Activity
From 06/20/2011 to 06/26/2011
06/26/2011
-
11:57 PM Bug #4445: ext/openssl の verify_callback が rb_protect で保護されていない
- Martin, can you handle this? Original reporter said that verify_cb does not use rb_protect to invoke a callback so an error raised from the callback passed directly to Ruby interpreter. Proposed patch looks good but we should check the r...
-
06:12 PM Bug #4445 (Assigned): ext/openssl の verify_callback が rb_protect で保護されていない
-
11:54 PM Feature #4168: WeakRef is unsafe to use in Ruby 1.9
- I'm particularly worried if current documentation of WeakRef states it is currently unreliable until this can be fixed.
I mean, 1.9.3 should change the documentation to reflect this issue so people know they shouldn't use it, while ke... -
04:44 PM Feature #4168 (Open): WeakRef is unsafe to use in Ruby 1.9
- I have to say this is an open problem. No one is responding comment # 43.
-
-
11:53 PM Feature #3793: No test for OpenSSL::PKCS12
- Hiroshi Nakamura wrote:
> Martin, thank you for your concern. I'll change 'Assignee' of ext/openssl tickets to you. Please send those ticket back to me as you like.
Ok, thanks! Once we're done with the tickets and reach the code free... -
- Martin, thank you for your concern. I'll change 'Assignee' of ext/openssl tickets to you. Please send those ticket back to me as you like.
For this ticket, Aaron added test_pkcs12.rb IIRC so you should be able to close it soon. -
- Hiroshi, I could take this if you'd like?
I'm basically through with 1.9.3 issues assigned to me, please let
me know how I can help further for the upcoming release!
Regards,
Martin - 11:01 PM Revision 8ce62003 (git): * ext/date/date_parse.c: should use ALLOCA_N.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32243 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- I agree to wait until this causes problems for anyone. The current solution seems fine for now.
-
-
10:17 PM Bug #4931 (Closed): debug.rbでvarコマンドを実行するとTypeErrorで落ちる
- =begin
辻本です。
debug.rbをロードしてvarコマンドを実行すると例外を吐いて落ちます。
$ ./ruby -Ilib -rdebug -e '0'
Debug.rb
Emacs support available.
-e:1:0
(rdb:1) var g
/home/k_tsj/work/ruby-trunk/lib/debug.rb:150:in `eval': can't convert Symbol in... -
08:23 PM Bug #4480: Thread-local variables issue: Thread#[] returns nil when called first time
- (2011/06/26 18:24), Yui NARUSE wrote:
> In ruby 1.9.2p180
> t = Thread.new do
> Thread.current[:var] = "var"
> Thread.stop
> end
>
> p t[:var] #=> nil
> p t[:var] #=> "var"
>
> In ruby 1.8.7
> p t[:va... -
-
08:06 PM Feature #4893: Literal Instantiation breaks Object Model
- Lazaridis Ilias wrote:
[...]
> I've to call "initialize" from "rb_str_new" instead from "str_new".
> ...
The v2 patch works on my site, tested with "nmake test" whilst adding a String running_counter (see file "bootstraptest_runner_ad... -
- Lazaridis Ilias wrote:
> John Higgins wrote:
> ...
Seems you were right, essentially it does not work (only in my small scope test)
> The error occurs when "rb_str_new_frozen" is used to create the string internally. Will look into ... -
- John Higgins wrote:
> The patch doesn't work....
The basic functionality works (see a simple test StringInit.rb), although there's work to do for a final version.
Remember that this is a draft version, mainly to demonstrate how it... -
12:56 PM Feature #4893: Literal Instantiation breaks Object Model
- Actually I stand slightly corrected - require/load seems to work on the empty file - but require_relative does not.
I apologize for the slightly misleading report. -
- The patch doesn't work....
The 'strings.rb' file I believe is about as simple as we can get. The 'empty_file' attempted to be required in the file is just that - a completely empty file with nothing in it at all.
ruby strings.rb yi... -
- OK, Ilias did his homework. It's your move matz. Though I don't like the call_initialize= thing (thread unsafe), the idea of calling initialize at string creation is clearly explained in the patch. I think it's worth considering.
-
- Yukihiro Matsumoto wrote:
> Your request has been too vague for me. Your definition of terms such as "object model" seems different from others. Probably the code will tell what you want. I will reopen this issue when a patch is post... -
- Deleted by myself, duplicate post.
-
07:54 PM Bug #4062: "gem uninstall" does not honor --program-suffix
- =begin
Following tests appear to test this feature.
* source:test/rubygems/test_gem_uninstaller.rb#L91
* source:test/rubygems/test_gem_commands_uninstall_command.rb#L59
But these files require 'rubygems/installer_test_case.rb' which ... -
- =begin
Reproducible with r32231.
Environment setup.
$ ruby-r32231 -v
ruby 1.9.3dev (2011-06-26 trunk 32231) [x86_64-darwin10.8.0]
$ mv ~/.gemrc ~/.gemrc.hidden
$ mv ~/.gem ~/.gem.hidden
Install a gem with suffix.
$ gem-r32231 ins... -
-
-
- Let's discuss after 1.9.3. Sorry.
-
-
- Net::SMTP has no maintainer.
-
- すみません，1.9.3 の後の課題とさせて下さい．
-
- Is it reproducible? If so, we'll reopen it.
-
- Do you have any information?
-
- I confirmed this on FreeBSD...
-
-
- =begin
$ sudo rm -rf /usr/local/lib/ruby
$ umask
022
$ sudo sh -c umask
0022
$ sudo make install-nodoc
./miniruby -I../lib -I. -I.ext/common ../tool/rbinstall.rb --make="make" --dest-dir="" --extout=".ext" --mflags="" --ma... -
-
- =begin
辻本です。
すみません、vm_clear_all_inline_method_cacheの動きについては勘違いしていました。
ちなみに、ポストしたパッチはキャッシュヒット率を向上させる面もあるので
別途取り込みを検討していただければと思います。
現在の実装では、以下のコードでのm呼び出しによる@ivへのアクセスはいずれもキャッシュミスしますが、
パッチを当てると初回のo0.m以外はキャッシュヒットするようになります。
cla... -
- ささだです．
なかなか見つけづらいバグを見つけて頂いてありがとうございます．
この問題は，vm_clear_all_inline_method_cache() が未実装だから，という
理由 *ではなく*，単に vm_getivar() が vm state を見ていないから，ではな
いかと思っています．そもそも，なんで見てないのか，記憶が確かではないんで
すが...．
vm_method_search() で行っている... -
- ささだです．
なかなか見つけづらいバグを見つけて頂いてありがとうございます．
この問題は，vm_clear_all_inline_method_cache() が未実装だから，という
理由 *ではなく*，単に vm_getivar() が vm state を見ていないから，ではな
いかと思っています．そもそも，なんで見てないのか，記憶が確かではないんで
すが...．
vm_method_search() で行っている... -
- =begin
辻本です。
自己レスです。
チケット登録してから気づいたのですが、これは本質的にはvm_clear_all_inline_method_cacheを
実装しないといけないという話でした。
vm.c=63=static void
vm.c:64:vm_clear_all_inline_method_cache(void)
vm.c-65-{
vm.c-66- /* TODO: Clear all inline cache... -
- =begin
辻本です。
--gc-stressオプションをつけてtest/ruby/enc/test_emoji.rbを実行するとテストに失敗します。
$ make RUBYOPT=-W TESTS='ruby/enc/test_emoji.rb --gc-stress' test-all
/home/k_tsj/work/ruby-trunk/test/ruby/enc/test_emoji.rb:154: warning: instance v... -
-
-
- Let's talk it after 1.9.3 release.
-
-
-
-
-
-
- Please close it.
-
-
- Behavior: It inherits current visibility (visibility of top-level is "private"). 1.8 also causes an exception. It seems to be a spec.
