Activity
From 03/31/2016 to 04/06/2016
04/06/2016
-
07:23 PM Bug #12255 (Closed): [DOC] Small fix for rdoc of ext/json/lib/json/add/*
- This patch removes some comments that otherwise would appear in the docs of the affected classes.
E.g. for Exception
```
$ ri Exception
Descendants of class Exception are used to [...]
[snipped class documentation]
Exceptio... -
06:41 PM Feature #12254 (Closed): Add SSLCiphers option to webrick ssl.rb
- Hi,
In order to secure a sinatra api using webrick i need to specify the ciphers to be used.
To do this i have added SSLCiphers option to webrick ssl.rb on the sslcontext.
I have created this pull request :
https://github.com/ruby/... -
06:26 PM Bug #12136: OpenStruct.new(format: :bar).send :format # => too few arguments
- > a) Keep behavior the same and rubyists can alleviate these by using `public_send` instead of `send`
It doesn't seem that swapping `send` by `public_send` has any effect, at least not in the example in #12251.
Is there a workaroun... -
04:16 PM Bug #12223: Backport r54304 (SEGV in rb_str_format())
- ruby_2_3 r54505 merged revision(s) 54304.
-
- * sprintf.c (rb_str_format): fix buffer overflow, length must be
greater than precision. reported by William Bowling <will AT
wbowling.info>.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_3@54505 b2dd... -
03:01 PM Bug #12253: tkとmechanizeを併用してTkButtonを押すと"TypeError: allocator undefined for TkEvent::Event"エラーが起こる
- はじめのコードより
require 'mechanize'
をはずして2.3.0p0で試した結果、エラー無く動作しました。
申し訳ありません、当方の早とちりでお手数をお掛けしました。
恐れ入りますがチケットのクローズをお願い致します。 -
- 申し訳ありません。いま気づいて上記のコードから
require 'mechanize'
をはずしたらどうなるかを試しました。
結果は同じで、2.1.8では動作しますが、2.2.4では上記と同様のエラーでした。
タイトルは「2.2.4ではTkRootを明示的にnewするとエラーになる」のほうが適切だったかもしれません。 -
- Rubyコミュニティ 御中
恐れ入りますがお尋ねしたいことがあり、チケットを切りました。
ruby/tkとmechanizeを併用した際に「`TypeError: allocator undefined for TkEvent::Event`」というエラーが発生します。
2.2.4ではエラーになるのですが、2.1.8では問題なく動作します。
2.3.0はいまのところ "gem install mechanize" が失敗する(mechanizeが依存して... -
02:54 PM Bug #10015: Performance regression in Dir#[]
- The problem seems to persist Mac OS X, although is better on 2.3.0, is still a lot slower than 2.1.x.
I have run the benchmark
```ruby
require "benchmark"
time = []
5.times do
result = Benchmark.realtime { 800000.times do Dir['... -
02:29 PM Bug #12220 (Closed): Why does Coverage keep previously-loaded files as empty arrays?
- Applied in changeset r54504.
----------
* ext/coverage/coverage.c: Fully reset coverage to not persist global state.
It was returning old file coverages as empty arrays to the user.
[ruby-core:74596] [Bug #12220]
* ext/coverage/cove... -
- Yusuke Endoh wrote:
> Xavier, thank you for your opinion. I also don't think that it is a good idea to change because of a spec, but I don't have a strong objection if the OP of #4796 is satisfied. So I agree with eregon's patch. Ere... -
- Thank you for your reply.
Xavier Shay wrote:
> Hello, I filed #4796. It was a long time ago, and I don't remember what I was trying to do. Likely something around a "coverage per test" feature.
> ...
Please read my argument fully, f... -
01:30 PM Bug #12220: Why does Coverage keep previously-loaded files as empty arrays?
- Xavier, thank you for your opinion. I also don't think that it is a good idea to change because of a spec, but I don't have a strong objection if the OP of #4796 is satisfied. So I agree with eregon's patch. Eregon, could you commit i...
-
12:57 AM Bug #12220: Why does Coverage keep previously-loaded files as empty arrays?
- Hello, I filed #4796. It was a long time ago, and I don't remember what I was trying to do. Likely something around a "coverage per test" feature.
