35478-38534 subjects 35906-36673

^ XmlParser, Expat help
35674 [chris.morris] I've got the 1.6.6 for Windows installed. It comes with xmlparser, but when

^ [ANN] Ruby dataquery shell (RDQS) 0.3.3
35677 [cochi uni-pa] RDQS (Ruby DataQuery Shell) version 0.3.3 was released

^ Deleting read only files on Windows
35678 [chris.morris] Ruby 1.6.6 built with VC++ cannot delete read only files, where as 1.6.5
+ 35685 [chris.morris] Here's what I came up with. I have a clutilwin.rb that I started to collect
+ 35722 [nobu.nokada ] require 'ftools'
  35733 [chris.morris] Cool, thanks!

^ [FXRuby] getting selection in FXTable
35679 [rasschaert i] how could i get the selection range from an FXTable object ?
35686 [ljohnson res] how

^ Windows File.mode doesn't exist
35682 [chris.morris] irb(main):008:0> RUBY_VERSION
35688 [ljohnson res] To directly access the mode bits, you can use the "mode" instance method of
35732 [chris.morris] D'oh! Got in a rush yesterday and didn't realize I was reading the docs

^ RubyCOM and Lotus Notes
35687 [martinkauer ] I'd like to use RubyCOM with Lotus Notes. The call to

^ Module Syntax
35692 [sean celsoft] Being primarily a C++ programmer, I'm happy with Ruby's Module concept
+ 35711 [matz ruby-la] No, it didn't work as you might expect.  It defined a new method of
| 35713 [sean celsoft] That's actually what I wanted to do.  My first declaration of MyFunction was
| 35714 [matz ruby-la] Yes.
+ 35717 [bob_sideboth] module A
  35752 [kentda stud.] module MyInsanelyLongModuleName
  35754 [decoux moulo] This is also usefull when you want to define class method, i.e.

^ [ANN] packaged modules
35693 [list chromat] I've started a repository of .rpk packages, the following is currently

^ [ANN] rpkg 0.3
35694 [list chromat] rpkg - Ruby Package Management System 0.3
35699 [james rubyxm] This looks great!  I wanted to try it on Windows 2000, where I have the
+ 35701 [Dave Pragmat] That should probably change to be testunit (which is in the Windows
| + 35728 [pdcawley bof] If you look at CPAN.pm under perl you'll see that it has about five
| | 35740 [list chromat] Errr... sorry if this is a stupid question, but how does this relate
| | 35744 [elanthis awe] I think the point was that CPAN can make use of a number of different
| | 35763 [list chromat] I see.  However, the issue here was not on the fetching method---rpkg
| + 35738 [list chromat] M-x doctor <RET>
|   35746 [Dave Pragmat] Basically, yes. Installing the packager should ideally be as simply as
|   35760 [list chromat] I will look into this, it is something that bugged me as well when
+ 35739 [list chromat] Thank you James, that is exactly the kind of feedback I need.  A
  35788 [gehlker fast] Welcome to Darwin!

^ [ANN] Ruby-Gnome 0.27 is out!
35696 [nconway klam] I've released version 0.27 of Ruby-Gnome. Ruby-Gnome is a collection of

^ Common usages for continuation
35706 [ysantoso jen] I was wondering how many people have used continuation and in what
35712 [pit capitain] I used continuations to convert an internal iterator to an external

^ testunit, was RE: [ANN] rpkg 0.3
35715 [tsondergaard] Am I the only one who is confused about the turn of events with unit test frameworks for ruby? It seems testunit is suddenly preferred by most everybody. Why?
35716 [dsafari xtra] Thomas,
35718 [tsondergaard] Strange answer :-) The next question is of course then: why has testunit been chosen for distribution?
+ 35743 [rich infoeth] It is the merging of Lapidary and RubyUnit (backward compatibility APIs for
| 35747 [nathaniel ta] Er, actually there's only backwards compatibility for RubyUnit... does
| 35762 [list chromat] Fine with me, I'll happily convert my Lapidary tests to Test::Unit
| 35769 [nathaniel ta] does
| 35830 [list chromat] Ok, then rpkg gets preinst scripts.
| + 35833 [chris.morris] Yeup. The installer associates the file extension with ruby.exe, so that's
| + 35842 [gehlker fast] Just assume Macs are running FreeBSD. The differences only matter to the
+ 35745 [Dave Pragmat] Because

