Activity
From 03/13/2014 to 03/19/2014
03/19/2014
-
08:24 PM Bug #9653: gc_sweep(): unknown data type 0x0
- Thanks for your advice. Yes, I will try ruby 1.9.3-p545, and post the findings.
Eric Wong wrote:
> hongbin034@gmail.com wrote:
> ...
-
06:41 PM Bug #9653: gc_sweep(): unknown data type 0x0
- hongbin034@gmail.com wrote:
> * ruby -v: ruby 1.9.3p125 (2012-02-16 revision 34643) [x86_64-linux]
Hi, can you try Ruby 1.9.3-p545? (And if possible, try 2.1.1)
Early versions of ruby 1.9.3 had a bad lazy sweeping bug which
affec... -
- The error message:
/usr/local/lib/ruby/1.9.1/fileutils.rb:828: [BUG] gc_sweep(): unknown data type 0x0(0x54fd1e8)
ruby 1.9.3p125 (2012-02-16 revision 34643) [x86_64-linux]
This error is not deterministic. It occurred about 5% in my ... -
- normalperson@yhbt.net wrote:
> 0001-doubly-linked-list-from-ccan-to-manage-vm-living_thr.patch (68.1 KB)
I'll de-duplicate the CC0 declaration files if allowed to commit this.
The original had symlinks, but I assume symlinks are no... -
- Thanks all, committed as r45364
Sorry my original patch set the ivar on the wrong object :x
But fortunately I was able to reproduce the GC error and test
for it. -
08:19 AM Bug #9652 (Closed): Marshal doesn't dump/load Time#zone correctly with too many time object - Applied in changeset r45364.
----------
time.c: freeze and preserve marshal-loaded time zone
We need to prevent vtm.zone from pointing to a GC-ed string buffer.
The rb_copy_generic_ivar call misses it because get_attr deleted it.
Thank... -
- normalperson@yhbt.net wrote:
> nobu@ruby-lang.org wrote:
> > Seems better to call `rb_str_new_frozen()` before setting `vtm.zone`.
>
> OK. Though rb_fstring() might be better because time objects
> with identical zones are c... -
- nobu@ruby-lang.org wrote:
> Seems better to call `rb_str_new_frozen()` before setting `vtm.zone`.
OK. Though rb_fstring() might be better because time objects
with identical zones are common. -
07:37 AM Bug #9652: Marshal doesn't dump/load Time#zone correctly with too many time object
- Seems better to call `rb_str_new_frozen()` before setting `vtm.zone`.
-
- I cannot reproduce the problem here, however it could be the
lack of reference for GC.
Can you add this `rb_ivar_set` call?
~~~diff
--- a/time.c
+++ b/time.c
@@ -4828,6 +4828,7 @@ end_submicro: ;
}
if (!NIL_P(zone... -
06:54 AM Bug #9652 (Closed): Marshal doesn't dump/load Time#zone correctly with too many time object
- Hi, there
I wrote a script to test Marshal dump/load with many time object ( see attachment ).
basic idea is:
1) A is container of time, it contains 100,000 time instance (all Time.now).
2) dump/load A by using Marshal, expected... - 08:19 AM Revision e0a0edf5 (git): time.c: freeze and preserve marshal-loaded time zone
- We need to prevent vtm.zone from pointing to a GC-ed string buffer.
The rb_copy_generic_ivar call misses it because get_attr deleted it.
Thanks to nobu for the rb_str_new_frozen suggestion.
* time.c (time_mload): freeze and preserve mar... -
- [GH-561] <https://github.com/ruby/ruby/pull/561> ?
- 12:42 AM Revision 5f00b999 (git): * 2014-03-19
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45363 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * .gdbinit (rb_method_entry, rb_ancestors): `super` moved to
RClass from rb_classext_t since r44294.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45362 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
03/18/2014
-
08:50 PM Feature #9347: Accept non callable argument to detect
- I agree, an optional argument would be best. I'll propose a patch. Could we use `if_none` though?
-
08:22 PM Bug #9605: Chaining "each_with_index.detect &lambda" raises ArgumentError
- @nobu I meant to tell you how honored I am that you and matz are taking so much time talking about this issue and even considering changing this small corner of ruby itself. The idea of changing the language because of something I notice...
