3293-3547 subjects 3729-4093

^ [PATCH] XMRPC WEBrickServlet
3544 [k.shutemov s] /usr/share/ruby/vendor_ruby/1.9/xmlrpc/server.rb:795:in

^ realtime Ruby GC
3545 [matju sympat] I work with PureData, a realtime audio/video application, and I made a

^ Bug report: CGI std lib, MSIE Mac OS X, multipart forms, and image buttons.
3548 [sera fhwang.] Over on ruby-talk some folks had said that this is the place to report

^ frame_dup(), frame_free() [ruby_core 3456]
3549 [decoux moulo] The original example is in [ruby-core:3456]
3550 [matz ruby-la] Yes!  I confirmed the problem was fixed.  Thank you!

^ ubygems missing?
3551 [transami run] I've never been one for compiling code, so I bet this is a simple fix, but
+ 3552 [hgs dmu.ac.u] Strange. I've not seen this.  Which platform? Did you pass options
| 3553 [transami run] ./configure
+ 3554 [nobu.nokada ] Doesn't environment RUBYOPT exist?  Or check the actual command
  3555 [rich infoeth] You have installed rubygems and set the RUBYOPT variable.  This is the
  3556 [transami run] Ah, that did it. Thanks!

^ ruby-1.8 / regex.c / gcc 3.4.2
3557 [decoux moulo] gcc 3.4.2 give some warnings for regex.c (not really important, but ...)

^ CSV, string handling.
3558 [hgs dmu.ac.u] Have I discovered a bug in CSV?  If data is read in and it is

^ Missing rdocs
3559 [B.Candler po] In 1.8.2-preview2, there are a number of cases where libraries seem to be
3560 [dave pragpro] If the owner of a file says it is complete, then we can sure add it.

^ 1.8.2 - what can we do to help?
3561 [dave pragpro] The new PickAxe assumed that 1.8.2 would be released by it's
3562 [matz ruby-la] Use stable snapshots, find bugs, report them (or fix them ;-), and keep
+ 3563 [hgs dmu.ac.u] [Dave Thomas asked what we can do for 1.8.2]
| 3564 [matz ruby-la] It's on ruby-dev mailing list.  Non Japanese speaker can use ruby-core
| 3565 [hgs dmu.ac.u] RubyForge has some bug/issue tracking facilities which might be less
| 3566 [matz ruby-la] I'm just too lazy to set up and try.  If some one help me, RubyForge
| 3567 [transami run] So there will be new mailing list? ruby-bug? And ruby-core twill be for what?
| 3568 [matz ruby-la] If I understand correctly, bugs and issues related to release will be
| 3570 [hgs dmu.ac.u] That was the way I was thinking.  I've never created a RubyForge
| + 3571 [rich infoeth] Working it...
| + 3572 [curt hibbs.c] I can do this. I'll wait about an hour to make sure someone else hasn't
|   3585 [curt hibbs.c] I knew I should wait... Rich Kilmer is already doing this.
+ 3569 [sera fhwang.] Maybe you're just talking about the core itself here, but this seems to
  + 3577 [gsinclair so] Slightly.  My perspective is that ruby-core is a perfectly suitable
  + 3578 [markus reali] Let's say you are understandably frustrated.  I saw your post and
  + 3592 [matz ruby-la] Which one?  Probably I missed it.
    3594 [tom infoethe] Francis, perhaps you could bless the new Ruby bug tracker with its first

^ [BUG] Small issues with Symbols
3573 [florgro gmai] Moin!
+ 3574 [jim freeze.o] Could be because $0 has a value and $1 is nil.
| 3579 [florgro gmai] irb(main):132:0> /(.)/.match("x"); $1
+ 3575 [florgro gmai] 6) It is possible to create an empty symbol literal via :"#{""}"
| 3576 [jim freeze.o] irb(main):007:0> :"#{nil}"
| 3580 [florgro gmai] Quite odd, because "#{nil}" works. Maybe it was an attempt to protect
+ 3582 [transami run] Should it?
  3583 [florgro gmai] Yes, it is a special syntax. But alias :foo :bar works, so I think the
  3584 [transami run] I see. I dod not realize you could use the symbols there.