^ RPMs...
35720 [sean chitten] Err..  just curious but are there RPMs for Linux available for Ruby?
+ 35724 [erik bagfors] Isn't ruby included in most distributions.  In the dists I use they are.
+ 35750 [ljohnson res] I don't know about an official RPM source for Ruby RPMs, but a search on

^ [ANN] Ruby/SMB beta 2
35725 [    f lck.nu] * Files are automatically reconnected if connection is lost

^ key comparision inside Hash
35731 [Vladimir.Doz] ehlo.
+ 35735 [erik bagfors] From what I understand you were wrong...
| 35753 [decoux moulo] One way (and *not* the best) is to use XOR (or ADD). Something like this
+ 35748 [matz ruby-la] Ruby is an opensource software.  You can always read the source.
| 35843 [Vladimir.Doz] ehlo.
| 35862 [matz ruby-la] Do you mean you're proposing binary search instead of hash algorithm
| 35878 [Vladimir.Doz] ehlo.
| 35880 [matz ruby-la] Because the default semantic of "eql?" is equivalence ("equivalence"
| 35885 [Vladimir.Doz] ehlo.
| + 35892 [kentda stud.] That kind of default behaviour sounds like it might have worse
| | 35895 [Vladimir.Doz] ehlo.
| + 35903 [matz ruby-la] I meant "same".  Two objects need to be same for "eql?" to be true.
|   35910 [Vladimir.Doz] ehlo.
+ 35767 [nobu.nokada ] This may not work as you want.  Generally <=> method returns,

^ how do i: add_observer( { small bit of code } )
35742 [mrp cre.cano] observing an object, i want to run a small bit of code when update is
35749 [matz ruby-la] How about using Proc?
35823 [mrp cre.cano] more context required from me!  i'm using the observer mixin.  putting
35836 [matz ruby-la] I guess a wrapper is the best solution, or you can define a "functional
35838 [Dave Pragmat] That is _sweet_!

^ Ruby and Makefiles
35751 [rubydev.dewe] I'm using SWIG to wrap some C code so I can access it from a Ruby script. My problem is that the swig-generated C requires to access the Ruby's ruby.h file.

^ process two arrays pairwise?
35757 [ronjeffries ] Is there somewhere in Ruby a way to process the elements of two arrays (of equal
35759 [paul atdesk.] Paul
+ 35765 [list chromat] Also the map2 method in the xbuiltins package.
+ 35780 [ronjeffries ] Hmm, well, those aren't it. Thanks! Maybe I like what I had best for now: it has
  + 35800 [vjoel PATH.B] There's also
  | 35808 [ronjeffries ] Well, it's cool. Seems like overkill though. I could be wrong ...
  + 35818 [pit capitain] Don't know exactly what you didn't like about those, but just in
    35863 [ronjeffries ] Thanks!

^ Development Efforts
35764 [stathy.toulo] I was wondering what the process is for getting involved with Ruby directly
35768 [tobiasreif p] Tobi

^ [ANN] Ruby/zlib 0.5.0
35766 [unnie blue.s] (Changes from 0.5.0-pre6)

^ Threads + Sockets == Instability?
35770 [matt lickey.] 100), each doing a little bit of network stuff (either TCP or UDP).
35778 [rubydev.dewe] ...

^ Ruby mention in April 2002 DDJ
35773 [ptkwt shell1] The article is entitled: Java's new Considered Harmful

^ Re: Method type 'abstract' (Module#abstract)
35774 [ptkwt shell1] The code I posted previously doesn't work, however I played with it and
35786 [kentda stud.] Ugh! Some Java/C++ idiom just reared its head in the back of my skull
+ 35789 [nat.pryce b1] No it shouldn't.  Ruby lets you define singleton methods on objects. You
+ 35821 [ptkwt shell1] Perhaps, but you could redefine it later (classes are always open) or add

^ Anyone using the WASTE editing library with Ruby?
35776 [ptkwt shell1] Is anyone using the WASTE text editor library with Ruby?
35779 [cjack cjack.] WASTE is a Mac OS text editing library, originally open source. It's not the
35791 [gehlker fast] I agree. What would be the point of a Carbon binding?