However, the following code:
```ruby
::String.send(:define_method, :x=){|v|}
s = ''
s.x = 10... -
-
-
- It seems mongrel's issue.
-
-
-
12:43 AM Bug #4651: Bus Error using continuation on x86_64-darwin11.0.0 (Lion)
- this seems to be fixed in trunk by https://github.com/ruby/ruby/commit/be953b4d1ce3f0dfc7f24c84ec7a51e027102557
(tested with the latest XCode Beta, Version 4.2 (4C114))
the commit applies cleanly on ruby_1_9_2 and also fixes `mak... -
-
-
-
-
-
-
-
-
- Is this still reproducible?
-
-
-
- Would you show a reproducible code?
-
-
-
- 1.9.0 is too old (and this is pre released one) and obsoleted.
Please reopen this if it still happens on the latest version of ruby. -
- It looks like mongrel's issue.
-
-
-
-
- I can't reproduce this with 1.9.2p188 on FreeBSD.
Can you reproduce with p180 (or latest ruby_1_9_2 branch)? -
- This issue was solved with changeset r32235.
Shyouhei, thank you for reporting this issue.
Your contribution to Ruby is greatly appreciated.
May Ruby be with you.
----------
* parse.y: comma at the end of line is no longer allowed.
A... -
- It seems ruby-debug's issue.
-
- It seems debian package's issue; please report this to ruby-deb/lucas.
-
-
-
-
-
-
-
- Matz, how do you think, 1.8's result is:
% ruby -ve 'p eval("raise", binding, "(eval)", 999)'
ruby 1.8.7 (2010-08-16 patchlevel 302) [amd64-freebsd8]
-e:1:in `eval': (eval):999: (RuntimeError)
from -e:1 -
- I close this as Rejected since no feedback provided.
Please reopen this if it still happens on the latest version of ruby. -
- I close this as Rejected since no feedback provided.
Please reopen this if it still happens on the latest version of ruby. -
- It seems ruby-debug's issue.
-
- Aaron, do you still gets a SEGV? Or do you want to someone take over this?
-
-
05:04 PM Bug #4223: GC.stress = true で謎の ArgumentError
- 私の手元の環境では変わってません。
$ cat foo.rb
# foo.rb
GC.stress = true
t = Time.local(2000)
File.utime(t + 1, t + 2, "foo.rb")
$ ruby19 -v foo.rb
ruby 1.9.3dev (2011-06-26 trunk 32231) [x86_64-freebsd8.2]
foo.rb:4:in `utime': time out of s... -
- 現在のtrunkで再現しませんがどうでしょう
-
- I close this as Rejected since no feedback provided.
Please reopen this if it still happens on the latest version of ruby. -
- ありがとうございます。修正済み扱いで閉じますので、再現したらご連絡をお願いします。
-
- trunk の r32231 で再現できませんでした
-
- もうこれ再現しないと思うんですがいかがでしょうが
-
-
- Would you show a small reproducible code?
-
-
- This seems resolved.
-
-
-
-
-
-
- I think current IO.popen should have enough doc.
> * [env, cmdname, arg1, ..., opts] : command name and zero or more arguments (no shell)
"no shell" means you must not quote command name and arguments. -
- It seems ruby-prof's issue.
-
-
-
- Contribution is welcomed.
-
- Spaceship operator doesn't have asociativity, so a <=> b <=> c is parse error.
-
- とりあえず現在FreeBSDではskipにしてます
-
- I discussed this with Shyouhei, Sasada, Tanaka and Naruse. And we don't have a consensus.
-
- Tanaka-san, please handle this.
-
04:34 PM Revision 61a34d62 (git): * test/etc/test_etc.rb (TestEtc#test_get{pw,gr}nam): skip entries
- start with + sign, which means NIS. these are returned in the
case that passwd and group entries in /etc/nsswitch.conf are set
to use "nis" explicitly on Debian. fixed #3683
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32242 ... -
- URI.encode is obsoleted.
Note that HTML5 seems specify the behavior of escaping and unescaping for addressbar.
Future URI module may have such methods. -
04:32 PM Bug #2460: RubySpecでFiberのSpecがおちる
- > 「単独では無理」というのはFiberに限った事でしょうか。
> ...
後者だと思います。
Fiber に限らず、Ruby プロセス終了時の挙動は実際に終了させなければテストでき
ないので。
この件に関しては、実行終了というよりは、「Fiber が途中状態のまま GC されたときに
ensure 節が実行される」という仕様を検査したいわけですが、確実に GC させるには
Ruby プロセスを終了させるくらいしか手がないですかね。
--
Yu... -
-
- Ruby cannot guarantee SystemStackError to be thrown.
-
- I close this as Rejected since no feedback provided.
Please reopen this if it still happens on the latest version of ruby. -
-
-
- I close this as Rejected since no feedback provided.
Please reopen this if it still happens on the latest version of ruby. -
- I close this as Rejected since no feedback provided.
Please reopen this if it still happens on the latest version of ruby. -
-
04:08 PM Bug #1066: Frequent failure: test_io.rb test 2 on OS X 10.5.6
- At least, My Mac OS X 10.6.7 can't reproduce this issue.
-
-
-
-
- (2011/06/26 10:43), Martin Bosslet wrote:
> Could you please verify that it works for you, too?
It works. Thanks!
--
// SASADA Koichi at atdot dot net -
- I added the wrapper solution in r32230 and tested it with OpenSSL 0.9.7m and 1.0.0d on Fedora 15 32bit.
Fortunately it turns out to be much less overhead than I initially thought.
Could you please verify that it works for you, too?
... -
10:32 AM Bug #4916 (Closed): [BUG] Segmentation fault - dyld: lazy symbol binding failed: Symbol not found: _ASN1_put_eoc - This issue was solved with changeset r32230.
Hiroshi, thank you for reporting this issue.
Your contribution to Ruby is greatly appreciated.
May Ruby be with you.
----------
* ext/openssl/extconf.rb
* ext/openssl/ossl_missing.h/.c: add ... -
- 1.9.3's YAML now dumps 9 digits of fractional (6 digits before).
So this seems resolved. -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32241 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- I close this as Rejected since no feedback provided.
Please reopen this if it still happens on the latest version of ruby. -
- I close this as Rejected since no feedback provided.
Please reopen this if it still happens on the latest version of ruby. -
- This looks rmagick's issue.
-
-
- 03:45 PM Revision 19ef07bb (git): * 2011-06-27
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32240 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- value. patched by Sho Hashimoto. [Bug #4511]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32239 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- without block returns enumerator.
patched by Michael Edgar [Bug #4847] [ruby-core:36800]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32238 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- We don't have loongson.
Contribution is welcomed. -
- Documentation contribution is welcomed.
-
- It seems ruby-debug-ide's issue.
Please reopen this if this is ruby's issue. -
- I close this as Rejected since no feedback provided.
Please reopen this if it still happens on the latest version of ruby. -
- We don't have Solaris.
Contribution is welcomed. -
- Contribution is welcomed.
-
- On 1.9.3, require gets faster than 1.9.2.
So I close this as duplicated. -
- I close this as Rejected since no feedback provided.
Please reopen this if it still happens on the latest version of ruby. -
- This issue is only happend w/ --gc-stress.
i.e. seems similar with Bug#4927.
-
- % time make test-all TESTS="-v -q --gc-stress ../test/dl/test_func.rb"
./miniruby -I../lib -I. -I.ext/common ../tool/runruby.rb --extout=.ext -- "../test/runner.rb" --ruby="./miniruby -I../lib -I. -I.ext/common ../tool/runruby.rb ... -
-
-
- Contribution is welcomed.
-
-
- timeout
-
-
-
-
- Unfortunately this patch seems to cause SEGV on os x. [#4927]
I cannot fix it because I have no os x. So I can't help but
revert it.
