridden `require` is not something one thinks about these days.
... -
- vo.x (Vit Ondruch) wrote in #note-88:
> Thanks. Very nice overview.
Thank you.
> ...
Nah, no prob, it has been a long one with so many angles.
> can you also elaborate how does it look with RubyGems? Do I need to `require "rubygems"` ... -
- bughit (bug hit) wrote in #note-89:
> If this is referring to instance state, like ivars, this is a corruption of the concept of an object. If the state is different it can't be the same object. It would be like two string references be... -
- > Technically, the reference is the same, the object ID is the same, but the builtin object state is per-namespace now.
If this is referring to instance state, like ivars, this is a corruption of the concept of an object. If the state... -
- fxn (Xavier Noria) wrote in #note-87:
> I have written a [blog post](https://gist.github.com/fxn/86ad8584d7813caf03dac9222f8dcf41) trying to present this feature in a digested way and provide a mental model for all Ruby programmers.
... -
- I have written a [blog post](https://gist.github.com/fxn/86ad8584d7813caf03dac9222f8dcf41) trying to present this feature in a digested way and provide a mental model for all Ruby programmers.
Hope it is all factually correct, and it ... -
07:40 PM Bug #21315: Finalizers violate the `rb_ractor_confirm_belonging` assertion
- it will be fixed with ractor-local GC patch.
-
06:13 PM Revision 04f538c1 (git): Remove dependency on sanitizers.h in default.c when BUILDING_MODULAR_GC
-
- Looks like it's not that simple. This smells of memory corruption because we end up in this loop:
```c
-> 230 while (*keeping_mutexes && *keeping_mutexes != mutex) {
231 // Move to the next mutex in the list:
23... -
- Looks like we're missing a null check:
```
* frame #0: 0x0000000100349cb4 miniruby`thread_mutex_remove(thread=0x0000000000000000, mutex=0x0000600001147f00) at thread_sync.c:225:12
frame #1: 0x000000010033c1f0 miniruby`rb_mutex... -
- Additionally, on macOS, the script crashes but then gets in a seemingly endless loop after printing the 'C level backtrace information' line, and can't be killed with ^C.
-
04:44 PM Bug #21342 (Closed): Segfault: invalid keeping_mutexes when using Mutex in Thread then Fiber after GC
- Ruby crashes with a `[BUG] invalid keeping_mutexes error` when attempting to GC locked mutex that was used in a Thread within a Fiber context after garbage collection. The error indicates an attempt to unlock a mutex that is not locked, ...
-
04:51 PM Revision 06a56a7f (git): [ruby/openssl] ssl: fix potential memory leak in SSLContext#setup
- If SSL_CTX_add_extra_chain_cert() fails, the refcount of x509 must be
handled by the caller. This should only occur due to a malloc failure
inside the function.
https://github.com/ruby/openssl/commit/80bcf727dc -
- OpenSSL::Cipher#encrypt and #decrypt have long supported a hidden
feature to derive a key and an IV from the String argument, but in an
inappropriate way.
This feature is undocumented, untested, and has been deprecated since
commit http... -
04:40 PM Revision 0b9644c2 (git): [ruby/openssl] AWS-LC has support for parsing ber constructed strings now
- https://github.com/ruby/openssl/commit/cdfc08db50
-
- We see this on CI that uses Ubuntu 24.04, since it uses autoconf 2.71 like the environment that makes the release tarballs.
https://github.com/ruby/ruby/actions/runs/15047656643/job/42294231514#step:8:159
> checking for stdbool.h that ... -
01:45 PM Bug #21340: Bump autoconf version to properly handle C23 bool/stdbool defines
- nobu (Nobuyoshi Nakada) wrote in #note-1:
> Isn't `HAVE_STDBOOL_H` defined?
> ...
Not if you build on GCC 15 (or C23) using the ./configure script packaged with the official ruby release tar.gz.
GCC 15 bumps the default C standard ... -
- Isn't `HAVE_STDBOOL_H` defined?
`HAVE_STD_BOOL_H` should be for `std_bool.h`.
- 03:21 PM Revision b2ab1b04 (git): Update bundled gems list as of 2025-05-15
-
-
03:05 PM Revision c3eb4068 (git): Update power_assert for `opt_new`
- https://bugs.ruby-lang.org/issues/21298#note-5
-
- Building that table will likely malloc several time which
can trigger GC and cause race condition by freeing objects
that were just added to the table.
Disabling GC to prevent the race condition isn't elegant,
but iven this is a depreca... -
- It doesn't make sense to set ivars or anything shape
related on a T_IMEMO.
Co-Authored-By: John Hawthorn <[email protected]> -
- The table is global so accesses must be synchronized.
-
- The previous implementation assumed `RBasic` size is `2 * sizeof(VALUE)`,
might as well not make assumption and use a proper `sizeof`.
Co-Authored-By: John Hawthorn <[email protected]> -
- Basically dependent or even caused by #21324 then
-
01:58 AM Bug #21339: Namespace: `RubyVM::InstructionSequence.load_iseq` isn't called for the root namespace
- It works as expected with `--disable=gems`.
```
$ RUBY_NAMESPACE=1 ./local/bin/ruby --disable=gems test.rb
./local/bin/ruby: warning: Namespace is experimental, and the behavior may change in the future!
See doc/namespace.md for kn... -
10:07 AM Bug #21341: `Namespace is not a module (TypeError)` without enabling the namespace
- This only triggers the TypeError for the top-level module Namespace, not when it’s nested inside another module.
```ruby
$ ruby -ve 'module Foo; module Namespace; end; end'
ruby 3.5.0dev (2025-05-15T08:25:56Z master 427ede2dde) +PRI... -
- This code comes from a Rails test case, so we can change it if needed. However, there might be other cases to consider.
-
- Ruby on Rails test case has the `module Namespace`.
https://github.com/rails/rails/blob/main/activemodel/test/validators/namespace/email_validator.rb#L5
```ruby
module Namespace
class EmailValidator < ::EmailValidator
end
e... -
- ```
# frozen_string_ltieral: true
hash["literal"] = value
``` -
-
- - Quote % inside `if` block
- Use short branch name - 06:50 AM Revision 87261c2d (git): Ensure that forked process do not see invalid blocking operations. (#13343)
-
-
- Get rid of hardcoded paths
- 05:51 AM Revision a4ce8639 (git): Add `continue-on-error` to failed Windows 2025 build.