-
- funny.falcon@gmail.com wrote:
> Can you test this
> + if (SYMBOL_P(a)) a = (a >> RUBY_SPECIAL_SHIFT) ^ (a >>
> (RUBY_SPECIAL_SHIFT + 3)); /* 3=ID_SCOPE_SHIFT */
Seems to hurt performance on x86 compared to my original. I don'... -
09:48 AM Feature #9425: [PATCH] st: use power-of-two sizes to avoid slow modulo ops
- Can you test this
+ if (SYMBOL_P(a)) a = (a >> RUBY_SPECIAL_SHIFT) ^ (a >>
(RUBY_SPECIAL_SHIFT + 3)); /* 3=ID_SCOPE_SHIFT */
2014-03-18 13:02 GMT+04:00 Eric Wong <normalperson@yhbt.net>:
> normalperson@yhbt.net wrote:
> > ... -
- normalperson@yhbt.net wrote:
> hash_aref_sym 1.000
Lack of improvement here was disappointing since symbol keys are common,
and this showed a regression on my x86 (32-bit) VMs.
I tweaked rb_any_hash to be symbol-aware:
htt... -
06:27 PM Feature #5663: Combined map/select method
- I came to think that, if this feature is to be added, then are other methods to be considered: `flat_map` and `to_h` in addition to `map`. I think it should be either the case that the corresponding counterpart to all of these methods sh...
-
02:46 PM Bug #9651 (Closed): __dir__ returns nil in instance_eval/module_eval
- `instance_eval`/`module_eval` 内の `__dir__` が常に `nil` を返します。
`eval` は r42103 で修正済みでしたが `instance_eval`/`module_eval` では `nil` のままだったようです。
trunk では r45360 および r45361 で修正済みです。
Backport 管理用に改めてチケット化しておきます。 -
02:22 PM Revision 524817bb (git): * vm_eval.c (eval_string_with_cref): Unify to use NIL_P.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45361 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- given. Related to [ruby-core:56099] [Bug #8662] and r42103.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45360 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 09:28 AM Revision 9d1279b8 (git): * 2014-03-18
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45359 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
09:28 AM Revision 7088bef3 (git): fix a typo [ci skip]
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45358 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Updated.
~~~diff
pathname.rb: separators
* ext/pathname/lib/pathname.rb (cleanpath_aggressive): make all
separators File::SEPARATOR from File::ALT_SEPARATOR.
[ruby-core:61402] [Bug #9618]
* ext... -
05:02 AM Feature #9647: File::Stat#birthtimeの追加
- Yui NARUSE wrote:
> Windowsの方はなるほど…なので、とりあえずWindowsもNotImpErrorにしてうささんに実装してもらいますかね（ゎ
いやあの、Windowsはそれでいいので付けといてください(笑)
今のctimeのことを考えると、多少の意味の違いなんかどうでもいい感があります。 -
04:53 AM Feature #9647: File::Stat#birthtimeの追加
- Motohiro KOSAKI wrote:
> Linuxの場合、媒体にはbirthtimeが記録されていて、syscallの追加が自転車置き場の議論で進まないという感じなのでctimeが帰ってくるのはうれしくないなあ。将来的に混乱が予想されるので。
> ...
はい。
なもんでLinuxが実装するの待っていたんですが、もう見捨てるという結論にしました（どーん
Windowsの方はなるほど…なので、とりあえずWindowsもNotImpErrorにしてうさ... -
- NTFSでは$STANDARD\_INFOMATIONと$FILE\_NAMEのそれぞれに日時系メタデータが入っていますが、後者を変更するAPIは存在しないので、たぶんそれのことかと思います。
しかし、「真のファイル作成日」というよりは、そのファイル名の名付け日(例えばハードリンクを作ったならその日時)なので、ちょっと意味合いは違いますね。 -
12:01 AM Feature #9613: Warn about unsafe ossl ciphers
- Single datapoint: r45274 will likely end up in Debian jessie's ruby 2.1, and by extension probably in Ubuntu's ruby 2.1.
03/17/2014
-
08:39 PM Feature #9647: File::Stat#birthtimeの追加
- 更に雑談をつづけてしまうと Windowsには file system tunnelling という機能があり、われわれが birthtimeという文脈で
話すファイル作成日とは若干セマンティクスが違う
http://stackoverflow.com/questions/661977/why-windows-sets-new-created-files-created-time-property-to-old-time
ので、名前が birthtime ... -
- Linuxの場合、媒体にはbirthtimeが記録されていて、syscallの追加が自転車置き場の議論で進まないという感じなのでctimeが帰ってくるのはうれしくないなあ。将来的に混乱が予想されるので。
いまでもrootならdebugfs使って作成日とれます。
この辺参照してください（一番下） http://stackoverflow.com/questions/5929419/how-to-get-file-creation-date-in-linux
... -
- File::Stat#birthtimeを追加しませんか。
以下の様なシステムの stat(2) には st_birthtimespec があり、
(ctime = change time ではなく) ファイルを作成した日時を得ることができます。
http://netbsd.gw.com/cgi-bin/man-cgi?stat+2+NetBSD-current
http://www.freebsd.org/cgi/man.cgi?query=stat&se... -
08:05 PM Bug #9521: [Doc] Fix error in Time.parse documentation (in lib/time)
- @zzak I suggest adding the return value of Time.parse("Thu ...") to make the examples a little more clear:
```
now = Time.parse("Thu Nov 29 14:33:20 GMT 2001") #=> 2001-11-29 09:33:20 -0500
```
(Or correspondingly for JST.)
