Activity
From 10/05/2014 to 10/11/2014
10/11/2014
-
09:00 PM Bug #10369 (Closed): [PATCH] array.c: missing RB_GC_GUARDs
- ~~~
[PATCH] array.c: swap volatile for RB_GC_GUARD
* array.c (ary_recycle_hash): add RB_GC_GUARD
(rb_ary_diff): remove volatile
The RB_GC_GUARD seems needed for most of the ary_recycle_hash
functions anyways.
~~~
-
08:58 PM Feature #10341: Fiber switch performance improvements
- Attached a revised version of patch 5 with fiber_mark_self replaced by rb_fiber_mark_self. Thanks for the hint.
-
06:09 PM Bug #10367 (Rejected): Net::HTTP read_timeout value behaves unexpectedly.
- John Thomas wrote:
> I would think that the expected behavior of the Net::HTTP.new().read_timeout property be that if that timeout was hit during a request, a Net::ReadTimeout exception would be thrown, and the request would be killed. ... -
07:49 AM Bug #10367: Net::HTTP read_timeout value behaves unexpectedly.
- This should be assigned to NARUSE, Yui (naruse) it looks like.
-
- I would think that the expected behavior of the Net::HTTP.new().read_timeout property be that if that timeout was hit during a request, a Net::ReadTimeout exception would be thrown, and the request would be killed. Currently, however, wh...
-
05:06 PM Feature #10365: `Rational#to_a` and `Array#to_r`
- Nobuyoshi Nakada wrote:
> Just a memo:
> ...
Thanks for the information. -
03:28 PM Feature #10365: `Rational#to_a` and `Array#to_r`
- Just a memo:
~~~ruby
[1, 2].inject(:quo) #=> (1/2)
~~~ -
01:17 PM Feature #10365 (Rejected): `Rational#to_a` and `Array#to_r`
- And without having Rational#to_a, Array#to_r does not mean much.
So I reject this.
Matz.
-
- I think Nobuyoshi Nakada and Jeremy Evans are right. I will withdraw `to_a` from the proposal.
-
04:39 AM Feature #10365: `Rational#to_a` and `Array#to_r`
- Wouldn't the proposed Rational#to_a break backwards compatibility, as well as consistency with other numeric classes:
Currently:
Array(1) # => [1]
Array(1.0) # => [1.0]
Array(BigDecimal.new('1.0')) # => [#<BigDecimal:12c032467878... -
- IMHO, splatting a rational doesn't feel to make sense, so `to_a` doesn't feel appropriate.
-
- Sorry, I meant
2/51r.to_a # => [2, 51] -
- I guess people would want to extract the numerator and the denominator of an array. `to_a` would perhaps be a good name:
(2/51).to_a # => [2, 51]
It also might be useful if `Array#to_r` were provided as an alternative way to cr... -
04:39 PM Revision 8980a9b4 (git): * file.c: include sys/time.h only if HAVE_SYS_TIME_H
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47888 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- For some more sophisticated examples, see https://github.com/nome/coroutines. The library does work with vanilla Ruby, but the patch improves performance.
Also, similar code can be simplified by using Fiber#raise. Compare e.g. the two... - 03:13 PM Revision 23abeeb7 (git): * 2014-10-12
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47887 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * file.c (sys/time.h): include after unistd.h to get rid of
mismatch on struct stat and some system call functions on Linux.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47886 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 01:33 PM Revision 0d1b2755 (git): * remove trailing spaces.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47885 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Fixes build error.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47884 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- extinit.$(OBJEXT) used to be generated by the builtin rule, thus
didn't accept custom $(CC) and caused linkage error for cross
compiling.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47883 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Note: Some of the fixes are for newlib in general but not NaCl-specific.
* include/ruby/intern.h (rb_fd_select): declare struct timeval, or the
struct gets local to the function in C99.
* file.c (#include): add nacl/stat.h for PNaCl.... -
10:58 AM Bug #10368 (Closed): GC mark miss on bindings
- 以下のコードでSEGVすることがあります。
mに渡したブロックがGCされてしまうのが原因です。
~~~ruby
def m
binding
end
GC.stress = true
b = nil
tap do
b = m {}
end
GC.start
0.times.to_a
b.eval('yield')
~~~
ブロックに対応するProcオブジェクトをrb_binding_tに持たせるようにするパッチを添付します。 - 06:33 AM Revision 3fedb512 (git): compile.c: fix RB_GC_GUARD usage
- * compile.c (iseq_build_from_ary_exception): move RB_GC_GUARD
(iseq_build_from_ary_body): use PRIsVALUE instead of RB_GC_GUARD
Currently unused functions, but we should set good examples for
using RB_GC_GUARD (and favoring PRIsVALUE i... - 06:02 AM Revision 58a73d96 (git): string.c (rb_str_intern): remove unnecessary RB_GC_GUARD
- Incorrectly placed, and no longer needed since r25351
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47880 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- On the other hand, maybe we should allow Ruby-level options to `system'
to raise, but continue hiding errors when running the command itself.
In other words:
# continue old 1.8 behavior if redirect is done via shell
# (n.b. th... -
- Yes, it's intentional spec, but to raise errors `Process.spawn` and `Process.wait` are needed.
It may be good to add an option for errors to `system`. -
- * parse.y (remove_duplicate_keys): remove duplicate literal keys,
i.e., symbols and strings. [ruby-core:65368] [Bug #10315]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47879 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * vm.c (kwmerge_i): override existing keys by new keys.
[ruby-core:65368] [Bug #10315]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47878 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r47877.
----------
parse.y: precedence of duplicated keys
* parse.y (assocs): concatenate splatted literal hashes. the
former key has precedence even if duplicated literal keys
follow. [ruby-core:65368] [Bug ... -
- * parse.y (assocs): concatenate splatted literal hashes. the
former key has precedence even if duplicated literal keys
follow. [ruby-core:65368] [Bug #10315]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47877 b2dd03c8-39d4-4d... -
- * configure.in: remove extra right-bracket.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47876 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * configure.in (RUBY_NACL): automatically locate pnacl-clang.
(RUBY_PLATFORM): pnacl instead of le32-nacl.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47875 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- platforms other than NativeClient.
* ChangeLog: ditto. add entries for the last two commits.
