5340-6077 subjects 5820-6240

^ Get Request
5568 [ruby-core ha] # get 1173-1176

^ Get Request
5569 [ruby-core ha] # get 1173-1176

^ [bug] wrong stacktrace, last entry
5570 [kero chello.] I have a reproducable stack trace of which the last entry is wrong.

^ [PATCH] Ruby 1.8.2 DTrace Support, Round Two
5575 [pabs pablotr] Richard is having problems subscribing to the list (ruby-talk.org

^ Wandering in the Wilderness Project
5576 [ruby-core ha] I recently became fascinated with Ruby.  I am starting a personal
5577 [drbrain segm] You're missing Fixnum representation.
5579 [nobu.nokada ] 1-a is Fixnum.

^ Wandering in the Wilderness Project
5584 [ruby-core ha] That leaves Question #2 and an Extra Thought
5594 [nobu.nokada ] Yes.

^ Wilderness Project
5598 [ruby-core ha] ==========================================================================================
5599 [nobu.nokada ] class Fixnum

^ Pathname#walk for traversing path nodes (patch)
5609 [ruby-ml magi] Here is a small addition to Pathname against 1.9, probably suited
+ 5612 [akr m17n.org] I'm not sure that "walk" is a good name.
| 5615 [dblack wobbl] Maybe "scan".
| + 5642 [akr m17n.org] I think "scan" is acceptable.
| | 5643 [dooby d10.ka] I've followed the recent protests that a Pathname is not
| | 5645 [dblack wobbl] I don't think any given class can monopolize a method name, unless
| | 5646 [dooby d10.ka] David A. Black
| | 5647 [dblack wobbl] My first choice would actually be split, but that's already in use for
| + 5644 [dooby d10.ka] It's iterating over each_dirpath isn't it ?
| + 5661 [ruby-ml magi] Coming up with a name for this method is a bit of a problem. #each_dir,
|   5667 [meta pobox.c] I'd like to suggest "descend", because it's really descending into the
|   + 5668 [paul luon.ne] I seem to use it the other way around as well.  Paths are just a bit like
|   + 5669 [dblack wobbl] The thing is, though, it isn't really descending.  The path may not
|   | 5670 [meta pobox.c] Only if you were doing the entire tree. It's still descent, even if you
|   | 5672 [dblack wobbl] Whoops.  I got that all screwed up in my mind.  Never mind, then....
|   + 5701 [akr m17n.org] Now I think "descend" (and "ascend") is acceptable.  So they are
|     5707 [Daniel.Berge] I can't remember - was it decided that the root directory should be
|     + 5709 [akr m17n.org] First, I asked it in [ruby-core:5641].  The reply, [ruby-core:5660],
|     + 5718 [meta pobox.c] My vote would be for including the root directory. If you include it,
+ 5641 [akr m17n.org] It seems that Pathname#walk doesn't yield the root directory.
  5660 [ruby-ml magi] You are absolutely correct. How about this version?
  5675 [akr m17n.org] I'd like to hear your usage of Pathname#walk.
  5678 [transfire gm] If so, will you also introduce "ascending" version? I can easily think

^ Re: [ ruby-Bugs-2290 ] TypeError: IO#reopen( anIO )
5640 [nobuyoshi.na] It is not deprecated.

^ File.extname edge case bug?
5651 [Daniel.Berge] Ruby 1.8.2
+ 5653 [tilman code-] File.extname(".some_app.conf") => ".conf"
| 5654 [Daniel.Berge] Hrm...I guess you're right Tilman.  We'll just have to accept the
+ 5655 [dblack wobbl] ./.mime.types
+ 5656 [meta pobox.c] I agree that extname("..") should return "", but I disagree with the
  + 5657 [Daniel.Berge] Roger that.
  + 5658 [Daniel.Berge] Based on this, which I think is correct, I submit the following patch.  I've
    5659 [nobu.nokada ] You don't consider the case no suffix but path contains a dot.
    + 5666 [Daniel.Berge] Oops, you are correct.
    + 5795 [matz ruby-la] Check in, please.

^ Postgrey
5662 [shugo ruby-l] I installed Postgrey (http://isg.ee.ethz.ch/tools/postgrey/)

^ [ ruby-Bugs-2298 ] Curses bkgd functions loose their high bits when converting
5664 [noreply ruby] Bugs item #2298, was opened at 2005-08-24 15:59
5677 [ocean m2.ccs] Umm, chtype in curses seems not to be `char' rather `long'.

