Activity
From 01/04/2020 to 01/10/2020
01/10/2020
-
10:36 PM Feature #16499 (Rejected): define_method(non_lambda) should not change the semantics of the given Proc
- From https://bugs.ruby-lang.org/issues/15973?next_issue_id=15948&prev_issue_id=15975#note-38
But I think we should... -
- Yes, sorry I should not have mentioned `define_method` here, even though it's related it's not the main topic.
I'll ... -
02:36 PM Misc #16487: Potential for SIMD usage in ruby-core
- Just to confirm, `__attribute__(target(""))` is how PHP does it:
- https://github.com/php/php-src/blob/7ce531f2... -
02:12 PM Misc #16487: Potential for SIMD usage in ruby-core
- > use __target__ attributes
thanks for the links, very helpful
> It's unfortunate the C compiler cannot do this... -
- > My concern here is who can read, understand and maintain that code?
> Such complexity is likely to have bugs or ne... -
- The code in https://github.com/Shopify/ruby/pull/2/files looks rather arcane to me.
I guess that's often the case fo... -
- > for maximum portability, you should use -march=x86-64
Oh I see what you mean now.
Apparently the solution for... -
- > Could you extend on that ? I'm not sure I understand what you mean here.
for maximum portability, you should use... -
- > You can't use for example AVX instructions generated from a regular C code, even if your processor supports it. You...
-
07:36 AM Misc #16487: Potential for SIMD usage in ruby-core
- Could we do something that allows us to install a gem that gives us the speedup by rerouting stuff in MRI?
Then y... -
- > Do you have any practical applications whose performance is significantly improved by the SIMD hacks? I'm unsure ab...
-
02:08 AM Misc #16487: Potential for SIMD usage in ruby-core
- Do you have any practical applications whose performance is significantly improved by the SIMD hacks? I'm unsure abo...
-
01:57 PM Revision b53d8230 (git): Fixed the wrong url for benchmark
-
-
01:18 PM Bug #16497 (Assigned): StringIO#internal_encoding is broken (more severely in 2.7)
-
11:18 AM Bug #16497 (Assigned): StringIO#internal_encoding is broken (more severely in 2.7)
- To the best of my understanding from [Encoding](https://docs.ruby-lang.org/en/master/Encoding.html) docs, the followi...
-
-
- Applied in changeset commit:git|1b4d406e3a04032b6d01e92b6d184a16945c6ac3.
----------
Hash#transform_values should re... -
11:23 AM Bug #16498 (Closed): Hash#transform_values in 2.7.0 sets new hash's default to old hash's default_proc
- The following is unexpected for me.
```ruby
Hash.new { }.transform_values { }.default
# => #<Proc:0x000055eecda0... -
- [Bug #16498]
-
12:17 PM Revision 135b533e (git): add missing #include
-
- ISO/IEC 9899:1999 section 6.5.7 states that "If the value of the right
operand is negative or is greater than or equa... -
- Replaces `#ifdef _MSC_VER` with more accurate version checks. Also,
`defined(_WIN64) && defined(__AVX2__)` is redund... -
- Fixing typo revealed that _BitScanReverse is BSR, which behaves
differently than LZCNT. What we want here is LZCNT s... -
-
06:34 AM Feature #5781: Query attributes (attribute methods ending in `?` mark)
- sudo (Sudo Nice) wrote:
> How about implementing it similarly to Crystal?
>
> attr_accessor? :foo
+1 -
- Like that!
sudo (Sudo Nice) wrote:
> How about implementing it similarly to Crystal?
>
> attr_accessor? :f... -
- `_1 *1` is parsed as `_1(*1)`, a call to a method `_1` with a variable-length argument with `1`. It is the same as `...
-
02:21 AM Bug #16496 (Rejected): Numbered Parameter not parsed properly in lambda
- The space between `*` and `1` changes the lambda behaviour, looks like numbered parameter `_1` is not parsed properly...
