Activity
From 04/24/2014 to 04/30/2014
04/30/2014
-
08:56 PM Bug #9790 (Closed): Zlib::GzipReader only decompressed the first of concatenated files
- There is a similar old issue in Node that I came across that perfectly describes the situation in ruby:
https://github.com/joyent/node/issues/6032
In ruby given the following setup:
```
echo "1" > 1.txt
echo "2" > 2.txt
gzip ... -
05:25 PM Bug #9787: r45698 causes a segv in Rails tests
- @nobu the segv seems fixed, but now we get an error:
~~~
/Users/aaron/git/rails/activesupport/lib/active_support/dependencies.rb:247:in `require': wrong argument type Symbol (expected Symbol) (TypeError)
from /Users/aaron/git/rails... -
08:07 AM Bug #9787 (Closed): r45698 causes a segv in Rails tests
- Applied in changeset r45756.
----------
parse.y: pin down dynamic symbol only
* parse.y (rb_id_attrset): pin down dynamic symbol only. it is
possibe that attrset ID can be registered as a static symbol
after the corresponding attrg... -
-
- Aaron Patterson wrote:
> ~~~
> ...
`73328908` is `0x45ee90c`, so it is a static symbol.
> ~~~
> ...
Seems "new" ID with "=" was not actually new, but it has been interned in C level.
Does this fix it?
~~~diff
diff --git a/pa... -
05:11 PM Feature #9781: Feature Proposal: Method#super_method
- (2014/05/01 1:35), nobu@ruby-lang.org wrote:
>
> It's an ordinary `Method` (or `UnboundMethod`) instance, same as `SuperClass.instance_method(:foo).bind(obj)`.
Oh, OK. I misunderstood the spec of Method class.
--
// SASADA ... -
- It's an ordinary `Method` (or `UnboundMethod`) instance, same as `SuperClass.instance_method(:foo).bind(obj)`.
-
- What should happen on `#call` method with method objects returned from `super_method`?
-
- * win32/win32.c (ustatfs): need a linkage to get rid of conflict.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45770 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- `IO::SEEK_DATA`自体はうごくんでしょうか。
~~~diff
diff --git a/test/ruby/test_io.rb b/test/ruby/test_io.rb
index d2abd9e..5ed2ddd 100644
--- a/test/ruby/test_io.rb
+++ b/test/ruby/test_io.rb
@@ -1715,7 +1715,7 @@ class TestIO < Test::Unit::Te... -
04:26 PM Bug #9789 (Closed): Error in TestIO#test_seek_symwhence and TestIO#test_seek on systems without File::Statfs#type (e.g. Solaris)
- File::Statfs#type が無い（NotImplementedErrorになる）環境にて、
TestIO#test_seek_symwhence と TestIO#test_seek がエラーになります。
Solaris 10 (64ビット)、r45760 (+ Bug #9788 のパッチ）にて確認しました。
~~~
52) Error:
TestIO#test_seek_symwhence:
NotImplementedError: ty... -
- * win32/win32.c (ustatfs): declare GetVolumePathNameW for old VC.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45769 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- > 過去のリリースでコンパイルしたバイナリでは動作するのかもしれません。
Bad System Call が出てエラーになったのは64ビットでコンパイルした場合で、32ビットでは未確認でした。32ビットでコンパイルしたら、SunOS時代のバイナリが今でも一応動くのがSolaris10の売りですし、動作する気がします。 -
- Applied in changeset r45768.
----------
* file.c (FSTATFS): check availability of struct statfs and
struct statvfs in addition to fstatfs(2) and fstatvfs(2).
This fixes error in Solaris. [Bug #9788] [ruby-dev:48145] -
02:48 PM Bug #9788: TestFile#test_statfs = Bad System Call on Solaris
- In message <redmine.issue-9788.20140430140222.829abc7694e7af28@ruby-lang.org>
on Wed, 30 Apr 2014 14:02:22 +0000,
ngotogenome@gmail.com wrote:
> という記述があり、現在のSolaris10では、削除はされていないものの、呼ん
> でもSIGSYSかENOSYSを返してまともに動作しないようです。（または、未確
>... -
- Solaris 10 にて、make test-all が以下のエラーで中断します。(r45759 で確認。)
~~~
TestFile#test_statfs = Bad System Call
make: *** [yes-test-all] Error 140
~~~
以下のように、statfsメソッド内にて呼んでいるfstatfsシステムコールがSIGSYSを発生させているようです。
~~~
% truss ruby -e 'f ... -
- struct statvfs in addition to fstatfs(2) and fstatvfs(2).
