Activity
From 06/01/2022 to 06/07/2022
06/07/2022
-
08:19 PM Bug #18170: Exception#inspect should not include newlines
- Sorry that the `#inspect` call actually came from the `ruby/debug`'s [own patch](https://github.com/ruby/debug/blob/master/test/support/utils.rb#L78). But I still found an `exception#inspect` occurrence in `test-unit` [here](https://gith...
-
09:58 AM Bug #18170: Exception#inspect should not include newlines
- Can you show where a change to `#inspect` will impact this code?
-
- I think this change will affect many users' tests because test frameworks generally use exceptions for assertion messages.
For example, this is how [test-unit](https://github.com/test-unit/test-unit) implements [assert_block](https://... -
06:15 PM Revision bf4684d9 (git): Remove duplicated prototype in header file
- rb_imemo_new is defined again later in the file.
-
06:07 PM Feature #18819 (Closed): Moving Strings between size pools
- [Github PR](https://github.com/ruby/ruby/pull/5986)
## Motivation
Using GC Compaction we can move objects around on a Ruby heap, this allows us to decrease memory fragmentation for more efficient memory usage and better copy on wri... -
04:15 PM Revision 19c6aaca (git): thread_pthread.c: trigger THREAD_EVENT_READY when going throuhg the fast path.
- 03:21 PM Revision bdc6e991 (git): * 2022-06-08 [ci skip]
- 03:20 PM Revision 1598c945 (git): Add special-case code for the String unary plus operator (#5982)
-
02:23 PM Bug #18816: Ractor segfaulting MacOS 12.4 (aarch64 / M1 processor)
- I've tried to isolate the problem with a few attempts:
First guess was that something GC related (cleaning up something that was still in use etc).
I've tried to disable GC with code below:
```
require 'socket'
# Set based o... -
02:17 PM Bug #18816: Ractor segfaulting MacOS 12.4 (aarch64 / M1 processor)
- I get the same segfault, but not as consistent as Gabriel. I had to first do a normal request in the browser before running "ab" the couple times it printed the segfault, but I haven't been able to reproduce as consistently. The localhos...
-
02:18 PM Revision deff9e26 (git): [rubygems/rubygems] Remove unnecessary string concatenation
- https://github.com/rubygems/rubygems/commit/81ccb3ab89
-
- Having a while loop over `heap_prepare` makes the GC logic difficult to
understand (it is difficult to understand when and why `heap_prepare`
yields a free page). It is also a source of bugs and can cause an infinite
loop if `heap_page` ... -
01:47 PM Bug #18061: Execshield test: libruby.so.N.N.N: FAIL: property-note test because no .note.gnu.property section found
- Here is a minimal reproducer to see the property-note test failure. The `./configure --enable-shared --with-gcc="gcc -fcf-protection -Wl,-z,now"` above was wrong. The `-Wl,-z,now` should not be in `--with-gcc="..". I checked the reproduc...
-
12:04 PM Revision f50432fb (git): [rubygems/rubygems] Relax performance spec limit
- https://github.com/rubygems/rubygems/commit/eab417d0ce
-
- `test_thread_instrumentation_fork_safe` has been failing occasionaly
on Ubuntu and Arch. At this stage we're not sure why, all we know is
that the child exit with status 1.
I suspect that something entirely unrelated cause the forked ch... -
08:40 AM Revision f075be3d (git): [ruby/error_highlight] Use Exception#detailed_message instead of overriding #message (https://github.com/ruby/error_highlight/pull/24)
- See https://bugs.ruby-lang.org/issues/18564.
Ref: https://github.com/ruby/did_you_mean/pull/177
https://github.com/ruby/error_highlight/commit/671b7c61b2 -
07:18 AM Bug #18811: PTY I/O not working on AIX 7.x
- mame (Yusuke Endoh) wrote in #note-2:
>
> ...
I found a way to fix this. The problem seems to be a known strangeness in the AIX pty implementation. It's sensitive to in which order things are done relative closing the descriptors. It'... -
06:24 AM Revision 11b9dd8c (git): Manually merged https://github.com/ruby/did_you_mean/pull/177
-
- > I sent the PR, https://github.com/ruby/ruby/pull/5938 .
The PR was merged.
> ...
The issue above is that the `./configure --enable-shared --enable-mkmf-verbose --with-gcc="gcc -O2 -fcf-protection" ldflags=-Wl,-z,now` was wrong. T... -
05:51 AM Bug #18781 (Closed): MJIT tests failing with Ubuntu focal with gcc-11 and some flags - Applied in changeset commit:git|cfcf33f1270f78c5fb928d7a0927e6be02ec8900.
