Activity
From 08/31/2024 to 09/06/2024
09/06/2024
-
09:55 PM Bug #20718: Objects created with Data_Make_Struct and the default free function are not freed
- I have added a pull request that should patch the issue. https://github.com/ruby/ruby/pull/11563 - I was able to build Ruby and run it against my example and it does work.
Presumably this would be a candidate for a backport :)
-
- I discovered a memory leak when using the FFI gem prior to version 1.16 and Ruby 3.3 and up.
During debugging I found that this earlier version of FFI uses Data_Make_Struct (https://github.com/ffi/ffi/blob/v1.15.5/ext/ffi_c/Pointer.c#... -
09:50 PM Bug #20710: Reducing Hash allocation introduces large performance degradation (probably related to VWA)
- I implemented @byroot's suggestion in this PR: https://github.com/ruby/ruby/pull/11562
It significantly improves the performance in your benchmark and brings it almost as fast as Ruby 3.2. - 09:26 PM Revision 6dc93846 (git): [ruby/rdoc] Add more space after magnifying glass
- (https://github.com/ruby/rdoc/pull/1173)
https://github.com/ruby/rdoc/commit/6a9cad4c54 -
08:51 PM Revision 731805dd (git): Add an explicit check for broken libunwind builds
- Libunwind as packaged in MacOS cannot actually unwind code which has
pointer authentication on, because _LIBUNWIND_IS_NATIVE_ONLY is not
defined. Add a check for this, and prefer building with working
unwinding over pointer authenticatio... -
- This reverts commit 6a746e1bc902d4245aac58db4e9ffc2f72d79629.
This patch breaks the C level backtrace on macOS ARM machines. -
07:13 PM Feature #20715: `Ractor.[]` and `Ractor.[]=` to access Ractor local storage
- > it would be nice to do the same for Fiber and Thread as well
Would also be the occasion for `Thread[]` to be actual thread local instead of fiber local. -
03:38 PM Feature #20715: `Ractor.[]` and `Ractor.[]=` to access Ractor local storage
- +1, makes perfect sense (IMO it would be nice to do the same for Fiber and Thread as well but out of scope of this issue).
-
06:52 PM Bug #20501: ruby SEGV
- Okay, I figured out what's happening. In compile.c, new LABELs are allocated from an arena, and this is using xmalloc, so it's not zeroed. Labels have a `position` field that is not set in the `new_label_body()` function, so it could be ...
-
- This is a weird way to reproduce, but you can see it on https://runruby.dev/ if you comment out the Gemfile and put this in main.rb:
```ruby
def my_method
var = nil
if false && defined? File::TMPFILE
end
raise "woops"
en... -
04:45 PM Bug #20501: ruby SEGV
- I'm not able to reproduce on Linux with this script using either master HEAD or 5613d6e95b.
-
- I'm on x86-64 linux so that might have to do with it. I'll investigate a bit more.
-
02:20 AM Bug #20501: ruby SEGV
- luke-gru (Luke Gruber) wrote in #note-5:
> For example, this is a minimal reproduction:
I can't reproduce it with master.
```ruby
def my_method
ivar = nil # seems to be needed to reproduce the issue
if false && defined? Fil... -
06:44 PM Revision b0adae7f (git): [rubygems/rubygems] Remove unnecessary "./" when appending string to uri
- https://github.com/rubygems/rubygems/commit/732679306d
-
- Every time a gem is not found in the Compact Index API, RubyGems will
fallback to the full index, which is very slow. This is unnecessary
because both indexes should be providing the same gems, so if a gem
can't be found in the Compact I... -
- https://github.com/rubygems/rubygems/commit/282dbb3d62
-
- This exception has not been raised for a long time.
https://github.com/rubygems/rubygems/commit/a6271a0b21 -
- Nothing is actually raising this at the moment.
https://github.com/rubygems/rubygems/commit/3b824ca7a6 - 04:58 PM Revision ad742de7 (git): [ruby/openssl] Fix test_provider.rb in FIPS.
- https://github.com/ruby/openssl/commit/7bdbc52100
-
02:48 PM Feature #20309: Bundled gems for Ruby 3.5
- Does this change mean that even if Ruby is installed, the irb command will no longer be available?
If that is the case, I oppose it for the following reasons:
- Existing learning materials will no longer work as-is, causing confusion... -
05:35 AM Feature #20309: Bundled gems for Ruby 3.5
- I will warn `benchmark`, `irb` and `reline` at Ruby 3.4. and make them and `readline` wrapper to bundled gems at Ruby 3.5 with September Dev Meeting. The current list is final proposal for Ruby 3.4.
I know `bundled_gems.rb` have so... -
- https://github.com/rubygems/rubygems/commit/2706acb271
-
- https://github.com/rubygems/rubygems/commit/b82e43fd54
-
- https://github.com/rubygems/rubygems/pull/7806#issuecomment-2241662488
https://github.com/rubygems/rubygems/commit/4f06ee234a -
-
-
- Applied in changeset commit:git|bc85c8d8529b58c5c649f418ca549569ba348caa.