> Right, there is a compatibility concern, and I would like to check simplecov about it... -
- It was returning old file coverages as empty arrays to the user.
[ruby-core:74596] [Bug #12220]
* ext/coverage/coverage.c (rb_coverages): remove unused static state.
* thread.c: Moved and renamed coverage_clear_result_i to reset_covera... -
01:41 PM Revision 6a79276e (git): configure.in: fix lgamma_r condition
- * configure.in (rb_cv_lgamma_r_m0): fix the condition for
lgamma_r(-0.0). [Bug #12249]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54503 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
11:49 AM Feature #11322: OpenUri: RuntimeError: HTTP redirection loop
- i've had a similar issue with `open('http://www.replayjeans.com/us/shop/product/women/jumpers-knitwear/neoprene-printed-sweatshirt/pc/48/c/61/sc/-1/1962')`
which i solved modifying D.E. Akers workaround a bit:
~~~
url = 'http://ww... -
11:17 AM Bug #12252 (Closed): Clarify Thread exception handling documentation.
- The current documentation for Thread exception handling in vm.c is somewhat unclear. I've tried to explain the behaviour more clearly.
- 08:40 AM Revision 3f1fc5e4 (git): * tool/downloader.rb (RubyGems.download): follow the change of the
- rubygems ssl_certs directory tree introduced by previous commit.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54502 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * bootstraptest/test_thread.rb: rescue resource limitation errors.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54501 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
06:01 AM Revision 469bac0f (git): * lib/rubygems.rb, lib/rubygems/*, test/rubygems/*: Update rubygems-2.6.3.
- Please see entries of 2.6.3 on
https://github.com/rubygems/rubygems/blob/master/History.txt
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54500 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * configure.in (rb_cv_lgamma_r_m0): check if lgamma_r(-0.0)
returns negative infinity. [Bug #12249]
* math.c (ruby_lgamma_r): define by the configured result.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54499 b2dd03c8-39d4-4d8f... -
- * test/openssl/test_ssl.rb (test_get_ephemeral_key): skip
unavailable public keys. [GH-1318]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54498 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * lib/logger.rb (Logger#level=): remove unnecessary local
variable.
* lib/logger.rb (Logger#initialize, Logger#reopen): [DOC] mention
the default values. cherrypicked from [GH-1319].
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trun... -
- And seems a usual "incompatible library version" error.
-
- * math.c (ruby_lgamma_r): missing/lgamma_r.c is used on Windows,
since msvcrt does not provide it.
* missing/lgamma_r.c (lgamma_r): fix lgamma(-0.0).
[ruby-core:74823] [Bug #12249]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@5...
04/05/2016
-
- Ruby 2.0.0 is EOL
-
03:41 PM Bug #12250 (Rejected): Ruby segfaults on `foodcritic .`
- Trace:
```
Downloads$ foodcritic .
/Users/andrew.pennebaker/.gem/ruby/2.0.0/gems/nokogiri-1.6.6.2/lib/nokogiri/nokogiri.bundle: [BUG] Segmentation fault
(snip)
Abort trap: 6
``` -
08:46 PM Bug #12249: Math.gamma(-0.0) should return -Infinity
- Regarding `Math.lgamma(-0.0)` (r54494,r54495):
It's actually caused by the `lgamma_r()` fallback implementation. A non-invasive fix (I don't think adding `if(x==0)` would be worthwhile):
~~~diff
diff --git a/missing/lgamma_r.c b/m... -
- Nobuyoshi Nakada wrote:
> Applied in changeset r54492.
> ...
Thanks nobu! -
- Also `Math.lgamma(-0.0)`.
-
- Applied in changeset r54492.
----------
math.c: fix tgamma
* math.c (ruby_tgamma): fix tgamma(-0.0) on mingw.
[ruby-core:74817] [Bug #12249] -
- On Windows, Math.gamma(-0.0) returns +Infinity instead of -Infinity,
which is the correct result according to tgamma(3) and the behavior on other platforms.
Is this OK to fix in Ruby?
