Activity
From 10/30/2016 to 11/05/2016
11/05/2016
-
11:58 PM Revision a2acbf4b (git): shell.rb: suppress warnings
- * lib/shell.rb: initialize class instance variables and remove
writer methods to be overwritten soon.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56616 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * lib/e2mmap.rb (bind): eval with the location. undefine
predefined method before overwriting to suppress a warning.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56615 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
09:16 PM Feature #12142: Hash tables with open addressing
- Yura Sokolov wrote:
> my st_table code is much shorter and simpler.
> ...
I think measuring lines after removing comments and single braces to
define what is simpler is not an accurate approach:
- I suspect you did not count tab... -
05:03 PM Feature #12142: Hash tables with open addressing
- Sorry again for sending a version after accepted decision.
It is my last attempt to jump in a train:
https://bugs.ruby-lang.org/attachments/download/6228/st_table_array4.mbox
I removed all siphash related code. (but I kept murmur fi... -
- Koichi Sasada wrote:
> > in my repository https://github.com/vnmakarov/ruby.git
> ...
Yes. The head of the branch has smaller memory footprint for tables than my April version. And it is a bit slower than April version. -
- Matz,
I apologize for this message,
I'm just asking for a bit more comments.
> So I chose Vlad's version as a baseline since it is slightly simpler.
> ...
Although my version contains a lot of non st_table related code (improvin... -
04:27 PM Feature #12142: Hash tables with open addressing
- > in my repository https://github.com/vnmakarov/ruby.git
Is it a small (but slightly slow) version?
I prefer it. -
- Yukihiro Matsumoto wrote:
> Performance-wise, the difference between two candidates is so small. I'm having a hard time making a decision. But to merge it before 2.4, we need to make a decision now.
> ...
Thank you.
I learned a lot ... -
03:22 PM Feature #12142: Hash tables with open addressing
- Performance-wise, the difference between two candidates is so small. I'm having a hard time making a decision. But to merge it before 2.4, we need to make a decision now.
So I chose Vlad's version as a baseline since it is slightly si... -
- I've uploaded new version.
It has one-line "security" fix in `murmur_finish` that doesn't change performance or memory footprint.
https://bugs.ruby-lang.org/attachments/download/6226/st_table_array2.mbox
- 06:09 PM Revision f0e2194b (git): * properties.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56614 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 06:09 PM Revision 84323167 (git): * remove trailing spaces.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56613 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
06:09 PM Revision 8c5ae893 (git): Add ChangeLog-2015 as an anchor of new ChangeLog see #12283
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56612 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
05:06 PM Feature #5446: at_fork callback API
- We had to get around missing at_fork in concurrent-ruby as well. Since we are avoiding monkey_patching we had to result in lazy-checking pid, when it changes we re-create a resource. The resource has to check on every usage though which ...
-
04:56 PM Revision 051ba7b0 (git): Alias regist_scanner to register_scanner
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56611 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
04:43 PM Bug #12670: Segmentation fault on `Magick::Image#get_pixels` on ruby since ruby-2.2
- The attached Dockerfile may help to reproduce.
-
- I'm also using "Ubuntu 14.04.5 LTS" with gcc version 4.8.4 (Ubuntu 4.8.4-2ubuntu1~14.04.3).
-
03:36 PM Bug #12670: Segmentation fault on `Magick::Image#get_pixels` on ruby since ruby-2.2
- Koichi Sasada wrote:
> On my linux VM, I can't reproduce with and without previous patch.
> ...
I have several PC and this problem reproduces even in TravisCi. But all my PC's and travis images are Ubuntu-based. It may be important. Wh... -
- On my linux VM, I can't reproduce with and without previous patch.
* ruby 2.4.0dev (2016-11-05 trunk 56574) [x86_64-linux]
* ruby 2.4.0dev (2016-11-01 trunk 56534) [x86_64-linux]
Sorry... -
- I think the status is changed, but there is another issue.
We can see a assertion fail when I set RGENGC_CHECK_MODE as 1 in gc.c.
```
$ bundle exec ruby test.rb
ruby: ../gc.c:1532: heap_page_allocate: Assertion `heap_allocated_page... -
02:34 AM Bug #12670 (Assigned): Segmentation fault on `Magick::Image#get_pixels` on ruby since ruby-2.2
-
04:42 PM Bug #12576 (Closed): SEGV when interrupting tail recursion
- Applied in changeset r56610.
----------
* compile.c (tailcallable_p): disable tail call optimization for
toplevel, eval, and load to avoid SEGV when interrupted by SIGINT.
[ruby-core:76327] [Bug #12576] -
- sorry for long absence. I agree with your suggestion. Could you commit it?
-
- toplevel, eval, and load to avoid SEGV when interrupted by SIGINT.
[ruby-core:76327] [Bug #12576]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56610 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
04:32 PM Feature #10298 (Rejected): Array#float_sum (like math.fsum of Python)
- Ruby 2.4 implements Array#sum and Enumerable#sum.
-
- Applied in changeset r56609.
----------
* vm_insnhelper.h: rename REG_(CFP|PC|SP|EP) to VM_REG_....
[Bug #12527]
* vm_exec.c: ditto. -
- [Bug #12527]
* vm_exec.c: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56609 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- We have Process.clock_gettime since Ruby 2.1.
```
% ruby -e 'p Process.clock_gettime(Process::CLOCK_REALTIME, :millisecond)'
1478362786099
```
This can be used as Time.now.to_i(:millisecond) which is shown in this issue.
... -
- Thank you for your investigation. Now, `RUBY_EVENT_SPECIFIED_LINE` should not be used and I'll remove this feature soon.
Thanks,
Koichi -
- Now I need to reconsider about this feature request. So I close it.
Thank you for your joining about this discussion. -
- Agreed.
Matz.
-
- Sorry for late response.
I'll ask matz again.
I and Nobu talked about this topic and we agree with:
(1) to be obsolete `absolute_path` method and alias with `path` method.
(2) `path` method returns absolute path, even if main scr... -
- Current behavior is intentional.
Time#utc? returns the mode of Time object, not the time zone offset is zero.
The mode affects several methods.
For example, Time#to_s generates "UTC" for UTC Time objects and
numeric timezone of... -
-
- * dir.c (rb_glob): merge rb_glob2 which is used only here.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56608 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * enumerator.c (InitVM_Enumerator): Enumerator::Lazy#chunk_while
defined. [Fix GH-1186]
https://github.com/ruby/ruby/pull/1186
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56607 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * numeric.c (int_pos_p, int_neg_p): raise TypeError if not an
Integer instead of returning nil.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56606 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
03:30 PM Revision 3f2f64db (git): * vm_eval.c (vm_call0_cfunc_with_frame): pass method id to
- dtrace hook. follow up r56592.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56605 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r56604.