----------
Implement Array#fetch_values
[Feature #20702]
Works the same way than `Hash#fetch_values` for for array. -
- [Feature #20702]
Works the same way than `Hash#fetch_values` for for array. -
08:06 AM Feature #20705: Should "0.E-9" be a valid float value?
- > But I think it's good to make Float/to_f to accept "0.e-9"
Changing `String#to_f` introduces incompatibility:
```
$ ruby -ve "p '1.e-9'.to_f"
ruby 3.3.4 (2024-07-09 revision be1089c8ec) [x86_64-linux]
1.0
``` -
02:48 AM Feature #20705: Should "0.E-9" be a valid float value?
- Implementation: https://github.com/ruby/ruby/pull/11559
Should we also accept "0." as @mame showed in #2? The implementation includes "0." support.
FYI: to_f already accepts "0.". Float didn't accept "0.". -
- In short, get rid of division and modulo of negative integers.
-
- https://github.com/ruby/digest/commit/76878e3201
-
-
-
-
- When we make irb to bundled gems, binding.irb with bundle exec is not
working without irb declaration of Gemfile.
This force_require makes to load irb without Gemfile. -
-
04:22 AM Misc #20717 (Closed): DevMeeting-2024-10-03
- # The next dev meeting
**Date: 2024/10/03 13:00-17:00** (JST)
Log: https://github.com/ruby/dev-meeting-log/blob/master/2024/DevMeeting-2024-10-03.md
- Dev meeting *IS NOT* a decision-making place. All decisions should be done at t... -
-
- As for the current SOCKS support at least, if the necessary library and header are found `socket.so` uses SOCKS functions instead of standard socket functions, and the built `socket.so` cannot be loaded where SOCKS library is not install...
-
03:42 AM Bug #20716: Different instance_method behavior in Ruby 2.7 and Ruby 3.x
- Yes, you are correct. This isn't specific to `define_method`. I modified the snippet a little bit
```
module A
def test(*args)
super
end
end
module B
def test(a)
puts a
end
end
B.prepend(A)
class C
i... -
03:32 AM Bug #20716: Different instance_method behavior in Ruby 2.7 and Ruby 3.x
- This appears to be a method caching bug in `super`. If you remove the `.call(1)`, then it works correctly. `super_method` returns the correct method. The `define_method(:test, a)` call should result in the method cache being cleared o...
-
- Hi, I am working on upgrading our project from 2.7 to 3.x and found a breaking change in `instance_method` behavior.
Here is the code snippet
```
module A
def test(*args)
super
end
end
module B
def test(a)
puts ... -
- I tried `make test-all TESTS=test/ruby/test_array.rb` with the proposed implementation, but several tests failed. They should be solved before discussing performance.
```
[ 47/447] TestArray#test_bsearch_in_find_any_mode = 0.00 s
... -
01:00 AM Revision 948ca04a (git): Allow failures for all Travis jobs
- Every job on Travis is unstable now.
-
- * Merge reline-0.5.8
* Merge reline-0.5.9
* Merge reline-0.5.10 -
- "Arm Neoverse (Ubuntu)" has been removed from https://rubyci.org/.
-
- https://github.com/ruby/ruby/pull/11558
- 12:32 AM Revision 81b74c9f (git): Update default gems list at e17d91459cc1f806b31feaff0cdbaa [ci skip]
-
- https://github.com/ruby/time/commit/a14fa00ef7
09/05/2024
-
- [Feature #20707]
Ruby 3.4 will define this method natively, so the time gem shouldn't
redefine it with a slower version.
https://github.com/ruby/time/commit/f05099ce38 -
08:34 PM Feature #20692: Rewrite Array#bsearch in Ruby
- I think this is a good idea, but I think we need to make a few changes. First, `to_enum` and `block_given?` are both method calls. It's possible for people to monkey patch Array and add those methods, but the original implementation of ...
-
-
- Implementation: https://github.com/ruby/ruby/pull/11555
-
08:46 AM Feature #20702: Add `Array#fetch_values`
- Looks like a good idea. Accepted.
Matz.
-
- The instance variable @tmpfile was removed in ddcfc9f so check_tempfile_leak
did not work. -
- I created a patch here: https://github.com/ruby/ruby/pull/11554. I'm new to the code in compile.c so perhaps someone could come up with a better solution.
-
- Applied in changeset commit:git|57e3fc32ea83e55d40f4cc6c3e437d485c506d34.
----------
Move Time#xmlschema in core and optimize it
[Feature #20707]
Converting Time into RFC3339 / ISO8601 representation is an significant
hotspot for appl... -
- Accepted.
Matz.
-
- [Feature #20707]
Converting Time into RFC3339 / ISO8601 representation is an significant
hotspot for applications that serialize data in JSON, XML or other formats.
