19385-19707

19142-21936 subjects 19582-21708

[Bug #657] Thread.new { fork }
19385 [redmine@ru y] Bug #657: Thread.new { fork }

[Bug #663] Benchmark.measure outputs different result when executed using command line  "ruby -e  ..."
19388 [redmine@ru y] Bug #663: Benchmark.measure outputs different result when executed using command line  "ruby -e  ..."
19389 [mame@ts .n .] Thank you for your report.
19390 [artem.vorozt] I'm really sorry.

[Bug #664] ruby 1.9 (trunk rev 19846) segfaults
19391 [redmine@ru y] Bug #664: ruby 1.9 (trunk rev 19846) segfaults

[Feature #666] Enumerable::to_hash
19397 [redmine@ru y] Feature #666: Enumerable::to_hash

Response Splitting Risk
19399 [shugo@ru y- ] The potential risk of cgi.rb has been notified at

Error in revision extraction in 1.9.0?
19402 [shortcutter@] I noticed a weirdness with extraction of SVN revisions in 1.9.0 when
19407 [nobu@ru y- a] This is the revision of the whole repository, including the
19408 [shortcutter@] Ah!

[Bug #670] select doesn't handle fd's > FD_SETSIZE very well
19403 [redmine@ru y] Bug #670: select doesn't handle fd's > FD_SETSIZE very well

hashing
19404 [curth@mi ro ] Is there an "official" way to get from the value returned by "hash" to the =
19409 [matz@ru y- a] I consider it a implementation-dependent.  FYI, CRuby truncate the

rb_errinfo() vs rb_rubylevel_errinfo()
19410 [pbrannan@at ] What is the difference between these two functions?
19411 [ko1@at ot ne] def m
+ 19412 [ko1@at ot ne] s/current method/current thread/, of course.
+ 19418 [pbrannan@at ] Is this just an optimization that e and $! are the same place on the
| 19420 [ko1@at ot ne] I missed.  $! looking rb_thread_t#errinfo.
| + 19421 [ko1@at ot ne] missed again.
| | 19426 [pbrannan@at ] Thanks, this is what I thought was happening.
| + 19425 [pbrannan@at ] Yes, 1.8 saves ruby_errinfo on the stack when it evaluates NODE_RESCUE,
+ 19419 [pbrannan@at ] Also, I had been curious whether rb_rubylevel_errinfo() should be

Is this expected, or should I report it?
19413 [dave@pr gp o] Given
19414 [ko1@at ot ne] It seems a bug.
19415 [dave@pr gp o] Shall I add it to redmine?
19416 [ko1@at ot ne] I have already fixed it :)
19417 [dave@pr gp o] ...
19442 [shortcutter@] comes around the corner and asks for a plan how long a certain fix

Now that lambda has more powerful arguments...
19422 [dave@pr gp o] is there anything that
19423 [ed.odanow@wo] If you include "it can be very confusing" into the list of possible
19424 [dave@pr gp o] p(-> a, b  { a + b }.call(1,2))
+ 19427 [pbrannan@at ] Can a lambda have default argument values?
| + 19428 [dave@pr gp o] Yes.
| + 19429 [dblack@ru yp] Yes; I believe it's got the full complement of method argument
|   19430 [pbrannan@at ] So long as the default value doesn't include a '|' character?
|   19431 [dblack@ru yp] Yes; then you'd have to do b=(x|y). I suppose that counts as a
|   19432 [jim.weirich@] You were on the other side of the fence initially, right?  What
|   + 19433 [dave@pr gp o] At the risk of butting in, my two cents is that we no longer need the
|   | + 19435 [dblack@ru yp] It's your thread; you can't be butting in :-)
|   | | 19436 [james@gr yp ] Yeah, I'm a small fish in this big pond, but I'm all with Dave and
|   | + 19437 [transfire@gm] lambda{||}
|   |   + 19438 [dave@pr gp o] Then deprecate lambda.=B9 Because having two totally disjoint ways of =20=
|   |   | 19444 [transfire@gm] =20
|   |   | 19445 [dblack@ru yp] ...
|   |   + 19441 [meinrad.rech] ...
|   + 19434 [dblack@ru yp] I mean I'm rooting for the removal.
|     + 19439 [quixoticsyco] I acknowledge the long discussion on this topic which has already
|     + 19440 [jim.weirich@] Ahh, the world makes sense again.
|       19456 [charles.nutt] I think I've gotten confused by this thread...we are talking about
|       19457 [dave@pr gp o] I don't believe that's valid MRI syntax.
|       19459 [charles.nutt] Ok, I think I misinterpreted some other emails then. So the argument
+ 19443 [ed.odanow@wo] My comment was more in the direction, that some words and examples like

confused by this catch table
19446 [pbrannan@at ] irb(main):001:0> require 'internal/proc'
19447 [ko1@at ot ne] See vm.c:1199
19448 [pbrannan@at ] Ok, that make sense.  The retry is triggered by the rescue clause, and
19449 [ko1@at ot ne] Absolutely.