----------
fix Resolv::LOC::Coord.create.
* lib/resolv.rb (Resolv::LOC::Coord.create): fixed.
[ruby-core:72567] [Bug #11912] fixed by Eric Wong and Kazuhiro
NISHIYAMA. -
- * lib/resolv.rb (Resolv::LOC::Coord.create): fixed.
[ruby-core:72567] [Bug #11912] fixed by Eric Wong and Kazuhiro
NISHIYAMA.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56604 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- fall through warnings but it is too noisy.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=7652
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56603 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 03:02 PM Revision e3142f88 (git): * 2016-11-06
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56602 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- warnings by -Wsign-compare.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56601 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r56600.
----------
* file.c (rb_home_dir_of): convert given username into filesystem
encoding. [ruby-core:76682] [Bug #12652]
patched by Dāvis Mosāns -
- encoding. [ruby-core:76682] [Bug #12652]
patched by Dāvis Mosāns
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56600 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r56599.
----------
fix vtm_add_offset yday on last day of year.
* time.c (vtm_add_offset): Fix yday on last day of year.
[ruby-core:72878] [Bug #11994] Fixed by Andrew White. -
- * time.c (vtm_add_offset): Fix yday on last day of year.
[ruby-core:72878] [Bug #11994] Fixed by Andrew White.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56599 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Yui NARUSE wrote:
> Agreed about Net::HTTP.post shouldn't support post(hostname, path, port) form because there's no reason to support to use HTTP.
> ...
Merged in r56597.
-
- Agreed about Net::HTTP.post shouldn't support post(hostname, path, port) form because there's no reason to support to use HTTP.
Could you commit it? -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56598 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56597 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
02:35 PM Feature #12902: How about Enumerable#sum uses initial value rather than 0 as default?
- Akira Tanaka wrote:
> I think [].sum should return 0.
> ...
I guess you mean Array#join
And the fact that Enumerable#sum could be used for string concatenation makes the default value of 0 pretty weird. Just imagine the following pi... -
04:01 AM Feature #12902: How about Enumerable#sum uses initial value rather than 0 as default?
- On 2016/11/05 04:28, aaron.lasseigne@gmail.com wrote:
> Issue #12902 has been updated by Aaron Lasseigne.
> In https://bugs.ruby-lang.org/issues/12217#note-3, Akira Tanaka, mentions that the default argument to sum is 0. This create... -
- I think [].sum should return 0.
Returning 0 is better than nil because programmers don't need to treat empty array specially.
This behavior is different with Enumerable#reduce which returns nil for empty array.
Also, string summatio... -
- module_function is the standard way to define module methods.
I think it is because "include mod" makes us possible to call module methods without module name,
as "popen3(...)" instead of "Open3.popen3(...)",
"sin(...)" instead of "... -
02:19 PM Bug #12904 (Feedback): Seg fault in openssl
- Do you reproduce with ruby_2_3 branch head?
I can't tell anything without a reproducer but I suspect this is related to [Bug #12292]. -
02:04 PM Bug #12904 (Rejected): Seg fault in openssl
- I'm using Threads and EventMachine in a weird way because I don't really know what I'm doing. The same code but single threaded does not produce an error. There were no ruby files in the crash report directories.
/usr/local/Cellar/ru... -
- I agree with performance advantages.
I'm negative to introduce this feature because it is too hard.
People who know well about thread programming can use this feature using concurrent-ruby and I don't want to recommend it for peopl... -
01:55 PM Feature #12607: Ruby needs an atomic integer
- OK, then let's discuss the needs of this class.
An obvious benefit of this class to be in core is that we don't even need Mutex in theory. Because we have GVL, if we do this in core we can just increment the backend integer and that ... -
- > So you mean we don't need atomic integer because concurrent-ruby sucks? I don't think that's a valid reason to reject this.
I didn't mean that. After "BTW" is only my impression about threading. -
- So you mean we don't need atomic integer because concurrent-ruby sucks? I don't think that's a valid reason to reject this.
-
-
- At first, we need a box of integer, not an atomic integer.
Like that:
```ruby
box = IntegerBox.new(0)
box.increment
box.increment
box.to_i #=> 2
```
This IntegerBox can be implemented by the following code:
```ruby
clas... -
- Applied in changeset r56596.
----------
* lib/net/http.rb (transport_request): other than HTTPContinue
in 1xx (HTTPInformation) also needs to continue. [Bug #12890] -
- in 1xx (HTTPInformation) also needs to continue. [Bug #12890]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56596 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * lib/time.rb (make_time): "now" argument as nil works again.
This is broken since Ruby 2.2.
Mathieu Jobin pointed a problem.
https://github.com/ruby/ruby/commit/e4b05d91eb0d48fd172abf015c493bb42d755d07#commitcomment-17421387
gi... -
02:02 PM Bug #12198: Hash#== sometimes returns false incorrectly
- The documentation of Set clearly states the following:
> * Set assumes that the identity of each element does not change
> ...
So it is not a bug, but that you are not supposed to modify an object once stored in a set. -
- It seems the proposed implementation treats the receiver as a glob pattern.
```
% ruby -rpathname -e '
class Pathname
def glob(pattern, flags = 0, &b)
Array(pattern).flat_map do |pat|
self.class.glob(join(pat), flags,... -
- Committed, thanks!
-
- Applied in changeset r56594.
----------
IPAddr#== and IPAddr#<=> no longer raise an exception if coercion fails
* lib/ipaddr.rb (IPAddr#==): If coercion fails, return false
instead of passing through the exception. [ruby-core:77451]
... - 01:53 PM Revision 2d884179 (git): IPAddr#== and IPAddr#<=> no longer raise an exception if coercion fails
- * lib/ipaddr.rb (IPAddr#==): If coercion fails, return false
instead of passing through the exception. [ruby-core:77451]
[Bug #12799]
* lib/ipaddr.rb (IPAddr#<=>): If coercion fails, return nil
instead of passing through the excep... -
- I missed `0.0` and `0`.
-
- Thank you, I committed.
-
- Applied in changeset r56592.
----------
* eval.c, method.h, proc.c, vm.c, vm_eval.c, vm_insnhelper.c, vm_method.c:
TracePoint#method_id should return method_id, not callee_id.
