Activity
From 07/17/2015 to 07/23/2015
07/23/2015
-
10:28 PM Bug #11384: multi-threaded autoload sometimes fails
- Currently testing this in a loop:
http://80x24.org/spew/m/94541be0225540e34f0196e9754ae0eb5c07a4b7.txt
Subject: [PATCH] variable.c: additional locking around autoload
[ruby-core:70075] [Bug #11384]
Note: this open-coding locki... -
10:10 PM Bug #10910 (Feedback): NoMethodError when opening SSL connection with OpenSSL::SSL::VERIFY_PEER set and anonymous ciphers allowed
- Hi,
When I apply just the test, it doesn't fail. Are you sure the bug is still present? If it's still present, can you make a test that fails without changes to the implementation? -
09:54 PM Bug #11393: segfault on trivial application, embedding in C app.
- Note that the file "abc" intentionally doesn't exist.
-
- ~~~c
#include <ruby.h>
#include <stdio.h>
int main()
{
ruby_setup();
rb_load_file("abc"); // AV here
ruby_cleanup(0);
}
~~~
~~~
rbtest1.exe!rb_vm_bugreport(const void * ctx) Line 1024 C
rbtest1.exe!rb_bug_context(co... -
08:31 PM Bug #11392 (Third Party's Issue): proj4_c_impl.bundle: [BUG] Segmentation fault at 0x00000000000440
- ~~~
DOTs-MacBook-Pro:sims moboyle$ rails s
/Users/moboyle/.rvm/gems/ruby-2.2.2/gems/rgeo-0.3.20/lib/rgeo/coord_sys/proj4_c_impl.bundle: [BUG] Segmentation fault at 0x00000000000440
ruby 2.2.2p95 (2015-04-13 revision 50295) [x86_64-dar... -
08:01 PM Bug #10968: [BUG] object allocation during garbage collection phase in /opt/rubies/ruby-2.2.1/lib/ruby/2.2.0/openssl/ssl.rb:177
- Problem still happens.
Any news with it?
It happens only on one machine on Debian 7.
Does not reproduce on same machine with same OS. -
- richard.schneeman@gmail.com wrote:
> You've mentioned the case statement optimization previously in a patch I sent to Rack. I agree that the difference is pretty negligible. Although I'm able to consistently see one running `benchmark/... -
06:45 PM Feature #11375: Decreased Object Allocation in Pathname.rb
- Thanks for reviewing! I added a new patch above: https://bugs.ruby-lang.org/attachments/download/5400/ruby-changes.patch
You've mentioned the case statement optimization previously in a patch I sent to Rack. I agree that the differenc... -
07:40 PM Bug #11389: Regression in method parameter parsing
- Thank you, it works fine now! <3
-
12:05 AM Bug #11389 (Closed): Regression in method parameter parsing
- Applied in changeset r51350.
----------
parse.y: pop cmdarg
* parse.y (lambda_body): pop cmdarg stack for lookahead
token. [ruby-core:70067] [Bug #11380] -
05:55 PM Feature #10949: Time is WB unprotected
- @Koichi Sanada: could you guide me some hints to make Time WB-protected ? I really can't figure it out from the current documentation...
-
- Ping ? Still happening on ruby 2.1.2p95
-
05:14 PM Feature #11390: Allow symbols starting with numbers
- But it goes against the principle of least surprise that ruby follows throughout the language. It's counter-intuitive and IMO from a layman ruby programmer's point of view the solution you propose does not confirm to the elegance of the ...
- 09:53 AM Revision a34343b3 (git): * remove trailing spaces.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51358 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
09:53 AM Revision eed613f4 (git): * vm_core.h: constify rb_iseq_constant_body::local_table and
- rb_iseq_param_keyword::table and
rb_iseq_param_keyword::default_values.
* compile.c: catch up this fix.
* iseq.c: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51357 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- rb_control_frame_t::pc.
* compile.c (rb_iseq_translate_threaded_code): catch up this fix.
* iseq.c: ditto.
* vm_exec.c (vm_exec_core): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
08:31 AM Bug #11383: Infinite loop in str_buf_cat triggered by str_gsub
- Nobuyoshi Nakada wrote:
> Very interesting.
> ...
I'll submit the content of `*(struct RString *)str` as soon as the issue reproduces. I'm sure it will since it has happened many times for those last weeks.
I had to kill the previou... -
08:18 AM Bug #11391 (Closed): chopped_length should return early for empty strings
- The current implementation only tries to check if the beginning of the string isn't coming after its end (I'm not even really sure if that's possible actually, is there case where RSTRING_LEN returns a negative number?).
In any case, ... -
07:38 AM Feature #10903: [PATCH] Matrix#zip returns a matrix
- Marc-Andre Lafortune wrote:
> Hi, sorry I missed your proposal until now, and thanks for the patch.
> ...
No worries! Thanks for taking a look.
