17391-25625 subjects 17890-18784

^ .Net and languages.
17655 [erik bagfors] I just read that ximian will be doing a .Net-implementation for linux

^ Old chestnut: invariants, pre/post conditions and so on
17656 [hgs dmu.ac.u] Given the discussion in the past on Eiffel-like features
17682 [hal9000 hype] so on
17685 [maki inac.co] TAKAHASHI Masayoshi (maki@inac.co.jp)
17689 [mikkelj-anti] I'm not familiar with Eiffel, so I may be off the track but...
+ 17691 [andy toolshe] I started on an implementation of Design by Contract for Ruby
| 17708 [hgs dmu.ac.u] I know of this one...
+ 17730 [aleksei.guze] Adiscussion on C-style method overloading had taken place at the list
  17738 [hgs dmu.ac.u] Pre- and post-conditions are not only for type checks, of course.

^ Environment variables with exec?
17664 [onge mcs.anl] Does the Kernel exec method support passing environment
17763 [tsyk yk.rim.] It doesn't.  Kernel::syscall might be an alternative.

^ vi and {}'s
17667 [aaron.barnet] i'm new to the ruby scene and have mostly been using perl/java/c++ up to
17699 [djkea2 mugca] Regards,

^ module_function
17668 [markn cs.mu.] Is there a document somewhere describing keywords or modifiers
+ 17670 [decoux moulo] These are not keywords, but private methods from Module. There are
+ 17673 [Dave Pragmat] You'll find the attr_ and module_function stuff in there. If you use

^ IO in a Ruby extension
17671 [Brett.W.Denn] I need to write a Ruby extension which reads binary data from a C FILE
17672 [decoux moulo] Look at ruby.h
17718 [kjana os.xax] And you missed useful macros defined in rubyio.h? :-P
17719 [decoux moulo] This is this, that sometimes, I don't like

^ Ruby warnings
17674 [pbrannan atd] I have just discovered Ruby's -w option, and I think it's really nifty.
+ 17679 [neumann s-di] Just leave the parens away and the warning disappears!
+ 17680 [tsyk yk.rim.] def foo(x); x + 2; end
  17717 [pbrannan atd] p foo (4 + 6) / 2

^ LinuxDevices 0.1
17678 [matju sympat] Announcing LinuxDevices 0.1

^ Ruby, with out-of-the-box XML support
17683 [dnm pobox.co] Although this applies directly to XML-RPC and not just XML, I think

^ YART (Yet Another Ruby Talk)
17686 [ jimm io.com] This talk was given to the New York City CTO Club on July 10, 2001.

^ Ruby jobs
17688 [peterhi shak] I was looking on Job Serve here in the UK (things aren't going too well down

^ Funky file.each_byte
17694 [nlper jungle] Howdy!  I was filtering some files and ran into an anomaly.

^ ANNOUNCE: ruby-tmpl-0.1...
17695 [sean-ruby-ta] Howdy.  ruby-tmpl-0.1 has been released for review.  It is
17696 [ruby jamesbr] What is ruby-tmpl-0.1?
17713 [sean-ruby-ta] Sorry I didn't make a better announcement earlier, I was a tad

^ ANN: libxml2 Rubified!
17698 [waisun.chia ] This is a multi-part message in MIME format.
17768 [maki inac.co] Have you ever see libgdome-ruby?
17769 [waisun.chia ] Yes, I've seen libgdome-ruby.
17784 [maki inac.co] Thanks. I see now (maybe).

^ sound editor + ruby
17710 [bil ccrma.St] I recently added Ruby as an embedded language in my sound editor

^ Null Pattern
17721 [K.Hodges fte] This is a multi-part message in MIME format.
+ 17725 [decoux moulo] One small problem with your implementation is (your class is in null.rb)
+ 17771 [matz ruby-la] There's only one way: hack the parser.  Pseudo variables: nil, false,
  + 17783 [aleksi.nieme] def null
  | 17788 [K.Hodges fte] I think that a special exception handling mechanism could be implemented to give this behaviour.
  + 17785 [K.Hodges fte] Dear Matz, thanks for your replies
  | 17818 [avi beta4.co] It might be worth pointing out that in Objective-C, like Ruby and unlike
  + 23428 [K.Hodges fte] Have you had any further thoughs on this matz?