This fixes error in Solaris. [Bug #9788] [ruby-dev:48145]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45768 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * parse.y (rb_id_attrset): show the name of broken ID if it is
still registered.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45767 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * parse.y (rb_id_attrset): all valid ID types are handled
properly, raise with numeric ID type which may be broken for
GCed static symbol.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45766 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 03:16 PM Revision 58fdffaa (git): * 2014-05-01
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45765 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * parse.y (rb_id_attrset): use rb_id2str to get rid of method call.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45764 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * ext/-test-/symbol/intern.c: Symbol.find is available now.
* test/-ext-/symbol/test_inadvertent_creation.rb (noninterned_name):
use Symbol.find instead.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45763 b2dd03c8-39d4-4d8f-98ff... - 02:53 PM Revision d46cd5cf (git): missed at r45748.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_0_0@45762 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- In message <redmine.journal-46331.20140426142125.1b2e3962cfe9fb0b@ruby-lang.org>
on Sat, 26 Apr 2014 14:21:25 +0000,
akr@fsij.org wrote:
> Issue #9772 has been updated by Akira Tanaka.
>
>
> Takahiro Kambe wrote:
>> 「NetBSD ... -
06:27 AM Feature #9772: IO#statfs and File::Statfs
- Akira Tanaka wrote:
> ruby-core:62137 で Eric Wong も尋ねていますが、なぜ POSIX で定義されている statvfs じゃなくて、statfs なのか、というのは私も最初に疑問に思いました。
> ...
statfs では f_type の類がとれないからですね。[ruby-core:62150]
> そして、その理由があったとしても、API としては statvfs という名前にして、
> ...
うーん... -
- * parse.y (next_id): use rb_str_ellipsize to preserve encoding and
get rid of incomplete multibyte sequence.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45761 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
10:58 AM Revision ca319081 (git): * gc.c (gc_after_sweep): suppress unnecessary expanding heap.
- Tomb heap pages are freed pages here, so expanding heap is
not required.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45760 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
10:34 AM Bug #9759: [TracePoint API] return event missing when raising exception
- Thanks!! So happy that this got fixed!! I hope it'll be added to 2.1 too.
-
- It can be reproduced by the following code:
```
class C9759
define_method(:foo){
raise
}
end
events = []
obj = C9759.new
TracePoint.new(:call, :return){|tp|
events << [tp.event, tp.method_id]
... -
- Applied in changeset r45758.
----------
* vm.c (invoke_block_from_c): add VM_FRAME_FLAG_BMETHOD to record
it is bmethod frame.
* vm.c (vm_exec): invoke RUBY_EVENT_RETURN event if rollbacked frame
is VM_FRAME_FLAG_BMETHOD.
[Bug #97... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45759 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- it is bmethod frame.
* vm.c (vm_exec): invoke RUBY_EVENT_RETURN event if rollbacked frame
is VM_FRAME_FLAG_BMETHOD.
[Bug #9759]
* test/ruby/test_settracefunc.rb: add a test for TracePoint/set_trace_func.
* vm_core.h: renmae rb_thread... - 08:06 AM Revision 9873f990 (git): * 2014-04-30
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45757 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * parse.y (rb_id_attrset): pin down dynamic symbol only. it is
possibe that attrset ID can be registered as a static symbol
after the corresponding attrget ID has been registered as a
dynamic, and then the latter may be collected.
... -
07:50 AM Bug #9560: Backport fix for: "SimpleDelegator, a wrapped object with an open method and send" from 2.1 to 2.0
- oops, it's reverted on ruby_2_0_0 because of incompatibility...
-
- backported r43682 (test), r43727 (test), r43752 and r43759 into ruby_2_0_0 at r45747.
# note: also implement assert_raise_with_message for these tests. - 07:45 AM Revision 03da71a9 (git): revert r45747 because it introduced severe incompatibility.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_0_0@45755 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- backported into ruby_2_0_0 at r45754.
- 07:39 AM Revision da20411b (git): merge revision(s) 45518: [Backport #9702]
- * ext/readline/extconf.rb (rl_hook_func_t): check pointer type.
[ruby-dev:48089] [Bug #9702]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_0_0@45754 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- backported into ruby_2_0_0 at r45753.