^ standard libraries [was: 1.8.2 - what can we do to help?]
3581 [sera fhwang.] I appreciate the offer, but I think I'm more concerned about the
3586 [ruby jamesbr] The was a post recently on ruby-talk on a related topic (it was yet
3587 [dave pragpro] I don't think so. The problem with the RDP, and I anticipate with any
+ 3588 [hgs dmu.ac.u] I think this is laudable,
+ 3593 [matz ruby-la] Let us discuss and define the standard for the core libraries, and the
+ 3602 [sera fhwang.] This addresses what has to happen before a library is accepted into the

^ Bug tracking project on RubyForge...
3589 [rich infoeth] All,
3590 [Daniel.Berge] Excellent.
3591 [rich infoeth] Sure...
3595 [matz ruby-la] I want any submissions to be mailed to me in any way, directly or via
3597 [rich infoeth] It should not be too noisy...we could only notify on the original submission
3599 [matz ruby-la] Tell me the address (from: field value) of the incoming mails from
3603 [tom infoethe] Currently the From: field contains "noreply@rubyforge.org".
3604 [batkins57 gm] +1 for the P.S.

^ Float and Bignum
3596 [transami run] What does Ruby currently use for FP and Bignum? I've been reading a little on
3598 [matz ruby-la] Ruby uses its own implementation of Bignum.  I'm not sure what FP
3601 [transami run] Sorry, Floating Point.
3605 [halostatue g] Because it places restrictions on embedding Ruby. Languages are too

^ Ruby Vs. ... might find comparison of interest.
3600 [transami run] ( o _  елеще┴
3606 [ruby jamesbr] He doesn't know enough Ruby to be making such comparisons.
3607 [transami run] True. But that's not the point. That kind of info is "out there". And this
+ 3608 [ruby jamesbr] Ah.  I see.
| 3609 [transami run] You da Man!
+ 3612 [timsuth ihug] Maybe mention that anything that can be done inside an object can be

^ Tadayoshi Funaba's Date2
3610 [transami run] Tadayoshi Funaba has a lib on RAA called Date2, the additions/improvements to
3629 [dblack wobbl] I haven't looked closely for implementation differences, but just
3647 [transami run] He doesn't seem to make a comparison between his work and Ruby's, except to
3648 [dblack wobbl] standard distribution.
+ 3649 [kjana dm4lab] <http://www.ruby-lang.org/cgi-bin/cvsweb.cgi/ruby/lib/date.rb>
| 3652 [transami run] Great! Thank you very much.
+ 3651 [transami run] Oh, I see. I misread (over looked the word "core"). Sorry.

^ [BUG] Memory leak in ruby_1_8
3611 [dross code-e] I have a small memory leak. I tried using cvs to get ruby_1_8 right now.
+ 3613 [dgtized gmai] would that be because the entire ruby file is client.rb and everytime
| 3614 [dross code-e] *sigh* Sorry, mozilla mail fails me again.
+ 3619 [matz ruby-la] Can you check the attached patch?
  3625 [dross code-e] bada boom! The patch works fine. Thank you Matz, now I can continue

^ [ ruby-Bugs-1000 ] TEST BUG
3615 [noreply ruby] Bugs item #1000, was opened at 2004-10-28 23:12

^ [ ruby-Bugs-999 ] CGI: multipart forms, image buttons, and MSIE Mac OS X
3616 [noreply ruby] Bugs item #999, was opened at 2004-10-28 22:18
3621 [matz ruby-la] Recent check-in (2004-10-27) fixed the MSIE multipart problem on OSX.

^ [ ruby-Bugs-1000 ] TEST BUG
3617 [noreply ruby] Bugs item #1000, was opened at 2004-10-28 09:12
3620 [transami run] That was supposed to be "test _comment_ to go with", but you get the idea.

