Activity
From 02/16/2014 to 02/22/2014
02/22/2014
-
09:07 PM Feature #9113: Ship Ruby for Linux with jemalloc out-of-the-box
- @nobusan I think that would be a reasonable approach
@eric / @ko1 / everyone
here are the results of runnin... -
06:05 PM Bug #9544 (Feedback): Ruby resolver not using autoport
- If system's resolver, getaddrinfo(), is usable, you can use it.
-
- Applied in changeset r45144.
----------
* lib/resolv.rb (bind_random_port): Rescue EPERM for FreeBSD which
securit... -
02:34 PM Bug #9544: Ruby resolver not using autoport
- So is it ruby's concern, or should the system developers make appropriate patches?
Currently you've reinvented th... -
- I found GNU/Hurd and Haiku also allocates ports sequentially.
GNU/Hurd:
% ./ruby -v -rsocket -e '10.times {... -
- Akira Tanaka wrote:
> I tested several OSs and I found NetBSD 6.1.3 still doesn't randomize port:
>
> % ruby ... -
02:03 PM Bug #9555 (Closed): Ruby 2.0.0 compilation fails to compile on cygwin. (curses.so)
- Using RVM we see that ruby fails to compile for 2.0.0. This seems to be fixed in 2.1.0.
Compiling the 2.1.0 as follo... -
01:27 PM Bug #9554: need read access for conout$ (io/console)
- ruby_2_1 r45139 .
-
01:13 PM Bug #9554: need read access for conout$ (io/console)
- r45130 was backported to ruby_2_0_0 at r45137.
-
11:48 AM Bug #9554: need read access for conout$ (io/console)
- backported into ruby_1_9_3 at r45131.
-
-
- 今まで気づかなかったとは不覚。
-
- ruby_2_1 r45129 merged revision(s) 45084.
-
- r45084 was backported to ruby_2_0_0 at r45103.
-
- backported into ruby_1_9_3 at r45092.
-
- ruby_2_1 r45128 merged revision(s) 45076.
-
- r45076 was backported to ruby_2_0_0 at r45096.
-
- backported into ruby_1_9_3 at r45091.
-
- `ruby_2_1` r44843 merged revision(s) 44527,44552,44553.
-
- ruby_2_1 r45126 merged revision(s) 45057,45059.
-
- ruby_2_1 r45054 merged revision(s) 44327.
-
- ruby_2_1 r44841 merged revision(s) 44517,44518,44519,44523.
-
- ruby_2_1 r45125 merged revision(s) 44637.
-
- `ruby_2_1` r45124 merged revision(s) 44628.
-
07:58 AM
Feature #9553: Make argument validation routine of a method an object
- Even more precisely, I did not fail to notice, but rather forgot about it =). In any case, built in `Arguments` objec...
-
07:39 AM
Feature #9553: Make argument validation routine of a method an object
- It is my fault that I was not aware of `Method#parameters` when writing this post. Too bad I did not participate in i...
-
07:16 AM
Feature #9553: Make argument validation routine of a method an object
- Does Method#parameters not solve your problem?
-
07:02 AM
Feature #9553 (Open): Make argument validation routine of a method an object
- Ruby methods have their acceptable argument configurations specified by special syntax. As argument fields has gotten...
-
- r45123
-
07:52 AM
Feature #9548: Module curry
- I have just learned about the existence of `Method#parameters` method, which would enable to write near-perfect `Modu...
-
05:29 AM
Feature #9548: Module curry
- @mame, I don't quite get these nuances between "currying" and "partial application". I could try and
learn them, but... -
01:46 AM Feature #9548: Module curry
- That is not "currying", but "partial application."
http://lambda-the-ultimate.org/node/2266
--
Yusuke Endoh <... -
12:22 AM Feature #9548: Module curry
- Firstly, Proc has a #curry, though not as flexible as you would like: http://rubydoc.info/stdlib/core/Proc#curry-inst...
-
- r45122
-
07:49 AM
Feature #9552: Module map!
- I failed to notice that `Method#parameters` was already available. This makes it possible for me to implement my desi...
-
06:52 AM
Feature #9552: Module map!
- Core-level syntax might be similar to that of `alias`,
```ruby
class FooBarBazCollection
map! fn foos, barn ba... -
06:44 AM
Feature #9552 (Feedback): Module map!
- I would like to beg for `map!` directive in `Module`. I can imitate it with this code:
```ruby
class Module
de... -
- r45121
-
- r45120
-
- r45066 was backported to ruby_2_0_0 at r45098.
