Activity
From 03/06/2014 to 03/12/2014
03/12/2014
-
10:54 PM Bug #9628: Simple and isolated segfault in 2.1.0 and 2.1.1 on Mac OS X 10.9.2 (via rbenv)
- Example code and output for 2.0.0-p451, 2.1.0 and 2.1.1:
https://gist.github.com/tmornini/9518056
Code:
~~~
#!/usr/bin/env ruby
module Subledger
class Error < StandardError
def to_s
"#{message}, HTTP status #{... -
- Example code and output for 2.0.0-p451, 2.1.0 and 2.1.1:
https://gist.github.com/tmornini/9518056
Code:
#!/usr/bin/env ruby
module Subledger
class Error < StandardError
def to_s
"#{message}, HTTP status #{sta... -
08:10 PM Bug #9627 (Closed): SMTP does not properly dot-stuff an unterminated last line
- ```
str = "Date: Wed, 12 Mar 2014 16:02:26 -0400\r\nFrom: someone@aol.com\r\nTo: someone@aol.com\r\nMessage-ID: <5320bd52be4d3_6d653fd3d8c339e458df@Joshs-MacBook-Pro.local.mail>\r\nMime-Version: 1.0\r\nContent-Type: text/plain;\r\n char... -
06:17 PM Bug #9593: Keyword arguments default argument assignment behaviour not consistent with optional argument
- Hi,
Before r45272 this code worked:
~~~ruby
require 'minitest/autorun'
class DefaultArg < MiniTest::Unit::TestCase
class Default
def foo; 'foo'; end
def bar foo = foo
foo
end
end
def test_defa... -
04:59 PM Bug #9622: Extra method arguments are magically swallowed after first successful call, possible method cache bug
- just for memorandum. r45320 is also required to be backported.
-
02:19 AM Bug #9622 (Closed): Extra method arguments are magically swallowed after first successful call, possible method cache bug
- Applied in changeset r45321.
----------
vm_insnhelper.c: disable fastpath if splat
* vm_insnhelper.c (vm_callee_setup_arg): disable fastpath if splat
argument, since argc may differ for each calls.
[ruby-core:61422] [Bug #9622] -
03:49 PM Bug #9618: Pathname#cleanpath creates mixed path separators
- Thanks for both of your answers!
Indeed I think I met the fact it did not work in Explorer, but if it works with the standard API that is what matters.
I would definitely go with something like nobu's solution then (the conversion sh... -
03:14 PM Bug #9626 (Rejected): alias, prepend and super cause unexpected SystemStackError
- ```ruby
module M
def foo
super
end
end
class C
def foo
'C#foo'
end
prepend M
alias_method :orig_foo, :foo
def foo
orig_foo
end
end
C.new.foo
```
I expect: `M#foo` -> `C#foo` -> `C#ori... -
12:58 PM Bug #9605: Chaining "each_with_index.detect &lambda" raises ArgumentError
- Nobuyoshi Nakada wrote:
> It conflicts with existing behaviors too much.
I'm not surprised this caused other problems but I was hoping it wouldn't.
Is the current behavior now the expected behavior of Ruby and should I let other i... -
- It conflicts with existing behaviors too much.
-
10:58 AM Feature #9625: Define `Pathname#/` to append to the path
- Didn't this exist once and was later taken out? I have a dim (possibly
false) memory of this.
--
Avdi Grimm
http://avdi.org
On Mar 12, 2014 5:12 AM, <alexey.muranov@gmail.com> wrote:
> Issue #9625 has been reported by Alexey M... -
09:12 AM Feature #9625 (Closed): Define `Pathname#/` to append to the path
- How about borrowing this definition from *Homebrew* project:
class Pathname
def / that
join that.to_s
end
end
Looks nice to me. -
08:33 AM Bug #9624: Unrecognized command-line option (CLANG/Xcode 5.1) "-multiply_definedsuppress" prevents installation of gems / ruby on OS X
- Adding a link to a specific commit instead of trunk to prevent further commits from changing the line number: [configure.in:L2707](https://github.com/ruby/ruby/blob/ef0ce95daffe32be9d89f82cce0e4000dbc3fc02/configure.in#L2707)
-
- Currently there is a single unrecognized command-line option [configure.in:L2707](https://github.com/ruby/ruby/blob/trunk/configure.in#L2707) that should be removed in order for the ruby/gems native extensions to compile on OS X with XCo...