There is already a similar fix for #10642. -
- This came from https://github.com/ruby-grape/grape/issues/1348
~~~
require 'delegate'
require 'ostruct'
class AttributeTranslator < DelegateClass(OpenStruct)
def initialize(attributes = {})
ostruct = OpenStruct.new(attrib... - 04:23 PM Revision bb7a4c2a (git): * math.c (ruby_lgamma_r): mswin's lgamma_r also seems to be wrong.
- cf. [Bug #12249]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54495 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * math.c (ruby_lgamma_r): fix lgamma(-0.0) on mingw and OSX.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54494 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 03:52 PM Revision 8c4c3508 (git): * 2016-04-06
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54493 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * math.c (ruby_tgamma): fix tgamma(-0.0) on mingw.
[ruby-core:74817] [Bug #12249]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54492 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- ruby_2_3 r54491 merged revision(s) 53419.
-
- * lib/rubygems/security.rb (DIGEST_ALGORITHM, KEY_ALGORITHM):
should check same name as the used constants.
[ruby-core:72674] [Bug #11940]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_3@54491 b2dd03c8... -
10:16 AM Bug #9198 (Closed): Segfault in TestException#test_machine_stackoverflow
- This test was disabled by r49550 => closing this issue.
-
- * string.c (rb_str_concat_literals): concatenate literal string
fragments.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54490 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
06:03 AM Bug #12202: nkf.c static void mime_putc the variable being used for loop used for outer loop too
- Anton Sivakov wrote:
> Yui NARUSE wrote:
> ...
fixed, thanks! -
- [Bug #12202] [ruby-core:74802]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54489 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Rubygems fixed bug that was introduced by Rubygems 2.4.5 on Ruby 2.2.0.
This bug is backported 2.4.x branch on rubygems upstream repository.
* https://github.com/rubygems/rubygems/issues/1256
* https://github.com/rubygems/rubyge...
04/04/2016
-
10:16 PM Feature #12247: accept multiple arguments at Array#delete
- +1 I needed this a few days ago
-
- With gcc version 5.3.0 20151204 (Ubuntu 5.3.0-3ubuntu1~14.04), following result:
```
% time ./miniruby54488 -e'i=1000000;while (i-=1)>0;[1,2,3,4,5].delete(1);[1,2,3,4,5].delete(1);[1,2,3,4,5].delete(1);[1,2,3,4,5].delete(1);[1,2,3,4... -
03:02 PM Feature #12247: accept multiple arguments at Array#delete
- Sebastian Skalacki wrote:
> Anyway, I think that the return value could be better described in docs.
Agreed.
Please discuss better behavior to decide what we should describe in docs :-P
-
- Tsuyoshi Sawada wrote:
> depending on the length of `array`, you get different types of results. It can be a gotcha.
At first, I also think so, but nobu told me a precedent, `p`.
In fact, compatibility and convenience would be more ... -
02:48 PM Feature #12247: accept multiple arguments at Array#delete
- If the return value has as many items as arguments passed to the method and the items order is preserved, then the return value decomposition will be possible, for example:
~~~ruby
ary = %w[c a e c]
a, b, c = ary.delete("a", "b", "c... -
02:45 PM Feature #12247: accept multiple arguments at Array#delete
- I like the idea, but I feel a slight inconsistency of the proposed output with the existing behavior. When the argument is single, you get the relevant element. But when the argument is multiple, the return value becomes an array.
~~~... -
- With my implementation (attached patch),
```ruby
ary = [1, 2, 3, 4, 5]
ary.delete(1, 6) # => [1]
ary # => [2, 3, 4, 5]
```
Because if you want to get `[1, nil]`, you can write:
```ruby
ary = [1, 2, 3, 4, 5]
ar... -
01:48 PM Feature #12247: accept multiple arguments at Array#delete
- I guess this is a simple change. I have nothing against it.
As for the last question:
```ruby
ary = [1, 2, 3, 4, 5]
ary.delete(1, 6) # => [1, nil]
ary # => [2, 3, 4, 5]
```
I guess that would be the most consist... -
- Given:
~~~
ary = [1, 2, 3, 4, 5]
~~~
What is going to be returned by:
~~~
ary.delete(1, 6)
~~~
? -
07:03 AM Feature #12247: accept multiple arguments at Array#delete
- +1 I needed this before too. For API consistency I would also want Hash#delete to accept multiple argumenrts as well.
