Activity
From 12/04/2014 to 12/10/2014
12/10/2014
- 08:38 PM Revision 0eebb8f1 (git): * lib/prime.rb: Remove useless loop and block capture.
- See [#10354]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48767 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
07:35 PM Bug #10511 (Feedback): fix for #9940 causes dramatic performance regression
- could you try r48765?
-
- Applied in changeset r48765.
----------
* vm_core.h: introduce new field
rb_thread_t::local_storage_recursive_hash... - 07:35 PM Revision 05f087c8 (git): * 2014-12-11
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48766 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- rb_thread_t::local_storage_recursive_hash_for_trace to store
recursive hash to avoid creating new recursive (nested... -
- Sorry for late.
> Benchmark results:
>
> vm2_struct_big_aref_hi* 1.305
> ...
how to read? seconds? compared... -
04:47 PM Feature #10575: [RFC] struct: avoid all O(n) behavior on access
- Couple of other struct optimizations in #10585
-
04:34 PM Bug #10583: Process.spawn stalls forever opening named pipes (fifo)
- While open(2) blocks, the spawn should NOT block. Only the spawned process itself should block waiting on the open()...
-
08:23 AM Bug #10583 (Rejected): Process.spawn stalls forever opening named pipes (fifo)
- `open(2)` either blocks until the file gets ready or fails with `ENXIO`.
Open in a thread. -
- 0001 - Define optimized setters for first 10 attributes.
0002 - Cache members definition in an subclasses - it is ... -
01:13 PM Feature #8566 (Open): [PATCH] Allow to configure additional preludes
- Unfortunately placing something like:
~~~
begin
require 'abrt'
rescue LoadError
end
~~~
into operating_s... -
- * string.c (rb_str_delete_bang): fill wchar terminator.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48764 b2dd... -
- * string.c (rb_str_sub_bang): fill wchar terminator.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48763 b2dd03c... -
11:21 AM Bug #10584: String.valid_encoding?, String.ascii_only? fails to account for BOM.
- This isn't as simple as you describe it. With respect to BOMs, there is a clear distinction between external data and...
-
12:55 AM Bug #10584 (Rejected): String.valid_encoding?, String.ascii_only? fails to account for BOM.
- IMO:
- A Unicode (UTF-16, UTF-32) string with a valid BOM should not be considered a valid encoding if endianness ... -
- * string.c (rb_str_splice_0): fill wchar terminator.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48762 b2dd03c... -
- * string.c (rb_str_chop_bang): fill wchar terminator.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48761 b2dd03... -
- * string.c (rb_str_chomp_bang): fill wchar terminator.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48760 b2dd0... -
- * string.c (rb_str_times): fill wchar terminator.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48759 b2dd03c8-3... -
- * string.c (rb_str_plus): fill wchar terminator.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48758 b2dd03c8-39... -
- * string.c (rb_str_plus, rb_str_splice_0): extract local
variables.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/tr... -
- * string.c (chompped_length, chomp_rs): extract from
rb_str_chomp_bang to share with rb_str_chomp.
git-svn-id: svn... -
- * test/-ext-/string/test_cstr.rb (assert_wchars_term_char): test
for all wide char encodings.
git-svn-id: svn+ssh:... -
- * struct.c (define_aref_method, define_aset_method): use iseq
VALUE instead of rb_iseq_t to prevent from GC, as RB_... -
- * test/-ext-/string/test_cstr.rb (test_wchar_long): show the
failed encoding name in error messages.
git-svn-id: s... -
- * thread.c (exec_recursive): use the same last method name as
recursive_push in the error message when recursive_po...
12/09/2014
-
08:47 PM Bug #10582: String#chomp strips off more newlines than its documentation states
- Looks good to me. Thanks!
-
12:27 PM Bug #10582: String#chomp strips off more newlines than its documentation states
- I merged https://github.com/ruby/ruby/pull/780
-
-
- Given an empty string as its parameter, String#chomp strips off as many trailing newlines as are present, plus a prec...