-
- * enum.c: revert r45284, r45286, r45292. it conflicts with
existing behaviors. [Bug #9605]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45322 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * vm_insnhelper.c (vm_callee_setup_arg): disable fastpath if splat
argument, since argc may differ for each calls.
[ruby-core:61422] [Bug #9622]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45321 b2dd03c8-39d4-4d8f-98ff-823fe69... -
- * vm_insnhelper.c (vm_callee_setup_arg): turn a macro into an
inline function.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45320 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
03/11/2014
-
- security fixではないですが、明らかなregressionなので1.9.3にも必要かもしれません。
-
11:37 PM Bug #9623 (Closed): warning: "named capture conflicts a local variable" does not allow loops with named captures
- Given the following example:
args = []
remaining = 'a,b,c'
begin
/(?<token>[^,]*),?(?<remaining>.*)/ =~ remaining
args << token
end until remaining.empty?
p args
Giv... -
10:28 PM Bug #9581 (Closed): `=~` defined on a subclass of `String` is sometimes ignored, and `String#=~` is called instead - Applied in changeset r45318.
----------
insns.def (opt_regexpmatch2): respect redefined match op
* insns.def (opt_regexpmatch2): respect redefined match op
Thanks to Sam Rawlins for the fix.
* test/ruby/test_string.rb: test based on ... - 10:28 PM Revision b3a613c8 (git): * 2014-03-12
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45319 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 10:28 PM Revision 4101a569 (git): insns.def (opt_regexpmatch2): respect redefined match op
- * insns.def (opt_regexpmatch2): respect redefined match op
Thanks to Sam Rawlins for the fix.
* test/ruby/test_string.rb: test based on Tsuyoshi Sawada's report
[Bug #9581]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45318 b2d... -
10:08 PM Bug #9622 (Closed): Extra method arguments are magically swallowed after first successful call, possible method cache bug
- The following code behaviour totally change since ruby 2.0.0
~~~
class Foo
def bar(*args)
attribute(:bar, *args)
end
def attribute(name)
puts name
end
end
foo = Foo.new
begin
foo.bar(100)
rescue... -
07:34 PM Feature #2567: Net::HTTP does not handle encoding correctly
- This still seems to be a problem in Ruby 2.1.1p76. Data ends up as ASCII-8BIT even if the server specifies that it's UTF-8 or ISO-8859-1, and I end up with errors when writing it out somewhere else as a result.
-
06:28 PM Bug #9618: Pathname#cleanpath creates mixed path separators
- I could not find anything official regarding UNC paths with forward slashes in the Win32 documentation for paths (http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247%28v=vs.85%29.aspx)
So I did a few tests: They work on ... -
- It works.
-
- There might be the special case of UNC paths, does anyone know if UNC paths with forward slashes work on Windows?
-
- I'm not sure if converting the separators in `#initialize` is the best way to go. I can see the advantages, but I can also imagine existing programs breaking because of that.
Personally I'd prefer a more conservative approach which wo... -
- Agree.
~~~diff
diff --git a/ext/pathname/pathname.c b/ext/pathname/pathname.c
index 3db97fc..a68ecd2 100644
--- a/ext/pathname/pathname.c
+++ b/ext/pathname/pathname.c
@@ -17,6 +17,12 @@ get_strpath(VALUE obj)
static void
set... - 04:52 PM Feature #9620: Rename "curry" to "schönfinkelize"
- +1 to mame's proposal
-
03:59 PM Feature #9620: Rename "curry" to "schönfinkelize"
- http://en.wikipedia.org/wiki/Moses_Sch%C3%B6nfinkel
> While Curry attributed the concept to Schönfinkel, it had already been used by Frege.[4]
#frege ...!
--
Yusuke Endoh <mame@tsg.ne.jp> - 02:11 PM Feature #9620: Rename "curry" to "schönfinkelize"
- The deep benefit of this improvement would be that in this way, the real first author of this operation, Moses Schönfinkel, would be commemorated, while "partial application" would gain a shortish name. Haskell Curry already has Haskell ...
- 01:59 PM Feature #9620 (Rejected): Rename "curry" to "schönfinkelize"
- I have just been able to understand the difference between currying and partial application. With this newly acquired knowledge, I would like to canvass in favor of making the symbol `:curry` incorrectly mean partial application all over...
- 02:00 PM Feature #9548: Module curry
- Having studied up the theory, I have added a feature request that would enable to misuse `curry` for partial application, #9620.