- 07:37 AM Revision e6c3b9cd (git): merge revision(s) 45471: [Backport #9688]
- * win32/win32.c (rb_w32_accept, open_ifs_socket, socketpair_internal):
reset inherit flag of socket to avoid unintentional inheritance of
socket. note that the return value of SetHandleInformation() is not
... -
- backported r45364, r45395, r45396, r45403 and r45406 into ruby_2_0_0 at r45752.
- 07:34 AM Revision fdec8bb0 (git): merge revision(s) 45364,45395,45396,45403,45406: [Backport #9652]
- * time.c (time_mload): freeze and preserve marshal-loaded time zone
* test/ruby/test_time.rb: add test for GC on loaded object
[Bug #9652]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_0_0@45752 b2dd03c... -
- backported into ruby_2_0_0 at r45751.
- 07:29 AM Revision 29fdecfe (git): merge revision(s) 45365: [Backport #9656]
- dl/extconf.rb: check for -fno-defer-pop option
* ext/dl/extconf.rb: check for -fno-defer-pop option, since clang
5.1 no longer support -fno-defer-pop option.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_0_0@457... -
- backported r45367, r45387, r45388 and r45389 into ruby_2_0_0 at r45750.
- 07:26 AM Revision 0b9f0164 (git): merge revision(s) 45367,45387,45388,45389: [Backport #9475]
- * vm_method.c (rb_method_entry_get_without_cache): me->klass is 0
for a method aliased in a module. [ruby-core:61636] [Bug #9663]
* vm_method.c (rb_method_entry_get_without_cache): get rid of
infinite recursion at aliases in a subc... - 07:18 AM Revision 16a10e0a (git): missed at previous commit.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_0_0@45749 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- backported into ruby_2_0_0 at r45748.
- 07:09 AM Revision 13ed57b4 (git): merge revision(s) 45320,45321: [Backport #9622]
- * vm_insnhelper.c (vm_callee_setup_arg): turn a macro into an
inline function.
* vm_insnhelper.c (vm_callee_setup_arg): disable fastpath if splat
argument, since argc may differ for each calls.
[rub... -
- memo: related commits: 45046,45047,45063,45087,45150,45151,45152
- 06:32 AM Revision 20d4a375 (git): merge revision(s) 43682,43727,43752,43759: [Backport #9560]
- * lib/delegate.rb (Delegator#send): override to get rid of global function interference.
[Fixes GH-449]
* lib/delegate.rb (Delegator#send): separate from method_missing so
that super calls proper method.
... - 06:04 AM Revision a2542330 (git): merge revision(s) 45488: [Backport #9631]
- * ext/readline/extconf.rb: fix typo, `$defs` not `$DEFS`.
[ruby-core:61756] [Bug #9578]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_0_0@45746 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
-
02:26 AM Bug #9763: Windows上のirbで「金」と入力できない
- This issue is not a bug of ruby or irb but a bug of rb-readline library.
I fixed this bug in https://github.com/phasis68/rb-readline.
You can try with https://github.com/phasis68/rb-readline/archive/master.zip
If you have issues a... -
- MVM branch has incomplete per-thread cwd, some methods are not implemented however, e.g., `File#rename`.
-
- Refinements does affects to the current scope only.
And `Symbol#to_proc` is independent from the callers' context.
A block works as you expect:
~~~ruby
puts a.map {|x| x.add_z}
~~~
04/29/2014
-
- I haven't been able to reduce the reproduction code yet, but r45698 introduced a segv. If you need any other info than what I've provided below, please let me know!
Here is the backtrace from the debugger:
~~~
(lldb) bt
* thread... -
09:26 PM Bug #9592: Fix segfault with old OpenSSL
- Nobuyoshi Nakada wrote:
> I suspect it occurs only with very old version OpenSSL, I can't reproduce it on other platforms at least.
Yes, Heroku is running OpenSSL 0.9.8k 25 Mar 2009 on their Cedar stack.
(heroku run openssl version)... -
- I suspect it occurs only with very old version OpenSSL, I can't reproduce it on other platforms at least.
It is planned to backport to 2.1, and the next 2.1 will be 2.1.2.
-
07:10 PM Feature #9777: Feature Proposal: Proc#to_lambda
- I find that using `throw` is useful in the situation you describe. It gives a nice explicit message about what's going: it's like a multi-level return to the point where you want to `catch` it.