- 05:51 PM Revision bf4b7f3b (git): * test/net/http/test_https.rb
- (TestNetHTTPS#test_certificate_verify_failure): on Windows,
Errno::ECONNRESET will be raised when the verify is fai... -
- Ruby's implementation of Process.spawn seems to attempt to send error/success information from the spawned process ba...
-
03:44 PM
Feature #10575 (Closed): [RFC] struct: avoid all O(n) behavior on access
- Applied in changeset r48748.
----------
struct: avoid all O(n) behavior on access
This avoids O(n) on lookups with ... -
- Looks fine.
- 03:44 PM Revision 9d237f43 (git): * properties.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48750 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 03:44 PM Revision 45210a36 (git): * 2014-12-10
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48749 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 03:43 PM Revision 65651b34 (git): struct: avoid all O(n) behavior on access
- This avoids O(n) on lookups with structs over 10 members.
This also avoids O(n) behavior on all assignments on Struct... -
- Patchby @stderr [ci skip][fix GH-780]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48747 b2dd03c8-39d4-4d8f-98f... -
- Applied in changeset r48746.
----------
object.c: [DOC] Revise documentation
* object.c: [DOC] Revise documentation... -
- * object.c: [DOC] Revise documentation by Marcus Stollsteimer at
[ruby-core:66368]. [Bug #10526]
* #inspect: be ... -
04:27 AM Bug #10450: multiple assignment in conditional
- Yukihiro Matsumoto wrote:
>
> But I am afraid it's meaningless, since multiple assignment always return an array a... -
- * vm_eval.c (rb_eval_string_wrap): [DOC] Fix `rb_eval_string_wrap`
documentation. It is referencing `require` inst... -
- Applied in changeset r48744.
----------
thread.c: get rid of invalid ID symbol
* eval.c (rb_frame_last_func): retur... -
- * eval.c (rb_frame_last_func): return the most recent frame method
name.
* thread.c (recursive_list_access): use th... - 01:05 AM Revision 1b13e1e6 (git): * 2014-12-09
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48743 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * ruby.c (ruby_init_loadpath_safe): get rid of modify shared
string directly.
git-svn-id: svn+ssh://ci.ruby-lang.o...
12/08/2014
-
- One of the simplest code to reproduce it is:
~~~
$ ruby -e '{["foo"]=>nil}; p Thread.current[:__recursive_key__]'... -
- Not only openssl.so, resolv.rb too.
The cause is `exec_recursive` used by `rb_hash_any` stores the current method ... -
02:12 PM Bug #10579: Segmentation fault at 0x00000000000000
- The root cause of this issue is openssl.so.
~~~
C:\>ruby -ropenssl.so -e "ObjectSpace.each_object{|obj| p obj}"
... -
12:54 PM Bug #10579 (Closed): Segmentation fault at 0x00000000000000
- Following script cause a Segmentation fault
~~~
$ ruby -rresolv -e 'ObjectSpace.each_object { |obj| p obj }' > te... -
10:20 PM Bug #10581 (Rejected): OpenSSL::ASN1 fails to decode DKIM public keys
- Recently I needed to decode a DKIM public key. I found that OpenSSL::ASN1 failed to decode the key, while Krypt::ASN1...
-
08:44 PM Feature #10499: Eliminate implicit magic in Proc.new and Kernel#proc
- Adding a deprecation warning would be easy if we can get buy-in from matz.
matz: ball's in your court, I think! -
08:20 PM Bug #10564: DelegateClass, method_missing, and instance_eval -- different behavior in ruby 2.1
- Okay, thanks.