-
01:33 PM Revision c1f1b931 (git): * ext/.document: add objspace/objspace_dump.c to document file.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45317 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- remove dependency on json library.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45316 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
11:43 AM Bug #9617: Windows7 issue with Kernel.spawn close_others => false
- Lionel PERRIN wrote:
> * Is there a technical reason why it is not supported ?
> ...
We do not have a way how specify which handle corresponds to which file descriptor to a child process.
First of all, we do not know whether a child p... -
11:14 AM Bug #9617: Windows7 issue with Kernel.spawn close_others => false
- Thanks for the answer.
Two last questions:
* Is there a technical reason why it is not supported ?
I've looked at the ruby source code. From my understanding, the CreateProcess call is made with inheritHandles set to true, which... -
05:23 AM Bug #8860: 64-bit x86 ruby 1.9.3-p448 on Solaris core dumps on bug4950 test case
- The latest version is Ruby 1.9.3-p545. Please try the latest version.
https://www.ruby-lang.org/en/news/2014/02/24/ruby-1-9-3-p545-is-released/
Could you try latest version of SolarisStudio 12.3 compiler?
I only use Solaris 10 on ... -
- * test/ruby/test_hash.rb (test_exception_in_rehash): do not run twice
since this test does not use @cls. making a hash may increase
memory usage and should be in preparation.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45315 ... -
- * test/ruby/envutil.rb (Test::Test#assert_no_memory_leak): do not
check RSS by default, enabled by rss optional keyword argument.
TestHash#test_exception_in_rehash sometimes fails otherwise.
git-svn-id: svn+ssh://ci.ruby-lang.org/ru... - 04:22 AM Revision 5543a55b (git): * test: get rid of warnings.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45313 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Applied in changeset r45312.
----------
README.EXT{,.ja}: update ruby_options and ruby_run_node
* README.EXT{,.ja} (Appendix B): update contents of `ruby_options`
and replace `ruby_run` with `ruby_run_node`. based on the patch
by ... -
- * README.EXT{,.ja} (Appendix B): update contents of `ruby_options`
and replace `ruby_run` with `ruby_run_node`. based on the patch
by Kaneko Yuichiro at [ruby-dev:48030] [Bug #9619].
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trun... -
01:25 AM Revision 89dff271 (git): r45298 fix trivial bug
- $initial_size must be a size, not a status
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45311 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45310 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45309 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
03/10/2014
-
- It is not supported on mswin/mingw to pass file descriptors (except 0,1,2) to other processes.
So, it is expected behavior. -
- I faced an issue while trying to inherit a file or pipe handle from a master process to its child on windows 7.
I tried this implementation:
* master.rb:
~~~
rd, wr = IO.pipe
puts "Pipe opened: #{rd.fileno} <= #{wr.fileno}"
c... - 09:55 PM Revision c8c539a1 (git): ext/objspace: Check fptr before trying to dump FILE object fd
- Patch by Scott Francis. Closes GH-562.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45308 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 05:11 PM Revision d8182739 (git): * 2014-03-11
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45307 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * vm_dump.c (procstat_vm): copied from FreeBSD.
http://svnweb.freebsd.org/base/head/usr.bin/procstat/procstat_vm.c?revision=261780
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45306 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
04:37 PM Feature #9602: Logic with `Enumerable#grep`
- I've updated my patch with Array#select, nobu's `break` suggestion, and documentation and tests:
* GitHub: https://github.com/srawlins/ruby/compare/select-to-accept-args
* patch file: https://github.com/srawlins/ruby/compare/select-t... -
01:29 PM Bug #9619 (Closed): Fix README.EXT
- Update contents of `ruby_options` and replace `ruby_run` to `ruby_run_node`.
README.EXTの`ruby_options`を更新し、`ruby_run`を`ruby_run_node`に変えました。 -
- I agree on using / only in Pathname (converting in #initialize).
-
- When using `Pathname#cleanpath` with a Windows path the resulting path contains a mixture of slashes and backslashes.
~~~
require 'pathname'
path = Pathname.new('c:\projects\ruby\bug\test.rb')
path.to_s # => "c:\\proj... -
09:11 AM Bug #9562: Cannot install gems from gzip-compressing gem server
- Eric, I think the problem is that Gem::RemoteFetcher#fetch_path fails in `Gem.gunzip data` if the `data` is already decoded by Net::HTTP of Ruby 2.0 and later.