By moving it into core we can optimize it much further than what `strf... - 04:53 PM Revision d4de8aef (git): `rake install` command is failed (#1170)
- * `rake install` command is failed
\### Problems
Several file paths were changed by following PR.
- https://github.com/ruby/rdoc/commit/4211292ffe80dd4737db2450d72df404a9d55051
- https://github.com/ruby/rdoc/commit/d7bca12c13b8b3f1632... -
02:36 PM Feature #20673: Enable native SOCKS support by default
- @nobu there has perhaps been a misunderstanding. The `socksify` library works adequately for HTTP today, but it patches a private method in the `Net::HTTP` class. This makes any code built using `socksify` very fragile and potentially su...
-
- At the developers' meeting, there was an opinion that we could drop the support for SOCKS if it works well with `socksify`.
- 12:49 PM Revision ffebc1c7 (git): Update default gems list at 721a2ceecfe3e3f943b7da50efe8f1 [ci skip]
-
- (https://github.com/ruby/reline/pull/745)
https://github.com/ruby/reline/commit/0ebd54f675 -
- frozen in the future`
(https://github.com/ruby/reline/pull/744)
https://github.com/ruby/reline/commit/69c95c8b6a -
10:57 AM Bug #20713: Ruby 3.3.5 triggers a deprecation warning with `require "json"`
- The `rdoc` warning is from `irb` itself: https://github.com/ruby/irb/pull/999
-
10:46 AM Feature #20684 (Closed): Add optimized instructions for frozen literal Hash and Array
- Applied in changeset commit:git|a99707cd9c6a1d53cf8ebc883dc210219bd67a28.
----------
Optimized instruction for Array#freeze
If an Array which is empty or only using literals is frozen, we detect
this as a peephole optimization and chan... -
- Makes sense, thank you Matz!
-
- Basically agreed. But the existence (and side effect) of this optimization should be implementation defined, especially regarding `object_id`.
Matz.
-
- If a Hash which is empty or only using literals is frozen, we detect
this as a peephole optimization and change the instructions to be
`opt_hash_freeze`.
[Feature #20684]
Co-authored-by: Jean Boussier <byroot@ruby-lang.org> -
- If an Array which is empty or only using literals is frozen, we detect
this as a peephole optimization and change the instructions to be
`opt_ary_freeze`.
[Feature #20684]
Co-authored-by: Jean Boussier <byroot@ruby-lang.org> -
- > Warning category option like
I think this is the better option as it build on top of an existing interface.
And probably some existing warnings could be moved into that category. The way I see it, a `pedantic` category would be ... -
03:02 AM Feature #15554: warn/error passing a block to a method which never use a block
- As @Eregon mentioned on #60 we need to discuss how to enable strict mode.
https://hackmd.io/CoLraFp_QrqyHBcv3g8bVg?view#Feature-15554-warnerror-passing-a-block-to-a-method-which-never-use-a-block-ko1
Summary of proposed ways:
* Environ... -
- Applied in changeset commit:git|63cbe3f6ac9feb44a2e43b1f853e2ca7e049316c.
----------
Proof of Concept: Allow to prevent fork from happening in known fork unsafe API
[Feature #20590]
For better of for worse, fork(2) remain the primary ... -
- [Feature #20590]
For better of for worse, fork(2) remain the primary provider of
parallelism in Ruby programs. Even though it's frowned uppon in
many circles, and a lot of literature will simply state that only
async-signal safe APIs ar... -
09:23 AM Bug #20690 (Rejected): URI.encode_www_form_component method escapes tilde when it's not supposed to
- `URI.encode_www_form_component` is designed for HTML form submission.
It is not whatt RFC3986 defines as "Percent-Encoding", but defined as "application/x-www-form-urlencoded".
(The latest definition is https://url.spec.whatwg.org/#url... -
- It seems a bug in missing/dtoa.c.
-
- We are not going to change the literal format of floating point values. But I think it's good to make Float/to_f to accept "0.e-9".
Matz.
-
- Since `StringIO` is a IO-like object, not String-like object, so it should not provide `to_str`. And string body of a StringIO tends to be long, so simple alias of `string` to `to_s` does not always work well.
So please use `string` met... -
- At this developers' meeting, no one was favor of this.
IMO, `StringIO` is not a `String`-like object.
- 08:41 AM Revision 2e5680d3 (git): [ruby/openssl] Fix test_pkey_rsa.rb in FIPS.
- * test_sign_verify
I created the signature text (`signature_encoded.txt`), that is used as a
text to create the `signature0` in the `test_sign_verify` by the following
steps with the `openssl` CLI on FIPS module.
```
$ OPENSSL_... -
- - Extract functions to check not-found conditions
- Set the length to the result of `rb_getlogin`
- Reentrant versions return an error numeber but not `errno`
- Check maybe-undefined macros with `defined` - 07:01 AM Revision 630bfd36 (git): Update bundled gems list as of 2024-09-04
-
-
- deivid (David Rodríguez) wrote in #note-6:
> Is it as simple as https://github.com/ruby/ruby/pull/11550?
