Activity
From 05/03/2014 to 05/09/2014
05/09/2014
-
11:48 PM Revision 08928bb5 (git): * signal.c (trap): Return "SYSTEM_DEFAULT" if SIG_DFL is set.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45894 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
11:44 PM Bug #7395: Negative numbers can't be primes by definition
- > ayumin
this issue seems incompatible changes. it should be add NEWS. -
11:36 PM Bug #9823 (Feedback): Segmentation fault using DL
- Hello,
There is a require to mwe that is not defined. Please include all the code or provide a gist with multiple files that describe the problem. -
11:34 PM Bug #9823 (Closed): Segmentation fault using DL
- I receive a segmentation fault when calling OpenProcessToken from DL, demonstrated by running the attached 'runner.rb' file with the additional files in the same directory. This is a simplified representation, but perhaps not minimal, th...
-
11:19 PM Feature #4788: resolv.rb refactoring
- I'm happy to se this has been completed. I just thought I'd mention that I reimplemented the DNS resolver in RubyDNS using EventMachine as I couldn't wait for this bug fix. The new EventMachine resolver is asynchronous which makes it use...
-
10:02 PM Bug #9822 (Assigned): Ruby doesn't respect system OpenSSL configuration
-
09:24 PM Bug #9822 (Closed): Ruby doesn't respect system OpenSSL configuration
- Hello.
I need to work with SSL (HTTPS) with GOST encryption, but ruby doesn't connect to the servers that requires GOST algorithms to be used for encryption.
The issue is in fact, that it is required to modify system OpenSSL config... -
05:04 PM Feature #9508: Add method coverage and branch coverage metrics
- Hi Yusuke, thanks for the comments! I want to first defend the Demand and Use Case. And thank you for the Review comments; I'll apply them ASAP.
## Demand in Ruby Core
I think that Ripper is inadequate for these new metrics for the... -
04:15 PM Revision 5d488225 (git): test_signal.rb: missing argv
- * test/ruby/test_signal.rb (test_hup_me): add missing argv.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45893 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 03:13 PM Revision 800c2d39 (git): * 2014-05-10
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45892 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * test/ruby/test_beginendblock.rb (test_propagate_signaled): run
with default handler.
* test/ruby/test_signal.rb (test_hup_me): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45891 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- GVL があるので変なことはまず起こらない気がしますが、起こるとしたらどういう状況ですかね。
-
- `SIG_IGN`ではないときだけ、というところでrace conditionになったりしないですかね。
-
- 自分自身に送るときは、即座に届かないと困ることがあった気がします。
例外が発生するはずがその先まで実行されてしまうとか。
思うに、自分自身(だけ)に送るときは signal を使わないで、直接 signal handler を呼ぶなり呼ばないなりすればいいんじゃないかという気がするんですが、どうですかねぇ。
-
- 元々シグナルがすぐ配送されるとは限らないわけで、とりあえずタイムアウトをつけてみるのはどうでしょうか。
~~~diff
diff --git i/thread.c w/thread.c
index dfa91a8..209aff1 100644
--- i/thread.c
+++ w/thread.c
@@ -5180,10 +5180,11 @@ ruby_kill(rb_pid_t pid, int sig)
* delivered ... -
- r39819で、自分自身にシグナルを送ったときにはすぐシグナルを処理するために受信するまで待つようになっていますが、`SIG_IGN`の場合は送られないのでそこでずっと待機してしまうということのようです。
シグナルの遅延が予測できない以上、`kill`する前にチェックするしかないですかねぇ。 -
- さらに調べた結果、どうも cron から動かしたプログラムでは SIGINT の signal handler が
SIG_IGN になっているようです。
以下のようにして、端末からでも再現できました。
以下を実行すると hang します。
```
./miniruby -e 'trap(:INT, "IGNORE"); Process.kill(:INT, $$)'
```
Ctrl-C で止められないのは少し面倒なので、SIGQUIT に変... -
- 最近、しばたさんが運用を始めた chkbuild で、
TestBeginEndBlock#test_propagate_signaled が hang しています。
http://chkbuild002.hsbt.org/chkbuild/ruby-trunk/log/20140509T060013Z.fail.html.gz
調べて再現コードを小さくしていった結果、
miniruby -e 'Process.kill(:INT, $$)'
... -
- * thread_{pthread,win32}.c (native_cond_timedwait): constify.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45890 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * signal.c (ruby_signal): turn multiple `if` statements into a
`switch`.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45889 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45888 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Seems it occurs in RMagick.
-
11:26 AM Bug #9821 (Third Party's Issue): Segfault when using gruff/RMagick/Rails 3.2.17
- Our Rails 3.2.17 project uses gruff, which uses RMagick. It worked fine with Ruby 1.9 (FreeBSD 8.3). Then we moved to another server with Ruby 2.1 (FreeBSD 10.0) and now the Ruby code that generates the image crashes.
Bug report as gi... -
06:12 AM Feature #9816: 文字列内の数字を数値として比較するメソッド
- メソッド名は分かりませんが、
a.foo(b)
という書き方は対称性を崩していて何となく気持ち悪いので、どうせなら
String.foo(a, b)
という書き方も見当していただけないでしょうか。あるいは、`<=>`と同様に、メソッドの前の`.`の要らないメソッドがあるといいんですけどね。 -
- Yui NARUSE wrote:
> > (あと numericcmp は cc という同じ文字の並びの間で単語が分かれるというのが、
> ...
`numcmp`とか。 -
05:32 AM Feature #9816: 文字列内の数字を数値として比較するメソッド
- tadayoshi funaba wrote:
> 名前はともかく俺が欲しいのは filevercmp のほうかもしれない。
仕様が明確に決まっているのと、バージョン比較というわたしのユースケースには影響がないので、
それでもいい感はありますが、どうなんでしょうね。
Akira Tanaka wrote:
> ...
それぞれ後半が逆な気がしますが、MS的にはlogicalcmpらしいですね。
> ではどういう名前にするかというと、あまりいい名前... -
02:48 AM Feature #9816: 文字列内の数字を数値として比較するメソッド
- 2014-05-08 7:54 GMT-04:00 Tadayoshi Funaba <tadf@dotrb.org>:
> 名前はともかく俺が欲しいのは filevercmp のほうかもしれない。
>
> ~~~
> x #=> ["2.1.10", "2.1.2", "8 layers", "8 layers 2", "8 layers 2.nki", "8 layers.nki", "a16", "a17"]
> puts x.sort{|a... -
- Feature#5861 の時と違って、色々とバックグランドの話が提示されたので私は説得されました。
+1
> ~~~diff
> + def test_numericcmp
> + assert_equal(-1, "2.1.0".numericcmp("2.1.1"))
> + assert_equal(-1, "2.1.9".numericcmp("2.1.10"))
> + assert_equal( 0... -
- * ext/extmk.rb (extmake): don't build nested libraries if parent
library isn't build.
* ext/{dl/callback,tk/tkutil}/extconf.rb: no longer need to check
if the parent is build.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45887... -
- * ext/extmk.rb: fix variable name for --extstatic option.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45886 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r45885.