-
07:02 PM Feature #9777: Feature Proposal: Proc#to_lambda
- Andrew Vit wrote:
> Would it work to just wrap it inside a lambda to get the semantics you want?
Like `lambda &proc`? That would be fine. As Nobu mentioned Proc and lambda behave differently, sometimes I want control over the behavio... -
- Would it work to just wrap it inside a lambda to get the semantics you want?
-
- `return` in a proc exits the defined method, unless it has returned already.
Turning a proc into a lambda disappoints that expectation.
It doesn't feel a good idea. -
06:57 PM Feature #9786 (Rejected): refinement & map(&:refinement_method) does not work
- Greetings!
Thank you for your hard work on ruby, I love the language.
I encountered an oddity, in which, I expected a refinement to be available and it wasn't.
Here is the sample code using: ruby 2.1.1p76 (2014-02-24 revision 45... -
04:49 PM Feature #9785: Feature Proposal: Dir.chdir Thread Safety
- richard.schneeman@gmail.com wrote:
> It's come to my attention that this is fairly hardcoded into the OS
> (changing CWD is a per-process operation rather than a per-thread
> one). I do not have a proposed implementation for how to c... -
03:38 PM Feature #9785: Feature Proposal: Dir.chdir Thread Safety
- If forking is an option for you, it would allow you to use chdir blocks the way you want I think.
-
- It's come to my attention that this is fairly hardcoded into the OS (changing CWD is a per-process operation rather than a per-thread one). I do not have a proposed implementation for how to change directory within a thread, perhaps we c...
-
- I am proposing that `Dir.chdir` with a block be local to the current thread and any threads that are created inside of that block. `FileUtils.cd` and `FileUtils.chdir` should also behave the same way.
Currently `Dir.chdir` will chang... -
- Bump, @ko1?
-
- I still failed to build trunk.
~~~
C:\work\ruby-2.2.0-r45738>nmake
...
generating psych-x64-mswin64_120.def
compiling psych.c
psych.c
compiling psych_emitter.c
psych_emitter.c
compiling psych_parser.c
psych_parser.c
compilin... -
08:42 AM Revision b6d7233d (git): fix style of ChangeLog
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45745 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
07:29 AM Feature #9783: [PATCH] Add Method#curry
- Updated the patch with documentation. There's an interesting side effect to be considered. `Method#to_proc` returns a lambda that checks its arity. `Proc#curry` takes an optional `arity` argument. If this doesn't match the method's arity...
-
- There is already Proc#curry, but to curry a method you need to go through to_proc. This patch adds `Method#curry` which delegates to `method.to_proc.curry`.
Looking forward to seeing this discussed. -
- To make Pathname and URI more consistent, it would be great to have a `join` instance method that behaves like `Pathname#join`. This could be a simple alias of `URI#merge`.
-
05:28 AM Bug #9782: Array#bsearch skip first match while using block with complex condition
- .
Understood. Thanks for explanation. -
05:14 AM Bug #9782 (Rejected): Array#bsearch skip first match while using block with complex condition
- Array#bsearch is binary search; it works correctly *only when* the array is sorted.
In other words, "ary.map { cond }" must return [false, false, ..., false, true, true, ..., true] if you want to use "ary.bsearch { cond }".
> [{a:1, ... -
- More accurate description: Array#bsearch cannot find first match while using block with equality condition:
Example:
$ irb
2.1.1 :001 > [1, 2].bsearch{|x| x == 1}
=> nil
(WRONG, expected 1)
2.1.1 :002 > [1, 2].bsearch{|x| x > 0... -
- Description: Array#bsearch cannot find first match when using block with condition by hash key.
Example:
$ irb
2.1.1 :001 > [{a:1, b:2}, {a:3, b:4}].bsearch{|x| x[:a] == 1}
=> nil
(WRONG, expected {a:1, b:2})
2.1.1 :002 > [{a:3... -
05:19 AM Revision 07f25637 (git): * (lib/tmpdir.rb): Rescue LoadError on etc.so for miniruby.
- Revert r45707, r45711, r45717.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45744 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
04:24 AM Feature #9781: Feature Proposal: Method#super_method - I would also find this feature very useful when debugging large code bases with complicated class/module hierarchies.
-
- A patch.
No tests yet.