[ruby-core:77241] [Feature #12747]
* test/ruby/test_set... -
- add TracePoint#callee_id. [ruby-core:77241] [Feature #12747]
* cont.c, eval.c, gc.c, include/ruby/intern.h, insns.def, thread.c,
vm.c, vm_backtrace.c, vm_core.h, vm_eval.c, vm_insnhelper.c, vm_trace.c: ditto.
* test/ruby/test_settrac... -
- TracePoint#method_id should return method_id, not callee_id.
[ruby-core:77241] [Feature #12747]
* test/ruby/test_settracefunc.rb: change accordingly.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56592 b2dd03c8-39d4-4d8f-98ff-823... -
12:59 PM Revision bbeeb253 (git): fix a typo [ci skip]
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56591 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Applied in changeset r56590.
----------
numeric.c: round to nearest even
* numeric.c (flo_round, int_round): support round-to-nearest-even
semantics of IEEE 754 to match sprintf behavior, and add `half:`
optional keyword argument f... -
- We had a discussion today, and the conclusions:
* add `half:` optional keyword argument to `round` methods,
* its value must be one of `:up`, `:even`, and `nil`,
* `:up` is compatible with the existing behavior,
* `:even` is the ne... -
- * numeric.c (flo_round, int_round): support round-to-nearest-even
semantics of IEEE 754 to match sprintf behavior, and add `half:`
optional keyword argument for the old behavior.
[ruby-core:76273] [Bug #12548]
git-svn-id: svn+ssh:... -
- Note that SortedSet#compare_by_identity is unimplemented when SortedSet uses rbtree.
`rbtree.readjust { |a, b| a.object_id <=> b.object_id }` comes pretty close, but rbtree treats string keys specially by storing a copy of each string... -
- Applied in changeset r56589.
----------
Add Set#compare_by_identity and Set#compare_by_identity?
* lib/set.rb (Set#compare_by_identity, Set#compare_by_identity?):
New methods. [Feature #12210] -
- I'd rather add Set#compare_by_identity.
- 09:23 AM Revision 76977611 (git): Add Set#compare_by_identity and Set#compare_by_identity?
- * lib/set.rb (Set#compare_by_identity, Set#compare_by_identity?):
New methods. [Feature #12210]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56589 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 09:18 AM Revision cb3b7fc0 (git): * properties.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56588 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 09:18 AM Revision 942bfc7b (git): * remove trailing spaces.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56587 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
09:18 AM Revision cdc527db (git): * lib/rdoc/*, test/rdoc/*: Update rdoc-5.0.0
- Release note: https://github.com/rdoc/rdoc/blob/b825775647f62c5b525e9780a28ff2fbb1d5bf6f/History.rdoc#500--2016-11-05
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56586 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- implemented at r56571
-
- fixed at r56583
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56585 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56584 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * lib/resolv.rb (Resolv::DNS#extract_resources): Use each_resource
instead of each_answer.
[ruby-core:75461] [Bug#12372] reported by Rafael Fernandez Lopez.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56583 b2dd03c8-39d4-4d8... -
- is enabled or not.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56582 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r56581.
----------
chomp option
* io.c (extract_getline_opts): extract chomp option.
[Feature #12553] -
- * io.c (extract_getline_opts): extract chomp option.
[Feature #12553]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56581 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Pathname#each_child is designed that easier access for child files by default (with_directory=true).
But it is also possible to obtain filenames only (with_directory=false) for applications which
remember the directory in the applicati... - 07:18 AM Revision cc7b3feb (git): * doc/standard_library.rdoc: remove "thread.rb" phrase because
- Queue and ConditionVariable were moved to built-in class.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56580 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * tool/mkconfig.rb: [DOC] remove trailing space from a generated
line.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56579 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * test/openssl/ut_eof.rb (OpenSSL::TestEOF): move TestEOF module
under OpenSSL to get rid of conflict with test/ruby/ut_eof.rb.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56578 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 06:59 AM Revision 639ccc40 (git): * tool/mkconfig.rb: [DOC] add rbconfig documentation.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56577 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Applied in changeset r56576.
----------
* lib/net/smtp.rb (tlsconnect): support timeout for TLS handshake.
[ruby-core:76893] [Bug #12678]
* lib/net/protocol.rb (ssl_socket_connect): new method to implement
timeout for TLS handshake... -
- [ruby-core:76893] [Bug #12678]
* lib/net/protocol.rb (ssl_socket_connect): new method to implement
timeout for TLS handshake.
* lib/net/http.rb (connect): use Net::Protocol#ssl_socket_connect.
git-svn-id: svn+ssh://ci.ruby-lang.org/... - 05:40 AM Revision 19c749f6 (git): Document the change in Shellwords.shellwords
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56575 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Applied in changeset r56574.
----------
parse.y: fix restored cmdarg_stack
* parse.y (brace_body, do_body): since cmdarg_stack is saved in
VALUE val, should restore from the same member. on big-endian
platforms where VALUE is larg... -
- * parse.y (brace_body, do_body): since cmdarg_stack is saved in
VALUE val, should restore from the same member. on big-endian
platforms where VALUE is larger than int, it restored 0 in the
upper word. [ruby-core:77920] [Bug #1290... -
- That is correct! I've fixed the implementation to conform to the standard.
-
- Applied in changeset r56573.
----------
Fix the handling of the backslash in double quotes
* lib/shellwords.rb (Shellwords#shellsplit): Fix the handling of
the backslash in double quotes to conform to the standard.
[ruby-core:63807... - 04:58 AM Revision 2da5ae42 (git): Fix the handling of the backslash in double quotes
- * lib/shellwords.rb (Shellwords#shellsplit): Fix the handling of
the backslash in double quotes to conform to the standard.
[ruby-core:63807] [Bug #10055]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56573 b2dd03c8-39d4-4d8f-98... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56572 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * ext/pathname/pathname.c (Pathname#empty?): New method.
[ruby-core:76404] [Feature#12596] Proposed by John Backus.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56571 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Yui NARUSE wrote:
> NetBSD pkgsrc and FreeBSD ports were fixed by adding volatile.
> ...
It seems to be a different issue.
I tried adding volatile to `cont` in `cont_new()`, but it didn't work.
-
- NetBSD pkgsrc and FreeBSD ports were fixed by adding volatile.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=206111 -
- Applied in changeset r56561.
----------
* cont.c (cont_new): disable optimization if clang's version is
3.8.0. [ruby-core:77894] [Bug #12893] -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56570 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- ruby_2_3 r56569 merged revision(s) 56416.
-
- * io.c (copy_stream_body): use IO to write to copy to duplex IO.
http://twitter.com/knu/status/786505317974585344
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_3@56569 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- The bug was fixed in Ruby 2.3, and need not to be backported to prior versions, I think.