Matrix#zip is useful for simply implementing any binary (or n-ary) operation on matrices... -
- * string.c (fstr_update_callback): create new string for fake
string, and pool shared target unless substring.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51355 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * string.c (rb_str_new_frozen, str_make_independent_expand):
trivial peephole optimizations.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51354 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
02:00 AM Bug #11386: taint flag about rb_fstring()
- むむ、早い。
で、このパッチだと、Stringのサブクラスのインスタンスだったり特異メソッドがあったりインスタンス変数を持ってたりするとfstringにならなくなってるわけですが、そもそもtaintフラグが立ってるものもfstringにならなくてもいいんじゃない？ 的なことも思ったりします。
というわけで、やっぱり、「fstringって本質的にはなんなの？」という仕様の問題に立ち返る気がするのですが、これを説明できるのは誰でしょう？
最初は笹田さんかと思っ... -
- ほいさ
-
- * include/ruby/ruby.h: add raw FL macros, which assume always the
argument object is not a special constant.
* internal.h (STR_EMBED_P, STR_SHARED_P): valid only for T_STRING.
* string.c: deal with taint flags directly across String in... -
- * string.c (str_shared_replace): split from rb_str_shared_replace
without argument check.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51352 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * ext/openssl/ossl_ssl.c (ossl_sslctx_setup): fix argument types
to suppress shorten-64-to-32 and shorten-64-to-32 warnings.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51351 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r51350.
----------
parse.y: pop cmdarg
* parse.y (lambda_body): pop cmdarg stack for lookahead
token. [ruby-core:70067] [Bug #11380] -
- * parse.y (lambda_body): pop cmdarg stack for lookahead
token. [ruby-core:70067] [Bug #11380]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51350 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
07/22/2015
-
- なるほど、とは思ったのですが、
「次にぼくは『特異メソッドを定義したStringオブジェクトをrb_fstring()に渡すと問題がある』と言うッ！」
という感じです。(そして次はインスタンス変数) -
- tainted/untaintedのfstringを分けるようにするパッチです。
-
- r51261以降、mswinのtest-allでfailureが出るようになった件を調査していて発見したのですが、
rb_fstring()にはtaintフラグを保存しないという問題があります。
原因は2つあって、
1. sharedなStringオブジェクトを登録する際にtaintフラグをコピーしないバグがある
2. 既に同じバイト列・エンコーディングで表現可能なStringオブジェクトが登録されている場合それを返すが、
引数のtaintフラグ... -
- This bug is a duplicate of #11380.
-
- The issue can be fixed by reverting r50402 which was introduced by #11107.
-
- There is a regression in parameter parsing with current ruby-2.2-head and ruby-head:
```ruby
# ruby 2.2.2p95 (2015-04-13 revision 50295) [x86_64-darwin14]
def foo fn, options, █ end
=> :foo
foo -> {}, foo: :bar do; end
=> n... -
09:13 PM Feature #11390: Allow symbols starting with numbers
- Permitting such syntax would needlessly complicate the parser for virtually no gain, and potentially break existing code that uses a numerical literal as the alternative of a non-padded ternary expression (`foo?1:2`). Just use `:'1twothr...
-
- Currently it is not possible to create a symbol that looks like `:1twothree`.
Converting to a string and then symbolizing causes hash lookup problems and proves counter-intuitive. What's also surprising is that ruby allows symbols to ... -
- OpenSSL [copies the string returned by the pointe](https://github.com/openssl/openssl/blob/9f040d6decca7930e978784c917f731e5c45e8f0/ssl/t1_lib.c#L1800-1809), so it should be safe to just return a pointer to the string object and not set ...
-
06:35 PM Feature #11356 (Closed): Add ECDH support to OpenSSL wrapper - Applied in changeset r51348.
----------
* ext/openssl/ossl_ssl.c: add ECDH callback support. [Feature #11356]
* test/openssl/test_pair.rb: test for ECDH callback support -
- * test/openssl/test_pair.rb: test for ECDH callback support
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51348 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 06:32 PM Feature #9390 (Closed): Support for the ALPN TLS extension
- Applied in changeset r51347.
----------
* ext/openssl/ossl_ssl.c: add ALPN support. [Feature #9390]
* ext/openssl/extconf.rb: detect ALPN support in OpenSSL
* test/openssl/test_ssl.rb: test for ALPN -
- * ext/openssl/extconf.rb: detect ALPN support in OpenSSL
* test/openssl/test_ssl.rb: test for ALPN
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51347 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
04:29 PM Feature #11388 (Open): SMTP Service Extension for Delivery Status Notifications
- Considering https://tools.ietf.org/html/rfc1891 and http://www.sendmail.org/~ca/email/dsn.html
TD;DR
SMTP session via Telnet:
`R: 220 Pure-Heart.ORG SMTP server here
S: EHLO Pure-Heart.ORG
R: 250-Pure-Heart.ORG
R: 250-DSN
R: 25... - 03:16 PM Revision 3b0d4905 (git): * 2015-07-23
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51346 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
03:16 PM Revision bef7ff00 (git): fix typos [ci skip]
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51345 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Applied in changeset r51344.
