Activity
From 08/18/2024 to 08/24/2024
08/24/2024
-
11:17 PM Bug #20680 (Closed): No "void context" warning for literals/etc. in ensure block
- Applied in changeset commit:git|620ce3807b51e126bfb41b9b87ff7b33584058f2.
----------
[Bug #20680] `ensure` block is ... -
-
10:04 PM Bug #20694: "try to mark T_NONE object" error for very large hash
- Ah yes, I meant hash.
I’ll upload the full back trace when I’m back online later today. -
09:47 PM Bug #20694: "try to mark T_NONE object" error for very large hash
- > very large arrays
You mean hashes right?
This smells like a missing write barrier in the compiler or iseq loa... -
06:15 PM Revision 519152db (git): [ruby/rdoc] Make the summary triangle appear in the correct place
- when the summary text overflows to next line.
(https://github.com/ruby/rdoc/pull/1160)
https://github.com/ruby/rdoc/... -
03:51 PM Revision fd0263d7 (git): [DOC] Tweaks for Array#concat
-
-
-
-
11:07 AM Revision 6717b690 (git): [ruby/rdoc] Modernize RDoc Darkfish template CSS
- (https://github.com/ruby/rdoc/pull/1157)
- Update color scheme with muted green tones and improved contrast
- Enhanc... -
08:08 AM Bug #20698 (Closed): Please backport fix for CVE-2024-43398
- This is similar request to #20667, but since that was already partially done, I have decided to open new ticket
-
01:53 AM Feature #20625: Object#chain_of
- @matz since we don't have a `not_nil?` method, what do you think about adding a `take_until` method to Enumerable or ...
08/23/2024
-
11:45 PM Bug #20693: Dir.tmpdir should perform a real access check before warning about writability
- I did think about this, but the way the stat methods work is that the `stat(2)` syscall once on the path and returns ...
-
05:59 PM Bug #20693: Dir.tmpdir should perform a real access check before warning about writability
- What about changing/fixing `stat.writable?` to behave like `File.writable?`
It seems to me a source of confusion and ... -
- I have a PR for doing this: https://github.com/ruby/ruby/pull/11403
But actually it breaks Bundler's tests, becaus... -
- @mame reported:
```ruby
Rational("1r") #=> invalid value for convert(): "1r" (ArgumentError)
```
https://gith... -
- I expected the following code captures the error message, but `err:` option is ignored.
```shell-session
$ ruby -... -
10:48 PM Bug #20691 (Closed): Use-after-free in WeakKeyMap#clear
- Applied in changeset commit:git|552f5cd50c8b622294e2dc87a73eb9e1102d78a0.
----------
Fix use-after-free in WeakKeyMa... -
- I opened a backport PR to 3.3: https://github.com/ruby/ruby/pull/11443
-
- Fix use-after-free in WeakKeyMap#clear
[Bug #20691]
If the WeakKeyMap has been marked but sweeping hasn't started y... -
10:03 PM Bug #20695 (Closed): Elevated GC allocations in parse.y parser
- Testing on the `lobsters` benchmark from [`yjit-bench`](https://github.com/shopify/yjit-bench), the latest `master` p...
-
- [Bug #20691]
If the WeakKeyMap has been marked but sweeping hasn't started yet and we
cann WeakKeyMap#clear, then th... -
-
07:18 PM Revision 3eb42054 (git): [ruby/prism] Pass Unicode escapes on to onigmo
- When we encounter an invalid unicode escape within a regular
expression, we now pass that error on to Onigmo as if it... -
- We can improve object ID performance by checking the FL_SEEN_OBJ_ID flag
instead of looking up in the table. - 11:44 AM Revision b51e1c07 (git): [DOC] Fix typos in ObjectSpace::WeakMap docs
- The value of variable key2 should be "bar". This way, when nil is assigned to val1 and garbage collection occurs, the...
-
- In https://bugs.ruby-lang.org/issues/20693, I'd like to have Dir.tmpdir
call `File.writable?` instead of `Stat#writab... - 07:00 AM Revision 73a946c6 (git): Update bundled gems list as of 2024-08-23
-
07:00 AM Revision 9f586040 (git): [rubygems/rubygems] Fix error message when Bundler refuses to install due to frozen being set without a lockfile
- https://github.com/rubygems/rubygems/commit/0857d62ca6
-
06:07 AM Revision 7812732e (git): [ruby/tempfile] File.new(fileno, mode: mode, path: path) is provided from Ruby 3.2
- https://github.com/ruby/tempfile/commit/67ce897727
-
-
-
- (https://github.com/ruby/fiddle/pull/143)
Pick
https://github.com/ruby/ruby/commit/92865d8760e22dc5035a67e636fab3fbd... -
02:19 AM Feature #20673: Enable native SOCKS support by default
- Enabling SOCKS support depends `SOCKS_SERVER` too now, I think it is unusual.