Anyone who wants this feature, please fix it by yourself or
please give me a Mac :-)
--
Yusu... -
-
- timeout
-
- 決定の問題だと思うので、決めてください
-
- This crash disappered when I revert r32071.
But instead, following test failure was shown.
----------------------------------------------
TestCoverage#test_restarting_coverage = 0.54 s = F
TestCoverage#test_result_with_nothing = 0.... -
- > Could you try this patch?
I have no luck. crashed again.
Thread 0 Crashed: Dispatch queue: com.apple.main-thread
0 libSystem.B.dylib 0x00007fff885225d6 __kill + 10
1 libSystem.B.dylib 0x00007fff8... -
- Hello.
2011/6/26 Motohiro KOSAKI <kosaki.motohiro@gmail.com>:
> test-all w/ --gc-stress got crash on may mac os x.
Could you try this patch?
diff --git a/ext/coverage/coverage.c b/ext/coverage/coverage.c
index 3a26aaa... -
- test-all w/ --gc-stress got crash on may mac os x.
% time make test-all TESTS="-v -q -j8 --gc-stress"
/Users/kosaki/ruby/test/coverage/test_coverage.rb:11: [BUG] Segmentation fault
ruby 1.9.3dev (2011-06-26 trunk 32229) [x86_64-... - 02:50 PM Revision a252c17e (git): * test/date/test_switch_hitter.rb: added a test.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32237 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
-
- Eric, how do we document $.?
-
-
-
- timeout
-
-
- What's happened on this, please explain.
-
-
- Graham, the patch is still available?
If so, I'll merge it. -
-
-
-
-
- This is intentional.
-
-
-
-
- //ignore is GNU libiconv's feature
- 01:23 PM Revision 51a0501a (git): * ext/date/date_core.c: refactoring.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32236 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- A patch from Yukihiro Matsumoto <matz AT ruby-lang.org>.
(fixed #3456).
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32235 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * ext/openssl/extconf.rb: Should check SSLv2_*method.
openssl compiled with "no-ssl2" the extconf don't fail
when running `make' having this compilation errors.
Patched by Laurent Arnoud. fixes #4562, #4556
... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8_7@32233 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8_7@32232 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Jeremy Evans wrote:
> =begin
> ...
The former encoding is that of @basic_constraints in test_x509ext.rb.
It is defined as
@basic_constraints_value = OpenSSL::ASN1::Sequence([
OpenSSL::ASN1::Boolean(true), # CA
OpenSSL::A... -
- on Mac. It should be placed after "-- C level backtrace" line.
Suggested by Endoh-san.
<before>
-- See Crash Report log file under ~/Library/Logs/CrashReporter or ---------
-- /Library/Logs/CrashReporter, for the more detail of ... -
- * ext/openssl/ossl_missing.h/.c: add ASN1_put_eoc if missing.
* ext/openssl/ossl_asn1.c: introduce ossl_asn1_object_size and
ossl_asn1_put_object to wrap functionality depending on OpenSSL
version in use.
Fixes [ Ruby 1.9 - Bug #4... -
12:05 AM Bug #4881 (Closed): super() in define_method which returns proc causes SEGV
- This issue was solved with changeset r32227.
Thomas, thank you for reporting this issue.
Your contribution to Ruby is greatly appreciated.
May Ruby be with you.
----------
* vm_insnhelper.c (vm_search_superclass): avoid control frame
...
06/25/2011
-
- This issue was solved with changeset r32226.
三村, thank you for reporting this issue.
Your contribution to Ruby is greatly appreciated.
May Ruby be with you.
----------
* thread.c (sleep_forever): now Kernel#sleep don't wakeup by
sign... -
10:23 AM Bug #4072: dRubyで作成したサーバプログラムがsleepしていてもexitしてしまう
- まつもと ゆきひろです
In message "Re: [ruby-dev:43895] [Ruby 1.9 - Bug #4072][Assigned] dRubyで作成したサーバプログラムがsleepしていてもexitしてしまう"
on Sat, 25 Jun 2011 09:20:27 +0900, Motohiro KOSAKI <kosaki.motohiro@gmail.com> writes:
|ただIRCで聞いたところに... -
- まつもと ゆきひろです
In message "Re: [ruby-dev:43895] [Ruby 1.9 - Bug #4072][Assigned] dRubyで作成したサーバプログラムがsleepしていてもexitしてしまう"
on Sat, 25 Jun 2011 09:20:27 +0900, Motohiro KOSAKI <kosaki.motohiro@gmail.com> writes:
|ただIRCで聞いたところに... -
- > 遠藤です。
>
> 2011年6月24日14:43 KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>:
>> ところで、話は変わるのですが thread.stopも process.sleep の「thread.run するまで
>> スリープする」と書いて有るのにシグナル受信したときの挙動が違うのはなぜなんでしょうね？
>> deadlock check作った遠藤さんに聞くべきなのかも知れません... -
- > 遠藤です。
>
> 2011年6月24日14:43 KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>:
>> ところで、話は変わるのですが thread.stopも process.sleep の「thread.run するまで
>> スリープする」と書いて有るのにシグナル受信したときの挙動が違うのはなぜなんでしょうね？
>> deadlock check作った遠藤さんに聞くべきなのかも知れません... -
- kosakiです
unblock functionはGVL持たずに呼ばれるので、そこがthread stateによって挙動を変えていたら「死ねばいいのに」レベルだと思っています。よってわたし的にはこのパッチはOK。
ただIRCで聞いたところによると mameさんの意識は
ruby -e 'trap(:INT) { }; sleep'
の時もsleepは解除されるという仕様だったそうです。ここの仕様はまつもとさんに確認
する必要があると思います。... -
- Hi,
Thank you for your reporting.
This issue seems similar to #2420 and #3136 .
According to discussion in #3136, fundamental fix needs tough work.
I've wrote quick fix patch for only avoid SEGV but raise an NoMethodError excepti... -
- Yukihiro Matsumoto wrote:
> Your request has been too vague for me. Your definition of terms such as "object model" seems different from others. Probably the code will tell what you want. I will reopen this issue when a patch is post... -
- fixed by r32221.
- 04:03 PM Revision 3380f2ff (git): * ext/date/date_core.c (date_strftime_internal): removed meaningless braces.
- * ext/date/date_core.c (gengo): the value should be int.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32229 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
03:51 PM Bug #3137: complex.rb changes exceptions of Math
- r32055 の修正だと、CMath.#log で底に数値以外を与える場合に Math.#log のように TypeError が発生しませんでした。
CMath.log(8, "2") # => ArgumentError - 03:05 PM Revision 64fd7d87 (git): * 2011-06-26
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32228 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- stack overrun. currently super() in Proc created in a method
defined by Module#define_method raise NoMethodError. [Bug #4881]
* test/ruby/test_method.rb t_super_in_proc_from_define_method):
add test for it.
git-svn-id: svn+ssh://ci.... -
- signal handler execution. [Bug #4072]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32226 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- backup. fix race condition which may results unexpected main thread's
status transition. see #4072
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32225 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
01:01 PM Bug #4925 (Rejected): Infinite recursion allowed in rescue clause
- The issue pertains to the following example:
def a
b
rescue NameError
a
end
a
The type of exception raised and rescued is not relevant; the bug occurs with any exception type.