[Bug #677] LoadError bug in irb/init.rb
19450 [redmine@ru y] Bug #677: LoadError bug in irb/init.rb
19451 [Daniel.Berge] =20

[Bug #649] Memory leak in a array assignment?
19452 [redmine@ru y] Issue #649 has been updated by Roger Pack.

[Bug #664] ruby 1.9 (trunk rev 19846) segfaults
19453 [redmine@ru y] Issue #664 has been updated by Stefan Lang.

[Bug #678] Socket.getservbyport should convert the port param to network byte order
19454 [redmine@ru y] Bug #678: Socket.getservbyport should convert the port param to network byte order
19460 [nobu@ru y- a] I think it should check if the port number fits with int16_t.
19461 [matz@ru y- a] I think it's a good idea.  Please check in.

[Bug #678](Closed) Socket.getservbyport should convert the port param to network byte order
19455 [redmine@ru y] Issue #678 has been updated by Yukihiro Matsumoto.

Should Method@instance_methods reveal protected methods?
19458 [dave@pr gp o] The RDoc says it just returns public methods, but
+ 19462 [matz@ru y- a] Code says list method that is callable from outside (i.e. public + protected).
+ 19463 [emiel@re no ] Having played around with this some more I've found the following
  19464 [emiel@re no ] I forgot about singleton methods...

[Bug #680] csv.rb: CSV.parse is too late when encoding is mismatch
19465 [redmine@ru y] Bug #680: csv.rb: CSV.parse is too late when encoding is mismatch

Default source encoding (Was: [Bug #680] csv.rb: CSV.parse is too late when encoding is mismatch)
19466 [michael.seli] This bug actually brings up an interesting issue - should the source
19471 [duerst@it ao] A default for the source encoding has been discussed quite a long
+ 19473 [michael.seli] I am not sure I understand your argument about not defaulting the source
| 19474 [matz@ru y- a] You are correct here.
+ 19480 [james@gr yp ] I think I understand what you are saying here, which is that it's
  + 19481 [james@gr yp ] The real issue here is that you can argue the user doesn't even know
  | 19482 [nobu@ru y- a] Though I don't know about TextMate at all, ruby-mode.el in 1.9
  | 19483 [james@gr yp ] But that's not for Ruby scripts emacs is using though, right?  You are
  + 19526 [michael.seli] I too have been concerned about 1.8 script compatibility regarding
    19530 [james@gr yp ] They sure could, yeah.  Our policy for TextMate development has always
    19543 [duerst@it ao] Wouldn't it be only these scripts (including those that contain
    19551 [james@gr yp ] That's correct, just those scripts.  I have no idea which ones they

[Bug:1.9] failures of test/minitest
19468 [nobu@ru y- a] $ ./ruby test/runner.rb test/minitest
19491 [ryand-ruby@z] Fixed. thanks for the patch!

Re: [ruby-dev:36896] * lib/csv.rb: Failure test_inspect_is_ascii_8bit_encoded(TestCSVRow)
19470 [duerst@it ao] [forwarded from ruby-dev to ruby-core, with translation(t)/comments(c)]
19493 [james@gr yp ] Thanks so much for doing this.  You are really helping us bridge the

[Bug #680] csv.rb: CSV.parse is too late when encoding is mismatch
19475 [redmine@ru y] Issue #680 has been updated by Takeyuki Fujioka.

Ruby 1.8.7 Throwing "Too many open files" Exception lately???
19478 [admin@ha th ] Group,
19479 [shortcutter@] Sounds like an environmental (ulimit) or compile issue.  Did you rule those out?
19486 [admin@ha th ] Well I have not changed it in at least 18 months,

[Bug #649] Memory leak in a array assignment?
19484 [redmine@ru y] Issue #649 has been updated by Roger Pack.

Strange noise values bound to block params (1.9)
19485 [dblack@ru yp] I'm trying to write an example that shows that block args are now

[ANN] Sipper 1.1.3 Released
19487 [rubylearner@] ...
19488 [ryand-ruby@z] please don't post these to ruby-core. ruby-talk is appropriate, ruby-
19499 [why@ru y- an] Whenever Ryan does this, it's like citizen's arrest.
19502 [vjoel@pa h. ] Whenever _why does this, it's like marshmallows falling from Mars.
19509 [rubylearner@] ...

[Bug #680](Closed) csv.rb: CSV.parse is too late when encoding is mismatch
19492 [redmine@ru y] Issue #680 has been updated by James Gray.

[Bug #680] csv.rb: CSV.parse is too late when encoding is mismatch
19494 [redmine@ru y] Issue #680 has been updated by James Gray.