It happens when the HTTP server is configured to set a `Content-Encoding: gz... -
06:58 AM Feature #8426: Implement class hierarchy method caching
- Eric Wong <normalperson@yhbt.net> wrote:
> I may try adding caching in the main method table itself;
> especially if we end up using the container_of-style of method tables
Tried and unimpressive on bm_so_binary_trees so far:
http... -
04:36 AM Feature #9612: Gemify OpenSSL
- :+1: Great step.
-
- * addr2line.c (fill_lines): show the line number in C backtrace if
ruby is built without --enable-shared (PIE) on Linux.
patch is originally by Shinichiro Hamaji
https://twitter.com/shinh/status/441957774264504321
NOTE: ld doesn'... -
-
- 01:51 AM Revision 3727e066 (git): * 2014-03-10
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45304 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * test/ruby/test_enumerator.rb (test_iterators): fix test for hash
iterators. [Fixes GH-558]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45303 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
03/09/2014
-
01:15 PM Bug #9597: Core dump on Solaris on test case runner.rb
- I recompiled ruby against libffi.so.4 and the problem disappeared, so I confirm this bug is linked somehow to libffi.
-
- Hi,
I confirm that this bug is still present with ruby 1.9.3 patch level 484:
Output of ruby -v: ruby 1.9.3p484 (2013-11-22 revision 43786) [i386-solaris2.10]
I attached the core dump (which is 6 MB, so it is split in two files) ... -
- Add r45301
-
10:30 AM Feature #8426: Implement class hierarchy method caching
- parallel/continuation of this issue is in https://bugs.ruby-lang.org/issues/9262
-
- I noticed this was reverted in r43027 for being too slow.
Is there a plan to improve and reintroduce it?
I may try adding caching in the main method table itself;
especially if we end up using the container_of-style of method table... - 06:45 AM Feature #9548: Module curry
- Indeed, there is much less use for currying than for chaining in the functional world. (In #9552, I now prefer `#chain` to `#map!`.) Yet there is already `Proc#curry` in Ruby (whose present behavior is horrible, imo), therefore it is fai...
- 05:51 AM Feature #9552: Module map!
- I apologize for late response. Again, I experienced that the method I'm proposing here has much practical use in my code. As for better name, I can think of:
`#compose` --- weight: 3 -- as in functional composition
`#apply` --- ... - 05:22 AM Bug #9616 (Closed): [PATCH] class.c: check for malloc failures
- Applied in changeset r45302.
----------
class.c: check for malloc failures
* class.c (rb_class_subclass_add): use xmalloc
* class.c (rb_module_add_to_subclasses_list): ditto
* class.c (rb_class_remove_from_super_subclasses): use xfree
... -
- Will commit attached patch to trunk. Note backport field for 2.1
- 05:22 AM Revision b39eabac (git): class.c: check for malloc failures
- * class.c (rb_class_subclass_add): use xmalloc
* class.c (rb_module_add_to_subclasses_list): ditto
* class.c (rb_class_remove_from_super_subclasses): use xfree
* class.c (rb_class_remove_from_module_subclasses): ditto
[Bug #9616]
xmal... -
- * ext/fiddle/function.c (function_call): fix memory leak when an
exception occurs at argument conversion or the function call.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45301 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * ext/fiddle/function.c (initialize): check argument number if the
temporary buffer exceeds size_t max.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45300 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * test/dl/test_{cptr,handle}.rb: tests for [Bug #9599].
* test/fiddle/test_{handle,pointer}.rb: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45299 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * test/ruby/envutil.rb (Test::Test#assert_no_memory_leak): compare
also RSS if available.
* test/ruby/memory_status.rb (Memory::Status.parse): read string
form of an instance.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45298... -
04:09 AM Bug #9615 (Closed): fix typo in comments of eval.c
- Applied in changeset r45297.
----------
* eval.c: remove needless space.
[ruby-dev:48026] [Bug #9615] -
- Delete unnecessary space from comments of eval.c .
不要に思われるスペースを削除しました。 -
- [ruby-dev:48026] [Bug #9615]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45297 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Hi matz, I would like your permission to remove the order preservation from
any or all of the following currently implemented using `st_table`:
* method tables
* global symbols (`Symbol.all_symbols`)
* constant tables
* instance v...
03/08/2014
-
09:50 PM Misc #9215: Maintenance Policy for Future Releases (2.1.0 & beyond)
- @nurse: ping!