----------
configure.in: check for `setjmp` type with `CCDLFLAGS`
* configure.in (`RUBY_SETJMP_TYPE`): check for `setjmp` type after
`CCDLFLAGS` is appended to `CFLAGS`, since `__builtin_setjmp` can... -
- You can use `--with-setjmp-type=_setjmp` `configure` option.
-
- * configure.in (RUBY_SETJMP_TYPE): check for setjmp type after
CCDLFLAGS is appended to CFLAGS, since __builtin_setjmp can be
affected. [ruby-core:62469] [Bug #9818]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45885 b2dd03c8-... -
04:35 AM Bug #9819 (Closed): changed documentation for enumerable methods find and detect
- The docs right now under the enumerable method "find" contains only 2 examples using detect, and no examples with find. The usual pattern throughout the rest of the docs is that when two methods are aliases, such as detect and find, then...
- 03:58 AM Revision 48a9d171 (git): * tool/release.sh: fix for neon.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45884 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
03:14 AM Feature #6074 (Rejected): Allow alias arguments to have a comma
- As Akinori stated, alias is special to be special.
In your case, forget alias and always use alias_method.
Matz.
-
02:34 AM Feature #6074: Allow alias arguments to have a comma
- Akinori MUSHA wrote:
> I believe the lack of comma is an important sign that alias is not really a method call in which directive arguments are not evaluated like with normal method calls.
Agreed, and this issue should be judged by M... -
02:05 AM Feature #6074: Allow alias arguments to have a comma
- I believe the lack of comma is an important sign that alias is not really a method call in which directive arguments are not evaluated like with normal method calls.
If you are allowed to, and should you put a comma in an alias statemen... -
- How about performance down of this change?
-
01:00 AM Bug #9800: Ship 2.1.2 with GC_HEAP_OLDOBJECT_LIMIT_FACTOR of 1.3
- Sorry for late response.
I can agree this proposal because current default "2" also has no evidence to use.
I think more survey are welcome to set ideal default setting. 1.7? 1.5? 1.3?
However, already 2.1.2 released?
05/08/2014
-
- Yui NARUSE wrote:
> なお、メソッド名は String#numericcmp としています。
"2.9" と "2.10" の比較を考えると、
整数の並びの辞書順の比較として 2.9 が大きいとするか、
数 (実数) の比較として 2.10 が大きいとするか、
という考え方の違いがあります。
numericcmp という名前は後者を想起させる気がします。
今回欲しいのは前者だと思うので、それを想起させる名前がいいんじゃない... -
- `versioncmp`は'-'と'.'だけ特別扱いって感じですかね。
あと、片方だけが数字で終わっているときの終端チェックが抜けているような。
~~~diff
diff --git i/string.c w/string.c
index 66f667f..855d74f 100644
--- i/string.c
+++ w/string.c
@@ -2639,6 +2639,11 @@ rb_str_numericcmp(VALUE str1, V... -
11:58 AM Feature #9816: 文字列内の数字を数値として比較するメソッド
- 名前はともかく俺が欲しいのは filevercmp のほうかもしれない。
~~~
x #=> ["2.1.10", "2.1.2", "8 layers", "8 layers 2", "8 layers 2.nki", "8 layers.nki", "a16", "a17"]
puts x.sort{|a,b| a.numericcmp(b)}
2.1.2
2.1.10
8 layers
8 layers 2
8 layers 2.nki
8 ... -
- Kazuhiro NISHIYAMA wrote:
> numericcmp だと複数の数値が入っている文字列を比較するものというのがわかりにくいと思いました。
> ...
名前の候補としてはいいんじゃないですかね。
String に生やしているわけではないので、puppetが使っているのは無視していいと思います。
> coreutils だと filevercmp という名前のようです。
> ...
rpmvercmpは論外として、filevercmpは... -
10:15 AM Feature #9816: 文字列内の数字を数値として比較するメソッド
- numericcmp だと複数の数値が入っている文字列を比較するものというのがわかりにくいと思いました。
最初に思いついたのは versioncmp という名前でしたが、
puppet などで使われているようです。
https://github.com/puppetlabs/puppet/blob/master/lib/puppet/util/package.rb
coreutils だと filevercmp という名前のようです。
https://... -
09:50 AM Feature #9816: 文字列内の数字を数値として比較するメソッド
- +1
自分でいちいち書くのも面倒だし、Gem::Versionをこのために持ってくるのも……という類のメソッドですよね。 -
- 文字列内の数字を数値として比較するメソッドを追加しませんか
そのような比較は一般的な用途としてはGUIシェルのファイラーが比較に用いており、
Windows では StrCmpLogicalW が、OS X では NSString:compare:options:へのNSNumericSearch定数が提供されています。
http://msdn.microsoft.com/en-us/library/windows/desktop/bb759947(v=vs.... -
09:13 PM Bug #9818 (Closed): __builtin_setjmp and __builtin_longjmp caused a build failure on PPC Linux with gcc 4.4.0
- From a certain revision around r45503 to r45515, Ruby began to use `__builtin_setjmp` and `__builtin_longjmp` on PPC Linux, instead of `_setjmp` and `_longjmp`. However, `__builtin_setjmp` and `__builtin_longjmp` cause a build failure be...
-
-
08:07 AM Bug #9814 (Closed): Improving documentation example for Enumerator#peek
- The existing example for **Enumerator#peek** doesn't effectively document the behavior when peeking after an enumerator has been entirely consumed. Rather, it documents the behavior of **Enumerator#next**.
This change more effectively... -
07:51 PM Feature #9781: Feature Proposal: Method#super_method
- I've only ever needed to do this in codebases like Rails, where there's an insane amount of inheritance. However, Rails also relies heavily on method_missing, which `super_method` would completely miss.
Anyway, I'd like this method to... -
07:02 PM Revision 2ff10784 (git): * lib/delegate.rb: Fix example of using delegator.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45882 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45881 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45880 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 06:22 PM Revision 9fdf3e3e (git): * 2014-05-09
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45879 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Applied in changeset r45878.
----------
* lib/prime.rb (Prime#prime?): negative numbers can't be primes
by definition. reported by Ivan Kataitsev. [Bug #7395]
* test/test_prime.rb: add test. -
- by definition. reported by Ivan Kataitsev. [Bug #7395]
* test/test_prime.rb: add test.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45878 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- I didn't notice this by now, because there was no problem for building mswin at all.
Are these lines really necessary on mingw? -
03:14 PM Bug #9817 (Closed): The extconf.rb for OpenSSL assumes MingW
- It appears that the extconf.rb file for openssl assumes mingw. The problem is that gdi32 and wsock32 need to be linked, even for Visual Studio.
The simple fix is to change "`if $mingw`" to "`if File::ALT_SEPARATOR`" on line 40.
-
- Akira Tanaka wrote:
> Yui NARUSE wrote:
> ...
まぁ採否も含めてまつもとさんに判断頂く際に、名前も決めていただくってことでいいかと。
> あるいは、もっと動機に率直なメソッドにしてしまうという手もあって、
> ...