-
02:47 AM Feature #16476: Socket.getaddrinfo cannot be interrupted by Timeout.timeout
- Dan0042 (Daniel DeLorme) wrote:
> +1
>
> This has been an issue for a very long time, and it's often been handled... -
- * [Feature #16495] Inconsistant Quotes in Error Messages (mame)
* This is the third time that we receive a ticket ... -
- alanwu (Alan Wu) wrote:
> I would like to see discussion about [Misc #16487] itself. My notes are a bit of an aside,... -
- To prevent from leaking the seed data.
-
- Expanding less random data with MT is not needed when it
succeeded. -
- This combination is very common, e.g., Emacs, TeX, etc.
-
12:36 AM Feature #16495: Inconsistent quotes in error messages
- Hmm. Is this encoding-related perhaps? I remember having had that when I used ISO (ISO-8859-1 specifically, and
a no... -
01/09/2020
-
11:35 PM Bug #16480: Ruby 2.7, C++, and rb_define_method compilation error
- nobu (Nobuyoshi Nakada) wrote:
> Doesn't `#undef rb_define_method` in that file work?
Ah yes it looks like it doe... -
- Doesn't `#undef rb_define_method` in that file work?
-
09:54 PM Feature #16495 (Closed): Inconsistent quotes in error messages
- Error messages use inconsistent pairs of quotes, for instance:
```
-e:1:in `<main>': undefined local variable or ... - 09:11 PM Revision 5bdc6a01 (git): * 2020-01-10 [ci skip]
-
09:11 PM Bug #16469 (Closed): Wrong warning message at `URI.encode` and `URI.decode`
- Applied in changeset commit:git|d3b28ebc7ac527724831a9fb9ec2f963235f9a21.
----------
Fix warnings for URI.encode and... -
- Use __callee__ to display the called method.
Fixes [Bug #16469] -
07:20 PM Misc #16454: DevelopersMeeting20200116Japan
- mame (Yusuke Endoh) wrote:
> alanwu (Alan Wu) wrote:
> > * [Misc #16487] Potential for SIMD usage in ruby-core (ala... -
- Okay updated.
-
- @mame Here is an updated version of the script which also works if there is no ` (author)` at the end of a comment:
... -
- * [Feature #16494] Allow hash unpacking in non-lambda Proc
* allow `map { |foo:, bar:| ... }` just like array unpa... -
03:01 PM Bug #13758: TestRubyOptions#test_segv_setproctitle segfaults on AARCH64
- Was there some change/improvement in this area? I am trying Ruby 2.7.0 and I have not seen this issue during several ...
-
- Same as 053f78e13988e9253d1f207bf5e23d9505112b32.
emscripten requires a prototype declaration of rb_gvar_readonly_set... -
-
- I wanted to add just a short comment - while byroot is quite possibly correct in regards to most users
using pre-pac... -
- I would like to support this. The linked pull request shows 20x speed up for `coderange_scan`, which is definitely w...
-
- Inside the block where `RB_METHOD_DEFINITION_DECL` family are
defined. -
11:37 AM Bug #16492: TestBugReporter#test_bug_reporter_add test failures
- > Could you attach the reported error log's build.log file?
I meant that people might want to compare the build en... -
- Note seeing Travis's latest 5 s390x jobs, the s390x jobs have done successfully.
This error does not happen on Travi... -
- Building Ruby packages on Fedora, in 90% of cases, I observe the following error on s390:
~~~
1) Error:
TestBu... -
- Copying from the discussion of the PR:
> @eregon IMO, the mutating version is enough here because it is not for a ... -
- First of all, I fully understand the value of separating "real" keyword arguments and disallowing implicit and unexpe...