I had something a little different in mind (trim `SINCE`) but I think that is a nice solution. The PR looks reasonable to me. -
- OK, I understand the reason and background of this behavior. The issue is withdrawn.
Matz.
-
- `String#append_as_bytes` looks good to me too. Accepted.
Matz.
-
04:03 AM Revision 32680f54 (git): Implement AND/OR NODE operator locations
-
- I want to add new features:
* `Ractor._activated` which is called when the first `Ractor.new` is called
* `Ractor._require(feature)` described in this issue
* `_activated` method uses `prepend` with new anonymous module to call `Rac... -
- Trivial proposal.
Now `Ractor#[]/#[]=` is supported to access current ractor local storage. However, it doesn't allow to access local storage of other ractors.
```ruby
Ractor.current[:foo] = 1
Ractor.new{
p Ractor.main[:foo...
09/04/2024
-
- It looks like it has to do with `defined?()` in an `if` expression and its catch table entries when the first part of the if expression has been eliminated and there is no then label to follow.
For example, this is a minimal reproductio... -
- esad (Esad Hajdarevic) wrote in #note-21:
> Is there a reason why this doesn't work with keyword arguments too?
> ...
Keyword arguments are not leading arguments. Trying to support what you want was not in scope. I think trying to s... -
09:04 PM Feature #16378: Support leading arguments together with ...
- Is there a reason why this doesn't work with keyword arguments too?
For example:
```
def foo(bar:, ...)
other(...)
end
```
results in syntax error
-
- > Drop warnings when ruby would throw an error here. When the require will raise, it doesn't need to warn. This is how I already imagined it works. I guess it doesn't for visibility?
I think this is the best solution. The warning is a... -
- Maybe a good solution is to make `$VERBOSE` thread-local or fiber-local at some point, then at least this concern and related ones would be solved once and for all.
-
- Thanks for you reply, I was about to write something similar about the older rubies.
At first I was writing about how I don't think a new api is really necessary since the compatibility code can simply be dropped once `required_ruby_v... -
- The problem is any new API would be rather cumbersome to use (in a way that still works on older Rubies), so probably this fix is good as a quick measure.
I think it would be good to think about a longer-term solution too though. -
- One potential problem with setting `$VERBOSE` to nil is that's not thread-safe, and so might hide valid warnings from other threads requiring around the same time.
Maybe it should be some explicit method in `Gem` or so to suppress that ... -
- I've encountered a few places around bundled gems where the library doesn't care if the gem is available, but will still provide some functionallity if it is.
The way to accomplish that right now seems to be by setting `$VERBOSE = nil... -
- Merge JSON 2.7.2
-
- Followup: a3f589640fd443deea230c27efd6bdfc92f5817f
Restarting scanning has a fixed cost. Since we build a string
from many small parts, that causes the scan to restart many times.
It's much faster to ignore the coderange, if it's neede... - 01:57 PM Revision 2e5a7d76 (git): [rubygems/rubygems] Improve `bundle add` manpage
- * Fix synopsis for mutually exclusive options.
* Add more descriptive examples.
Signed-off-by: Takuya Noguchi <takninnovationresearch@gmail.com>
https://github.com/rubygems/rubygems/commit/c0f1efda69
Co-authored-by: Jerome Dalbert <je... -
-
01:13 PM Revision da0460a2 (git): [ruby/prism] Update config.yml
- https://github.com/ruby/prism/commit/8a3fa9f46e
-
- https://github.com/ruby/prism/commit/369b2c1304
-
- https://github.com/ruby/prism/commit/9c68c01bcd
-
- https://github.com/ruby/prism/commit/fa65b66237
-
- Use a classic doubling of capacity rather than only adding
twice as much capacity as is already known to be needed.
```
compare-ruby: ruby 3.4.0dev (2024-09-04T09:21:53Z opt-strftime-2 ae98d19cf9) +YJIT [arm64-darwin23]
built-ruby: ruby... -
- I believe reline requires a PR like that as well: `bundle exec ruby -e "require 'reline'"`
> /home/user/.rbenv/versions/3.3.5/lib/ruby/3.3.0/reline.rb:9: warning: fiddle was loaded from the standard library, but will no longer be part o... -
- https://github.com/ruby/ruby/pull/11541
-
06:43 AM Bug #20713: Ruby 3.3.5 triggers a deprecation warning with `require "json"`
- I think that if Bundler is used, that means gems are used and they should always fully specify their dependencies. The other way is to not use Bundler.
-
05:06 AM Bug #20713: Ruby 3.3.5 triggers a deprecation warning with `require "json"`
- Ah is the default `json` gem no longer supported too? This broke REXML CI: https://github.com/ruby/rexml/actions/runs/10695343850/job/29648614890 because `rdoc` uses `json` from the standard library and I see the same warning from runnin...
-
- You can suppress that warning with `json-2.7.2`.