^ Null Pattern
17722 [K.Hodges fte] and
17770 [matz ruby-la] Sorry, there's no such way.  I've learned users often go beyond
17786 [ugly-daemon ] Would wrapping it in another if statement work??
17789 [K.Hodges fte] I am afraid I dont think so Chris,

^ Fwd: Re:  Re: Problems compiling FOX for Ruby (FXRuby) on NetBSD
17724 [ser germane-] charset="iso-8859-1"

^ File.tail?
17727 [lists.ruby-t] Just curious as to whether anybody has already cooked up a File.tail
18331 [lists.ruby-t] Well, since no responses, I just hacked together the following.  I'm

^ Array#sort! returns nil when array empty
17729 [mike lepton.] Array#sort! returns nil if the array is empty, whereas ri says it should
17731 [decoux moulo] pigeon% ruby -ve 'p [].sort!'
17732 [hfulton pop-] says it should
+ 17733 [hal9000 hype] Ah, now *that* is interesting!
| + 17735 [decoux moulo] pigeon% ruby -ve 'p "a".gsub!(/c/, "")'
| + 17764 [matz ruby-la] (a) to provide the cheapest way to detect modification.
+ 17734 [decoux moulo] This is the goal of the null pattern, but you can always write
+ 17736 [pbrannan atd] I almost agree.  I don't like this feature, but I have used it.  In an irc
  17739 [decoux moulo] pigeon% ruby -e 'p ("a b c :d e f".split(/:(\S.*)|\s+/) - [""])'
  + 17742 [hal9000 hype] I'm now thinking that perhaps the "right" way to do this
  | 17745 [pbrannan atd] foo = "This is a test"               #=> "This is a test"
  + 17746 [pbrannan atd] Very cool!
    17748 [hal9000 hype] Of course, Paul. Thank you.

^ Ruby/C extension for module with nested classes
17740 [Brett.W.Denn] I am trying to create a Ruby extension that consists of a module containing
17741 [decoux moulo] ruby call only *one* Init_ method for the extension, and in your case this

^ What is Array#- ?
17747 [jim freeze.o] While following the Array thread, I noticed the minus
17752 [Dave Pragmat] Forgive the doubled minus sign - wee LaTeX -> HTML conversion
17753 [jim freeze.o] Yes, but why remove duplicates? Doesn't seem to make sense.
+ 17755 [Dave Pragmat] It's been a long time for me, but aren't sets defined _not_ to have
| 17757 [jim freeze.o] Hmm..  :)
| + 17759 [Dave Pragmat] It's just that you can't have 5 apples in a set. They'd need to be in
| | 17761 [avi beta4.co] Ahh, you beat me to it.
| + 17760 [avi beta4.co] Well, if you have 2 pairs of apples that makes 4, and if you take away 4
| + 17776 [Stephan.Kaem] But not that much confusing. Array#+ can't be used, it's already given waway
| + 25645 [Stephan.Kaem] But not that much confusing. Array#+ can't be used, it's already given waway
+ 17775 [Stephan.Kaem] Yes.
+ 17805 [Stephan.Kaem] I didn't intend to say that I do not like it the way it is. In fact, I do.
+ 25649 [Stephan.Kaem] Yes.
  + 17779 [matz ruby-la] Both behaviors are rational, I think.  Last time I made decision, I
  | + 17794 [Dave Pragmat] Wouldn't it be clearer to define a new Set type, rather than new
  | | + 17823 [hal9000 hype] I agree... but there still might be decisions to be made.
  | | | 17825 [maverick cs.] If we implement sets using balanced trees, then we can support the natural
  | | + 17847 [matz ruby-la] Yes.  When Set is what you want, Set is the cleanest way.
  | |   17851 [billk cts.co] Does this look reasonable, performance-wise ?
  | + 17796 [ jimm io.com] I have toyed with the idea of implementing Set, Bag, OrderedCollection, and
  + 17790 [jim freeze.o] Thanks for the clear description. If we analyze this from a 'unit'
    17795 [hgs dmu.ac.u] for now?

^ Colorado Software Summit
17754 [b redhookgro] Are any Ruby-ists going to the Colorado Software Summit in October

^ Re: .Net, JVM and languages.
17756 [matju sympat] I think Ruby for JVM is much more important, because the list of
+ 17797 [mikkelj-anti] I think it is important for Ruby to be running on .Net and JVM.
| + 17799 [hgs dmu.ac.u] Does the existence of the various front ends and back ends for GCC
| | + 17800 [armin approx] Yes. gcc3.0 can even compile java classes into native machine code now. [gcj]
| | + 17831 [mikkelj-anti] [gcj]
| | + 25616 [mikkelj-anti] [gcj]
| + 18859 [matju sympat] What you are saying is that languages should be giving up features until
|   18874 [mikkelj-anti] in
|   18921 [frankm bayar] I've done a little reading into .NET's Common Language Runtime, and I'll
+ 17855 [dblack candl] That makes sense to me.  But it also raises the question (going back