[Bug #641] GC patch to cache the most recent heap for is_pointer_to_heap
19497 [redmine@ru y] Issue #641 has been updated by Narihiro Nakamura.

Need a ruling: binding with a block?
19498 [charles.nutt] I need a ruling on this one.
19501 [nobu@ru y- a] Unspecified.
19513 [charles.nutt] Thanks!

[Bug #682] Rogue values bound to block parameters
19503 [redmine@ru y] Bug #682: Rogue values bound to block parameters

Is the stabby proc gone? broken?
19504 [dblack@ru yp] $ ruby19 -ve '->(){}'
19506 [matz@ru y- a] I am sorry, it was my mistake, and Nobu fixed it.
+ 19508 [dblack@ru yp] Does that mean it's gone? :-)
+ 19537 [shyouhei@ru ] No, not yet.
  19539 [nobu@ru y- a] It's a different story.
  19548 [dave@pr gp o] Can we just remove it? (Now would seem to be the perfect time, before

Drafting of Ruby International Standard
19505 [shugo@ru y- ] I have news to tell you today.

Possible memory leak when using $KCODE='u'
19507 [romanbsd@ya ] ...
19514 [nobu@ru y- a] What's the version (and the revision if possible)?
19519 [romanbsd@ya ] ...
19574 [romanbsd@ya ] Good day,
19587 [nobu@ru y- a] I can't see the leak.

[Bug #683] rdoc fails on singleton method of Object
19512 [redmine@ru y] Bug #683: rdoc fails on singleton method of Object

String literal encoding (Was: Default source encoding (Was: [Bug #680] csv.rb: CSV.parse is toolate when encoding is mismatch))
19515 [michael.seli] Sorry, perhaps I have been giving a (bad) solution, rather than stating
19517 [nobu@ru y- a] I'd prefer to default the internal encoding to the source
19518 [michael.seli] But then how do you tell Ruby NOT to set "default_internal"?
19522 [nobu@ru y- a] I think defaulting the internal encoding to something other is
19525 [michael.seli] Yes you are right, and I was not suggesting doing that.
+ 19531 [nobu@ru y- a] I don't like the idea setting default_internal from source
| 19532 [michael.seli] OK, I don't use Emacs, and no one told me that before, thanks! I assumed
| + 19533 [nobu@ru y- a] No.  Magic comments without -*- markers are for VIM, like
| | 19535 [michael.seli] Why then can I write a ruby 1.8 script which does a "puts" of a Shift_JIS
| | 19538 [nobu@ru y- a] Because you are avoiding troublesome chars.  Without such
| | 19540 [michael.seli] OK, I'm sure you know more about Japanese encodings that I do.
| | + 19544 [duerst@it ao] To give you the details, these characters, in Shift_JIS, are
| | + 19546 [nobu@ru y- a] It just seemed working by chance.
| |   + 19550 [james@gr yp ] I believe so, yes.
| |   | 19552 [matz@ru y- a] -Ku (or -KU) specifies to
| |   | 19553 [james@gr yp ] OK, well this meet's TextMate's need then.  Thank you.
| |   + 19562 [michael.seli] Actually I think that any ruby program which uses ASCII everywhere except
| |   + 19655 [nobu@ru y- a] ===================================================================
| |     + 19656 [nobu@ru y- a] Sorry, missed these 2 lines.
| |     + 19657 [duerst@it ao] Great work!
| |       19661 [nobu@ru y- a] -E equals to --encoding.
| + 19542 [duerst@it ao] The simplest case is a script in US-ASCII only, but where you want
|   19545 [michael.seli] But as UTF-8 or any other default_internal setting has to be ascii
+ 19541 [duerst@it ao] I'm not sure what you mean by "Japanese". It's no problem at all
  19547 [nobu@ru y- a] I don't do it of course, but know that some people love to do

[Bug #685] LoadError warning on debug mode
19520 [redmine@ru y] Bug #685: LoadError warning on debug mode

path to library directory to RDoc and RubyGems should use
19521 [yugui@yu ui ] ...

Too Many Files Error -- Test Case Produced.
19523 [admin@ha th ] ...
19524 [tommy.nordgr] A d.close here will fix the problem.
19527 [admin@ha th ] OK -- I will try that!
19529 [matz@ru y- a] We didn't change the Dir class in 1.8.7 as far as I remember, so I

failure of MiniTest#test_filter_backtrace and test_filter_backtrace_unit_start
19528 [yugui@yu ui ] Ryan, two tests will fail when Ruby is built at a directory except
19534 [yugui@yu ui ] ...
19536 [ryand-ruby@z] backported. thanks for the heads up...

[Bug #682](Closed) Rogue values bound to block parameters
19549 [redmine@ru y] Issue #682 has been updated by Yuki Sonoda.