Expected behavior: A SystemStackError shoul... - 09:20 AM Revision eae50817 (git): * 2011-06-25
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32224 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32223 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
06/24/2011
-
- > RUBY_VM_CHECK_INTS() を抜けた後で th->status が THREAD_RUNNABLE になっていることがあるようで
この原因は rb_threadptr_check_signal() でタイマースレッドがメインスレッドの status を変更しているためではないかと思います。
mth->status = THREAD_RUNNABLE にセットしてから rb_threadptr_interrupt() でメインスレッドへシグナル送信し... -
- 遠藤です。
2011年6月24日14:43 KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>:
> ところで、話は変わるのですが thread.stopも process.sleep の「thread.run するまで
> スリープする」と書いて有るのにシグナル受信したときの挙動が違うのはなぜなんでしょうね？
> deadlock check作った遠藤さんに聞くべきなのかも知れませんが。
... -
- 遠藤です。
2011年6月24日14:43 KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>:
> ところで、話は変わるのですが thread.stopも process.sleep の「thread.run するまで
> スリープする」と書いて有るのにシグナル受信したときの挙動が違うのはなぜなんでしょうね？
> deadlock check作った遠藤さんに聞くべきなのかも知れませんが。
... -
- すみません、調査不足でした。再度 open します。
kosaki さんのご指摘の通り引数なしの Kernel#sleep がシグナルによって起きてしまうという問題は残っていました。 trunk で以下のように空のシグナルハンドラを設定して sleep すると darwin10.7.0 では SIGINT で sleep から目覚めてしまいました。
Signal.trap(:INT){}
sleep
これは spurious wakeup 対策不足という... -
- 何度もすみません。SIG_DFL じゃなくて SIG_IGN ですね。
あと引数なしの sleep の場合は sleep_forever() で眠りますが、こちらは deadlockable 引数が 0 だと while の条件部が抜けるようになっているのでそのため一度起床するとそのまま抜けてしまってたようです。 -
- そうですね、EINTR でのリトライじゃなく SIGPIPE が SIG_DFL になったことの影響かもしれません。
-
- いま、Macのpthread_cond_wait()見てるんですが、
http://www.opensource.apple.com/source/Libc/Libc-594.9.4/pthreads/pthread_cond.c
/*
* Suspend waiting for a condition variable.
* Note: we have to keep a list of condition variables which ar... -
- I played around a bit and it seems like this could be some kind of race condition.
At first I thought that the custom Fedora OpenSSL would probably reject some
algorithm and so I wanted to see what cipher is negotiated in the tests and... -
- Strange. I tried with 0.9.8r, couldn't reproduce it. Then I tried with 1.0.0d, again the tests succeeded. It is only the customized version of 1.0.0d that ships with Fedora 15 that causes the error. I know that they disabled ECC support ...
-
- Hiroshi NAKAMURA wrote:
> Martin Bosslet wrote:
> ...
Infinite length support has been introduced with 1.9.3, in 1.9.2 decoding infinite length values already
worked, but they were re-encoded using a definite length value (cf. http://... -
- Martin Bosslet wrote:
> The missing ASN1_put_eoc is not the only problem with OpenSSL 0.9.7, unfortunately. ASN1_put_object and
> ...
Thanks for investigation. Hmm. I don't remember if there was such a case in ext/openssl development.
... -
- The missing ASN1_put_eoc is not the only problem with OpenSSL 0.9.7, unfortunately. ASN1_put_object and ASN1_object_size behave differently in >= 0.9.8 regarding the length parameter being passed to them when dealing with infinite length...
-
- I close it since I replied to the original reporter at ruby-core.
-
- Hi,
On Fri, Jun 24, 2011 at 02:13, Aaron Patterson <aaron@tenderlovemaking.com> wrote:
>> And the second part, can you remember some more details of SEGV you got? If the error you got is timeout or something, it could be caused by in... -
02:23 AM Bug #4237: SSL_shutdown should be called until it returns 0
- On Thu, Jun 23, 2011 at 02:27:41PM +0900, Hiroshi NAKAMURA wrote:
>
> Issue #4237 has been updated by Hiroshi NAKAMURA.
>
>
> Thanks for the explanation. I think r30294 (fix for lib/net/smtp.rb) is good. It solved the original p... -
12:09 PM Feature #4924 (Assigned): mkmf have_header fails with C++ headers
- =begin
When a user calls (({have_header('some_cpp_header.h')})), and then header includes a line such as(({ #include <string>})), mkmf will fail.
An example run follows:
+ extconf.rb
require 'mkmf'
have_library('stdc++')
... - 11:05 AM Revision 635d13a8 (git): * lib/webrick/httprequest.rb (setup_forwarded_info): Parsing request
- header failed when the request is from 2 or more Apache reverse
proxies. It's said that all X-Forwarded-* headers will contain more
than one (comma-separated) value if the original request already
contained one of these headers. S... -
- twice. fixed Bug#4920.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32221 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 07:06 AM Revision dd91beb3 (git): * Typo fixed. Is it reallyl a typo? 2*2 == 2+2 == 4.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32220 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 07:01 AM Revision 369b0950 (git): * ext/openssl/ossl_ssl.c (ossl_ssl_shutdown): Try to shutdown SSL
- connection more gracefully. Call SSL_shutdown() max 4 times until it
returns 1 (success). Bi-directional SSL close has several states but
SSL_shutdown() kicks only 1 transition per call. Max 4 is from
mod_ssl.c of Apache httpd that... -
- (2011/06/24 1:00), Benoit Daloze wrote:
> On 22 June 2011 19:55, Marc-Andre Lafortune <ruby-core@marc-andre.ca> wrote:
>> The method `to_ary` is for classes that can be implicitly converted to an Array. This doesn't apply to NilClas... -
01:23 AM Feature #4917: `NilClass#to_ary`
- On 22 June 2011 19:55, Marc-Andre Lafortune <ruby-core@marc-andre.ca> wrote:
> The method `to_ary` is for classes that can be implicitly converted to an Array. This doesn't apply to NilClass.
>
> I'm also highly sceptical as to th... -
12:25 AM Feature #4910: Classes as factories
- Charles Nutter wrote:
> I'm not sure I agree with adding `to_proc` to `Class` instances, since it seems questionable that `#new` is what you'd always want to be called.
Hmm, but what else? I think it is a reasonable default.
> ..... -
- とりあえずFeatureに移動しますかね
06/23/2011
-
- Hiroshi NAKAMURA wrote:
> Aaaaarhg! Sorry, 2**31 is not in a signed long value. Miserable me...
> ...
No problem, works fine now. Happens to the best of us ;)
> >>>>2) Anyone else getting the "sslv3 alert illegal parameter"? I g... -
- 2011/6/23 Martin BoÃlet <martin.bosslet@googlemail.com>:
>> I modified it a little at r32211.
>
> This gives me: (Ubuntu 10.04 32bit, OpenSSL 0.9.8k)
Aaaaarhg! Sorry, 2**31 is not in a signed long value. Miserable me...
Fixed at... -
- 2011/6/23 Martin BoÃlet <martin.bosslet@googlemail.com>:
>> I modified it a little at r32211.
>
> This gives me: (Ubuntu 10.04 32bit, OpenSSL 0.9.8k)
Aaaaarhg! Sorry, 2**31 is not in a signed long value. Miserable me...
Fixed at... -
- 2011/6/23 Hiroshi Nakamura <nakahiro@gmail.com>:
> 報告ありがとうございます。32bit環境で落ちるとか考えてませんでした。。。
> Thanks for the report. I was not aware that the code has 64bit dependency...
>
> On Wed, Jun 22, 2011 at 18:45, Tomoyuki Chikanaga <naga... -
- 2011/6/23 Hiroshi Nakamura <nakahiro@gmail.com>:
> 報告ありがとうございます。32bit環境で落ちるとか考えてませんでした。。。
> Thanks for the report. I was not aware that the code has 64bit dependency...
>
> On Wed, Jun 22, 2011 at 18:45, Tomoyuki Chikanaga <naga... -
- 報告ありがとうございます。32bit環境で落ちるとか考えてませんでした。。。
Thanks for the report. I was not aware that the code has 64bit dependency...
On Wed, Jun 22, 2011 at 18:45, Tomoyuki Chikanaga <nagachika00@gmail.com> wrote:
>>1) Is it OK if I apply the f... -
- 報告ありがとうございます。32bit環境で落ちるとか考えてませんでした。。。
Thanks for the report. I was not aware that the code has 64bit dependency...