~~~diff
diff --git a/proc.c b/proc.c
index 8153cc9..d1db478 100644
--- a/proc.c
+++ b/proc.c
@@ -1481,11 +1481,17 @@ method_owner(VALUE obj)
return defined_class;
}
-void
-rb_method_name_error... -
- rescue LoadError on 'etc'.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45743 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
01:35 AM Revision cce5fe1a (git): * test/ruby/test_symbol.rb (TestSymbol#test_symbol_gc_1): fix typo.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45742 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45741 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
04/28/2014
-
- I also have this problem especially when debugging code where modules are mixed in at runtime, so I have to do something like this:
~~~ruby
class << obj; self; end.ancestors.find_all { |klass|
klass.method_defined? :foo
}.map { |... -
When `super` is called in a method the Ruby VM knows how to find the next ancestor that has that method and call it. It is difficult to do this manually, so I propose we expose this information in Method#super_location.
Ruby Method ...-
- We saw this error in production with Ruby 2.1.1p76 on Heroku, but I don't know how to reproduce it. Is a backport to 2.1.1 planned? Or including the fix in 2.1.2?
```
Apr 22 08:03:01 app/worker.1: /app/vendor/ruby-2.1.1/lib/ruby/2.1... -
- * file.c (HAVE_STRUCT_STATFS_T_F_FSTYPENAME): Add new macro for
statfs_t.
* file.c (HAVE_STRUCT_STATFS_T_F_TYPE): ditto.
* file.c (rb_io_statfs): check FSTATFS macro only instead of
HAVE_FSTATFS and HAVE_FSTATVFS.
* file.c (statfs_... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45739 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Sorry for the delayed response. I was not "watching" this issue by default.
> Why can't you use break and next?
I wanted this to make a public api based on anonymous functions: https://github.com/opro/opro#password-token-exchange (... -
- kcburge@gmail.com wrote:
> Ruby 1.9.3-p286 does not do this.
Sorry, I might have introduced this bug in 2.0
Can you try the following patch? Thanks.
http://80x24.org/0001-io.c-try-to-fix-9780.patch -
04:03 PM Bug #9780 (Closed): sysread blocks when reading some files (AIX)
- This hangs when running on AIX 7 build of Ruby 2.1.1:
test_sysread:
~~~ruby
filename = "/proc/#{Process.pid}/as"
File.open(filename, 'rb') do |f|
p f.sysread(4)
end
~~~
Ruby 1.9.3-p286 does not do this. If I simply chang... - 03:20 PM Revision 5aa94c6c (git): * 2014-04-29
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45738 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * bootstraptest/test_literal.rb: fix typo of "dynamic". [ci skip]
* regexp.rdoc: fix typo of "organized". [ci skip]
* lib/session.rb: fix typo of "recognized". [ci skip]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45737 b2dd03c... -
- * file.c (FSTATFS): fix typo and commit miss.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45736 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * file.c (rb_io_statfs, rb_statfs_init, statfs_inspect): fix
compilation errors on Windows.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45735 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- to support OpenBSD.
* file.c (statfs_type): use above macro to switch.
* file.c (statfs_inspect): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45734 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * configure.in: on NetBSD fstatfs is obsoleted.
* file.c: support NetBSD for File::Statfs.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45733 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 08:46 AM Revision 1782a16e (git): * 2014-04-28
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45732 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45731 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45730 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
07:39 AM Bug #9648: Dir.glob('../*') shows unexpected result in a symbolic linked directory on Windows
- Usaku NAKAMURA wrote:
> Nobuyoshi Nakada wrote:
> ...
I thought 1. is spec on Windows because old version ruby, `dir` command and `FindFirstFile` behavior is same.
I agree to change behavior unixen systems.
-
- Nobuyoshi Nakada wrote:
> In that case, what should it return?
> ...
First I think it should returns the directory that the symlink exists (maybe 1 ?),
but now I think it should be 3.
> > Skipping `READDIR` usage for this case so... -
- Hiroshi Shirosaki wrote:
> If current directory is C:\foo\bar (linked to V:\), `READDIR` of '.' (i.e. `FindFirstFile(".\\*", &d)`) globs linked directory (V:/), but the result doesn't have '..'.
> ...
In that case, what should it retur... -
- If current directory is C:\foo\bar (linked to V:\), `READDIR` of '.' (i.e. `FindFirstFile(".\\*", &d)`) globs linked directory (V:/), but the result doesn't have '..'.
So we cannot trace parent directory.
Skipping `READDIR` usage for... -
- Nobuyoshi Nakada wrote:
> Is there any files at `V:/`?