-
- Applied in changeset r56568.
----------
* test/ruby/test_refinement.rb (test_refine_alias_in_subclass):
add a test to check that alias in subclasses can be refined.
[ruby-core:69374] [Bug #11186] -
- add a test to check that alias in subclasses can be refined.
[ruby-core:69374] [Bug #11186]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56568 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- ruby_2_3 r56567 merged revision(s) 53533.
-
- * sprintf.c (rb_str_format): format exact number more exactly.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_3@56567 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56566 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- ruby_2_3 r56565 merged revision(s) 55074.
-
- * ext/openssl/ossl.c (Init_openssl): register an ex_data index for
X509_STORE and X509_STORE_CTX respectively. Since they don't share
the ex_data index registry, we can't use the same index.
(ossl_verify_cb)... -
- ruby_2_3 r56564 merged revision(s) 53449.
-
- * thread.c (rb_thread_pending_interrupt_p): no pending interrupt
before initialization.
* thread.c (thread_raise_m, rb_thread_kill): uninitialized thread
cannot interrupt. [ruby-core:72732] [Bug #11959]
gi... -
- Thanks for your confirmation! It's just as intended that the URLs being rendered in bold.
-
- * test/logger/test_logger.rb (test_datetime_format): suppress
warning when strftime called with empty format string.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56563 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * ext/bigdecimal/bigdecimal.c: Import changes from ruby/bigdecimal
repository.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_3@56562 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Nobuyoshi Nakada wrote:
> Do you mount with `noatime` option?
I didn't mean to, but yes.
> ...
It might to be better to check the mount option.
-
- Applied in changeset r56559.
----------
test_file.rb: fix noatime
* test/ruby/test_file.rb (TestFile#test_stat): fix noatime case.
[ruby-core:77943] [Bug #12903] -
- As commented by Herwin W and replied to the submitter in a personal mail, Set has no sense of index where elements are theoretically unordered. It's just that a Set happens to respond to `find_index` via the Enumerable module and have a...
-
- 3.8.0. [ruby-core:77894] [Bug #12893]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56561 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 01:07 AM Revision 9d708e05 (git): * 2016-11-05
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56560 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * test/ruby/test_file.rb (TestFile#test_stat): fix noatime case.
[ruby-core:77943] [Bug #12903]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56559 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
11/04/2016
-
- Do you mount with `noatime` option?
-
- TestFile#test_stat fails on FreeBSD 11.0 as follows:
```
CC = cc
LD = ld
LDSHARED = cc -shared
CFLAGS = -O3 -fno-fast-math -ggdb3 -Wall -Wextra -Wno-unused-parameter -Wno-parentheses -Wno-long-long -Wno-missing-field-initializ... -
- Shugo Maeda wrote:
> static VALUE
> ...
I'm not sure whether the version of clang should be checked as follows:
```diff
diff --git a/cont.c b/cont.c
index 8119b5c..85d3766 100644
--- a/cont.c
+++ b/cont.c
@@ -471,6 +471,10 @@ ... -
- Reproduced on my FreeBSD box:
```
./ruby -r continuation -e 'cont = nil; ary = [1,2,3]; ary.combination(2) { callcc {|k| cont = k} unless cont }; cont.call'
ruby 2.4.0dev (2016-11-04 trunk 56558) [x86_64-freebsd11.0]
-- Control f... -
- FreeBSD failure log is here:
http://rubyci.org/logs/rubyci.s3.amazonaws.com/freebsd11zfs/ruby-trunk/log/20161104T063002Z.fail.html.gz -
- Seems related to segfaults on FreeBSD 11.
I tried with MacPorts clang 3.8.1, but it didn't reproduce.
Maybe clang 3.8.0's issue?
-
- It should work.
```
$ ruby2.0 -v -rmatrix -e 'p = { Vector[1.0, 2.0, 0.0] => 1}' -e 'p p.key?(Vector[1.0, 2.0, 0.0])'
ruby 2.0.0p648 (2015-12-16 revision 53162) [universal.x86_64-darwin15.3.0]
true
```
Is your matrix library th... -
05:18 PM Bug #12897 (Rejected): Hash failed to recognize the existed key and send not found when using Vector as key
- This is not a bug. `Hash#[]` uses `eql?` and not `==` to match keys.
0 == 0.0 # => true
0.eql? 0.0 # => false
{0 => 42}[0.0] # => nil
{0.0 => 42}[0.0] # => 42 -
08:52 AM Bug #12897: Hash failed to recognize the existed key and send not found when using Vector as key
- also affected version ruby 2.3.1p112 (2016-04-26 revision 54768) [x86_64-linux].
-
- When I was using Vector as the key to the Hash, Hash sometimes failed to recognize the key and send nil, while the key-value pair actually existed.
Example as followed:
irb(main):002:0> require"matrix"
=> true
irb(main):003:0> Ve... -
10:42 PM Feature #12515: Create "Boolean" superclass of TrueClass / FalseClass
- Samuel Williams wrote:
>
> ...
It's just one of the issues, but it's the one backed by the same numbers you seem to think are the be-all and end-all. Anyone can munge code to not conflict with existing implementations. (`Kernel::Boole... -
- Just a side question:
Samuel Williams wrote:
> class Kernel::Boolean
> ...
Why do you use Kernel::Boolean and not just Boolean?
-
12:37 PM Feature #12515: Create "Boolean" superclass of TrueClass / FalseClass
- Matthew, I ask you to find one of those examples that would break by introducing the following to core Ruby:
~~~
if $VERBOSE and defined? Boolean
warn "Top level Boolean is deprecated, please update your code"
end
class Kern... -
-
09:45 AM Bug #12899 (Closed): OptParse documentation : example is broken
- Hello,
The complete example code in OptParse documentation at http://ruby-doc.org/stdlib-2.3.1/libdoc/optparse/rdoc/OptParse.html#class-OptParse-label-Complete+example does not run at all.
There are several problems in this code. c... -
- I've uploaded refined mbox patch:
https://bugs.ruby-lang.org/attachments/download/6225/st_table_array.mbox
Corresponding github branch:
https://github.com/funny-falcon/ruby/commits/st_table_with_array5
It almost the same version Vl... -
- Vladimir Makarov wrote:
> Koichi's investigation reported that small hash tables are majority during
> ...
-
- Koichi's investigation reported that small hash tables are majority during
work of Ruby on Rails:
https://gist.github.com/ko1/dda8243134bc40f7fc05e293abc5f4b2#file-report-md
Memory footprint of cloud applications like Ruby on Rail... -
- Vladimir is right.
