Activity
From 01/21/2022 to 01/27/2022
01/27/2022
-
10:42 PM Feature #18552: Expose `VALUE rb_singleton_class_get(VALUE)` to extensions
- > The name rb_singleton_class_get() doesn't seem great though.
Yeah, it could definitely be expose under another n... -
09:15 PM Feature #18552: Expose `VALUE rb_singleton_class_get(VALUE)` to extensions
- A new function seems better than that flag check (which doesn't work on TruffleRuby currently, will always return `fa...
-
- > Which does seem very clean.
I of course meant "doesn't". Oops.
> In this case, it's exposing yet another oper... -
05:12 PM Feature #18552: Expose `VALUE rb_singleton_class_get(VALUE)` to extensions
- > `!SPECIAL_CONST(obj) && FL_TEST(RBASIC(obj)->klass, FL_SINGLETON);`
> Which does seem very clean.
Beyond cleanl... -
- Right now the only way to check wether an object has a singleton class is to do something akin to:
```c
!SPECIAL_... -
04:49 PM Feature #18554 (Open): Move unicode_normalize to a default gem
- Could we move the rest of unicode_normalize to a default gem?
The recent updates were mostly updating the Unicode ... -
- I opened a draft for it: https://github.com/ruby/ruby/pull/5500
@ko1 if you'd like to have a look. -
02:30 PM Bug #18553 (Closed): Memory leak on compiling method call with kwargs
- The following code produces a memory leak:
```ruby
p(foo: 1)
```
It comes from the allocation in `compile.c`:... -
11:54 AM Revision d66e7ec7 (git): [ruby/reline] Add a comment for VK_MENU
- https://github.com/ruby/reline/commit/ee307c2b01
-
- On European keyboards.
https://github.com/ruby/reline/commit/75fe9759a4 -
- I took the liberty to set the backport field.
I was able to repro on 2.6, 2.7,3.0 and 3.1. However 2.6 is security... -
06:46 AM Bug #18516 (Closed): Memory leak on aliasing method to itself
- Applied in changeset commit:git|e89d80702bd98a8276243a7fcaa2a158b3bfb659.
----------
Fix memory leak at the same nam... -
- The error here is confusing for users because JRuby does not use
libyaml and installing it will not help. Instead, JR... -
- When aliasing a method to the same name method, set a separate bit
flag on that method definition, instead of the ref... -
05:58 AM Feature #18551 (Closed): Make `Range#reverse_each` to raise an exception if endless
- https://github.com/ruby/ruby/pull/5498
Currently, `Range#reverse_each` for an endless range never returns.
```
... -
-
- Applied in changeset commit:git|6a6227e0168b059c3ed34c9f0ace2e5dc2364221.
----------
Shifting zero always results in... -
-
04:51 AM Bug #18514: ruby 2.6.9's Date#jisx0301 does not support Reiwa
- I released date-2.0.3 with Reiwa support.
@kouji Can you confirm with [date-2.0.3](https://rubygems.org/gems/date/... - 03:33 AM Revision 99d02cae (git): [ruby/logger] Fix log rotation inter-process lock failed.
- Issue only occurs in JRuby 9.3.0.0 and Windows and the full
console output is:
log rotation inter-process lock faile... -
- https://github.com/ruby/ruby/pull/5496
01/26/2022
-
- Seems reasonable to me.
-
01:46 AM Feature #13110: Byte-based operations for String
- Are there any objections to introduce only byteindex and byteoffset?
-
08:28 PM Revision d3d888b9 (git): [lldb] Handle MacOS 64Kb heap pages in the lldb helpers
-
- @shenningz
> Is there any way to replace the yajl-ruby C extension with another json extension?
Depending on y... -
07:07 PM Feature #18408: Allow pattern match to set instance variables
- Eregon (Benoit Daloze) wrote in #note-17:
> In practice it should rarely matter because if a clause isn't matched th... -
05:02 PM Revision 1a180b7e (git): Streamline cached attr reader / writer indexes
- This commit removes the need to increment and decrement the indexes
used by vm_cc_attr_index getters and setters. It ... -
04:29 PM Bug #18389 (Closed): `binding.irb` can fail in some classes that implement `context` and `print` methods.
- Applied in changeset commit:git|b6b2e489f10bfc771beae8fa0ee5237613acd18f.