Sat Oct 11 11:12:00 2014 Yuki Yugui Sonoda <yugui@yugui.jp>
* signal.c (install_signalhandler, init_sigchld): allow failure because i... -
- * signal.c (install_signalhandler, init_sigchld): allow failure because it
always fails with ENOSYS on NaCl.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47873 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * configure.in (RUBY_NACL and others): Supports PNaCl.
* dln.c: replace the old hacky dynamic loading over HTTP with nacl_io.
* file.c: tenatively use access(2) instead of eaccess.
(rb_file_load_ok): weaken with attribute but not by po... -
- Currently, rational inspection is expressed with parentheses:
(2/51)
If this were taken as a Ruby expression, it would mean integer division, whose value is `0` in this case. It does not make much sense to express that it is in... -
12:33 AM Revision f21431e5 (git): * ext/socket/unixsocket.c: [DOC] Fix example to render in HTML
- properly, with a patch by @eval [Fixes GH-733]
https://github.com/ruby/ruby/pull/733
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47871 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
10/10/2014
-
- Yes, `autoload` can do the trick, but it needs a separated file and warns only once.
-
- Eric Wong <normalperson@yhbt.net> wrote:
> http://80x24.org/spew/m/opt-str-lit-v1%40m.txt
Benchmark suite so far:
http://80x24.org/spew/m/opt_str_lit-v1-bench%40whir.txt
Note: this increases iseq size and can slow some perfo... -
- Current patch (named "opt_str_lit" insn):
http://80x24.org/spew/m/opt-str-lit-v1%40m.txt
One VM instruction now optimizes away object allocation for string
literals in the following cases:
* "lit" % obj
* str << "lit"
* "li... -
09:59 PM Bug #9681: miniruby Bus Error at 0x3d44c4
- Hello, are there any plans to fix this issue in Ruby 2.1.4?
-
07:16 PM Feature #10361 (Closed): Support open_timeout option in open-uri - Applied in changeset r47869.
----------
open-uri: accept :open_timeout option
* lib/open-uri.rb (OpenURI::Options): add :open_timeout default
* (def OpenURI.open_http): check :open_timeout option
* (module OpenURI): rdoc for :open_time... -
09:46 AM Feature #10361 (Closed): Support open_timeout option in open-uri
- I found it weird that open-uri accepts read_timeout option, but not open_timeout.
I believe this option should be added. - 07:16 PM Revision b394484a (git): * 2014-10-11
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47870 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 07:15 PM Revision 9559b04e (git): open-uri: accept :open_timeout option
- * lib/open-uri.rb (OpenURI::Options): add :open_timeout default
* (def OpenURI.open_http): check :open_timeout option
* (module OpenURI): rdoc for :open_timeout
* test/open-uri/test_open-uri.rb (test_open_timeout): new test
[Feature #1... -
- bdimych@narod.ru wrote:
> irb(main):013:0* system 'cmd', '/c', 'echo', 'aaa', :out => File.open('bad/file.txt', 'w')
> Errno::ENOENT: No such file or directory - bad/file.txt
> from (irb):13:in `initialize'
> from (i... -
10:04 AM Bug #10362 (Closed): spawn doesn't raise exception on redirection error
- ~~~
irb(main):002:0*
irb(main):003:0*
irb(main):004:0* system 'cmd', '/c', 'echo', 'aaa'
aaa
=> true
irb(main):005:0>
irb(main):006:0*
irb(main):007:0*
irb(main):008:0* system 'cmd', '/c', 'echo', 'aaa', :out => ['bad/file.txt',... -
03:53 PM Bug #10364 (Rejected): Malformed CSV Header Causes NoMethodError
- The method CSV#readline raises a "NoMethodError: undefined method 'encode' for nil:NilClass" exception when the CSV file's header row contains a trailing delimiter (a comma in my case). The following CSV data can be used to recreate thi...
-
- * thread.c (only_if_constant): get rid of __builtin_choose_expr()
on older gcc.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47868 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * configure.in (__builtin_choose_expr_constant_p): in gcc 4.8 or
earlier, __builtin_choose_expr() does not consider
__builtin_constant_p(variable) a constant expression.
* include/ruby/ruby.h (RUBY_SAFE_LEVEL_CHECK): fix for older g... -
- * include/ruby/ruby.h (RUBY_SAFE_LEVEL_CHECK): eliminate function
call for warning/error if not match to get rid of unconditional
warning/error by a certain compiler option.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47866 b2... -
- * io.c (rb_io_make_open_file): remove useless backslash.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47865 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
09:42 AM Revision 5a77a4e7 (git): Update doc. and warnings.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47864 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
07:16 AM Feature #10355: Feature request: Module#prepended?(mod)
- 提案する名前について `prepended?` は正しくありませんでした。`prepend?` が正しいです。訂正します。
-
- メソッド定義が上書きされているかどうかを調べたいというより、誰がownerでもいいけど確実にprependされていてほしい、という方が要求として正確なように思います。
-
- このメソッドに反対なわけではないですが、メソッド定義が上書きされているかどうかを調べたいのであれば、`UnboundMethod#owner`などを使うほうがいいのではないでしょうか。
-
- 特定メソッドの出力のフィルタをprependによって実現したい場合、親クラスで既にprependされていても子クラスでメソッド定義が上書きされればprependが実際には効かないことになります。
このような状態が起きてないかどうかを調べる必要がある(起きていればprependしなおす)、というケースで Module#prepended? 相当の機能が必要となります。
フレームワークなどの記述において、DI等で挿入されたオブジェクトの機能を調べてコントロールする用... -
- ある機能が prepend によって追加されているかどうかを調べたいときに Module#include? では不十分です。
継承関係の親クラスにおいて prepend されたモジュールのメソッドは、子クラスでも定義が存在する場合、実際には呼び出されません。
~~~ruby
module P
def myname
str = super
"p" + str
end
end
class X
def myname
... -
- あるモジュール X に対してモジュール P1 がprependされているかどうかを調べる方法があると嬉しく思います。
`Module#include?` を使うと継承関係に含まれているかどうかはわかりますが、それが対象モジュールに対してprependされたものかは分かりません。
~~~ruby
module P1; end
module P2; end
class X
prepend P1
end
class Y< X
prepend ... -
05:24 AM Feature #10356: [PATCH 2/2] Remove unused internal functions
- @Nobuyoshi Nakada san
I'm so sorry... I just did grep by `rb_math_cos`, `rb_math_sin`, and so on... forgot C-lang has macro.