^ [BUG] CVS problem
3618 [dross code-e] There is a problem with the CVS, not sure if its cvs. It may be the CIA
3622 [matz ruby-la] I'm not sure what is CIA bot and what I can do with CVSROOT/loginfo,
3623 [dross code-e] oh, okay. I was unaware of the ruby-cvs. I always watched CIA(the cvs
3624 [matz ruby-la] If you (or others) find out what we can do to fix, tell us please.

^ Licensing (was: Float and Bignum)
3626 [transami run] Is that right, matz?  LGPL is too restrictive in this way?  Would a more
+ 3627 [matz ruby-la] That is right.  Ruby 1.8 comes with regex.c which is covered by LGPL.
+ 3628 [shiba mail2.] (This is my first English post; Sorry for my broken English ;| )
  3632 [transami run] Mine's probably worse and I'm English ;|  No worries... :)

^ [ ruby-Bugs-999 ] CGI: multipart forms, image buttons, and MSIE Mac OS X
3630 [noreply ruby] Bugs item #999, was opened at 2004-10-28 13:18

^ [ ruby-Patches-998 ] XMLRPC and WEBRick
3631 [noreply ruby] Patches item #998, was opened at 2004-10-28 07:09

^ [ ruby-Patches-1004 ] Typo and grammar/style fixes for ext/win32ole/win32ole.c
3633 [noreply ruby] Patches item #1004, was opened at 2004-10-29 09:56

^ [ ruby-Patches-1004 ] Typo and grammar/style fixes for ext/win32ole/win32ole.c
3634 [noreply ruby] Patches item #1004, was opened at 2004-10-30 01:56

^ Fwd: Re: GMP
3635 [transami run] I contacted Tomasz Wegrzanowski <taw@users.sf.net> concerning is work with

^ [ ruby-Patches-1005 ] eval.c: xstr literal can be changed
3636 [noreply ruby] Patches item #1005, was opened at 2004-10-29 20:14

^ MBignum in the core (was: Fwd: Re: GMP)
3637 [evan falling] Just to chime in quickly. I have done another implementation of Bignums

^ Ruby, pthreads, and HPUX 11
3638 [jgb3 email.b] I'm finally trying to delve into the issue of Ruby not compiling
+ 3640 [jgb3 email.b] Incidentally, I just tried compiling Ruby 1.9, and it segfaults with the
+ 3650 [decoux moulo] No idea, can you try to compile it with
  3657 [jgb3 email.b] Thanks for fielding a suggestion, Guy. Unfortunately, I still get the
  3658 [binary42 gma] That would be. Have you tried changing any of the optimization
  3659 [jgb3 email.b] -g -DYYMAXDEPTH=300
  3660 [decoux moulo] Well, try to look if it don't exist a bad interaction on HP/UX between
  3662 [jgb3 email.b] I haven't seen any such reports, which surprised me. In particular, it
  3663 [jgb3 email.b] I did just encounter the following statements in a man page on

^ [ ruby-Bugs-999 ] CGI: multipart forms, image buttons, and MSIE Mac OS X
3639 [noreply ruby] Bugs item #999, was opened at 2004-10-28 13:18

^ [ ruby-Patches-1005 ] eval.c: xstr literal can be changed
3641 [noreply ruby] Patches item #1005, was opened at 2004-10-29 18:14

^ [ ruby-Patches-1006 ] TEST PATCH: More email address munging
3642 [noreply ruby] Patches item #1006, was opened at 2004-10-29 19:12
3643 [tom infoethe] I apologize for all the test emails coming to the list.  I am trying to

^ Consolidation of Ruby projects on RubyForge
3644 [tom infoethe] Right now we have two Ruby-core-ish projects on RubyForge - the
3646 [matz ruby-la] It's OK for us.  I will update the links.
3677 [tom infoethe] Thanks,

^ [PATCH] dl.txt explaining PtrData#ptr,ref
3645 [timsuth ihug] I posted this patch last night under ruby-talk:118202.