-
04:45 PM Bug #10642: 2.2: TestMath#test_gamma test failure (MinGW)
- This issue keep failing in ruby_2_2 branch.
Please backport r49913. -
04:12 PM Revision 8fba800f (git): fix typos [ci skip]
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54488 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * test/openssl/test_ssl.rb (test_get_ephemeral_key): should use
assert_instance_of instead of comparison of classes.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54487 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 03:06 PM Revision cd2eb1f9 (git): * 2016-04-05
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54486 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * ext/openssl/ossl_ssl.c (ossl_ssl_tmp_key): Access to ephemeral
TLS session key in case of forward secrecy cipher. Only
available since OpenSSL 1.0.2. [Fix GH-1318]
* ext/openssl/extconf.rb: Check for SSL_get_server_tmp_key.
git-... -
- Applied in changeset r54484.
----------
at_exit list
* vm_core.h (rb_vm_struct): make at_exit a single linked list but
not RArray, not to mark the registered functions by the write
barrier. based on the patches by Evan Phoenix.
... -
- `at_exit` functions should be called in the inverse order, LIFO.
-
- * vm_core.h (rb_vm_struct): make at_exit a single linked list but
not RArray, not to mark the registered functions by the write
barrier. based on the patches by Evan Phoenix.
[ruby-core:73908] [Bug #12095]
git-svn-id: svn+ssh://c... -
- I think that would make sense.
I usually add an extra method call afterwards to round it manually (or, if not .round then to use it as a string and sprintf, % or .ljust / .rjust output it).
I should also note that newcomers often a... -
10:35 AM Bug #12202: nkf.c static void mime_putc the variable being used for loop used for outer loop too
- Yui NARUSE wrote:
> Applied in changeset r54423.
> ...
Hi. I saw your changes and now it's:
...
for (j=0;i<mimeout_state.count;j++) {
(*o_mputc)(mimeout_state.buf[j]);
...
Now you're using 'j' but on condition you still using... -
- escalating URL spec: https://github.com/whatwg/url/issues/112
- 08:49 AM Revision dc58fbea (git): * remove trailing spaces.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54483 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
08:49 AM Revision 02f75074 (git): * gc.c: change deafult value of
- RUBY_GC_HEAP_FREE_SLOTS_MIN_RATIO 0.3 -> 0.2
RUBY_GC_HEAP_FREE_SLOTS_MAX_RATIO 0.8 -> 0.65
These values are same as Ruby 2.0.0.
This change cause GC counts.
However, generational GC reduced each (minor) GC time and
increase m... -
- And also take an accept_zero flag which allow to accept zero
even if lower_bound is set.
* gc.c (ruby_gc_set_params): fix parameters.
RUBY_GC_HEAP_FREE_SLOTS_MAX_RATIO set 0.9 as *lower_bound*, so that
it should be upper_bound.
... - 07:42 AM Revision 60f14c29 (git): * 2016-04-04
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54480 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54479 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Does this help? RBasic and RBasicRaw does in fact occupy identical memory region so making them union seems the most natural way.
~~~patch
diff --git a/internal.h b/internal.h
index 3970431..498a7e3 100644
--- a/internal.h
+++ b/i... -
- Usaku NAKAMURA wrote:
> One more result of mswin64 (Visual C++ 2013):
> ...
Thank you for benchmark.
Hmm Visual C++'s optimization is not well developed at 2013 (and [Ruby doesn't support VC2015](https://bugs.ruby-lang.org/issues/1111...
04/03/2016
-
- ruby_2_3 r54478 merged revision(s) 53495.
-
- * compile.c (compile_massign_lhs): when index ends with splat,
append rhs value to it like POSTARG, since VM_CALL_ARGS_SPLAT
splats the last argument only. [ruby-core:72777] [Bug #11970]
git-svn-id: svn+ssh://ci.ru... -
- I found that it's very useful if `Array#delete` accepts multiple arguments.
```ruby
ary = [1, 2, 3, 4, 5]
ary.delete(1, 3) #=> [1, 3]
ary #=> [2, 4, 5]
``` -
- * numeric.c (flodivmod): round division if it is a finite number
and module is required.