-
- https://github.com/ruby/ruby/pull/13344
05/14/2025
-
11:59 PM Bug #21338: TracePoint Not Triggered for Kernel#block_given?
- ruby-prof does listen for C calls. See:
https://github.com/ruby-prof/ruby-prof/blob/master/ext/ruby_prof/rp_profile.c#L337
``` c
void prof_install_hook(VALUE self)
{
prof_profile_t* profile = prof_get_profile(self);
VALUE even... -
- It's a C method in 3.4, so you can listen for `:c_call`:
```
ruby -ve 'TracePoint.new(:c_call) { p _1 }.enable { block_given? }'
```
Wether something is a C or Ruby method is not something we guarantee to be stable, though, so it... -
- When updating ruby-prof for Ruby 3.4, various tests now fail because a method enter/leave tracepoint is not triggered for `Kernel#block_given?` For example:
https://github.com/ruby-prof/ruby-prof/blob/master/test/line_number_test.rb#L... -
-
-
- `struct RTypedData` was changed significantly in https://github.com/ruby/ruby/pull/13190
which breaks many extensions.
Bumping the ABI version might save some people from needlessly
investigating crashes. -
- Yep, the original ticket was about incorrectly allowing `require` in Ractors and that could segfault.
https://github.com/ruby/ruby/pull/11142 is the proper fix. -
- I hope I'm not misinterpreting the ticket.
Since https://github.com/ruby/ruby/pull/11142 you can now require from a Ractor. `$"` is of course still not readable from a background ractor, but I think the ticket was really about `require`. -
- https://github.com/ruby/setup-msys2-gcc/pull/26#issuecomment-2877229861
The root cause is that the extension is compiled against a newer version of `libwinpthread-1.dll` and headers from msys2, but at the runtime ruby always load `$ruby... -
02:09 AM Bug #21329: `date_core.so` is broken with the recent MSYS2 update
- We need to backport fix if the problem is our side. This ticket is reminder for that.
-
-
-
- Fixed in 1d4822a
-
05:02 PM Bug #21037: Ractors hang with multiple threads
- I didn't tag the commit message properly, but this can be closed. PR https://github.com/ruby/ruby/pull/12633 was merged. Thanks!
-
- Fixed in 1d4822a
-
- I didn't tag the commit message properly, but this can be closed. PR https://github.com/ruby/ruby/pull/12633 was merged. Thanks!
-
- I agree with John, I think we need to find the next eligible ractor for finalizers after ractor termination.
-
05:13 PM Feature #21279: Bare "rescue" should not rescue NameError
- I wondered how it looks in the real world (not really, I only ran against tests) so I wrote some code:
```rb
require "prism"
by_message = Hash.new(0)
by_location = Hash.new(0)
class Visitor < Prism::Visitor
attr_reader :hit
def in... - 05:11 PM Revision ee7dcef0 (git): Update default gems list at 57f8dde0f2228dbc67503403d740a7 [ci skip]
-
- https://github.com/ruby/erb/commit/42f389dc45
-
- This reverts commit cb88edf0bfdc2ce6bfbe3b4e0463a4c2dc5d2230.
It didn't help. You need to go to a different repository (ruby/ruby.wasm)
to see meaningful backtraces.
https://github.com/ruby/ruby.wasm/actions/runs/15000135135/job/4214467... -
- Right now if I install a recent release using rbenv or asdf, or manually download the latest release 3.4.3 or the preview 3.5.0, the included `./configure` seems to have been generated by an old version of autoconf. After running `./conf...
-
- Eregon (Benoit Daloze) wrote in #note-85:
> What are the advantages of running multiple Rails apps in one process instead of one process per Rails app?
One advantage is operational simplicity, only one process to boot / monitor. Nam... -
- What are the advantages of running multiple Rails apps in one process instead of one process per Rails app?
That seems similar to the "Mounting multiple server-less applications in one process" motivation, and so I see the same disadvan... -
-
- This commit allows building YJIT and ZJIT simultaneously, a "combo
build". Previously, `./configure --enable-yjit --enable-zjit` failed. At
runtime, though, only one of the two can be enabled at a time.
Add a root Cargo workspace that c... -
- This no longer reproduce with either parser, so I'm going to assume it was fixed.
-
- ```ruby
File.write("/tmp/compile-cache.rb", <<~'RUBY')
class << RubyVM::InstructionSequence
def load_iseq(path)
p [:load_iseq, path]
RubyVM::InstructionSequence.compile_file(path)
end
end
RUBY
File.write(... -
- This report being about CI failures, I'm going to assume it was fixed otherwise we'd still see it today.
-
- The reproduction steps aren't very clear to me.
But the backtrace shows `fork` and `finalizers` are involved. Since Ractors received fixes around this area recently and the issue date back to 3.0.1, I'm going to assume it has been fix... -
- I can't reproduce on master nor on any version since 3.0.3.
I'm going to assume this was fixed. -
- If the object isn't shareable and already has a object_id
we can access it without a lock.
If we need to generate an ID, we may need to lock to find
the child shape.
We also generate the next `object_id` using atomics. -
-
-
- The rest parameter is always a rb_cArray, even when anonymous. (This is
different from kw_rest, which can be nil.) -
- Use total parameter size instead of lead parameter size when parsing
iseq into hir. Also, copy over IntoUsize for compile-time checked
u32->usize cast. -
-
- As well as associated functions, this should make it more obvious
what the purpose is. -
- Given classes and modules have a different set of fields in every
namespace, we can't store the object_id in fields for them.
Given that some space was freed in `RClass` we can store it there
instead. -
-
- By making `super_classdepth` `uint16_t`, classes and modules can
now fit in 160B slots again.
The downside of course is that before `super_classdepth` was large
enough we never had to care about overflow, as you couldn't
realistically c... -
07:16 AM Bug #21333: heap-use-after-free caused by rehash during update
- byroot (Jean Boussier) wrote in #note-3:
> So `#update` with block isn't considered iteration, so it doesn't increase the `iterlevel` hence `#rehash` is mistakenly allowed.
> ...
byroot (Jean Boussier) wrote in #note-2:
> ```ruby
> .... -
- @nobu given you reverted your fix, should we re-open?
