Activity
From 04/12/2021 to 04/18/2021
04/18/2021
-
03:02 PM Feature #12543: explicit tail call syntax: foo() then return
- mame (Yusuke Endoh) wrote in #note-9:
> And I have another concern. If tail call optimization is on by default, som... -
- x3qt (Yury Polyakov) wrote in #note-6:
> shyouhei (Shyouhei Urabe) wrote in #note-4:
> > Tell us how is it better t... -
10:06 AM Bug #17723: autoconf 2.70+ is not working with master branch
- And 5a9b5b8e953dc36f4f7f7639e8aeaa8132241c79 too.
-
- ruby_3_0 a55ba76617f240337854668c8b5dd9961873f5af merged revision(s) 4250d7c910fd67c0da07c6213901009330e3cd51,83f9664...
-
-
- autogen.sh: Added
Currently it just runs "autoreconf --install"
---
aclocal.m4 | 4 ++--
... -
- spec/ruby/core/hash/transform_keys_spec.rb: Fix the failure of
ruby_3_0
https://github.com/ruby/spe... - 04:49 AM Revision aae812d6 (git): * 2021-04-18 [ci skip]
-
04:47 AM Revision 6943ffc3 (git): Split revision.h rule
- GNU-make specific rule is defined in defs/gmake.mk.
04/17/2021
-
11:43 PM Feature #17808: Feature Request: JS like splat of Object properties as named method parameters
-
> But ultimately your whole feature request rely on a very specific JS "feature" (many people would call it a cruft... -
09:00 PM Bug #17812 (Closed): Timeout::Error doesn't let two-argument raise() set a new message
- Shouldn't this:
```
martind@pizzagate:~$ ruby -we 'require "timeout"; raise(Timeout::Error.new("hello"), "world")'
... -
05:43 PM Feature #17795: Around `Process.fork` callbacks API
byroot (Jean Boussier) wrote in #note-21:
> I must admit I don't understand this argument. There are plenty of adv...-
11:02 AM Feature #17795: Around `Process.fork` callbacks API
- > some committers agreed with Jeremy's concern (easy to misuse).
I must admit I don't understand this argument. Th... -
- This thicket was discussed in the dev meeting, but no conclusion was reached. The current situation is not so good, b...
-
05:06 PM Bug #11230: Should rb_struct_s_members() be public API?
- At the April 2021 developer meeting, it was decided to remove this C-API function after Ruby 3.2.
-
- At the April 2021 developer meeting, it was decided to make this change in Ruby 3.2.
-
-
- Autoconf 2.70 seems to omit the check for the given CC,
`AC_COMPILE_IFELSE` does not know which is the cause of the
f... -
- # The next dev meeting
**Date: 2021/05/21 13:00-17:00**
Place/Sign-up/Agenda/Log: https://github.com/ruby/dev-mee... -
-
- This ticket was discussed in the dev meeting, but no conclusion was reached.
BTW, Mathematica's "Accumulate" is di... -
08:04 AM Feature #17753: Add Module#namespace
- In my view, the way to implement the use case that matches Ruby is to go downwards.
Module has many constants, that ... -
- This ticket was discussed on the dev meeting. @matz said that (1) the use case is not clear to him, and that (2) he w...
-
- Thank you! Then, I think the patch should use const_source_location info instead of adding `file` and `line` fields.
... -
- This ticket was discussed on dev-meeting. A method (or proc) created in an `eval` context should be marked and `AST.o...
-
06:44 AM Bug #17810 (Closed): addr2line.c: Some symbol names missing in C backtrace with Clang/LLVM
- When Clang/LLVM is used for compilation, some symbol names are missing in C level backtrace information. An example:
... - 02:54 AM Revision ed9ac039 (git): * 2021-04-17 [ci skip]
-
02:54 AM Revision d953ac11 (git): Fix writing XDG_CONFIG_HOME in test-bundler
- https://github.com/ruby/actions/actions/runs/756591173
```
Errno::EACCES: Permission denied @ dir_s_mkdir - /home/run...
04/16/2021
-
09:46 PM Bug #17809 (Closed): Ruby 2.6.7 backported C99 code that breaks older compilers
- Hey folks,
It looks like the 2.6.7 release inadvertently backported some C99 code from HEAD onto the Ruby 2.6 bran... -
09:05 PM Feature #17808: Feature Request: JS like splat of Object properties as named method parameters
- I believe the answers given so far should help (i.e. `(**rest)` and defining `to_hash`).