----------
string.c: fix coderange of reverse
* string.c (rb_str_reverse): reversed string is not a substring,
and should not set coderange of the original string.
[ruby-dev:49189] [Bug #11387] -
12:53 PM Bug #11387 (Closed): String#force_encoding と String#reverse の後にSEGV
- 以下のコードでSEGVします。
```ruby
broken_str = "abcd\xf0"
broken_str.force_encoding("utf-8").reverse
p broken_str.reverse
```
また、以下のコードでは `String#split` で `ArgumentError (invalid byte sequence in UTF-8)` が
発生することが期待されますが、実際は例外が発生しません。
... -
- * string.c (rb_str_reverse): reversed string is not a substring,
and should not set coderange of the original string.
[ruby-dev:49189] [Bug #11387]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51344 b2dd03c8-39d4-4d8f-98ff-823f... -
01:43 PM Bug #11379: Difference in `Array#sort_by` method between Ubuntu Linux and OS X
- All clear, thank you.
-
- Because `qsort_r()` of glibc is used instead of `ruby_qsort()`.
-
- > You may not expect order is same even with two subsequent sort invocation.
Non deterministic algorithm... makes sense. Is that somewhere stated in documentation?
Just curious, do you happen to know what has changed in ruby 2.2 li... -
12:50 PM Bug #11379: Difference in `Array#sort_by` method between Ubuntu Linux and OS X
- > I understand that the order cannot be guaranteed but I would expect it to work the same on all platforms.
And you will be wrong.
You may not expect order is same even with two subsequent sort invocation.
Every "same order" produce... -
01:26 PM Revision 7389ac15 (git): * ChangeLog: fix a typo for r51324. [ci skip]
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51343 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * include/ruby/ruby.h (ASSUME): hint for optimization, the
expression is assumed to be true always.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51342 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Move frequent accesssing fields to upper part.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51341 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- iseq_compile_data_ensure_node_stack.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51340 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- rb_iseq_variable_body (rb_iseq_t::variable_body).
rb_iseq_variable_body can be modified after compilation.
* compile.c: use rb_iseq_t::variable_body.
* iseq.c: ditto.
* thread.c: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/tr... - 09:04 AM Revision 2f5897e6 (git): * test/webrick/test_utils.rb (TestWEBrickUtils#test_nested_timeout_outer): seems
- to be too short for mswin CI. fixed occasional failure introduced at r51235.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51338 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
08:56 AM Bug #11385: `==` with bidirectional/cyclic dependency
my version is "ruby 2.3.0dev (2015-07-21 trunk 51319) [x86_64-linux]"
and it does return true for me too-
08:34 AM Bug #11385: `==` with bidirectional/cyclic dependency
- Should the result not be **false** instead?
These two instances each have a different @friend and it cannot be determined if they are equal so "false" seems a much safer answer.
Also, this seems inconsistent to how Comparable#== trea... -
07:16 AM Bug #11385: `==` with bidirectional/cyclic dependency
- Hi Nakada-san:
~~~
D:\> ruby -v
ruby 2.2.3p147 (2015-07-04 revision 51143) [x64-mswin64_120]
D:\> ruby test.rb
test.rb:8:in `==': stack level too deep (SystemStackError)
from test.rb:8:in `=='
from test.rb:8:in... -
- `true` is the expected result, and the same with x64-mswin64_120.
-
02:54 AM Bug #11385 (Rejected): `==` with bidirectional/cyclic dependency
- ```ruby
class Something
attr_accessor :friend
def initialize(friend)
self.friend = friend
end
def ==(other)
friend == other.friend
end
end
a = Something.new([])
b = Something.new([a])
a.friend = [b]
... -
08:50 AM Revision eb0e81fc (git): * lib/matrix/eigenvalue_decomposition.rb: refine code style.
- [fix GH-959][ci skip] Patch by @bogdanvlviv
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51337 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
-
- Very interesting.
An empty string is usually embedding, and I can't tell why there is an empty non-embed string.
Couldn't you show its content, `*(struct RString *)str`? -
- * common.mk: fix macro name, exec but not EXEC. pointed by
@nagachika.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51336 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- exclude_end true case. [fix GH-968] Pach by @yui-knk
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51335 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- ~~~
I get this failure once in a blue moon:
#8 test_autoload.rb:46:in `<top (required)>':
open("zzz.rb", "w") {|f| f.puts "class ZZZ; def self.ok;:ok;end;end"}
autoload :ZZZ, "./zzz.rb"
t1 = Thread.new {ZZZ.ok}
... -
- * numeric.c (num_coerce): [DOC] fix doc for Numeric#coerce,
missing '+'. [Fix GH-974]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51334 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * Makefile.in (lex.c): make position independent table.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51333 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * parse.y (parse_ident): call rb_intern2 with the known length
instead of strlen().