Despite all my believes, I suggest again to concentrate on open-addressing variants.
I'll prepare new mbox with last changes accurately squashed into right commits and further simplified code.
I'll remove my p... -
- To decrease sizes of very small tables I tuned some parameters. Now minimal table size is 4 (instead of 8) and bins starts with 16 (instead of 8):
https://github.com/vnmakarov/ruby/commit/f4d2bf3ab78e59118e5dd89bcc5c3d8c4371ed35
I... -
- A testing endpoint is available at https://nghttp2.org/?103-eh (see https://lists.w3.org/Archives/Public/ietf-http-wg/2016OctDec/0408.html).
-
07:28 PM Feature #12902: How about Enumerable#sum uses initial value rather than 0 as default?
- In https://bugs.ruby-lang.org/issues/12217#note-3, Akira Tanaka, mentions that the default argument to sum is 0. This creates problems with non-numeric summations (e.g. strings). I would like to suggest using the first enumerable value. ...
-
- In https://bugs.ruby-lang.org/issues/12217#note-3, Akira Tanaka, mentions that the default argument to sum is 0. This creates problems with non-numeric summations (e.g. strings). This would make the method more flexible. It also makes it...
-
06:03 PM Bug #12900: TestSyntax#test_block_after_cmdarg_in_paren fails on PPC64
- The same error is observed on sparc64-solaris2.10, running on 64-bit big-endian CPU.
-
05:06 PM Bug #12900: TestSyntax#test_block_after_cmdarg_in_paren fails on PPC64
- It actually passes on PPC64LE. Isn't it something related to different endianness of PPC64?
-
- Trying to build r56558 on Fedora Rawhide, the build fails on PPC64 with following message:
```
1) Failure:
TestSyntax#test_block_after_cmdarg_in_paren [/builddir/build/BUILD/ruby-2.4.0-r56558/test/ruby/test_syntax.rb:897]:
[ruby-... -
06:01 PM Feature #12901: Anonymous functions without scope lookup overhead
- Much +1 for this feature, I used anonymous functions extensively in graphql-ruby until I learned about this overhead. We still have lots of APIs where the user input is a callable, and I would love to be able to advise procs in those cas...
-
05:30 PM Feature #12901: Anonymous functions without scope lookup overhead
- It would probably be better if ruby could do analysis and if there are no references to local variables in the outer scope(s), automatically optimize such procs, so that you don't need to specify `scope: false`. I believe JRuby already ...
-
04:54 PM Feature #12901 (Open): Anonymous functions without scope lookup overhead
- Right now if you are writing performance sensitive code you can get a performance bump by not using anonymous functions (procs, lambdas, blocks) due to the scope lookup overhead.
https://twitter.com/apotonick/status/791167782532493312... -
06:01 PM Feature #12236: Introduce `mmap` managed heap
- Yura Sokolov wrote:
> > I believe we don't have any drawback with 256MB virtual memory allocation. (but it can depend on OSs)
> ...
Well, as long as your Ruby process never actually needs more than 64MB of memory, that's not a problem,... -
05:15 PM Bug #12666: Fatal error: glibc detected an invalid stdio handle
- For both BSD and Linux, I want to suggest
~~~ ruby
when /bsd|dragonfly|linux/
libc_so = "libc.so"
libm_so = "libm.so"
~~~
Works for me on OpenBSD:
~~~
$ make test-all TESTS=fiddle
...
# Running tests:
Finished test... -
- Ping ... any chance to get this fixed? PPC was recently added into primary Fedora builder, so this is troublesome :/ Also, it will make issues to Fedora PPC users as soon as Fedora 25 is release, which should be next Tuesday if I am not ...
-
11:57 AM Bug #12780: BigDecimal#round returns different types depending on argument
- We need matz's approval to change `Float#round`.
-
07:31 AM Bug #12780: BigDecimal#round returns different types depending on argument
- Good point, I didn't catch that. If it was up to me I'd rather have `Float#round` always return `Float` as well for consistency. I could work on a patch for that too if you agree.
-
- `Float#round` returns Fixnum value when the argument is `0` or omitted.
```
irb(main):005:0> 12.34.round.class
=> Fixnum
irb(main):006:0> 12.34.round(0).class
=> Fixnum
```
If we will fix `BigDecimal#round`, we should change t... -
11:20 AM Feature #12898: String#match? method in addition to Regexp#match?
- Shyouhei Urabe wrote:
> You mean a String instance method named "match?" that takes a Regexp instance?
Yes, exactly. The methods to match a string with a regexp (or vice versa) are present in both classes:
`Regexp#=~` and `String#... -
- Stefan Schüßler wrote:
> Ruby 2.4 introduces [`Regexp#match?`](https://bugs.ruby-lang.org/issues/8110). There should be an accompanying `String#match?` method.
Can you be a bit more verbose here? You mean a String instance method na... -
- Ruby 2.4 introduces [`Regexp#match?`](https://bugs.ruby-lang.org/issues/8110). There should be an accompanying `String#match?` method.
-
- Koichi Sasada wrote:
> I commit a fix on trunk. Could you check it?
Seems nothing changed. Errors looks like same
ruby -v
ruby 2.4.0dev (2016-11-04 trunk 56558) [x86_64-linux]
Attached a terminal output file. -
- I commit a fix on trunk. Could you check it?
-
- Applied in changeset r56558.
----------
* gc.c (heap_page_resurrect): do not return tomb_pages when
page->freelist == NULL.
[Bug #12670] -
- page->freelist == NULL.
[Bug #12670]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56558 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- This is working memo. Now I can't find out the solution.
(1) auto splat for lambda with `yield ary` is allowed [Bug #9605]. So #4 is intentional change.
(2) However, lambda generated by Method#to_proc is not allowed (reported by this... -
- Applied in changeset r56557.
----------
util.c: round to even
* util.c (ruby_dtoa): round to even, instead of rounding to
nearest. [ruby-core:77864] [Bug #12889] -
- * util.c (ruby_dtoa): round to even, instead of rounding to
nearest. [ruby-core:77864] [Bug #12889]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56557 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r56556.
----------
* configure.in: Add compiler version message into rbconfig
as RbConfig::CONFIG['CC_VERSION_MESSAGE']. [Feature #12896] -
- When a extension library developer receives bug reports, they sometimes want to know what compiler is used for building the ruby binary.
We can already know its compiler option but cannot know compiler name and version.