-
- ruby_3_4 commit:862480a316c014d2b4659b22cd12e311c0810105 merged revision(s) commit:bb180b87b43c45e17ff49735a26d7a188d5c8396.
-
-
- https://github.com/rubygems/rubygems/commit/5d5e37bf23
-
06:13 AM Revision a89460a8 (git): [rubygems/rubygems] Reenable skipped specs in truffleruby since they should be fixed
- https://github.com/rubygems/rubygems/commit/a7cbec95c1
-
- https://github.com/rubygems/rubygems/commit/b9a4722d5e
-
-
-
-
- Closed by commit:b66c5c3b1b31581960bcb69d49b618d69ae2a87f
-
- The patch for [Bug #21331] https://github.com/ruby/ruby/pull/13317 introduced a new bug for WASM build.
```
basictest/test.rb:array ...........................basictest/test.rb:53: [BUG] Cannot malloc during GC
ruby 3.5.0dev (2025-0... -
04:57 AM Bug #21337 (Closed): Using `not` on the RHS of a logical operator becomes valid syntax with Prism
- The following syntax behavior differs between Ruby 3.3 (parse.y by default) and Ruby 3.4 (Prism by default).
```ruby
if true && not true; end
```
## Expected
In Ruby 3.3 and earlier, the following code results in a syntax erro... -
-
-
-
-
-
- This reverts commit bb180b87b43c45e17ff49735a26d7a188d5c8396, which
caused "malloc during GC" error on wasm. -
-
- mame (Yusuke Endoh) wrote in #note-2:
> As the current behavior of the backtrace, if the receiver class is completely bound to the constant names, they are prefixed to the method names in the backtrace.
Ah, ok, I didn't know that. T... -
01:08 AM Feature #21335: Namespaces should be present in the backtrace
- The `ns_id` seems good to be shown in backtrace (I'm unsure about this format is good or not):
```
foo.rb:7:in namespace(1) 'Foo.test'
``` -
- It is possible to change how to display the backtrace to show class names that are not completely bound to constants.
I have conservatively designed the current spec of the backtrace to avoid potential confusing output, but I'm not sure... -
- As the current behavior of the backtrace, if the receiver class is completely bound to the constant names, they are prefixed to the method names in the backtrace. So, if you do:
```ruby
NS1 = Namespace.new
NS1.load "foo.rb"
NS2 =... -
-
- This reverts commit bb180b87b43c45e17ff49735a26d7a188d5c8396, which
caused "malloc during GC" error on wasm. -
- ruby_3_4 commit:65e02ab1a6e4482f417ec1d1cb4453958ad36ca3 merged revision(s) commit:3e47e7a499acd256be549935fcb559d3c82e556c, commit:b48b841378f80e16378ceb83f3b78e52df9ae023, commit:2fe8b9cd3d308d754f3d33a948dfb1dd782a10dc.
-
- Applied in changeset commit:git|65e02ab1a6e4482f417ec1d1cb4453958ad36ca3.
----------
merge revision(s) 3e47e7a499acd256be549935fcb559d3c82e556c, b48b841378f80e16378ceb83f3b78e52df9ae023, 2fe8b9cd3d308d754f3d33a948dfb1dd782a10dc: [Backpo... -
- Fix redefinition of `clock_gettime` and `clock_getres`
winpthreads-git 12.0.0.r720 provides `clock_gettime` and
`clock_getres` as inline functions.
digest.so needs ruby/digest.h which is installed by build-ext
... -
- I would have appreciated something like, the two last examples are clear, could you provide code for the first one?
By rejecting this way, you have lost me. -
- This is not about whether the ticket is "normal" or not. The fact is, your explanation is vague and incomplete. You may feel it’s obvious, but it isn’t.
Do not assume others are smart enough to understand your explanation based on nat... -
- [Bug #21331] Prohibit modification during stlike loop
-
- ruby_3_4 commit:ac311437d3c4d60fe4ac100e9652b0cbc3e5f481.
-
- ruby_3_4 commit:50b1759be00713535c41f5650feb3967c533450a.
-
- https://github.com/ruby/ruby/actions/runs/15008767265/job/42173424631
"error while executing at wasm backtrace" doesn't have meaningful
symbols. We can't debug this from just looking at the "[BUG] Cannot
malloc during GC" message.
As su...
05/13/2025
-
- Apparently `$LOAD_PATH.resolve_feature_path('erb/escape')` returns true
for miniruby but `require 'erb/escape'` fails on it.
I still don't want to check it and rescue LoadError at the same time
because the code looks too complicated. Le... -
- I have spent countless hours of my free time on namespaces in the last days. I experimented with the feature again tonight, and wrote this ticket with care, looking for edge cases, and presenting them, past midnight.
While rejecting on ... -
- > If you call something like ref.http_client.fetch, the method fetch is executed in the context of the namespace where http_client lives. That's the namespace where blank? is defined. So in that case, the call should work.
`blank?` was ... -
- I've lost count of how many times I've said this, but here it is again:
**You must provide reproducible code, your expectation, and actual results.**
The current implementation of namespaces is incomplete, and code using real-world... -
- > Maybe redefine communication across namespaces in a very restrictive way that make these issues by design
Maybe redefine communication across namespaces in a very restrictive way that **prevents** these issues by design -
- # Implications related to builtin classes and modules
As we know, when a namespace is created, builtin classes in the namespace are as in the root one. This applies to all references to them, explicit or implicit.
This is at the co... -
- (https://github.com/ruby/erb/pull/61)"
This reverts commit https://github.com/ruby/erb/commit/1c9200aab071.
Now that we've bumped BASERUBY, we shouldn't need to rescue the
LoadError.
https://github.com/ruby/erb/commit/846b20fe0e -
- Namespaces can call in to other name spaces. I think it might be helpful if namespace information is in the backtrace.
For example:
```ruby
File.binwrite "foo.rb", <<-eorb
class Foo
def self.test(ns1, ns2, count)
if count... -
- Eregon (Benoit Daloze) wrote in #note-82:
> @matz What real-world use cases do you see for Namespace?
One thing I can think of is running multiple Rails applications in one process. Right now Rails applications are basically singlet... -
- mame (Yusuke Endoh) wrote in #note-51:
> This feature proposal is quite exceptional. Because this is a feature strongly driven by @matz himself, and Ruby is matz's language.
> ...