Could you review my comments? I'd like to move on from this :) - 09:44 PM Revision e9328e69 (git): * 2014-03-09
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45296 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 09:44 PM Revision eb227b25 (git): variable.c (struct global_variable): shrink by 8 bytes on 64-bit
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45295 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45294 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
10:06 AM Bug #8716: segmation fault 正規表現で大量のグループを利用時
- > https://github.com/k-takata/Onigmo/commit/b9fba1dc63ccb42a86e934011b468e6022fabb74
Rubyにはおそらく影響しないのですが、上記リンク先を見ていただければ分かるとおり、
マルチスレッド環境でこの変更がうまく動かない場合があるようなので、修正方法を再度検討中です。 - 08:56 AM Revision fa043eb4 (git): vm.c: cleanup to use rb_method_entry_at
- * vm.c (add_opt_method): cleanup to use rb_method_entry_at
Easier experiments with a non-st method entry tables.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45293 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * enum.c (each_val_i): svalue is no longer used.
[ruby-core:61340] [Bug #9605]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45292 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r45291.
----------
ext/dl, ext/fiddle: fix memory leak
* ext/dl/cptr.c (dlptr_free), ext/dl/handle.c (dlhandle_free),
ext/fiddle/handle.c (fiddle_handle_free),
ext/fiddle/pointer.c (fiddle_ptr_free): fix memory... -
- * ext/dl/cptr.c (dlptr_free), ext/dl/handle.c (dlhandle_free),
ext/fiddle/handle.c (fiddle_handle_free),
ext/fiddle/pointer.c (fiddle_ptr_free): fix memory leak.
based on the patch Heesob Park at [ruby-dev:48021] [Bug #9599].
git-... -
- Applied in changeset r45290.
----------
process.c: expand buffer on ERANGE
* process.c (obj2uid, obj2gid): now getpwnam_r() and getgrnam_r()
may need larger buffers than sysconf values, so retry with
expanding the buffer when ERANG... -
- * process.c (obj2uid, obj2gid): now getpwnam_r() and getgrnam_r()
may need larger buffers than sysconf values, so retry with
expanding the buffer when ERANGE is returned.
[ruby-core:61325] [Bug #9600]
git-svn-id: svn+ssh://ci.ruby... -
- * process.c (obj2uid, obj2gid): need capacity as buffer size, not
length.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45289 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * process.c (GETPW_R_SIZE_INIT, GETGR_R_SIZE_INIT): sysconf values
are not maximum sizes, but initial sizes.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45288 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * process.c (OBJ2UID1, OBJ2GID1): separate from OBJ2UID and
OBJ2GID respectively, need given buffers.
* process.c (OBJ2UID, OBJ2GID): no longer need PREPARE_GETPWNAM
and PREPARE_GETGRNAM.
* process.c (obj2uid, obj2gid): use tmp buf... -
- As of r45274, we now have sane whitelist of available OpenSSL ciphers. However, this patch breaks backwards compatibility for any apps that use any ciphers not whitelisted.
## Solution
* Implement a new class: OpenSSL::SSL::Ciphers... -
- Closing in favor of #9612
-
- Previously in #9439 we discussed removing OpenSSL from Ruby, however this wasn't the right approach.
I would like to suggest Aaron's approach from [ruby-core:60075](http://blade.nagaokaut.ac.jp/cgi-bin/scat.rb/ruby/ruby-core/60075)
...
03/07/2014
-
- * enum.c (find_i): yield multiple values instead of a packed
array, so that lambda block can work. with tests by Alex
Rothenberg. [ruby-core:61340] [Bug #9605]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45286 b2dd03c8-39d4... -
- Applied in changeset r45284.
----------
enum.c: yield multiple values
* enum.c (find_i): yield multiple values instead of a packed
array, so that lambda block can work. with tests by Alex
Rothenberg. [ruby-core:61340] [Bug #9605] -
- @nobu Wow, you fixed this issue so quickly. I wrote a few additional test cases that fail on trunk but pass on your branch in case you want to merge them in https://github.com/nobu/ruby/pull/1
Thank you! -
- There are other methods which have same behavior
I've thought it would have compatibility issues, but `make check` didn't fail at least.