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51332 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * common.mk: add exec to quoted commands.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51331 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * thread.c (rb_wait_for_single_fd): return the error when ppoll
failed. fix r51319.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51330 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
07/21/2015
-
11:24 PM Revision 8c1a75a3 (git): follow r51064
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51329 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51328 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Applied in changeset r51327.
----------
* make rb_iseq_t T_IMEMO object (type is imemo_iseq).
All contents of previous rb_iseq_t is in rb_iseq_t::body.
Remove rb_iseq_t::self because rb_iseq_t is an object.
RubyVM::InstructionSequ... -
- All contents of previous rb_iseq_t is in rb_iseq_t::body.
Remove rb_iseq_t::self because rb_iseq_t is an object.
RubyVM::InstructionSequence is wrapper object points T_IMEMO/iseq.
So RubyVM::ISeq.of(something) method returns differ... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51326 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- rb_control_frame_t::iseq and rb_control_frame_t::block_iseq.
* iseq.c (iseq_free): catch up this fix.
* vm.c: ditto.
* vm_dump.c: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51325 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51324 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * vm.c, vm_insnhelper.c: catch up this fix.
* iseq.c (iseq_data_to_ary): constify the first iseq parameter.
* vm_insnhelper.c (vm_make_proc_with_iseq): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51323 b2dd03c8-39d4-4d8f-... -
- * proc.c (rb_method_entry_min_max_arity): catch up this fix.
* vm_insnhelper.c (def_iseq_ptr): constify.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51322 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 07:07 PM Revision fd70f7c4 (git): * 2015-07-22
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51321 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Applied in changeset r51320.
----------
* gc.c (internal_object_p): Now a singleton classes appear by
ObjectSpace.each_object. [Bug #11360]
* test/ruby/test_objectspace.rb: add a test about it. -
03:04 PM Bug #11360: Singleton class doesn't appear by ObjectSpace.each_object
- It was intentional, since in the old days singleton classes are merely internal data structure. After intoduction of `#singleton_class` method, singleton classes are objects visible to Ruby world. In that sense, now it's natural for th...
-
- ObjectSpace.each_object. [Bug #11360]
* test/ruby/test_objectspace.rb: add a test about it.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51320 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- I've got a Ruby program which reads some RSS/Atom feeds to load each of them into a relational database. It makes use of a modified version of SimpleRSS v1.2.
After moving to Ruby 2.2.2, I observe 100% CPU usage from time to time. Usi... -
03:16 PM Bug #11380: Parser regression in 2.3?
- I found out that this regression is caused by 50402 (https://github.com/ruby/ruby/commit/fd0e9d2580bf757071683581d34354064c6ceb6a)
```
# prase.rb
def foo(pr, options, &blk)
p pr.call
end
foo -> { :hello }, a: 1 do end
```
... -
12:56 PM Bug #11380 (Closed): Parser regression in 2.3?
- This code works with 2.2 but causes SyntaxError with 2.3.
~~~
def foo(pr, options, &blk)
p pr.call
end
foo -> { :hello }, a: 1 do end
~~~
#=> t.rb:5: syntax error, unexpected keyword_do, expecting end-of-input -
03:09 PM Bug #11382: Keyword arguments randomly nullified
- My ruby version: "ruby 2.2.2p95 (2015-04-13 revision 50295) [x86_64-darwin14]"
-
- Actually, this has nothing to do with inheritance. Here is simplified example:
~~~
class Bug
def initialize(a:, b: nil, **others)
puts("a: #{a.inspect}, b: #{b.inspect}, others: #{others}")
end
end
require "yaml"
Bu... -
- Here is the example:
~~~
class Parent
def initialize(a: nil, b: nil, **properties)
puts("parent: { a: #{a.inspect}, b: #{b.inspect}, properties: #{properties}")
end
end
class Child < Parent
def initialize(args={})... -
02:23 PM Bug #11381 (Closed): String のサブクラスをハッシュのキーに指定した時に hash メソッドが呼ばれない
- String のサブクラスとして大文字小文字を同一視するようなクラスを作ろうとしましたが、
そのオブジェクトをハッシュのキーに指定しても期待通りに動作しませんでした。
どうやら hash メソッドが呼ばれていないようです。
```ruby
class CIString < String
def eql?(other)
self.casecmp(other) == 0
end
def hash
self.to_s.down... -
- * thread.c (do_select): replace switch and goto with a loop to
suppress maybe-uninitialized warnings by gcc6.