Messages Being Blocked
19554 [james@gr yp ] I've sent a query to Ruby Core twice now and it has never shown up.
19556 [james@gr yp ] James Edward Gray II

Managing 1.9 threads in extensions
19555 [dave@pr gp o] I'm trying to pin down the rules for folks who write extensions for
19557 [ko1@at ot ne] Such macros are not provided.  Please use rb_thread_blocking_region().
19558 [dave@pr gp o] Ah, OK.
19559 [ko1@at ot ne] a comment in thread.c.
19563 [dave@pr gp o] So, if folks have time...

stable snapshot has no readme
19560 [ryand-ruby@z] The stable snapshot has no README file. README.EXT is there, but not

Was there a feature freeze on October 25th?
19561 [dave@pr gp o] Curious authors want to know... :)
19567 [yugui@yu ui ] Yes, the feature of branch ruby_1_9_1 was already frozen. But the fork
19568 [duerst@it ao] Is it okay (or maybe even better) to commit just to HEAD and to
19631 [yugui@yu ui ] It's ok.

Ruby 1.9.1 preview1 is out
19564 [yugui@yu ui ] I released Ruby 1.9.1-preview 1. You can download it from the following

[Bug #641] GC patch to cache the most recent heap for is_pointer_to_heap
19565 [redmine@ru y] Issue #641 has been updated by Roger Pack.

GC thought
19566 [roger.pack@l] Here is a recent patch I've been experimenting with--for any advice. [1]
+ 19569 [kbloom@gm il] Letting the program continue execution during the mark phase could cause
| 19575 [roger.pack@l] Hmm. Good points.  I suppose that would err.
| + 19576 [roger.pack@l] if you ever call ObjectSpace.each_object [or dereference a weak hash]
| + 19577 [pbrannan@at ] Requiring all ObjectSpace methods to grab a mutex and wait for the
|   19596 [shortcutter@] What about scenarios like this?
|   + 19597 [charles.nutt] Yes please :)
|   + 19600 [hongli@pl n9] There's at least one important use case for ObjectSpace: closing all
|   | + 19602 [nobu@ru y- a] Did you want to write like this?
|   | | + 19603 [hongli@pl n9] Yes that's what I mean.
|   | | + 19613 [vjoel@pa h. ] What about something like
|   | |   19614 [hongli@pl n9] This won't do us any good. The goal isn't to close operating system file
|   | + 19605 [charles.nutt] ObjectSpace is more than just inefficient...it and features like it are
|   | | + 19616 [kbloom@gm il] ObjectSpace shouldn't go.
|   | | | 19626 [charles.nutt] id2ref is just about as bad, since in order to support id2ref we have to
|   | | + 19643 [kurt@ca hn t] The sooner we can get rid of ObjectSpace.each_object, the sooner we can
|   | |   19644 [matz@ru y- a] CRuby already does.
|   | + 19635 [roger.pack@l] This is very interesting.  I may have run into something similar to it
|   |   19638 [hongli@pl n9] You mean an example of the problem? I can't seem to be able to come up
|   |   19641 [roger.pack@l] Which OS's might those be?  I know that when a forked child process
|   |   19642 [hongli@pl n9] Varies. I'm pretty sure that it's not because of the OS, but because of
|   + 19615 [roger.pack@l] Hmm.  Thought.
|     19617 [hongli@pl n9] I actually have thought about aligning the heaps for fast lookup, as you
|     19629 [roger.pack@l] Good point.  I'd suppose you'd have to 'overallocate' slightly or
+ 19570 [pbrannan@at ] I wonder how (or if) this is any different from a parallel collector.
+ 19707 [roger.pack@l] On second thought one problem that could occur is that if a program is

[Bug #690] Ruby 1.9.1 preview1 build error
19571 [redmine@ru y] Bug #690: Ruby 1.9.1 preview1 build error

Should HEAD report itself as 1.9.0 or 1.9.1?
19572 [dave@pr gp o] Currently, we have
19573 [usa@ga ba ec] I think 1.9.2 is better because both 1.9.0 and 1.9.1 will never

[Bug #691] Time::zone_utc? does not follow rfc2822
19578 [redmine@ru y] Bug #691: Time::zone_utc? does not follow rfc2822

[Bug #692] non-ascii comment with Encoding.default_internal set raises Encoding::InvalidByteSequenceError
19579 [redmine@ru y] Bug #692: non-ascii comment with Encoding.default_internal set raises Encoding::InvalidByteSequenceError

1.9 gem weirdness question
19580 [roger.pack@l] I was wondering if anyone could shed some insight onto this
19586 [drbrain@se m] There's something weird with const_missing and defined? that hasn't
19588 [ryand-ruby@z] that's prolly their magic/clevar const_missing in action.

[Bug #693] Thread#set_trace_func seems not to work
19581 [redmine@ru y] Bug #693: Thread#set_trace_func seems not to work
threads.html
top