ふむ、確かに今のわたしのユースケースだとこちらのほうが適している気はしますね。
こちらはこちらでいじってみます。
こっちだと Etc.mntinfo とか Etc.df とかですかねぇ -
- Yui NARUSE wrote:
>
> ...
とすると、statvfs があるときは少なくとも statvfs の情報は提供することになるような気がしますね。
statvfs は POSIX で標準化されているので、結局、提案されているメソッドの実際の動作は
statvfs + 環境依存の様々な情報、ということになりそうな気がします。
そうすると常に提供されると期待できるのは statvfs の部分なわけで、やっぱ名前は statvfs のほうがい... -
- Akira Tanaka wrote:
> Yui NARUSE wrote:
> ...
他の方が互換実装を提供することに反対はしませんが、わたしがわざわざ実装する動機は欠けますね。
> > なお、NetBSDとOpenBSDでは動くようにした…はずです。
> ...
r45873で直しました。 -
- どういう意図でしたっけ。
ちょっとググったけど見当たらず。
-
- 意図的なものです。
`rb_ivar_get`と`rb_attr_get`の差です。 -
attr_reader で定義したメソッドでは、-w 時に、未設定の警告「warning: instance variable @bar not initialized」みたいなのが出ないんですが、そういうもんでしょうか。
~~~ruby
class C
attr_reader :foo
def bar; @bar; end
end
C.new.foo
C.new.bar #=> t.rb:4: warning: instance vari...- 06:31 AM Revision 2a83260a (git): fixed previous commit mistake.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_0_0@45876 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * object.c (init_copy): no longer copy tables of classes/modules,
since r45874 rb_mod_init_copy() does it instead.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45875 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r45874.
----------
class.c: always clear tables first
* class.c (`rb_mod_init_copy`): always clear instance variable,
constant and method tables first, regardless the source tables.
[`ruby-dev:48182`] [`... -
- `Module#initialize_copy`で定数やインスタンス変数を持たない`Module`をコピーしても、元の定数やインスタンス変数が残っています。
~~~ruby
m = Module.new do
def x
end
const_set(:X, 1)
@x = 2
end
p m.instance_methods, m.instance_variables, m.constants
#=> [:x]
#=> [:@x]... -
- * class.c (rb_mod_init_copy): always clear instance variable,
constant and method tables first, regardless the source tables.
[ruby-dev:48182] [Bug #9813]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45874 b2dd03c8-39d4-4d8f-98... -
- sys/mount.h. [ruby-dev:48167]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45873 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * lib/webrick/httpserver.rb (WEBrick::HTTPServer#run): stop
handling requests on shutdown, even if the socket is readable
and IO.select() returns true. [Fixes GH-607]
* lib/webrick/server.rb (WEBrick::GenericServer#start): IO.select... -
12:28 AM Feature #9777: Feature Proposal: Proc#to_lambda
- Beware! The answer on Stack Overflow (http://stackoverflow.com/a/2946734) is wrong, because it does not preserve the value of `self` in the block.
This is better, but still wrong:
~~~
def convert_to_lambda &block
obj = block.bi...
05/07/2014
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45871 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- backported a part of r45453 into ruby_2_0_0.
I decided to update only libyaml and not psych itself. - 05:30 PM Revision 30340160 (git): merge revision(s) r45453: [Backport #9798]
- * ext/psych/yaml/config.h: ditto.
* ext/psych/yaml/scanner.c: ditto.
* ext/psych/yaml/yaml_private.h: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_0_0@45870 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- backported into ruby_2_0_0 at r45869.
I couldn't reproduce SEGV with wanabe's script, but I believe the fix is right, or at least has no bad side effect :) - 05:07 PM Revision 9bececb7 (git): merge revision(s) 45015: [Backport #9657]
- * ext/bigdecimal/bigdecimal.c (BigDecimal_initialize): Insert GC guard.
* ext/bigdecimal/bigdecimal.c (BigDecimal_global_new): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_0_0@45869 b2dd03c8-39d4-4d8f-98f... -
- backported into ruby_2_0_0 at r45868.
- 04:59 PM Revision b786887a (git): merge revision(s) 45595: [Backport #9743]
- * ext/openssl/ossl_pkey.c (ossl_pkey_verify): as EVP_VerifyFinal()
finalizes only a copy of the digest context, the context must be
cleaned up after initialization by EVP_MD_CTX_cleanup() or a
memory leak wi... -
04:22 PM Bug #9586: Unable to build Ruby 2.1.1 on NetBSD
- r45190, r45202, r45204 and r45713 were backported into `ruby_2_1` at r45867.
-
- configure.in: define SET_THREAD_NAME
* configure.in (SET_THREAD_NAME): define according to
pthread_setname_np variations.
* configure.in: correct pthread_setname_np's prototype on NetBSD.
[Bug #9586]
git-sv... -
- test_gc.rb: relax assertion
* test/ruby/test_gc.rb (test_profiler_total_time): GC time may be
shorter than the timer resolution.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_1@45866 b2dd03c8-39d4-4d8f-98ff-823f... - 03:43 PM Revision 87083e76 (git): * 2014-05-08
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45865 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * gc.c ({free,mark}_{method,const}_entry_i): fix argument types as
st_data_t for st_foreach().
* gc.c (rgengc_unprotect_logging_exit_func_i): add third argument.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45864 b2dd03c8-39d4-4... -
- I think this introduces no incompatibility and should be backported.
-
- Applied in changeset r45863.
----------
* ext/openssl/lib/openssl/ssl.rb (OpenSSL::SSL::SSLServer#accept):
Consider Socket#accept as well as TCPServer#accept.
Reported by Sam Stelfox. [ruby-core:62064] [Bug #9750] -
- Request for syntax change.
-
- Salvaged. https://github.com/ruby/ruby/pull/609
-
- Ulrich Kramer wrote:
>
> ...
No.
caller() doesn't return the current stack frame unless 0 is specified as the argument.
caller() returns an array which contains the caller stack frame and ancestor stack frames.
caller of the to... -
- Consider Socket#accept as well as TCPServer#accept.
Reported by Sam Stelfox. [ruby-core:62064] [Bug #9750]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45863 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45862 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- In the meantime, third parties can work around this problem by a monkey patch like this:
https://github.com/sparklemotion/nokogiri/commit/c98745d9098487c51685a732f1c5a21d8d07cdad -
- Applied in changeset r45861.
----------
numeric.c: check keyword arguments
* numeric.c (`num_step_scan_args`): check keyword arguments and fail
if they conflict with positional arguments.
[`ruby-dev:48177`] [`Bug #9811`] -
- Numeric#stepにキーワード引数を指定すると、終了値が無視されます。
無効なキーワードでもエラーになりません。
~~~ruby
1.step(10, foo: nil)
~~~
-
- * numeric.c (num_step_scan_args): check keyword arguments and fail
if they conflict with positional arguments.