----------
.github/workflows/compilers.yml: annocheck: Fix a linker flag to pass MJIT tests.
Set the linker flag `-Wl,-z,now` properly.
Co-authored-by: Takashi ... -
05:51 AM Bug #18294 (Closed): error when parsing regexp comment
- Applied in changeset commit:git|ec3542229b29ec93062e9d90e877ea29d3c19472.
----------
Ignore invalid escapes in regexp comments
Invalid escapes are handled at multiple levels. The first level
is in parse.y, so skip invalid unicode esca... -
- Applied in changeset commit:git|c85d1cda86d75ee2c3f7b42f22c543409cb5a186.
----------
Fix Module#const_source_location for autoload constants with direct requires
If an autoload exists for a constant, but the path for the autoload
was r... -
- Applied in changeset commit:git|7cda7fbbdc14f4262afaa94cdeb5a5987f1eb01a.
----------
Add Module#undefined_instance_methods
Implements [Feature #12655]
Co-authored-by: Nobuyoshi Nakada <nobu@ruby-lang.org> -
- This reverts commit 9d927204e7b86eb00bfd07a060a6383139edf741.
-
02:44 AM Revision 082c2d1b (git): [ruby/rdoc] [DOC] Undocument internal constants [ci skip]
- https://github.com/ruby/rdoc/commit/6d7bf24bb8
-
- ... only when the message string has a newline.
`p StandardError.new("foo\nbar")` now prints `#<StandardError: "foo\nbar">'
instead of:
#<StandardError:
bar>
[Bug #18170] -
- Currently only literal `0` and `1` are accepted as `read`/`write`
flags.
This patch allows other boolean arguments, C macros (`FALSE`/`TRUE`),
Ruby `VALUE`s (`Qfalse`/`Qtrue`), and C99 `bool`s (`false`/`true`), as
well.
https://github.c...
06/06/2022
- 10:25 PM Revision cfcf33f1 (git): .github/workflows/compilers.yml: annocheck: Fix a linker flag to pass MJIT tests.
- Set the linker flag `-Wl,-z,now` properly.
Co-authored-by: Takashi Kokubun <takashikkbn@gmail.com>
Fixes [Bug #18781] -
- Invalid escapes are handled at multiple levels. The first level
is in parse.y, so skip invalid unicode escape checks for regexps
in parse.y.
Make rb_reg_preprocess and unescape_nonascii accept the regexp
options. In unescape_nonascii,... -
06:31 PM Bug #18818: Thread waitq does not retain referenced objects, can lead to use after free.
- There might be a good reason we don't want to do this (or want to do it differently). But I can confirm that marking the fibers in the wait queues makes the SEGV go away! :D
Patch also at https://github.com/ruby/ruby/pull/5979 -
- Aha! Thanks, that makes perfect sense. And it does indeed fix it. I knew this toy scheduler wasn't *good*, and my original version did retain references to the waiting fibers, but I was slowly golfing it down to the smallest readable v...
-
- @ko1 I saw this problem because fiber is not retained while waiting, because we have waiting threads but not waiting fibers at VM level IIRC. Probably we need to make mutex/queue mark the wait list correctly? Is there performance issue?
-
- Unfortunately Mutex, Queue and probably other objects don't take a reference to the waiting fiber, so your fiber is being garbage collected after it's in the wait list. Then some other object is allocated and unblock is called with it (o...
-
06:17 PM Misc #18803: DevMeeting-2022-06-16
- * [Feature #18788] Support passing Regexp options as String to Regexp.new (jeremyevans0)
* Do we want to add support for `Regexp.new(code, options)`, where options is a string (e.g. 'im').
* @nobu pointed out it is already possible... -
- If an autoload exists for a constant, but the path for the autoload
was required, const_source_location would return [false, 0] instead
of the actual file and line. This fixes it by setting the appropriate
file and line in rb_const_set, ... - 05:47 PM Revision 653e517e (git): Use bindgen to import Ruby constants wherever possible. (#5943)
- Constants that can't be imported via bindgen should have
a comment saying why not. -
05:38 PM Feature #18183: make SecureRandom.choose public
- One option would be to extend `Random::Formatter#alphanumeric` to have an optional "alphabet":
```ruby
def alphanumeric(n=nil, alphabet: ALPHANUMERIC)
n = 16 if n.nil?
choose(alphabet, n)
end
``` -
04:45 PM Feature #18183: make SecureRandom.choose public
- My bad for not searching the old topics.