* thread.c (set_unblock_function, rb_wait_for_single_fd): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51319 b2d... -
- [fix GH-969] Patch by @yui-knk
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51318 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * win32/Makefile.sub (config.h): define UNREACHABLE by __assume.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51317 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * ext/io/console/console.c (console_key_pressed_p): raise
undefined virtual key code.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51316 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * cont.c (fiber_initialize_machine_stack_context): fallback
definition of CreateFiberEx for VC6.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51315 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * win32/win32.c (rb_w32_wreadlink): define symlink constants for
VC6.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51314 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * include/ruby/missing.h (M_PI, M_PI_2): fallback definitions for
VC6.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51313 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * include/ruby/win32.h: include windows.h before winsock2.h,
because mswsock.h included by the former uses SOCKET defined
after it in the latter. fix a build failure with VC6.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51312... -
- * iseq_add_mark_object()
* iseq_add_mark_object_compile_time()
* iseq.c, iseq.h (rb_iseq_add_mark_object): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51311 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- What do you mean that it's a stability issue? I understand that the order cannot be guaranteed but I would expect it to work the same on all platforms.
Also this code works fine with 1.9.3 and 2.1.6 and only fails with 2.2.
-
- It *is* a stability issue.
`Enumerable#sort_by` compares only the results of the given block, so can't differentiate "ac" and "AC".
Maybe you may want to write as:
~~~ruby
['az', 'ac', 'AC'].sort_by {|s| [s.downcase, s]}
~~~ -
- Sorry, it is not a stability issue.
-
- Not a bug.
`Array#sort_by` is not stable sort. -
- The sort order (for the same objects) is different depending on the operating system. It looks like the Linux version of sorting has changed versus the previous ruby versions.
### On Linux:
~~~
piotr@pi-virtualbox:~$ uname -a
Lin... -
- It's irrelevant to array, but a precedence problem.
An expression of `and` keyword can't be an argument without parentheses.
-
- Passing Array boolean logic into a method is yielding a syntax error when the opening parenthesis is not separated from the method name by a space. It works fine if there is a space or if the boolean logic is enclosed in parenthesis.
... -
- Jeremy Evans <code@jeremyevans.net> wrote:
> This turns out to be a bug in the lower levels of the OpenBSD kernel,
> specific to SEQPACKET. I've tested a patch that fixes it, it will
> probably be in the next release.
Good to kno... -
- Jeremy Evans <code@jeremyevans.net> wrote:
> Looks like my initial analysis was partially wrong. OpenBSD's default
> buffer for SEQPACKET sockets is 4096. However, ancdata.c needs a patch
> to handle EMSGSIZE.
This looks like a ... - 08:17 AM Revision 04b7a801 (git): test/socket/test_nonblock.rb: increase buffer sizes to OpenBSD limits
- * test/socket/test_nonblock.rb: increase buffer sizes to
OpenBSD limits. Thanks to Jeremy Evans <code@jeremyevans.net>
[ruby-core:70058]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51310 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 07:09 AM Revision e3af9cea (git): load.c (ruby_dln_librefs): make static
- This isn't used outside of load.c, so there appears to be no
reason to expose it to other modules.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51309 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * lib/optparse.rb (complete): [DOC] fix typo. [Fix GH-973]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51308 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * test/ruby/test_process.rb (test_execopts_popen): split into 3
tests, no other redirections, stdio, and other fds.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51307 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
07/20/2015
-
11:05 PM Feature #4897: Define Math::TAU and BigMath.TAU. The "true" circle constant, Tau=2*Pi. See http://tauday.com/
- As I wrote in https://github.com/ruby/ruby/pull/644#issuecomment-123082639, I closed the pull-request for the present time.
-
- merch-redmine@jeremyevans.net wrote:
> Bug #11364: Use smaller buffer for sendmsg
> https://bugs.ruby-lang.org/issues/11364
>
> OpenBSD's limit is less than 128.
Btw, this is pretty disappointing since I want to use SOCK_SEQPACK... - 08:35 PM Revision f3dbd84f (git): * 2015-07-21
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51306 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 08:34 PM Feature #11377 (Closed): [PATCH] IO.copy_stream uses poll on Linux
- Applied in changeset r51305.
----------
io.c: IO.copy_stream uses poll on Linux
poll and ppoll have a superior API which doesn't require the
kernel to scan a potentially large bitmap to find a high-numbered
FD [ruby-core:35572]. So fa... -
- poll and ppoll have a superior API which doesn't require the
kernel to scan a potentially large bitmap to find a high-numbered
FD [ruby-core:35572]. So favor using poll in case IO.copy_stream
encounters a non-blocking FD.
We canno... - 08:33 PM Revision f7c3b6ff (git): io.c: IO.copy_stream uses poll on Linux
- poll and ppoll have a superior API which doesn't require the
kernel to scan a potentially large bitmap to find a high-numbered
FD [ruby-core:35572]. So favor using poll in case IO.copy_stream
encounters a non-blocking FD.
We cannot rel... -
- Eric Wong <normalperson@yhbt.net> wrote:
> I started working on this patch which causes pipes to always
> be closed during timer thread shutdown.
>
> http://80x24.org/spew/m/839ba256acbf1a5b17ad60953fc3ecf1df6cec50.txt
OK, need ... -
- ngotogenome@gmail.com wrote:
> With the patch, when exec system call fails, signals that are sent
> after stopping the timer thread before re-starting the timer thread
> may be ignored.