I took some time before replying here to try to find the ... -
- > (In fact, the overhead appears to be only about 5%.)
That's when namespaces aren't used. Since creating a namespace start some sort of de-optimization phase, I assume the impact is way worse with `RUBY_NAMESPACE=1`, but I couldn't ... -
- jhawthorn (John Hawthorn) wrote in #note-79:
> Sorry, I didn't get a chance to review this before it was merged. I really don't think adding this level of indirection to RCLASS_EXT access and similar is a good idea.
Since I was the o... -
- This ticket was kind of oriented to docs.
I believe the emphasis has to be different and have created https://bugs.ruby-lang.org/issues/21334.
This one can be close if you will. -
- So `#update` with block isn't considered iteration, so it doesn't increase the `iterlevel` hence `#rehash` is mistakenly allowed.
At first sight the fix is to treat `#update` like an iteration when yielding. -
- ```ruby
$a = (1..1337).to_h { |k| [k, k] }
$b = (1..1337).to_h { |k| [k, k * 2] }
$a.update($b) { |k, o, n|
$a.rehash
}
```
Seem to do the trick. -
- The reproduction script seem missing something? `$b` is `nil` here so the script simply fail.
-
- Hi, we found a heap-use-after-free caused by rehash during update
```
$a = (1..1337).to_h { |k| [k, k] }
$a.update($b) { |k, o, n|
$a.rehash
}
```
```
$ git log | head -n4
commit a6435befa76c2ae0525147f934bd9cd1914ffb8a
... -
-
-
- Applied in changeset commit:git|1ee4b43a56e9539d3b09c08114c59cb49857b956.
----------
Set#merge: raise if called during iteration
[Bug #21332] -
- `Set#merge` is missing a check for iteration: https://github.com/ruby/ruby/pull/13322
-
- Hi, we found a heap-use-after-free caused by mutating the set.
hash() returning 0 is important to create the collision and then trigger eql?
```
$flag = false
$c = 0
class C
def hash
return 0
end
def eq... -
- [Bug #21332]
- 08:23 PM Revision 1d4822a1 (git): Get ractor message passing working with > 1 thread sending/receiving values in same ractor
- Rework ractors so that any ractor action (Ractor.receive, Ractor#send, Ractor.yield, Ractor#take,
Ractor.select) will operate on the thread that called the action. It will put that thread to sleep if
it's a blocking function and it needs... -
- It currently fails like: Can't find 'action.yml'
-
- https://github.com/ruby/erb/commit/c2d1f82817
-
- (https://github.com/ruby/erb/pull/61)"
This reverts commit https://github.com/ruby/erb/commit/1c393aa738f3.
https://github.com/ruby/erb/commit/1c9200aab0 - 06:10 PM Revision f82dce14 (git): Update default gems list at 9db0704e67c8af3b47429e28e1da5e [ci skip]
-
- https://github.com/ruby/erb/commit/08b544cdb8
-
- Even cgi.gem publicly defines CGI::VERSION today. It's just weird that
ERB::VERSION is kept private at this point.
https://github.com/ruby/erb/commit/46801cbd47 -
06:02 PM Bug #21327: Windows builds seem broken after clock_gettime changes?
- I just tested this with the ruby_3_4 branch. It's failing with:
```
../ruby/win32/win32.c:4795:1: error: redefinition of 'clock_gettime'
4795 | clock_gettime(clockid_t clock_id, struct timespec *sp)
| ^~~~~~~~~~~~~
In file i... -
- It's fixed at https://github.com/ruby/ruby/pull/13302
But cross build is still broken.
* https://rubyci.s3.amazonaws.com/crossruby/crossruby-master-mingw32/log/20250513T000922Z.fail.html.gz
* https://rubyci.s3.amazonaws.com/crossr... - 06:00 PM Revision 4a9d46ce (git): [ruby/erb] Refactor the logic of require 'erb/escape'
- (https://github.com/ruby/erb/pull/61)
https://github.com/ruby/erb/commit/1c393aa738 - 05:50 PM Revision 735f2838 (git): [ruby/erb] Support all cgi.gem versions
- (https://github.com/ruby/erb/pull/60)
https://github.com/ruby/erb/commit/de9bb8c3cc -
- RUBY_FL_USER3
(https://github.com/ruby/stringio/pull/133)
This way when someone removes these flags from Ruby or update them,
they'll find this reference when greping.
Followup: https://github.com/ruby/stringio/pull/128
https://github... -
-
- Applied in changeset commit:git|bb180b87b43c45e17ff49735a26d7a188d5c8396.
----------
[Bug #21331] Prohibit modification during stlike loop -
- https://github.com/ruby/ruby/pull/13317
-
- Hi, we found a heap-use-after-free caused by rehash during transform_values!.
```
$a = (1..1337).to_h { |k| [k, k] }
$a.transform_values! { |v|
$a.rehash
v * 2
}
```
```
$ git log | head -n4
commit a6435befa76c2ae05... -
- I was able to reproduce the bug, and rebasing @luke-gru 's patch does indeed fix the hang (I reduced 100M to 10M though cause it's very long otherwise).
@jhawthorn or @tenderlovemaking could you review the PR? The patch make sense to ... -
- Related to https://bugs.ruby-lang.org/issues/21037
-
- I was able to reproduce this on 3.3, but not on 3.4 not master. So I'll assume it was fixed and close.
-
- 02:16 PM Revision 1e2a67e1 (git): Fix typo in `NEWS.md`.
-
01:54 PM Bug #21166: Fiber Scheduler is unable to be interrupted by `IO#close`.
- I've renamed `rb_thread_io_interruptible_operation` to `rb_thread_io_blocking_operation` which better suits the naming convention internally.
-
- https://github.com/ruby/ruby/pull/13127 has been merged, allowing us to move forward with this feature.
-
- The `includer` field is only used for `T_ICLASS`, so by moving
it into the existing union we can save one `VALUE` per class
and module. -
- The frozen flag is what I missed. I think we should move the flag to `rb_classext_t`.
-
- ```ruby
File.write("/tmp/test.rb", <<~'RUBY')
Hash.freeze
RUBY
ns = Namespace.new
ns.require("/tmp/test.rb")
class Hash
def monkey_patch
end
end
```
Expected behavior:
Since the monkey patch is in a different na... -
- @Eregon It has already been fixed. Thank you.