In this module, (most of) the method name describes the output, not the procedure, not how to generate the output.
For example:
Method `hex` outputs a hexadecimal number.
Method `uuid` out... -
05:38 AM Feature #18183: make SecureRandom.choose public
- I don't think `SecureRandom.random_string(args, len)` describes the behavior, neither.
Matz.
-
- @matz The OP proposed `random_string`. What do you think about it?
-
- My bad. args.sample(random: SecureRandom) is not a replacement for SecureRandom.choose(*args,len).
But still, choose does not describe the behavior.
Matz.
-
- `SecureRandom.choose(*args)` is not intuitive from my POV. Use `args.sample(random: SecureRandom)`.
Matz.
-
-
-
- Previously `(2..).cover?("2"..)` was false, but
`(..2).cover?(.."2")` was true. This changes it so both are false,
treating beginless ranges the same as endless ranges in regards to
type checks.
This also adds documentation to #cover? ... -
- Implements [Feature #12655]
Co-authored-by: Nobuyoshi Nakada <nobu@ruby-lang.org> -
03:37 PM Revision b737998d (git): [ruby/fileutils] [DOC] Enhanced RDoc for FileUtils (https://github.com/ruby/fileutils/pull/78)
- Treats:
::rm
::rm_f
::rm_r
::rm_rf
::remove_entry_secure
https://github.com/ruby/fileutils/commit/ce2a438d75 - 03:07 PM Revision f2f8c0b3 (git): * 2022-06-07 [ci skip]
- 03:06 PM Revision 6fc16e74 (git): [ruby/open-uri] [DOC] Fix markup for `URI.open`
- * Add missing slash.
https://github.com/ruby/open-uri/commit/40023e63da -
- https://github.com/ruby/rdoc/commit/fec1ab2e41
-
- It was being explicitly required from `Gem::Specification` but also a
strange autoload was set for it at `Gem::Version`. The autoload was non
standard because it should've been done in the `Gem` module, not in
`Gem::Specification`, since... -
- 09:13 AM Revision 4bc7cef8 (git): [ruby/cgi] jruby support
- https://github.com/ruby/cgi/commit/93326fb622
-
- The tests fail randomly on some platforms.
http://rubyci.s3.amazonaws.com/ubuntu/ruby-master/log/20220605T213004Z.fail.html.gz
http://rubyci.s3.amazonaws.com/arch/ruby-master/log/20220605T210003Z.fail.html.gz
```
[15737/21701] TestThrea... -
- "choose" is named just for an internal API. We need another good name if we make it public. Continue to discuss in #18183.
-
03:00 AM Revision 52678296 (git): Fix a typo [ci skip]
- 02:40 AM Revision 93555e01 (git): * 2022-06-06 [ci skip]
-
-
- THREAD_MODEL is exported already, so this matches that. Exporting this
is simpler than inspecting configure_args and arch and matching that up
with a specific configure.ac.
Fix GH-5976 -
12:41 AM Bug #18790: cannot load such file -- digest (LoadError)
- There is a similar issue with Red Hat Enterprise Linux 8. To install `libyaml-devel` package, it needs 'codeready-builder-for-rhel-8-x86_64-rpms' repository enabled. Refer to https://github.com/rbenv/ruby-build/discussions/1976 .
06/05/2022
-
- The attached script (and/or others like it) can cause SEGV in 3.0, 3.1, and master. It has always behaved as expected when I use `optflags=-O0`.
When I use it with `make run` on `master`:
```
./miniruby -I../lib -I. -I.ext/common ... -
- I read through [the doc of `SecureRandom`](https://ruby-doc.org/stdlib-3.1.2/libdoc/securerandom/rdoc/SecureRandom.html) and found that it has the `choose` method.
So, I spin up `irb` for a quick test.
~~~
> ::SecureRandom.choose([*... - 06:18 AM Revision 5ce0be02 (git): * 2022-06-05 [ci skip]
-
-
- schneems (Richard Schneeman) wrote in #note-16:
> I am worried about the memory implications for enabling this by default. Is that a valid concern? If we are storing the contents of all files parsed in memory and are loading many files ... -
- Reproduction server:
```
require 'socket'
# Set based on CPU count
CONCURRENCY = 8
server = TCPServer.new(8080)
workers = CONCURRENCY.times.map do
Ractor.new do
loop do
# receive TCPSocket
session = Ract...