Also looks into `Struct` ... -
- > Is there a way to modify existing declared methods adding that parameter
Well, I suppose you can decorate the me... -
- byroot (Jean Boussier) wrote in #note-3:
> Well, the ruby equivalent to your example JS function would be:
>
> ```ru... -
- Well, the ruby equivalent to your example JS function would be:
```ruby
def test(param1: nil, param2: nil, **)
e... -
- Thanks for the quick update.
That almost works but to automate this the to_hash function needs to be aware of the m... -
- Doesn't the implicit `to_hash` answer your demand?
```ruby
def foo(bar:)
p bar
end
class MyObject
def i... -
- I'm pretty sure there is no equivalent Ruby for a very convenient JS way to enter named function parameters as below:...
-
05:59 PM Feature #17749: Const source location without name
- mame (Yusuke Endoh) wrote in #note-8:
> @tenderlove I'm not sure about the use case. What purpose do you want this f... -
- @tenderlove I'm not sure about the use case. What purpose do you want this for? Memory profiling?
`const_source_lo... -
01:28 PM Feature #17795: Around `Process.fork` callbacks API
- At Datadog, we use our monkey patching (mentioned in the description) to restart background threads that are used to ...
-
- > It doesn't need system call unlike getpid.
It would be a win for sure, but in my opinion we might as well go all... -
08:03 AM Feature #17795: Around `Process.fork` callbacks API
- Another idea is introducing Process.fork_level which can be used to detect fork instead of getpid.
```
Process.fo... -
- Hey there! I'm in [Datadog's ddtrace gem](https://github.com/DataDog/dd-trace-rb) team and one of our customers was r...
-
08:51 AM Bug #16608: ConditionVariable#wait should return false when timeout exceeded
- Previous dev-meeting (March), there is no objection to change the return value of `Mutex#sleep`.
> matz: agreed to... -
08:05 AM Feature #17762: A simple way to trace object allocation
- It's OK to add this feature. The side effects are acceptable (for this is only for debugging anyway). I am OK with an...
-
- Applied in changeset commit:git|b6bb4623ebcf1335cf0a81d279cde24a9e97bcce.
----------
NEWS for [Feature #15198] [ci s... -
07:03 AM Feature #15198: Array#intersect?
- We discussed this in today's developer meeting. Here are some remarks:
- It takes one argument for now.
- It means ... -
- Accepted.
Matz.
-
- Applied in changeset commit:git|b6bb4623ebcf1335cf0a81d279cde24a9e97bcce.
----------
NEWS for [Feature #15198] [ci s... -
- 07:06 AM Revision 55d91a09 (git): Add Array#intersect?
-
-
- It's intentional.
Matz.
-
05:41 AM Revision 97cee1cb (git): rb_hash_free(): has never existed.
- This is just a hoax. Nobody have ever implemented a function named as
such. Also the functionality implied by the n... -
05:25 AM Bug #17777: 2.6.7 fails to build on macOS: implicit declaration of function 'rb_native_mutex_destroy' is invalid in C99
- Fixed at https://github.com/ruby/ruby/pull/4360
-
- Pros
* More intuitive / consistent / natural
Cons
* Duplicated syntax rules
* Even more complex syntax
If I were yo... -
04:12 AM Feature #17403: Remove Fixnum and Bignum
- No at least 3.1.
Don't introduce incompatibility in 3.1. -
03:25 AM Bug #17806 (Closed): Bad interaction between method cache, prepend, and refinements
- I'm running into a couple of issues with Ruby 3's new method cache and
refinements.
The first script raises `Syst... - 02:05 AM Revision dd3eca64 (git): * 2021-04-16 [ci skip]
-
- https://github.com/ruby/spec/pull/833
https://github.com/ruby/spec/commit/8290e5ad8952b14ee9a5069651d9864c66681112
04/15/2021
-
08:30 PM Bug #17802: backport fbbc37dc1d5b329777e6d9716118db528ab70730 to 2.7 and 3.0
- ruby_2_7 67f1cd20bfb97ff6e5a15d27c8ef06cdb97ed37a merged revision(s) fbbc37dc1d5b329777e6d9716118db528ab70730.