What I don't understand is, if `method_missing` is not defined in the `< DelegateClass(Object)` cla... -
- Testing with Fedora Rawhide:
~~~
1) Error:
TestProcess#test_deadlock_by_signal_at_forking:
Timeout::Error: ex... -
01:40 PM Feature #10510: Remove REXML instead of patching it
- Erik Michaels-Ober wrote:
> I believe semantic versioning prevents doing this until Ruby 3 is released (many years f... -
- Was this resolved somehow? Testing this on Fedora 20, where the FP mangling was enabled [1], Ruby 2.0.0 fails:
htt... -
10:09 AM Bug #10578 (Closed): Allow undefining methods in refinements
- Since refinements are locally scoped monkey patches, and it is possible and useful to undef a method globally, I thin...
-
- Multiple assignment returns the object on the right hand side, *before* `to_ary` is called on it:
~~~Ruby
irb(mai... - 04:18 AM Revision 3af8e150 (git): * 2014-12-08
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48741 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * struct.c (rb_struct_set): get rid of calling rb_frame_this_func
twice.
git-svn-id: svn+ssh://ci.ruby-lang.org/ru... -
02:00 AM Bug #10466: rb_eval_string_wrap does not actually wrap in a module binding
- You can reproduce this without C code also.
```ruby
Module.new {X = 5} #=> #<Module:0x000000012186c0>
Module.ne...
12/07/2014
-
- Yukihiro Matsumoto wrote:
>
> But I am afraid it's meaningless, since multiple assignment always return an array a... -
03:34 PM Bug #10450 (Rejected): multiple assignment in conditional
- It's a limitation of LALR syntax defined by yacc.
Wrap the multiple assignment in parentheses.
```
if (a,b = 1,2... -
05:40 PM Bug #10317: JSON::State is not using the parameter space_before
- Sent pull request to the JSON implementation repo:
https://github.com/flori/json/pull/225 -
02:01 PM Revision f975d115 (git): fix a typo [ci skip]
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48739 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 01:59 PM Revision f53a1931 (git): * test/net/imap/test_imap.rb (ImapTest#imaps_test): forced error on SSL sockets
- causes Errno::ECONNRESET or Errno::ECONNABORTED instead of Errno::EPIPE on
Windows.
git-svn-id: svn+ssh://ci.rub... -
01:03 PM Bug #10577 (Rejected): Confusing behavior of File.open with bad filenames in windows
- Google "NTFS Alternate Data Streams".
-
12:44 PM Bug #10577 (Rejected): Confusing behavior of File.open with bad filenames in windows
- I've used a colon in a filename in my script and confronted with a confusing behavior when ran program in windows. Ru...
-
10:40 AM Revision eebb6bfc (git): * ext/socket/basicsocket.c, ext/socket/sockssocket.c:
- remove code for $SAFE=4.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48737 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * lib/rdoc/known_classes.rb: reverted regression changes of rdoc
known class.
git-svn-id: svn+ssh://ci.ruby-lang.o... - 02:20 AM Revision 512705e6 (git): io.c: Typo close -> closes.
- Patch by @cirosantilli [Fixes GH-757]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48735 b2dd03c8-39d4-4d8f-98f... - 01:22 AM Revision b75be6b7 (git): * properties.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48734 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 01:22 AM Revision 4b0ebe8f (git): * remove trailing spaces.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48733 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
01:22 AM Revision 4260aa1d (git): * lib/rdoc: Update to RDoc 4.2.0.
- * test/rdoc: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48732 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 12:53 AM Revision 7d2d6d0a (git): * remove trailing spaces.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48731 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 12:53 AM Revision 04a79e46 (git): * 2014-12-07
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48730 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * test/rubygems: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48729 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
12/06/2014
-
07:20 PM Bug #10576: [DOC] Several fixes for the BigDecimal documentation
- Patch 3:
Improve documentation for #round and #truncate. -
- Patch 1:
Clarify the return value types of abs, fix, and frac.
(They return the results as a BigDecimal; target... -
07:14 PM Bug #10450: multiple assignment in conditional
- It seems wierd to me that this is a syntax error.
~~~
def foo
a,b = (1,2)
end
if foo
put 'hello'
end
~~... -
- I submitted a patch that clarifies the documentation accordingly, see issue #10576 (and provides some additional impr...