* numeric.c (dbl2ival): do not round here.
* numeric.c (flo_ceil): use dbl2ival.
* numeric.c (flo_round): round explicitly.
git-svn-id: svn+ssh:/...
04/02/2016
- 10:38 PM Revision 2b829860 (git): * 2016-04-03
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54476 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * numeric.c (fix_lshift, fix_rshift, flo_truncate): move forward
declaration to the top.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54475 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- The patch provides several language improvements for doc/extension.rdoc.
-
- Why this new method is too slow seems because of keyword argument.
-
- We sometimes calculates `integer - integer % num`.
For example time series events into time partitions, we write code like
```ruby
event # {time: 1459580435, name: "hoge", text: "Rawr!"}
partition = event[:time] - event[:time] % ... -
- Yusuke Endoh wrote:
Thank you for your reply.
> Benoit Daloze wrote:
> ...
Yes, I already read it but it seems hard to understand the wanted behavior.
The particular test case shows nothing of the wanted behavior unfortunately.
... -
- Related to [Feature #12244].
I think that `Integer#floor`, `Integer#ceil`, `Float#floor`, and `Float#ceil` should have optional parameter `ndigits`, as well as `round` methods.
https://github.com/ruby/ruby/compare/trunk...nobu:featur... -
02:49 PM Bug #12243: Segmentation fault at 0x0000000000001c
- Nobuyoshi Nakada wrote:
> How to reproduce it?
It happens when I parse a large XML File with libxml-ruby and write the results with rails active record in the database. I would upload the xml for you to test but its compressed with g... -
- How to reproduce it?
-
-
08:42 AM Feature #12241: super end
- NOOOO GOOD!! is ruby want become to perl?
-
- Robert A. Heiler wrote:
> Edit: Or perhaps "FINAL_END"
It isn't always "FINAL".
```ruby
module MyModule
class MyClass
def my_method
10.times do
if rand < 0.5
p :small
!end
def another... -
- Martin Dürst wrote:
> I think this should be end!, to look like a Ruby method name :-).
Since it conflicts with the existing syntax, and ruby-mode.el considers it an ordinary method call, autoindent doesn't work.
-
-
- Martin Dürst wrote:
> In addition, a proposal for adding black and white circled numbers
> ...
HOLY UNICODE!!
-
- * numeric.c (flo_round, int_to_s, int_round): use rb_check_arity
instead of rb_scan_args for a simple optional argument.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54474 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 06:24 AM Revision 9731a745 (git): * 2016-04-02
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54473 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * include/ruby/intern.h (rb_check_arity): returns argc.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54472 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
04/01/2016
-
- /Users/fero46/.rvm/gems/ruby-2.2.3/gems/activerecord-4.2.6/lib/active_record/relation/query_methods.rb:574: [BUG] Segmentation fault at 0x0000000000001c
(snip bug-12243.log)
terminated by SIGIOT
------------------------ AND THE DU... -
- * ext/ripper/lib/ripper/lexer.rb (on_heredoc_dedent): Fix
Ripper.lex error in dedenting squiggly heredoc. heredoc tree is
also an array of Elem in the outer tree. [Fix GH-1234]
git-svn-id: svn+ssh://ci.ruby-lang.o... -
- You mean you suggest to have the documentation better?
The UI can be a bit confusing I agree, I'd like to have a separate tracker for documentation, since docu is definitely not a "bug", but also not 100% a real "feature" (since the l... -
01:16 PM Feature #12242: Is it worth adding collision probability of SecureRandom functions in RubyDoc?
- Apologies if I opened this in a wrong format. I'm pretty new to redmine tracker, and unfamiliar with the UI :(
-
- Is it feasible and/or even useful to document the probability of
collisions for the various random number generation methods in the
SecureRandom module?
I feel having a note to read about the possibility of collisions
happening for... -
- You almost got me with this one!
Also see the "endddd" suggestion. :)
Edit: Or perhaps "FINAL_END" -
06:32 AM Feature #12241: super end
- Nobuyoshi Nakada wrote:
> This proposal will introduce "super end", `!end`, which ends the block began at the same column with `!`.