Thank you for your time. I should learn more. -
- Aren't they used in complex.c?
-
- I wondering which is preferred, remove or comment-out.
I came to the conclusion that we can invoke such a function when the time we need come. -
03:40 AM Feature #10354: Optimize Integer#prime?
- Oups, it would help if I gave the right implementation for Bignum#prime? See below
Here are some benchmarks:
Benchmark over 1..1_000_000
Current prime lib: 15.54
Fixed prime lib: 7.13
Nick's: 2.49
Benchmark for... -
- Nick Slocum shows in https://github.com/ruby/ruby/pull/736 that Integer#prime? can be optimized quite a bit.
First, that's because there are some basic things to avoid in the current lib, like needlessly capturing blocks and there's a... -
03:31 AM Feature #10351: [PATCH] prevent CVE-2014-6277
- On 10/10/2014 02:16 AM, Eric Wong wrote:
> shyouhei@ruby-lang.org wrote:
>> This patch does not add a new feature, nor delete anything. It just
>> changes the default behaviour when ruby spawns subprocesses.
>>
>> Process.spaw... -
- Akira, isn't it obvious this patch introduce a backward incompatibility?
I'd be surprised if it could passed the tests, because that means our tests are insufficient. Is it that important to pass make check at this point?
I don't... -
-
02:58 AM Bug #10353 (Closed): weird behavior when dynamically adding method using instance_eval
- Earlier this week I cross by a weird bug using a gem and after digging into the problem I found this issue with MRI.
I could not reproduce it on 1.9.3-p545 nor on the 2.2.0-preview1, the issue occurs on the 2.1.2
~~~
2.1.2 :001 > cl... -
- * io.c (rb_io_make_open_file): move from include/ruby/io.h, and
hide too detailed implementations.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47863 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 02:27 AM Revision 0d3d17b9 (git): * 2014-10-10
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47862 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * include/ruby/io.h (rb_io_mode_flags, rb_io_modenum_flags):
deprecate old macros for compatibility for ruby 1.8 and older.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47861 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
10/09/2014
-
11:44 PM Bug #8543: new rb_iseq_load crash
- Eric Wong wrote:
>
> ...
Good call -- cygwin indeed allowed me to roll back to bison 2.7.x
I wasn't able to fully automate `git bisect` as there was various build
breakage on some of the commits (usually to do with changes in enc... -
- billk@cts.com wrote:
> Sorry to be the squeaky wheel, but I was wondering if there still
> might be a chance to look into this before 2.2 is released?
Not speaking for the rest of ruby-core, but I welcome occasional
reminders like... -
- Hi,
Koichi Sasada wrote:
> Thank you for your report.
> ...
Sorry to be the squeaky wheel, but I was wondering if there still
might be a chance to look into this before 2.2 is released?
I attempted a `git bisect` this evening in... -
-
- shyouhei@ruby-lang.org wrote:
> This patch does not add a new feature, nor delete anything. It just
> changes the default behaviour when ruby spawns subprocesses.
>
> Process.spawn('/usr/bin/printenv') # -> prints nothing
T... -
- It is very aggressive.
At least, you should try "make check".
```
% make check
...
./miniruby -I./lib -I. -I.ext/common ./tool/runruby.rb --extout=.ext -- --disable-gems "./test/runner.rb" --ruby="./miniruby -I./lib -I. -I.ext... -
- ~~~
From 4636ca0308f1933c9b191f36e808a8d3bcf5e88e Mon Sep 17 00:00:00 2001
From: "Urabe, Shyouhei" <shyouhei@ruby-lang.org>
Date: Wed, 8 Oct 2014 15:44:27 +0900
Subject: [PATCH] prevent CVE-2014-6277
ShellShock was about bash. I ... -
- Give alias Vector#inner_product and Vector#cross_product which are of two method used frequently.
Vector#inner_product is aliased to Vector#dot.
Vector#cross_product is aliased to Vector#cross.
Take your time. -
02:50 PM Bug #8000: "require 'tk'" segfaults on 64-bit linux with Tk 8.6
- To generalize further, is it even necessary to do platform tests? Could the simple change:
+ Tcl_Eval(interp, "rename namespace __orig_namespace_command__");
Tcl_CreateObjCommand(interp, "namespace", ip_rbNamespaceObjCmd,
... -
- If this patch fixes the crash on Ubuntu, can it be generalized to support Ti 8.6 across platforms? Tk 8.5 is going to be EOL'ed at some point soon and it is very important that Ruby work with 8.6.
-
02:36 PM Bug #10350 (Closed): compile error in bignum.c line 3297 with old fcc on Solaris
- Applied in changeset r47860.
----------
* bignum.c (absint_numwords_generic): set an array element after
definition of a variable to fix compile error with older version
of fcc (Fujitsu C Compiler) 5.6 on Solaris 10 on Sparc.
[Bug... -
- Solaris上の古い fcc (Fujitsu C compiler) 5.6 にて、以下のコンパイルエラーが発生します。
~~~
"bignum.c", line 3297: error: initialization: constant expression is expected for variable: `nlz_bits_in_msbyte_bary'
~~~
以下のように変数の宣言と初期化を別々にしたらエラーなくビルドできました。
... -
- definition of a variable to fix compile error with older version
of fcc (Fujitsu C Compiler) 5.6 on Solaris 10 on Sparc.
[Bug #10350] [ruby-dev:48608]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47860 b2dd03c8-39d4-4d8f-98ff-... -
10:16 AM Feature #10344: [PATCH] Implement Fiber#raise
- Yes: it is missing piece to make "greenlet" port, in other words - green threads that mimics real threads.
To accomplish that there is a need to raise exception in a fiber waiting on event if event fails.
As Knut said it could be imple... -
02:58 AM Feature #10344: [PATCH] Implement Fiber#raise
- Do you have good example to use it?
-
- While it is possible to implement this in pure Ruby (by wrapping Fiber.yield and Fiber#resume), this feels like a low-level feature that ought to be provided out of the box. Also, the C implementation is more straight-forward, and more e...
-
- * ext/tk/tcltklib.c (tcltkip_type): use typed data.
* ext/tk/tcltklib.c ({call,eval,invoke}_queue_handler): hide
internal objects.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47859 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * internal.h (rb_ary_new_from_args): optimization by expanding
arguments in caller to get rid of va_list if possible.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47858 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 07:16 AM Feature #10321 (Closed): [PATCH] test st_foreach{,_check} for packed-to-unpack change
- Applied in changeset r47856.
