Activity
From 02/21/2019 to 02/27/2019
02/27/2019
-
11:59 PM Feature #15626: Manual Compaction for MRI's GC (`GC.compact`)
- Really interesting, but what are the benefits for users who writes Ruby?
Did the Rails app get faster enough? Or you have some other motivations? -
10:57 PM Feature #15626: Manual Compaction for MRI's GC (`GC.compact`)
- Adding updated patch.
-
- I've tested the compactor with a Rails application, and I want to share my results.
I added the middleware below:
```ruby
# This file is used by Rack-based servers to start the application.
require_relative 'config/environment'... -
- Hi,
I've attached a patch that implements a compactor for MRI. I'll try to describe the patch in detail below, but the biggest user facing changes are:
1. A new method `GC.compact` that can be called from Ruby
2. C extensions hav... -
08:11 PM Bug #15625: Module#name performance has exponential-time worst case by aliased constants
- This may be useful to mention at an upcoming developer meeting.
Aliased constants are probably quite common; I use them in a few of my
projects myself, although this is very minor compared to me using aliased
methods (I use a LOT o... -
05:36 PM Bug #15625 (Closed): Module#name performance has exponential-time worst case by aliased constants
- It's well-known (c.f #11119) that `Module#name` has poor performance on anonymous classes, due to searching the entire constant namespace of the VM.
However, we recently discovered that it's even worse than that. In the presence of al... -
02:41 PM Misc #15614: DevelopersMeeting20190311Japan
- * [Feature #14799] Startless range
* What kind of situation is it? I want it quickly :) -
01:55 PM Feature #15624 (Rejected): Allow net/http Response to close before reading entire body
- As far as I understand HTTP doesn't provide such close without reading.
You know HTTP protocol itself doesn't provide such early close mechanism itself.
And though TCP, HTTP's underlying protocol, provides its error handling packet f... -
03:25 AM Feature #15624 (Rejected): Allow net/http Response to close before reading entire body
- Currently net/http has:
```
def reading_body(sock, reqmethodallowbody) #:nodoc: internal use only
@socket = sock
@body_exist = reqmethodallowbody && self.class.body_permitted?
begin
yield
self.body #... - 12:40 PM Revision 6f2c516d (git): * properties.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67139 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
12:40 PM Revision 19408c78 (git): skip to not support color tty environment.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67138 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Applied in changeset trunk|r67137.
----------
Test Bundler examples and bundled gems tests with Azure Pipeline.
* They are invoked on Linux and macOS environment.
* Try to build with vs2017. But it's only binary build.
https://d... -
- * They are invoked on Linux and macOS environment.
* Try to build with vs2017. But it's only binary build.
https://dev.azure.com/rubylang/ruby/
[Feature #15181][ruby-core:89222]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@... -
08:44 AM Bug #15622: Default version of Bundler incorrectly invoked when using binstubs
- Forgot to comment here. I'm pretty sure this is the same as https://bugs.ruby-lang.org/issues/15582? The fix for it should be released in the next patch level version
For the moment `gem update --system` indeed workarounds it, since i... - 03:26 AM Revision c2ee3389 (git): * 2019-02-27
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67136 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 03:26 AM Revision 8092d571 (git): * expand tabs.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67135 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
03:26 AM Revision 5e2a8cb7 (git): Remove stale arguments
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67134 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
03:08 AM Bug #15623: Ruby 2.6.1 Segmentation Fault in on Phusion Passenger server boot in dev
- Forgot to add crash log (from one of the crashes)
```
/Users/pikachuexe/.rvm/gems/ruby-2.6.1/gems/activesupport-5.2.2/lib/active_support/core_ext/time/conversions.rb:55: [BUG] Segmentation fault at 0x000000010260aa3a
ruby 2.6.1p33 (... -
- Ruby is install via RVM
Passenger is install via gem install/bundle install
It's not always crashing the same thread
Happens in 2.6.0 too
But not in 2.5.x
02/26/2019
-
- I guess this may be up to the release manager; perhaps he is a little bit
busy right now. You could consider adding it to the next upcoming developer
meeting ( https://bugs.ruby-lang.org/issues/15614 ) to get feedback whether
there ma... -
-
-
08:22 AM Bug #15622: Default version of Bundler incorrectly invoked when using binstubs
- I was able to reproduce this with the steps provided, and it seems to work after a `gem update --system`. Possibly a bug in a particular version of Rubygems?