^ range.c
3653 [transami run] I have been working on improving Range class today per discussions on
+ 3654 [transami run] BTW, Matz, are you still planing to alias #member? to #include?  That fits in
| 3665 [matz ruby-la] Yes.  They are aliases already in 1.9.
+ 3664 [matz ruby-la] Hmm, I don't know why EXCL() is summed up twice.  How about

^ [BUG] autoload
3655 [vjoel PATH.B] (I'm not sure if this is still the right place to report bugs. I didn't
+ 3656 [gsinclair so] I like to see bugs reported here.  The auto-generated emails from
+ 3661 [decoux moulo] This is this [ruby-list:40085]
  + 3666 [matz ruby-la] It's a bug in autoload.  Joel, Thank you for reporting.  And Thank you,
  + 3667 [vjoel PATH.B] The issue seems to be whether or not foo.rb should be added to $" before
  + 3668 [nobu.nokada ] Sorry, I forgot to modify rb_provided().
    3669 [matz ruby-la] @@ -6625,3 +6625,17 @@ rb_provided(feature)
    + 3670 [vjoel PATH.B] Works for me. Thanks!
    + 3671 [nobu.nokada ] Ah, rb_feature_p() should not be modified, maybe.

^ Including a module twice
3672 [Peter.Vanbro] I know Ruby tests whether a module is already included, and doesn't
3673 [matz ruby-la] @@ -414,3 +414,2 @@ rb_get_method_body(klassp, idp, noexp)
3674 [Peter.Vanbro] Thanks. I wasn't sure it was this simple. This has the downside that a
3675 [matz ruby-la] I didn't know neither.

^ GNUish build process, suggest allowing `make check`.
3676 [hgs dmu.ac.u] Given that the build process for ruby is very like many GNU

^ at_exit, trap sigint, and sleep. no good in 1.8, ok in 1.6
3678 [homernile ya] Any ideas about why the following code hangs in 1.8
3679 [usa garbagec] This was bug of ruby-1.8, and already fixed on CVS.

^ [ ruby-Bugs-1013 ] RSS: Can't make children of RSS::Rss
3680 [noreply ruby] Bugs item #1013, was opened at 2004-11-01 23:17

^ [ ruby-Bugs-1015 ] latest ruby cvs will not compile if old c compiler used.
3681 [noreply ruby] Bugs item #1015, was opened at 2004-11-02 06:08

^ x86_64 -- warning and errors
3682 [decoux moulo] uln% gcc -v
3683 [matz ruby-la] Does this patch improve the situation?
+ 3684 [decoux moulo] yes and no
+ 3685 [decoux moulo] apparently it need also this

^ [ ruby-Bugs-1020 ] win32ole bug - segfault when using ole_get_methods
3686 [noreply ruby] Bugs item #1020, was opened at 2004-11-03 17:44

^ lib/ostruct.rb 1.12 - wrong commit?
3687 [eivind FreeB] The above mentioned commit makes OpenStructs into singletons, which
3689 [matz ruby-la] We know the problem.  Wait for a while.  We will consider your patch.
3690 [batsman.geo ] Does the following make sense?
3691 [matz ruby-la] It does, except for marshal data incompatibility.

^ [ ruby-Bugs-1021 ] problem with break points on non-executable lines
3688 [noreply ruby] Bugs item #1021, was opened at 2004-11-04 12:57

^ "Old" CVS-revisions of Ruby fail to compile
3692 [holmberg iar] I just tried to compile some old revisions of Ruby, by checking out
+ 3764 [holmberg iar] I haven't seen any answer yet.
+ 3799 [nobu.nokada ] A cvs 1.11.x issue.  It ignores default branches in the past.
  3815 [holmberg iar] Thanks for the reply.
  3817 [nobu.nokada ] It depends the server, not the client.