On Wed, Jun 22, 2011 at 18:45, Tomoyuki Chikanaga <nagachika00@gmail.com> wrote:
>>1) Is it OK if I apply the f... -
- Hi,
I was finally able to reproduce this with Fedora 15 32 bit, OpenSSL 1.0.0d.
The error occurs only when running make test-all TESTS="openssl",
if I run test_ssl.rb independently the test succeeds.
Here is the output I get:
... -
- これですがおそらく r31482 のあたりで native_cond_wait() が EINTR でリトライするようになったので
直っているのではないかと思います。
手元では ruby 1.9.2dev (2010-05-10 revision 27709) で再現していたのが trunk では起きなくなっていました。
ruby_1_9_2 にもマージ済みなので 1.9.2-head でも同じではないかと思います(未確認ですが)。
というわけでcloseさ... -
- On Thu, Jun 23, 2011 at 20:54, Martin Bosslet <Martin.Bosslet@googlemail.com> wrote:
>> For that purpose, algorithm should be fixed so they don't get Digester
> ...
Ah, X509::Name#digest sounds good. I didn't know OCSP impl already us... -
- Hiroshi NAKAMURA wrote:
> They're using the hash of name for c_rehash. You see files something
> ...
OK, I see, thanks for the info! X509_NAME_hash(_old) has a special meaning
internally. I checked, there also exists X509_NAME_dige... -
10:24 PM Revision ea9c30be (git): * lib/rake/version.rb: Fixed VERSION to work with tool/rbinstall.rb
- * bin/rake: Import bin/rake from 0.9.2
* tool/rbinstall.rb (install): Rake::VERSION is now in
lib/rake/version.rb. Fixes `make install`
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32218 b2dd03c8-39d4-4d8f-98f... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32217 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Benoit Daloze wrote:
> A quick `ack -l --invert-file-match -G 'rdoc|rubygems' MiniTest test` gives:
> ...
Thanks for the analysis, Benoit! As I don't want to fiddle around in other people's domains anyway, I suppose it's fine to agr... -
- On Jun 22, 2011, at 5:41 AM, Benoit Daloze wrote:
> On 22 June 2011 07:46, Martin Bosslet <Martin.Bosslet@googlemail.com> wrote:
>> Some tests for OpenSSL inherited from Minitest::Unit::TestCase instead of
>> Test::Unit::TestCase. To... -
- Hiroshi NAKAMURA wrote:
> I agree with DH interface is confusing.
Adding to the confusion is that DH implements the PKey interface in OpenSSL (OpenSSL itself, not Ruby's ext/openssl), but it conceptually is not really like the other ... -
07:55 PM Bug #4579: SecureRandom + OpenSSL may repeat with fork
- I don't understand why /dev/urandom fallback should be removed.
Is your reason the "theoretically vulnerable to a cryptographic attack on the algorithms used by the driver" (from Linux man page)?
-
- Hi,
On Thu, Jun 23, 2011 at 08:15, Akira Tanaka <akr@fsij.org> wrote:
>> We should avoid using
>> /dev/urandom every time on the env w/o OpenSSL in the future.
>
> I'd like to say "Please install OpenSSL" for such request.
... -
- NaHi:
> We should avoid using
> ...
I'd like to say "Please install OpenSSL" for such request.
Cryptographic algorithms should be implemented/maintained by cryptographic experts but I am not a cryptographic expert.
-
- Hi,
2011/6/20 Jarosław Skrzypek <skrzypek.jarek@gmail.com>:
> You might be also interested in my very short patch: https://gist.github.com/1035322 Basically it reorders conditions to start with fastest ones and execute slower ones o... -
- Hi,
2011/6/23 Hiroshi Nakamura <nakahiro@gmail.com>:
> Anyone can imagine a downside of this? It could not work as expected
> if the result of rb_file_expand_path *changes* during require (adding
> a new file during require s... -
- Hi all,
On Wed, Jun 1, 2011 at 09:14, Urabe Shyouhei <shyouhei@ruby-lang.org> wrote:
>> This process creates many objects, i.e., strings. Â Typically,
>> $LOADED_FEATURES are already expanded, so the process is not
>> needed in no... -
- See also: 'Reverse Proxy Request Headers' in http://httpd.apache.org/docs/2.2/en/mod/mod_proxy.html
Be careful when using these headers on the origin server, since they will contain more than one (comma-separated)
value if the or... -
- (From [ruby-core:33872])
Hi.
This is my first post to this mailing list, so please tell me, if I posted this the wrong place, or I in any other way should do something different.
I have been playing around with Webrick for some ... -
- I guess that WEBrick::Server#{start,stop,shutdown} is not designed to be thread-safe and there's no simple fix for your usage. When I used WEBrick::Server for such purpose, I created my own lock outside of WEBrick::Server instances.
I... -
- 直接の原因はこのコミットのようです。
commit afbd5661a0ca041ac818c9e5732911c5db21c9f9
Author: nobu <nobu@b2dd03c8-39d4-4d8f-98ff-823fe69b080e>
Date: Tue Jul 13 12:31:17 2010 +0000
* process.c (rb_daemon): split from proc_daemon.
gi... -
- I take this since Fujioka-san must be busy for the meanwhile.
Can somebody point the new location of http://dev.rubyonrails.org/ticket/10886 ? I once posted a patch related to multipart form boundary parsing (http://blade.nagaokaut.ac... - 03:45 PM Revision 3fbc9440 (git): * ext/date/date_core.c (c_valid_{julian,gregorian}_p): fixed the range of month.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32216 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- The sentence "defines" essentially the terminology:
Yukihiro Matsumoto wrote:
> I agree providing a method to tell whether it is loaded as a library or
^^^^^^^^^
"is loaded"
load... - 03:15 PM Revision 838af29c (git): * 2011-06-24
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32215 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 03:15 PM Revision 2e83b823 (git): * ext/date/date_core.c: trivial changes on text.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32214 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Thanks for the explanation. I think r30294 (fix for lib/net/smtp.rb) is good. It solved the original problem, right? Of course, ext/openssl should try to shutdown the connection gracefully to the extent possible though.
And the second... -
- On Wed, Jun 22, 2011 at 11:08:08PM +0900, Hiroshi NAKAMURA wrote:
>
> Issue #4237 has been updated by Hiroshi NAKAMURA.
>
>
> Aaron, I'm sorry for late response, but can you point the original report of this issue? The link 'htt... -
- なるほど。
ファイルを削除してしまうと問題があるケースもありえなくはないので、
実際にはruby.hをincludeするだけのファイルにする等を考えています。 -
- intern.hは「RubyインタプリタAPI(ruby.h)ではない、メソッド実装とその周辺関数だが、他のファイルから参照されている関数」プロトタイプのために用意したヘッダでした。なし崩し的に公開APIっぽくなりましたが。なくすぶんには反対しませんが、1.9の途中でなくして互換性問題が発生しませんか？
-
- まず、intern.h って何者なんでしょう。
推測としては internal の意だと思うんですが、その場合 include/ruby にいるのは
よろしくないですし、一方で公開 API っぽいのが名実ともに多い気がします。
思うに 1.9 において、intern.h の中身は include/ruby/ruby.h か、
最近新設された internal.h のどちらかにあるべきで、include/ruby/intern.h は
もう必要ないのではない... - 01:51 PM Revision bf2e60cd (git): * ext/openssl/ossl_x509name.c: Add X509::Name#hash_old as a wrapper
- for X509_NAME_hash_old in OpenSSL 1.0.0. See #4805
* test/openssl/test_x509name.rb (test_hash): Make test pass with
OpenSSL 1.0.0.
* NEWS: Add it.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32213 b2dd03c8-39d4-4d8f-98ff-823f... -
- r32192 is a fix for No.1. User can set HTTPResponse#chunked = true but it might not be a choice (Some client does not support chunked encoding for example.)