@... -
- I agree, the behaviour of Time.parse *is* confusing, at least when different time zones are involved.
It can be understood if you consider these points:
1\. Time.parse assumes local time for all time specifications, unless explicit... -
- Marcus Stollsteimer wrote:
> @zak Just to consider: I used EST because
> ...
TZ environment variable must be used like
irb(main):001:0> require"time"
=> true
irb(main):002:0> ENV["TZ"]="EST+5"
=> "EST+5"
irb(main):003:0> Time.pa... -
07:05 PM Bug #8358: TestSprintf#test_float test failure
- I maintain ruby package in Linux Arch and our users reported an issue with it https://bugs.archlinux.org/task/39470
We compile 32-bit binaries for old machines (PentiumPro+ CPU). The old CPUs do not have SSE2 support, but we've found ... -
- It would probably cause a lot of breakage.
More importantly, why would it be an improvement? What's wrong with using `String#bytesize` in the rare case where you actually want the byte length? - 04:01 PM Feature #9650 (Rejected): String#size versus String#length
- Hi guys !
A quick idea for the next major version :
String#size => bytes length
String#length => chars length
Could this be a good move or too much BC break ? -
10:27 AM Bug #9648 (Closed): Dir.glob('../*') shows unexpected result in a symbolic linked directory on Windows
- `Dir.glob('../*')` shows unexpected result if current directory is a symbolic linked directory on Windows.
This issue seems to be introduced by r44802.
The following is a test case.
~~~
C:\foo>mklink /D bar V:\
symbolic link c... -
- * process.c (rb_execarg_new, rb_execarg_init): constify argv.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45357 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
07:42 AM Bug #4008: Rails 3::declarative_authorization::authorization_rules request => segvault ruby 1.9.2
- I am concerned about what Nahi said regarding infinite recursion handling. Also, after 2.0 Koichi asked for reproduce on trunk which makes me think he anticipates this issue > 1.9.2.
I'll leave this ticket for ko1 to determine whether... -
- Seems stack overflow, probably.
1.9.2 has died long ago, and there are many changes for it.
I think there is no chance for new release.
-
04:59 AM Bug #9646 (Closed): Infinite loop at Hash#each
- Hashでキーを同じにすると、無限ループしてしまいます。
再現コードは
ruby -e 'h={};h[a=[]]=1;a<<1;h[[]] = 2;a.clear;h.each{|i| p i}'
です。
`st_foreach_check`で`callcc`対応のために
`find_packed_index`を毎回呼んでおり、そこで`i`が進まなくなっています。
そもそも、同じキーの状態を作り出せるのがどうなんだという話はあるかもしれませんが、
... -
- * ext/json/parser/prereq.mk (parser.c): use `enum` instead of
`static const int` to get rid of unused-const-variable warnings.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * complex.c (nucomp_to_s): use rb_String.
* complex.c (nucomp_inspect): use rb_inspect.
* complex.c: use PRIsVALUE flag not to use an intermediate string
which can be collected by GC.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trun... -
- * complex.c, rational.c: remove unused functions, which are warned
by clang 5.1, and also variables only used by removed functions.
* ext/date/date_core.c: ditto.
* enc/utf_16be.c, enc/utf_16le.c: comment out constants only used
by... -
04:20 AM Revision 57e55790 (git): * enumerator.c (enumerator_block_call): use PARRAY_CONST_PTR()
- instead of RARRAY_PTR().
* io.c (rb_io_s_popen): ditto.
* numeric.c (num_step_size): ditto.
* vm_eval.c (rb_apply): ditto.
* vm_eval.c (rb_eval_cmd): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45353 b2dd03c8-39d4-4d8f-98... -
- Eric Wong wrote:
> Running this (on top of current trunk) to serve my (mostly static sites)
> ...
Thanks!
> I noticed vm1_gc_short_with_complex_long got very slow with this
> ...
OMG. Thank you for reporting.
Yes. It will do ... -
- * configure.in (os_version_style): get default style from
`/usr/bin/ruby` if possible.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45352 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Seems no wildcard tests.
- 01:24 AM Bug #9645 (Closed): [PATCH] variable.c: avoid memory leak on const redefinition
- Applied in changeset r45350.
----------
variable.c: avoid memory leak on const redefinition
* variable.c (rb_const_set): delete existing entry on redefinition
[Bug #9645]
* test/ruby/test_const.rb (test_redefinition): test for leak -
- I noticed this leak because I was making constants use ihash from Feature 9614.