-
- Can be reproduced with `bundle exec ruby -e "require 'json'"`:
```
/opt/hostedtoolcache/Ruby/3.3.5/x64/lib/ruby/3.3.0/json/common.rb:3: warning: ostruct was loaded from the standard library, but will no longer be part of the default ... -
- https://github.com/rubygems/rubygems/commit/0a6c1c53ce
-
11:58 AM Revision 9a766777 (git): [ruby/reline] Split render_differential to LineEditor's rendering
- logic and actual differential rendering part
(https://github.com/ruby/reline/pull/731)
https://github.com/ruby/reline/commit/c90f08f7e3 -
10:03 AM Misc #20013: Travis CI status
- > So, I will drop all the pipelines on the 2nd commit on the following PR.
Dropped Travis as temporary workaround. The infra issue is still ongoing. The Travis customer support replied with the following message.
> ...
-
-
-
- Fixed https://github.com/rubygems/rubygems/pull/7996
https://github.com/rubygems/rubygems/commit/16bfcac883 -
- * [Feature #20707] Move `Time#xmlschema` (AKA `iso8601`) into core (byroot)
* Usability wise, this is such a common format that I'd expect to have it without needing to require anything.
* On a more general point, `require 'time'` ... - 07:02 AM Revision 7fc1fd89 (git): Update bundled gems list as of 2024-09-04
-
- Although `pend` in test-unit gem expects the block does not pass if it
is given, our `pend` should ignore the block silently as same as it
just skips the test with no message by default. Add an unused
parameter, to suppress the warning ... -
-
-
- @peterzhu2118 This is totally different from the issue we talked about in Slack, which was entirely a micro-benchmark of the speed of generating empty hashes. This problem is much more complex.
This problem is that “reducing the numbe... -
- Applied in changeset commit:git|37d7ae06afb03ae5508bfd81033961559886bd6b.
----------
[Bug #20708] Retry `open` on EINTR
Co-Authored-By: Martin Dorey <martin.dorey@hds.com> -
- https://github.com/ruby/ruby/pull/11537
-
- Co-Authored-By: Martin Dorey <martin.dorey@hds.com>
-
-
01:21 AM Revision 08f14b8d (git): Allow Errno::EACCES when testing connection timeout
- Some packaging systems that include support for running tests,
such as OpenBSD's, do not allow outbound network connections
during testing for security reasons. EACCES is the error raised by
OpenBSD in this case.
09/03/2024
-
-
09:47 PM Bug #20711: readline/reline duplicate display of prompt
- tompng (tomoya ishida) wrote in #note-1:
> https://github.com/ruby/reline/pull/742 will fix it.
Great - thanks :-) -
- https://github.com/ruby/reline/pull/742 will fix it.
-
- ```ruby
#!/usr/bin/env ruby
require 'readline'
prompt = "asdf: "
default = "true"
Readline.pre_input_hook = -> do
Readline.insert_text "#{default}"
Readline.redisplay
Readline.pre_input_hook = nil
end
Readline.readli... -
06:28 PM Feature #20709 (Closed): Improve String#rindex performance on OSX
- Merged in https://github.com/ruby/ruby/commit/e7cb70be4eb7411204f73ee748e317fefaa0410a
-
-
- The object could be garbage (e.g. T_NONE, T_MOVED, or T_ZOMBIE) so we
don't want to traverse those. -
- 05:40 PM Revision 675529b9 (git): [rubygems/rubygems] standardize pretty-print output for Gem::Source and subclasses
- https://github.com/rubygems/rubygems/commit/6d5fbf82f1
-
-
-
- run without RDoc
(https://github.com/ruby/irb/pull/998)
* EasterEgg no longer depend on RDoc
* Run most of the input-method tests even if RDoc is not avialable
https://github.com/ruby/irb/commit/30fa1595d9 -
03:37 PM Feature #20525: Percent string literal with indentation support or String#dedent
- Oh interesting, I thought that was a typo 😂.
I was playing around with characters after I wrote this and found something that I think looks a little better than the "}" thing:
```ruby
markdown <<~___
# Hello
How are you do... -
12:28 PM Feature #20525: Percent string literal with indentation support or String#dedent
- @bradgessler I am not sure if you've missed the suggestion by @shyouhei in https://bugs.ruby-lang.org/issues/20525#note-5:
```ruby
markdown <<~"}"
# Hello
How are you doing?
}
```
This doesn't need naming anything and has no... -
- Same reason you can conjure up a Proc via `-> {}` — the syntax looks cleaner and you don't have to stop and try to name it.
Not having to name a string is a pretty big boost in terms staying in the flow.