No.3 and No.4 should be discussed in the future... - 01:03 PM Revision 3e01a8d5 (git): * test/openssl/test_ssl_session.rb: Test parameter bug fixed. 2**31 is not in
- a signed long in 32bit long env.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32212 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
12:41 PM Bug #4658: Minitest - invalid return code when using MiniTest::Unit.after_tests
- Just in case a repro comes up, the above code doesn't cover all cases correctly. I'm currently going with:
status = $!.respond_to?(:status) ? $!.status : ($! ? 1 : 0)
- 10:36 AM Revision 544daf1f (git): * ext/openssl/ossl_ssl_session.c (ossl_ssl_session_set_time): Check
- argument type with NUM2LONG if the arg is not a Time object.
See #4919.
* ext/openssl/ossl_ssl_session.c (ossl_ssl_session_set_timeout): Check
type with NUM2LONG. Time as an arg is not allowed. See #4919.
* test/openssl/test_ssl_se... - 09:44 AM Revision 9fe3a7c2 (git): * 2011-06-23
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8_7@32210 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 09:44 AM Revision 2f50fc74 (git): backported r26281 from ruby_1_8
- * lib/webrick/accesslog.rb (WEBrick::AccessLog.format): log parameter
embedding did not work. See #4913.
* test/webrick/test_accesslog.rb: Add for test it.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8_7@32209 b2dd03... -
09:20 AM Revision a02504ba (git): * ChangeLog: Fix typo
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32208 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
07:34 AM Feature #4917: `NilClass#to_ary`
- Perhaps if an optimization is needed, it could just be adding a nil check to Kernel#Array.
-
02:55 AM Feature #4917 (Rejected): `NilClass#to_ary`
- The method `to_ary` is for classes that can be implicitly converted to an Array. This doesn't apply to NilClass.
I'm also highly sceptical as to the actual real life impact of such an optimization. - 07:22 AM Revision 45f164ea (git): * lib/webrick/accesslog.rb: Cosmetic fix of mismatched indentation.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32207 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 04:33 AM Revision b9cbe522 (git): * 2011-06-23
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32206 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32205 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- I'm not sure I agree with adding `to_proc` to `Class` instances, since it seems questionable that `#new` is what you'd always want to be called. Dodging that debate for now, there is another way to get the result you seek:
```ruby
cl...
06/22/2011
-
- Aaron, I'm sorry for late response, but can you point the original report of this issue? The link 'http://intertwingly.net/blog/2010/12/07/SQLite3-CorruptException-database-disk-image-is-malformed' looks not related to openssl. Am I misu...
-
- r32204で、ご提案いただいたとおりの修正を入れました。Rubyオブジェクトからの参照があり、OpenSSLに何を返しても、session_new_cbに渡されたセッションはクリアされないのですから、0や1を返し分けても無意味、なので、OpenSSL側での整合性を取るために、常に0を返して参照数を適切に調整すべき、ということですね。やっと理解しました。ありがとうございました。
結果として、session_new_cbのコールバックについて、「返す値に意味はない」... -
- この参照追加はRubyオブジェクトからの参照で、Rubyオブジェクトの開放時にSSL_SESSION_freeを呼ぶために必要なんですが、逆に0（Rubyレベルでfalse）を返したときのOpenSSLの挙動がよくわかりませんね。。。うーむ。
- 10:28 PM Bug #4919 (Closed): OpenSSL::SSL::Sesssion#time= に Bignum を渡すと ArgumentError が発生する
- This issue was solved with changeset r32203.
Tomoyuki, thank you for reporting this issue.
Your contribution to Ruby is greatly appreciated.
May Ruby be with you.
----------
* ext/openssl/ossl_ssl_session.c: Respect T_BIGNUM time value... -
- Hi, Martin
>1) Is it OK if I apply the fix?
Yes, please apply it. :)
> ...
I can't get this error on Ubuntu 10.4 with openssl(0.9.8k-7ubuntu).
Regards, -
- I get the same result for OpenSSL 0.9.8k plus an additional error when running
make test-all -TESTS="openssl"
on revision 32199.
1) Error:
test_connect_and_close(OpenSSL::TestSSL):
OpenSSL::SSL::SSLError: SSL_connect return... -
- 掲題のような理由で make test-all TESTS="-v openssl/test_ssl_session.rb" を実行すると 1E になります。
1) Error:
test_session(OpenSSL::TestSSLSession):
ArgumentError: unknown type
/opt/ruby-trunk/src/ruby/test/openssl/test_ssl_session.rb:21:in `tim... -
- Hello,
On 22 June 2011 07:46, Martin Bosslet <Martin.Bosslet@googlemail.com> wrote:
> Some tests for OpenSSL inherited from Minitest::Unit::TestCase instead of
> Test::Unit::TestCase. To be consistent, all tests should probably... -
- Some tests for OpenSSL inherited from Minitest::Unit::TestCase instead of
Test::Unit::TestCase. To be consistent, all tests should probably inherit from
Test::Unit::TestCase (cf. [ruby-core:37275]).
I would scan through all tests in... -
- ささださんとakrさんが発見してくださったのですが、現在 trunkで
$ ./ruby -e 'Process.daemon(true,true); p Dir.entries("/proc/self/task")'
と実行すると、スレッドが３匹います(Linuxで実行してください）。何故かというと Process.daemon()
を契機にタイマースレッドが２つ出来てしまうからです。
原因は以下のコードです
---------------... -
-
- Sure! I would put it in openssl_missing.c ?
I'll find out whether there is more than ASN1_put_eoc missing in 0.9.7.
Regards,
Martin -
- openssl-0.9.7 does not have ASN1_put_eoc. Martin, can you handle this?
Dropping 0.9.7 support could be an option if there're other missing functions hardly replace. Users? -
- $ ruby -v -ropenssl -e 'p OpenSSL::OPENSSL_VERSION'
ruby 1.9.3dev (2011-06-22 trunk 32194) [i386-darwin9.8.0]
"OpenSSL 0.9.7l 28 Sep 2006"
$ uname -a
Darwin ... 9.8.0 Darwin Kernel Version 9.8.0: Wed Jul 15 16:55:01 PDT 2009; root:xn... -
- I just committed the fix for http://redmine.ruby-lang.org/issues/4885
in revision 32199.
This should solve the issue that was discovered here, and I was also
able to run the tests without Yui's fix.
Could you please confirm that it... - 05:41 PM Bug #4885 (Closed): [ext/openssl] Use BIO_reset and ERR_get_error in conjuntion
- This issue was solved with changeset r32199.
Martin, thank you for reporting this issue.
Your contribution to Ruby is greatly appreciated.
May Ruby be with you.
----------
* ext/openssl/ossl.h: Introduced OSSL_BIO_reset macro for PEM/D... - 03:04 PM Feature #4672 (Closed): [PATCH] openssl: enable SSL_MODE_RELEASE_BUFFERS if available
- This issue was solved with changeset r32198.
Eric, thank you for reporting this issue.
Your contribution to Ruby is greatly appreciated.
May Ruby be with you.
----------
* ext/openssl/ossl_ssl.c: Use SSL_MODE_RELEASE_BUFFERS if availab... -
02:20 PM Feature #4917 (Rejected): `NilClass#to_ary`
- `Kernel#Array`, when passed `nil`, first tries to send `to_ary`, which ends up calling `method_missing`, and then tries to send `to_a`, which finally succeeds. When `Kernel#Array` is used frequently, for example in library/gem code, this...
- 01:39 PM Revision dea3113a (git): * ext/openssl/ossl_ssl.c (ossl_sslctx_session_new_cb): Return 0 to
- OpenSSL from the callback for SSL_CTX_sess_set_get_cb().
Returning 0 means to OpenSSL that the the session is still valid
(since we created Ruby Session object) and was not freed by us with
SSL_SESSION_free(). Call SSLContext#remov... -
- Tomoyuki Chikanaga.