-
- backported into ruby_1_9_3 at r45093.
-
- 個別にbackportチケットを切るのがめんどいので手抜いてこっちに作ります。
-
- r45119
-
06:49 AM Bug #9356: TCPSocket.new does not seem to handle INTR
- [email protected] wrote:
> > http://bogomips.org/ruby.git/patch?id=d8241102f54
> > git://80x24.org/ruby sock... -
05:34 AM Bug #9356: TCPSocket.new does not seem to handle INTR
- Eric Wong wrote:
> [email protected] wrote:
> > Eric Wong wrote:
> > > We should guard against sockerr settin... -
- [email protected] wrote:
> Eric Wong wrote:
> > We should guard against sockerr setting errno to
> > EINTR, E... -
- r45118
-
- r45117
-
- r45116
-
- r45115
-
- r45114
-
- r45113
-
- r45112
-
- r45111
-
- r45110
-
05:20 AM
Feature #9528: mathn.rb library
- Hi Umair,
You should attach a .patch file and wait for feedback. -
- r45109
-
- r44568 was backported to ruby_2_0_0 at r45108.
-
- r45105.
-
- r45107.
-
- r44720 was backported to ruby_2_0_0 at r45106.
-
- r44858.
-
- r45104.
-
03:41 AM Bug #9551 (Closed): [DOC] Fix for documentation of Kernel::catch
- The current documentation for `Kernel::catch` contains complicated examples and confusing language.
*"when arg is gi...
02/21/2014
-
- r45090.
-
- I'd like to mention that you've got a hardcoded range of ports that can be used to _bind()_, however every system ime...
-
- EPERM may happen if you're using (like me) bind port filter policy. For instance, FreeBSD mac_portacl (http://www.fre...
-
- I tested several OSs and I found NetBSD 6.1.3 still doesn't randomize port:
% ruby -rsocket -e '10.times { p A... -
- bind_random_port chooses more random than using the port 0.
The choosen ports by the port 0 is guessable from an a... -
- ### Problem
On one of my production servers I've noticed that customers were failing to install anything using gem... -
09:11 PM Feature #9549 (Rejected): Improvements to Time::strptime
- I opened [a pull request on GitHub](https://github.com/ruby/ruby/pull/540) a few days ago but wanted to open a parall...
-
07:48 PM
Feature #9548 (Feedback): Module curry
- I would like to beg for either `Module#curry` method with syntax
```ruby
module Foo
curry( :sym2, :sym1, 0 => ... -
06:54 PM
Feature #7939: Alternative curry function creation
- Thanks to Yui Naruse for reopening this issue. As a matter of fact, I hate the current behavior of `#curry` method wi...
-
- r45086.
-
- r45086.
-
- Shugo Maeda wrote:
> Shugo Maeda wrote:
> > On FreeBSD 10, Errno::ENOTCONN isn't raised, but Ruby goes in an infini... -
- Shugo Maeda wrote:
> On FreeBSD 10, Errno::ENOTCONN isn't raised, but Ruby goes in an infinite loop because getsocko... -
- Eric Wong wrote:
> > > However, I still prefer my v2 if possible:
> > > http://bogomips.org/ruby.git/patch?id... -
- [email protected] wrote:
> Eric Wong wrote:
> > Anyways, I have an alternative (v3) patch here which retries co... -
- Eric Wong wrote:
> Anyways, I have an alternative (v3) patch here which retries connect()
> on EINTR and ERESTART... -
- Applied in changeset r45084.
----------
* ext/socket/init.c (wait_connectable): break if the socket is
writable to... -
- The script attached in #9356 goes in an infinite loop on FreeBSD (and other platforms conforming to SUSv3).
Naruse-s... -
- r45083.
-
- r45082.
-
- r45080.
-
- すみません、間違えてruby-devにしてしまったので、ruby-coreの方にチケットを作成し直しました。
-
- The script attached in #9356 goes in an infinite loop on FreeBSD (and other platforms conforming to SUSv3).
Naruse-s... -
- r45079.
-
- r45078.
-
11:42 AM Bug #9535 (Closed): Kernel.send(:initialize_copy, Kernel) で segmentation fault
- Applied in changeset r45076.
----------
class.c: do nothing if copying self
* class.c (rb_mod_init_copy): do nothin... -
- r45075.