The second thing: it looks ... -
- (https://github.com/ruby/reline/pull/741)
* Fix reset variables
* Add assertion for reload
* Add helper method to get instance variable value of Reline::Config
https://github.com/ruby/reline/commit/386f619ff5 -
- (https://github.com/ruby/reline/pull/742)
* Fix redisplay/insert_text called from pre_input_hook
* Rename insert_pasted_text to insert_multiline_text
It is now used from Reline.insert_text which is not inserting pasted text
https://g... -
- This commit removes the need for the GC implementation to implement heap
walking and instead Ruby will implement it. -
-
-
- `module` and `class` don't return `nil`, they return the value of the last statement in the scope:
```ruby
x = module Foo
42
end
puts x #=> 42
x = class Bar
def foo = 42
end
puts x #=> :foo
```
There might be code that r... -
- Ok I was too quick on this one. module keyword returns last expression in the block, so just ending module declaration with self works:
```
using(module Foo
refine ...
self
end)
```
so this request can be closed -
- Currently module keyword returns nil after defining a new module:
```
x = module Foo
end
# x is nil
```
It would be more consistent with `Module.new` if module keyword returned the newly defined module. Then we could do:
```... -
-
- Profiling of `JSON.dump` shows a significant amount of time spent
in `rb_enc_str_asciionly_p`, in large part because it fetches the
encoding.
It can be made twice as fast in this scenario by first checking the
coderange and only falling... -
07:21 AM Revision 245ed2fc (git): Range#step: restore legacy behavior for String ranges
-
-
- This reverts commit 6986536488ba23c033c043ef8765438f22df7bf5.
We were just missing the secrets. -
-
-
-
- Removed them at https://github.com/ruby/www.ruby-lang.org/pull/3347
-
- This issue happens only when ruby is installled but erb is not installed. 6ab591f80aa19d63ecd1e1df3c09c391efb318a6 checks if not only ruby but also erb is installed. So I think the issue itself is already resolved. Using miniruby for reb...
-
- vo.x (Vit Ondruch) wrote in #note-4:
> nobu (Nobuyoshi Nakada) wrote in #note-1:
> ...
Now miniruby is used if erb is not available. -
05:25 AM Revision e7cb70be (git): Improve String#rindex performance on OSX
- On OSX, String#rindex is slow due to the lack of `memrchr`.
The fallback implementation finds a match by instead doing
a `memcmp` on every single character in the search string
looking for a substring match.
For OSX hosts, this changese... -
- https://github.com/ruby/optparse/commit/bb08cd47a8
-
03:24 AM Misc #20660: DevMeeting-2024-09-05
- * [Feature #20673] Enable native SOCKS support by default (znz)
* I think enable by default is better than depend on an environment variable. -
- https://github.com/flori/json/commit/148afef84c
-
- `json` requires Ruby 2.3, so `HAVE_RUBY_ENCODING_H` and `HAVE_RB_ENC_RAISE`
are always true.
https://github.com/flori/json/commit/5c8dc6b70a -
- It is loaded by `Fetcher` so in most case it's fine.
But if using `bundler/inline` and a gem need to be fetched,
`securerandom` will be loaded and cause a conflict.
Can be reproduced with:
```ruby
require 'bundler/inline'
gemfile do
... -
- Co-authored-by: Jean Boussier <jean.boussier@gmail.com>
-
- The default version of securerandom (0.2.2) gets activated by RubyGems,
but does not match Rails requirements (>= 0.3), leading to an error like
this:
```
$ gem exec rails new repro
/Users/deivid/Code/rubygems/rubygems/lib/rubygems/spec... -
-
- Fix symlinks to executable as relative paths from `bin` directory.
09/02/2024
-
11:40 PM Bug #20695 (Closed): Elevated GC allocations in parse.y parser
- Applied in changeset commit:git|c93d07ed7448f332379cf21b4b7b649b057e5671.
----------
[Bug #20695] Do not create needless string object in parser
`set_parser_s_value` does nothing in parser therefore no need to
create string object in p... -
- `set_parser_s_value` does nothing in parser therefore no need to
create string object in parser `set_yylval_node`.
# Object allocation
Run `ruby benchmarks/lobsters/benchmark.rb` with the patch
```diff
diff --git a/benchmarks/lobsters... -
- It's possible for a GC to run between the calls of GC.latest_gc_info,
which would cause the test to fail. We can disable GC so that GC only
triggers manually. -
- Since `Gem.open_file` no longer locks the target file and is same as
`File.open` now, simply `Gem.read_binary` should read in binary mode.
Also the body of `Gem.write_binary` is same as `File.binwrite`.
https://github.com/rubygems/rubyg... -
- I still think free pages should be in a global pool rather than tied to a specific pool size. I believe that would solve this issue.
And yes we don't see it on macro benchmarks, but it might still cause more frequent GC than necessary. -
- Thank you for the detailed report.
I had a discussion about this with @mame and @byroot last year in the Ruby core slack: https://ruby.slack.com/archives/C02A3SL0S/p1702910003614609
One of the big differences is that VWA eagerly al... -
06:24 AM Bug #20710 (Closed): Reducing Hash allocation introduces large performance degradation (probably related to VWA)
- I found a surprising performance degradation while developing RBS.