-
- ruby_3_0 d5a34e1b142eccc54971494be2243a0c6ac01d94 merged revision(s) fbbc37dc1d5b329777e6d9716118db528ab70730.
- 08:30 PM Revision 67f1cd20 (git): merge revision(s) fbbc37dc1d5b329777e6d9716118db528ab70730: [Backport #17802]
- test/drb/test_drb.rb: Specify the host of DRbServer
to try fixing the following error.
http://rubyc... -
08:10 PM Bug #17777: 2.6.7 fails to build on macOS: implicit declaration of function 'rb_native_mutex_destroy' is invalid in C99
- https://github.com/ruby/ruby/pull/4385
-
07:31 PM Bug #17805: Segmentation fault at 0x0000000000000010 ruby 2.7.3p183
- peterzhu2118 (Peter Zhu) wrote in #note-1:
> Thank you for the report. It's a duplicate of #17780 which has been fla... -
07:30 PM Bug #17805 (Closed): Segmentation fault at 0x0000000000000010 ruby 2.7.3p183
-
- Thank you for the report. It's a duplicate of #17780 which has been fixed and flagged for backport (so it will land i...
-
- 02:10 PM Revision f1f08f5b (git): Remove useless attribute set in init_mark_stack
- init_mark_stack already clears the mark stack so we do not need to
set the attribute cache_size to zero. -
-
-
-
-
-
- c.f. https://opensource.apple.com/source/ruby/ruby-145.40.1/patches/tool_mkconfig.rb.diff
* moved to "includedir" fr... -
- merged into ruby_3_0
https://github.com/ruby/ruby/commit/6add6014d8626c4e5f6816030ea651321e6aa044
https://github.co... -
-
- Update for autoconf 2.70
---
configure.ac | 232 +++++++++++++++++----------... -
- Removed unused AC_CHECKING
---
configure.ac | 7 -------
1 file changed, 7 deletions(-) -
- [Bug #17521] won't be backported until 3.1.0.
https://github.com/ruby/spec/pull/832 -
- Hi @xtkoba
Are you ineterested in becoming a Ruby commiter?
At least, @nobu and I want you to have a commit bit.
... -
- xtkoba (Tee KOBAYASHI) wrote in #note-2:
> Maybe reproducible by spawning hundreds of threads each of which calls de... -
- Although I am not sure, this looks like a thread-safety issue. Maybe reproducible by spawning hundreds of threads eac...
-
- Since `matrix` is a gem that can be updated independently from Ruby, I think backports are not required. In addition,...
-
- Even though nobody filled the Backport field yet, I will set "WONTFIX".
Such a long long standing bug could be thoug... -
02:14 AM Revision 92545977 (git): [Doc] Add Hash#value? into call-seq (#4293)
-
- FWIW, that code was introduced by commit commit:c995315153f7403636ec0d6119b7885668b32039.
-
01:17 AM Bug #17804 (Closed): Math.cbrt(0.0) returns NaN
- Math.cbrt(0.0) returns NaN. It should return 0.0.
From reading the implementation, I expect this bug to happen onl... -
- ruby_3_0 84d9a9afc0b49d095541acb9832f8b12fb506e19 merged revision(s) 31e0382723bfb35cffe3ca485dd0577668cafa07,5e5fb72...
-
- Keep non evaluated keys in `Hash#transform_keys!` [Bug #17735]
---
hash.c ... -
- I think this is a performance thing.
If there are any real world application affected by this issue, please let me k... -
- The fix was already backported at 3e67bfe202d6b410375af990ac806f15cb2df762.
-
- test/drb/test_drb.rb: Specify the host of DRbServer
to try fixing the following error.
http://rubyc... -
04/14/2021
-
07:16 PM Bug #17801: repeated read from PTY blocks all Fibers
- ioquatix (Samuel Williams) wrote in #note-1:
> Investigating.
Hello Samuel, if found the following:
**A read on a PT... -
06:28 PM Feature #17790: Have a way to clear a String without resetting its capacity
- > Maybe String#capacity and String#capacity= would make sense?