- 01:34 AM Revision ff074304 (git): * 2019-02-26
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67133 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
01:34 AM Revision e2d1bb1f (git): add exceptions for indenting conventions for files related to regular expressions
- Ruby uses the Oniguruma/Onigmo regular expression engine, including the underlying
character encoding framework. In contrast to other code implementing Ruby, the
regular expression engine related code is formatted with two spaces per ind...
02/25/2019
-
10:38 PM Bug #15622 (Rejected): Default version of Bundler incorrectly invoked when using binstubs
- When trying to run a binstub with Ruby 2.6.1 with Bundler 2.0.1 on the system, the wrong version of Bundler gets loaded which then fails when the Gemfile.lock points to a 2.x version (since the default version that ships with Ruby is Bun...
-
08:25 PM Feature #15619: Support blacklisting certain dependency versions
- Thanks for your suggestion. Moved to https://github.com/rubygems/rubygems/issues/2659.
-
- This request makes a lot of sense to me.
However, while the gem library is part of Ruby, they are developed separately. I suggest you submit your request to https://github.com/rubygems/rubygems/issues. -
08:51 AM Bug #15621 (Closed): system({}, ...) breaks nonascii ENV["PATH"] on Windows
### Problem
It seems that `system({}, ...)` breaks nonascii `ENV["PATH"]` after execution.
`system(...)` isn't affected. Only `system(env, ...)` is affected.
OS: Windows 7
Ruby: https://github.com/oneclick/rubyinstaller2/releas...-
- Yes. It's a matter of precedence. The "Pickaxe" book describes it as
> Single terms in an expression may be any of the following:
> ...
That means that in ```+foo.gsub!("bar", "car")```, the `.` of method invocation has higher preced...
02/24/2019
-
06:57 PM Bug #15620 (Closed): Block argument usage affects lambda semantic
- The following snippet demonstrate the issue:
``` ruby
def pass_after_use(&block)
raise unless block
lambda(&block).call
end
def direct_pass(&block)
lambda(&block).call
end
pass_after_use do |_arg|
puts "fine, beca... -
- # Abstract
This feature request proposes introducing a new dependency constraint "!=", which will allow to blacklist a specific buggy version of some gem dependency without dropping support for older releases.
# Background
I am ... -
- I have no particularly strong pro or con opinion on the functionality itself.
I have only one comment about syntax, though. While { |line| y << line } }
may or may not be redundant (let's leave that open for the moment), and is
defi... -
03:55 PM Feature #15618 (Closed): Implement Enumerator::Yielder#to_proc
- When writing an Enumerator block, you often want to delegate iteration to another method like this:
```ruby
enum = Enumerator.new { |y|
Dir.glob("*.rb") { |file|
File.open(file) { |f| f.each_line { |line| y << line } }
}
... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67131 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 05:27 AM Revision a582007e (git): * 2019-02-24
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67130 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
05:27 AM Revision 8b8285e2 (git): configure.ac: remove check for broken memmem
- the bug of memmem(3) was fixed in glibc 2.1 released in 1999.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67129 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
02/23/2019
-
- とりあえず`NotImplementedError`とも一番近そうな`ENOSYS`を投げるようにしましたが、本来セットされていない`errno`を装うよりは`RuntimeError`あたりのほうがいいのではないかと思います。
もしくは次点で`nil`。 -
- Linux and AIX have statx() with different prototype.