----------
st: test packed-to-unpacked transitions during iteration
The st_foreach and st_foreach_check functions support transitioning
from a packed to an unpacked state during iteration. However, this
fu... - 07:16 AM Revision 9629e7ef (git): * properties.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47857 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 07:16 AM Revision f8b219e4 (git): st: test packed-to-unpacked transitions during iteration
- The st_foreach and st_foreach_check functions support transitioning
from a packed to an unpacked state during iteration. However, this
functionality did not get exercised by the current test suite until
now. This should help us prevent... -
- SASADA Koichi <ko1@atdot.net> wrote:
> On 2014/10/09 11:04, Eric Wong wrote:
> > A new putstring_for instruction may replace all current uses of:
> Sorry, I can't find a `putstring_for' instruction. Which patch should I see?
put... -
- On 2014/10/09 11:04, Eric Wong wrote:
> A new putstring_for instruction may replace all current uses of:
Sorry, I can't find a `putstring_for' instruction. Which patch should I see?
I'm happy how ruby sources compiled to bytecode... -
- Eric Wong <normalperson@yhbt.net> wrote:
> > Maybe start moving existing iseq_compile_each optimizations to the
> > peephole optimizer (work-in-progress):
> > http://80x24.org/spew/m/ee49aae645e0953fc16fc1557dce6a09b4de4324.txt
>
... -
- nobu@ruby-lang.org wrote:
> Mine is https://github.com/nobu/ruby/tree/peephole-opt
Thanks. I also noticed your `peephole_opt_send' branch which implements
opt_aset_with. Any reason it was not commited?
I'll take a closer look ... -
- Eric Wong <normalperson@yhbt.net> wrote:
> Maybe start moving existing iseq_compile_each optimizations to the
> ...
Part #2, generic putstring_for instruction:
http://80x24.org/spew/m/5a77be4e211c81a509573e3e1ca3bc3ca2383e68.txt
A ... -
- * array.c (ary_new): allocate buffer in heap after new object get
allocated successfully, to get rid of potential memory leak at
object allocation failure.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47855 b2dd03c8-39d4-4d8f-9... - 03:44 AM Revision 098b0ba5 (git): benchmark/bm_hash_aref_sym*.rb: force static symbols
- Dynamic symbols hash more slowly because they need extra method
dispatch in rb_any_hash. I am not sure if dynamic symbols are
a realistic use case as hash keys, so this commit only
restores performance when comparing against versions of... - 03:39 AM Revision 3cb39668 (git): hash.c (rb_any_hash): remove unnecessary dsym check
- gcc (Debian 4.7.2-5) 4.7.2 on x86 generates smaller code this
way, as it does not seem to be able to reuse the result of
SPECIAL_CONST_P.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47853 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
10/08/2014
-
- Cool, I can confirm the performance results on one of my systems.
I do not know the fiber code well, but it seems correct.
I think fiber_mark_self (and any non-static functions) needs to be
prefixed with "rb_" even if it is an inte... -
- The attached patches increase performance of switches between Fibers by ~17% on my test system (Linux, gcc 4.8.2).
Patches 1-3 are purely cosmetic, but included here because submitting them separately would cause conflicts between the... -
10:34 PM Feature #10090: Display of program name in process listing under AIX
- Ryan,
The previous patch was slightly (but critically) wrong, so please apply r47852. -
10:05 PM Feature #10090: Display of program name in process listing under AIX
- Rei,
I can confirm that this patch works as expected and that `ps auxww` now reports all of argv, as expected!
Thank you for taking a look at this!
Ryan -
- Rei,
thank you for that patch! I will test it on AIX today. -
- Ah, I almost understood. After Process.setproctitle is called, argv[1], argv[2], etc. are no longer valid, so we must set argv[1] to NULL.
However, they should be valid until the first call to Process.setproctitle. I thinks the follow... -
- I would like to fix this problem. It seems there is no reason this line is necessary (`missing/setproctitle.c:compat_init_setproctitle`), as Ryan pointed out:
~~~
argv[1] = NULL;
~~~
Does anyone know why it is needed?
It is int... - 10:31 PM Revision f3754f57 (git): missing/setproctitle.c: Avoid invalidating argv[1], argv[2], etc. until the first call to Process.setproctitle, because the ps command of AIX refers to the argv array. [Bug #10090]
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47852 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
06:43 PM Misc #10339: normalize reference to Timeout::Error
- > The doc/ChangeLog-1.9.3 modification is incorrect.
Whoops, sloppy of me!
> ...
Thanks! -
- @Marc-Andre Lafortune
Thank you for the thorough explanation.
OK, I understand what you said, this satisfied me now.
I realize we'd better think about efficiency more than maintainability or brevity of codes.
(There might be so... -
- Thank you for your comments.
I want to point out that I've added both `x.hstack(y)` and `Matrix.hstack(x, y)`, so both interfaces are available now, the same way I would love for Array#product to have a corresponding Array.product (se... -
- @Marc-Andre Lafortune san
It’s hard for me to say this, but I have confidence in my implementation more than merged one.
My implementation
```
def row_merge(*matrices)
if matrices.any?{ |m| m.row_size != row_size }
rais... -
- The same effect can already be achieved using `autoload`, right?
-
- Proposal for `Module#deprecate_constant`, which marks constants referred by arguments as deprecated.
~~~ruby
class C
FOO = 1
deprecate_constant :FOO
end
C::FOO #=> warning: constant C::FOO is deprecated
~~~
https://github... -
- I'm not sure how we can implement this.. but it looks not bad for me.
What I afraid of most is ..being able to parse this syntax well?
Aside from that, we can achieve what you want for now.
```
case foo
when some_very_long_c... -
- In a `case` statement, the condition part (`when ...`) in the branches do not have the same length in general, and especially, `else` is much shorter than the conditions. So when we write the condition and the return value in a single li...
-
04:16 PM Bug #10248: Possible missing test or bug for Module#include and Module#prepend
- I've created a patch to the MRI test suite used in JRuby to check for this behavior.
https://github.com/tduehr/jruby/commit/5bf0da4e4fff1a9e122ff427ff34586025db9955
I'm now using #9112 for discussion on new behavior with a patch to... -
- Thanks for the input. I think that's enough that I can start working on it.