Yes, there are some files and directories at `V:/`.
If no files and directories exist at `V:/`, the result is same. -
07:28 AM Feature #9779: Add Module#descendents
- s/descendents/descendants/
-
- Now classes have links to their subclasses, so how about to add Module#descendents?
```ruby
class X
end
class Y < X
end
class Z < Y
end
p X.descendents #=> [X, Y, Z]
module A
end
module B
include A
end
mo... -
-
01:13 AM Feature #9778: Bring shortcut methods to Hash
- `Hash#+` rejected in #6225. Also see #7739 for `Hash#|`.
04/27/2014
-
10:48 PM Feature #9778 (Rejected): Bring shortcut methods to Hash
- My obsession with easy syntax annoys me when I see `Hash#merge` does not have a shortcut in the form of a mathematical operation (like `-` or `+`).
~~~ruby
h_one = { :foo => "bar", :this => "that", :all => "nothing" }
h_two = { :foo... -
03:34 PM Bug #9657 (Closed): Segfault in bigdecimal/util.rb Float#to_d
- I close the ticket for backporting r45015.
I've confirmed that this issue is solved on trunk, by following code.
~~~ruby
require 'bigdecimal'
GC.stress = true
(200..400).each do |i|
p BigDecimal.new(10 ** i) * BigDecimal.new(... - 10:45 AM Revision a3a6da5e (git): trivial change
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45729 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 10:44 AM Revision bb05bcde (git): * ext/date/date_strptime.c (date__strptime_internal): do not overwrite century.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45728 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 07:35 AM Revision 264b4150 (git): * 2014-04-27
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45727 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45726 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Why can't you use `break` and `next`?
-
- Richard Schneeman wrote:
> Currently different block objects such as a lambda can be converted into to a proc: http://www.ruby-doc.org/core-1.9.3/Proc.html#method-i-to_proc
It changes nothing, but returns `self` as-is, without any ef...
04/26/2014
-
- Currently different block objects such as a lambda can be converted into to a proc: http://www.ruby-doc.org/core-1.9.3/Proc.html#method-i-to_proc
However you cannot turn a Proc instance into a lambda. Since a Proc and lambda behave di... -
- Takahiro Kambe wrote:
> 「NetBSD はどうなのかなぁ。」とは、どのような意味でしょう??
NetBSD で動かないのはよろしくないんじゃないか、という意味です。
-
- In message <redmine.journal-46305.20140424231120.f745bb8a50ac0750@ruby-lang.org>
on Thu, 24 Apr 2014 23:11:20 +0000,
akr@fsij.org wrote:
> 214 Others have statvfs.f_fstypename[_VFS_NAMELEN] (NetBSD 3.0).
> 215 Others have ... -
08:36 AM Feature #9711: Remove test-unit and minitest from stdlib.
- I update my patch. this patch copy test-unit and minitest to test/lib and separate to dependecy from stdlib.
Could someone review this patch? -
- Hi,
In message "[ruby-core:62170] [ruby-trunk - Bug #9648] Dir.glob('../*') shows unexpected result in a symbolic linked directory on Windows"
on Apr.26,2014 16:22:34, <nobu@ruby-lang.org> wrote:
> 'ok' means it globs as if it ... -
- Usaku NAKAMURA wrote:
> I've tested on 1.9.3-p545, 2.0.0-p451, 2.1.1 and found they are ok.
‘OK’ means it globs as if it is a real directory but not a symlink?
> ...
It doesn’t glob anything, neither the parent of the symlink, or ... -
- I've tested on 1.9.3-p545, 2.0.0-p451, 2.1.1 and found they are ok.
This problem is only with trunk. -
- Is there any files at `V:/`?
-
- Contribute from @igaiga. [fix GH-598]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45725 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r45724.
----------
compile.c: non-destructive keyword splat
* compile.c (compile_array_): make copy a first hash not to modify
the argument itself. keyword splat should be non-destructive.
[ruby-core:62161] [B... -
- * compile.c (compile_array_): make copy a first hash not to modify
the argument itself. keyword splat should be non-destructive.
[ruby-core:62161] [Bug #9776]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45724 b2dd03c8-39d4-4d... -
- It seems this issue has diverged so far from the original.
I reject it for now. -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45723 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
12:07 AM Bug #9757: Segfault in 2.1.1 from RSpec expect_any_instance_of
- Sorry I missed the notification about your message. I've retried this with trunk (ruby 2.2.0dev (2014-04-25 trunk 45720) [x86_64-darwin13]) and it just gives a SystemStackError, with message "stack level too deep" now, so it does seem l...