-
- I can see quite often `TestThreadQueue#test_thr_kill` failing on AArch64:
~~~
1) Failure:
TestThreadQueue#test... -
- The role of this is a so-called "salt" but not "seed", rename to
get rid of confusion with other "seed" of PRNG. -
- Try this regex:
```
/<testsuites(?:(?!<\/testsuites>).|\n)+<\/testsuites>/m
```
`String#scan` returns an arra... -
07:31 AM Feature #16491 (Rejected): ruby string scan can not support full regex string feature
- I have a complex regex string which works fine with match method, but fails with scan. and I think it is because the ...
-
-
- On macOS, GNU coreutils `id` is limited to NGROUPS_MAX groups,
because of the backward compatibility of getgroups(2). -
05:16 AM Feature #13383: [PATCH] Module#source_location
- We now have `Module#const_source_location` #10771, which tells us where the module was named as a constant (which is ...
-
03:59 AM Feature #13383: [PATCH] Module#source_location
- I might have a go at this over the next month or two.
-
03:17 AM Bug #16490 (Third Party's Issue): mkmf.rbの非互換な変更
- tmailのrubygem 1.2.7.1をインストールする際に、
```
make "DESTDIR="
make: don't know how to make /usr/pkg/include/ruby-2.7.0/def... -
01:22 AM Feature #16489 (Open): Make rb_warn_deprecated a public API
- For extension libraries developers, `rb_warn_deprecated` should be useful to warn the API deprecation.
How about mak... -
- `Binding#source_location` returns the `__FILE__` when created, and
may not be an absolute or real path. And in the `... -
01/08/2020
-
- A memo for the dev-meeting discussion
* What we should provide as `Hash.ruby2_keywords!`? mutating version, non-mu... -
- @Dan0042 Very good point, I agree there should be no way to change that flag inplace, since there isn't currently (`*...
-
- I've pushed https://github.com/ruby/ruby/commit/65c2c75e162ebc8c4b35b0823967eeb132c00749. Could you please check if ...
-
- Applied in changeset commit:git|65c2c75e162ebc8c4b35b0823967eeb132c00749.
----------
lib/net/imap.rb: use `&blk` ins... -
10:41 PM Bug #16488 (Closed): 2.7 spews warnings about net/imap
- I just upgraded from 2.6.5 to 2.7.0. Some of my scripts are now emitting warnings that weren't there before. Eg., #...
-
- [Bug #16488]
-
- alanwu (Alan Wu) wrote:
> * [Misc #16487] Potential for SIMD usage in ruby-core (alanwu)
> * It would be great if... -
- * [Misc #16487] Potential for SIMD usage in ruby-core (alanwu)
* It would be great if we can come up with a stance... -
- * [Feature #16463] Fixing *args-delegation in Ruby 2.7: ruby2_keywords semantics by default in 2.7.1 (eregon)
* Th... -
- Okay, thanks!
-
- The test case was relaxed by [1] similarly to my initial proposal. If this is desired fix and not just workaround, th...
-
08:23 PM Bug #9573: descendants of a module don't gain its future ancestors, but descendants of a class, do
- Attached is a work-in-progress patch that includes similar support for `Module#prepend`. It does work in terms of `M...
-
07:53 PM Revision 1d09acd8 (git): [DOC] Improve docs for String#match
- Fix invalid code to make it syntax highlighted; other small fixes.
- 07:51 PM Revision 841a945d (git): * 2020-01-09 [ci skip]
-
- Move existing example to the corresponding paragraph and
add an example for `string =~ regexp` vs. `regexp =~ string`... -
04:13 PM Misc #16487: Potential for SIMD usage in ruby-core
- The topic is
* portability between architectures
* portability in a architecture
* maintenanceability
## Portab... -
- SIMD Everywhere seem very interesting, but from a quick check it seems that all the fallbacks are applied during comp...
-
- > The downside of SIMD instructions is that they are not universally available.
So it means maintaining several vers... -
- ### Context
There are several ruby core methods that could be optimized with the use of SIMD instructions.