[ruby-dev:48177] [Bug #9811]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45861 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * numeric.c (num_step_scan_args): fix merge miss.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45860 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * numeric.c (num_step_scan_args): turn a macro into a function.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45859 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * numeric.c (ruby_num_interval_step_size): adjust indent.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45858 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
07:03 AM Bug #9677: OpenSSL::PKey::EC#to_text causes a segmentation fault
- I found that the segmentation fault occurs when a public_key is not set.
Here is a patch.
~~~
diff --git a/ossl_pkey_ec.c b/ossl_pkey_ec.c
index 5e419bd..070bfc4 100644
--- a/ossl_pkey_ec.c
+++ b/ossl_pkey_ec.c
@@ -574,6 +574,13... -
06:58 AM Bug #5328: Splat args are treated wrongly in instance_exec and class_exec
- Thanks. I'm still trying to wrap my head around it but that's quite helpful.
-
05:54 AM Feature #9807: String.new with block
- Michael Kohl wrote:
> After a discussion in our team chat today, I wondered if it would be a good idea to have a version of String.new that accepts a block and works as a string builder. Something like
> ...
Like Matthew, I'd also like... -
- * vm_eval.c (rb_f_local_variables): hide intermediate hash and use
rb_hash_keys() directly.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45857 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * hash.c (rb_hash_keys): make an internal public function.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45856 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * vm_eval.c (local_var_list_add): use symbol identity, and get rid
of method calls.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45855 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * vm.c (vm_collect_local_variables_in_heap): use struct local_var_list
instead of VALUE directly.
* vm_eval.c (local_var_list_update): add valid name IDs only.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45854 b2dd03c8-39d4-4d8... -
- * vm.c (rb_f_local_variables): clear intermediate hash after
extracting its keys.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45853 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- specified file.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45852 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- raw results into FILE.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45851 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * vm_eval.c (rb_f_local_variables): exclude variables hidden by
shadowing. [ruby-core:60501] [Bug #9486]
* vm.c (collect_local_variables_in_iseq): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45850 b2dd03c8-39d4-4d8f-98ff... - 02:00 AM Revision c8dbc43d (git): * 2014-05-07
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45849 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * vm.c (vm_collect_local_variables_in_heap): rename an argument.
* vm_eval.c (rb_f_local_variables): rename a local variable.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45848 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- This is a repeated question, so it feels nice to add examples.
-
-
- If a user wanted to write it as a proc, then it means that he/she doesn't want it to raise error, doesn't it?
05/06/2014
-
- I can remember several documents.
* The document of Proc#lambda?.
* The thread from ruby-core:47453 "yield plus splat unwraps too much"
(I don't recommend reading it in bugs.ruby-lang.org because messages are unintentionally forma... -
- Thanks for taking the time to explain.
This surprises me a bit because I've followed ruby development closely for a number of years and this is the first time I've heard of this particular proc vs lambda difference (I know of others l... -
- Ruby 1.9 changes block arguments behavior.
Non-lambda block arguments always expands single array.
(Ruby 1.8 expands sometimes.)
If you needs block arguments behaving as methods, use lambda.
```
% cat splat_args
def foo(arg_1, ... -
11:01 PM Bug #5828: Non anonymous classes can't be frozen, cloned and then inspected
- I'm happy for this bug to be closed, as I don't have a need to freeze and clone a class.
-
- The name of anonymous class is searched and cached when it is first inspected.
The cache is implemented as a hidden instance variable of the class.
If the class is frozen, the cache is failed as the exception because the instance varia... -
09:13 PM Bug #9806: URI#encode doesn't encode characters '[' and ']'. They should be encoded as %5B and %5D respectively.
- Notes:
* Per RFC 2396 section 2.4.3 "Data corresponding to excluded characters **must** be escaped in order to be properly represented within a URI."
* Per RFC 2396 section 2.2 reserved characters are ";" | "/" | "?" | ":" | "@" | "&" ... -
- The subject says it all.
IRB session demonstrating the problem:
charlez$ irb
head :001 > RUBY_VERSION
=> "2.2.0"
head :002 > require 'uri'
=> true
head :003 > my_str = '[ futsal club ]'
=> "[ futsal club ]"
head :004 > URI.... -
05:25 PM Bug #9810 (Closed): Numeric#step behavior with mixed Float, String arguments inconsistent with documentation
- The `Numeric#step` documentation states:
"If any of the arguments are floating point numbers, all are converted to floats..."
https://github.com/ruby/ruby/blob/trunk/numeric.c#L1921
Is the following code intended to be described... -
05:21 PM Bug #9809: Date.new(2000,1,1).cweek should == 1
- This looks like correct behaviour. From a previous issue #9372:
A period of time of seven days within a calendar year, starting on a
Monday and identified by its ordinal number within the year; the first
calendar week of the year is... -
04:54 PM Bug #9809: Date.new(2000,1,1).cweek should == 1
- I see https://bugs.ruby-lang.org/issues/9372
If this is not a bug, please close it, thank you. -
- I run `Date.new(2000,1,1).cweek`, the result is 52.
My Ruby version is 2.1.1 -
05:17 PM Feature #9777: Feature Proposal: Proc#to_lambda
- Another possible reason to convert a Proc to a lambda is to for raising error on arguments
```
foo = -> { puts "hello" }
foo.call(1)
ArgumentError: wrong number of arguments (1 for 0)
from (irb):4:in `block in irb_binding'
f... -
- Could you explain what it does?
-
02:00 PM Feature #9808 (Rejected): Enumerable#argmin_by, #argmax_by
- What about introducing `Enumerable` `#argmin_by` (and `#argmax_by` too, of course) method?
Use cases are wide from mathematical problems to a general tasks. Here are two very simple examples:
~~~ruby
(1..2).step(0.01).argmin_by{|x| ... -
- Applied in changeset r45847.
----------
missed ref. [ruby-core:60501] [Bug #9486] -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45847 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * parse.y (new_bv_gen): no duplicated names, if already added in
shadowing_lvar().
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45846 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * parse.y (local_tbl_gen): remove local variables duplicated with
arguments.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45845 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * parse.y (arg_var_gen, local_var_gen): make void since the result
values are not used.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45844 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
-
-
11:38 AM Feature #9807: String.new with block
- >
> That could be a nice solution to avoid all the Array#join and "".tap
hacks for string creation.
>
Which hacks are these? Also, I don't see how it's different from
´"foo".tap{|s|s<<"bar"}´
Can you give some examples? -
11:22 AM Feature #9807 (Open): String.new with block
- After a discussion in our team chat today, I wondered if it would be a good idea to have a version of String.new that accepts a block and works as a string builder. Something like
string = String.new("foo") do |s|
s << "bar... -
- https://github.com/ruby/ruby/pull/608
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45843 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 09:37 AM Revision d85c226f (git): * io.c (io_{read,write}_nonblock): use rb_get_kwargs instead of
- rb_hash_aref.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45842 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Time.parse and Time.strptime.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45841 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * eval.c (setup_exception): add the method name to system stack
error message.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45840 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45839 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- It seems the original document of that part written by gsinclair used GMT.
But I mistakenly modified to use JST.
I agree that most examples uses EST.