^ File.size()
17765 [joe vpop.net] What does everyone think of making File.size an instance method as
17766 [Dave Pragmat] Try f.stat.size
17767 [joe vpop.net] Yes that is what I am using.  Just curious if f.size might be a good
17782 [aleksi.nieme] This is somewhat discussed idea. Please see the threads starting at

^ What happened to Ruby-Fltk?
17773 [adamon MailA] I heard about Ruby bindings to FLTK on this newsgroup, but the Ruby-Fltk

^ Regex Groups in String#split (Re: Array#sort! returns nil when array empty)
17777 [pit capitain] I've never seen this feature documented anywhere. Very tricky!
+ 17778 [decoux moulo] One use of '- [""]' is for
| 17792 [jim freeze.o] ruby -e 'p " a b c :d e f".split(/\s+:(\S.*)|\s+/) - [""]'
+ 17780 [tobiasreif p] With
  17781 [decoux moulo] Typo, who have forgotten |

^ Overriding IO#close
17791 [pit capitain] Is there a way to reliably override IO#close ?
17896 [matz ruby-la] No, sorry.  You can use finalizers instead, if you don't need them

^ GC mark
17798 [aleksei.guze] "The mark routine will be called by the garbage collector during its
17801 [aleksei.guze] ... Can I increase the reference counter while not in "mark phase".Just
17802 [decoux moulo] Can you give a small example of what you are trying to do, because in ruby
17803 [aleksei.guze] I know of Ruby GC having not counting :))))
17804 [decoux moulo] * window class
17807 [aleksei.guze] #     * window class
17813 [ljohnson res] You're going to have to maintain some mapping from those C struct pointers

^ bug or feature: yield calling method object
17808 [pit capitain] class C
17809 [decoux moulo] pigeon% cat b.rb

^ Re: Old chestnut: invariants, pre/post conditions
17811 [andy toolshe] Time is the fire in which we burn...
17812 [hal9000 hype] Ah, Dr. Sorin, we meet again.

^ ANN: FXRuby 0.99.173
17814 [ljohnson res] I am pleased to announce the latest release of FXRuby, the Ruby

^ Re: Ruby-enabled web hosting services?
17815 [toddg linux1] Perhaps very like the Perl Mongers?

^ Ruby bug?
17817 [   xm w3d.ru] xm@ns:~$ ruby -v
17826 [matz ruby-la] Fixed in the latest CVS.

^ func-menu additions for Ruby
17819 [hotfusionman] I got frustrated with func-menu.el not supporting Ruby, so I added the

^ RUBY NEWSGROUP FAQ -- Welcome to comp.lang.ruby!  (Revised 2001-05-26)
17827 [schneiker ju] RUBY NEWSGROUP FAQ -- Welcome to comp.lang.ruby!  (Revised 2001-05-26)

^ Class variables
17830 [aleksei.guze] Why class variables are shared by subclasses?
17870 [senderista h] Class variables seem to mar the uniformity of the Ruby class model,

^ Creating methods on the fly
17832 [harryo zipwo] I would like to create a new method on the fly.  Assuming I have a
+ 17835 [neumann s-di] methName = "myMethod"
+ 17838 [b redhookgro] You could use the operation "send" on an object to invoke a method at
+ 17854 [harryo zipwo] Yep; that should work.  What if I want to create a global function?  I
+ 17858 [harryo zipwo] That worked a treat!  Thanks very much for the help.
| 17859 [harryo zipwo] Just out of interest, what's the difference between using "eval %{ ...}"
+ 25622 [harryo zipwo] That worked a treat!  Thanks very much for the help.
| 25623 [harryo zipwo] Just out of interest, what's the difference between using "eval %{ ...}"
| + 17860 [Dave Pragmat] 'self' is available.  Remember that in Ruby, global methods aren't
| | 17861 [harryo zipwo] Can someone explain why the following code fails to define the new method
| + 17863 [harryo zipwo] That works OK if I do it by having a method in the class that then calls
| + 17878 [dblack candl] None -- I just use the here document out of habit for multi-line evals.
| + 17891 [neumann s-di] %{ ... } is just a multi-line string, and I like it more than "here" documents.
| + 25630 [harryo zipwo] Can someone explain why the following code fails to define the new method
| + 25634 [harryo zipwo] That works OK if I do it by having a method in the class that then calls
|   + 17865 [Dave Pragmat] If I understand what you're after, you want to define an instance
|   + 17867 [harryo zipwo] Actually, what I want to be able to do is something like ...
|   + 17868 [harryo zipwo] Eureka!  While that code you gave as an example wasn't doing it the way I
|   + 25631 [harryo zipwo] Actually, what I want to be able to do is something like ...
|   + 25656 [harryo zipwo] Eureka!  While that code you gave as an example wasn't doing it the way I
+ 25652 [harryo zipwo] Yep; that should work.  What if I want to create a global function?  I
  17856 [dblack candl] meth = "talk"   # using string interpolation just to illustrate

^ Extending objects
17833 [aleksei.guze] What about extending object's memory representaion. Like adding some
17837 [decoux moulo] Can you give an example for this ?
17887 [aleksei.guze] After [ruby-talk:17870] I undertand class object instance variables and

^ Ruby-newbie seeks help with Rubywin starting IRB
17834 [lucid users.] Once I fire up Rubywin, and then invoke _R_uby _I_rb from the
17839 [feoh fourfuz] The 1.6.4 release for Windows is broken, several other people have reported
17841 [Dave pragmat] Andy's back and is looking at it.