I ex... -
- This was very likely resolved via #14340
-
-
-
-
09:09 AM Revision 592d7cee (git): Speeds up fallback to Hash#default_proc in rb_hash_aref by removing a method call
-
- 03:30 AM Revision f518b608 (git): * 2020-01-08 [ci skip]
-
03:26 AM Revision beb59c3b (git): Add GC guard
- Try to fix infrequent error:
https://rubyci.org/logs/rubyci.s3.amazonaws.com/solaris10-sunc/ruby-master/log/20200108... -
12:38 AM Misc #16483: How about stopping new *.tar.bz2 releases?
- If bz2 removal goes forward, it might be nice to deprecate now but not remove until later. It'd be ideal to have work...
01/07/2020
-
- Unfortunately, this bug report does not provide enough information to diagnose this issue. If you can provide a repr...
-
- We can probably use `__callee__` instead of hard coding the method names, that should fix this issue:
```diff
dif... -
- Attached is a patch that implements support for this for `Module#include`, but not `Module#prepend`. It passes `make...
-
03:37 PM Feature #16476: Socket.getaddrinfo cannot be interrupted by Timeout.timeout
- +1
This has been an issue for a very long time, and it's often been handled by installing an asynchronous DNS reso... -
- I would like to recommend this be a non-mutating method. Instead of `Hash.ruby2_keywords!(hash)` it should be somethi...
-
- IIRC this has to do with zero-length matches being ignored in certain conditions, in particular having to do with rep...
-
- The problem is not just for `partition`, but also involves `split` and `scan`.
I think your regex `/^=*/` is unnec... -
-
- RVM also primarily uses bz2.
-
- I prefer to remove bz2 from release packages. I can change to use gz from bz2 at ruby-build.
- 03:23 AM Revision f132825f (git): Disable IPv6 on Travis s390x case. (#2819)
- This fixes following error that sometimes happens once in a few times
on Travis s390x environment.
```
$ tool/travis... -
02:18 AM Bug #16367 (Third Party's Issue): stdin not support ibm866 encoding for windows
- Given that https://github.com/piotrmurach/tty-prompt/issues/115#issuecomment-557919189 says that the `tty-prompt` lib...
-
01:39 AM Feature #16461: Proc#using
- Eregon (Benoit Daloze) wrote:
> shugo (Shugo Maeda) wrote:
> > It doesn't mutate the Proc, but the block, and if Ot...
01/06/2020
-
- @jeremyevans0 @eregon Thanks, I put this on the next dev meeting.
-
- +1, sounds good to me as class methods.
Could also be under ExperimentalFeatures, but not a great fit for this. -
- I'm fine with adding these methods, and I agree that making them class methods instead of instance methods discourage...
-
- Hash's ruby2_keywords flag is designed to be as implicit as possible, but unfortunately, sometimes we need to handle ...
-
- * [Bug #16486] `Hash.ruby2_keywords?(hash)` and `Hash.ruby2_keywords!(hash)` (mame)
* It turned out that we someti... -
- @eregon I agree with your points. I respect that you've studied a lot in your thesis so ultimately I'll defer to your...
-
- `Fiber#[]` and `Fiber#[]=` sounds fine, but what if somebody does:
`some_fiber = Fiber.new { ... }; Thread.new { s... -
08:41 PM Revision b0bf654c (git): always expand ivar arrays to max width
- If the instance variable table hasn't been "expanded", allocate the
maximum size of the ivar table. This operates un... -
- `ruby-install` also uses `.tar.bz2` currently, so changing that would mean that existing ruby-install releases (and i...
-
- I like it. I’m biased though since I’m a sucker for opportunities to simplify like this one.
If we do go ahead, we... -
- I don't mind as in I don't need .tar.bz2.
Personally I use .tar.xz; all local archives of source files that I
kee... -
- Current ruby releases generate `*.tar.gz`, `*.tar.bz2`, `*.tar.xz`, and `*.zip`.
But I think we can stop generating ... -
- > Later I will send the pull-request to disable IPv6 on only Travis s390x-linux case.