^ [ ruby-Bugs-1028 ] ruby crash while installing rake gem
3693 [noreply ruby] Bugs item #1028, was opened at 2004-11-07 10:22

^ Ruby 1.8.2 preview3
3694 [matz ruby-la] I have _finally_ prepared the ruby 1.8.2 preview3 at
+ 3696 [matz ruby-la] No.
| 3697 [matz ruby-la] Oops, I meant yes.  I.e. the CVS head (1.9) no longer deletes the
| + 3698 [florgro gmai] Is this behavior unchanged?
| | 3699 [matz ruby-la] Yes.
| + 3708 [matt lickey.] Yes, sorry -- I didn't notice that my CVS tree became 1.9 instead of
+ 3700 [tom infoethe] Yours,
| 3705 [tom infoethe] Also, Matz, the link from the text "[...] and its rubyforge mirror" on
| + 3706 [dblack wobbl] I've gone ahead and made the change for you.
| | 3709 [tom infoethe] Cool, thanks!  I wasn't sure who else had access to change that page...
| + 3707 [matz ruby-la] done.
+ 3701 [hgs dmu.ac.u] Did you want to know about warnings? Sun Solaris, with GCC
  3702 [hgs dmu.ac.u] I'm getting
  3703 [dave pragpro] You've likely got the most recent one - that version number hasn't
  3704 [hgs dmu.ac.u] OK, thanks. I've still not got into CVS in depth yet, so can't

^ [ ruby-Bugs-1036 ] Segment fault is occued when load iconv on mingw32.
3710 [noreply ruby] Bugs item #1036, was opened at 2004-11-09 23:28

^ Converting nil to NULL
3711 [Daniel.Berge] I asked this on ruby-talk, but I thought I should ask here as well.
+ 3712 [cmills frees] char *
+ 3713 [matju sympat] In addition to Charles' solution, which is good, I'd like to say that the
  3714 [nobu.nokada ] unless/end
  3715 [matju sympat] Thank you. So there are _nine_ of them. The rest of my message stays

^ possible bug in method dictionary
3716 [ryand zenspi] installed.
3717 [nobu.nokada ] It silently redefines and discards old one.
3720 [ryand-ruby z] OK. I think I have a more clarified issue now. Thank you very much Nobu
+ 3721 [vjoel PATH.B] Isn't this behavior an expected consequence of the method search order
+ 3722 [nobu.nokada ] Mixed-in methods can't override the methods defined in the
  3723 [ryand zenspi] That is exactly what I needed to know. Thank you for the clarification.

^ File.link on Win32
3718 [Daniel.Berge] I noticed that win32.c in 1.8.2 p3 contained a link() function (or maybe
3724 [timsuth ihug] File.link uses CreateHardLinkA which creates a "hard link" (using
3725 [Daniel.Berge] I see, and Windows doesn't like it when I use '.lnk' as the extension.

^ ruby-mode.el patch
3719 [ysantoso-rub] for_foo do |bar|

^ Fixnum#clone and Float#clone raise different exceptions
3726 [dblack wobbl] irb(main):001:0> 3.clone
3749 [dblack wobbl] Just replying to myself, in the hope of triggering response from Matz
3751 [matz ruby-la] Hmm, do you mean the fact that clone fails for Numbers etc?
3752 [dblack wobbl] No, I mean it fails in different ways.  Fixnum#clone gives a
+ 3819 [dblack wobbl] Following up, based on some examination of source and some chatting w/
| 3821 [gsinclair so] 4.clone -> 4 sounds better to me, but I must be missing something.
| 3825 [matz ruby-la] Do you accept the following behavior?
| + 3826 [florgro gmai] I think this is okay. I think getting an exception when doing obj.clone
| | 3833 [batsman.geo ] Why is the attached patch unacceptable? I can see two problems, but
| | 3835 [dblack wobbl] That probably has a bearing on my earlier claim, too -- namely, that
| + 3827 [gsinclair so] Yes, I do accept that behaviour, because setting instance variables of
| | 3828 [matz ruby-la] No, it's impossible to distinguish two immediate values.
| + 3830 [dblack wobbl] I don't like it -- but even more fundamentally, "clone" really is
|   3837 [matju sympat] Well, it depends what kind of formal definition you give to clone. Pick
+ 3829 [matz ruby-la] They both raise TypeError on the latest CVS.  We'll see how it works.

^ [ ruby-Bugs-1047 ]  rss/maker is requiring <image> for rss/0.9
3727 [noreply ruby] Bugs item #1047, was opened at 2004-11-13 00:41

^ 1.8.2-preview3 warnings
3728 [decoux moulo] Not really important