Porting the implementation I have for JRuby would yield O(N) method search time with recursion up to the deepest include/prepend. I'm pretty sure the current ... -
- The reasons behind the current behavior are as you have guessed. Besides that, it was quite difficult to implement this more dynamic inclusion efficiently (both space-wise and time-wise).
So with smarter implementation, with proper ti... -
- Are you just looking for the following test to pass then for 2?
~~~
def test_simple_include
obj = Object.new
class << obj
class A; def foo; 1; end; end
module B; end
class C < A; include B; def bar; f... -
- No.
`Module#include` doesn't change superclass of the receiver's metaclass. - 03:53 PM Revision ecb57e43 (git): * 2014-10-09
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47851 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Applied in changeset r47850.
----------
dir.c: fix rdoc [ci skip]
* dir.c (dir_s_aref): fix rdoc. `Dir.glob` allows an array but
`Dir[]` not. the former accepts an optional parameter `flags`,
while the latter accepts arbitrary n... -
-
- * dir.c (dir_s_aref): fix rdoc. `Dir.glob` allows an array but
`Dir[]` not. the former accepts an optional parameter `flags`,
while the latter accepts arbitrary number of arguments but no
`flags`. [ruby-core:65265] [Bug #10294]
... -
- * variable.c (generic_ivar_set): use st_update to insert object
which does not have generic instance variables yet.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47849 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
12:44 PM Revision b61cb94e (git): ext/win32ole/win32ole_variable.c: use typed data.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47848 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
08:44 AM Bug #10335: ssl connect using net/http on solaris x86 fails with core dump
- Naohisa Goto wrote:
> FYI, in my environment (SPARC Enterprise M5000), compiling OpenSSL 1.0.1i with "-xO5" and with "-m64" by using SolarisStudio 12.3 works fine.
thats true. i just compiled openssl on a sparc system (T5-2) and it w... -
- * variable.c (sv_i, cv_i): use st_update to insert non-existing
entries.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47847 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * constant.h (RB_CONST_PRIVATE_P, RB_CONST_PUBLIC_P): macros to
predicate constant visibilities.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47846 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47845 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
07:37 AM Feature #10340 (Closed): [PATCH] Improve test_syntax.rb in terms of underscore arg
- Applied in changeset r47844.
----------
* test/ruby/test_syntax.rb: added syntax tests of underscore arguments.
[Feature #10340][ruby-core:65496] -
- [Feature #10340][ruby-core:65496]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47844 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
07:11 AM Feature #7793: New methods on Hash
- > the block for these methods should take a hash rather than an array.
Do you mean the input should be a single argument with a hash? I don't think that is very consistent for `|k, v|` expansion.
> ...
Everything inside the block i... -
- Just like there are `map` and `map!`, there should be both a non-desctructive and a destructive version for this method.
~~~ruby
h = {a: "foo"}
h.non_destructive_one{|k, v| [k.to_s, v.upcase]} #=> {"a" => "FOO"}
h #=> {a: "foo"}
... -
- Mine is https://github.com/nobu/ruby/tree/peephole-opt
-
- Eric Wong <normalperson@yhbt.net> wrote:
> ko1@atdot.net wrote:
> > Another idea is to make more general approach to indicate arguments
> > (and a receiver) are string literal. It is called specialization.
> > Specialized instructio... -
-
01:48 AM Bug #8348: ruby hangs randomly in daemonized processes on Debian GNU/kFreeBSD
- I can't reproduce this bug with ruby 2.1.3 anymore. Feel free to close it.
-
- * lib/mkmf.rb (create_makefile): add rules to translate to
assembler sources.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47843 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
10/07/2014
-
- @Martin Dürst @Marc-Andre Lafortune
I agree it. Thank you so much. -
- Martin Dürst wrote:
> Why not allow both? Ruby is flexible.
Indeed.
In this case though, it would be a bit of an overkill I think.
Let's use named arguments. If `basis` becomes really popular and we want to make a shortcut we stil... -
- Sure, no problem.
-
- I've used hstack and vstack, as suggested in https://github.com/ruby/ruby/pull/344
I hope this satisfied everyone :-)
Committed as r47769 -
- Thank you, too.
-
- Done.
Thanks for catching this omission, and for the patch. -
- URI is considered that it doesn't require id.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47842 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Add some examples in terms of underscore given as argument.
Please check it, whenever you want. - 08:25 PM Revision 7fcf5741 (git): ext/etc/etc.c (etc_systmpdir): set default tmplen correctly
- Fixup r47826.
This fixes the following test failure for me:
TestRequire#test_tainted_loadpath [test/ruby/test_require.rb:331]:
SecurityError expected but nothing was raised.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47841 b2dd0... -
- Done.
Thanks for catching this omission, and for the patch. - 08:18 PM Revision a2ae7bcc (git): * lib/matrix.rb: Add @- and @+ for Matrix and Vector.
- patch by gogo tanaka [#10068] [#10069]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47840 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 08:16 PM Revision 49e9f471 (git): * NEWS: Move Matrix changes to the right place
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47839 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- No feedback, so closing.
I can't see how this could possibly "fail". -
- Great, thanks for the updated patch.
- 08:00 PM Revision 9bb7dfa2 (git): normalize reference to Timeout::Error
- From: John Bachir <j@jjb.cc>
* bootstraptest/test_io.rb (assert_finish):
normalize rescue for Timeout::Error
* lib/net/ftp.rb (Net#read_timeout): ditto for doc
* lib/resolv.rb (Resolv::ResolvTimeout): ditto for subclass
* lib/webrick/... -
- Better test with latest releases before making a bug report...
Duplicate of #9622.
- 07:40 PM Revision 55457701 (git): test/ruby/test_process.rb (TestProcess#test_setsid): AIX does not allow Process::getsid(pid) when pid is in a different session
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47837 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 07:33 PM Revision 185c85b2 (git): test/ruby/test_rubyoptions.rb (TestRubyOptions#test_encoding): On AIX, locale_charmap is ISO-8859-1 with LANG=C. This means the source encoding of stdin is ISO-8859-1, so "invalid multibyte char" error does not occur
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47836 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 07:30 PM Revision 56d57285 (git): * lib/matrix.rb: Add Matrix#laplace_expansion.