-
- @make_now_just In the first two cases, it should be `.dump_in_ns` not `.dump`
-
- I found an issue on `Marshal` and `Namespace` maybe related to this ticket.
When dumping an object defined in a namespace using Marshal, the result will vary depending on whether the namespace is held as a variable or constant, and whet... -
- byroot (Jean Boussier) wrote in #note-1:
> ### Half-way solution
> ...
I think this would be a pretty good solution actually, it seems very unlikely to have class names which can't be encoded in UTF-8. -
- In my opinion, we need to introduce a new format for dumping classes/modules correctly.
Marshal uses `c` and `m` (`TYPE_CLASS` and `TYPE_MODULE`) as format prefixes currently, so the format is the following:
```
| 1 byte | ... | ... -
- Loading RubyGems creates more constants than just `Gem`:
```
$ ruby -v --disable-gems -e 'a=Object.constants; require "rubygems"; b=Object.constants; p b-a'
ruby 3.4.1 (2024-12-25 revision 48d4efcb85) +PRISM [x86_64-linux]
[:Monitor,... -
06:47 AM Bug #21324: Namespace loads RubyGems in root Namespace but it should not
- I've identified the issue and prepared a fix.
The problem is that RubyGems is being loaded into the root namespace during Ruby's boot process, and then all user namespaces inherit this same Gem constant through the Copy-on-Write mecha... -
- Applied in changeset commit:git|c941fced211c7472d6a61e7469b70282ac506ea6.
----------
Throw RuntimeError if getting/setting ractor local storage for non-main ractor
[Bug #19367] -
- [Bug #19367]
-
- Some debugging happening at https://github.com/ruby/date/pull/125 already! Should this be moved to the date repo, by the way?
-
- From https://github.com/ruby/psych/issues/730 and https://github.com/nobu/rdoc/actions/runs/14969443434/job/42046842539#step:5:35
```
Error: test_accept_blank_line(RDocMarkupToAnsiTest): LoadError: 127: The specified procedure could ... -
10:28 AM Revision 7517d762 (git): [DOC] Fix a link and sort links in NEWS.md
-
- 10:02 AM Revision 425fa0ae (git): Make `waiting_fd` behaviour per-IO. (#13127)
- - `rb_thread_fd_close` is deprecated and now a no-op.
- IO operations (including close) no longer take a vm-wide lock. -
-
07:57 AM Feature #21308: Replacing the Float#to_s (dtoa.c) implementation with a modern algorithm
- The change in https://github.com/ruby/json/pull/768 was an easy win, because the previous implementation was calling into question`Float#to_s` via `rb_funcall`. The overhead of doing so is so extreme, that any implementation that would s...
-
- @akr the test is insufficient, you should probably update RubySpec too.
-
-
-
- 05:13 AM Revision c4055709 (git): Update default gems list at cd7495a1d0e003360c96bb9746c1a8 [ci skip]
-
- Report EOF when applicable instead of an empty fragment.
Also stop fragment extraction on first whitespace.
https://github.com/ruby/json/commit/cc1daba860 -
- https://github.com/ruby/json/commit/f3dde3cb2f
-
- https://github.com/ruby/json/commit/41f1f6939d
-
- https://github.com/ruby/json/commit/30e35b9ba5
-
- https://github.com/ruby/json/commit/832b5b1a4c
-
- 18d395e0784401585b5c14300e689de55e208647
-
03:22 AM Bug #21280 (Closed): StringIO#set_encoding warns when backed by chilled string literal
- Fixed by commit:18d395e0784401585b5c14300e689de55e208647
-
03:20 AM Revision 5e59ae18 (git): [ruby/stringio] Fix Ruby 3.4 check
- https://github.com/ruby/stringio/commit/a27c5d5e2e
Co-authored-by: Sutou Kouhei <[email protected]> -
- StringIO does not warn for unchilled unfrozen string or for frozen
string, so it should not warn for chilled string.
https://github.com/ruby/stringio/commit/4ac33b8c70 -
- Fix C23 (GCC 15) WIN32 compatibility for rb_define_* functions
Fixes [Bug #21286] -
- ruby_3_2 commit:53c09781495b0096eec88e4268a48a13b245097a merged revision(s) commit:719486a642f0e282b02b958069b8b39b85b3aa1e.
-
-
- Fixed by commit:203199251f723d02f23f5782523d16f91277ea82
-
- This overrides Enumerable#to_h, but doesn't handle a block, breaking
backwards compatibility.
Set#to_h was added in the marshalling support commit, but isn't
necessary for that, as the underlying function is called. Remove
the method de... -
05/12/2025
-
- Sorry, I didn't get a chance to review this before it was merged. I really don't think adding this level of indirection to RCLASS_EXT access and similar is a good idea.
Let's look at `rb_class_get_superclass` as an example. This shoul... -
-
- As another datapoint on Linux (Ubuntu 24.04, AMD Ryzen 9800X3D), it looks like it is slower (around 0-5%) but also has higher memory usage by about 5-10%.
```
pre-namespace: ruby 3.5.0dev (2025-05-11T03:09:26Z master 49742414f6) +PRI... -
- byroot (Jean Boussier) wrote in #note-75:
> > Please open another ticker for further problems, issues, bugs, etc.
> ...
I had a plan to add CI configuration to add `RUBY_NAMESPACE=1` checks and I thought it would not be a part of this.... -
- > Please open another ticker for further problems, issues, bugs, etc.
Right now `RUBY_NAMESPACE=1 make -j btest` fails, are you actively working on adding a CI step that run with `RUBY_NAMESPACE=1` or should I open another ticket to t... -
- I've merged the change into master. So, let me close this issue.
Please open another ticker for further problems, issues, bugs, etc. -
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
- I will :) I'm planning to cut an early Ruby 3.4 release this week.
-
10:02 AM Bug #21286: Windows - MSYS2 just updated to GCC 15.1.0, builds failing
- Will you create new ruby stable releases in the next days? I'm asking because I've seen many CI failures and bug reports due to this GCC-15 issue and I wonder if I should release a new RubyInstaller series with this patch or if there wil...
-
- 04:22 PM Revision 03a7f1ff (git): Update default gems list at 957473d87fc7741c8d36caed8f05db [ci skip]
-
- https://github.com/ruby/psych/commit/b9dec9f811
-
- In Ruby < 3.0, the superclass of StringIO was actually already `Data`,
but it doesn't have the expected shape. So, on these earlier versions it errors:
> NoMethodError: undefined method `members' for #<StringIO:0x00005641dd5f2880>
> ...