----------
[ruby/irb] Changed to call Kern... - 04:29 PM Revision 3ce97a18 (git): * 2022-01-27 [ci skip]
-
- If you call `binding.irb` on a class defined `#print`, it will crash, so call `Kernel.print`.
Fix [Bug #18389] `bind... -
- https://github.com/ruby/ruby/pull/5494
-
- I've been looking at Integer#{<<,>>} with RHS bignum and long and there are several issues.
This seems a clear bug... -
- https://github.com/ruby/ruby/pull/5493
-
03:20 PM Bug #18516: Memory leak on aliasing method to itself
- ibylich (Ilya Bylich) wrote in #note-1:
> but if we stop increment `alias_count` on `alias foo foo` the warning appe... -
- As @nobu mentioned the following code gives no warnings (in verbose mode):
```ruby
class A
def foo; end
ali... -
- The following code produces a memory leak:
```ruby
class A
1.upto(Float::INFINITY) do |i|
define_method(:... -
- Repro:
```ruby
exp = 2**40 # also fails with bignum e.g. 2**64
def exc
begin
yield
rescue NoMemoryErr... -
02:45 PM Revision e7141630 (git): Fix typo in assertion in gc.c
-
- As MinGW has the declaration, the `dllimport` attribute difference
is warned when compiling missing/*.c without inclu... -
- Thanks @tenderlove, your explanation makes sense to me.
From your description we should be able to reproduce with... -
-
- PR is ~~https://github.com/ruby/ruby/pull/5489~~ https://github.com/ruby/ruby/pull/8525
Current `Range#reverse_eac... -
-
04:09 AM Revision c925d3b6 (git): [rubygems/rubygems] Fix `force_ruby_platform` ignored when lockfile includes the current specific platform
- https://github.com/rubygems/rubygems/commit/9ca371adf8
-
- I prepared to fix this regression.
https://github.com/ruby/date/pull/47
I will release date-2.0.3 for Ruby 2.6 ... -
02:11 AM Bug #18514 (Closed): ruby 2.6.9's Date#jisx0301 does not support Reiwa
-
01:45 AM Bug #18514 (Closed): ruby 2.6.9's Date#jisx0301 does not support Reiwa
- In ruby 2.6.9 (ruby 2.6.9p207 (2021-11-24 revision 67954) [x86_64-darwin20]).
$ ruby -rdate -e 'p Date.new(2019, 5, ... -
01:52 AM Feature #18513: Hide patchlevel from `ruby -v`
- :+1:, patch levels no longer serves us very well.
-
01:05 AM Revision cc8064ba (git): Do not need to print to stderr
-
- don't duplicate the message, but prepare `out` var to choose
stdout/err. -
- Now all failure messages are printed to stdout. This option
makes all failure messages printed into stderr.
01/25/2022
-
11:57 PM Bug #18501: [BUG] try to mark T_NONE object in RubyVM::InstructionSequence. load_from_binary
- ## Hash is writing T_NONE references
I think it's possible that T_NONE objects are being written in to a hash. Ba... -
- Iteration should top if the replace function returns ID_TABLE_STOP.
-
- These places never replace the value, so call rb_id_table_foreach_values
instead of rb_id_table_foreach_values_with_r... -
- Renames rb_id_table_foreach_with_replace to
rb_id_table_foreach_values_with_replace and passes only the value to the
... -
- This reverts commit 530e485265dac6e2aea1d587a4a79a314cc772cf.
`rb_id_table_foreach_with_replace` is used during GC co... - 05:24 PM Revision ac574052 (git): * 2022-01-26 [ci skip]
- 05:24 PM Revision 42ef3fca (git): [rubygems/rubygems] Use Gem::Platform.local instead of RUBY_PLATFORM
- In certain places, we want to display the platform name with
`Gem::Platform.local` instead of `RUBY_PLATFORM`.
Fixes... -
- The if statement is redundant since if `index == 0` then
`BITS_BITLENGTH * index == 0`. -
-
- And not depend on the state of rack's master branch, in particular, on
their Ruby support range.
https://github.com/... -
- https://github.com/rubygems/rubygems/commit/ef4e5c6169
-
- We still show patch-level like `3.1.0p0` with `ruby -v`. But it has no benefit for users.
In fact, I heard "What ... -
11:29 AM Bug #18512 (Feedback): MacOS 12.1 Monterey Bug
- Could you try this? https://github.com/ffi/ffi/issues/800#issuecomment-1008308614
-
- I guess this is related to #18496, and maybe the cause is FFI with M1.