Using `capacity=` for the method name would set the... - 05:51 PM Revision d9a52813 (git): * 2021-04-15 [ci skip]
-
05:51 PM Revision b190f6ef (git): [ruby/matrix] Fix a typo in a error message
-
- You should add this as an issue to the zlib repository: https://github.com/ruby/zlib/issues
However, before doing ... -
02:19 PM Bug #17803 (Closed): Segmentation fault during deflate
- Hi there,
I've been running on a lot of crash when using permessage_deflate gem for websocket that uses the zlib d... -
- I updated the description with:
- Added the third hook (after fork in parent) on @ko1's demand.
- Added links... -
- For debugger, I also want to use.
I want to use 3 hooks like `pthread_atfork`. -
- Ruby 1.9 is EOL today.
-
- ruby_2_7 and ruby_3_0 snapshot test fails during drb test suite.
https://github.com/ruby/actions/actions/runs/7457... -
- Thanks to user:shyouhei's work, we are now able to notice undefined identifiers evaluated in `#if` directives using G...
04/13/2021
-
- ~~@mame I'm afraid I don't have the permission to edit my own tickets.~~ Nevermind.
-
- @byroot I see, thanks! I think that the intended behavior of the proposed APIs is valuable to be explained in the tic...
-
- > what process calls the after_fork hook? Only parent, only child, or both?
The `child`. But I agree that a better... -
02:46 PM Feature #17795: Around `Process.fork` callbacks API
- > Do people actually fork in the middle of a transaction?
I think the most likely case is if the transaction is un... -
- I'd like to make sure: what process calls the `after_fork` hook? Only parent, only child, or both?
In the previous... -
11:16 AM Feature #17795: Around `Process.fork` callbacks API
- @byroot Thanks for the new issue, this seems much clearer than the previous one :)
jeremyevans0 (Jeremy Evans) wrote... -
- @jeremyevans0 I understand your point, and I agree that in an ideal world this wouldn't be necessary.
However prag... -
- The main issue I see with this is the potential for misuse. If I could be sure this would only be used by applicatio...
-
01:32 AM Feature #17795: Around `Process.fork` callbacks API
- Python provides `os.register_at_fork` API for registering callbacks which are called on both before and after fork.
... -
- Looks like you are referring to the mail gem, which is not part of Ruby itself: https://github.com/mikel/mail/blob/8f...
-
07:14 PM Bug #17800 (Third Party's Issue): Cannot disable SMTP TLS through mail library
Net::SMTP now defaults to automatic TLS; the **mail** interface hasn't been updated to reflect this. It will turn T...-
- expected output from script on cli:
date
Di 13 Apr 2021 21:51:20 CEST
test
date
Di 13 Apr 2021 21:51:21 CEST
te... -
07:51 PM Bug #17801: repeated read from PTY blocks all Fibers
- Investigating.
-
- Ruby version:
ruby 3.0.1p64 (2021-04-05 revision 0fb782ee38) [x86_64-linux]
used Scheduler:
https://github.com/... -
- My explanation in #note-1 is partly incorrect. What is wrongfully 0 is the value of the class itself, not the class s...
-
- My observation is that the function rb_class_clear_method_cache is called with argument klass=0, which clearly causes...
-
05:28 PM Bug #17799 (Feedback): Seg fault in rb_class_clear_method_cache
- Recently our CI tests have been intermittently failing with segmentation faults at random points, such as:
```
/b... -
- BTW, you all know AS has this concept right? https://github.com/rails/rails/blob/f1e00f00295eb51a64a3008c7b1f4c4f46e9...
-
- The lexical parent module happens to be just the object from which you set the name, which does not even reflect the ...
-
- I see, the name `namespace` is what we're disagreeing on.
Maybe you have another suggestion for the name of that met... -
- Oh, let me say something explicitly: You guys are Ruby committers, you are the ones that have the vision for what mak...
-
- To me, the ability of a namespace being namespace of itself
```ruby
m = Module.new
m::M = m
```
is one clear indica... -
- BTW, we were discussing yesterday with Aaron that the flag I am raising is about the name `namespace`. What we are de...
-
- > It's also a way to inform the user where that constant lives. The contents of the string have meaning.
The numerou... -
03:07 PM Misc #17734: DevelopersMeeting20210416Japan
- * [Feature #17016] Add `Enumerable#accumulate` (parkerfinch)
* Is the name `#accumulate` acceptable?
* Is this ... -
- I like `#accumulate`, and thank you @nobu for an implementation [here](https://github.com/ruby/ruby/pull/3361)!