... -
-
-
03:13 PM Feature #21258: Retire CGI library from Ruby 3.5
- CGI may be an "old" protocol but that maturity brings stability. It means I don't need to run an app server process (like puma) at all. I don't need to worry about open ports, memory leaks, resource leaks, connection pooling, etc. It's e...
-
-
- It's already defined in internal/sanitizers.h.
-
-
- https://github.com/ruby/erb/commit/9152ce8db4
-
-
-
-
-
- Fix Visual C warnings:
```
iseq.c(3793): warning C4312: 'type cast': conversion from 'int' to 'void *' of greater size
iseq.c(3794): warning C4312: 'type cast': conversion from 'int' to 'void *' of greater size
``` -
- Fix Visual C warnings:
```
regenc.h(121): warning C4067: unexpected tokens following preprocessor directive - expected a newline
``` -
- When the test is repeated 20 or more times in the same process
it's not that unlikely for `rand(100_000)` to return the same thing
twice, causing `TestObjectIdTooComplexClass` to fail.
```
1) Failure:
TestObjectIdTooComplexClass#test_... -
- 02:09 AM Revision 4464cbe5 (git): [rubygems/rubygems] Fix doctor command parsing of otool output
- I have several gem dylibs that have a line matching "(compatibility "
with no file path preceding it.
https://github.com/rubygems/rubygems/commit/de9dc2bdc4 -
- e.g.
```
JSON.dump(1746861937.7842371)
```
master:
```
"1.https://github.com/ruby/json/commit/746861937784+9"
```
This branch and older json versions:
```
https://github.com/ruby/json/commit/1746861937.7842371
```
In the end it's shor...
05/11/2025
-
- Or, use GCC 14 for a while, if possible. (I am not familiar with MSYS2)
-
- I believe this is the same issue as #21286.
Please apply the patch of https://github.com/ruby/ruby/pull/13202 yourself, or wait for Ruby 3.3.9. -
03:27 PM Bug #21328 (Closed): Gem::Ext::BuildError: ERROR: Failed to build gem native extension.
- We are facing an issue that starts in 05/03/2025
We have an Azure Devops Pipeline that build and test a project, but in the test step we have the following error:
```
Gem::Ext::BuildError: ERROR: Failed to build gem native extension... -
- ../namespace.c: In function ‘current_namespace’:
../namespace.c:221:48: warning: ‘proc_ns’ may be used uninitialized [-Wmaybe-uninitialized]
221 | if (permit_calling_builtin || (proc_ns && NAMESPACE_USER_P(proc_ns))... -
- GC can trigger while we build the table, and if it sweeps an object
with an ID, it may not find it in the `id_to_obj` table. -
-
-
- This reverts commit 2a9236366d6016738a756caecab03263565a20c7.
spec/ruby/optional/capi/spec_helper.rb doesn't work well for mingw. -
- winpthreads-git 12.0.0.r720 provides `clock_gettime` and
`clock_getres` as inline functions. -
- The intial complex shape implementation never allowed objects
other than T_OBJECT to become too complex, unless we run out of
shapes.
I don't see any reason to prevent that.
Ref: https://github.com/ruby/ruby/pull/6931 -
05:02 PM Feature #21219: `Object#inspect` accept a list of instance variables to display
- ℹ️ In case it's of interest/inspiration, I released [Inspectable](https://alchemists.io/projects/inspectable) 0.2.0 which tackles what is described in this issue by making it easy to redact sensitive information, slim down verbosity, or ...
-
04:32 PM Misc #21325: make ruby more middle-aged man friendly
nobu (Nobuyoshi Nakada) wrote in #note-2:
> `format` is an alias of `#sprinf` inherited from `Kernel`.
> ...
if format is not ok,
to_s will be good, like Integer, `255.to_s(16)`-
- `format` is an alias of `#sprinf` inherited from `Kernel`.
We can't rename it. -
- > wanted to make sure everyone is aware of this
What I want all in the thread to realize is that if somewhere in the tree of your object, some code relies on a core extension, that object is automatically not cross-namespace-safe, and y... -
- Example:
```ruby
# main.rb
class C
attr_reader :target
def initialize(target)
@target = target
end
end
ns = Namespace.new
ns.require_relative 'foo'
c = C.new(Object)
ns::Foo.new.m(c)
# foo.rb
class Foo
def m(c)
(c.... -
- fxn (Xavier Noria) wrote in #note-70:
> Maybe you have the C code in your head, I am looking at it as a Ruby programmer would. I have one object reference, and I pass it in method call, no namespace-level constant resolution involved in... -
- > That should not happen, and if it happens, it's a bug. Could you write and show the script to reproduce it?
Exactly, as far as I can tell it does not happen, which to me it means we do not have (1) in place.
It was a way to say that,... -
- fxn (Xavier Noria) wrote in #note-68:
> @tagomoris when I said that (1) would break Ruby semantics I meant that **in the same namespace** you would have two objects with the same ID and different state. That does not happen as far as I ... -
- @tagomoris when I said that (1) would break Ruby semantics I meant that **in the same namespace** you would have two objects with the same ID and different state. That does not happen as far as I can tell.
-
- @tagomoris From the point of view of the Ruby programmer, I have an object, I pass it, and the callee receives a different object. To me, that is (2). Somehow, the object itself has mutated.
We are not talking about constant lookup in t... -
- @fxn I'm unsure I correctly understand what you mean... Let me check your point.
You said:
* The current namespace implementation behaves like (2), not (1).
* The passed value to `ns::C.x` (in the callee namespace side) will be a di... -
- Some of the issues I have noted seem to be addressable and have been added to the TODO. (I have written tickets for them.)
But there is [one](https://bugs.ruby-lang.org/issues/21322) that seems fundamental and worries me. Since this tic... -
- @byroot Thank you for working on the feature. I'll merge the patches soon (had merged one already).
About the crash with `RUBY_NAMESPACE=1`, I'll work on it ASAP after merging the branch once.
I know that the 3-5% performance degrada... -
- Alright, so I can't benchmark with `RUBY_NAMESPACE=1` because I keep running into that crash and can't figure it out.