In short, I tried to remove unnecessary Hash allocations for RBS. Then, it made the execution time 2x slower.
VWA for Hash probably causes this degradation. I'd be ha... -
-
- Instead, just open in CREATE and APPEND mode.
Also, move the workaround for old Solaris as fallback to retry.
https://github.com/rubygems/rubygems/commit/2daad257be -
03:36 PM Revision b539b43b (git): [ruby/rdoc] [DOC] Fix links (https://github.com/ruby/rdoc/pull/1169)
- https://github.com/ruby/rdoc/commit/a576ff890f
-
- fails
(https://github.com/ruby/reline/pull/734)
https://github.com/ruby/reline/commit/66291b5e3c -
- Ruby ships with empty directories for default gems. If Ruby
installations has unsafe world-writable permissions, we will complain
when about to install a gem that happens to be also a default gem,
because we'll start by removing the prev... -
- https://github.com/rubygems/rubygems/commit/7c71dac22e
- 01:36 PM Revision 4303a02f (git): CI: windows.yml: Drop pull-request - path-ignore.
- It's to avoid the stucking "Windows result" job.
It seems only the files included in the list of the pull-request - path-ignore,
GitHub Actions "Windows result" permanently waits for the jobs that don't exist. - 01:36 PM Revision c853e48f (git): .travis.yml: Drop all the pipelines due to jobs not starting.
- Remove `.travis.yml` as a temporary workaround.
- 01:36 PM Revision 3d83df2b (git): .travis.yml: Allow failures for arm64/arm32 pipelines.
- Right now, the arm64 servers on Travis took around 7 hours to start the jobs.
Therefore, allow failures for the arm64 and arm32 pipelines
https://app.travis-ci.com/github/ruby/ruby/builds/272105049 -
- I found it useful to be able to use `GEM_COMMAND` outside of a ruby-core
env, so I made the ruby-core env detection not depend on it.
https://github.com/rubygems/rubygems/commit/6cb389c290 -
- Update bundled gems list as of 2024-08-02
-
- It seems necessary to pass test-bundled-gems with REXML.
-
- ruby_3_3 commit:cf9a6c2b63e6337a3f6ce76527446739e5aceb67 merged revision(s) commit:a3562c2a0abf1c2bdd1d50377b4f929580782594.
-
- Remove incorrect setting of KW_SPLAT_MUT flag
Fixes [Bug #20701]
Co-authored-by: Pablo Herrero <pablodherrero@gmail.com> -
- Please consider filing PRs to branches like `ruby_3_3` next time.
-
- ruby_3_3 commit:e0e23e7d5e.
-
- ruby_3_3 commit:ac8d50e52ebc2d2684914e56548a64a65830c16a.
-
- ruby_3_3 commit:d83b5633b16f4ddcece4ff924f21c5a5851470cf merged revision(s) commit:992596fb7af18a7f472589a607d0eb3fbb03b49a.
-
- Fix next inside block argument stack underflow
[Bug #20344]
Fix compile_next adding removable adjust label -
- ruby_3_3 commit:b210c86a0201f6a97c4da8266908260746f53ae0 merged revision(s) commit:97449338d6cb42d9dd7c9ca61550616e7e6b6ef6.
-
- [Bug #20649] Allow `nil` as 2nd argument of `assign_error`
Fallback to the last token element in that case, for the backward
compatibilities. -
- ruby_3_3 commit:6a4e79533b4454392f3afdfa0c4d35b763b68466.
-
- ruby_3_3 commit:6a4e79533b4454392f3afdfa0c4d35b763b68466 merged revision(s) commit:1870505f478cc75993b296b7144a45137ace6937.
-
- Fix wrong unreachable chunk remove when jump destination label is unremovable
-
- Please consider filing a backport PR to stable branches next time.
-
- ruby_3_3 commit:e0e23e7d5eb4da42c490b1d3408bd6e5047e8f83 merged revision(s) commit:29500e30346.
-
- Update bundled gems list as of 2024-08-22
-
- Those jobs are way too unstable and provide very little benefit.
e.g.
https://app.travis-ci.com/github/ruby/ruby/jobs/624611955
https://app.travis-ci.com/github/ruby/ruby/jobs/625551620 -
- ruby_3_3 commit:65cca99b30c65d462aa67969e0f16f6e45db4705.
-
- ruby_3_3 commit:ce565cd4b851977bf37a470bee54e441bb60486d.
-
- ruby_3_3 commit:4667f8ec10269b0b5deca459f098abbdf3bae4ec.
-
- ruby_3_3 commit:66312ad913d67bfd3c2c83b174eabf537f5def84.
-
- The message was auto-generated by a script. The script only matches the last commit mentioning the ticket number, and there were two matching commits. Sorry for a delayed reply but I got distracted to fix it first (commit:7d47f3c94f4c38f...