-
10:41 AM Bug #18512 (Third Party's Issue): MacOS 12.1 Monterey Bug
- Steps to Reproduce
1. Create an Zendesk App: https://developer.zendesk.com/documentation/apps/app-developer-guide/za... -
08:57 AM Bug #18507: Incorrect target_os detection in configure script
In case 'arm-unknown-linux-gnueabihf' GNU Autotools recognizes target_os as 'linux-gnueabihf' (and 'linux-gnu' if '...-
- Applied in changeset commit:git|a51e1cd682daeb76f201e62eaedcf6d51a18cffe.
----------
enable `RUBY_ON_BUG` on release... -
- [Feature #18483]
-
- Another name proposal:
When I write `array.min_with_value { foo }`, I mean "minimize foo with an element in array"... -
02:15 AM Feature #18490: MakeMakefile.pkg_config should accept multiple options
- @nobu, I wanted to check if there was any other feedback on this feature? Thank you again for your time.
-
12:33 AM Bug #18437 (Closed): 3.1.0 install stucks on FreeBSD & NFS
-
12:23 AM Bug #18437: 3.1.0 install stucks on FreeBSD & NFS
- Oh, I see. Thanks for the fix.
I think this issue has been resolved.
＃ 自分で Issue 閉じれないんですね……。
01/24/2022
-
- The gem doesn't even install on old rubies, but since the gemspec claims
it's supported, `gem install pathname` will ... -
- Applied in changeset commit:git|530e485265dac6e2aea1d587a4a79a314cc772cf.
----------
`ID` in `rb_id_table_foreach_wi... -
- Pass the `ID` from `rb_id_table_foreach_with_replace` to callback
functions. -
10:40 PM Feature #18273: Class#subclasses
- I see, and tend to agree—to some extent, at least!
Not sure how much can be done here—some amount of "semantically... -
03:45 PM Feature #18273: Class#subclasses
- More than the existence of a real collection (which is a private implementation detail), it is the existence of a cle...
-
- @zverok Sorry, we were typing at the same time.
Exactly! That is my point of view! -
- @zverok I think our disagreement is clear. Perhaps your position is also the one of @byroot.
To me, `Class#subclas... -
- @fxn You know what?
I think I actually got your point:
* `Class#ancestors` represent a "really existing thing", ... -
- > In the sense that, as a programmer, if I do not change the collection, it should stay the same.
Of course, removin... -
- @zverok sure!
I see your point, it works as it is supposed to be.
My point is, `Class#subclasses` should be det... -
- We are going in circles here :( It is "non-deterministic" all right, but also reflects the reality.
It is more or ... -
- @zverok one thing is "subclasses can vanish", because the language is dynamic (like ancestors can change), and a diff...
-
- TBH, I still fail to see the point of how the API clearly reflecting the reality should be considered a "broken windo...
-
- Hmm, now that I think, maybe you don't see how this affects Zeitwerk directly.
In Zeitwerk, "reloading" means issuin... -
- @zverok Absolutely, the API of a dynamic language should be dynamic.
For example, take this:
```ruby
C = Class... -
- @fxn Isn't it an inherent property of a dynamic language, that whatever you can introspect about the code structure, ...
-
- To me, this feature makes sense in a language where subclasses cannot disappear once created. In Ruby, they are regul...
-
- > Sure, and you are entitled to it, I just don't understand how you reconcile it with the existence of weak reference...
-
- Sure, and you are entitled to it, I just don't understand how you reconcile it with the existence of weak references.
-
- That is, _since_ objects are not gced as soon as they are unreachable, this implementation of `Class#subclasses` is n...
-
- To me it behave just like weak references do:
```ruby
weakmap = ObjectSpace::WeakMap.new
weakmap[Object.new] = O... -
- What I mean is that `Class#subclasses` is effectively a cache that does not have proper invalidation.
The red flag... -
- I don't think I understand what your issue is. `subclasses` returns all the *live object* (not yet GCed) that are dir...
-
- > This is perfectly fine, Ruby isn't concerned about wether a Class is still present in the constant table or not
... -
- > In my view, this method does not play well with object lifetime, semantically.