-
- Samu Voutilainen wrote:
> When I’m trying to floor number with line “"%.2f" % 4.225”, I get 4.22 as result, instead ... -
- r45074
-
- [email protected] wrote:
> Good call Eric. I've carried out your suggestion [1], and attached the cumulative pa... -
- Good call Eric. I've carried out your suggestion [1], and attached the cumulative patch.
[1] this commit: https://... -
- I'm absolutely against including external libraries into ruby repository itself, e.g., libyaml.
It may not be the wo...
02/20/2014
-
- Thanks for taking a look, r45064
-
11:46 PM
Bug #7805 (Closed): ruby 2.0rc2 core on solaris
- Applied in changeset r45064.
----------
gc.c: RB_GC_GUARD should be robust enough for any compiler
* include/ruby/r... -
- Seems better.
Let's try. -
- [email protected] wrote:
> I'd like to hear others' thoughts.
Likewise. We need to hear folks with more exper... -
- Eric Wong wrote:
> I'm not sure about implementations of getsockopt(SO_ERROR),
> there are too many differences f... -
07:35 PM Bug #9543: [Doc] Fix example in GServer documentation (lib/gserver)
- Now the patch...
-
- The code example does not match its description: the server should return time in seconds since 1970.
Patch append... -
04:35 PM Bug #9542 (Rejected): Delegator does not delegate protected methods
- The Delegator of the "delegate" standard library has changed its behaviour since 2.0 because of the change in "respon...
-
- It's a spec.
Aliased methods share the body, so `alias_method_chain` can work.
It's not related to `prepend`. -
- I think it's intended behavior. nobu, how do you think?
-
- Eric Wong wrote:
> * backwards for loop for list is confusing to me,
> any reason for not reversing list decl... -
08:58 AM Bug #9525 (Closed): Stuck with Socket.pack_sockaddr_in
-
- [email protected] wrote:
> I tried to workaround this issue at r45047.
> However I don't have an environment to reprodu... -
- I tried to workaround this issue at r45047.
However I don't have an environment to reproduce the problem.
Would a... -
11:53 AM Bug #9541 (Closed): Markedly increased Hash memory usage if Hash is cleared before reinsertion.
- After a recent upgrade to Ruby 2.1.0 I noticed a significant increase in memory usage when using hashes of any size w...
-
09:34 AM Bug #9540 (Rejected): Number formatting with 4.225 fails
- Hi,
When I’m trying to floor number with line “"%.2f" % 4.225”, I get 4.22 as result, instead of 4.23. Rounding fo... -
09:10 AM Bug #9539 (Closed): REXML XPath UTF8 encoding problem
- I found some problems in REXML when processing XPath expressions with Unicode not ascii characters. I attached a samp...
-
- r45061 on ruby_2_1
-
- COMMITS: 44687,44706,44727
-
- I wrote https://bugs.ruby-lang.org/projects/ruby/wiki/HowToRequestBackport
-
02:27 AM
Bug #9533 (Closed): ruby 2.1.0のHash#flattenが以前のバージョンと異なる挙動をする
- Applied in changeset r45057.
----------
* hash.c (rb_hash_flatten): fix behavior of flatten(-1).
[ruby-dev:47988] ... -
01:59 AM Bug #9533 (Assigned): ruby 2.1.0のHash#flattenが以前のバージョンと異なる挙動をする
-
12:35 AM Bug #9534: bundle exec rake assets:precompile fails
- Hey,
I noticed that you are on ruby 2.1.0p0. I could install ffi gem successfully for ruby-2.1.0 on Maverics. Just...
02/19/2014
-
- [email protected] wrote:
> Eric Wong wrote:
> > > Ah, I see. However, if getsockopt() returns no error, we can... -
- Eric Wong wrote:
> > Ah, I see. However, if getsockopt() returns no error, we can know
> > that at least connect... -
- Eric Wong <[email protected]> wrote:
> Ah, I forget the outer for(;;) loop. Maybe it's better to not loop,
> ... -
- [email protected] wrote:
> Eric Wong wrote:
> > > Could you describe such a race condition in detail?
> >
>... -
- Eric Wong wrote:
> > Could you describe such a race condition in detail?
>
> getsockopt(SO_ERROR) => no error... -
- [email protected] wrote:
> Eric Wong wrote:
> > OK. I wonder if we should even use getsockopt(SO_ERROR) at all... -
- Eric Wong wrote:
> OK. I wonder if we should even use getsockopt(SO_ERROR) at all.