I sent the pull-request to f... - 04:40 PM Revision 95d23f36 (git): * 2020-01-07 [ci skip]
-
- ```
.../ruby/test/ruby/test_keyword.rb:3509: warning: assigned but unused variable - bug8993
.../ruby/test/ruby/test_... -
-
- Tainting was made a noop in 2.7.0, this issue can now be closed.
-
12:50 PM Revision 7392083c (git): Support history-size in .inputrc correctly
-
09:52 AM Revision e44e3716 (git): Slightly change the job wording [ci skip]
- mame-san said it's weird
-
- Since 8c9450e7b875db846b19cc631af0d7fee66db5c6, we increased the chance
to run GitHub Actions on a fork, as we usuall... -
- instead of unnecessarily defining env vars
-
- I don't think I have used either one of them so I may not be the primary
target audience either way.
I only reme... -
- I added net-telnet(2015) and xmlrpc(2016) to bundled gems. They are the old protocol in 2020.
Should we remove the... -
- I discussed about promote(demote?) rexml and rss to the bundled gems with @kou
Pros: @kou can handle to maintain a... -
- Applied at the following commits.
* https://github.com/ruby/ruby/commit/2e5ef30cb9f56e5a7a8139e0f1d75bbcf5ee8362
... -
- Applied at https://github.com/ruby/ruby/commit/1658e6b5db0380c39d2423281e10acc5b6c6a8bd
-
- https://github.com/flori/json/commit/2f3f44c180
-
- https://github.com/flori/json/commit/3845491d92
-
- https://github.com/flori/json/commit/4ede0a7d19
-
- This text used to be true in older versions of json, but has not
been true for a number of years (since json version ... -
06:06 AM Revision 33d86655 (git): support RUBY_ON_BUG envval on assert failure.
- Check RUBY_ON_BUG env val also on rb_assert_failure().
-
- Note that 2.6 (.4-?) has same issue.
-
- Applied in changeset commit:git|ce072fe5689184cba5e4a86968367c525cb22a72.
----------
script_compiled event on compil... -
- script_compiled event for TracePoint should not be invoked on
compile error (SyntaxError) because it is not "compiled... -
- It indeed seems to be a bug. For C extensions, that arity argument can be a variable. There is a fallback code. Bu...
01/05/2020
-
- Yes, using 2.7.0:
```
> git clone https://github.com/ioquatix/covered
> cd covered
> COVERAGE=Summary rspec
... -
- @matz I agree with adding all three APIs, `Fiber#[]`, `Fiber#[]=` and `Fiber#backtrace`. Can you let me know if you a...
-
- Can you report this to the administrator of https://ruby-doc.org/? It's NOT the official site.
-
05:52 PM Feature #16482 (Closed): net/http should support TLS connection to proxies
- Right now net/http forces the user to use a clear text connection to a proxy. This massively reduces security as the ...
- 04:24 PM Revision 46845d03 (git): * 2020-01-06 [ci skip]
-
- def foo
''.upca[TAB]
This will be completed to be:
def foo
''.upcase
The indent was gone. This commit fixe... -
- [The log for developers' meeting 20191128](https://docs.google.com/document/d/1AZ74HXEedKksJwhEUPIlnRxAUgchndZPZYAKjG...
-
- If C-d is pressed before IRB is ready, IRB crashes because EOFError occurs.
-
- In https://github.com/ruby/ruby/commit/9be3295d53b6fd9f8a3ad8157aa0655b1976d8ac,
OpenStruct's documentation stopped t... -
- ref 44a164c26f5371519636585d8ba7aa59f489442e and beae6cbf0fd8b6619e5212552de98022d4c4d4d4
-
- ref c7f01d889becbeffc4254e1b0b7faecb80ea3f3e and beae6cbf0fd8b6619e5212552de98022d4c4d4d4
-
-
- It is no longer used due to beae6cbf0fd8b6619e5212552de98022d4c4d4d4.