I think this should be end!, to look like a Ruby method name :-). -
- This is a refinement of [Feature #5054].
It required to count the depth of count to end, and tiresome.
This proposal will introduce "super end", `!end`, which ends the block began at the same column with `!`.
~~~ruby
module MyM... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54470 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- swapcase functionality for titlecase characters. Swapcase isn't defined
by Unicode, because the purpose/usage of swapcase is unclear anyway.
The implementation follows a proposal from Nobu, swaping the case of
each component of a t... -
- One more result of mswin64 (Visual C++ 2013):
```
Speedup ratio: compare with the result of `ruby 2.4.0dev (2016-04-01 trunk 54468) [x64-mswin64_120]' (greater is better)
name built-ruby
loop_whileloop 1.058
vm1_attr_ivar... -
- mswin64 result:
```
Speedup ratio: compare with the result of `ruby 2.4.0dev (2016-04-01 trunk 54468) [x64-mswin64_100]' (greater is better)
name built-ruby
loop_whileloop 0.880
vm1_attr_ivar* 0.995
vm1_attr_ivar... -
07:13 AM Feature #12142: Hash tables with open addressing
- > average improvement on MRI hash benchmarks is close to 40% on Intel Haswell and >55% on ARMv7
What about realworld applications like Redmine? -
03:41 AM Feature #12142: Hash tables with open addressing
- Eric Wong wrote:
> vmakarov@redhat.com wrote:
> ...
Thank you for sharing this. I am still working on the hash tables when I have spare time.
I already implemented variable indexes (8-, 16-, 32-, 64-bits). It gave about 3% average... -
- On 2016/04/01 15:10, nobu@ruby-lang.org wrote:
> ~~~ruby
> [[1, 2], [3, 4]].map {①+②}
> ~~~
>
> `①`..`⑳` denote the arguments upto 20.
I admire your modesty. Unicode's circled numbers go up to 50 (see
http://www.unicode.org/... -
- Often improvements about implicit block argument have been proposed, but in vain.
Such as:
~~~ruby
[[1, 2], [3, 4]].map {|a1, a2| a1+a2} #=> [3, 7]
~~~
to
~~~ruby
[[1, 2], [3, 4]].map {_1+_2}
~~~
Now, we (ko1 and nobu) f... -
- Applied in changeset r54468.
----------
improve git repository detection
* configure.in (AC_CONFIG_FILES): $srcdir/.git can be a file pointing
the real git_dir, such as when the git working tree is a "linked
working tree" (a workin... -
04:30 AM Bug #12239 (Closed): configure script does not detect Git when in a working tree with non-directory .git
- `make up` doesn't work well on a Git working tree created by `git-worktree add`.
~~~sh
% git clone https://github.com/ruby/ruby.git ruby-src
% cd ruby-src
% git worktree add -f test-tree trunk
% cd test-tree
% autoconf && ./confi... -
- * configure.in (AC_CONFIG_FILES): $srcdir/.git can be a file pointing
the real git_dir, such as when the git working tree is a "linked
working tree" (a working tree created by git-worktree). So use
git-rev-parse --git-dir to check ... -
- Seems intended,
> Logger#add(severity, message = nil, progname = nil) { ... }
> ...
but maybe a bug.
```diff
diff --git a/lib/logger.rb b/lib/logger.rb
index 8393806..9a05518 100644
--- a/lib/logger.rb
+++ b/lib/logger.rb
@@ ... - 03:27 AM Revision 34c0c840 (git): * tool/release.sh: BRE accepts \{m,\} expression. use it for future many-digits
- version numbers.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54467 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
03/31/2016
-
08:33 PM Bug #12126 (Closed): [PATCH] openssl: accept moving write buffer for write_nonblock - Applied in changeset r54466.
----------
openssl: accept moving write buffer for write_nonblock
By setting the SSL_MODE_ACCEPT_MOVING_WRITE_BUFFER flag.
This flag was introduced at the same time as
SSL_MODE_ENABLE_PARTIAL_WRITE in OpenS... - 08:33 PM Revision 280f7322 (git): openssl: accept moving write buffer for write_nonblock
- By setting the SSL_MODE_ACCEPT_MOVING_WRITE_BUFFER flag.