I hope nobody is really affected by this leak.
- 01:24 AM Revision c7e88f76 (git): * 2014-03-17
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45351 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 01:24 AM Revision 3e6fe414 (git): variable.c: avoid memory leak on const redefinition
- * variable.c (rb_const_set): delete existing entry on redefinition
[Bug #9645]
* test/ruby/test_const.rb (test_redefinition): test for leak
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45350 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
01:23 AM Bug #7492: Segmentation fault at DL::TestDL#test_call_double on x64 Windows 8
- I uploaded the patch for the trunk.
03/16/2014
-
10:46 PM Bug #9644 (Closed): ssl hostname verification security bug: verify_certificate_identity wildcard matching allows to much
- Hi,
I'm not a ruby developer but the maintainer of the IO::Socket::SSL module in Perl. While comparing the state of the SSL implementations in various languages I've noticed, that your validation of the hostname inside the certificate i... -
- ko1@atdot.net wrote:
> File gc.patch added
Running this (on top of current trunk) to serve my (mostly static sites)
on yhbt.net. Memory usage seems stable at ~31M (from ~49M)
I noticed vm1_gc_short_with_complex_long got very s... -
- You raise a good point, I was only trying to avoid extra work :)
To be honest, these examples are confusing.. I'm not sure what the point of passing tz is, for example:
`now = Time.parse("Thu Nov 29 14:33:20 GMT 2001")`
Is clea... -
- @zak Just to consider: I used EST because
* the docs for `Time` state: _All of these examples were done using the EST timezone which is GMT-5._
* AFAIK it's not easily possible to set JST on the local machine (via environment variabl... -
- Applied in changeset r45349.
----------
* lib/time.rb: [DOC] Fix timezone in example of Time.parse [Bug #9521]
Based on patch by @stomar [ci skip] -
- @heesob Could you reupload your patch?
It seems the RTE migration ate your old patch :( -
- No feedback for over a year, should we close this?
-
-
02:59 AM Feature #9642 (Third Party's Issue): Allow CCM as an authenticated encryption mode in OpenSSL::Cipher
- Support for GCM authenticated encryption mode has already been introduced (https://bugs.ruby-lang.org/issues/6980#change-32884). However, CCM is not currently supported as an authenticated encryption mode.
This would require changes ... -
- Based on patch by @stomar [ci skip]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45349 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
11:09 AM Feature #9625: Define `Pathname#/` to append to the path
- I am not sure, however, if the type conversion `.to_s` is a good idea: i think it will be easier to remember if `#/` will be just a synonyme of `#join`, and would help to catch some bugs early.
-
10:15 AM Bug #9505: Bug that should cause SystemStackError segfaults under Ruby 2.1
- I can reproduce almost exactly this issue on 2.1.1p~~97~~ p76. Full details and steps I used to reproduce are in https://github.com/danielfone/rspec-memoization-segfault
-
10:00 AM Bug #9624: Unrecognized command-line option (CLANG/Xcode 5.1) "-multiply_definedsuppress" prevents installation of gems / ruby on OS X
- Mika Mannermaa wrote:
> Yup, installed rbenv and compiled ruby 2.1.1 from source.
> ...
And just to confirm with 2.0.0-p247:
$ ruby -e 'p RbConfig::CONFIG.values_at(*%w[GCC DLDFLAGS])'
["yes", "-Wl,-undefined,dynamic_lookup... -
- Nobuyoshi Nakada wrote:
> Mika Mannermaa wrote:
> ...
Yup, installed rbenv and compiled ruby 2.1.1 from source.
$ ruby -e 'p RbConfig::CONFIG.values_at(*%w[GCC DLDFLAGS])'
["yes", "-Wl,-undefined,dynamic_lookup -Wl,-multipl... -
- Mika Mannermaa wrote:
> Please note that the ruby is the system ruby compiled and installed by Apple before the clang version was updated.
> ...
Thank you, I could confirm it, but when configured from the p247 source:
$ grep -w ... -
- Mika Mannermaa wrote:
>
> ...
Downloaded the tar files for ruby-2.0.0-p451 and ran the configure script, the following line is from the generated Makefile:
DLDFLAGS = -Wl,-undefined,dynamic_lookup -Wl,-multiply_defined,suppress... -
- Nobuyoshi Nakada wrote:
> I might have misread this issue.
> ...
$ ruby -e 'p RbConfig::CONFIG.values_at(*%w[GCC DLDFLAGS])'
["", "-undefineddynamic_lookup -multiply_definedsuppress"]
$ clang -v
Apple LLVM version ... -
-
- Applied in changeset r45348.