06/04/2022
-
- So that an exception raises by non-existent command, not via shell.
https://github.com/ruby/rdoc/commit/fd94dce69d -
- https://github.com/ruby/rdoc/commit/6b1a011243
-
- https://github.com/ruby/rdoc/commit/83051403d6
-
- https://github.com/ruby/rdoc/commit/ce63794fde
-
- `IO.popen` does that job.
https://github.com/ruby/rdoc/commit/3bbbc5ac84 -
- https://github.com/ruby/rdoc/commit/47a1aef447
-
- matz (Yukihiro Matsumoto) wrote in #note-24:
> When investigating #18435, I realized the visibility of a method is not an attribute of the method. Instead, each class should have lists of methods for each visibility. Otherwise, we would...
06/03/2022
-
08:07 PM Feature #18159: Integrate functionality of syntax_suggest gem into Ruby
- Having `SyntaxError#file_path` would be hugely helpful, there are some non-trivial edge cases I have to handle when reading the file name from the message: https://github.com/zombocom/dead_end/blob/main/spec/unit/pathname_from_message_sp...
-
- The attribute `file_path` has been written at #note-7 and answered at #note-8 already.
The input from STDIN can be got from `SCRIPT_LINES__` hash if it is set. -
- Is it useful if `SyntaxError` (or `ScriptError`) has an attribute for the failed locations?
`SyntaxError` often contains multiple failures.
Is it useful an attribute for such list? - 05:50 PM Revision c19a842e (git): * 2022-06-04 [ci skip]
-
05:50 PM Revision e4fe3473 (git): Prevent printing crash report in a loop
- In the event that we are crashing due to a corrupt Ruby stack, we might
re-enter rb_vm_bugreport() due to failed assertions in
rb_backtrace_print_as_bugreport() or SDR(). In these cases we were
printing the bug report ad infinitum with u... -
- @matz I don't think so, the request here was for all `to_s` methods, so `Object#to_s`, `Module#to_s`, `Symbol#to_s`, etc.
The semantic being: "I know I won't need to mutate that string, so you can give me your internal representation ... -
- Probably #18595 addresses this request. Any opinion?
Matz.
-
- Merged as https://github.com/ruby/ruby/commit/9125374726fbf68c05ee7585d4a374ffc5efc5db after several rounds of review with @ko1
-
- Ref: https://bugs.ruby-lang.org/issues/18339
Design:
- This tries to minimize the overhead when no hook is registered.
It should only incur an extra unsynchronized boolean check.
- The hook list is protected with a read-write lock as... -
- https://github.com/ruby/nkf/commit/b386ddc11c
-
- When investigating #18435, I realized the visibility of a method is not an attribute of the method. Instead, each class should have lists of methods for each visibility. Otherwise, we would face issues like #18729 and #18751. With this u...
-
- I thought it was OK to accept this behavior, but it caused issues like #18729 and #18751. At the time of the decision, I haven't noticed those corner cases. Although it has already been shipped with 3.1, I proposed to revert this change....
-
05:35 AM Revision a4721ec6 (git): [DOC] Fix documentation of Numeric#div: Complex does not have #div
- 05:13 AM Revision 512d8a25 (git): * 2022-06-03 [ci skip]
-
-
06/02/2022
-
- zverok (Victor Shepelev) wrote in #note-3:
> @jeremyevans0 I don't think it is a "bug", but maybe a nice feature to have
That makes sense. Switching to feature request. -
04:56 PM Feature #18815: instance_{eval,exec} vs Proc#>>
- @jeremyevans0 I don't think it is a "bug", but maybe a nice feature to have (composite procs being a bit smarter than just a really thin syntax sugar over what you could've done manually). The realistic case I have is a constant with def...
-
- This is basically the same as #18067. I don't think it's a bug, because you get the exact same behavior with manual composition (see example in #18067). However, I welcome feedback from other committers.
-
- ```ruby
measure = proc { p "self=#{self}"; size }
multiply = proc { '*' * _1 }
'test'.instance_eval(&measure)
# "self=test"
# => 4
'test'.instance_eval(&measure >> multiply)
# "self=main"
# NameError (undefined local variabl... -
01:35 PM Misc #18803: DevMeeting-2022-06-16
- * [Feature #18773] Pass an optional range object to deconstruct
* It can be very expensive to compute the array for matching against deconstruct
* By passing a range object we can quickly dismiss matches that won't work
* This c... -
- * [Bug #18813] Module#autoload isn't strict about the autoloaded constant (fxn)
* `module M; autoload :OpenSSL, "openssl"; end` works but is inconsistent.