This convention is inherited from time.c.
time.c has a sentence:
"All of these... -
- Applied in changeset r45837.
----------
* lib/time.rb (Time.parse): [DOC] Fix an example in the documentation.
Reported by Marcus Stollsteimer.
[ruby-core:60778] [Bug #9521] and [ruby-core:61718] [Bug #9682] -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45838 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Applied in changeset r45837.
----------
* lib/time.rb (Time.parse): [DOC] Fix an example in the documentation.
Reported by Marcus Stollsteimer.
[ruby-core:60778] [Bug #9521] and [ruby-core:61718] [Bug #9682] -
- Reported by Marcus Stollsteimer.
[ruby-core:60778] [Bug #9521] and [ruby-core:61718] [Bug #9682]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45837 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
05/05/2014
-
08:43 PM Feature #9804: File::CREATE as a synonym for File::CREAT
- I do not think this increases human overhead to learn: CREATE is an English word, CREAT is not. This would allow me to avoid writing in my code
File::CREATE = File::CREAT -
07:34 PM Feature #9804: File::CREATE as a synonym for File::CREAT
- alexey.muranov@gmail.com wrote:
> I would like to have `File::CREATE` as a synonym for `File::CREAT`.
At least give a reason for wanting this...
Fwiw, I am against adding more aliases/synonyms to core.
They increases human overh... -
- I would like to have `File::CREATE` as a synonym for `File::CREAT`.
-
- The following code:
def a; a end
begin
a
rescue => e
puts e.backtrace.inspect
end
only returns a single line of backtrace like:
/tmp/scratch:3:in `a'
but this is not helpful when debugging.... - 07:38 PM Revision 08e0c31d (git): * file.c (rb_f_test): removed meaningless "case 'a'".
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45836 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- I'm negative.
Such convention makes naming new method more difficult.
Naming is already very difficult.
I don't want to it more difficult. -
- I think this is not a problem of Ruby.
-
- Please avoid Timeout.
Even if it doesn't hang as Ruby 1.8, it doesn't clean child processes.
Thread#join's argument (timeout) is usable to implement to timeout.
```
Open3.popen3("sleep 100 && echo hi", :pgroup=>true) {|i, o, e, w... -
- r45453, r45455 and r45577 were backported into `ruby_2_1` at r45812.
See #9798 -
04:46 PM Misc #9803 (Rejected): document of Removed C API is missing in NEWS
- Sorry, I mistook NEWS in ruby_2_1 for trunk...
-
- What's the version of NEWS you are seeing?
It is described since r44963.
```
% svn ann NEWS|egrep '9502|rb_check_safe_str'
44963 akr * Deprecated APIs removed. [Feature #9502]
44963 akr rb_check_safe_str -> S... -
- I found that `rb_check_safe_str()` is removed in ruby-2.1 and trunk.
And I found #9502 and [CAPI\_obsolete\_definitions](https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/CAPI_obsolete_definitions) page, but these changes is not in NE... -
- the StringIO object is not closed yet.
Reported by Jordi Massaguer Pla. [ruby-core:42538] [Bug #6010]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45835 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
04:29 PM Revision 5ab5d921 (git): * ext/tk/lib/tkextlib/tkDND/tkdnd.rb: fix typo and missing definition.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45834 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * gc.c (gc_after_sweep): suppress unnecessary expanding heap.
Tomb heap pages are freed pages here, so expanding heap is
not required.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_1@45833 b2dd03c8-39d... -
- is available to run benchmark driver on ruby 1.9.2.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45832 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
-
- I reject this issue because no one implement it until now.
It is too difficult to discuss concretely without implementation. -
- It seems glibc-2.12 provides IP_TRANSPARENT.
So there is no problem to use it now. -
- Applied in changeset r45830.
----------
* ext/socket/mkconstants.rb: Add IP_TRANSPARENT.
IP_TRANSPARENT is provieded since glibc-2.12.
Reported by Eliezer Croitoru. [ruby-core:50372] [Bug #7476] - 03:15 PM Revision da35e754 (git): * 2014-05-06
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45831 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- IP_TRANSPARENT is provieded since glibc-2.12.
Reported by Eliezer Croitoru. [ruby-core:50372] [Bug #7476]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45830 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45829 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- This issue is not acceptable because it is very inconsistent that only FIX2UINT on 64bit environment rejects negative numbers.
Changing the behavior of other functions, NUM2Uxxx and FIX2Uxxx, is too incompatible.
If people needs rigi... -
- Applied in changeset r45828.
----------
* process.c (check_exec_redirect): Open the file in write mode for
redirect from [:out, :err].
Proposed and implemented by Yusuke Endoh.
[ruby-dev:41430] [Feature #3348] -
- redirect from [:out, :err].
Proposed and implemented by Yusuke Endoh.
[ruby-dev:41430] [Feature #3348]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45828 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r45827.
----------
* ext/pathname/lib/pathname.rb (cleanpath_aggressive): make all
separators File::SEPARATOR from File::ALT_SEPARATOR.
Reported by Daniel Rikowski.
Fixed by Nobuyoshi Nakada. [Bug #9618]
* e... -
- separators File::SEPARATOR from File::ALT_SEPARATOR.
Reported by Daniel Rikowski.
Fixed by Nobuyoshi Nakada. [Bug #9618]
* ext/pathname/lib/pathname.rb (cleanpath_conservative): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby... -
- Applied in changeset ruby-trunk:r45826.
----------
* ext/pathname/lib/pathname.rb (Pathname#/): Aliased to Pathname#+.
Suggested by Alexey Muranov. [ruby-core:61432] [Feature #9625] -
- Suggested by Alexey Muranov. [ruby-core:61432] [Feature #9625]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45826 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 08:28 AM Revision a210f7c2 (git): * math.c (rb_math_sqrt): omitted exporting an unused function,
- anyway.
* internal.h: follows the above change.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45825 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45824 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
-
> どうダメなのかわからないのでなんともいいようがない、というのが私の正直な感想です。
俺にはもう関係ない事ですし、余計な事でした。
田中さんのほうで気になってなかったのなら別に構わないと思います。
-
- tadayoshi funaba wrote:
> とりあえず、'%z' のみとかはダメらしいですから。
そうですね。
Time.strptime("+09:00", "%z") も受け付けていたので、原作者の意図するものではないということで変えてみました。
> ...
どうダメなのかわからないのでなんともいいようがない、というのが私の正直な感想です。
-
- Erik Michaels-Ober wrote:
>
> ...
I'm not sure that is a good idea.
For example, -4712-01-01 is not a special date in Time.
> * Raise ArgumentError if time passed to `Time::strptime` is invalid. This fixes a Ruby bug and adds a te... -
- doesn't extract date information.
Reported by tadayoshi funaba. [ruby-core:62349]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45823 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
05/04/2014
-
- tadayoshi funaba wrote:
> 変更しました。
ありがとうございます。
> ...
要らなくなることはないと思います。義務はないのでまた手伝ってください。
> 今後は横断的にちょっかいを出したり、
> ...