----------
configure.in: insert a space for non-GCC
* configure.in (DLDFLAGS): insert a space between option and its
argument for non-GCC compilers. [ruby-core:61429] [Bug #9624] -
- I might have misread this issue.
Doesn't `DLDFLAGS` in rbconfig.rb start with "-Wl,"?
Then it means CLANG 5.1 is not detected as GCC.
Could you show the result of:
ruby -e 'p RbConfig::CONFIG.values_at(*%w[GCC DLDFLAGS])'
-
- If `build` never accept `-C` option, just `"-r#{siteconf.path}"` is enough.
-
- Seems it is caused by `Gem::Ext::ExtConfBuilder.build`.
It insert `-r#{siteconf_path}` to `RUBYOPT`, and the path is cwd, which is under the installed gem directory.
I'd suggest to use just base name in `RUBYOPT` and pass cwd by `RUB... -
06:00 AM Bug #9643 (Closed): error on install dir contain space .
- invalid switch in RUBYOPT: -F (RuntimeError)
error on install dir contain space .
http://stackoverflow.com/questions/16898286/error-invalid-switch-in-rubyopt-f-runtimeerror-is-shown-while-install-gems -
- * configure.in (DLDFLAGS): insert a space between option and its
argument for non-GCC compilers. [ruby-core:61429] [Bug #9624]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45348 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
03/15/2014
- 11:06 PM Revision f5c2e89c (git): gc.c (objspace_xcalloc): fix GC accounting
- This hopefully works on all platforms with malloc_usable_size.
This may also trigger bugs in places which did not expect GC, too;
so maybe some existing code will need RB_GC_GUARD.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45347... -
- address. Therefore noreturn function with tail call's return
address may be in another function.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45346 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r45344.
----------
* lib/gserver.rb: [DOC] Fixed typo in example by @stomar [Bug #9543] [ci skip] - 08:52 PM Revision cfab6e0f (git): * 2014-03-16
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45345 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45344 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
08:17 PM Bug #9544: Ruby resolver not using autoport
- Just a little follow-up:
I still think that OS should take care of that, this is a wrong layer for me.
For instance: if a system is trying to choose a free port - either random or sequential - it will randomize from a pool of *free... -
04:49 PM Feature #9585: Add Object#in? to make ruby easier to read
- This probably would not be a big addition and not bother many.
I think it can be readable:
array = [1,2,3,4,5]
if 3.in? array
puts 'Yup, is in the array!'
end
I think one problem may be that it is on a Fixnum here t... -
- * ext/.document: remove refinement from documentable directories.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_0_0@45343 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
02:41 PM Bug #9617: Windows7 issue with Kernel.spawn close_others => false
- Documentation patch is welcome.
-
- I like normalization in cleanpath only.
I feel normalization in initialize is too drastic.
I feel mixed separators is not clean, so it is natural to normalize in cleanpath. -
- I feel this proposal is acceptable.
-
12:45 PM Bug #9641 (Closed): Digest libraries are built incorrectly due to ambiguous location of "extconf.h"
- This is essentially the same issue as bug #3231, "Digest Does Not Build", which has reappeared for me on CentOS 6.5 using gcc 4.4.7.
The attached patch changes the generated Makefiles so they explicitly specify the location of `extcon... -
-
11:19 AM Bug #9636: Fiddle::Importer で Segmentation fault が発生する
- Nobuyoshi Nakada wrote:
> Integerからポインタへはそのまま渡されます。
> ...
ご指摘ありがとうございます。
ご指摘の通り、修正した結果。正常に動作しました。
お騒がせをしてしまい、すみませんでした。
ご指摘ありがとうございました。 -
- Integerからポインタへはそのまま渡されます。
Array#packでStringにしないと、参照渡しにはできません。
~~~ruby
result = @@synthe.call(input, ispeed, size = 0, phont) # ←このコードが実行された時点でコアダンプ
~~~
を
~~~ruby
size = [0].pack('i!')
result = @@synthe.call(... -
09:56 AM Revision 1cde3077 (git): * ext/.document: remove refinement from documentable directories.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45342 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
04:21 AM Feature #9634: [PATCH]Symbol GC
- Eric Wong wrote:
> volatile is not always enough, and tends to generate bad code.
It make sense for me.
I've removed the volatile declaration of rb_check_id_without_pindown.
https://github.com/authorNari/ruby/commit/5d5f9a63cc0594... -
- * vm_method.c (rb_add_method, rb_attr): cast between pointer and
ID via VALUE.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45341 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * configure.in (DLDFLAGS): fix typo, missing comma.
[ruby-core:61429] [Bug #9624]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45340 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- It needs architecture dependent header files directory.
Seems an issue in `Makefile` generated by swig. -
- Applied in changeset r45339.