04/25/2014
- 11:07 PM Revision bb565a20 (git): * 2014-04-26
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45722 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- RLIMIT_NOFILE too small.
This fix sporadic "[ASYNC BUG] thread_timer: select" on GNU/Linux.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45721 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
08:30 PM Bug #9776: Ruby double-splat operator unexpectedly modifies hash
- Origin:
http://stackoverflow.com/questions/23282342/ruby-double-splat-operator-destructive-vs-nondestructive -
- I noticed what I find to be a very surprising behavior with the double-splat (`**`) operator in Ruby 2.1.1.
When key-value pairs are used before a `**hash`, the hash remains unmodified. However, when key-value pairs are only used afte... -
06:40 PM Feature #9772: IO#statfs and File::Statfs
- 2014-04-24 19:11 GMT-04:00 <akr@fsij.org>:
> Issue #9772 has been updated by Akira Tanaka.
>
>
> こういうときは先人の知恵ということで探してみると coreutils の src/stat.h になにかそれっぽいコメントがありました。
>
> ```
> 212 /* Return the type of the specified file... -
03:46 PM Feature #9704: Refinements as files instead of modules
- Ah, attack the problem from the other way round. That's a great idea!
-
- Monkey-patching style has another problem that super cannot be supported in monkey patching.
I prefer another approach to extend classes both globally and locally in the same manner.
How about to provide a way to activate refinements... -
- Applied in changeset r45720.
----------
* lib/fileutils.rb (rmdir): rescue Errno::EEXIST in addition to
ENOTEMPTY (and ENOENT), because SUSv3 describes that "If the
directory is not an empty directory, rmdir() shall fail and set
e... -
- SUSv3 に以下の記述があり、Solaris以外でもエラーになるOSがあるかもしれません。
> If the directory is not an empty directory, rmdir() shall fail and set errno to [EEXIST] or [ENOTEMPTY].
http://pubs.opengroup.org/onlinepubs/000095399/functions/rmdir.html
-
- ENOTEMPTY (and ENOENT), because SUSv3 describes that "If the
directory is not an empty directory, rmdir() shall fail and set
errno to [EEXIST] or [ENOTEMPTY]" and Solaris uses EEXIST.
[Bug #9571] [ruby-dev:48017]
git-svn-id: svn+... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45719 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
08:23 AM Bug #9775 (Closed): Segmentation Fault
- Segmentation Fault while running RoR 4 in debugging mode with 'thin' server.
See attachment for backtrace. -
- * node.c (dump_node): adjust indent to "ruby-style.el".
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45718 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * lib/mkmf.rb (try_link0): workaround for miniruby as 'tmpdir.rb'
always requires 'etc.so' now.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45717 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Tsuyoshi Sawada wrote:
> Seeing that these custom methods are used only in the context of a certain module/class, I request for a way to define a method (`foo`) on a module/class (`A`) so that it will be visible only from within a speci... -
- It differs from #7842, since you overwrites `C#foo` itself.
`C#orig_foo` is an alias of `M#foo`, but not of the former `C#foo`. -
- Applied in changeset r45716.
----------
`configure.in`: refine `__builtin_unreachable` check
* `configure.in` (`rb_cv_func___builtin_unreachable`): try with an
external variable not only by a warning, which might not be
show... -
- With MacPorts apple-gcc42, `__builtin_unreachable` seems detected as `no`, successfully.
-
- * configure.in (rb_cv_func___builtin_unreachable): try with an
external variable not only by a warning, which might not be
shown due to the optimization. [ruby-core:61647] [Bug #9665]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/tru... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45715 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Applied in changeset r45713.
----------
* configure.in: correct pthread_setname_np's prototype on NetBSD.
[Bug #9586] - 04:06 AM Revision 1944f1d6 (git): * 2014-04-25
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45714 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- [Bug #9586]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45713 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
04/24/2014
-
- こういうときは先人の知恵ということで探してみると coreutils の src/stat.h になにかそれっぽいコメントがありました。
```
212 /* Return the type of the specified file system.