- patch by gogo tanaka [#10073]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47835 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 07:30 PM Revision 19a1257d (git): * 2014-10-08
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47834 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 07:29 PM Revision eb9c3e71 (git): * lib/matrix.rb: Add Vector.basis.
- Based on patch by gogo tanaka [#10072]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47833 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- If descending order is to adopted, a thing that has to be made clear is how to order different objects that are in a tie with respect to the measure in question.
~~~ruby
[0, 1, 2, 3, 4, 5, 6, 7, 8].max_by(6){|e| e % 3}
~~~
Which ... -
05:39 PM Feature #8887: min(n), max(n), min_by(n), max_by(n)
- A strong case can be made for returning the elements in their original order. That ordering information is otherwise lost; whereas, the client can always sort the resulting elements in whatever order she wants. This behavior would also...
-
03:20 AM Feature #8887: min(n), max(n), min_by(n), max_by(n)
- Hello. I think the Ruby team should reconsider the ordering of the array returned by `max` and `max_by` when the `n` argument is provided. It makes much more sense to me that it would be sorted in descending order, meaning that the mos...
-
- The doc/ChangeLog-1.9.3 modification is incorrect. I will remove it and
commit the rest:
normalize reference to Timeout::Error
From: John Bachir <j@jjb.cc>
* bootstraptest/test_io.rb (assert_finish):
normalize rescue for T... -
- normalize references to Timeout::Error
TimeoutError is a legacy constant, Timeout::Error is the canonical constant.
This patch normalizes all code and comments to reference Timeout::Error. -
- I understand 1. I'm missing something on 2. Does that mean my example code above should raise on including `m2`?
-
06:37 PM Feature #9112: Make module lookup more dynamic (Including modules into a module after it has already been included)
- See #1586
~~~
1. allow multiple inclusion
1. propagate when a module includes a module
Both have been (basically) accepted
~~~ -
- I'm working on a patch for this to use a live tree for module ancestry. I had already implemented this as part of Module#prepend before coming across this discussion. I wrote a test for this behavior as a starting point but it fails unde...
-
03:30 PM Bug #10338 (Third Party's Issue): ruby script running gnome-open will crash vim, if the file open by gnome-open is handled by vim
- I hit this bug recently.
Code to reproduce the error under my environment: -
02:43 PM Bug #10337 (Third Party's Issue): sqlite3/statement.rb:37: [BUG] Segmentation fault
- C:/Ruby200-x64/lib/ruby/gems/2.0.0/gems/sqlite3-1.3.9-x64-mingw32/lib/sqlite3/statement.rb:37: [BUG] Segmentation fault
ruby 2.0.0p481 (2014-05-08) [x64-mingw32]
-- Control frame information ------------------------------------------... -
- * signal.c (rb_f_kill): get rid of deadlock as unhandled and
discarded signals do not make interrupt_cond signaled.
based on the patch by Kazuki Tsujimoto at [ruby-dev:48606].
[Bug #9820]
git-svn-id: svn+ssh://ci.ruby-lang.org/rub... -
- Thanks.
FYI, in my environment (SPARC Enterprise M5000), compiling OpenSSL 1.0.1i with "-xO5" and with "-m64" by using SolarisStudio 12.3 works fine.
-
- Toni Schmidbauer wrote:
> i still have to verify this with a more recent openssl version.
so i'm able to confirm the solution, this is a problem with compiling openssl and not ruby:
ruby 2.0.0p576 / openssl 1.0.1i -> works
ruby 2... -
- Naohisa Goto wrote:
> The function "tls1_PRF" exists in libssl.so, and I still doubt if the OpenSSL is correctly installed.
> ...
seems like i found the solution... the culprit seems to be the default optimization flags the sun studio ... -
- The function "tls1_PRF" exists in libssl.so, and I still doubt if the OpenSSL is correctly installed.
Could you please provide more information about OpenSSL?
~~~
% ldd /opt/puppet/lib/libssl.so.1.0.0
% ldd /opt/puppet/lib/libcry... -
- Naohisa Goto wrote:
> % ldd /opt/puppet/lib/ruby/2.1.0/i386-solaris2.10/openssl.so
libssl.so.1.0.0 => /opt/puppet/lib/libssl.so.1.0.0
libcrypto.so.1.0.0 => /opt/puppet/lib/libcrypto.so.1.0.0
libsoc... -
- Could you please provide results of the following commands and
check if these are linked with intended OpenSSL and zlib libraries?
~~~
% ldd /opt/puppet/lib/ruby/2.1.0/i386-solaris2.10/openssl.so
% ldd /opt/puppet/lib/ruby/2.1.0/... -
- we tried to compile ruby 2.1.3 on solaris x86 with the current sun studio compiler 12.3 (for running puppet).
we used the following configure options to compile openssl:
~~~
./Configure --prefix=/opt/puppet shared solaris-x86-cc... -
- ＞ なかださん
修正ありがとうございます。
ただ、gdbでProcess.kill(:TRAP, $$)によるシグナルを受けた後にcontinueすると
ruby_kill内のnative_cond_waitで待機したままになります。
Rubyでsignal handlerを設定していない場合にはruby_killではなくkillを呼び出すようにすればよさそうですが、
何か問題あるでしょうか。
~~~patch
diff --git a/sig... -
- So Wieso wrote:
> Nobuyoshi Nakada wrote:
> ...
If you don't want to create a module by that name, you don't need to use the name.
Should not introduce implicit conversion between a name and a module.
You can use anonymous module too... -
12:27 PM Feature #10320: require into module
- Michael Gee wrote:
> I don't like changing `require` with one argument to mean something else. It would break too much legacy code.
You are definitely right here, we should not do that.
Nobuyoshi Nakada wrote:
> ...
Is it? If th... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47831 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- OS's limitation.
Not Ruby's limitation. -
01:34 PM Bug #10336: limit of number of arguments passed to system
- fine, however is the limitation imposed by my system (and tunable) or by ruby?
-
- POSIX systems can have a limitation on the size of arguments.
Use spawn() to raise the execption on the error of execve(2):
```
% ruby -v -retc -e 's = "true " + "a" * Etc.sysconf(Etc::SC_ARG_MAX); p s.length; Process.wait spawn(s... -
- just adding the join() to my example.