This is perfectly fine, Ruby isn'... -
- Said in a different way:
```ruby
C = Class.new
-> { Class.new(C); 1 }.call
pp C.subclasses
pp C.subclasses
``... -
- In my view, this method does not play well with object lifetime, semantically.
Take for example:
```ruby
C = Class.... -
07:57 PM Bug #18511: ruby in infinite loop consuming 100% CPU
- Right now we cannot reproduce this systematically, it just occurs with our Fluentd workload sooner or later.
Fluentd... -
04:06 PM Bug #18511 (Feedback): ruby in infinite loop consuming 100% CPU
- This happens inside the yajl-ruby C extension. Can you reproduce this without yajl-ruby?
-
- After running Fluentd (Td-agent) workload for several hours or days, Ruby hangs in an infinite loop.
gdb stack tra... -
- NUM_IN_PAGE could return a value much larger than 64. According to the
C11 spec 6.5.7 paragraph 3 this is undefined b... - 04:11 PM Revision 97ab2599 (git): * 2022-01-25 [ci skip]
-
- https://github.com/rubygems/rubygems/commit/68bef90339
-
- WebAssembly doesn't support signals so we can't use read
barriers so we can't use compaction. -
-
- Even since 0.1.0, other than +true+ or +false+ may be returned.
https://github.com/ruby/io-wait/commit/d0721e300a -
03:12 AM Revision e7b573e5 (git): [ruby/io-wait] Document mandatory require for using io/wait methods
- https://github.com/ruby/io-wait/commit/2beb05bd48
- 01:57 AM Revision c6a19b77 (git): Update default gems list at 328e6bf3b3a167529e5c64a281e773 [ci skip]
- 01:56 AM Revision 328e6bf3 (git): [ruby/bigdecimal] Version 3.1.2
- https://github.com/ruby/bigdecimal/commit/57e2194135
- 01:56 AM Revision 7db195d5 (git): [ruby/bigdecimal] Fix the maximum precision of the quotient
- Fixes https://github.com/ruby/bigdecimal/pull/220
https://github.com/ruby/bigdecimal/commit/127a1b5a31 - 01:56 AM Revision 506728d0 (git): [ruby/bigdecimal] Fix typo
- https://github.com/ruby/bigdecimal/commit/79c09b4dac
- 01:17 AM Revision d77f2ff5 (git): [ruby/rdoc] Fix typo in RDoc::Task example
- https://github.com/ruby/rdoc/commit/7a77e55c2a
01/23/2022
- 10:00 PM Revision 52caeb47 (git): * 2022-01-24 [ci skip]
-
10:00 PM Revision c1554457 (git): [ruby/reline] Ignore global constants when checking if Fiddle::VERSION exists
- If a top-level `VERSION` constant exists, or if a module containing a `VERSION` constant is included into the top-lev...
-
02:57 PM Feature #18481: Porting YJIT to Rust (request for feedback)
- Numba uses llvmlite https://github.com/numba/llvmlite for its JIT, but seems to need many patches. They also primaril...
-
02:53 PM Bug #18510 (Open): Unexpected waiting for console when starting ruby on windows
- # ruby version
C:\Ruby31-x64\bin>"c:\Ruby31-x64\bin\ruby.exe" -v
ruby 3.1.0p0 (2021-12-25 revision fb4df44d16) [x64... -
- That's great to hear. Yes, we fixed this differently by avoiding flock on non Windows platforms, since it didn't seem...
-
- I tried installing the latest development version of Ruby and RubyGems, and successfully completed it.
```
% rben... -
12:56 PM Revision bb955096 (git): Fix a typo [ci skip]
-
- I sent the core files to Aaron privately.
01/22/2022
-
- Interesting. Are you able to get a core file? I can poke around that code, but a core file would be very helpful.
... -
- It seems like intentional, as there is the second substitution.
Is arm-linux-eabihf incorrect? -
- Nice find!
Agreed this should be fixed, and `Kernel#binding` should never provide access to anything but its direct ... -
- I submitted a pull request to make `Kernel#binding` only look up a single frame, which fixes the issue: https://githu...
- 04:02 PM Revision ab1aa90c (git): * 2022-01-23 [ci skip]
-
-
- Autoconf 2.69 fails to detect `-Werror=old-style-definition` due
to the old style definition of `main`. -
-
07:26 AM Feature #16757: Add intersection to Range
- I'd like to see this explored, as I find myself solving a number of problems that involve Range merging, and potentia...