213 Some systems have statfvs.f_basetype[FSTYPSZ] (AIX, HP-UX, and Solaris).
214 Others have sta... -
- POSIX的には、statvfsで取れるのは以下のメンバで他はoptional。つまり一番欲しいfs typeが
取れないわけです。
なので、statvfsはmisleadingだと思います。
それよりもstatfsで有益な情報がとれるOSは結局どのぐらいあるのか。が焦点ではないかと思います。
プラットフォーム依存なのは最初からわかっているんだし。
unsigned long f_bsize File system block size.
u... -
- ruby-core:62137 で Eric Wong も尋ねていますが、なぜ POSIX で定義されている statvfs じゃなくて、statfs なのか、というのは私も最初に疑問に思いました。
理由を参照できる形で残しておくのは重要だと思うので、説明を書いていただけないでしょうか。
そして、その理由があったとしても、API としては statvfs という名前にして、
statfs が存在する環境ではより多くの情報がなぜか取得できるとした方がいいので... -
- IO#statfs and File::Statfs を追加しませんか。
（テストで statfs.f_type が必要だったのでとりあえず追加してしまっていますが）
statfs(2) は Unix 系 OS でそこそこ普及している、あるパスが属するファイルシステムについての情報を得るためのシステムコールです。
OS によって多少差がありますが、だいたい以下の様な情報が得られます。
struct statfs {
uint32_t... -
06:25 PM Feature #9770: Etc.uname
- If you don't want to use WMI, you could get the same type of information using a combination of GetVersionEx and GetSystemInfo.
-
- Already available in the sys-uname gem if anyone is interested. On Windows I return much more information using WMI.
-
- As far as the current usages are guraded by platform test such as /linux/ =~ RUBY_PLATFORM,
it seems that emulation on Windows is not so important for non-Windows platforms.
If someone want to check Windows version, Etc.uname may be ... -
02:39 PM Bug #9774 (Closed): Net::HTTP failure to validate certificate
- I'm trying to make an https connection to a host that uses a self-signed certificate. I've downloaded the certificate into a directory of my project and named it based on the fingerprint of the certificate. Using the openssl command line...
- 02:19 PM Revision 3c00fe9f (git): * lib/fileutils.rb (fu_get_uid, fu_get_gid): Etc.getpwnam/getgrnam may
- returns nil.
* lib/webrick/utils.rb (su): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45712 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45711 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45710 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- usable from miniruby.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45709 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Two broken link `https://github.com/languages/Ruby` exist in `./man/ruby.1`.
I do not know what it was, and I can't replace newer URL. -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45708 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45707 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45706 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45705 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- assume those values won't be larger than LONG_LONG_MAX.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45704 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
06:54 AM Bug #7807: [PATCH] Remove duplicated load paths when empty version string is configured
- Recent version of the patch:
http://pkgs.fedoraproject.org/cgit/ruby.git/tree/ruby-2.1.0-Prevent-duplicated-paths-when-empty-version-string-i.patch -
- * parse.y (lookup_id_str): resurrect the dynamic symbol before
accessing its content, and its name before returning.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45703 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * parse.y (rb_sym2id, rb_sym2id_without_pindown): reject
non-symbol values.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45702 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r45701.
----------
ossl_asn1.c: fix undefined behavior
* ext/openssl/ossl_asn1.c (ossl_asn1_initialize): SYMID on a value
other than Symbol is an undefined behavior. fix up r31699.
[ruby-core:62142] [Bug #9771] -
- `ext/openssl/ossl_asn1.c:ossl_asn1_initialize()` allows `Qnil` as `tagging` since r31699.
But `SYM2ID()` is defined for T_SYMBOL values only.
-
- * ext/openssl/ossl_asn1.c (ossl_asn1_initialize): SYMID on a value
other than Symbol is an undefined behavior. fix up r31699.
[ruby-core:62142] [Bug #9771]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45701 b2dd03c8-39d4-4d8f-... -
- * parse.y (rb_sym2id, rb_sym2id_without_pindown): return 0 for
non-symbol values, for the time being.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45700 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45699 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * parse.y (rb_pin_dynamic_symbol): resurrect the dynamic symbol
which may be unmarked already and swept at next time, before
pindown.
* parse.y (dsym_node_gen): now rb_intern_str() resurrects dynamic
symbols, via lookup_sym_id() a... -
- * parse.y (rb_str_dynamic_intern): branch by same condition as
rb_id2sym to reduce unnecessary branch.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45697 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45696 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * file.c (statfs_inspect): add File::Statfs#inspect method.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45695 b2dd03c8-39d4-4d8f-98ff-823fe69b080e