WIP: https://github.com/nobu/ruby/compare/enum-yield_values?expand=1 - 10:35 PM Revision cfe3d989 (git): * 2014-03-08
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45285 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * enum.c (find_i): yield multiple values instead of a packed
array, so that lambda block can work. with tests by Alex
Rothenberg. [ruby-core:61340] [Bug #9605]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45284 b2dd03c8-39d4... -
09:20 PM Bug #9606: Ocassional SIGSEGV inTestException#test_machine_stackoverflow on OpenBSD
- It fails on x86 as well - #9198
-
04:18 AM Bug #9606: Ocassional SIGSEGV inTestException#test_machine_stackoverflow on OpenBSD
- Same result with the nightly snapshot.tar.gz (ruby -v: ruby 2.2.0dev (2014-03-07 trunk 45279) [x86_64-openbsd])
-
- Thank you, and could you try it with the latest trunk?
-
- Unfortunately, I lost the original core dump, so this core dump is slightly different, but the basic problem is the same:
(gdb) bt
#0 0x000014779593419a in kill () at <stdin>:2
#1 0x000014779599452a in abort () at /usr... -
- Jeremy Evans wrote:
> It appears that ruby's stack overflow handling is not working correctly in this case. Any pointers for how to fix this issue?
Yes, it is the test for machine stack overflow handling, as its name.
It very depen... -
08:00 PM Bug #9600: sysconf(_SC_GETGR_R_SIZE_MAX) is just a hint for getgrnam_r()
- I don't think there's a way to get it. You'll have to rescue and retry with a realloc'ed buffer.
One bit of trickiness is that different platforms seem to set errno to different values on different platforms. In my own sys-admin gem, ... -
- Yes, this is well-known (but often misunderstood) ruby's spec.
-
07:28 PM Bug #9611: Arithmetic operator following parentheses-less method gives ArgumentError
- Even though this also happens with +, * and /, I guess it is on purpose to pass negative numbers, not a bug. My mistake.
-
- It would be better to explain this with an example:
> 'a'.length -1
ArgumentError: wrong number of arguments (1 for 0)
... -
02:47 PM Bug #9610 (Rejected): Report by terminal error
- I use a rails application (rails 4.0.1) in mode development.
I start the server (rails s), and use the site.
When i kill the server (ctrl+c), this message appear in my terminal:
###################
[NOTE]
You may have encountere... -
11:10 AM Bug #9609: [PATCH] vm_eval.c: fix misplaced RB_GC_GUARDs
- (2014/03/07 20:03), SASADA Koichi wrote:
>> > Should we ignore WB unprotect for frozen objects?
> No relation between frozen and wb unprotected flag. We can use
> `RARRAY_CONST_PTR()` for this case. I'll commit it with other patch... -
- (2014/03/07 19:38), Eric Wong wrote:
> Thanks (r45283). Btw, I noticed rb_apply uses `RARRAY_PTR`, too,
> ...
No relation between frozen and wb unprotected flag. We can use
`RARRAY_CONST_PTR()` for this case. I'll commit it with ... -
- SASADA Koichi <ko1@atdot.net> wrote:
> And also it should be `RARRAY_CONST_PTR()`.
Thanks (r45283). Btw, I noticed rb_apply uses `RARRAY_PTR`, too,
but also with `OBJ_FREEZE`.
Should we ignore WB unprotect for frozen obj... -
- (2014/03/07 19:09), normalperson@yhbt.net wrote:
> `RB_GC_GUARD` needs to be placed after `RARRAY_PTR` usages to portably
> prevent compilers from optimizing the VALUE away.
And also it should be `RARRAY_CONST_PTR()`.
--
... -
- RB_GC_GUARD needs to be placed after RARRAY_PTR usages to portably
prevent compilers from optimizing the VALUE away.
-
- sam.rawlins@gmail.com wrote:
> Hi Eric, thanks for that test. It revealed my poor choice for `#define
> RUBY_EVENT_MCOVERAGE 0x040000` (`RUBY_INTERNAL_EVENT_SWITCH` is the
> same). I've updated my pull request with better choices in
... - 10:30 AM Revision 9c236f11 (git): vm_eval.c: use RARRAY_CONST_PTR to aid RGenGC performance
- Pointed out by ko1.