~~~
system("ls #{files_to_be_p_as_a.join(' ')}")
# => nil
system("ls #{files_to_be_p_as_a[0..2273].join(' ')}")
... # list of files
system("ls #{files_to_be_p_as_a[0..2274].join(' ')}")
... -
- The following code snipped stops to work with argument bigger than 2274:
~~~
files_to_be_p_as_a = Dir[File.join(".", '**', '*.c')]
files_to_be_p_as_a.size
# => 43152
system("ls #{files_to_be_p_as_a}")
# => nil
system("ls #{fi... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47830 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 12:45 PM Revision cfd359ca (git): * remove trailing spaces.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47829 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- olemethod_data_get_struct to wrap Data_Get_Struct.
ext/win32ole/win32ole_method.h: ditto.
ext/win32ole/win32ole_param.c (oleparam_ole_param):
call olemethod_data_get_struct instead of Data_Get_Struct.
git-svn-id: svn+ssh://ci.ruby-la... -
- Can you try `configure` with `--with-setjmp-type=setjmp`?
-
06:18 AM Bug #10334 (Closed): Installation Error
- Trying to install
-
11:00 AM Feature #10084: Add Unicode String Normalization to String class
- This feature is going to add one or more methods to class String (String#unicode_normalize and probably String#unicode_normalize! and String#unicode_normalized?).
The implementation also internally needs various additional methods and... -
- ko1@atdot.net wrote:
> Comments for this ticket and the following tickets:
>
> > 1) [Feature #10326] optimize: recv << "literal string"
> > 2) [Feature #10329] optimize: foo == "literal string"
> To continue this kind of h... -
- How about to add .freeze to string literals?
```
i = 0
foo = "literal"
while i<6_000_000 # benchmark loop 2
i += 1
foo == "literal".freeze
end
```
It reduce object allocations and may have some performance gain (without ... -
- Comments for this ticket and the following tickets:
> 1) [Feature #10326] optimize: recv << "literal string"
> ...
To continue this kind of hack, we need tons of instructions for each methods.
What do you think about it?
Basicall... -
- This is a follow-up-to:
1) [Feature #10326] optimize: recv << "literal string"
2) [Feature #10329] optimize: foo == "literal string"
This can be slightly faster than: (string == "literal") because
we can guaranteed the "yoda lite... -
- BTW, we provide "bitblt" and "answer" insturctions as a official (joke) instructions without enabling SUPPORT_JOKE flag.
People can use them by assembling YARV codes (with some tricks) manually.
With your measurement, we paid 736 byt... -
- * configure.in (RUBY_UNIVERSAL_ARCH): remove arch options from
CXXFLAGS, not only CFLAGS.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47827 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * ext/etc/etc.c (etc_systmpdir): try user temporary directory by
confstr() on Mac OS X.
c.f. http://www.opensource.apple.com/source/ruby/ruby-104/patches/ext_etc_etc.c.diff
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47826 b2d... -
- * configure.in (struct stat.st_size): prefer off_t over int, long,
and so on. inspired by
http://www.opensource.apple.com/source/ruby/ruby-104/patches/config.h.ed
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47825 b2dd03c8-39d... - 01:37 AM Revision 8a0b7f64 (git): * 2014-10-07
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47824 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * configure.in (RUBY_UNIVERSAL_ARCH): fix missing quoting
brackets. incorporated from
http://www.opensource.apple.com/source/ruby/ruby-104/patches/configure.diff
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47823 b2dd03c8-39d4... -
- normalperson@yhbt.net wrote:
> Future optimizations:
>
> * recv << "interpolated #{dstr}"
> (immediate free + force recycle)
Tried, not worth it from what I can tell:
http://80x24.org/spew/m/7c7dd29f45920fb8fc9d2c72858334c60e...
10/06/2014
-
08:53 PM Feature #9112: Make module lookup more dynamic (Including modules into a module after it has already been included)
- +1 over here.
I remember I discovered this the hard-way extracting some code from Rails into a gem.
I believe the current behavior is a leak of the implementation. Conceptually, in my opinion, a method or constant lookup should fol... -
08:33 PM Feature #9112: Make module lookup more dynamic (Including modules into a module after it has already been included)
- nobu's point about reincluding the same module brought up a different concern.
Many modules perform very specific actions at include time. If they did not go through a reinclude process for every hierarchy, those changes might not hap... -
- @Nobuyoshi Nakada san
It was an honor to encounter such a underground holy war.
Sure. I'll follow what you said.
Thank you. -
- Welcome to the underground holy war. ;)
You may know that ruby-style.el has `(indent-tabs-mode . t)`. -
- @Nobuyoshi Nakada san
Thank you for your comment.
Starting to implement this issue, I find spaces and tabs being mixed.
And I suppose this might make us confusion and there are no reason spaces and tabs be mixed, but I am afraid... -
- @Tsuyoshi Sawada san
Thank you for your comment. I got your point.
As you said (and I mentioned also) it has much pain, so I can't implement right now.
How do you think about the case giving a argument?
I mean like that.
`... -
- Why expanding tabs?
-
- When there is more than one parameter passed to a block, your proposal will create a problem.
When the number of parameters passed to a block is more than the number of block variables in `||`, then destructive assignment should apply... -
- So Wieso wrote:
> I chose the symbol `:Lib`, as I thought Ruby would complain if the constant `Lib` would not exist at this time. The keyword `in` would define it, if it would not exist. I would prefer if we could solve it without using... -
01:46 PM Feature #10320: require into module
- So Wieso wrote:
> ~~~ruby
> ...
I don't like changing `require` with one argument to mean something else. It would break too much legacy code.
If you want to require a feature into your current namespace how about this:
~~~ruby
... -
- I chose the symbol `:Lib`, as I thought Ruby would complain if the constant `Lib` would not exist at this time. The keyword `in` would define it, if it would not exist. I would prefer if we could solve it without using symbols, but writi...
-
- Current behavior of rational literal and `String#to_r` does not recognize mixed fractions. Mixed fraction is not rare, and is frequently used in places such as US length measurement.
![drill bits](http://www.strongtie.com/graphics/anc... -
- Current behavior or `String#to_r` does not recognize negative denominators. This can lead to confusing results:
~~~ruby
"1/-3".to_r
#=> (1/1)
~~~
I propose negative denominators to be recognized.