----------
configure.in: check symbol resolution options
* configure.in (DLDFLAGS): check for each options to control
symbol resolution. [ruby-core:61429] [Bug #9624] -
- * configure.in (DLDFLAGS): check for each options to control
symbol resolution. [ruby-core:61429] [Bug #9624]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45339 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * gc.c (rb_obj_id): remove unused rdoc.
* object.c (rb_obj_hash): [DOC] merge unused rdoc from rb_obj_id()
in gc.c.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45338 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
03/14/2014
-
11:02 PM Feature #9047: Alternate hash key syntax for symbols
- Looks like there is a patch already for this:
https://bugs.ruby-lang.org/issues/4276
- 10:26 PM Revision 1784c9a0 (git): * 2014-03-15
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45337 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 10:25 PM Revision b50780ea (git): * st.c (st_update): remove unnecessary assignment
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45336 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
09:53 PM Bug #9639 (Rejected): libyaml not found, rubygems not functional on freebsd with ruby 2.1.0
- Please see the full report here: https://github.com/sstephenson/ruby-build/issues/509
Note that ruby -v is not functional either per below:
~~~
stealth% ruby -v
/home/me/.rbenv/versions/2.1.0/lib/ruby/2.1.0/yaml.rb:4:in `<top (re... -
08:44 PM Bug #9629: GC::Profiler.total_time under-reports GC time compared to dtrace GC probe measurement
- One minor follow-up: it's actually not that getrusage takes a 'long' time (relative to the cost of each lazy sweep invocation), it's the dtrace probes themselves firing. The conclusion remains the same, though GC::Profiler seems correct.
-
- sparse array is described in ruby-core:55079
-
- This should allow better use of cache-friendly lookup mechanisms such as
funny_falcon's sparse array in [ruby-core:55079]
Also limits symbol space to prevent OOM.
Some structs may also be made smaller as a result (rb_method_entry_... -
06:14 PM Bug #9637: File "ruby.h" includes "config.h" which is not in proper location
- I am afraid this is by design. On Fedora, we have overcome this issue by adding header [1] and renaming the original file(s), but I know that upstream is not happy about this ([2], my 4th point) and they will suggest you to use mkmf to s...
-
05:36 PM Bug #9637 (Third Party's Issue): File "ruby.h" includes "config.h" which is not in proper location
- In Ruby 2.0.0p451, I using SWIG to convert some C/C++ code into Ruby. However, when using GCC to compile the C++ code into Ruby and pointing GCC to the "ruby.h" file (i.e. gcc -fPIC -c test_wrap.cxx -IC:\Ruby200-x64\include\ruby-2.0.0), ...
-
- Fiddle::Impoter で、 Segmentation fault が発生します。
Ruby 1.8.7-p375
Ruby 1.9.3-p545
Ruby 2.1.1-p76
いずれのバージョンでも、同様のコアダンプを発生させています。
Ruby 2.1.1-p76 で実行時のコアダンプを添付します。
実行環境は
Ubuntu 13.10 x86_64 です。CPU の Byte Order は Lit... -
-
- @naruse could you verify that r44896 fixed this on 32bit linux?
-
04:08 PM Bug #9624 (Assigned): Unrecognized command-line option (CLANG/Xcode 5.1) "-multiply_definedsuppress" prevents installation of gems / ruby on OS X
-
- Narihiro Nakamura wrote:
> Thank you! I fixed it and rebased.
> ...
New symbol_gc branch works fine. Thanks! -
- Cool! I benchmarked your original version and it didn't notice obvious
regressions.
I noticed rb_check_id_without_pindown still takes a volatile arg. Is
this for GC-safety? Can we encourage RB_GC_GUARD instead for new APIs?
vol... -
- Kazuki Tsujimoto wrote:
> `make test-all` sometimes causes segmentation fault.
> ...
Thank you! I fixed it and rebased.
https://github.com/authorNari/ruby/commit/9cd060aab6ca9cf55971b8d8881b30f0204f71be
https://github.com/authorNar... -
- symbol table if it is built with cc -g and not stripped.
Now ruby can show static symbols on Linux though glibc's
backtrace_symbols(3) don't show them.
* addr2line.c (rb_dump_backtrace_with_lines): use dladdr(3) to
detect what obj... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45334 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * misc/rdoc-mode.el (rdoc-fill-paragraph): fill indented block by
list.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45333 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- of backslash as well, patch by @idupree [Fixes GH-553] [ci skip]
https://github.com/ruby/ruby/pull/553
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45332 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- おおっと、間違えて変な変更をSubmitしてしまいました。無視してください
-
- Shugo Maeda wrote:
> hsbt (Hiroshi SHIBATA) wrote:
> ...