^ [ANN] Ruby/zlib 0.5.1 (was: Re: [ANN] Ruby/zlib 0.5.0)
35781 [unnie blue.s] Sorry, there is a serious bug about GzipReader in 0.5.0.
+ 35795 [dempsejn geo] Not sure if the zlib flaw is what you're referring to, but to those who
+ 35807 [list chromat] Will this go into the standard distribution?  I had rpkg working with
  35951 [unnie blue.s] Yes. The other day, I imported Ruby/zlib into rough/ext for review

^ runtime lookup of constants
35782 [slumos yahoo] Is it possible to get the value of a constant given its name as a
35783 [gotoken notw] That is Module#const_get(const).

^ testunit - setup -> set_up ?
35787 [chris.morris] I'm just starting to use testunit instead of rubyunit ... I noticed with an
+ 35792 [matt lickey.] As you probably know, you can use the RubyUnit compatibility layer if
+ 35793 [nathaniel ta] To answer the first question, Test::Unit uses #set_up because it's
  + 35796 [Dave Pragmat] I agree with you about aliasing.
  | 35797 [nathaniel ta] Ah, but JUnit uses setUp... two words. So which is the "xUnit way?"
  | + 35804 [Dave Pragmat] Sigh... good point.
  | | + 35812 [matt lickey.] Believe it or not, I've used inheritance and done things like call
  | | | 35817 [Dave Pragmat] Nothing stops you from using inheritance here - the per_test_setup and
  | | + 35860 [nathaniel ta] Don't sigh... we can get through this if we all stick together :-)
  | |   35868 [Dave Pragmat] Yes! I like that.
  | |   + 35887 [ronjeffries ] We did this very thing in sUnit lo! these many years ago when all this was being
  | |   + 35939 [nathaniel ta] of
  | |     36008 [zohr im.eth.] I want to unsubscribe from ruby talk. What is the way?
  | |     36105 [mark.firesto] <SARCASM>You know.. what I really wanted was about fifteen messages from you
  | + 35816 [matz ruby-la] I just miss "setup" just for its conciseness.
  |   35865 [nathaniel ta] Woah... pressure comes to bear ;-)
  |   35888 [ronjeffries ] Doesn't sUnit use setup?
  |   35936 [nathaniel ta] way?"
  |   35985 [matt lickey.] I don't parse Ruby statements into English ones before I understand
  + 35801 [matt lickey.] I like this best, though I'd only warn if set_up is also absent.  It

^ Re: testunit - setup -> set_up or setUp?
35799 [pete mcbreen] As a recovering C++/Java programmer I have to admit to having written a
35810 [nathaniel ta] Have no fear, I didn't mean to imply that I was considering using

^ What does RDoc do with aliases?
35802 [ljohnson res] The RDoc home page says that RDoc "handles" aliasing. I was interested in

^ Problem cross compiling Ruby
35803 [kero a26074.] When I cross compile ruby 1.6.7 on i386 for arm, I encountered the

^ Ruby from ASP
35814 [ralph.mason ] This is a multi-part message in MIME format.
35815 [james rubyxm] Try ActiveScriptRuby

^ Why is 1.6.7 the latest stable?
35819 [stesch no-sp] I thought 1.6.7 has a serious bug with class variables?
+ 35820 [matz ruby-la] It's not that serious.  Plus all know prominent bugs are fixd in the
+ 35824 [m.rokos sh.c] But also "not serious"... :-)

^ [FXRuby] advanced questions and bug
35822 [leon ugcs.ca] One of my pet projects is using FXRuby extensively and I'm now
35837 [ljohnson res] I haven't downloaded the program yet, I'm just going to try to hit these
35855 [leon ugcs.ca] In 0.99.188 on Linux, I got an error saying that <=> wasn't defined for FXList

^ Test::Unit
35825 [Steve.Merric] Can someone explain to me the *details* of how to use Test::Unit,

^ Newbie  -- Please help!!
35826 [mark.firesto] First an apology for asking what are probably lamer questions.
+ 35827 [kentda stud.] Turbo Pascal rocked... Back in the stone age, that is :-)
+ 35828 [decoux moulo] Why ? there is nothing wrong with global variable when you write a small
  35829 [decoux moulo] I've a bug in what I've send :-(, write it
  35834 [mark.firesto] Thanks for those guys!!  That has really helped me understand a little more!
  35835 [olonichev sc] so
  35839 [mark.firesto] Well.  That puts me back in the same boat I was before.  I have started to
  35841 [olonichev sc] Easier saing you must not have objects which can be modified at the same
  35883 [mark.firesto] Ok.  I understand WHAT you mean... I just don't know how to get there (nor