[ Ruby 1.9 - Bug #4919 ] [ruby-dev:43869]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32203 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- $(topdir). sorry!
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32202 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Moved target version to 1.9.x
-
- On clang -O, it is needed to avoid the optimization.
With this and llvm/clang's recent fix, clang 3.0 can
build ruby-trunk with -O option.
* cont.c (cont_capture): use for-loop.
* array.c (rb_ary_each): add volatile and use it.
* ... - 09:24 AM Revision d8af9c85 (git): * ext/openssl/ossl_ssl.c (ossl_sslctx_session_remove_cb):
- OpenSSL::SSL::SSLContext#session_remove_cb was broken. It wrongly
tried to call the session_*new*_cb callback.
* test/openssl/test_ssl_session.rb (class OpenSSL): Test it.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32200 b2dd... -
- fallback scenarios.
* ext/openssl/ossl_pkey_dsa.c
* ext/openssl/ossl_x509req.c
* ext/openssl/ossl_pkey_rsa.c
* ext/openssl/ossl_pkey_ec.c
* ext/openssl/ossl_ssl_session.c
* ext/openssl/ossl_x509crl.c
* ext/openssl/ossl_pkey.c
* ext/open... -
- I can't replicate with ruby 1.8 or 1.9.2:
require 'rubygems'
gem 'minitest' # ensure latest gem in 1.9
require 'minitest/autorun'
class TestX < MiniTest::Unit::TestCase
def test_x
flunk
en... -
- mail-to-rubymine truncated my response... Fixed:
I'm still broken:
7107 % svn st
7107 % svn info io.c | grep Revision
Revision: 31902
7108 % make -j 4 ruby19
make: `ruby19' is up to date.
7109 % make TESTS="-v testunit/test_parallel.rb... -
06:35 AM Bug #4821: Random Segfaults (in start_thread?)
- Did some further tests to find out where this (for me) hard to understand segfaults came from. I still don't know the root cause, but now I do know that using devise together with meta_search (1.1.0.pre) resulted in the segfaults. Disabl...
-
- Thanks, Eric Wong, for providing the patch.
[ Ruby 1.9 - Feature #4672 ] [ruby-core:36127]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32198 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
05:53 AM Bug #4914: Array#index fails when used in if statement assignment
- Ruby parses your code like this:
if i = (a.index(k.to_sym) && k.to_s =~ /b/)
You can fix this by using "and" or explicit parenthesis:
if i = a.index(k.to_sym) and k.to_s =~ /b/
if (i = a.index(k.to_sym)) && k.to_s =~ /b/... -
- not a bug. run the following:
a = [:a, :b, :c]
for k in 'a'..'c'
if i = a.index(k.to_sym) && k.to_s =~ /b/
puts i
end
end
for k in 'a'..'c'
if i = (a.index(k.to_sym) && k.to_s =~ /b/)
puts i
end
end
for k... -
05:37 AM Bug #4914: Array#index fails when used in if statement assignment
- It "works" for me if you change the `&&` to `and`; so I think it's a precedence issue you're facing here.
I think this blog post here has some good examples which I think directly apply to your case: http://blog.jayfields.com/2007/08/... -
05:28 AM Bug #4914 (Rejected): Array#index fails when used in if statement assignment
- cat /etc/issue: Ubuntu 11.04
rvm -v: rvm 1.6.2
rvm 1.9.2, ruby -v: ruby 1.9.2p180 (2011-02-18 revision 30909) [x86_64-linux]
Also tested with: rvm 1.9.2-head, ruby -v: ruby 1.9.2p274 (2011-06-06 revision 31932) [x86_64-linux]
Test ... -
- * test/openssl/test_pkcs12.rb: Inherit from Test::Unit::TestCase
instead of Mintest::Unit::TestCase. [ruby-core:37275]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32197 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 03:43 AM Revision bda678dc (git): * ext/openssl/ossl_ssl_session.c (ossl_ssl_session_to_der):
- OpenSSL::SSL::Session#to_der was broken. Fix buffer handling.
* test/openssl/test_ssl_session.rb (test_session): Test it.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32196 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 03:40 AM Revision 02ab6d95 (git): * test/openssl/test_ssl_session.rb: Split out SSL::Session related
- tests from test_ssl.rb
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32195 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- mathew: Interesting read. Thanks. I decided to not decide and let users choose, by allowing to include query part as a request URI.
-
- r32189
* test/openssl/test_x509store.rb (test_set_errors): Redhat is
distributing a patched version of OpenSSL that allows multiple CRL
for a key (multi-crl.patch.) Make test pass on such env. See #4122,
...
06/21/2011
-
- 動くようになっているので、closeします。usaさん++
-
- r26299 overrides RbConfig.ruby at test/ruby/envutil.rb for test environment. I think that fix is enough for the original intent of the patch by Aaron. Sorry for this result.
-
- Hi,
thank you for reporting. but we need some additional informations to fix.
Could you provide the following things?:
* which openssl are you using?
* crash log under ~/Library/Logs/CrashReporter
* is same issue happen withou... -
06:51 PM Bug #4912 (Rejected): Segmentation fault while HTTPS request
- Hi,
I'm throwing a SOAP request to a third party using soap4r (git://github.com/noeticpenguin/soap4r-noeticpenguin.git) and I'm getting this stacktrace. It's weird because it seems some kind of incompatibility with another gems, but i... -
- objects. Thanks to CvX for reporting the bug and a test case.
* test/psych/test_object.rb: test for cyclic object references.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32194 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32193 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 05:44 PM Revision 800c313b (git): * lib/net/http.rb (Net::HTTP.post_form): Do not ignore query part of
- the given URI to post. See #655.
* test/net/http/test_http.rb, test/net/http/utils.rb: Test it.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32192 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 04:34 PM Revision 7d112260 (git): * Add ref to tickets.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32191 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 04:33 PM Revision 0fd46532 (git): * 2011-06-22
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32190 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 04:33 PM Revision 5e8aa0a1 (git): * test/openssl/test_x509store.rb (test_set_errors): Redhat is
- distributing a patched version of OpenSSL that allows multiple CRL
for a key (multi-crl.patch.) Make test pass on such env.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
02:05 PM Bug #3827: ext/readline can't handle cursor keys and cp932 chars
- r32181にてパッチをコミットしました。お手数をおかけしますが、確認いただけますでしょうか。
-
- Koichi Sasada wrote:
> 高尾さん
> ...
こちらで試せるようなWin32環境がありません。一旦取り込みますので、試してみていただけないでしょうか。 > Akio Tajima さん -
- This issue was solved with changeset r32179.
Shyouhei, thank you for reporting this issue.
Your contribution to Ruby is greatly appreciated.
May Ruby be with you.
----------
* ext/readline/extconf.rb: fixed bug, specify --disable-libed... -
- =begin
対応が遅くなりました。
以下の修正を入れたいと思います。
動作確認をしてからコミットします。
Index: ext/readline/extconf.rb
===================================================================
--- ext/readline/extconf.rb (revision 32178)
+++ ext/readline/extco... - 12:58 PM Revision 4ce15814 (git): * lib/webrick/httpresponse.rb (HTTPResponse#setup_header): Close
- HTTP/1.1 connection when returning an IO object as response body
without setting HTTPResponse#chunked to true. See #855 no.1.
* test/webrick/test_httpserver.rb: Test it.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32188 b2dd03... -
- from intern.h to internal.h. It's still experimental API and
need more discussion. [ruby-dev:43698]
* include/ruby/intern.h: ditto.
* ext/socket/rubysocket.h: include internal.h.
* ext/socket/depend: add internal.h dependency.
* ext/s... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32186 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 11:42 AM Revision 5b16ddf5 (git): * ext/date/date_core.c (datetime_s_*): canonicalize 24 o'clock.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32185 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32184 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- from intern.h to internal.h. It's still experimental API and
need more discussion.