In the process of trying to find a fix I found multiple other bugs:
- https://github.com/tagomoris/ruby/pull/7
... -
- I submitted a fix for the crash here: https://github.com/tagomoris/ruby/pull/5
But now it's crashing in `require`
```
/Users/byroot/.gem/rubies/head-namespaces/gems/activesupport-7.2.2.1/lib/active_support/inflector/inflections.rb... -
- headline benchmarks (in non-ideal conditions, I was on battery, etc) and **whithout** `RUBY_NAMESPACE=1`:
```
master: ruby 3.5.0dev (2025-05-11T03:09:26Z master 49742414f6) +PRISM [arm64-darwin24]
namespace: ruby 3.5.0dev (2025-05... -
- Trying to run benchmarks with `RUBY_NAMESPACE=1` crashes during init:
```
opt/rubies/head-namespaces/bin/ruby: [BUG] Segmentation fault at 0x0000000000000008
ruby 3.5.0dev (2025-05-11T00:30:16Z namespace-on-read-.. 6629795fff) +PRIS... -
- I opened a PR on yjit-bench with workarounds for both the fiddle issue and the CGI issue: https://github.com/Shopify/yjit-bench/pull/363
I'll see if I can benchmark your branch. -
- > bin/rails aborted!
> ...
Yes, the `cgi` changes broke Rails. You can add `gem "cgi"` to the benchmark's Gemfile to workaround that.
Both Rails nightly CI and Shopify nightly CI are still broken from the recent Set change and got br... -
- `make yjit-bench` is now failing even on master HEAD.
> bin/rails aborted!
> ...
I saw commits to re-organize cgi.
Hmm, I can wait for the fix and will make benchmark outputs if others don't make changes which conflict with the name... -
- We have followed the discussion in https://bugs.ruby-lang.org/issues/21311.
I may contribute a doc patch to explain the scope to which root object references behave differently in different namespaces.
A consequence of this design is t... -
- The third option is more aligned with my intuitive idea of a new entity with new rules. Instead of a container of methods and constants that `::` can work with, you may need a new kind of language-level entity an "isolate", a "cell", a "...
-
- @eregon maybe, but the fact is that you pass an object, and the callee receives a different object. This is not consistent with the current language.
Ideas:
* Introduce the concept of "namespace-level root objects", documented to be sw... -
-
- https://github.com/tagomoris/ruby/pull/7
RUBY_TYPED_FREE_IMMEDIATELY can be added because namespace_entry_free does no IO nor
things to block. -
- If we're referencing the namespace object we have to mark it.
-
-
- ```
/opt/rubies/head-namespaces/bin/ruby(sigsegv+0x84) [0x104e897e8]
/usr/lib/system/libsystem_platform.dylib(_sigtramp+0x38) [0x18de56de4]
/opt/rubies/head-namespaces/bin/ruby(object_id+0x80) [0x104d7d420]
/opt/rubies/head-namespaces/bi... -
-
-
-
-
-
-
-
-
- The macro RCLASS_EXT() accesses the prime classext directly, but it can be
valid only in a limited situation when namespace is enabled.
So, to prevent using RCLASS_EXT() in the wrong way, rename the macro and
let the developer check it i... -
- To make RClass size smaller, move flags of prime classext readable/writable to:
readable - use ns_classext_tbl is NULL or not (if NULL, it's readable)
writable - use FL_USER2 of RBasic flags -
-
-
-
- `rb_define_private_method` performs strict type checking on the function
pointer. As a result, we cannot pass the function a generic signature.
```
/github/workspace/src/namespace.c:1097:72: note: expected 'VALUE (*)(void)' {aka 'long... -
- ```
/github/workspace/src/proc.c:2023:65: error: format specifies type 'void *' but the argument has type 'const rb_namespace_t *' (aka 'const struct rb_namespace_struct *') [-Werror,-Wformat-pedantic]
2023 | rb_bug("Unexpected ... -
-
-
- ```
internal/class.h:158:20: warning: ‘RCLASS_SET_CLASSEXT_TABLE’ declared ‘static’ but never defined [-Wunused-function]
158 | static inline void RCLASS_SET_CLASSEXT_TABLE(VALUE obj, st_table *tbl);
| ^~~~~~~~... -
- ```
namespace.c: In function ‘rb_namespace_available’:
namespace.c:55:1: warning: old-style function definition [-Wold-style-definition]
55 | rb_namespace_available()
| ^~~~~~~~~~~~~~~~~~~~~~
``` -
-
- This reverts commit 585598623da949c92d0f2ea94029a863142ec908.
This broke Windows CIs
```
linking miniruby.exe
Creating library miniruby.lib and object miniruby.exp
process.obj : error LNK2019: unresolved external symbol clock_gettim...
05/10/2025
- 10:07 PM Revision 915d0277 (git): Fix typo (newsString -> new String)
-
10:01 PM Bug #21327 (Closed): Windows builds seem broken after clock_gettime changes?
- https://github.com/ruby/ruby/actions/workflows/windows.yml
not sure how stable windows normally is, but this looks consistent since this commit (5855986)
Assigning to nobu to investigate. -
- Diff instructions using:
```sh
DUMP="puts RubyVM::InstructionSequence.compile(ARGV.first).disasm.gsub(/^\d+ /, '')"
RUBY="def a(x, ...); b(...); end"
env -u RUBYOPT \
diff -U99 \
<(./build/ruby -v --parser=parse.y -e "$... -
- after 35, I lose all interests about programming.
no linux, no vim, no iterm2, and no much much.
when I want to format a ruby time to string.
I will try `Time.now.format("yy-MM-dd")`,
but I get `private method 'format' calle... -
05:32 PM Revision 98667f82 (git): [DOC] Update documentation for ObjectSpace#each_object
- Co-authored-by: Benoit Daloze <[email protected]>
- 05:32 PM Revision 29b3d683 (git): [DOC] Make clear that current behavior is not ideal
- 05:32 PM Revision a51b4a86 (git): [DOC] ObjectSpace#each_object behavior in multi-Ractor mode
- This behavior of ObjectSpace#each_object has been around since Ruby 3.0
when Ractors were first introduced, but was never documented and has
caused some amount of confusion:
https://bugs.ruby-lang.org/issues/17360
https://bugs.ruby-lang... -
- kou (Kouhei Sutou) wrote in #note-54:
> > building fiddle fails
> ...