-
02:08 AM Feature #5825: Sweet instance var assignment in the object initializer
- Does anyone have a link to discussions/logs that lead to the decision to remove instance/global variables from block parameters?
-
- Tyler Rick wrote:
> I think CoffeeScript solved this problem quite nicely. Many constructors in CoffeeScript end up being beautiful, simple one-liners!
> ...
Sorry, it doesn look beautiful to me.
-
- * range.c (SET_EXCL): remove no longer used function.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45331 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * bignum.c (maxpow{16,32,64,128}_{exp,num}): adjust preprocessor
condtions to maxpow_in_bdigit_dbl().
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45330 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- add links to `Object#hash` to each #`hash` methods rdocs.
[Fixes GH-567]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45329 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * string.c (rb_str_hash_m): [DOC] hash value depends on the
encoding too.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45328 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
-
12:08 AM Feature #8572: Fiber should be a Enumerable
- Koichi Sasada wrote:
> Can I close it?
Sorry for delay. Close this please.
03/13/2014
-
- I think I've figured out the discrepancy here: the dtrace probes wrap around the getrusage(2) calls that GC::Profiler bases its timings on for Mac OS X. The average lazy sweep time is quite short (single-digit microseconds per lazy sweep...
-
- I realized that these are actually measuring different things: dtrace's timestamps measure wall clock time, whereas GC::Profiler on Mac OS X uses getrusage, which measures user CPU time. It still seems weird that the two would be so dive...
-
- Worth noting: a quick read through gc.c suggested that the missing GC sweep time might be due to me not having built with GC_PROFILE_MORE_DETAIL, but even after building with that, I still see the same discrepancy where GC::Profiler is s...
-
- Uploading the two files from that gist, just to keep everything in one place.
-
- I'm trying to square the numbers that I'm getting from GC::Profiler.total_time against those that I'm getting out of instrumentation with the GC dtrace probes embedded in Ruby, and having a hard time getting the two sources to agree.
... -
09:29 PM Bug #9573: descendants of a module don't gain its future ancestors, but descendants of a class, do
- Jeremy Evans wrote:
>
> ...
Conceptually Ruby does have multiple inheritance, an object `is_a?`(all included modules).
Is MRI's iclass snapshotting an implementation detail? Can it theoretically be done differently (e.g. tree that you... -
08:34 PM Bug #9573: descendants of a module don't gain its future ancestors, but descendants of a class, do
- First Last wrote:
> What is the objection to explaining how this works?
nobu explained how it works. However, as he is not a native English speaker, let me attempt to clarify.
In ruby, there exist pseudo-copies of modules called iclas... -
- First Last wrote:
> First Last wrote:
> ...
What is the objection to explaining how this works?
-
09:08 PM Bug #9627: SMTP does not properly dot-stuff an unterminated last line
- If I'm understanding this properly I think this monkey-patch is all it takes to fix it:
```
module Net
class InternetMessageIO
def using_each_crlf_line
@wbuf = ''
yield
if not @wbuf.empty? # untermina... -
09:07 PM Bug #9594: Segfault in 2.1.1 follows cont from debugger (Rails 4.0.3, debugger 1.6.6, rspec, os x mavericks 1.9.2)
- This is also happening to me when using debugger
Ruby 2.1.0, Mavericks 10.9.3 (Build 13D12), rails 4.0.3, debugger 1.6.6, rspec-rails 2.14.1
/Users/Adrian/.rvm/gems/ruby-2.1.0/gems/activerecord-4.0.3/lib/active_record/attribute_metho... -
07:10 PM Feature #5825: Sweet instance var assignment in the object initializer
- I would *love* to see this feature in Ruby. Assigning an argument to an instance variable in a constructor is something that we do in *almost every constructor we write*, so I think this should be made as easy and simple as possible, by...
-
- Yes, it's a very old bug.
Optional arguments have same semantics as assignments. -
05:52 PM Bug #9593: Keyword arguments default argument assignment behaviour not consistent with optional argument
- Nobu, are you *sure* this is a bug? This test has worked since 1.9.3, and this code was committed in 2008:
https://github.com/rails/rails/blob/master/activesupport/lib/active_support/values/time_zone.rb#L285
It's been the behavio... -
- Ok! I will fix the code in Rails. Thank you!
-
- Aaron Patterson wrote:
> Was this expected to break?
Yes. It was a bug.
An assignment creates a variable and it hides same name method in its RHS, as you can't call `foo` method inside `bar` method:
~~~ruby
def foo; 'foo'; en... -
- * vm.c (invoke_block_from_c): add splattable argument.
* vm.c (vm_invoke_proc): disallow to splat when directly invoked.