```diff
diff... -
- as RbConfig::CONFIG['CC_VERSION_MESSAGE']. [Feature #12896]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56556 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Close this because it duplicates #10286
-
- r53533をバックポートしてください。
-
- I've fixed this on r56554
- 04:29 AM Revision b7c93a3c (git): ChangeLog: cite issue on redmine
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56555 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
-
03:54 AM Bug #12894 (Third Party's Issue): Seems to be some issue with sqlite3
- Trying to run the console, but I get a massive error.
I think I should see development.sqlite3 and text.sqlite3 in my /db folder, but they're not there.
I'm attaching the bug report. - 04:05 AM Revision 95e9ccbb (git): numeric.c (rb_int_digits): Fix exception message
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56554 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * ext/pathname/pathname.c (path_hash): fix unnormalized Fixnum
value bug on mingw/mswin.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56553 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r56551.
----------
fileutils.rb: fix error for non-existent entry
* lib/fileutils.rb (FileUtils::Entry_#copy): ensure that the
source entry exists first, to fix error for non-existent entry.
[ruby-core:77885] [... - 01:35 AM Revision 61982b15 (git): * 2016-11-04
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56552 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * lib/fileutils.rb (FileUtils::Entry_#copy): ensure that the
source entry exists first, to fix error for non-existent entry.
[ruby-core:77885] [Bug #12892]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56551 b2dd03c8-39d4-4d8f-9...
11/03/2016
-
- Other segfault on `TestMarshal#test_gc`
~~~
$ make test-all TESTS='-n !/callcc/' # and I edited test/ruby/test_continuation.rb to no-op
CC = clang
LD = ld
LDSHARED = clang -shared
CFLAGS = -O3 -fno... -
- Test fails on linux with clang.
It doesn't fails with gcc.
OS: Ubuntu 16.04 x64_64
clang version 3.8.0-2ubuntu4 (tags/RELEASE_380/final)
configure: `../configure CC=clang CXX=clang++`
test:
$ make test-all TESTS='-n /callcc/'
... -
- Koichi,
HUGE makes difference for `st_table_entry` only for version with chaining.
Version with open addressing has same `st_table_entry` regardless of HUGE.
It differs only in sizeof `st_table`.
You had tested this version wit... -
- My assumption is HUGE enables to extend 64bits entries and HUGE entries
are increase constantly. And table size should be proportion of entry
count. However st_memsize() returns only constant difference compare
with HUGE=0.
So I a... -
- Well, chaining is not faster with HUGE=1. It is only faster with HUGE=0.
-
- Koichi,
Here is patch that returns chaining in my implementation.
It is smaller than I expected :-)
https://bugs.ruby-lang.org/attachments/download/6212/return-chaining-into-implementation.patch
(it is applied after last additi... -
- Yura Sokolov wrote:
> Vladimir:
> ...
To be honest I am not glad. I am working on RTL VM insns and JIT and it is a big distraction for me.
Instead of accepting one (original) variant and then improving it (as I proposed yesterday),... -
- Koichi,
Difference between HUGE=0 and HUGE=1 is small and constant cause they
differs only in header ( `struct st_table` ).
HUGE=1 has 48 byte header and HUGE=0 has 32 byte header.
Difference were larger when "chaining" were used... -
- As small table size seems so important now, I did some small changes to my branch to decrease it.
https://github.com/vnmakarov/ruby/commit/a99e4e0e929af225c2b7475fa6b492e989bc3b07
Before the change, my code maintained 4:1 ratio of ... -
- Yura:
Could you check that there are no memory usage (`st_memsize()`) between HUGE=0 and HUGE=1.
Is it reasonable?
Here is a raw data <https://gist.github.com/ko1/dda8243134bc40f7fc05e293abc5f4b2#file-size-csv>. -
01:21 AM Feature #12142: Hash tables with open addressing
- Thank you. This is very interesting, informative and educational. Good
look with getting your patch accepted.
On 11/02/2016 06:41 PM, funny.falcon@gmail.com wrote:
> Issue #12142 has been updated by Yura Sokolov.
>
> File hash... -
- Koichi, thank you for investigation.
Your graphs give a lot of useful information.
Expected behaviour:
- memory usage is lesser with my patch
(cause it has more reallocation steps, and smaller header)
- performance of huge table... -
- Samuel Williams wrote:
> I feel like this should work.
Feelings are not enough. As Matthew already said, https://tools.ietf.org/html/rfc3986.html#section-5.2 doesn't define this. I can see at least the following problems:
1) Based... - 12:46 PM Revision 03c9bc2b (git): Import from ruby/bigdecimal repository
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56550 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
10:21 AM Bug #12892 (Closed): FileUtils::cp_r() raises RuntimeError "unknown file type" instead of Errno::ENOENT "No such file or directory" for non-existent source path
- In Ruby 1.9.3 both FileUtils.cp(src, dst) and FileUtils.cp_r(src, dst) raised Errno::ENOENT "No such file or directory" for a non-existent src path (which is correct behaviour imho).
At least since Ruby 2.1.0 FileUtils.cp_r(src, dst) ... -
08:26 AM Bug #12890: Net::HTTP should treat unexpected 1XX responses as non-final.
- Apologies, it was noted that it would be more helpful to provide the example server in Ruby instead:
~~~ ruby
require 'socket'
server = TCPServer.new 8080
loop do
client = server.accept
client.puts "HTTP/1.1 103 Ear... -
08:07 AM Bug #12891: armv7l-linux-eabihf/openssl.so: [BUG] Illegal instruction at 0xb6659cfc
- Nobuyoshi Nakada wrote:
> Can't you reproduce it with `gdb` and get the stack trace?
If you could give me a hint on how to do that I certainly could.
-
- Can't you reproduce it with `gdb` and get the stack trace?
> ```
> ...
Seems failing inside libcrypto.so.
-
- Samuel Williams wrote:
> [...] But what evidence do you have to support NOT adding Boolean? I can't think of any good reason why you shouldn't add it.
Ignoring most of the "bad" reasons in this thread, how about reiterating this one:... -
- Matz has explained the reasons.
-
- > It seems specific to the context, and should be defined there.
Yeah but it's not. 2,080,658 instances of code on GitHub disagree with you. It's not my personal opinion, it's tangible evidence. But what evidence do you have to suppor... -
- * file.c: include sys/sysmacros.h for ArchLinux which deprecated
use of major() and minor() in sys/types.h.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56549 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 05:46 AM Revision f0baa122 (git): * 2016-11-03
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56548 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * file.c (rb_stat_dev_major, rb_stat_dev_minor): use DEVT2NUM as
well as rdev_major and rdev_minor.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56547 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Sorry for late response.
I missed incompatibility.
How about to commit it before next preview and ask feedback from users?