So I agree to enable by default.
I ... -
-
12:50 AM Bug #20694: "try to mark T_NONE object" error for very large hash
- Can you share the full backtrace, not only "Control frame information" section but also "C level backtrace informatio...
-
- In CI, we've been occasionally seeing `[BUG] try to mark T_NONE object`. I don't have any way to reproduce it but I'm...
08/22/2024
-
- 11:34 PM Revision 8800127d (git): Skip some tests which don't work under permissionless containers
- When running as UID 0 but without CAP_DAC_OVERRIDE (for example, in a
docker container running with --uid 0 but --cap... -
- The code in `Dir.tmpdir` attempts to warn the user if their temp directory is deficient for some reason:
```ruby
... -
- * Add struct weakmap_entry for WeakMap entries
* Refactor wmap_foreach to pass weakmap_entry
* Use wmap_foreach for... -
09:04 PM Feature #20692 (Open): Rewrite Array#bsearch in Ruby
- inspired by https://bugs.ruby-lang.org/issues/20182
# Proporsal
Rewrite Array#bsearch
```ruby
class Array
... -
- matz (Yukihiro Matsumoto) wrote in #note-31:
> It seems this change breaks step over string ranges (e.g. `"a".."z").... -
01:04 PM Feature #18368: Range#step semantics for non-Numeric ranges
- @mame It doesn’t take long, but I’ll have time for that on the weekend (most probably Sunday).
I hope that some cl... -
- @zverok Please fix the incompatibility first before having such a discussion. If it takes long, I will have to revert...
-
- @matz I can certainly implement the specialization, but just to clarify: are there any evidence that people use `("a"...
-
03:11 AM Feature #18368: Range#step semantics for non-Numeric ranges
- If we take this compatibility as important and go with it, should we give up `("a".."aaaa").step("a")` or support it?
-
- We have little choice but to specialize it for strings as we don't want to add support for `"a" + 3` that will most c...
-
01:29 AM Feature #18368 (Open): Range#step semantics for non-Numeric ranges
- It seems this change breaks step over string ranges (e.g. `"a".."z").step(3)`). We need to handle string ranges speci...
-
- I opened a PR to backport to Ruby 3.3: https://github.com/ruby/ruby/pull/11439
-
- Applied in changeset commit:git|df9a6aa94330cbf414afcd957d1b87defc67e1c5.
----------
Fix WeakMap use-after-free
[Bu... -
- Using gc_impl.h inside of gc/gc.h will cause gc/gc.h to use the functions
in gc/default.c when builing with shared GC... -
- When hash_foreach_replace_value returns ST_REPLACE, it's guaranteed that
the value has moved in hash_replace_ref_value. -
- GitHub PR: https://github.com/ruby/ruby/pull/11437
If the WeakKeyMap has been marked but sweeping hasn't started y... -
- https://github.com/ruby/prism/commit/7f7620b6d5
- 03:23 PM Revision 0ef703b9 (git): [ruby/prism] Render node comments into their `ast.h`
- https://github.com/ruby/prism/commit/b17c3241d3
-
- https://github.com/ruby/prism/commit/098b3f08bc
-
- [Bug #20688]
We cannot free the key before the ST_DELETE because it could hash the
key which would read the key and ... -
- [Bug #20688]
We cannot free the weakmap_entry before the ST_DELETE because it could
hash the key which would read th... -
-
-
-
-
-
- The original implementation of this flag was too naive and all it did
was restricting gems to locally installed versi... -
- https://github.com/rubygems/rubygems/commit/bea4c1ad79
-
10:24 AM Misc #20013: Travis CI status
- jaruga (Jun Aruga) wrote in #note-37:
> I am seeing Travis infra errors. It started at least since 13th August 2024 ... -
- Sent a PR to allow failures for ppc64le/s390x on ruby/prism.
https://github.com/ruby/prism/pull/3005
-
- I am seeing Travis infra errors. It started at least since 13th August 2024 until now. Now the errors only happen on ...