^ Problems with Ruby, apache, and cgi
35831 [mycorner6 ya] I'm a systems admin for a webhosting company.  We had a request for
+ 35832 [tobiasreif p] perhaps
+ 35853 [tom.hurst cl] mod_ruby provides some extra interfaces to apache, but I don't think
| + 35904 [ngawor lucen] Great! I had needed this answer yesterday, but discovered today,
| + 35916 [mycorner6 ya] Thanks for the comments.  Removing the $stdin.gets did the trick,
|   35925 [tobiasreif p] If you want to advertise your Ruby featuring webhosting company, check
+ 35932 [bob_sideboth] Fastcgi is an excellent alternative to mod_*. It's considerably

^ String.dump and back
35840 [petr.fischer] (probably stupid question) - how I get back the same string as before String.dump.
35844 [matz ruby-la] eval?

^ Debugger problem
35845 [nathaniel ta] def a
35864 [nahi keynaut] trace_func is removed at at_exit.  Workaround;
+ 35873 [nobu.nokada ] Shorter, this may also work?
| 35882 [nakahiro sar] klass, *rest|
| 35886 [nobu.nokada ] Oops,
| 36085 [nahi keynaut] I didn't know Method#to_proc.  What a waste!
+ 35933 [nathaniel ta] *rest|

^ Re: Some potential RCRs  [Comments Summarized]
35846 [bobalex attb] Following are summarized responses to the passle of RCR candidates I posted
+ 35847 [matt lickey.] I assume this is different from puts in that it always prints an extra
| 35848 [Dave Pragmat] And also that
| 35850 [paul atdesk.] def println(*x); x.each { |y| print y }; print "\n"; end
| 35856 [balexander r] I'm not quite sure what point you were making in your post -- if it's a
+ 35854 [sean celsoft] argument.to_s)
  + 35857 [balexander r] "abort" is exactly equivalent to "exit(1)", as far as I can tell. An optional
  | 35859 [sean celsoft] optional
  + 35870 [bobalex attb] That's pretty much how it works already. "%s" means "apply to_s", "%d" and
  | 35875 [sean celsoft] each
  + 36493 [dblack candl] Just FYI: this is somewhat similar (though not identical) to a
    36507 [nobu.nokada ] In 1.7, now Dir.glob takes second argument modifies only the

^ Win32 build error in latest CVS?
35851 [nathaniel ta] I just checked out the latest from CVS and got these errors when

^ Ruby 1.7: "long long" not valid for Visual C++ 6.0
35852 [ljohnson res] Just did a CVS update of the code and compilation with Microsoft Visual C++

^ RCR: unify cgi handling of normal and multipart forms
35858 [patrick-may ] rubygarden link => http://www.rubygarden.org/article.php?sid=186

^ RE: Test::Unit [long]
35861 [nathaniel ta] Hmmm... big question. Let me see if I can give you a little piece of the
+ 35877 [tobiasreif p] ... a great howto!
+ 35889 [ronjeffries ] Well done, Nathaniel! That needs to be an article somewhere!
  35893 [pate eylerfa] I agree.  I've been struggling with how to use test::unit (and rubyUnit

^ new user confused by multidimensional Ruby hashes
35866 [ian google.c] [ianmacd@caliban ~]$ irb
35867 [tom.hurst cl] *blink*
35884 [tom.hurst cl] Er, <cue explanation *not* written at 3am with a rat trying to tear off
35918 [ian google.c] Thank you very much. Your postings made things much clearer.

^ RE: Ruby version 1.6.6
35869 [w3gat bellso] I just wanted to thank everyone concerned with this version.  Finally all

^ Thank you...
35871 [patrick-may ] I feel possessed by a desire....