This flag was introduced at the same time as
SSL_MODE_ENABLE_PARTIAL_WRITE in OpenSSL 0.9.4 and makes usage
with non-blocking sockets much easier.
Before this, a Rubyist would nee... -
07:32 PM Feature #12142: Hash tables with open addressing
- vmakarov@redhat.com wrote:
> I think I'll have a few patches when I am done with the hash
> tables: the hash table itself, hash functions, code for
> recognizing a denial attack and switching to stronger hash
> functions. I am not ... -
07:13 PM Bug #12238 (Rejected): Logger writes progname if log message is nil
- Please consider the following example:
``` ruby
logger = Logger.new($stdout)
# Log message is nil
logger.info
# I, [2016-03-31T22:04:36.684636 #4019] INFO -- : nil
logger.progname = 'prog'
# Now log message is changed to ... -
- Yusuke Endoh wrote:
> Then, you want `Coverage.result` to remove all arrays from the Hash once, keep them in back, and automatically restore an array when the corresponding code is executed?
I am not sure what you mean by "keep them ... -
- Then, you want `Coverage.result` to remove all arrays from the Hash once, keep them in back, and automatically restore an array when the corresponding code is executed? This is the third idea to make it "intuitive" that I thought of but...
-
- Yusuke Endoh wrote:
> I'm unsure what behavior you expect. Do you mean that it should pass the following two tests?
I expect the second `Coverage.result` to return:
~~~RUBY
{"/tmp/test.rb" => [0, 1, nil]}
~~~
In other words,... -
- I'm unsure what behavior you expect. Do you mean that it should pass the following two tests?
~~~~
Coverage.start
require tmp + '/test.rb'
Coverage.result
Coverage.start
assert_equal {}, Coverage.result
~~~~
~~~~
Coverage.s... -
- Yusuke Endoh wrote:
> It was actually intentional. Please read the test case of #4796.
> ...
That is because the method `coverage_test_method` actually executes some lines in `"/tmp/test.rb"`, and so the corresponding array should not... -
- Benoit Daloze wrote:
> Tsuyoshi Sawada wrote:
> ...
It was actually intentional. Please read the test case of #4796.
~~~~
Coverage.start
require tmp + '/test.rb'
Coverage.result
Coverage.start
... -
- Tsuyoshi Sawada wrote:
> So, was it or was it not a bug that it returns empty arrays?
I consider it a bug in that it seems to be unintentional and it is just a side-effect of the implementation to disable coverage when reset.
I at... -
- So, was it or was it not a bug that it returns empty arrays?
-
- After looking into this, I discovered a bug!
And I also found the reason why it's returning empty arrays: it is leaking implementation details.
It is however easy to fix but let's start with fixing the bug first. -
- ko1@atdot.net wrote:
> * We need to manage x4 more pages. We need to avoid O(n) (n:
> page number) operations from the GC process.
I think we can use ccan/list with O(1) insert/delete without
iterations instead of relying on f... -
- # Abstract
I propose `mmap` (or similar API on Windows) managed heap.
Now, each heap pages are allocated by posix_memalign(), 16KB for each pages.
I wrote a patch to manage heap pages allocated by bulk `mmap` (256MB),
and provide... -
- Applied in changeset r54465.
----------
* thread.c (update_coverage): Do not track coverage in loaded files
after Coverage.result. Avoids out-of-bounds access. [Bug #12237]
* ext/coverage/coverage.c (coverage_clear_result_i): document. -
- Benoit Daloze wrote:
> Can I commit the fix in thread.c?
Sure, thank you very much!
--
Yusuke Endoh <mame@ruby-lang.org> -
- Attached proposed fixing patch.
-
- While looking at #12220, I found a bug.
Attached is a patch that reveals it.
Running
$ make test-all TESTS="test/coverage -n /test_restarting_coverage/"
gives
[1/1] TestCoverage#test_restarting_coverage
Updating cov for /tmp/d201... -
- after Coverage.result. Avoids out-of-bounds access. [Bug #12237]
* ext/coverage/coverage.c (coverage_clear_result_i): document.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54465 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 03:42 PM Revision 674725f1 (git): * test/rubygems/test_gem_commands_environment_command.rb (test_execute):
- version number may be two-digit.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_1@54462 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Akira Tanaka wrote:
> In the yesterday's meeting,
> ...