^ [ ruby-Bugs-2301 ] ruby182-15.exe CTD's on installation
5671 [noreply ruby] Bugs item #2301, was opened at 2005-08-24 19:55
5673 [tom infoethe] I've closed this with a note to the submitter that he should post this

^ Wilderness Project
5674 [ruby-core ha] Can someone tell me the CONCEPT behind the
5679 [Daniel.Berge] Can you please tag these subject lines differently so we have more of a

^ uri test failures.  (Re: [ruby-cvs] ruby/lib, ruby/lib/uri: Lovely RDOC patches from mathew (metaATpoboxDOTcom) on URI/* and getoptlong.rb)
5676 [akr m17n.org] It causes following test failures.
5689 [meta pobox.c] The test code is incorrect. As documented in RFC 1738 section 3.2.2 (and
5690 [meta pobox.c] One other thing: I did definitely mention the bug fix was included in
5691 [warrenbrown ] mathew,
5720 [meta pobox.c] Yes, it does.

^ [ ruby-Bugs-2308 ] Problem with mkmf and spaces in directory names?
5680 [noreply ruby] Bugs item #2308, was opened at 2005-08-25 13:42
5682 [djberg96 yah] I brought this up back in ruby-talk:104505.  It
5683 [usa garbagec] Ruby building process uses mkmf.rb.
5684 [nobuyoshi.na] Does this patch help?
5710 [matz ruby-la] If it does, commit the fix.

^ CoW [Was: Re: Wilderness Project]
5681 [drbrain segm] They are for copy-on-write.

^ Wilderness Project
5685 [ruby-core ha] OK - I see where ELTS_SHARED is used to implement COPY-ON-WRITE
5688 [halostatue g] I'll repeat someone else's request. Please start using alternate
5692 [ruby-core ha] I Will Do That ---  If it appears I am distupting this list I will go away!
5693 [Daniel.Berge] No, the questions are fine.  Just need a more concise subject line is

^ STR_ASSOC [Was: Re: Wilderness Project]
5686 [drbrain segm] Please keep the Subject sensible.

^ [PATCH] native extention by rubygems
5687 [nobuyoshi.na] Rubygems 0.8.11 fails to compile a native extention which

^ WIlderness::  Why do strings when created also seem to create a SHARED object (rb_str_new3)
5694 [ruby-core ha] While exploring the ELTS_SHARED flag, I  would think the shared object
5772 [cmills frees] This saves having to create a new char array each time the 'a =

^ Wilderness:: The purpose of "ifnone" entry in RHash Structure
5695 [ruby-core ha] Appears to be associated the Procedure Block ?
5696 [ mfp acm.org] It stores either the default value or the Proc invoked when a

^ Gateway comp.lang.ruby -> ruby-talk   (was: Postgrey)
5697 [dooby d10.ka] Thanks Shugo.
5699 [shugo ruby-l] Oh, I'm sorry.
+ 5700 [dooby d10.ka] ====================================================================
+ 5702 [matju arteng] I wonder how many months this postgrey thing will stay effective. Isn't it
  5703 [drbrain segm] Postgrey has auto-whitelisting where well-behaved hosts will no

^ net/http question - trace and options
5704 [Daniel.Berge] I'm curious - why aren't the "trace" and "options" methods implemented in
5705 [aamine lover] net/http CVS HEAD already has TRACE/OPTIONS and other WebDAV methods.

^ File.join edge cases
5706 [Daniel.Berge] Ruby 1.8.2
5708 [nobu.nokada ] Originally, it was just a wrapper for
5794 [matz ruby-la] Can you commit this fix?

^ [ ruby-Bugs-2371 ] Lexic confusion: method/local variable distinction works strange
5711 [noreply ruby] Bugs item #2371, was opened at 2005-09-04 00:40
+ 5712 [transfire gm] It has something to do with the fact that it is a symbol parameter
+ 5713 [nobu.nokada ] Which version do you use?  Worked fine for me.
  5714 [transfire gm] Mine is 1.8.2 and it does raise syntax error.
  5715 [matz ruby-la] Then try newer one, or wait for 1.8.3 release.
  5716 [transfire gm] No, really? :-P

^ [ ruby-Bugs-2383 ] uninitialized constant REXML::Parsers::SAX2Parser::Text
5721 [noreply ruby] Bugs item #2383, was opened at 2005-09-05 21:14

^ [BUG?] SystemExit & abort_on_exception in latest "ruby_1_8"
5724 [holmberg iar] I have noticed a change in the effect of Thread.abort_on_exception in
5725 [nobu.nokada ] Sorry, a bug.
5726 [matz ruby-la] Can you commit the patch?

^ executing linux shell commands with ruby??
5727 [praneeth147 ] Is there ne module in ruby that executes linux shell commands
5728 [chneukirchen] Use

^ [PATCH] Valgrind support
5729 [tilman code-] one way to be able to valgrind ruby applications is to use a suppression file, which tells Valgrind to ignore all the errors it thinks it found in Ruby.
5730 [nobu.nokada ] How better than suppression files?
5740 [tilman code-] Once you have a *perfect* suppression file there might not be any difference at all.