今後ともよろしくお願いします。 -
10:16 PM Feature #9794: DateTime.strptime() doesn't work correctly for '%s %z'
- Well, I still don't understand what was discussed here, but it seems we have a fix (r45822).
Thanks for doing the sensible thing :) -
- 変更しました。
今すぐコミッターの活動を停止するわけではありませんが、
今日で date の担当は降ります。
現行の date の実装はこのまま使ってもいいし、要らなくなったら遠慮なく捨てて下さい。
今後は横断的にちょっかいを出したり、
行末の空白を削ったりして、のんびり余生を過そうと思います。
-
> 「対応しなかった」理由はなんとなくわかりましたし、実際の利用例がなく、単に一貫性のためとかだったらこのまま放置でよかったと思いますが。
git については過大評価じゃないかなと思います。
もうどうでもいいですが。
> ...
まつもとさんは大変だと思います (俺も原因になってるし)。
-
> Time.parse は現在時刻で補完するという仕様だったからです。
> ...
DateTime でも補完はしますけどね。
> Time.strptime('', '') もそうで、これはまつもとさんが書いた最初からですが、
> ...
とりあえず、'%z' のみとかはダメらしいですから。
ついでだけど、Time の変更がこのまま進むなら to_time とかも直したほうがいいよ。
元々無理のある仕様だったけど、もう完全にダメになりそう。
-
- 元々のリポートが「これはバグだ」、「当然対応すべきだ」という態度だったのは置いといて、実際に利便性があり、ほかに実害がない以上、機能拡張として対応したほうがいいと思います。
私が見落としてるのかもしれないけど、今までのふなばさんの主張で「対応すべきでない」理由はわかりませんでした。
「対応しなかった」理由はなんとなくわかりましたし、実際の利用例がなく、単に一貫性のためとかだったらこのまま放置でよかったと思いますが。
これを決定権がないとおっしゃるのであれば... -
- tadayoshi funaba wrote:
> Time のほうは
> ...
Time.parse は現在時刻で補完するという仕様だったからです。
そういう仕様だったので、その仕様を変えるまでは、まったく時刻情報がなくても受け付けていました。
Time.strptime('', '') もそうで、これはまつもとさんが書いた最初からですが、
r8068 をみると、Time.parse のコードを複製したところがそのままだから、
という感じかなぁ... -
- ずっと俺には何の決定権もないって言われ続けているような気がしたけど
今日は本当にそれを実感した。 -
- 私もさすがに"%z"単体を許すべきとは思いません。それ日付でも時刻でもないから。
それはそれとして、"%s %z"はgitがコミット時間をログに記録するのに使ってるフォーマットなんだそうです。
たしかに、手元の .git/logs/HEAD とかに、すべてのコミットについてこの形式でチェックイン時間とタイムゾーンが記録されている。
これくらい実際に広く使われているツールで頻繁に使われているフォーマットであれば、曜日の場合のように矛盾する情報というわけで... -
- なんか本質に近いところを突いたような気がしてきた。
DateTime は日付の核になる所から積み上げられないと、そこは無効になるんだけど、
Time にはそういうのがなくて、仮令、核が虚になってても関係ない。
俺の見方だと、そもそも Time は日付 (時刻) になってなくてもいいって事なんだと思う。
そう考えると納得出来るような気がする。
-
> そうすると、%s %z がなんでいけない理由はいまだに私にとって謎だなぁ。
少し違うかもしれないけど、
もっと基本的なパターンとして '%z' があって、
'%s %z' を受け入れる事になったら、次は '%z' だなって以前に思ったのを思い出した。
Felipe Contreras の主張だとこれも出来ないとバグでしょ。
彼は別に自分が困っててここに来てるわけじゃないからね。
DateTime.strptime ではこれは許さない事になって...-
- tadayoshi funaba wrote:
>
> ...
関係あると思われないなら私の推測がまちがっていたのでしょう。
そうすると、%s %z がなんでいけない理由はいまだに私にとって謎だなぁ。
-
- > ここで興味の対象は "%s %z" がなぜいけないのか、という点です。
> ...
質問の意図が理解しきれていないかもしれませんが、望ましくはないと思います。
ただ、それと俺の判断に関係があるのかな。
> 読み飛ばすという使い方については私には関係性が見出せませんでした。
Date.strptime('Sun May 4 00:00:00 2014', '%a %b %d %T %Y')
これは典型的な日付ですが、田中さんのいう点を打つには ... -
- tadayoshi funaba wrote:
> Akira Tanaka wrote:
> ...
うぅむ。期待したものとは違う方向で回答されてしまった気がします。
ここで興味の対象は "%s %z" がなぜいけないのか、という点です。
私の思ったのは、もし、ふなばさんが DateTime.strptime に与えるフォーマットに時刻 (時間軸上の点) を
同定するのに関係しない要素は与えるべきでない、と感じるとすれば、
フォーマットに "%s ... -
- Nobuyoshi Nakada wrote:
> It doesn't seem there is absolute reason that you have to use `Time.strptime`.
Why wouldn't I? It does **exactly** what I want:
Time.strptime('0 +0100', '%s %z').strftime('%s %z')
=> "0 +0100"
... -
- Akira Tanaka wrote:
> tadayoshi funaba wrote:
> ...
極端な話、単に読み飛ばすだけでも意味があると思ってます。
`s = 'Sunday 4 May <unknown garbage> 2014; blob'
h = Date._strptime(s, '%a %d %b')
s2 = h[:leftover].gsub(/ *<.*> */, '')
h2 = Date._strptime(s2, ... -
- It doesn't seem there is absolute reason that you have to use `Time.strptime`.
Particular methods feel more suitable, like `rfc2822`, `httpdate`, and `xml`.
~~~ruby
class Time
def self.git(s)
/\A(\d+)\s+([-+]\d\d)(\d\d)\z/ =... -
- tadayoshi funaba wrote:
>
> ...
はい。Time や DateTime のインターフェースで hour がないとかは許しがたいので、
なにかを選ばざるを得ないと思います。
> %s %z は「今x秒です」「オレ %z の時差で生活してます」という感じで
> ...
ふむ。これは示唆的です。
もしかして、DateTime.strptime に与えるフォーマットに時刻 (時間軸上の点) を同定するのに関係しない要素は与える... -
- tadayoshi funaba wrote:
> /* Gr. strptime is crap for this; it doesn't have a way to require RFC2822
> ...
This has already been agreed: many options in strptime() are not portable, but that includes '%s' as well. -
- git は良く知らないんだけど、今、git-1.9.2 を見たところでは、
少なくとも現行の核で strptime は使ってないように見えます。
自前で解析して time_t 相当のものを組み立ててるようですね。
git-1.9.2/date.c から引用:
/* Gr. strptime is crap for this; it doesn't have a way to require RFC2822
(i.e. English) day/mon... -
- Akira Tanaka wrote:
> tadayoshi funaba wrote:
> ...
しかし、それは Time の場合は自明である地方時、DateTime では UTC という事になっているし、そういう形式なんで仕方がないと思います。
そういう積極的でない時間帯の選択は俺は勝手に「匿名」と読んで許容しています。厳密にはそういうのを別に識別する仕掛けがあるほうがいいのでしょうが。
%s %z は「今x秒です」「オレ %z の時差で生活して... -
- tadayoshi funaba wrote:
>
> ...