* `M.constants(false) # => [:OpenSSL]`
* `M::OpenSSL # => ::OpenSSL`
*... -
- https://github.com/rubygems/rubygems/commit/a20bac7924
-
- I found that the current test cases did not cover the bitwise AND
performed on modified words after each iteration
(https://github.com/rubygems/rubygems/blob/7e5765a66c9fe5187b167f619f34db5db121f2df/bundler/lib/bundler/digest.rb#L50)
ht... -
11:59 AM Feature #18814: Ractor: add method to query incoming message queue size
- For instance in GO it is possible to query the current size of a channel:
``` go
c := make(chan int, 100)
for i := 0; i < 34; i++ {
c <- 0
}
fmt.Println(len(c)) // => 34
```
-
- ## Abstract
A simple method to query the current size of a Ractor's incoming queue from outside. Can be used to decide on the sender's side if a message is sent or postponed.
## Background
Ractors have an infinite incoming mess... -
- Another discovery... sleep *after* the commands makes it work as well:
``` shell
# ./ptytest.rb 'echo foo; sleep 1'
Got line "foo"
Exit status 0
#
```
So apparently, it's when the child process is exiting too soon the problem ari... -
- https://github.com/rubygems/rubygems/commit/7c6f15040d
-
- Changing this from Misc to bug, and I'll add it to the upcoming dev-meeting.
-
-
- Define `GC.verify_compaction_references` as a built-in ruby method,
according to GC compaction support via `GC::OPTS`. -
-
- 01:13 AM Revision f35c5a28 (git): * 2022-06-02 [ci skip]
-
06/01/2022
-
- I support raising an error when this fails to load the right constant. I think we can add a warning in 3.2 and make it an error in 3.3.
-
- Thanks for opening the ticket @fxn.
I'm in favor of this change, however I believe that a deprecation cycle is preferable.
This behavior is a bit of a cruft, but at the same time it's not really breaking much things, and it's not ... -
10:55 PM Bug #18813 (Closed): Let Module#autoload be strict about the autoloaded constant
- ## Introduction
Let's consider
```ruby
module M
autoload :X, 'x'
end
```
The constants API does not distinguish existing constants from potential constants:
```ruby
M.constants(false) # => [:X]
M.const_define... -
- Here's some more information to round out this thread. I was a bit sloppy
in [ruby-core:108449] and the output I posted was misleading. The script
was observing return events from `tp.enable`. Tracing is tricky!
Here is an updated ver... -
02:42 PM Feature #18812 (Closed): Add ability to trace exit locations for YJIT
- Currently, when running yjit with `--yjit-stats` you are able to see method call exit reasons and the top 20 most frequent exits. This is useful to know where to spend time investigating whether an exit should be fixed, but in a larger c...
- 02:22 PM Revision 9d18661e (git): Revert incorrect string-guard optimisation. (#5969)
- Also add jhawthorn's test to for this bug.
Fix String#to_s invalidation test - 10:01 AM Revision 1177665e (git): [rubygems/rubygems] Fix `bundle remove` by invalidating cached `Bundle.defintion`.
- Prior to this commit, `bundle add GEM_NAME` updated the lockfile,
but `bundle remove GEM_NAME` left GEM_NAME in the lockfile.
By invalidating the cached `Bundle.definition`, the existing code
handles that without a problem.
https://git... -
- The installer is actually rewriting the spec's full gem path to be the
one of the newly installed gem, however the accessor was not properly
working for `StubSpecification` instances, and default gems are always
of this type, because the... -
- This change was never covered with a spec, and we have recently covered
the case of partially deleted gems with specs and it works fine
(installation is "auto-healed").
https://github.com/rubygems/rubygems/commit/6e66ee4235 -
- This issue was not detected because when all traces of old YAML parser
and emitter `Syck` were removed, this null-type.gemspec.rz marshalled
gemspec was updated to no longer load `YAML::Syck::PrivateType` but load
`Psych::PrivateType` in... -
- This test is making sure that RubyGems is able to load old marshalled
gemspecs that include a field loading `YAML::PrivateType` instances in
the `rubyforge_project` field instead of `nil` due to a bug in old YAML
emitters.
At some point... -
- This old bug used to affect the `rubyforge_project` field in serialized
gemspecs. However, this field has been removed and it's no longer
present in marshaled loaded gemspecs.
So, while the constant is still present in these marshalled ... -
02:01 AM Misc #18803: DevMeeting-2022-06-16
- * [Bug #18806] protected methods defined by refinements can't be called (shugo)
* I prefer "1. Treat defined methods as though they were defined on the refined class. "
* The pull request looks fine: https://github.com/ruby/rub...