-
09:33 AM Bug #20690 (Rejected): URI.encode_www_form_component method escapes tilde when it's not supposed to
- ## Problem
As the title says, `Uri.encode_www_form_component` is escaping tilde (`~`) where it should not according... -
08:03 AM Bug #20682: Slave PTY output is lost after a child process exits in macOS
- @akr
I'd appreciate any ideas you may have. -
- Here is the summary that we know so far.
## Current PTY's behavior
The slave PTY output is lost after a child p... -
- @mame and I have noticed that there is a problem in the workaround solution: https://github.com/ruby/ruby/pull/11404
... -
08:03 AM Bug #20586: Some filesystem calls in dir.c are missing error handling and can return incorrect results if interrupted
- jeremyevans0 (Jeremy Evans) wrote in #note-11:
>
> I've submitted a pull request to add error checking to `getlogi... -
-
- Applied in changeset commit:git|86c2724e75375ceb165c378ed4261c551136e593.
----------
Don't emit ELF notes on non-ELF... -
- These apparently break compilation on old MacOS toolchains, because the
MachO section name is capped to 16 chars (alt... -
- Linux calls it aarch64, but MacOS calls it arm64; pac-ret works on both.
-
- This wasn't looking at the right macro name for pac-ret support, so if
Ruby was compiled with pac-ret but NOT BTI, th... -
- It is expected that reading from command with offset fails by ESPIPE
and the pipe will be closed immediately. While ... - 07:01 AM Revision 29500e30 (git): Update bundled gems list as of 2024-08-22
-
- `make test-all TESTS="test/win32ole/test_folderitem2_invokeverb.rb"` shows the following warnings.
```
2024-08-22... -
- Dependabot will update matching version comments.
- 03:25 AM Revision 0846bcbd (git): Bump github/codeql-action from 3.26.3 to 3.26.4
- Bumps [github/codeql-action](https://github.com/github/codeql-action) from 3.26.3 to 3.26.4.
- [Release notes](https:... -
- ```
[ 14/191] TestParallel::TestParallel#test_hungup = 11.02 s
1) Failure:
TestParallel::TestParallel#test_hungup [... -
02:20 AM Revision a3f5a043 (git): Handle getlogin failure in PTY.spawn
- getlogin is only called if USER environment variable is not set,
but if getlogin returns NULL in that case, then do n... -
- getlogin can return NULL, and this can avoid a segfault
in that case.
Mentioned as an issue in comment to bug 20586. -
-
- ```
Leaked file descriptor: TestResolvDNS#test_multiple_servers_with_timeout_and_truncated_tcp_fallback: 15 : #<TCPSo...
08/21/2024
-
10:52 PM Bug #20586: Some filesystem calls in dir.c are missing error handling and can return incorrect results if interrupted
- jeremyevans0 (Jeremy Evans) wrote in #note-10:
> ivoanjo (Ivo Anjo) wrote in #note-9:
> > The only (potential?) iss... -
-
- nobu (Nobuyoshi Nakada) wrote in #note-1:
> In that case, the system ruby is used to re-check encoding sources and u... -
- Applied in changeset commit:git|6ab591f80aa19d63ecd1e1df3c09c391efb318a6.
----------
[Bug #20687] Check if base ruby... -
03:14 AM Bug #20687: Does tarball require baseruby?
- Thanks for looking into this, @hsbt and @nobu! ❤️
-
- In that case, the system ruby is used to re-check encoding sources and update encdb.h if needed.
Encodings under enc... -
- From https://github.com/rbenv/ruby-build/issues/2435
I'm wondering why tarball use system ruby with above environm... -
-
- If an array element is a static literal that does not result in a
intermediate array, it still needs to be compiled n... -
-
-
- * Make sure to always use the right `warn`
* lib/bundled_gems.rb: more reliable caller detection
The `2` skipped fr... -
-
-
-
-
-
03:37 PM Revision 7ad74d16 (git): [PRISM] Implement unused block warning
- Related: ruby/prism#2935
-
- We don't need to build a linked list from the finalizer table and
instead we can just run the finalizers by iterating... -
- The `file://` protocol does not really work with relative paths and it's
not necessary anyways. So restore support fo... -
- Doesn't really change the test outcome, but I noticed it was missing.
https://github.com/rubygems/rubygems/commit/ea... -
-
- GitHub PR: https://github.com/ruby/ruby/pull/11421
We cannot free the weakmap_entry before the ST_DELETE because i... -
-
- https://github.com/rubygems/rubygems/commit/29357a5dd6
-
- https://github.com/rubygems/rubygems/commit/9d41c4088a
-
- https://github.com/rubygems/rubygems/commit/417319ecb1
-
- On Windows, an executable file in the current directory has the
priority to the PATH environment variable always. -
-
-
-
-
- This reverts commit 3da7e440e9fca835f5475a98f1c0afb4d2ac71db.