ちゃんと定義すればそういうことも可能なんじゃないかと思います。
年月日時分秒が2つ出てくるのでわかりにくいとは思いますが。
> 俺は今、日本にいてUTCの時刻を扱っているとします。
> ...
後で出てくる時間軸の話でいえば、
ISO 8601 でいえば 2001-02-03T04:05:06Z という名前をもつ時間軸上の点に
2001-02-03T04:05:06 UTC (+0900... -
- On Sat, May 3, 2014 at 8:45 AM, <nikkoara@hates.ms> wrote:
> Issue #6634 has been updated by L Nicoara.
>
>
> L Nicoara wrote:
>> nhm tanveeer hossain khan wrote:
>> > Hi there,
>> >
>> > I've faced similar problem with ruby 2.... - 08:56 PM Revision a72f9f34 (git): * ext/date/date_core.c (rt_rewrite_frags): a new feature (not a
- bug fix) of strptime. applies offset even if the given date is
not local time (%s and %Q). This is an exceptional feature and
I do NOT recommend to use this at all. Thank you git community.
git-svn-id: svn+ssh://... -
08:44 PM Feature #9799: change behavior of Math::atan2 if y and x are both Float::INFINITY
- Marcus Stollsteimer wrote:
> Wait, wait...
> ...
I don't consider it wrong for a programming language. Maybe these special cases make less sense than the other ones, but they exist in many other languages. Even in numerical ones. That ... -
04:05 PM Feature #9799: change behavior of Math::atan2 if y and x are both Float::INFINITY
- Wait, wait...
So you introduce mathematically wrong behaviour with the argument that there already is some other wrong behaviour??? -
- I've searched for a rationale. It can be found in <http://www.open-std.org/jtc1/sc22/wg14/www/C99RationaleV5.10.pdf> on the bottom of page 188 (or real page 181):
>The specification of `atan2(∞,∞)` as `π/4` indicates the reasonable quad... -
- Yusuke Endoh wrote:
> Interesting. I'm not against the change since the proposed behavior looks prevailing, but I wonder if it is useful that the following case returns pi/4.
Yes, I wondered it too
But `Float::INFINITY == Float::IN... -
- Applied in changeset r45805.
----------
math.c: C99-like atan2
* math.c (math_atan2): return values like as expected by C99 if
both two arguments are infinity. based on the patch by cremno
phobia <cremno AT mail.ru> in [ruby-core:... -
12:55 AM Feature #9799: change behavior of Math::atan2 if y and x are both Float::INFINITY
- Interesting. I'm not against the change since the proposed behavior looks prevailing, but I wonder if it is useful that the following case returns pi/4.
~~~ruby
x = Float::INFINITY
Math.atan2(x, 2 * x) #=> Math::PI/4, not Math.atan... -
06:32 PM Feature #8667: Unable to set OpenSSL GCM iv_length in Ruby
- Alternate patch which makes iv length changes implicit in iv= when using GCM.
This issue currently requires me to require a patched ruby for my software to run. -
06:31 PM Bug #9657: Segfault in bigdecimal/util.rb Float#to_d
- Thanks chikanagasan and wanabesan!
-
- Thank you for notice this, wanabe san.
r45015 was backported into ruby_2_1 at r45815. -
- r45595 was backported into ruby_2_1 at r45821.
-
- * ext/openssl/ossl_pkey.c (ossl_pkey_verify): as EVP_VerifyFinal()
finalizes only a copy of the digest context, the context must be
cleaned up after initialization by EVP_MD_CTX_cleanup() or a
memory leak wi... -
- r45291, r45299, r45314 and r45325 were backported into ruby_2_1 at 45820.
-
- * 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:4... -
- test/ruby/envutil.rb: compare RSS to check memory leak
* 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 ... -
- r45638 was backported into ruby_2_1 at r45818.
according to #9796 -
- * gc.c (rb_gc_writebarrier): drop special case for big hash/array
[Bug #9518]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_1@45818 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- r45468 was backported into ruby_2_1 branch at r45817.
-
- * gc.c (gc_before_sweep): cap `malloc_limit' to
gc_params.malloc_limit_max. It can grow and grow with such case:
`loop{"a" * (1024 ** 2)}'
[Bug #9687]
This issue is pointed by Tim Robertson.
... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45816 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * ext/bigdecimal/bigdecimal.c (BigDecimal_initialize): Insert GC guard.
* ext/bigdecimal/bigdecimal.c (BigDecimal_global_new): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_1@45815 b2dd03c8-39d4-4d8f-98ff-... - 03:51 PM Revision e1fefb28 (git): * 2014-05-05
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45814 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45813 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * ext/psych/lib/psych.rb: Merge psych-2.0.5. bump version to
libyaml-0.1.6 for CVE-2014-2525.
* ext/psych/yaml/config.h: ditto.
* ext/psych/yaml/scanner.c: ditto.
* ext/psych/yaml/yaml_private.h: ditt... -
- * ext/psych/lib/psych/visitors/yaml_tree.rb: support dumping Encoding
objects.
* ext/psych/lib/psych/visitors/to_ruby.rb: support loading Encoding
objects.
* test/psych/test_encoding.rb: add test
... -
- * gc.c: Fix up default GC params by @csfrancis [fix GH-556]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_1@45810 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * vm.c (Init_vm_objects): initialize VM internal objects, after
heap initialized.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45809 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- expected offset from UTC.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45808 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- (Time.make_time): Use Time.force_zone!.
(Time.strptime): Ditto.
(Time.rfc2822): Ditto.
(Time.xmlschema): Ditto.
* lib/rss/rss.rb (Time.w3cdtf): Use Time.force_zone!.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45807 b2dd0... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45806 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * math.c (math_atan2): return values like as expected by C99 if
both two arguments are infinity. based on the patch by cremno
phobia <cremno AT mail.ru> in [ruby-core:62310]. [Feature #9799]
git-svn-id: svn+ssh://ci.ruby-lang.org/...
05/03/2014
-
- * test/ruby/test_math.rb (assert_float): rename with `assert_`
prefix so that the caller frame would appear in backtraces.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45804 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- It don't understand the details of the discussion, but it seems to me Tadayoshi is arguing that dates cannot be represented correctly in "%s %z".
If that was the case why did Git chose it as the format to store dates? Shouldn't Git be... -
- Akira Tanaka wrote:
> tadayoshi funaba wrote:
> ...
問題があるとかでなく、別のものなってるという事です。
たとえば、可能性としては、UTCで書いた日付を地方時にという事も出来るという事ですよね。
今回 %s だけですが。
俺は今、日本にいてUTCの時刻を扱っているとします。
それを例えば以下のように記述する事にします。
与えられた時刻はUTCだけど、それを +09:00 で参照する事にします。
... -
- tadayoshi funaba wrote:
>
> ...