>
> I know there's much li... -
05:59 PM Bug #9535 (Closed): Kernel.send(:initialize_copy, Kernel) で segmentation fault
- 以下のコードを実行すると segmentation fault が起きます。
Kernel.send(:initialize_copy, Kernel)
trunk (45047)、2.1.0p0、2.0.0p35... -
05:56 PM Bug #9534 (Closed): bundle exec rake assets:precompile fails
- Hi all
Sorry for the title, I really don't know what to put.
I don't know anything about ruby but when trying to ru... -
- r44082 was partially (only for ext/syslog/lib) backported to ruby_2_0_0 at r45052.
-
- sorry for my late response. r44175 and r44179 were already backported to ruby_2_0_0. see #7993
-
- r44527 was backported to `ruby_2_0_0` at r45051.
-
- r42844 and r42845 were backported to ruby_2_0_0 at r45050.
-
- r43148, r43149, and r43152 were backported to ruby_2_0_0 at r45049.
-
03:27 PM Bug #9533 (Closed): ruby 2.1.0のHash#flattenが以前のバージョンと異なる挙動をする
- ruby 2.1.0で、Hash#flattenに0以下の引数を与えた場合、ruby 2.0.0以前と
異なる結果を返します。例えば、以下のコード
h = {:a=>1, :b=>[2,3], :c=>4}
... -
11:11 AM Feature #9528: mathn.rb library
- Please guide how can I contribute, I have code written on my local.
-
- Kaneko Yuichiro wrote:
> Sorry, I misunderstood behavior of prepend. Shugo and Boris are right. Thanks for your quic... -
-
04:55 AM Bug #8770 (Assigned): [PATCH] process.c: avoid EINTR from Process.spawn
-
-
01:05 AM Bug #9127 (Assigned): URI::Generic#normalize documentation is vague.
-
- Resolved by r44729
Thanks Eric! -
- @Eric
sure bench needs a bit more love to be totally representative of a rails request. Also this test will do ko... -
- [email protected] wrote:
> An artificial test is:
>
> @retained = []
>
> MAX_STRING_SIZE = 100
>
> de...
02/18/2014
-
- Note, this pattern of
1. Retaining large number of objects
2. Allocating a big chunk of objects (and releasing) ... -
- I can confirm 2 findings.
When heaps are small you barely notice a different.
When heaps grow and general memory... -
- Motohiro KOSAKI wrote:
> I'm not an expert this area. But I guess we don't need to call getaddrinfo() in this case... -
- Akr-san,
I'm not an expert this area. But I guess we don't need to call getaddrinfo() in this case
because the na... -
- I checked a glibc code. I don't think fa3fc0fe5f fixed this issue. It is a mere optimization patch.
I'm not sure why... -
- I think we can accept a workaround if there is a good patch.
-
05:24 AM Bug #9525: Stuck with Socket.pack_sockaddr_in
- Thank you! Let me write a note here.
On my current environment, the glibc version was glibc-2.12-1.47.el6.x86_64.... -
- gdb says __check_pf is guilty.
https://gist.github.com/sonots/9047653#file-gistfile3-txt
(gdb) bt
#0 0x... -
-
- Please note: a number of libraries are using this class: https://github.com/search?l=ruby&q=LogDeviceMutex&ref=cmdfor...
-
10:11 PM Feature #9532 (Closed): Expose logger monitor
- I was wondering if we could do this type of change on the logger class:
https://github.com/arthurnn/ruby/commit/37... -
-
- Thank you for the patch!
-
- Eric Wong <[email protected]> wrote:
> The design is based on st, but uses linked-list of cache-sized arrays
>... -
- We don't need to care upstream kernel specific bugs. Nobody except developer uses it.
I've closed this. -
- I think we can accept a workaround if there is well commented patch.
-
03:33 PM Bug #9529 (Assigned): TarHeader (Gem::Package) doesn't parse size correctly for +8GB entries
-
01:33 PM Bug #9529 (Third Party's Issue): TarHeader (Gem::Package) doesn't parse size correctly for +8GB entries
- * The current TAR header parsing code assumes the size is represented as an octal string
* Because this is a 12-byte... -
01:46 PM Bug #9526: Method#owner returns wrong owner if using prepend.
- Sorry, I misunderstood behavior of prepend. Shugo and Boris are right. Thanks for your quick responce. Please close t...
-
05:16 AM
Bug #9526: Method#owner returns wrong owner if using prepend.
- Also for me, this is the expected behavior. `MyMod` is prepended before `Blah`, and therefore `Blah.new.method( :x )`...