-
04:55 AM Revision ea828c0b (git): Skip failing tempfile library tests
-
- 01:10 AM Revision 014708ad (git): [ruby/logger] Guarantee level_override exists
- Some Ruby apps subclass Logger without running the superclass
constructor, which means that `@level_override` isn't i...
08/20/2024
-
- https://github.com/ruby/tempfile/commit/f677941ea2
-
- Using `-rtempfile` requires the tempfile built into Ruby, not the
currently developed one, so the tests aren't testin... -
-
- As @jeremyevans pointed out for commit eb2d8b1:
> Each Tempfile instance has a separate File instance and file descr... -
- The Closer and Remover finalizers are defined on different objects in
Tempfile. The Closer is defined on the Tempfile... -
- See https://github.com/rubygems/rubygems/pull/7939
-
- https://github.com/ruby/reline/commit/d44fbf9f7b
- 09:36 AM Revision 927a44b4 (git): Rewrite #test_redefinition_mismatch to use a dedicated test class
- This test is checking what happens if you try and define a class in a C
extension where that constant is already not ... -
-
- I've finished account setup for ono-san.
-
- @hsbt
Yes. I sent the email to the address. -
- @ono-max Can you give Required and Optional information of https://github.com/ruby/ruby/wiki/Committer-How-To#how-to-...
-
-
- 02:48 AM Revision edd3fc62 (git): Bump github/codeql-action from 3.26.2 to 3.26.3
- Bumps [github/codeql-action](https://github.com/github/codeql-action) from 3.26.2 to 3.26.3.
- [Release notes](https:... -
- Applied in changeset commit:git|ea7ceff82cec98d0c419e9807dcb33dcc08b56fa.
----------
Avoid hash allocation for certa... -
- Previously, proc calls such as:
```ruby
proc{|| }.(**empty_hash)
proc{|b: 1| }.(**r2k_array_with_empty_hash)
```
bo... -
08/19/2024
-
- @matheusrich In my opinion `take_until` might be an interesting method to add, but I think we are unnecessarily compl...
-
- > The one problem we don’t currently have neither Enumerable#take_until, nor Object#not_nil?, to write something like...
-
- jhawthorn (John Hawthorn) wrote in #note-2:
> jeremyevans0 (Jeremy Evans) wrote in #note-1:
> > It appears RFC 3986... -
10:59 PM Bug #20686: URI::HTTPS can build URI with blank, invalid host
- Interestingly RFC2396_PARSER seems to allow nil for a host but not empty string, so the newer behaviour is at least m...
-
- jeremyevans0 (Jeremy Evans) wrote in #note-1:
> It appears RFC 3986 allows empty hosts (https://datatracker.ietf.org... -
- ronricardo (Roniece Ricardo) wrote:
> In Ruby 3.4.0+, calling `URI::HTTPS.build(host: "")` does not raise `URI::Inva... -
07:40 PM Bug #20686 (Open): URI::HTTPS can build URI with blank, invalid host
- In Ruby 3.4.0+, calling `URI::HTTPS.build(host: "")` does not raise `URI::InvalidComponentError` as expected. Instead...
-
- Anonymous wrote in #note-8:
> `rb_getpwdirnam_for_login` currently does not have any explicit
> handling for `EIN... -
- Hey! As usual thanks for picking up these thorny bugs :)
The fix looks good! In particular I've re-tested it with... -
- jeremyevans0 (Jeremy Evans) wrote in #note-3:
> Definitely related to the `ruby2_keywords` optimization. I'll keep d... - 07:21 PM Revision e398e84c (git): [rubygems/rubygems] Bump rb-sys
- Bumps [rb-sys](https://github.com/oxidize-rb/rb-sys) from 0.9.100 to 0.9.101.
- [Release notes](https://github.com/ox... - 05:55 PM Revision d1b2479b (git): [rubygems/rubygems] Bump rb-sys
- Bumps [rb-sys](https://github.com/oxidize-rb/rb-sys) from 0.9.100 to 0.9.101.
- [Release notes](https://github.com/ox... -
- When assertions are enabled, the following code triggers an assertion
error:
GC.disable
GC.start(immediate_m... -
- Avoids warnings like
```
/path/to/ruby/3.3.4/lib/ruby/gems/3.3.0/gems/rbs-3.4.0/lib/rdoc/discover.rb:10: warning: me... -
-
10:13 AM Misc #20685: Ruby 3.2.4 tag mentions unrelated changes (CVE-2024-27280)
- Thank you for your greate catch! You are totally right. ruby-3.2 seriese contains stringio-3.0.4 from the beginning.