I'm not seeing it, did you notice/reproduce ... - 06:05 AM Revision 8754f619 (git): parse.y (parser_initialize): avoid redundant zero-ing
- TypedData_Make_Struct already initializes fields to zero by default,
so there's no need to waste code re-zeroing them.
Size reduction on 32-bit x86:
text data bss dec hex filename
187231 2372 80 189683 2e4f... - 03:27 AM Revision 12652453 (git): parse.y (ripper_initialize): fixup r51302
- Missed eofp bitfield change in ripper_initialize in r51302 :x
("parse.y (struct parser_params): pack to 4 cache lines on x86-64")
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51303 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 03:22 AM Feature #11371 (Closed): [PATCH] parse.y (struct parser_params): pack to 4 cache lines on x86-64
- Applied in changeset r51302.
----------
parse.y (struct parser_params): pack to 4 cache lines on x86-64
This reduces the struct from 256 (from 288) bytes on my x86-64
system, allowing the struct to fit cleanly into 4 cache lines
(if a ... - 03:21 AM Revision 3114f4a2 (git): parse.y (struct parser_params): pack to 4 cache lines on x86-64
- This reduces the struct from 256 (from 288) bytes on my x86-64
system, allowing the struct to fit cleanly into 4 cache lines
(if a malloc implementation allows it).
* parse.y (struct parser_params): pack: 88 => 256 bytes on 64-bit
[ru... -
- Looks like you sent a backwards patch. We will need to use -R with
patch/git-apply to apply the patch.
Relying on "literal".freeze in "when" statements is pointless,
case_when_optimizable_literal in compile.c already does that for... -
12:39 AM Bug #11376 (Assigned): Stop using SSLv3 methods
-
- * thread.c (rb_thread_terminate_all): reuse the tag pushed before
a loop, instead of pushing/popping same tag repeatedly.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51301 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * vm_core.h (ruby_vm_throw_flags): constants for throw.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51300 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
07/19/2015
-
08:54 PM Bug #11376 (Closed): Stop using SSLv3 methods
- If openssl is compiled using the OPENSSL_NO_SSL3_METHOD you can't compile ruby anymore since it will still try to use the SSLv3_*_method()s.
Please stop using those method at least when they're not available.
It would also be nice ... -
- Pathname.rb has many string literals that are not mutated while being called. We can reduce object allocation and increase program speed by freezing these string literals.
I've attached a patch that adds `.freeze` to all non-mutated s... - 03:21 PM Revision 46b70c8d (git): * 2015-07-20
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51299 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * thread.c (ppoll): fix the limit, timeout argument of poll(2) is
an int but not a time_t.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51298 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
01:37 PM Feature #11361: proposal for easy method to nil-guard for generated variable name.
- Isn't this simpler (and easier to read):
@cache ||= {}
@cache[id] ||= User.find(1)
I can't recall ever needing to do what you want to do. -
11:27 AM Bug #11336: TestProcess#test_exec_fd_3_redirect failed on Solaris 10
- With the patch, when exec system call fails, signals that are sent after stopping the timer thread before re-starting the timer thread may be ignored.
Currently, except Mac OS X and Haiku, the signals could be invoked after the failur... -
- I think it's OK if intended, and you may want to remove all 0-assignment lines before this commit.
-
- nobu@ruby-lang.org wrote:
> Why remove this line?
>
> > - parser->parser_in_defined = 0;
Oops, I was going to MEMZERO-ing the whole struct (or rely on
TypedData_Make_Struct) but got distracted. I'll keep that line
if I comm... -
- Why remove this line?
> ~~~diff
> ...
-
09:15 AM Bug #11374: crash when using Thread.new{Tk.mainloop}
- When set $DEBUG, sometimes crash and sometimes exit without an abort message.
I add logs of both cases.
-
- I add the crash-log of serverTk2.rb on Linux x86_64 (Vine Linux 6.3) + ActiveTcl8.6.2.0.
-
- serverTk2.rb just existed silently, on OSX 10.10.
Couldn't you show the dump message? -
- What's your platform and tcltk's version?
-
- * ext/pty/pty.c: [DOC] fix example typo, an old name at move from
PTY.open. [Fix GH-972]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51297 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 12:26 AM Revision ce6ee740 (git): * 2015-07-19
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51296 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * eval.c (rb_jump_tag): paranoiac check of tag value.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51295 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
07/18/2015
-
08:16 PM Bug #11374 (Closed): crash when using Thread.new{Tk.mainloop}
- ServerTk2.rb crashed and core dumped when running this program
clientForkTk1.rb tested ok
please help -
03:38 PM Bug #11366: Don't force SSLv3 in test, as it is insecure and may not be supported
- > I think this change may require #11367 to work correctly.
You were right. This patch alone fails:
~~~
1) Failure:
OpenSSL::TestSSLSession#test_ctx_server_session_cb [/builddir/build/BUILD/ruby-2.2.2/test/openssl/utils.rb:317... -
12:21 PM Bug #11366: Don't force SSLv3 in test, as it is insecure and may not be supported
- This was reported as #10046, but I am not sure if this patch actually works everywhere ...