Thansk,
Koichi
11/02/2016
-
- I believe the fix is relatively simple; the main change is to `lib/net/http.rb`:
~~~diff
--- a/lib/net/http.rb
+++ b/lib/net/http.rb
@@ -1439,7 +1439,7 @@ def transport_request(req)
begin
res = HTTPRespons... -
- Long story short
Net::HTTP's client does not tolerate non-100 status codes from the 1XX block in a sensible manner: it treats them as final responses, rather than as provisional ones.
Expected behaviour
When Net::HTTP receives... -
- While installing GitLab on my Raspberry 2, at the end I'm executing the command: `sudo gitlab-ctl reconfigure`
This leads to the reported bug. The complete error message is as follows:
```
/opt/gitlab/embedded/lib/ruby/2.3.0/armv7... -
- Koichi Sasada wrote:
> Sorry for my lazy-ness, I evaluated your two implementations (ffalcon has two versions, HUGEHASH or not).
> ...
Koichi, thank you for your detail investigation of the two tables. Looking
at this report, I got a... -
- Sorry for my lazy-ness, I evaluated your two implementations (ffalcon has two versions, HUGEHASH or not).
Now, I depicted all of my evaluations in charts, so I introduce this report: <https://gist.github.com/ko1/dda8243134bc40f7fc05e2... -
03:38 PM Bug #12889 (Closed): Regression of Ruby 2.4.preview2 string interpolation
- ~~~
$ ./ruby -v
ruby 2.4.0dev (2016-07-10 trunk 55620) [x86_64-linux]
$ ./ruby -e 'puts ("%01.0f" % 1234567892.0)'
1234567892
~~~
https://github.com/ruby/ruby/commit/295f60b94d5ff6551fab7c55e18d1ffa6a4cf7e3
~~~
$ ./ruby -v
r... -
03:21 PM Bug #12888 (Rejected): Require 'thread' crash on recent 2.2.x
- By just calling the Thread class, a ruby script or just IRB crash :
(French install)
LoadError: 127: La procédure spécifiée est introuvable. - Init_Thread
It works on ruby for windows 1.9.3-p327, 2.0.0-p353
It fails on ruby f... -
- What about giving it an obvious name like `yield_self` and making a syntax sugar `object.{|x| ... }` for calling it like we do in Proc#call?
-
02:46 AM Feature #5446: at_fork callback API
- ping, any plans to add these apis, they will help clean up a lot in the Ruby ecosystem, just hit it again today.
I think what is really needed here is a conceptual Yes or No from Koichi or Matz -
- * test/lib/test/unit/assertions.rb (assert_all_assertions): prefix
assert to be filtered out.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56546 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * doc/extension.rdoc: set ANYARGS as arguments for func1 and func2.
[ci skip]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56545 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
11/01/2016
-
- * doc/extension.rdoc: set ANYARGS as arguments for func1 and func2.
* doc/extension.rdoc: mention the exception object which is passed
to func2 as the second argument. [Fix GH-1471]
[ci skip]
git-svn-id: svn+ssh://ci.ruby-lang.org/... -
- Applied in changeset r56543.
----------
numeric.c: bit op with non-integer
* numeric.c (rb_num_coerce_bit): enable bit operations with
coercing by non-integer object. [ruby-core:77783] [Bug #12875] -
- * numeric.c (rb_num_coerce_bit): enable bit operations with
coercing by non-integer object. [ruby-core:77783] [Bug #12875]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56543 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
10:03 PM Bug #8004: spawn does not honor its passed PATH
- On my computer, the reason for this is:
It should have found the path here: https://github.com/ruby/ruby/blob/c7e99cbfc8dea00ffa5d0a286ea484e90e789df4/process.c#L2193
But notice the second argument, 0, which causes `dln_find_exe_r`... -
08:29 PM Bug #12832: Calling Object#method hangs for private method defined on module then made public once it's been used to extend class
- @George: Thanks for explaining. Makes sense.
-
- Ruby defers signals. At safe moments, Ruby checks if a signal arrived, then handles it. If Ruby is stuck in an infinite loop (because of a bug like this one), then Ruby would never check if a signal arrived, so the signal would be ignore...
-
- @nobu: Is it possible there is a separate problem with the code such that interrupt signals are incorrectly being ignored...?
-
- It's not expected but a bug is always what causes unexpected behavior.
-
- @nobu: Thank you for fixing this. Is it expected that a bug like this would cause the interpreter not to accept interrupt signals?
- 03:47 PM Revision c7e99cbf (git): * 2016-11-02
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56542 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56541 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
02:56 PM Feature #12515: Create "Boolean" superclass of TrueClass / FalseClass
- Nobuyoshi Nakada wrote:
> Rodrigo Rosenfeld Rosas wrote:
> ...
This doesn't seem to work as I just tested in irb: [true, false] === true => false -
- Samuel Williams wrote:
> > It's a choice by that library author. They are not classes but methods, there can be `Boolean` method in the library.
> ...
It seems specific to the context, and should be defined there.
-
- Rodrigo Rosenfeld Rosas wrote:
> Maybe something like:
```ruby
Boolean = [true, false]
```
> ...
-
- Joel on Software "Don't Let Architecture Astronauts Scare You" by Joel
Spolsky http://www.joelonsoftware.com/articles/fog0000000018.html
-
- Rodrigo Rosenfeld Rosas wrote:
> I'd just like to add a real situation where I missed a `Boolean` class recently.
> ...
There's a potential religious argument here about validating against the schema before/after deserialising. Pragmat... -
- * test/ruby/test_integer.rb (MimicInteger, CoercionToInt): extract
common parts.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56540 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * numeric.c (int_and, int_or, int_xor): use rb_num_coerce_bit to
reduce repeated code.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56539 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
12:48 PM Bug #12887 (Closed): Object#extendを引数無しで呼び出すとErrorとなるが、Module#includeとModule#prependはエラーとならない
- Object#extend を引数無しで呼び出すと、ArgumentErrorとなります。 https://bugs.ruby-lang.org/issues/6085
~~~
$ ruby -e 'Module.new { extend }'
-e:1:in `extend': wrong number of arguments (given 0, expected 1+) (ArgumentError)
from -e:1:in `block in <... -
- Fixed by r55831 and r55833. Backport requested in #12657.