* include/ruby/intern.h: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32183 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- this test again. Current GVL respect thread priority rather
than past.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32182 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- どうして、これを先月気づかなかったのだろう。という罪悪感があるのですが、
target 1.9.x で起票します。
timer_thread_function()で vm->running_thread にアクセスするのは
thread unsafe な気がします
０．スレッドＡがrunning_threadである
１．タイマースレッドがvm->running_threadをレジスタにのせる
（レジスタにスレッドＡのアドレスがのっかる）
２... -
09:08 AM Bug #4875: Segmentation fault in http.rb
- I get this any time I try any sort of HTTP call. There is some suggestion that it might relate to the use of the 'pg' gem for Postgres, see first few results:
http://www.google.com/search?q=http.rb+678+pg
/Users/abailey/.rvm/rubi... -
- Sure.
I have to merge in my announced changes to Matrix & Vector in 1.9.3 as soon as I get the time, so I could include this too. Are we after the official feature freeze yet (I don't see the 1.9.3 branch)? -
- #3827 by by Akio Tajima <artonx AT yahoo.co.jp>. (see #3827)
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32181 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 04:20 AM Revision d2e61d22 (git): * 2011-06-21
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32180 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- then disable libedit, does not specify then check readline and
libedit if failed checking readline. (fixes #3375)
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32179 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- ささだです．
(2011/06/21 1:25), Masaya TARUI wrote:
> 同意します。
> というか、いままでmaskされてないと知らなかった。
mask するための新しい仕様が必要になると思いますが，さてこれを 1.9.3 に
入れますか？ それとも，trap 中はどんな trap を再入禁止？
--
// SASADA Koichi at atdot dot net -
01:29 AM Bug #4909: trapハンドラは再入されてはいけないのではないか？
- 同意します。
というか、いままでmaskされてないと知らなかった。
2011年6月20日18:46 Motohiro KOSAKI <kosaki.motohiro@gmail.com>:
>
> Issue #4909 has been reported by Motohiro KOSAKI.
>
> ----------------------------------------
> Bug #4909: trapハンドラは再入さ... -
- I also get the segfault when running "rails generate" and when running tests. So it does not seem to be solely a problem with webrick? Maybe the causes are multiple?
06/20/2011
-
- Benoit Daloze wrote:
> Hello,
> ...
No, it was meant exactly as stated. Advantage is that you can provide parameters to #new if needed while mapping the parameterless call of `to_proc` easily to the parameterless call of `Class#new`.
... -
- Hello,
Robert Klemme wrote:
> I suggest to add these two to class Class:
> ...
Did you want to mean:
```ruby
def to_proc
lambda { |*args| new(*args) } # or maybe lambda { |args| new(*args) }
end
```
?
`#to_proc` is call... -
- I suggest to add these two to class `Class`:
```ruby
class Class
alias call new
def to_proc(*args)
lambda {|*a| new(*args)}
end
end
```
Then we can use class instances where blocks are needed and can easily use t... -
11:23 PM Feature #4893: Literal Instantiation breaks Object Model
- On Fri, Jun 17, 2011 at 5:12 PM, Lazaridis Ilias <ilias@lazaridis.com> wrote:
> "Object Model" as in
>
> * Design and implementation of a programming language's OO behaviour (classes, object, inheritance, mixin, methods, attributes... -
- Excellent result. It looks like a gift from Tor project. There's no reason I can imagine not to apply this. Please apply this.
I think it's OK to commit this w/o writing a test. :) -
- わたしは構いませんが高尾さん的にはどうなんでしょう。
ていうか、#1823 #3827 #4635 が放置されてて何とかしてくれって感じなんですが -
- これ結局自分が間違ってたと思うんですが、どうすればいいんでしょう。勝手に直していい?
-
- ありがとうございました。r32176でパッチを当てました。
-
- 以下のプログラムは
if intrap == 1
raise "trap nested"
end
が真になってしまって例外終了してしまうのですが、これは起きてはいけないのではないでしょうか。
以下の二点が問題だと考えます。
１）このプログラムのようにシグナルを連続して配送されるとスタックオーバーフローを引き起こせる
２）Rubyレベルでsigprocmask()に相当するシグナルブロッキング操作が提供されていないため、
... -
- I agree with DH interface is confusing.
* DH#p ... DH parameter p
* DH#g ... DH parameter g
* DH#public_key ... DH parameter (DHParameterSpec in Java)
* DH#priv_key ... private value: S
* DH#pub_key ... exchange value: g^S mo... -
- The feature is changed on 1.9.2; now nil is not a valid argument.
-
04:59 PM Bug #4908 (Rejected): #force_encoding(nil) fails on Windows
- I read from The Ruby Programming Language that nil can be specified as "binary" or "ascii-8bit".
The book give a sample code on Page 61:
text = stream.readline.force_encoding("utf-8")
bytes = text.dup.force_encoding(nil) # nil e... -
06:01 PM Bug #3924: Performance bug (in require?)
- You might be also interested in my very short patch: https://gist.github.com/1035322 Basically it reorders conditions to start with fastest ones and execute slower ones only if needed. Here is also some description: http://www.lunarlogi...
-
- Sorry for keeping this ticket open.
I thought
logger.info ["User-input: %{input}", input: data]
looks good, though I like
logger.info "User-input: %{input}", input: data
better. But I cannot find the compatible way to ... -
- すいませんruby-devだった。orz
成瀬さん、NEWSになんか書きませんか。 -
- Naruse-san, would you please add an explanation of this feature to NEWS file?
-
- On Mon, May 16, 2011 at 11:47, NARUSE, Yui <naruse@airemix.jp> wrote:
> I have another comment: nahi and emboss's commit message has extra
> header and indent like:
>
> Mon May 16 05:13:20 2011  Martin Bosslet  <Martin.Bosslet@goo... -
- for avoiding a race.
* process.c (sig_do_nothing): new function.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32178 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Sorry for late reply, and marking it as 'Rejected'. I'm just want to say that 'it's not a bug'. (We should have suitable label in the future.)
net/http requires to set 'use_ssl' flag manually before starting a connection.
requir... -
- time_quantum. it's no longer 10ms.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32177 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Hi,
On Thu, Jun 9, 2011 at 06:14, Martin Bosslet
<Martin.Bosslet@googlemail.com> wrote:
> I think your patch is good! But there is one thing I don't like
> about OpenSSL itself here - why do they hardcode the digest algorithm
> i... -
- Adding '-r openssl' would affect autloading but the location where an exception thrown is OpenSSL::SSL::SSLSocket#connect.
Does the SEGV always happens for you? It would help to show us a minimized reproducible code or concrete steps ... -
11:06 AM Feature #4907 (Rejected): enumerable#permutation and combination
- Hello
Methods permutation and combination are defined for array but it make more sense to define them for enumerable.
Here is sample implementation which for simplicity works only with blocks.
Note that implementation works lazily.
... - 09:48 AM Revision c1028212 (git): * ext/openssl/ossl_cipher.c, ext/openssl/lib/openssl/cipher.rb:
- Documentation fix by Ippei Obayashi. See #4419.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32176 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 06:48 AM Revision 15e7e472 (git): * lib/webrick/cookie.rb (WEBrick::Cookie.parse): Revert r31228.
- r31228 was for allowing the 'Cookie:' header which did not have no
SP after ';' for separating cookie-pairs but RFC6265 requires single
SP after ';' there. We allow multiple SPs here for compatibility
with older WEBrick version.
... - 01:00 AM Revision 3fc0e0c1 (git): * 2011-06-20
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32174 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32173 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
12:21 AM Feature #4906: rb_w32_add_socket / rb_w32_remove_socket
- Thank you Greg for the work on this.
I've assigned to Mr. Nobuyoshi Nakada, hoping this can be cleared out as a w32 API specific and be allowed in 1.9.3.
I don't think Yugui's schedule has been followed (or at least was delayed) so...