I couldn't find the log when fiddle is built during `make yjit-bench`...
Anyway, I'll retry yjit-bench on Linux. -
- @mame
> I am personally excited to see Ruby adding a bold new feature that brings real energy and discussion to the community. This is Ruby. Let's enjoy this moment and move forward together with excitement.
Absolutely.
I would have w... -
09:03 AM Feature #21311: Namespace on read (revised)
- > building fiddle fails
Could you share the error message? -
- > So I ran benchmark/app_fib.rb
Fibonnaci isn't a very interesting benchmark here. It's just the same method being called over and over.
`yjit-bench` contains some macro benchmark that are more likely to have inline cache misses he... -
- On my laptop, building fiddle fails and it prevents running `make yjit-bench`.
So I ran benchmark/app_fib.rb and got results (The result after several times continuous runs):
```
# Ruby 3.4.2
$ time ruby benchmark/app_fib.rb
... -
- This allow to get or set fields without having to worry about the
shape type. -
- To be consistent with `rb_obj_field_set`.
-
- > It means CGI is not promoted bundled gems.
Is there a reason to skip this step?
For previous extractions we could rely on warnings to keep track of gems that need to be updated while continue to test ruby-head, with this immedi... -
- RubyGems has tons of mutable state, isn't core library and isn't "builtin classes" either, so it should not be in root Namespace, but it is currently:
```
$ RUBY_NAMESPACE=1 ruby -ve 'ns = Namespace.new; p ns::Gem.equal?(Gem)'
ruby 3.... -
- Looks somewhat similar, `benchmark` loads fine in main namespace but not in other namespace:
```
$ RUBY_NAMESPACE=1 ruby -ve 'require "benchmark"; p Benchmark::VERSION; ns = Namespace.new; File.write "ns.rb", "require :benchmark.to_s; ... -
- ```
$ irb
irb(main):001> 6 * 7
=> 42
$ RUBY_NAMESPACE=1 irb
/home/eregon/prefix/ruby-master/bin/ruby: warning: Namespace is experimental, and the behavior may change in the future!
See doc/namespace.md for know issues, etc.
/hom... -
- Seems very related to https://bugs.ruby-lang.org/issues/21322 so let's keep this one closed and discuss there.
-
- Full reproducer, showing that even though `Object` is the same reference in all namespaces it can have different constants in different namespaces, but only if calling `constants` in that namespace:
```
$ RUBY_NAMESPACE=1 ruby -ve 'ns ... -
-
- @byroot, if you will, I believe this one can be closed too.
-
- I suspect this one was an error of mine. I think we can close.
-
- Did you actually test the code? `mod1.constants == mod2.constants` returns `true`.
```
$ cat main.rb
ns = Namespace.new
mod1 = Object
mod2 = ns::Object
p mod1 #=> Object
p mod2 #=> Object
p mod1.object_id == mod2.object_id #=... -
- Right, IOW it's the duality between builtin classes and other classes.
Builtin classes' constants, methods, ivar and cvars are all a copy per Namespace (and so potentially differ), even though for all builtin classes they are `equal?` w... -
- For example, if I instead do
```ruby
ns = Namespace.new
ns.require_relative 'foo'
c = Class.new
c::X = 1
ns::C.x(c)
```
then `obj::X` resolves as expected in the method.
Method calls somehow seem to be swapping `VALUE`s that correspo... -
- > I believe this is consistent with the feature
I say so in the sense that objects with the same ID are equal in the namespace.
However, it could be argued that it is not consistent with Ruby.
`Object` is a constant that stores a clas... -
- @tagomoris thanks for the docs under `doc/namespace.md`.
Unless I missed it, I believe there is an edge case related to builtin classes (using the vocabulary there). Consider:
```ruby
# test.rb
ns = Namespace.new
ns.require_rela... -
- Full repro:
```
# Copy the description program to program.rb, e.g. $ xsel -b > program.rb
$ RUBY_NAMESPACE=1 ruby -ve 'puts "in main"; require "./program"; puts "in namespace"; Namespace.new.require "./program"'
ruby 3.5.0dev (2025-... -
- This is probably a consequence of https://bugs.ruby-lang.org/issues/21318.
-
- Right I think in the namespace defining the class/module the `Module#name` needs to not have a prefix, or it will break many gems.
OTOH when used outside, it could be very confusing without a prefix.
> suggested that Class#name shoul... -
- Main problem here is that there are many programs that depend on that name.
They may store it somewhere for later `const_get`. For example, [polymorphic Active Record associations](https://guides.rubyonrails.org/association_basics.html#... -
- This is definitely not ideal. `Class#name` could end up referring to a different constant.
```ruby
# main.rb
NS = Namespace.new
NS.require "./sub"
# sub.rb
class C; end
p C #=> expected: C
#=> actual: NS::C ← This could r... -
- Complete reproducer:
```
$ RUBY_NAMESPACE=1 ruby -ve 'p Module.nesting; ns = Namespace.new; File.write "ns.rb", "p Module.nesting"; ns.require "./ns"'
ruby 3.5.0dev (2025-05-10T07:50:29Z namespace-on-read-.. bd4f57f96b) +PRISM [x86_64... -
- > and sorry for the noise.
It's alright, even if the feature is experimental, the sooner bugs are noticed the better. So thank you for experimenting. -
- Yes, please, and sorry for the noise.
-
- > cannot repro anymore.
Closing then? -
- I was yesterday running a checkout from May 6. Updated the branch now and cannot repro anymore.
-
- @mame I was here till 4 AM experimenting with this feature.
To be organized, I reported different things in different tickets. These two could be related, but I don't know it. If they are, they will be closed together. -
- I don't think it's really necessary to create a separate ticket for this from #21319, is it?
-
-
- I was yesterday running a checkout from May 6. Updated the branch now and cannot repro anymore.
-
- I cannot repro the issue.
```ruby
# main.rb
ns = Namespace.new
ns.require "./sub"
p :success
```
```ruby
# sub.rb
module M
end
class C
include M
end
```
```
$ RUBY_NAMESPACE=1 ./miniruby main.rb
./miniruby: w... -
- winpthreads-git 12.0.0.r720 provides `clock_gettime` and
`clock_getres` as inline functions. -
- 06:59 AM Revision 3c37d6ff (git): Update bundled gems list as of 2025-05-09
-