Is... -
- byroot (Jean Boussier) wrote in #note-52:
> @mame I created https://bugs.ruby-lang.org/issues/17795, let me know if ... - 01:37 PM Revision 4eefb057 (git): Add RSymbol struct back into RVALUE
- Commit 0ca714fa1aa3fbe4fb60ae1e5b730e544dabc27b removed RSymbol from
RVALUE. This commit adds RSymbol back into RVALUE. -
- https://docs.ruby-lang.org/ja/latest/method/ObjectSpace/m/define_finalizer.html に
> proc の呼び出しで発生した大域脱出(exitや例外)は無... -
- @xtkoba Your 4 patches look good to me, could you commit them?
-
- Not only NUM2LL, also LL2NUM, ULL2NUM and NUM2ULL.
-
- Previous code failed to compile on MSVC. Log:
https://github.com/ruby/ruby/pull/4371/checks?check_run_id=2304484466
... -
-
- Use of TOKEN_PASTE was a bad idea at the first place. Just use ##
everywhere. Nobody practically lacks token pasting. - 04:12 AM Revision 6878d517 (git): * 2021-04-13 [ci skip]
-
- Could you please commit it yourself?
-
- In MIPS ELF, DWARF sections are of type `MIPS_DWARF`, not `PROGBITS`.
The attached patch modifies `addr2line.c` so... -
04/12/2021
-
- I think those edge cases are pretty rare.
Module#namespace would refer to the lexical parent when the module is crea... -
- > Yes, c.name is "X::C", but as I said above, that is just a string.
It's also a way to inform the user where that... -
05:07 PM Bug #17796 (Third Party's Issue): ruby does not compile with openssl-alpha14
- Yes. Work for this is being tracked at: https://github.com/ruby/openssl/issues/369
-
- This is probably a duplicate of #17628.
-
04:42 PM Bug #17796 (Third Party's Issue): ruby does not compile with openssl-alpha14
- ruby does not compile with openssl-alpha14
make[2]: Entering directory '/usr/src/photon/BUILD/ruby-2.7.1/ext/-test... -
- byroot (Jean Boussier) wrote in #note-2:
> > Afaik the proper way to do this is to close the connection after the fo... -
- > Afaik the proper way to do this is to close the connection after the fork.
No before. Otherwise the connection i... -
- > Most database clients, ORMs or other libraries keeping a connection pool might need to close connections before the...
-
- Replaces: https://bugs.ruby-lang.org/issues/5446
### Context
Ruby code in production is very often running in a... -
-
- @mame I created https://bugs.ruby-lang.org/issues/17795, let me know if it makes sense to you, in which case I'll upd...
-
- byroot (Jean Boussier) wrote in #note-50:
> @mame I'll do it.
Thanks! -
- @mame I'll do it.
-
- Can anyone summarize the latest proposal? I cannot understand what is needed precisely. I'm afraid if some people see...
-
- Applied in changeset commit:git|a5688b5ce6e87a5b881a4d52d4e59b6fe807e807.
----------
Support non-standard `struct st... -
- With the patch in #note-6 applied, `file.c` is compiled without any error or warning (including `-Wundef`) for armv7a...
-
- Patch updated.
-
- It makes no difference because neither `SIZEOF_DEVT` nor `SIZEOF_DEV_T` is defined on Android (both for 32- and 64-bi...
-
- All `SIZEOF_DEVT` were typo.
-
- Sorry, the second patch had a typo, `SIZEOF_DEVT` should be `SIZEOF_DEV_T`.
Is it fine? -
- user:nobu Thanks, it builds for armv7a-android without error or format warning.
-
- Do these work?
```diff
diff --git i/configure.ac w/configure.ac
index 6fe43bfc8dd..f08597a007c 100644
--- i/configur... - 06:58 AM Revision 59171378 (git): * 2021-04-12 [ci skip]
-
- On 32-bit Android:
* `st_dev`/`st_rdev` are not `dev_t`
* `st_mode` is not `mode_t` -
- When built without zlib (`./configure ac_cv_lib_z_uncompress=no`), `ruby` aborts immediately after C level backtrace ...
-
- What about `buffer.clear(capacity: 1024)`
Or maybe even `buffer.clear(capacity: 1024..8192)`
I think that's more st...
Also available in: Atom