... -
09:21 AM Misc #20685 (Closed): Ruby 3.2.4 tag mentions unrelated changes (CVE-2024-27280)
- # Problem
According to https://github.com/ruby/ruby/releases/tag/v3_2_4,
it mentions "CVE-2024-27280: Buffer over... -
09:20 AM Feature #20684 (Closed): Add optimized instructions for frozen literal Hash and Array
- # Context
Methods that take empty arrays or empty hashes as default values allocate a new object each time the met... -
-
- https://github.com/ruby/ruby/pull/11405 - I think this should fix it. It seems new toolchains truncate the section na...
-
- > deciding if BTI/PAC is available.
Silly me, this is about amd64.
So what you're running into is that I added ... -
- Approved.
Matz.
-
- +1
Our test suite and CI are complex and complicated. We need more maintainer for that. -
- I forgot to mention that he has expressed his willingness to accept committer's proposal here: https://bugs.ruby-lang...
-
- I'd like to propose @ono-max as a committer.
He has contributed to Ruby's test framework, starting with the introd... -
- mame (Yusuke Endoh) wrote in #note-2:
> The PR looks good to me.
>
> @ono-max Do you want to merge it yourself? I... -
- The PR looks good to me.
@ono-max Do you want to merge it yourself? If you are willing, I'd like to propose you as... -
- I created the PR: https://github.com/ruby/ruby/pull/11404
This change works in my macOS environment. -
- According to Launchable, the following PTY tests are flaky only on macOS.
https://app.launchableinc.com/organizati... -
12:04 AM Revision 70613595 (git): [rubygems/rubygems] Change new gem README template to have copyable code blocks
- https://github.com/rubygems/rubygems/commit/c805e9b558
08/18/2024
-
01:22 PM Bug #19374: Issue with Ractor.make_shareable with curried procs
- Hey y'all -
I'm commenting here to say that this bug seems to be the only way I was able to pass a Proc to a Ractor... -
- Applied in changeset commit:git|d450f9d6a28f01b7ca6030a925921dbf35cee439.
----------
Make Range#step to consistently... -
- Make Range#step to consistently use + for iteration [Feature #18368]
Previously, non-numerics expected step to be in... -
- ruby_3_2 commit:21c708ee802e1a59901eccc6448e40e8f72189b8 merged revision(s) commit:992596fb7af18a7f472589a607d0eb3fbb...
-
- Fix next inside block argument stack underflow
[Bug #20344]
Fix compile_next adding removable adjust... -
- Merged at https://github.com/ruby/ruby/commit/65fed1c3e439bc47bcf6ec884431a86cb9ebd1dc. Thank you!
-
09:02 AM Revision 65fed1c3 (git): Allow waitpid(-1, Process::WNOHANG) to be woken if a waitpid(pid) is
- ...pending
If two threads are running, with one calling waitpid(-1, Process::WNOHANG),
and another calling waitpid($... -
08:30 AM Bug #20681: Regular expression warnings are treated as compiler warnings during runtime
- nobu (Nobuyoshi Nakada) wrote in #note-1:
> You may want to re-define `Warning.warn`.
Thank you, that works. Apol... -
- You may want to re-define `Warning.warn`.
-
03:51 AM
Bug #20586: Some filesystem calls in dir.c are missing error handling and can return incorrect results if interrupted
- On 2024-08-16 21:59:04, "jeremyevans0 (Jeremy Evans) via ruby-core" <[email protected]> spake thus:
>Issue... -
- ruby_3_2 commit:a8a25291447e3630d455e82851ca58fc8ac372ec merged revision(s) commit:97449338d6cb42d9dd7c9ca61550616e7e...
-
- ruby_3_2 commit:50399eebd96c76ce808ea4d84fe39693f585a531 merged revision(s) commit:1870505f478cc75993b296b7144a45137a...
-
- ruby_3_2 commit:50399eebd96c76ce808ea4d84fe39693f585a531 merged revision(s) commit:1870505f478cc75993b296b7144a45137a...
-
- Fix wrong unreachable chunk remove when jump destination label is unremovable
-
- Thank you for preparing the backport PRs. They were merged.
-
- [Bug #20649] Allow `nil` as 2nd argument of `assign_error`
Fallback to the last token element in that case, ...
Also available in: Atom