^ Specialised data structures - good or bad?
17836 [spwhite char] There's been a move towards implementing sets, bags, red-black trees and
+ 17844 [Dave Pragmat] I agree with all that you say, with one possible exception. In
| + 17848 [spwhite char] I like the idea of creating a separate data structures package for those
| + 17883 [matz ruby-la] but don't take my words too seriously, especially when I said
+ 17849 [maverick cs.] I don't know if this is the correct reasoning or not. To me, it seems it is
| 17850 [spwhite char] I considered the possibility of inheriting from Array to create Set, which
| 17857 [harryo zipwo] That is certainly true.  Logically, the only reason one would want to
+ 17866 [harryo zipwo] I agree with that.  There are cases where you really want the semantics
+ 25639 [harryo zipwo] I agree with that.  There are cases where you really want the semantics
  25795 [mchahn facel] The way I make a set is to use the keys of a hash as a set.  I just set the

^ Another bug (now segfaults)
17840 [   xm w3d.ru] The following fails on ruby 1.7.1 (2001-06-22) [i686-linux] and ruby 1.7.1 (2001-07-03) [i686-linux] - seems some sort of buffer
+ 17843 [decoux moulo] Guy Decoux
+ 17845 [Dave Pragmat] There's a bug reporting system. You can add entries over the web at
+ 17885 [matz ruby-la] oops, I'll add buffer length check to rb_file_s_expand_file().

^ Bug ... probably another one
17842 [   xm w3d.ru] def expand_path(path, _path_info = '')
17846 [decoux moulo] pigeon% cat b.rb
17884 [matz ruby-la] You're right.  I should give you CVS account.  Will you accept?
17886 [decoux moulo] I've a problem, like I've said previously, with this patch. It's not
17895 [matz ruby-la] All other occurrence of rb_str_update() is called after

^ Ruby Idioms
17852 [spwhite char] to sort the values into key order.
17853 [spwhite char] Sigh. I thought of this after posting...

^ Path name strangeness under Ruby/Win32
17862 [davidkrider ] to
17956 [lucid users.] The easiest way would be to replace "LightPoint" with a new

^ Trapping Ruby exceptions from C
17864 [senderista h] I'm trying to solve the problem of how to trap a Ruby exception from a
+ 17871 [decoux moulo] Probably I've not understood what you want to do, but I'll write something
+ 17882 [matz ruby-la] Use pointer to the variable, i.e.
+ 17892 [senderista h] Thanks!  This sounds like just what I was looking for.
+ 17893 [senderista h] That was my first thought, until I saw that the fourth argument to
  17894 [matz ruby-la] Yes, I admit.  I want these arguments to be something like (void*),
  17903 [matt lickey.] In Unix98 (or some similar standard) has intptr_t and uintptr_t just
  17904 [matz ruby-la] Thank you for interesting information.  I'll check them

^ ruby question
17869 [jobeicus hot] class Foo

^ Overloaded methods
17872 [harryo zipwo] A while ago, someone asked whether it was possible to overload methods in
+ 17873 [decoux moulo] The approach of the P language is weird (bad syntax for me) and *slow*.
+ 17874 [harryo zipwo] *slow*.
  + 17875 [decoux moulo] Java (TM), but I don't want to have problem with Sun's lawyers if I forget
  | 17888 [aleksei.guze] Everyone can check types af the actual parameters at the beginning of
  | 17889 [decoux moulo] Look at [ruby-talk:16957] and try to do it with your test.
  | 17953 [harryo zipwo] I figured I'd learn something about Ruby in the attempt, which was
  | + 17969 [decoux moulo] This is not a problem default values
  | + 17975 [harryo zipwo] Yes, that was what I was getting at.  Ie, Ruby already provides the kinds
  + 17876 [harryo zipwo] Fair enough.  I have nothing they could weasle out of me :-).
    + 17877 [decoux moulo] Guy Decoux
    + 17879 [harryo zipwo] Thanks. I've printed it out.
    + 17880 [pulsar qks.c] I'd not heard that term before either. It is a rather weird although