-
- Well, the tests fail on OpenBSD regardless of the patch, as OpenBSD crypt(3) does not support the historical, insecure DES-based crypt, only bcrypt, and you need to pass a bcrypt salt as the second argument. I will be adding a local pat...
-
12:01 PM Bug #11363: Fix tests for String#crypt so they pass on OpenBSD
- Do the tests cause any actual issue? Failed on OpenBSD?
Ruby's tests are not conformance test. They should include a test of implementation-defined behavior and it is actually useful to detect an unintentional change of behavior.
... -
- * load.c (rb_require_internal): use TAG_RETURN not a magic number.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51294 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
02:28 PM Feature #11373 (Open): Add command line option to query valid values for --dump
-
- * thread.c (vm_check_ints_blocking): gather common statements at
the end, and prefer LIKELY for Visual C optimization.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51293 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * load.c (rb_load_internal0): do not raise any exceptions but
return the result tag state.
* load.c (rb_load_protect): reduce nested EXEC_TAGs.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51292 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Hi,
Although I can understand how you feel, the change will break millions of Ruby programs.
There could be milder option, like warning for certain usege of equal sign in expressions.
Matz.
-
10:21 AM Misc #11372: Confusing a+b=9 ... unintented behavior, valid statement
- Just small error: Number 11 is wrong, 7 is right. :)
-
- Hello, I found that when = is written instead of ==, it can cause unintended effects and it is hard to notice this typo, because it is valid statement. For example valid statement a+b=7 ,due to the everything is a expression, is treate...
-
- * gc.c (run_finalizer): set and restore safe level here to reduce
nested EXEC_TAGs.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51291 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * eval.c (ruby_cleanup): error_handle() returns exit status to the
system, not internal error state, do not convert the exit status
again.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51290 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- parse.y (struct parser_params): pack to four cache lines on x86-64
This reduces the struct from 256 (from 288) bytes on my x86-64
system, allowing the struct to fit cleanly into 4 cache lines
if a malloc implementation allows it.
... -
- I started working on this patch which causes pipes to always
be closed during timer thread shutdown.
http://80x24.org/spew/m/839ba256acbf1a5b17ad60953fc3ecf1df6cec50.txt
However, I may also eliminate timer thread entirely (on POS... - 01:29 AM Revision 3661dae1 (git): test/ruby/test_process.rb: test thread+sigs work after failed exec
- Preparation for possible upcoming changes to timer thread. We need
to ensure signal handling and thread scheduling works after an exec
failure.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51289 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
07/17/2015
- 10:21 PM Bug #11364 (Closed): Use smaller buffer for sendmsg
- Applied in changeset r51288.
----------
test/socket/test_nonblock: use smaller buffer for sendmsg
OpenBSD's limit is less than 128.
* test/socket/test_nonblock: use smaller buffer for sendmsg
Patch-by: Jeremy Evans <code@jeremyevans... -
- OpenBSD's limit is less than 128.
- 10:21 PM Revision 1d9cf30d (git): test/socket/test_nonblock: use smaller buffer for sendmsg
- OpenBSD's limit is less than 128.
* test/socket/test_nonblock: use smaller buffer for sendmsg
Patch-by: Jeremy Evans <code@jeremyevans.net>
[ruby-core:70016] [Bug #11364]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51288 b2dd... - 10:08 PM Revision a3dca0f5 (git): signal.c (trap_handler): cleanup to use RSTRING_GETMEM + memcmp
- strncmp is unnecessary since the switch/case statement already
checks length of the string; so use memcmp.
This makes for a small reduction in binary size on 32-bit x86:
text data bss dec hex filename
2847473 12360 ... -
- On OpenBSD a delta of 1 leads to occassional test failures:
1) Failure:
TestGc#test_expand_heap [/home/jeremy/code/ruby/test/ruby/test_gc.rb:292]:
invalid heap expanding (base_length: 286, GC.stat[:heap_eden_pages]: 288).
Expecte... -
- This fixes a failure when running on LibreSSL, where an exception
wouldn't be raised if the callback returned a selected protocol
that was too long (more than 255 bytes). Instead, it would silently
truncate the selected protocol by c... -
- In many cases in the test suite, server_connect is being called
without a block inside an assert_raises block. This leads to
misleading error messages, such as:
[OpenSSL::SSL::SSLError, Errno::ECONNRESET] exception expected, not.
... -
- This fixes the test when using LibreSSL and possibly some
configurations of OpenSSL. -
- LibreSSL disables SSLv3 by default, and there's no reason this code
should require SSLv3. -
- This changes Webrick to default to creating SHA htpasswd files.
It also changes Webrick to support reading SHA htpasswd files,
in addition to supporting the previous crypt password files.