-
-
- Kaneko Yuichiro wrote:
> In this case, a class or module that defines a method is Blah.
> So I think Blah.new.metho... -
- I want to add factorial method mathn.rb file as feature of Math module.
-
10:59 AM Feature #9527 (Open): make Net::HTTP.get_print not only to $stdout but to an IO as a parameter
- e.g.
url = 'http://server/path/large-file.zip'
fh = File.new(File.basename(url), 'wb')
Net::HTTP.get_print(URI(u... -
- "U.Nakamura" <[email protected]> wrote:
> In message "[ruby-core:60819] [ruby-trunk - Bug #9356] TCPSocket.new ... -
- Hi,
In message "[ruby-core:60819] [ruby-trunk - Bug #9356] TCPSocket.new does not seem to handle INTR"
on Fe... -
- Eric Wong wrote:
> Eric Wong <[email protected]> wrote:
> > Thanks for trying. This is probably specific to *... -
- Can you please try the following patch?
This is hopefully robust enough for all future compilers:
http://bogomips... -
- Sorry about the breakage. Instead of reverting completely and using
volatile again, we should try to make RB_GC_GU... -
07:12 AM Bug #7805 (Open): ruby 2.0rc2 core on solaris
- Please completely revert r44994 and r45025.
r45025 partly reverts r44494 and cause regression.
02/17/2014
-
10:46 PM Feature #8850: Convert Rational to decimal string
- I've updated the patch by rebasing and fixing syntax.
https://github.com/mrkn/ruby/compare/64634f92f3474babab93264b3... -
- rdoc indicates that Method#owner will return a class or module that defines a method.
http://www.ruby-doc.org/core-2... -
- I think that PF_NETLINK is not a socket to communicate to another host.
-
- Following command on CentOS release 6.2 x86_64 returned a line
$ strace -e socket /usr/lib64/fluent/ruby/bin/r... -
- I guess getaddrinfo() doesn't block if an IP address is given.
At least, following command on Debian GNU/Linux (je... -
- I think it is very difficult to add timeout to getaddrinfo() function in C.
-
- Are there any ways to timeout Socket.pack_sockaddr_in?
If so, this problem should be resolved.
EDIT: And, I was... -
- Socket.pack_sockaddr_in can block with a name resolution.
If you think the name resolution doesn't block, please exp... -
- We met this trouble with Fluentd https://github.com/fluent/fluentd.
Fluentd is sometimes stuck at `Socket.pack_so... -
- r43148, r43149, and r43152 are backported into ruby_1_9_3 at r45026.
-
- 1.9.3では諦めます。
-
- Applied in changeset r45025.
----------
marshal.c: do not recycle wrapper objects
* marshal.c (marshal_dump, marsha... -
05:15 AM Feature #9502: Remove deprecated definitions
- Koichi Sasada wrote:
> I copied this table to the following page:
> https://bugs.ruby-lang.org/projects/ruby-trunk/... -
- I copied this table to the following page:
https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/CAPI_ObsoleteFunction...
02/16/2014
-
- Ruby dumps core as follows.
% ./ruby -rcontinuation -e '
class C
def marshal_dump
callc... -
- r44637 was backported to ruby_2_0_0 at r45014.
-
- r44549 was backported to ruby_2_0_0 at r45013.
-
- r40830 and r40848 were backported to ruby_2_0_0 at r45012.
-
- r43942, r43957 and r43975 were backported to `ruby_2_0_0` at r45011.
-
- r44628 was backported to `ruby_2_0_0` at r45010.
-
- r44514 was backported to ruby_2_0_0 at r45009.
-
- r44731 and r44751 were backported to ruby_2_0_0 at r45008.
-
- r44772 was backported to ruby_2_0_0 at r45007.
-
- r44670, r44671, r44673 and r44675 were backported to ruby_2_0_0 r45006.
-
03:26 PM Feature #9362 (Rejected): Minimize cache misshit to gain optimal speed
- In the last developer meeting we agreed that sacrificing memory
consumption to gain speed is not a contemporary nice... -
- I updated the patch to use rb_check_arity: https://github.com/mrkn/ruby/commit/a6837fab190fed2707ec7e13c494cd4b642ff630
-
-
05:36 AM
Feature #9522 (Open): Float( "NaN" ), Float( "Infinity" )
- I have noticed that 0.0 / 0.0 returns NaN, but Float( "#{0.0 / 0.0}" ),
or simply Float( "NaN" ) does not return Flo...
Also available in: Atom