~~~ruby
"1/-3".to_r
#=> (-1... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47822 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * symbol.h (struct RSymbol): move from internal.h.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47821 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
12:49 PM Bug #10330 (Closed): 例外クラス階層のドキュメントの誤り
- バックポート用の備忘録です。
r47820で修正しました。 - 12:43 PM Revision 8474ebad (git): * error.c: update exception tree. [DOC]
- reported by @hemge via twitter.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47820 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 09:45 AM Revision 01a7a62d (git): configure.in: Fix typo. [Bug #9914]
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47819 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Exactly.
Please fix it. -
- I am not familiar with autoconf, but shoundn't this be `ac_cv_func_posix_fadvise`, not `ac_cv_posix_fadvise`?
~~~diff
--- configure.in (revision 47818)
+++ configure.in (working copy)
@@ -1207,7 +1207,7 @@
[
... -
- * signal.c (rb_f_kill): should not ignore signal unless the
default handler is registered. [ruby-dev:48592] [Bug #9820]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47818 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r47817.
----------
test_string.rb: enable huge test only on possible platforms
* test/ruby/test_string.rb (test_LSHIFT_neary_long_max): enable
only on platforms where string size range is smaller than memory
sp... -
- * test/ruby/test_string.rb (test_LSHIFT_neary_long_max): enable
only on platforms where string size range is smaller than memory
space. this test does not make sense but just wastes memory and
time on other platforms, as it is har... -
- * symbol.c (rb_enc_symname_type): allow junk-base attrset ID
unless the stem is not an attrset name.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47816 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * tool/unicode_norm_gen.rb: freeze strings and tables.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47815 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * tool/unicode_norm_gen.rb (Integer#to_UTF8): escape non-ascii
code.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47814 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * tool/unicode_norm_gen.rb: optimization by string literal
concatenation instead of runtime concatenation.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47813 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- @Marc-Andre Lafortune
Thank you for your great proposal. -
- Follow-up-to: [Feature #10326] (optimize: recv << "literal string")
This speeds up some string comparisons greatly, vm2_streq1 results:
[["loop_whileloop2",
[[0.21688890410587192,
0.21697380719706416,
0.216852264944463... -
- ko1@atdot.net wrote:
> or remove them?
Sure, even better.
Anybody have objections to removing joke VM options?
Ruby 2.2 can officially become a SERIOUS VM :) -
- or remove them?
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47812 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47811 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 01:27 AM Revision a3a2a6c7 (git): * properties.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47810 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 01:27 AM Revision cc8a255d (git): * remove trailing spaces.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47809 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- https://github.com/duerst/eprun/blob/master/lib/generate.rb
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47808 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47807 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47806 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
10/05/2014
-
- Ruby 自身が signal handler を設定していたとき以外は実際に kill(2) したほうがよさそうですね。
-
- r45911から以下のコードを実行してもkill(2)が呼ばれないようになっています。
これは意図された変更でしょうか。
~~~
$ ruby -e 'Process.kill(:TRAP, $$)'
~~~
# これまでの挙動に戻すべきだと主張するほどのものではないですが
# これでシグナルが飛ばせるとデバッグの際に便利です。
-
- This is not a joke :)
I worry about executable size with things like [Feature #10326],
and I noticed unused instructions are emitted even with jokes disabled
This reduces executable size slightly on x86-64-linux:
text | data... -
04:09 PM Feature #10327 (Open): Bool/False/True module for '==='
ruby-core:0237 から始まる Boolean Class の話題で Boolean Module が提案されていました。
それは必要とされているのか？という疑問とともに .true? メソッドなどが議論されていました。
それを読んで、=== メソッドを利用する Bool, False, True Module を書いてみました。
これを使うと case 式で when TrueClass, FalseClass ではなく when Bool と...- 03:31 PM Revision eb4e5247 (git): * 2014-10-06
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47805 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * io.c (get_kwargs_exception): rb_get_kwargs() expects IDs, but
not Symbols.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47804 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47803 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Since we'll need it for st_reverse_foreach_check ([ruby-core:65408]),
I've implemented list_for_each_rev_safe to ccan/list:
https://lists.ozlabs.org/pipermail/ccan/2014-October/thread.html
It applies on top of two of my others intend... -
- nobu@ruby-lang.org wrote:
> I'm not against that check, but we may enable that test only on 64bit
> Windows, since it is nearly impossible to reproduce on other
> platforms, because of memory usage.
OK, can you add the 64-bit Wind... -
- I'm not against that check, but we may enable that test only on 64bit Windows, since it is nearly impossible to reproduce on other platforms, because of memory usage.
-
- I've been skipping this test for a while on my weaker systems,
but this makes it automated so it avoids triggering a swap storm
for Linux users less familiar with the test suite.
I think this only affects Linux which defaults to ove... -
- Concatenating literal strings to existing strings seems fairly common
practice, so avoid extra garbage when doing it. I don't know if escape
analysis is coming to Ruby, but this seems easier-to-do for now:
vm2_strcat
~~~ruby
i ... -
- On 2014/10/03 05:51, ruby-core@marc-andre.ca wrote:
> Issue #10072 has been updated by Marc-Andre Lafortune.
>
>
> Looks good, but I'm wondering if we should use named arguments as it's really difficult to remember the order without... -
02:15 AM Misc #10312: Give people more control over how the ruby parser sees code and lexical code elements (valid/invalid - toggle options) + macros
- > That sounds like heredoc syntax, which ruby already does.
Hmm I guess you are right - I can get it into a string
that way. But the problem is - I can already get it into
a String if I do this:
'jquery +'
so your suggestion w... -
- Applied in changeset r47802.
----------
digest/*/extconf.rb: do not link openssl when using bundled
* ext/digest/{md5,rmd160,sha1,sha2}/extconf.rb: configure OpenSSL
only if bundled libraries is not used, so that OpenSSL is not
lin... -
- * ext/digest/{md5,rmd160,sha1,sha2}/extconf.rb: configure OpenSSL
only if bundled libraries is not used, so that OpenSSL is not
linked unnecessarily. [ruby-core:65404] [Bug #10324]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@... -
- Applied in changeset r47801.
----------
rmd160/extconf.rb: fix function name
* ext/digest/rmd160/extconf.rb: fix transform function name to
check. [ruby-core:65091] [Bug #10252] -
- * ext/digest/rmd160/extconf.rb: fix transform function name to
check. [ruby-core:65091] [Bug #10252]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47801 b2dd03c8-39d4-4d8f-98ff-823fe69b080e