^ cygwin port system() problem
35872 [nick sarion.] When executing a windows command containig output redirection via ruby's
35874 [nobu.nokada ] system() under cygwin uses /bin/sh.exe, so you need to include

^ CGI.new & cookies error... please help...
35876 [evan_leonard] I'm experiencing an odd error with CGI and sessions.
35890 [    s xss.de] Just how do you call CGI.new on the other page?
36126 [evan_leonard] It seems that it was just an error with the old version of ruby I had

^ language addition: automatic 'new'
35879 [vardhan.v pa] and it occured to me, it would be great if
+ 35881 [matz ruby-la] Hmm.. you are lazy enough to choose C(args) over  C.new(args).
+ 35899 [paul atdesk.] I'm not sure if this is a good idea, but it sounded like a challenge, so
  35905 [nobu.nokada ] It's too expensive to scan all objects, and a nested class name

^ Re: estunit - setup -> set_up ?
35891 [hipster xs4a] Extract the testunit archive and copy directory
35934 [nathaniel ta] (Or,

^ XmlParser: New lines in attribute values getting lost
35894 [dcabo hotmai] I'm currently trying to parse some Xml files using Ruby, but I cannot get
35896 [Vladimir.Doz] ehlo.

^ Simple Ruby mail server source?
35897 [kentda stud.] I was wondering if there are any simple mail servers out there,

^ camelCase and underscore_style
35898 [chris.morris] First, a question. If underscore_style is the Ruby norm for methods and the
+ 35900 [andy toolshe] Not really (at least, not that I rememeber), but I find it an issue
| + 35901 [chris.morris] What a loss to the world of ASCII art that would be ...
| | 35902 [andy toolshe] Nah, as long as we still have \ and / I'm okay :-)
| + 35915 [ronjeffries ] That's good enough for me. Underbars are gone from my code asap. Now I need a
|   35971 [W.L.Kleb LaR] ruby -pi -e "gsub(/underscore_style/, 'camelCase') *.rb
|   + 35972 [W.L.Kleb LaR] Sorry, I always forget the closing " ... it should be
|   + 35974 [tobiasreif p] does that work?
|     36544 [w.l.kleb lar] Yes. I got it from Ruby's "man" page and the "Ruby and Its World"
|     36550 [tobiasreif p] I meant, for the line to work, there must be a real regex, no?
|     36564 [w.l.kleb lar] Right. For example, if I wanted to change all occurances of
|     36569 [tobiasreif p] So if I wanted to change all underscore to camelCase, the line could
|     36570 [horst hduche] "Tobias Reif" <tobiasreif@pinkjuice.com> schrieb im Newsbeitrag
+ 35913 [matt lickey.] I personally find underscore_style easier to read, especially with
+ 35922 [ptkwt shell1] Personally, I tend to use camelCase for variable and class names and
  35924 [gnhurst hurs] Actually, there was a discussion about this last year somewhere and
  + 35930 [nathaniel ta] I agree completely with those two (classes and methods)... the hard ones
  | + 35935 [vjoel PATH.B] I use underscore for variables because of prefixing and other symbolic
  | + 35938 [in6x059 publ] Last fall there was a discussion on Ruby Garden, spawned by a proposed
  |   + 35940 [nathaniel ta] Marko Schulz [mailto:in6x059@public.uni-hamburg.de]
  |   | 35965 [ronjeffries ] This was kind of where I was coming from with my question. I much prefer
  |   | + 35966 [dblack candl] I kind of had the same reaction to what MatzHimself said about using
  |   | | + 35967 [tom.hurst cl] Well, for those who like weirdMethodNames, you can always add something
  |   | | + 35990 [ronjeffries ] Well, I asked in hope that there was a "standard". I certainly think there is a
  |   | + 36095 [Dave Pragmat] But of course, Andy is obviously totally mistaken in this preference
  |   |   36122 [akolarik son] This message is in MIME format. Since your mail reader does not understand
  |   + 35959 [matz ruby-la] Most of Ruby standard libraries using non camel case for names except
  |     35963 [in6x059 publ] I only meant, that make this statement so absolute seemed wrong to
  + 35941 [matz ruby-la] I use CamelCase for class and modules names, ALL CAPS for other
    35978 [jeremy chaos] Am I the *only* person who uses Underscored_Camels?
    35984 [kentda stud.] I do use it in testing, i.e. MyClass_UnitTest and similar. Not for much
    36022 [jeremy chaos] That's a good tip that I'll bear in mind.  But I wouldn't apply it to