Coverity Scan found this. -
- Due to beae6cbf0fd8b6619e5212552de98022d4c4d4d4, the variable last_idx
is no longer changed and always -1. This chan... -
02:28 AM Revision 9b928fa4 (git): [ruby/reline] Sort completion list
- #### Legacy mode:
```console
$ irb --legacy
irb(main):001:0> l[TAB][TAB]
lambda load l... -
- Sorry, those tags ware mixed into the ruby/readline-ext repository when I cherry-pick-ed from ruby/ruby. I removed it.
-
02:05 AM Feature #16468: Switch to Miller-Rabin for Prime.prime?
- On second thought, I think Marc is right, we can't ruin someones day with a composite without a warning that theres a...
-
- It can fall back to APR-CL primality test when Miller-Rabin does not work. In my personal opinion, it would be best ...
-
- marcandre (Marc-Andre Lafortune) wrote:
> Interesting. We might as well always return the correct result, i.e. apply... -
- @byroot, yeah.
`rb_funcall(str, rb_intern("-@"))` works and is backwards compatible, but I entirely agree that it...
01/04/2020
-
11:09 PM Feature #16468: Switch to Miller-Rabin for Prime.prime?
- Interesting. We might as well always return the correct result, i.e. apply the fast algorithm for integers < 318,665,...
-
- Attached is an implementation against master....
-
-
- I now think that writing as below is good enough:
```ruby
6.times.with_index(1){|_, i| puts "Chapter#{i}"}
```... -
-
- I do not feel this proposal to be natural any more. Please close this issue.
-
-
- This proposal was adopted in Ruby 2.6 (and is mentioned in the context of #14944, which is actually a duplicate of th...
-
-
- This is another duplicate of #10797. Please close this issue.
-
-
- Indeed this issue is a duplicate. Please close this.
-
-
- This proposal is indeed a duplicate of #8804. Please close this.
-
-
- I should use `each_with_index` and `each_with_object` as marcandre suggests. I withdraw this. Please close this issue.
-
- Actually, this proposal is a duplicate of #10797
-
- When you expand the "Core API" section in https://ruby-doc.org/, you can see that there is version 2.6 (instead of 2....
-
- I maintain the Rice library which provides constructs for exposing C++ libraries into Ruby. With Ruby 2.7.0 I've got ...
- 03:53 PM Revision ded072c2 (git): * 2020-01-05 [ci skip]
-
-
- From @jeremyevans0:
> Previously it wasn't possible because there wasn't a way to go from the module to all iclass... -
- Is there a clean way to fix this issue?
-
06:27 AM Feature #5321: Introducing Numeric#exact? and Numeric#inexact?
- I really think this is a very overblown issue that is being handled incorrectly.
First, we have what appears to be... -
- * [Bug #16383] TracePoint does not report calls to attribute reader methods (jeremyevans0)
* Do we want to support... - 04:13 AM Revision b594005c (git): * 2020-01-04 [ci skip]
-
- Applied in changeset commit:git|04eb7c7e462d1fcbab9efc7022c1b43636ab878a.
----------
Call initialize_clone with free... -
- Applied in changeset commit:git|0eeed5bcc5530edb0af2af2ccff09d067c59e8f9.
----------
Make eval(code, binding) use (e... -
- This makes it possible to initialize_clone to correctly not freeze
internal state if the freeze: false keyword is pas... -
- This removes the warning that was added in
3802fb92ff8c83eed3e867db20f72c53932f542d, and switches the behavior
so tha... -
- Whoopsie, changing to the Feature tracker (from Bug)
-
- References https://github.com/ruby/ruby/pull/2812
I stumbled onto these 2 tables (`vm_opt_method_table` and `vm_op... -
- References https://github.com/ruby/ruby/pull/2814
Back fiber local variables with an ID table instead of `st_table...
Also available in: Atom