Among other things, this allows Webrick's ht... -
- The String#crypt documentation states "the format and the result
are system and implementation dependent", so the tests should not
be checking for specific results. We should only be checking that
the method returns a string, that th... -
- And I forget we probably need to deal with SYSTEM_DEFAULT, too.
Ugh. I almost favor my alternative patch to use sched_yield
instead of overriding SIGCONT.
My alternative patch would allow removing th->interrupt_cond
entirely and ... -
- Alternative patch, but I relying on sched_yield() may be unreliable.
Anyways, both patches are hacky; but so is the expected behavior
of immediate delivery when a process signals itself...
~~~diff
--- a/thread.c
+++ b/thread.c
@@... -
- make Process.kill(:STOP, $$) resumable
Self-inflicted signals are delivered immediately. This is fine
for most signals which are catchable, but SIGSTOP and SIGKILL
are special and cannot be caught by a userspace process.
SIGKILL... - 06:03 PM Feature #11358 (Closed): [PATCH] ARGF.read_nonblock should support `exception: false'
- Applied in changeset r51285.
----------
io.c (argf_read_nonblock): support `exception: false'
This is a preparation for [ruby-core:69892]
("io.c: avoid kwarg parsing in C API")
since I noticed ARGF.read_nonblock did not properly catch ... - 06:03 PM Revision f2feaafe (git): * 2015-07-18
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51286 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 06:02 PM Revision d4454d01 (git): io.c (argf_read_nonblock): support `exception: false'
- This is a preparation for [ruby-core:69892]
("io.c: avoid kwarg parsing in C API")
since I noticed ARGF.read_nonblock did not properly catch up to
the `exception: false' change.
* io.c (argf_read_nonblock): support `exception: false'
... -
- @Nobu,
Thanks for the feedback! I've attached a new patch that refactors those parts. -
- * vm_eval.c (rb_eval_cmd): $SAFE=4 has been deprecated.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51284 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * compile.c (iseq_compile_each): use enum ruby_tag_type names.
* vm_core.h (ruby_tag_type): move from eval_intern.h for compiling
break/next/redo/return.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51283 b2dd03c8-39d4-4d8f-98ff-... -
- * compile.c (iseq_compile_each): remove unused big flag, `while`
loop always leaves a result on the stack.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51282 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Koichi Sasada wrote:
> From Ruby 1.6, singleton classes are not appeared.
> ...
I guess the reason is exposing the last layer of singleton classes would expose "VM-internal singleton classes/metaclasses".
The documentation of `rb_sing... -
- From Ruby 1.6, singleton classes are not appeared.
Is it intentional?
-
- Singleton class doesn't appear by ObjectSpace.each_object.
The following is reproducible code.
```ruby
class C
class << self
p [self, self.class]
$c = self
end
end
ObjectSpace.each_object(Class){|o|
exit if ... -
08:31 AM Feature #11361 (Open): proposal for easy method to nil-guard for generated variable name.
- It's easy to 'nil-guard' for normal variable.
~~~
def user
@user ||= User.find(1)
end
~~~
but it's not simple for generated variable name.
~~~
def user(id)
variable_name = "@user_#{id}"
instance_variable_set(var... -
- * vm_core.h (RUBY_VM_CHECK_INTS): evaluate the argument only once
to get rid of inadvertent side effects.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51281 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * thread.c (RUBY_VM_CHECK_INTS_BLOCKING): move from vm_core.h for
the static function rb_threadptr_pending_interrupt_empty_p.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51280 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * vm_core.h (EXEC_EVENT_HOOK_ORIG): evaluate each arguments only
once to get rid of inadvertent side effects. fix use of `th`
variable in the second `if` statement.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51279 b2dd03c8-3... -
- * include/ruby/encoding.h (ENC_CODERANGE_CLEAN_P): predicate that
tells if the coderange is clean, that is 7bit or valid, and no
needs to scrub.
* re.c (rb_reg_expr_str): use ENC_CODERANGE_CLEAN_P.
* string.c (enc_strlen, rb_enc_cr_s... -
- * test/ruby/test_range.rb (test_first_last): Add test for
`Range.new`. [Fix GH-971]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51277 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * test/ruby/test_range.rb (test_first_last): Add assertions to
test of `Range#last` with exclude_end true case. [Fix GH-970]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51276 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- > Sorry, but I think rb_divert_reserved_fd seems a racy fix.
Yes, it might be. Comparison and replacement of the fd is not atomic there, and calling dup2 after the rb_divert_reserved_fd is also a source of race.
Indeed, I implicitl... - 01:42 AM Revision 3ee156c7 (git): thread.c (rb_thread_alone): simplify
- This function now also works in more places if ruby_current_thread
is unset.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51275 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 12:59 AM Revision 3e5c21f7 (git): remove unnecessary enumerator requires
- * lib/rinda/tuplespace.rb: remove enumerator require
* test/pathname/test_pathname.rb: ditto
Related to https://bugs.ruby-lang.org/issues/10902
but not a complete fix.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51274 b2dd03c8-39...