-
- I checked that the patches fix the warnings in Ruby 2.2. BTW r55833 is required alongside of r55831
-
- * numeric.c (num_zero_p): should return true if zero.
rb_bigzero_p returns 1 or 0.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56538 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * numeric.c (rb_num2ulong_internal): adjust indent.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56537 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * numeric.c (rb_num2ulong_internal, rb_num2ull): fix typos in
comments.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56536 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * bignum.c (rb_big_and, rb_big_or, rb_big_xor): use dedicated
macro RB_INTEGER_TYPE_P instead of combination of FIXNUM_P and
RB_BIGNUM_TYPE_P.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56535 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Today I learned that [PHP caches realpath](http://jpauli.github.io/2014/06/30/realpath-cache.html), and causes troubles when people use symlink to deploy scripts.
The situation is not exactly the identical to ours, but we should avoid...
10/31/2016
-
- Samuel Williams wrote:
>
> ...
You keep saying things like "polymorphic," but you still haven't answered the question which is fundamental to this whole debate: what method would you define on `class Boolean` that applies identically ... -
- > It's a choice by that library author. They are not classes but methods, there can be `Boolean` method in the library.
Nobu, you are not looking deep enough.
https://github.com/jeremyevans/sequel/blob/c39594dd35b3f8a8975b05be156ba... -
- Maybe something like:
~~~ruby
input.zip([String, String, String, Boolean, String, Boolean]).all?{|(i, c)| c === i }
~~~ -
- Actually, this isn't a good example as it would return `TrueClass` or `FalseClass` rather than `Boolean`...
-
- I'd just like to add a real situation where I missed a `Boolean` class recently.
A web request takes an array of values in a JSON encoded string and before processing the data the server-side handler will try to validate the input by ... -
- Samuel Williams wrote:
> Having a class named `Boolean` would make things more readable, for example here: http://sequel.jeremyevans.net/rdoc/files/doc/schema_modification_rdoc.html#label-Column+types - you can see that because there is... -
- Samuel Williams wrote:
> > But that's how things already are, and the market deals with it just fine.
> ...
It's not a metaphor, it's the real world. Ruby gems exist in open competition, and developers have the opportunity to choose b... -
- The terms "truthy" and "falsey" were used by Douglas Crockford in his
series of JavaScript lectures at Yahoo about 5 years ago. A value is
"truthy" if treated as true by a conditional. A value is "falsey" if
treated as false by... -
- It's also interesting to note, so many people are working around this:
Searching on github for:
"is_a? TrueClass FalseClass" gives 90,000 results.
https://github.com/search?p=2&q=is_a%3F+TrueClass&ref=searchresults&type=Code&utf... -
- > But that's how things already are, and the market deals with it just fine.
Well, I know you are being metaphorical ("market deals with it"). But this is not really a case of a market, and I prefer DRY for many reasons. There should... -
- Applied in changeset r56534.
----------
configure.in: no round in x64-mingw
* configure.in (ac_cv_func_round): round(3) in x86_64-w64-mingw32
is not accurate in an edge case. [ruby-core:77794] [Bug #12878] -
- JFYI if we stick to 64bit representation, 0.49999999999999994 is `0011111111011111111111111111111111111111111111111111111111111111`.
-
- * configure.in (ac_cv_func_round): round(3) in x86_64-w64-mingw32
is not accurate in an edge case. [ruby-core:77794] [Bug #12878]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56534 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * dir.c (to_be_ignored): remove duplicate inline modifier, and
suppress a warning by VC.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56533 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 03:17 PM Revision 64116ea9 (git): * 2016-11-01
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56532 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56531 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- I get the same odd behavior with:
ruby --version ruby 2.3.1p112 (2016-04-26 revision 54768) [x86_64-linux]
./ruby --version ruby 2.4.0preview2 (2016-09-09 trunk 56129) [x86_64-linux]
It looks like the combination ( foo = nil, b... -
- Applied in changeset r56530.
----------
* test/test_forwardable.rb (TestForwardable#test_on_private_method): add test for [Bug #12782] [Bug #12840] -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56530 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
08:39 AM Feature #7360: Adding Pathname#glob
- The patch still applies to latest trunk.
-
06:41 AM Feature #7360: Adding Pathname#glob
- ping?
-
- Seems like a garbage issue.
-
- Seems a design issue. Let me assign this to the library's maintainer.
-
- Dan Erikson wrote:
> I believe this has recently been implemented as `Hash#dig`.
Indeed. Closing peacefully. -
- Closing. Follow-up issue is #12142.
-
- (Just to be clear) It has nothing to do with HWIA. I can reproduce the situation without it.
```ruby
class X < Hash; end
def test(params = X.new, disrupter: nil)
return params
end
x = X.new
x['foo'] = 'bar'
p test(x).cla... - 02:08 AM Revision b37af621 (git): * 2016-10-31
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56529 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * lib/uri/generic.rb (URI::Generic#merge): merge merge0.
[GH-1469]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56528 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Lee H wrote:
> -- C level backtrace information -------------------------------------------
> ...
It dies inside of libsybdb. Are you using Sybase? Please ask tiny_tds's authors first. If they say it's ruby-core, reopen this ticket ...
10/30/2016
-
- This ticket should be re-cast as a feature request, to allow merging of two relative references.
Incidentally:
Samuel Williams wrote:
>
> ...
As discussed elsewhere, it's a "relative reference", not a "URI." It parses to a `URI:... -
- I feel like this should work.
~~~
> URI.parse("/base/uri") + URI.parse("relative")
URI::BadURIError: both URI are relative
~~~
The result should be URI with path = "/base/relative".
But it doesn't. It fails with an exception.... -
- Samuel Williams wrote:
> The fact that so many gems are introducing "`class Boolean`" is an argument FOR it not AGAINST it, IMHO. Because when this code is loaded together, it might behave strangely if there is no shared meaning for "`c... -
- The fact that so many gems are introducing "`class Boolean`" is an argument FOR it not AGAINST it, IMHO. Because when this code is loaded together, it might behave strangely if there is no shared meaning for "`class Boolean`".
Having ... -
07:59 PM Bug #12885 (Third Party's Issue): Large Record Set: [BUG] Segmentation fault at 0x00000000000000
- [BUG] Segmentation fault at 0x00000000000000
ruby 2.2.4p230 (2015-12-16 revision 53155) [x86_64-darwin15]
-- Crash Report log information --------------------------------------------
See Crash Report log file under the one of fol... -
- * lib/rubygems/specification.rb: reapply 56225.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56527 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56526 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- rubygems to 2.6.8.
Release note of 2.6.8: https://github.com/rubygems/rubygems/commit/9fb8880976f5ab998912898b091d88aa10eb1d4a
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56525 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- This is a follow-up. I tested for the same issue with Ruby 2.4.0 preview
2. I got slightly different behavior, but essentially the same issue.
Here are the details. I wasn't sure if I could add them to the issue
tracker without ma...