-
01:13 AM Revision a0b80a44 (git): * lib/net/imap.rb: Fix undefined variable usage & refactor/DRY
- code. Patch by @aledovsky. [Fixes GH-770]
* test/net/test_imap.rb: related test.
git-svn-id: svn+ssh://ci.ruby-lan... - 01:10 AM Revision f9e3213f (git): thread.c (do_select): rename parameters to avoid shadowing
- Sometimes we may use write(2) syscalls for debugging inside
this function.
git-svn-id: svn+ssh://ci.ruby-lang.org/ru... -
12:31 AM Feature #10575 (Closed): [RFC] struct: avoid all O(n) behavior on access
- This avoids O(n) on lookups with structs over 10 members.
This also avoids O(n) behavior on all assignments on Struc... -
- * test/rake: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48726 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
12/05/2014
- 09:49 PM Revision e70210ca (git): compile.c (rb_iseq_build_from_ary): remove misc handling
- iseq->local_size is calculated, too, along with stack_size and
param.size, meaning there is no need to blindly load v... -
- * ruby.c (process_options): safe_level 4 was removed.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48724 b2dd03... -
- Applied in changeset r48723.
----------
configure.in: fix --with-ruby-version [ci skip]
* configure.in (RUBY_LIB_VE... -
- ~~~
$ configure --with-ruby-version=foo
...
snip
...
$ make
...
snip
...
In file included from loadpath.... -
- * configure.in (RUBY_LIB_VERSION): Fix --with-ruby-version
configuration option. get rid of quoting in config.stat... -
- * io.c (Init_IO): ARGF is not a class.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48722 b2dd03c8-39d4-4d8f-98... -
- With the expected behavior, i.e.
~~~
irb(main):013:0> 'foobar'.lchomp('foo')
=> "bar"
irb(main):014:0> 'foobar'... - 07:48 PM Revision d0352860 (git): * test/test_pp.rb (PPTestModule::PPFileStatTest): test for r48719.
- is this the right place here?
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48721 b2dd03c8-39d4-4d8f-98ff-823f... - 07:35 PM Revision b614474e (git): * 2014-12-06
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48720 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 07:35 PM Revision d2cee782 (git): * lib/pp.rb (File::Stat#pritty_print): some platforms (such as Windows)
- does not have major/minor parts of device.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48719 b2dd03c8-39d4-4d... -
05:05 PM Bug #10573 (Rejected): bigdecimal, io-console and json gems update error
- There is a known issue of Ruby 1.9.3 and RubyGems 2.4.x:
- https://groups.google.com/forum/#!topic/rubyinstaller/2... -
04:59 PM Bug #10573 (Rejected): bigdecimal, io-console and json gems update error
- Environment:
OS: Win7 Ent SP1 x64
Ruby: 1.9.3p545
gem: 2.4.5
DevKit: DevKit-tdm-32-4.5.2-20111229-1559-sfx.exe
... -
- * ext/psych/psych.gemspec: ditto.
* ext/psych/psych_emitter.c: ditto.
* ext/psych/psych_parser.c: ditto.
git-svn-id:... -
- * test/ruby/test_rubyoptions.rb (test_command_line_glob_noncodepage):
more assertion for fallback replacement on no... -
- * test/ruby/test_rubyoptions.rb (test_command_line_glob_noncodepage):
test for command line globbing on non-codepag... -
09:57 AM Feature #10571: Keyword arguments with local variable
- @Yukihiro Matsumoto san
Thank you so much. It does make sense, I need to more deepen understanding your language.
... -
- You can define a method shares the scope with the outside by `define_method`.
-
-
- `var` is a local variable in the class scope, which cannot be accessed from the method. a method has independent scop...
-
- Ops.. not only Keyword arguments also other arguments with default value has same behavior.
I suppose there must b... -
- It's kind of curiosity. As far as I searched, I couldn't get sensible reason.
Please give me information if you have... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48715 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
09:08 AM Bug #10568: segmentation fault after pack & ioctl & unpack
- Sorry, forgot to explicitly mention that this is running on linux and that sudo is required to run the snippet (becau...