* vm_insnhelper.c (vm_callee_setup_arg_complex, vm_callee_setup_arg):
relax arity check of yielded lambda. [ruby-... - 04:18 PM Revision 2deb6e8e (git): * 2014-03-14
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45326 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * test/dl/test_{cptr,handle}.rb, test/fiddle/test_{handle,pointer}.rb:
relax memory leak criteria.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45325 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- If I understand correctly what you're saying is that it would behave as below.
~~~ruby
plus = ->(x,y) {puts x+y}
~~~
1. This would continue to work as it does today
~~~ruby
def test
yield 1,2
end
test... -
- Today, matz and I chatted about this issue, and he decided to relax the arity check of lambda blocks.
Iff:
* just one argument is yielded
* it is an array, and
* its length is same as the number of the formal argument
the argume... -
- I created a jruby issue https://github.com/jruby/jruby/issues/1559 to track the odd "lambda" vs "->" difference.
-
- Finally found another implementation that behaves identically to MRI. RubyMotion acts the same with "each_with_index.map" accepting 2 args while "each_with_index.detect" raises an ArgumentError.
~~~
$ rake
Build ./build/iPhoneS... -
- A little more digging and I found Rubinius is consistent with the lambda and -> syntax but neither raises the ArgumentError. It seems like they remove the arity check when the lambda is converted to a block
~~~
$ ruby -v -e '[1].each... -
- Wow this gets weirder and weirder. It seems to be happening when jruby turns a lambda created with "->" syntax into a block. MRI does consistently raise ArgumentError in all 4 cases.
$ jruby -v -e 'def test(l); l.call(1); end; test(l... -
- Hmmm, in JRuby, `->(){}` and `lambda{}` are different?
~~~
$ jruby -v -e '[1].each(&lambda{p :ng})'
jruby 1.7.4 (1.9.3p392) 2014-03-13 fffffff on Java HotSpot(TM) 64-Bit Server VM 1.6.0_65-b14-462-11M4609 [darwin-x86_64]
:ng
$ j... -
- Ruby raises `ArgumentError` at arity mismatch when yielding to a lambda, as `test/ruby/test_lambda.rb`.
```sh
$ ruby -v -e '[1].each(&lambda{p :ng})'
ruby 2.2.0dev (2014-03-09 trunk 45302) [universal.x86_64-darwin13.0]
-e:1:in `blo... - 02:52 PM Revision 8851056b (git): * 2014-03-13
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45324 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 02:52 PM Revision 1df0e183 (git): * ext/-test-/win32/dln/libdlntest.c (dlntest_ordinal): no need to
- specify export in the source file because .def file do it.
get rid of warning on linking.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@45323 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- `make test-all` sometimes causes segmentation fault.
I attached the backtrace log.
-
09:41 AM Feature #9634: [PATCH]Symbol GC
- Wow, great work! Congrats :-)
-
- I've written a patch to collect most symbols.
PATCH: https://github.com/authorNari/ruby/compare/4a91fb7a45f0e3c...symbol_gc.patch
## Summary
* Most symbols in Ruby level are GC-able（generated by #to_sym, #intern, etc..）
* Exclu... -
- At the risk of being terse, https://bugs.ruby-lang.org/issues/7793
-
12:59 PM Feature #9635: Map a hash directly to a hash
- Ok, what i suggest is not a valid Ruby syntax. It should be
~~~
.hashy_map { |key, value| {key => value} }
~~~
which is hardly sweeter than
~~~
.map { |key, value| [key, value] }.to_h
~~~
and also goes against common pra... -
- There should be a way to directly map a hash to a hash.
The current way to do it is not very graceful:
~~~
{foo: 'foo', bar: 'bar'}.map { |key, value| [key.upcase, value.reverse] }.to_h # => { :FOO => 'oof', :BAR => 'rab' }
~... -
09:11 AM Bug #9618: Pathname#cleanpath creates mixed path separators
- No problem, I'm glad to help.
To be honest I'm a little concerned that my report of such a minor problem might result in such a drastic change.
But lets hope for the best :)
-
- すみません、 ruby-devに投げちゃいました。
#9634 移動します…。 -
- I've written a patch to collect most symbols.
PATCH: https://github.com/authorNari/ruby/compare/13834fb3c...symbol_gc.patch
## Summary
* Most symbols in Ruby level are GC-able（generated by #to_sym, #intern, etc..）
* Exclude a s... -
- This imports the ccan linked-list (BSD-MIT licensed version of the Linux kernel
linked list). I cut out some of the unused str* code (only for debugging),
but it's still a big import of new code. Modifications to existing code is
mi... -
- Hello,
Thank you for your report.
I can shrink the reproducible script.
And I've confirmed that this issue was reproduced only with 2.1.x. It seems already fixed on trunk.
~~~
class MyError < StandardError
def to_s
messa...