-
- https://rubyci.s3.amazonaws.com/debian-riscv64/ruby-master/log/20220122T050018Z.log.html.gz#miniruby
```
compiling vm... -
01:00 AM Revision c1a6ff04 (git): [ruby/psych] Add strict_integer option to parse numbers with commas as strings
- Authored-by: Seth Boyles <[email protected]>
https://github.com/ruby/psych/commit/75bebb37b8
01/21/2022
- 10:35 PM Revision b815a0bd (git): * 2022-01-22 [ci skip]
-
10:34 PM Revision fc6fd4c3 (git): Accurately report VM memsize
- Currently the calculation only counts the size of the struct. This commit adds the size of the associated st tables, ...
-
- This code doesn't come from Ruby, it comes from ruby-yajl (an external project): https://github.com/brianmario/yajl-r...
-
09:36 PM Bug #18509 (Third Party's Issue): yajl.rb - (# TODO: this code smells, any ideas?) - "ArgumentError"
- Please clarify - can we trust this code in yajl-ruby gem yajl.rb file?
class Encoder
def self.encode(obj, *a... -
- Could you give us how to reproduce the issue?
I guess this is related to #18496, and maybe the cause is FFI. -
01:27 PM Bug #18505 (Third Party's Issue): Crash reported while trying to run Cucumber via ruby on macbook air M1 processor
- Crash reported while trying to run Cucumber via ruby on macbook air M1 processor
-
- Looks like `irb --noautocomplete` turns off autocomplete. Or add `IRB.conf[:USE_AUTOCOMPLETE] = false` to `.irbrc`.
-
05:27 PM Feature #18508 (Closed): Is it possible to turn off irb autocomplete, it doesn't work very well in my visual studio code terminal.
- Is it possible to turn off irb autocomplete, it doesn't work very well in my visual studio code terminal.
When the c... -
- > How about 3.2.0.12345 then?
> It is a bug of any software to assume exactly 3 digits for ABI version, for instance... -
- Also from experience in TruffleRuby I can say this is a solution working very well to avoid users unintentionally reu...
-
- naruse (Yui NARUSE) wrote in #note-9:
> For dev builds, Ruby switchers has responsibility to handle the ABI compatibi... -
- > Is this to tell if an installed gem set is compatible, at switching ruby versions?
Yes
> If so, https://githu... -
- Regarding ruby/setup-ruby, can’t it use `hashFiles('include/**/*.h')` as the cache keys?
-
08:02 AM Bug #18249 (Rejected): The ABI version of dev builds of CRuby does not correspond to the ABI
- For dev builds, Ruby switchers has responsibility to handle the ABI compatibility breakage.
CRuby developers don't w... -
- Is this to tell if an installed gem set is compatible, at switching ruby versions?
If so, https://github.com/ruby/ru... -
configure.ac incorrectly detect target_os by target triplet such as arm-unknown-linux-gnueabihf. In case arm-unknow...-
01:28 PM Bug #18506 (Open): make and make install rebuild items every time unnecessarily - sometimes causing races in parallel installs
- The makefiles are written in a way that a few items are regenerated on every invocation of make, even if nothing chan...
-
- I know what is going on. About the only thing we can do is to settle for that warning. It works nonetheless.
The... -
- ```
$ sudo apt-get install gcc-aarch64-linux-gnu
$ ./configure --host=aarch64-linux-gnu
downloading config.guess .... -
- Do not print anymore except errors.
- 03:23 AM Revision a15c50d6 (git): * 2022-01-21 [ci skip]
-
- https://github.com/ruby/etc/commit/c8a133fe95
-
- https://github.com/ruby/etc/commit/e8ecce3442
-
- https://github.com/ruby/etc/commit/5cabc3996a
-
- Thanks. Most probably you have OpenSSL higher version (OpenSSL version >= 1.0.1 and < 3.0.0) pre-installed by default...
-
- On our Solaris, it is possible to build without any option about OpenSSL. Maybe you need setup your system properly. ...
-
- Thanks, Yusuke. Can you also help us to direct gmake openssl extension to the right version of OpenSSL?
We see er... -
- Well, I don't think `ln -s /opt/csw/lib/libgmp.so.3 /usr/lib/libgmp.so.3` is a right solution. Have you tweaked your ...
-
- Duplicated with https://github.com/ruby/openssl/issues/494
Also available in: Atom