-
08:55 AM Bug #20633: compile error at vm_insnhelper.c when HAVE_DECL_ATOMIC_SIGNAL_FENCE is 0
- Hey @k0kubun I don't think 66312ad913d67bfd3c2c83b174eabf537f5def84 is related to this one? 👀
(Although I think this could be closed, as the fix + both backports have landed) -
- ruby_3_3 ~~commit:66312ad913d67bfd3c2c83b174eabf537f5def84~~, commit:425e468d25a70740cef3ed676e9b82f7902e077a.
-
- with `done` command
-
- The `rev && has_commit(rev, "ruby_#{TARGET_VERSION.tr('.','_')}")`
case seems to be used if and only if it's manually specified in the
argument to the done command. It was hard to notice it with the previous
code. - 03:41 AM Revision 73c39a5f (git): Bump actions/upload-artifact from 4.3.6 to 4.4.0
- Bumps [actions/upload-artifact](https://github.com/actions/upload-artifact) from 4.3.6 to 4.4.0.
- [Release notes](https://github.com/actions/upload-artifact/releases)
- [Commits](https://github.com/actions/upload-artifact/compare/834a14... - 02:30 AM Revision 0642f71a (git): Update default gems list at 745b4775c5d8b28f6e5fa26b928239 [ci skip]
-
- https://github.com/ruby/logger/commit/bda937b4cc
09/01/2024
-
-
-
- https://github.com/ruby/io-console/commit/504292b487
-
- https://github.com/ruby/io-console/commit/cad8169568
-
-
- * [Bug #20693] Dir.tmpdir should perform a real access check before warning about writability
* A minor paper-cut, but I ran into this while running Ruby's tests in an unprivileged Docker container
* `File::Stat#writable?` is reall... - 02:47 AM Revision 44771e4a (git): Bump github/codeql-action from 3.26.5 to 3.26.6
- Bumps [github/codeql-action](https://github.com/github/codeql-action) from 3.26.5 to 3.26.6.
- [Release notes](https://github.com/github/codeql-action/releases)
- [Changelog](https://github.com/github/codeql-action/blob/main/CHANGELOG.md...
08/31/2024
- 06:30 PM Revision d6fc8f3d (git): [rubygems/rubygems] fix `@license` typo preventing licenses from being correctly unmarshalled
- https://github.com/rubygems/rubygems/commit/d6ba7ef79f
-
- `String#rindex` is much slower on OSX than on Linux hosts. This appears due to the lack of a `memrchr` implementation on OSX. The [fallback solution](https://github.com/ruby/ruby/blob/e69945fc57efba5c8895c21235e109145865952d/string.c#L43...
-
10:10 AM Revision b6f7d544 (git): merge revision(s) 29500e3034681a30045dea462d6bb653e8600738: [Backport #20698]
- Update bundled gems list as of 2024-08-22
-
- Added explicitly begin-end block for Ruby 2.4.
strscan, ipaddr and some default gems still support Ruby 2.4.
After this, I extract this CoreAssertions to their repositories. -
- Improve base time of assert_linear_performance (#11369)
Remove `.ceil` from base time calculation that makes 10x gap.
This will make the assertion more strict and also less flaky. -
- Suppress warning for shadowing outer local variable
-
- core_assertions.rb: Support old rubies
Some symbol argument might not be accepted by Process.clock_gettime. -
- core_assertions.rb: Prefer CPU time clocks
To prevent influence from other processes. -
- core_assertions.rb: Extract common code block
-
- Skip assert_linear_performance for RJIT
-
- core_assertions.rb: Tweak timeout limit
Increase the timeout limit when variance at rehearsal is small. -
- https://github.com/ruby/io-console/commit/f10c946ac7
-
- Ractor requires a shareable class has shareable constants only, but IO
is not shareable unless frozen.
https://github.com/ruby/io-console/commit/65e0ff895c -
- All the Travis partner pipelines (non-x86_64) jobs don't start. So, I will drop all the pipelines on the 2nd commit on the following PR.
https://app.travis-ci.com/github/ruby/ruby/builds/272123937
https://github.com/ruby/ruby/pull/11509 - 05:46 AM Revision 5dbb67a5 (git): Update default gems list at 365ededfdf4fa33000411e63d7a1c2 [ci skip]
-
- https://github.com/ruby/io-console/commit/b1adc15af7
- 05:22 AM Revision 2afde63e (git): * remove trailing spaces. [ci skip]
-
- https://github.com/ruby/stringio/issues/81
-
- 05:15 AM Revision a4ede0c4 (git): Update default gems list at 7900128ab2d8c1d3c52e9c3e14df3c [ci skip]
-
- https://github.com/ruby/uri/commit/31ec9cea66
- 05:05 AM Revision 53d8dca3 (git): Update default gems list at 818e3037ed8635ae2ed63a5d783cb3 [ci skip]
-
- https://github.com/ruby/zlib/commit/d756bb0a0f
-
- (https://github.com/ruby/reline/pull/740)
https://github.com/ruby/reline/commit/59e4ade807 -
-
-