-
- * ext/socket/option.c (inet_ntop): link aliased inet_ntop in
libruby on mswin not rb_w32_inet_ntop which fails to l... -
07:28 AM Feature #10570: Allow `+=` etc. to be methods
- Yukihiro Matsumoto wrote:
> `Array#[]=` can be defined by alternation of elements of an array, which does not contai... -
- `Array#[]=` can be defined by alternation of elements of an array, which does not contain any assignment to variables...
-
- David MacMahon wrote:
> Yukihiro Matsumoto wrote:
> > For example, how can we define `Integer#+=`?
>
> Immutable... -
- Thanks for your thoughtful reply!
Yukihiro Matsumoto wrote:
> `a+=b` is a short hand form of `a = a + b` by defin... -
- `a+=b` is a short hand form of `a = a + b` by definition. It's fundamentally an assignment. Target of assignment (inc...
-
- nobu: any comment? Should I commit my original patch? Thanks.
We'll also need to investigate why Alex's test co... -
03:21 AM Bug #9580: Refinements regression in IRB
- I'm embarrassed that I missed these responses to my bug report. I had forgotten that I had created this issue. Sorry!...
-
02:58 AM Misc #10541: Remove shorthand string interpolation syntax
- Recursive Madman wrote:
> Did you also check what character followed those interpolations? e.g. "#{@foo}bar" wouldn'... -
02:10 AM
Feature #10565 (Closed): [cleanup] remove unused rb_iseq_new_with_bopt
- Applied in changeset r48713.
----------
iseq.c: remove unused rb_iseq_new_with_bopt
* iseq.c (prepare_iseq_build): ... - 02:10 AM Revision d71c745b (git): iseq.c: remove unused rb_iseq_new_with_bopt
- * iseq.c (prepare_iseq_build): remove unused block_opt param
(rb_iseq_new_with_bopt_and_opt): remove
(rb_iseq_new... -
- * parse.y (parser_yylex): fix lex_state after tLABEL_END, should
be EXPR_LABELARG to be followed by "paren with arg...
12/04/2014
-
10:49 PM Feature #10561: Improve function of Thread::Backtrace::Location #path and #absolute_path
- I think the name #path should always refer to a #path of sorts using it as #filename is kind of odd. Which is a big r...
-
07:22 PM Feature #10561: Improve function of Thread::Backtrace::Location #path and #absolute_path
- I'd like to have some extra clarification on this matter as well. Right now in Rubinius we return _just_ the filename...
-
- In MRI, it is currently possible to create `#+=` methods in C extensions and even `+=` aliases for existing methods i...
-
07:28 PM Feature #4276: Allow use of quotes in symbol syntactic sugar for hashes
- Hi, any reason it doesn't work with inner hashes or arrays?
```
{ foo: {} } # => {:foo=>{}}
{ 'foo': {} }
# Syn... -
- Brian Hempel wrote:
> Sometimes code libraries forget to make a separate object for every error
This was, in fact... -
- You’re right there may be some risk for anti-patterns, but from my experience this example in the original proposal i...
-
- Any update on this? Is this just waiting for 2.2 to make it out the door?
-
- David MacMahon wrote:
> Maybe this could become a feature request to allow `<OP>=` to be methods instead of just syn... -
- I imagine that Python also lets one override the `+=` method. I think this is impossible on Ruby because I think it ...
-
- Of note, Python does implement `+=` this way. `+=` is (almost) an alias for `list.extend`:
~~~
>>> x = []
>>> y ... -
- It returns the integer part as a `BigDecimal`.
-
04:14 PM Bug #10569: BigDecimal#fix returns a BigDecimal object instead of a Integer
- Seems that Python works the same way. Probably is not a bug:
~~~
>>> decimal.Decimal(7).imag
Decimal('0')
>>> d... -
- BigDecimal, sorry for the typo :)
I'm pasting again the example because some chars were deleted in the original re... -
- According docs (http://www.ruby-doc.org/stdlib-2.1.5/libdoc/bigdecimal/rdoc/BigDecimal.html#method-i-fix): fix metho...