* vm_eval.c (vm_call0_body): use RARRAY_CONST_PTR
(check_funcall_exec): ditto
[ruby-core:61360]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45283 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 10:21 AM Revision bd1104f3 (git): vm_eval.c: fix misplaced GC guard
- * vm_eval.c (vm_call0_body): fix RB_GC_GUARD location
(check_funcall_exec): ditto
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45282 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * complex.c (parse_comp): replace ALLOCA_N with ALLOCV_N/ALLOCV_END
* rational.c (read_digits): ditto
This fixes segfaults on:
Rational("1" * 16 * 1024 * 1024)
Complex("1" * 16 * 1024 * 1024)
On my system where my stack size... -
- Closing, since this will be fixed in Kernel. Sorry for the noise.
http://patchwork.ozlabs.org/patch/326572/ -
09:33 AM Feature #9602: Logic with `Enumerable#grep`
- Sam Rawlins wrote:
> This idea should be extended to Enumerable#reject, and Array's #select and #reject, and probably Array's #select! and #reject!, and maybe other classes that extend Enumerable, and override #select and #reject (lik... -
- Hi Nobu, I was just using the same warning from enum_count, enum_find_index, and enum_inject. But it could `raise` if that is still desired. I'll fix with `break`.
-
- Sam Rawlins wrote:
> %w{foo bar baz}.select(/b/) {|e| e['f']} # warns "given block not used", returns ["bar", "baz"]
It should raise a "wrong number of arguments" exception, IMO.
> ...
It should `break` after once matched. -
- +1 to Matz's idea. I think this would work like `Enumerable#count`, taking `*args` _or_ a block, but not both:
%w{foo bar baz}.select #=> an Enumerator
%w{foo bar baz}.select {|e| e['b']} #=> an Array ["bar", "baz"]
%w... -
- Patch is added.
-
- Abstract
========
Generational GC (called RGenGC) was introduced from Ruby 2.1.0. RGenGC
reduces marking time dramatically (about x10 faster). However, RGenGC
introduce huge memory consumption. This problem has impact especially ... -
05:53 AM Revision 537ee14f (git): * parse.y (ENC_SINGLE): Unused macro removed.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45281 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
03:15 AM Revision 92a5ebb4 (git): * test/openssl/test_ssl.rb: Reuse TLS default options from
- OpenSSL::SSL::SSLContext::DEFAULT_PARAMS.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45280 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
03/06/2014
-
- ruby 2.1.1 on OpenBSD seems to occassionally suffer from a stack overflow when running TestException#test_machine_stackoverflow (about 1 every 3-4 times):
$ make test-all TESTOPTS="-q test/ruby/test_exception.rb"
Reading spec... -
- Naohisa Goto wrote:
> Please show compiler version, configure option and config.log, and libffi version.
Compiled was: gcc version 4.8.2 (GCC)
No configure option was given.
libffi is 3.0.9, soname is libffi.so.5.
There is ... -
- Please show compiler version, configure option and config.log, and libffi version.
-
- I found an odd edge case where "detect" and "select" behave differently from other methods of Enumerable.
Normally these methods yield a single argument to a block but when you chain them after "each_with_index" they yield two argumen... -
07:39 PM Bug #9604 (Closed): Comparing ranges or range includes range
- Hello guys,
First I got to say it's not like real bug - but will be good to have. Then in some case I I need compare to ranges on including each other: so I did like
(1..10).include? (2..7)
what kinda logical, but it always *... -
07:06 PM Bug #9603: unusual reference class-variable with cloned class.
- maybe...
<del>
first cloned-class write to original-class's class-variable; but new instance reference cloned-class's class-variable.
second cloned-class reference original-class that class-variable updated by first cloned-class.
</d... -
- description
-----------
Maybe panic reference to class-variable in cloned class. Not really sure about bug. But hang over my head.
I think minor irritant it which whether problem or not. because impractical code.
sample code
-----... -
- If the server does not return a gzip Content-Type for a .gz file it is misconfigured. This is not a bug in rubygems.
-
05:04 PM Bug #9525: Stuck with Socket.pack_sockaddr_in
- I applied following consecutive patches for ruby 2.1.1, and tried.
https://github.com/ruby/ruby/commit/948ce9decb97e5ff0833e53a392aa9f1d42c9b0d
https://github.com/ruby/ruby/commit/dd1c3a75096b97c1ebcb8597c001761ddfb3c1bf
https://gi... - 04:42 PM Revision 83299b67 (git): * 2014-03-07
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45279 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- RUN TESTS BEFORE COMMIT!!!
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45278 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
04:26 PM Bug #9582 (Rejected): CSV#headers not giving expected results.
- Yeah, I also looked at making this change, but the behavior of returning just the first line is documented:
https://github.com/ruby/ruby/blob/trunk/lib/csv.rb#L672
I'm not sure if users of this library count on that or not. This c... -
03:20 PM Feature #9602 (Feedback): Logic with `Enumerable#grep`
- I am afraid that proposed behavior is too far away from the original 'grep'.
We should use #select for more complex filtering.