^ Ruby development issue tracking will go to basecamp
5731 [matz ruby-la] Dear Ruby CVS Committers, and watchers,
+ 5732 [ville.mattil] I'm not committer, but active watcher. Could I have account to basecamp. I
| + 5733 [jqshenker gm] Could there be an unprivileged "guest" account for people like me wholike to watch but don't actually develop?
| | 5734 [binary42 gma] From what I recall, basecamp supports RSS feeds. That might (I am not
| + 5735 [matz ruby-la] For all watchers, I will set up "guest" account or publish RSS feed,
|   5757 [matz ruby-la] Don't even try to hack it.
+ 5736 [cboos bct-te] The day you switch to SVN, you may want to consider Trac again...

^ returning strings from methods/instance_methods
5737 [transfire gm] I was just wondering why with #methods and #instance_methods, it was
+ 5738 [matz ruby-la] Those methods predate Symbols.
| + 5741 [chneukirchen] When were Symbols introduced to Ruby?
| | 5745 [matju arteng] I started with Ruby 1.4.3 and I remember that calling String#intern
| | 5746 [transfire gm] BTW, am I double posting? Or is it just gmail making it look like I am?
| | 5751 [jqshenker gm] Gmail threading stuff shows both your "reply" and the reply actuallyrecieved from the ml server. Annoying, but actually not that bad.
| + 5742 [transfire gm] I see. So symbols came later on. What brought that on?
|   + 5743 [dominique.br] irb(main):001:0> s = "blahhoo"
|   | 5744 [transfire gm] "blahhoo".to_sym
|   + 5749 [matz ruby-la] Symbols are more lightweight objects than strings, since they are
+ 5739 [florgro gmai] I think Ruby used to return a Fixnum for :foo in the past meaning that

^ YAML.load hang-up
5747 [zn mbf.nifty] YAML.load hangs with some string.
5748 [ruby-core wh] Thankyou.  The new versions of Syck fix this.  I am merging right now

^ File.split edge cases
5750 [Daniel.Berge] Ruby 1.8.2
5754 [nobuyoshi.na] See [ruby-talk:78960].
5755 [Daniel.Berge] From what I can gather from my research, the current behavior is not
5756 [matz ruby-la] Then show me your reasoning.  I wish you are correct.
5758 [Daniel.Berge] path == File.split(File.join(path))
5759 [matz ruby-la] This sounds good, but it's kinda hard to define exact behavior for
+ 5760 [halostatue g] And this is *mostly* true on Windows, too.
| 5761 [matz ruby-la] Currently, there's officially no timeframe in 1.9 development.  I hope
+ 5762 [Daniel.Berge] Another interesting difference is the way Python interprets a trailing
| 5763 [halostatue g] Assuming that File.split should be implemented as [ File.dirname,
| 5764 [Daniel.Berge] Yeah, the closer I look at posixpath, the more I realize it's....wrong.
| 5765 [halostatue g] One might. ;)
| + 5766 [Daniel.Berge] I don't know that just because "ls" handles it one can conclude that
| | 5767 [halostatue g] #include <sys/types.h>
| | 5771 [akr m17n.org] //... may be implementation-defined.
| + 5769 [matz ruby-la] I don't say it's correct.  It just happens to work (on Unix).  I'd
+ 5778 [Daniel.Berge] File.join(File.split("foo")) # "./foo"

^ assert_raises
5752 [drbrain segm] From ruby-talk 155793
5773 [transfire gm] Why is assert_raises  being removed?

^ Wilderness: AutoLoad Feature (variable.c)- Need General Description
5753 [ruby-core ha] When determining whethre a Constant is defined - It calls 'autoload_file' .
5770 [florgro gmai] ...

^ Wilderness: AutoLoad Feature (variable.c)- Need General Description
5768 [ruby-core ha] When determining whethre a Constant is defined - It calls 'autoload_file' .

^ YAML broken
5774 [zn mbf.nifty] 4 failures and 1 error in recent chkbuild logs.
5775 [matz ruby-la] YAML is not broken.  Tests are not updated yet.  Thank you.
5776 [zn mbf.nifty] I confused YAML test failures and NameError of YAML::Store.
5777 [ruby-core wh] You're right.  I am working on these warnings right now.  Thank you, kazu!