はい。
> これの Sat を Fri 書き換えると間違った日付です。
はい。
"Fri May 3 00:00:00 2014" に書き換えると確かに間違っています。
> ...
曜日と年月日は独立ではないので、
片方を設定するともう一方も変化してしまう破壊的操作の列で設定を行うことをコマンド言語と表現しているのですよね。
これはわかります。
> 田中さんの説明からも... -
- Akira Tanaka wrote:
> tadayoshi funaba wrote:
> ...
伝わらないのでちょっと別の話をします。
Date.today.ctime #=> "Sat May 3 00:00:00 2014"
これは普通の日付ですよね。
これの Sat を Fri 書き換えると間違った日付です。
たしか、php のライブラリだったと思うけど、こういうのを入力すると
2014-05-02 とか 2014-05-0... -
- tadayoshi funaba wrote:
>
> ...
%s に UTC からの時差の情報が入っていないことに同意します。
しかし、Time オブジェクトを作るには時差 (あるいは地方時、UTC) を選ばないといけないので、
%z の情報を使ってもいいと思います。
> 田中さんはそこにオブジェクトを見てるから不自然とは感じないという事ですかね。
「そこにオブジェクトを見てる」というのがどのような意味なのかわかりません。
> ...... -
- Akira Tanaka wrote:
> tadayoshi funaba wrote:
> ...
見慣れないという意味ではないです。ユリウス日のようなものも日付だと思ってますし、%s 単体では特に問題に思っていません。
全然理解できないのですが、%s に時差って意味ないですよね。
田中さんはそこにオブジェクトを見てるから不自然とは感じないという事ですかね。
> もしそうだとすると、私はそのようには考えていません。
> ...
やっぱりおかしいで... -
- tadayoshi funaba wrote:
> 日付としておかしいからです。
おかしい、というのはどういう意味でしょう?
たしかに "%s %z" は人間が日常的に扱う形式ではありません。
でも、%s が過去から未来への流れの中の 1点を指定し、
%z がその 1点の表現方法を UTC からの offset という形でを指定する、
という時刻の形式だと解釈できると思います。
時刻の表現として、(閏秒の話を除いて) とくに問題があるとは... -
- At some point it would be nice to have something in English so that the international community can comment on that.
-
- Akira Tanaka wrote:
> tadayoshi funaba wrote:
> ...
日付としておかしいからです。
> Ruby の strptime は struct tm じゃなくて DateTime や Time のオブジェクトを生成し、
> ...
主張の根拠になっているのがバグだから直せだからです。
ruby 側の都合の話ではありません。
> %s と %z に対応する値を s と z として、
> ...
このあたりは... -
- tadayoshi funaba wrote:
> 言語の問題もあるんだろうけど、俺が英語に堪能でも Felipe Contreras に俺の考えを解らせるのは無理じゃないかという気がする。
私も "%s %z" がなんでそんなにいけないのか理解していません。
> ...
Ruby の strptime は struct tm じゃなくて DateTime や Time のオブジェクトを生成し、
DateTime や Time のオブジェクトは自分がどん... -
-
- 言語の問題もあるんだろうけど、俺が英語に堪能でも Felipe Contreras に俺の考えを解らせるのは無理じゃないかという気がする。
誤解のないよう少し補足しておくと、strcut tm で %s や %z を表現できない、という直接的な問題でもない。
C の時刻関数群はオブジェクト指向では書かれてないけど、データや関数の集合によってどんなモノを表現しているのか
考えないと理解できない。C の時刻関数には、まず、地方時 (あるいは局所時) と世界時の区別... -
- Ryan Davis wrote:
> I'd like to see this escalated to a neutral 3rd party (matz? ko1?) that can evaluate Charlie's code and weigh it against tadf's Japanese argument. Regardless of any final outcome, adding an English explanation would ... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45803 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45802 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
06:37 PM Misc #9801: Regexp documentation for Regexp::EXTENDED is incomplete
- I have submitted a pull request on GitHub to make this easier. (Before, I didn't know I could!)
https://github.com/ruby/ruby/pull/606 -
- The documentation for Regexp::EXTENDED, i.e. the `x` flag says that when it is used, Regexp will ignore all whitespace in the pattern. This is not completely true.
All whitespace is ignored, except for whitespace which has been escape... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45801 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Produce fixed-offset time object if appropriate.
(Time.xmlschema): Produce fixed-offset time object if appropriate.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45800 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Sorry for the very late response. I tried and read through your patch.
However, at first, I'd like to discuss the proposal itself.
## Demand
In fact, I think we can virtually implement this feature by using a Ruby code parser, ... -
03:35 PM Bug #5659: bug releasing a gem created with rails 3.1
- @vincius do you get this error on the latest release? I can't reproduce it on any of the machines I've got handy.
- 03:23 PM Revision f5c91e57 (git): * 2014-05-04
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45799 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- appropriate.
(Time.strptime): Use d[:zone] instead of d[:offset].
* lib/rss/rss.rb (Time.w3cdtf): Produce fixed-offset time object if
appropriate.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45798 b2dd03c8-39d4-4d8f-98ff-82... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45797 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
12:45 PM Bug #6634: Deadlock with join and ConditionVariable
- L Nicoara wrote:
> nhm tanveeer hossain khan wrote:
> ...
For the record, the test case is malformed. Bummer. I think the one I based it on (from khan) is malformed as well. My apologies if you spent time on it. -
- given.
Reported by tadayoshi funaba. [ruby-core:62322]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45796 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- It's worth noting that this segfault happens on not only arch, but also OSX, debian, and fedora. That makes it seem like a generic issue which is not isolated to arch.
-
- Thanks for the report! I'm looking into this now.
-
05:56 AM Bug #9800 (Closed): Ship 2.1.2 with GC_HEAP_OLDOBJECT_LIMIT_FACTOR of 1.3
- Many users are complaining about memory bloat in 2.1 series of Ruby
As denoted in https://bugs.ruby-lang.org/issues/9607 there is an excellent plan by ko1 to address this.
In the mean time I suggest we reduce GC_HEAP_OLDOBJECT_LIM... -
- Applied in changeset r45795.
----------
math.c: check domain of base argument
* math.c (math_log): check domain of base argument too. raises
Math::DomainError instead of returning NaN if the base is less
than 0, and returns NaN in... -
- `Math::log` with a negative `base` should raise a `Math::DomainError`.
~~~ruby
Math.log(-1.0) # raises Math::DomainError
Math.log( 1.0, -1.0) # doesn't (but calls `log(-1.0)` internally!)
~~~ -
- * math.c (math_log): check domain of base argument too. raises
Math::DomainError instead of returning NaN if the base is less
than 0, and returns NaN instead of -infinity if both of two
arguments are 0. [ruby-core:62309] [Bug #97... -
- * math.c (math_log): adjust prototype as method function.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45794 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- The current behavior when y and x are either negative or positive infinity is:
~~~ruby
Math.atan2(Float::INFINITY, Float::INFINITY) # raises Math::DomainError
~~~
The attached diff changes it to:
~~~ruby
Math.atan2(Float::IN... -
- Ask debugger gem developers.