Any opinion?
Matz.
-
- `Enumerable#grep` is useful to filter things:
[nil, {}, [], 1, :foo, "foo"].grep(String)
# => ["foo"]
1. Often, the condition cannot be expressed as a single object on which `===` is applied, but as a disjunction over `===... -
10:43 AM Bug #9600: sysconf(_SC_GETGR_R_SIZE_MAX) is just a hint for getgrnam_r()
- Off topic.
Yui NARUSE wrote:
>
> ...
There is no Single Unix Specification version 7 (yet).
The issue number is not the version number.
SUSv3:
| The Single UNIX Specification, Version 3 is made up of the Base Specifications, I... -
- Rei Odaira wrote:
> but actually this value is just a hint, so obj2gid() should gradually extend the buffer until getgrnam_r() no longer throws ERANGE.
SuSv7 changes it.
* http://pubs.opengroup.org/onlinepubs/009695399/functions/... -
- Isn't there any way to get the real number?
-
07:15 AM Bug #9600 (Closed): sysconf(_SC_GETGR_R_SIZE_MAX) is just a hint for getgrnam_r()
- When there is a group that has a lot of members, TestProcess#test_execopts_gid fails. Following is a more simple example:
$ ruby -e 'system("true", gid: "largegroup")'
-e:1:in `system': Numerical result out of range - getgrn... -
- Hi Eric, thanks for that test. It revealed my poor choice for `#define RUBY_EVENT_MCOVERAGE 0x040000` (`RUBY_INTERNAL_EVENT_SWITCH` is the same). I've updated my pull request with better choices in `include/ruby/ruby.h`: https://github.c...
-
07:14 AM Bug #9599: Fiddle::Function#call leaks memory
- Because Fiddle::Pointer class was alloced with TypedData_Make_Struct macro, it must be freed with ruby_xfree.
Same applies to ext/fiddle/handle.c, ext/dl/cptr.c and ext/dl/handle.c
Here is a patch.
diff --git a/pointer.c b/pointer.c... -
- `Fiddle::Function#call` seems leaking memory.
With the following code:
```ruby
# fiddle-memleak.rb
require 'fiddle'
n = 10
a = ["a"] * n
f = Fiddle::Function.new(Fiddle.dlopen(nil)["rb_obj_tainted"], [Fiddle::TYPE_VOIDP]*n, Fi... -
- By @idupree [Fixes GH-555] https://github.com/ruby/ruby/pull/555
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45277 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- By @idupree [Fixes GH-554] https://github.com/ruby/ruby/pull/554
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45276 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
04:15 AM Bug #9594: Segfault in 2.1.1 follows cont from debugger (Rails 4.0.3, debugger 1.6.6, rspec, os x mavericks 1.9.2)
- Nobuyoshi Nakada wrote:
> On MacOS X, a crash log file under `~/Library/Logs/DiagnosticReports` is important.
Attached here. -
- Kenta Murata wrote:
> Is this reproduced with byebug instead of debugger?
No, using byebug (2.7.0) in the same way *does not* reproduce the bug (thanks for pointing this gem out btw). -
- On MacOS X, a crash log file under `~/Library/Logs/DiagnosticReports` is important.
-
- The patch has been committed. After discussing the issue with Dirkjan, the decision was made to additionally add
ECDHE-ECDSA-RC4-SHA
ECDHE-RSA-RC4-SHA
RC4-SHA
to the end of the list because RC4 has been widely deploye... - 01:44 AM Bug #9424 (Closed): ruby 1.9 & 2.x has insecure SSL/TLS client defaults
- Applied in changeset r45274.
----------
* lib/openssl/ssl.rb: Explicitly whitelist the default
SSL/TLS ciphers. Forbid SSLv2 and SSLv3, disable
compression by default.
Reported by Jeff Hodges.
[ruby-core:59829] [Bug #9424] - 01:43 AM Revision 27a50dd1 (git): * 2014-03-06
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45275 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- SSL/TLS ciphers. Forbid SSLv2 and SSLv3, disable
compression by default.
Reported by Jeff Hodges.
[ruby-core:59829] [Bug #9424]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45274 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
12:18 AM Bug #9598 (Closed): Bug while installing a c-extensions gem
- I was trying to install a c-extension gem, the installation failed on first attempt. But succeeded on the second attempt.This was right after installing the ruby-2.1.1
(ruby 2.1.1p76 (2014-02-24 revision 45161) [x86_64-linux]) on Manjar...