^ Strange 'ambiguous arguments' warning
5779 [ntalbott gma] Why does line 5 in the following program trigger the 'ambiguous first
5780 [drbrain segm] t / 5 / 4

^ array sharing
5781 [eric_mahurin] This is my first time poking around in the ruby source code, so
5782 [meta pobox.c] You can use realloc to make an existing allocated chunk of memory
+ 5783 [eric_mahurin] What I was thinking was that the first #unshift would yield
+ 5785 [matju arteng] Most of the time realloc will fail to extend and instead will
  5809 [eric_mahurin] I've been looking at the code a bit more regarding element

^ Documentation for 1.8.3
5784 [drbrain segm] resolv.rb and uri.rb are missing from 1.8's lib/.document.  Can they

^ Difference between class declarations
5786 [calamitas ad] I would have expected these pieces of code to do the same, but it
+ 5788 [nobu.nokada ] The scope is determined lexically now.  Older versions till 1.6
| + 5792 [calamitas ad] Just curious: why was this changed? As it is now, self of a block can be
| + 5881 [transfire gm] So how can one get around this? I need to do so for my #testcase. i.e.
|   5885 [nobuyoshi.na] What do you want to get around, precisely?  Do you want a
|   5886 [transfire gm] def testcase(name=nil, &block)
|   5962 [matju arteng] I've also encountered a similar problem trying to use Class.new{}. It
|   5972 [transfire gm] Peter and I have discusses how one might create blocks the close off
+ 5789 [matju arteng] When I originally asked for Class.new{} (one of the first RCR's if I
+ 5790 [transfire gm] I have to concur that the current behavior isn't optimal. It means the

^ Wilderness: How does __autoload__ work??
5787 [ruby-core ha] <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
5791 [nobu.nokada ] rb_autoload() in variable.c

^ syck.so compile error
5793 [s-urabe par.] I noticed that make fails in ext/syck. See

^ proposed attr writer patch
5796 [Daniel.Berge] attr_accessor :foo?, :bar!
+ 5797 [Daniel.Berge] Just to clarify, this is actually Evan's code which I'm simply proposing
+ 5799 [nobu.nokada ] It doesn't work.
| 5803 [djberg96 yah] <snip updated patch>
| 5810 [matz ruby-la] I don't think so.  Its release day is too close.  There's chance for
| 5811 [nobu.nokada ] Will I commit it to 1.9?
| 5819 [matz ruby-la] Only when we are sure that's the right behavior.  I'm not sure yet.
+ 5812 [ruby jamesbr] Why would someone write
  + 5813 [matju arteng] It can be said that Ruby has so many exceptions that just another
  | 5814 [djberg96 yah] The current behavior is broken.  So, it should either
  | 5817 [ruby jamesbr] Broken? Does that not depend on a certain point of view?  Is it broken
  + 5816 [chneukirchen] IMHO, `foo?' makes sense (if @foo is meant to be boolean), but calling
    5827 [matju arteng] IMHO, FOO$ also makes sense, if it is meant to be a string.
    + 5828 [chneukirchen] Nice troll. :P
    + 5829 [djberg96 yah] Oh, let's not go over the top, shall we?
      5830 [ruby jamesbr] I see.
      + 5831 [transfire gm] Though I had done so myself in the past with I no longer think it
      + 5832 [jeremy bitsw] I, too, run into this a lot and wish attr_accessor :foo? "just worked."
      + 5833 [djberg96 yah] Why should it?  The use of "?" and "!" in method names

^ [ ruby-Bugs-2472 ] Makefile error in OpenSLL extension (on Windows)
5798 [noreply ruby] Bugs item #2472, was opened at 2005-09-16 18:56
5800 [nobu.nokada ] Which version is it exactly?
5801 [curt.hibbs g] This is the just released 1.8.3 preview2.
5802 [nobu.nokada ] Hmmm, does win32/Makefile.sub contain following two lines?
5805 [curt.hibbs g] No, win32/Makefile.sub doe not contain those two lines.
5815 [nobu.nokada ] Strange.
5821 [curt.hibbs g] I assume you mean "ruby\win32\config.h". I did that and tried rebuilding,=
5822 [nobu.nokada ] That file in both preview2 and preview3 surely has them.  Check
5823 [curt.hibbs g] I cannot find them. I attached a copy of my win32\Makefile.sub.
5824 [nobu.nokada ] It is version in preview1.  Preview2 contains
5825 [curt.hibbs g] I figured it out, this was my fault. I apologize for wasting your time.

^ Wilderness: Is there any way to invoke rb_define_class() from the Ruby Level
5804 [ruby-core ha] <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">

^ 5**(-1/2) gives incorrect results under mathn.
5806 [erlercw gmai] With ruby 1.8.1,
5808 [sheepman she] The following patch will fix this problem.

^ Wliderness: Ruby Code to force AUTOLOAD sequence to run
5807 [ruby-core ha] <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">

^ Weird SIGCHLD-related problem
5818 [benschumache] I'm experiencing an odd issue with a fork'd child process and I was