I am strongly in favor of having something like Thread#report_on_exception, defaulting to true.
If a Thread can support known exceptions, it can rescue them explicitly.
If a Thr... - 03:13 PM Revision f0060b81 (git): * tool/release.sh: two-digit version number support.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54461 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 03:04 PM Revision 69eee71c (git): * 2016-04-01
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54460 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 03:04 PM Revision a517e6f9 (git): * tool/merger.rb (version): version number may be two-digit.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54459 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 03:02 PM Revision 03a4a200 (git): * version.h: commit miss
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_1@54457 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 03:01 PM Revision ed3dedae (git): * version.h: bump to 2.1.10. welcome to two-digit version!
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_1@54456 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * gc.c (heap_extend_pages): fix format specifiers for size_t.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54455 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54454 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
09:53 AM Bug #12235 (Closed): URI.encode issue with square brackets
- URI::Encode doesn't correctly handle square brackets in the path of a URL:
> URI.encode('http://example.com/resource[1].doc')
=> "http://example.com/resource[1].doc"
It should return "http://example.com/resource%5B1%5D.doc... -
- Introduce new environement variable
GC_HEAP_FREE_SLOTS_GOAL_RATIO (goal_ratio) to calculate the ratio
of additional memory.
Before this change, we add pages with the following formula
(when free_slots < total_pages * RUBY_GC_HEA... -
- Martin Dürst wrote:
> Benoit Daloze wrote:
> ...
That's my point. Enumerable methods should work on any collection implementing #each.
Not only on #each returning a numeric-type or sth with a #+/#- method (and then the result of #- sh... -
- Benoit Daloze wrote:
> It seems to me Enumerable is not the right place for this, because it expects more than just #each.
The code is currently written in terms of #length and #[], but this can easily be fixed to use #each.
> ...... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54452 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Before this change, heap_pages_min_slots are calculated at the
beggining sweeping phase. And this value is used at the end of
*next* marking phase.
To simplify it, we use this value at the end of this marking phase.
It means tha... -
- * RUBY_GC_HEAP_FREE_SLOTS_MIN_RATIO:
allocate additional pages when free slots is lower than
the value (total_slots * (this ratio)).
* RUBY_GC_HEAP_FREE_SLOTS_MAX_RATIO:
allow to free pages when free slots is greater thatn
... -
- It's wrong code.
The second argument must be '/C'.
```ruby
exec(['cmd.exe', '/C'], '/C', 'echo', 'argv_zero')
``` -
05:07 AM Bug #12234 (Rejected): Kernel#exec with a command array broken on Windows as of Ruby 2.2.4
- On Ruby 2.0.0 this worked as expected:
```
C:\Users\Administrator\GitHub\spec>ruby -ve "exec(['cmd.exe', '/C'], 'echo', 'argv_zero')"
ruby 2.0.0p247 (2013-06-27) [i386-mingw32]
argv_zero
```
On Ruby 2.2.4 it is broken - it seems ... - 07:02 AM Revision fb296fe6 (git): * remove trailing spaces.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54449 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * gc.c (gc_verify_heap_pages): check also tomb heap.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54448 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * gc.c (gc_sweep_step): use returned free slots count.
* gc.c (gc_sweep_step): change variable name `next'
to `next_sweep_page'.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54447 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * complex.c (get_dat1, get_dat2): turn into pure variable
declarations only, not mixed code and declarations.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54446 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * ext/date/date_core.c (d_lite_strftime, dt_lite_strftime): [DOC]
fix indent not to be a big sole verbatim.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54445 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r54444.
----------
date_core.c: fix DateTime rdoc
* ext/date/date_core.c (Init_date_core): [DOC] fix misplaced doc
of DateTime. [ruby-core:74729] [Bug #12233] -
- * ext/date/date_core.c (Init_date_core): [DOC] fix misplaced doc
of DateTime. [ruby-core:74729] [Bug #12233]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54444 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Fixed at r53722.