-
06:25 PM Bug #10566: Failure to compile ruby-trunk via mingw-w64 on linux
- Nobuyoshi Nakada wrote:
> Applied in changeset r48711.
>
> ----------
> extmk.rb: no EXTOBJS to main if enable-s... -
- Ah, one second. Some of the redundancy is due to archlinux shipping a pair of configure wrapper scripts to make loopi...
-
- Applied in changeset r48711.
----------
extmk.rb: no EXTOBJS to main if enable-shared
* ext/extmk.rb: as all extens... -
- Could you show your exts.mk file?
And...
* --enable-static is not a valid option
* --disable-install-doc impli... -
- my configure command:
configure_args=' '\''--host=i686-w64-mingw32'\'' '\''--target=i686-w64-mingw32'\'' '\''--build... -
- =begin
Greetings;
In my quest for optimizing my network usage (very slow connection), I've been migrating my
arc... -
05:38 PM Bug #10558: failing require digest.so on OSX statically compiled ruby
- it's few libraries, most important here `openssl 1.0.1j`:
~~~
CFLAGS="-fPIC" LDFLAGS="-fPIC" ./Configure darwin6... -
- * ext/extmk.rb: as all extension objects including initializations
of ext and enc shouild be linked to libruby if e... - 05:10 PM Revision 34c3ac0e (git): * 2014-12-05
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48710 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * ext/socket/option.c (inet_ntop): the fallback implementaion is
always available on Windows.
git-svn-id: svn+ssh:... -
- This snippet produces a segmentation fault on rubies on ruby_2_1 branch from git commit [1]
onwards and ruby_2_0_0 b... -
11:30 AM Bug #10511: fix for #9940 causes dramatic performance regression
- Koichi Sasada wrote:
> Ok, I understand the reason.
>
> For each hash access, new Hash object are created each ti... -
- Ok, I understand the reason.
For each hash access, new Hash object are created each time.
You can check this beha... -
- Perhaps there is some VM optimization? As for the test: I've just mimic the code we have in debase native code, I do...
-
- I understood.
First hash (Hash.new) contains empty st table.
Second hash ({}) does not have st table (NULL).
S... -
- I found strange behaviour.
Using Hash.new, it is slow as you say:
```
Hash.new
ruby 2.2.0dev (2014-12-02 trun... -
- Eric Wong wrote:
> Oops, [Feature #9425] (power-of-two hash sizes :)
If the problem is in implementation of hash ... -
- Oops, [Feature #9425] (power-of-two hash sizes :)
-
- Eric Wong wrote:
> Oops, OK. I was suspicious of [Feature #9325] hurting performance
> (for cases where a hash fu... -
- Oops, OK. I was suspicious of [Feature #9325] hurting performance
(for cases where a hash function needs tuning). -
- Eric Wong wrote:
> [email protected] wrote:
> > I think I've found one more source of performance degradation:
> ... -
10:28 AM Bug #10567 (Closed): A fiber issue with `autoload`: `fiber called across stack rewinding barrier (FiberError)`
- We have two ruby files as following:
test.rb
autoload :User, './user.rb'
f = Fiber.new do
p U... -
- No problem.
Thanks. -
- OK to remove? Or is block-inlining to be resurrected in the future?
rb_iseq_new_with_bopt seems unused and block_... -
- I'm no longer able to reproduce the issue on trunk, so this looks fine. I tested both patches and Eric's does indeed ...
-
01:13 AM Feature #4147: Array#sample で重みを指定したい
- Yui NARUSE wrote:
> わたしの知る限りでは進捗ないと思います。
> うらべさん、あるいは他のどなたか何かありますか？
進捗ダメです。
このスレッドのコメント#1でも書きましたが実装があると話が大幅に進...
Also available in: Atom