Linux: int statx(int, const char*, int, unsigned int, struct statx*)
AIX: int statx(char, struct stat*, int, int)
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67128 b2dd03c8-39d... -
- https://github.com/flori/json/releases/tag/v2.2.0
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67127 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
02:22 AM Revision f22ffd83 (git): spec/../rbconfig_spec.rb: skip spec not working on MinGW
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67126 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 02:17 AM Revision 0dab1fce (git): * 2019-02-23
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67125 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- due to MJIT worker's known race condition.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67124 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
02/22/2019
-
- We started carrying code like this in Discourse due to #14634 breaking Queue after fork.
https://review.discourse.org/t/dev-disable-async-logging-in-development-on-broken-ruby/1829
I was wondering if there is any chance we can fa... -
04:07 PM Bug #15613 (Open): Enumerator::Chain#each doesn't relay block signature
-
03:51 PM Bug #15616: Chained destructive methods fail when using +@ to unfreeze a frozen string
- ```
+foo.gsub!("bar", "car")
```
stands for
```
+(foo.gsub!("bar", "car"))
```
and because foo is frozen, you'll get an error, see #15606
try this
```
foo = "bar".freeze
foo = (+foo).gsub!("bar", "car")
```
or
```
foo... -
02:52 PM Bug #15616 (Rejected): Chained destructive methods fail when using +@ to unfreeze a frozen string
- Using the +@ syntax to unfreeze a string does not work when chaining destructive methods
Consider the following;
``` ruby
foo = "bar".freeze
+foo.gsub!("bar", "car")
```
This raises;
``` ruby
FrozenError: can't modify frozen ... -
- [Misc #15615] File.birthtimeがLinux環境で有効なファイル作成時刻を得られなかった場合の挙動について
-
10:20 AM Misc #15614: DevelopersMeeting20190311Japan
- * [Feature #14609] `Kernel#p` without args shows the receiver (aycabta)
* carry over -
- Please comment your favorite ticket numbers you want to ask to discuss with your *SHORT* comment or summary.
(your summary/comment will help us because we don't need to read all of ticket comments)
*DO NOT* discuss then on this ticke... -
- `File.birthtime`はファイルの作成時刻を返すメソッドです。
これまでLinux環境ではこのメソッドは実装されておらず、呼び出すと`NotImplementedError`が発生していましたが、r67088から一部のLinux環境でも`File.birthtime`が動作するようになりました。
Linux環境での`File.birthtime`は、`statx()`というLinux 4.11から導入されたシステムコールを使うことで実装されています。
... - 07:25 AM Revision f53548a8 (git): * expand tabs.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67123 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
07:25 AM Revision b1c1ee71 (git): change `cfunc->invoker` type for opt.
- * vm_insnhelper.c: change `call_cfunc_*` parameters order
and specify a function type for the passed func ptr.
This fix reduce the number of asm instructions, such as:
# before this patch
0000000000000110 <call_cfunc_0>:
... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67121 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67120 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67119 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67118 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67117 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67116 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67115 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
02/21/2019
- 08:59 PM Revision 67032770 (git): * expand tabs.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67114 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * compile.c: refacetoring:
* initialize `branches` with Qfalse intead of 0.
* make compile_call* functions from `iseq_compile_each0()`
to make modifying them easy.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67113 b2dd03c... -
08:02 PM Feature #12745: String#(g)sub(!) should pass a MatchData to the block, not a String
- sawa (Tsuyoshi Sawada) wrote:
> What about having `Enumerator#with_m`, such that
> ...
This doesn't make sense to add to Enumerator in my opinion, since most Enumerators will not be created from `gsub`/`sub`:
```ruby
[1].each.with_... -
07:17 PM Feature #12745: String#(g)sub(!) should pass a MatchData to the block, not a String
- What about having `Enumerator#with_m`, such that
```ruby
gsub(pattern).with_m{|match, match_data| block} → new_str
sub(pattern).with_m{|match, match_data| block} → new_str
```
And if we want to expand that to `scan`, we can intr... -
03:38 PM Revision da797623 (git): Update to ruby/spec@7a16e01
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67112 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 03:38 PM Revision b8e389a0 (git): * 2019-02-22
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67111 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 03:38 PM Revision 203d9467 (git): * properties.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67110 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67109 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
03:28 PM Feature #15605: json library needs more frequent releases
- > Perhaps a better general solution may be to allow the ruby main team to decide who is co-owner of gems IF these gems are part of the official ruby distribution.
This seems like a good rule of thumb. I know we've had trouble finding ... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67108 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67107 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 06:12 AM Revision 32f3961f (git): * 2019-02-21
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67106 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67105 b2dd03c8-39d4-4d8f-98ff-823fe69b080e