95489-96104 subjects 95901-96707

ruby 1.9 (cvs) installation problem...
95664 [charleshixsn] This is on a Debian Sid system.  FWIW, I ran rdoc against the source
95670 [charleshixsn] I solved this problem by copying over the Ruby1.8 ruby/rdoc folder to
+ 95672 [dave@pr gp o] Sounds like you may have an old RDoc installed which the new one is
+ 95675 [nobu.nokada@] Due to the change on ruby.h at 10 March, you have to recompile
  95702 [charleshixsn] I did recompile them.  I assume that the problem was caused by the

mmap question
95668 [djberg96@ho ] Ruby 1.8.1
+ 95692 [ahoward@fa t] i know the system call is safe across forking - but it seems like there might
+ 95714 [decoux@mo lo] Like said it has internal state, try this

Rdoc templates
95677 [mdavis@se as] What paths does rdoc check when looking for templates?
95681 [dave@pr gp o] It is now... :)
95699 [mdavis@se as] Excellent!  Thanks.

ruby-talk web interface broken?
95682 [ptkwt@ar cn ] I don't subscribe to ruby-talk, I usually read c.l.r, but sometimes I've

Ruby's module search path
95683 [harryo@qi so] Is there an easy way to discover the path to a "require"-ed module (assuming it's a .rb file)?
95686 [neoneye@ad l] You may want to look at rextra.. I found this
+ 95687 [assaph@av ya] def which(libname)
| 95688 [harryo@qi so] Excellent!  Works a treat.  Thanks.
| 95689 [assaph@av ya] but just
+ 95691 [harryo@qi so] From the output, I'm guessing $" is the list of requires that have been executed so far?
| 95698 [assaph@av ya] been executed > so far?
| 95741 [vadimn@re ha] $" is aliased to $LOADED_FEATURES, but the $: variable does not seem
| + 95742 [ksibilev@be ] $: == $LOAD_PATH
| | 95745 [vadimn@re ha] And also,  $-I == $LOAD_PATH
| + 95754 [ptkwt@ar cn ] $: is aliased to $LOAD_PATH
+ 95701 [lists@za a. ] might have been manipulated in between. Also the code above did not
  95704 [harryo@qi so] True.

Limits of Time
95684 [cpine@he lo ] We are making a website to handle all of our family pictures.  The earliest
95685 [wjl@ic ca er] charset="utf-8"

proposal - enforce lower_case for variables and methods
95693 [neoneye@ad l] Ruby does already enforce CamelCase for class/modules.
+ 95694 [jamesUNDERBA] #-- Begin example ---
| + 95697 [neoneye@ad l] initial letter must be upcase. Neither lowercase nor underscore.
| + 95726 [neoneye@ad l] One of the first language I programmed in was AMOS for the amiga500
|   + 95727 [michael_s_ca] Perhaps, but taking away freedom is not guaranteed to reduce
|   | 95732 [app1tam@up .] -> -----Original Message-----
|   | 95734 [walter@mw ew] You can have it both ways.  There is a module in RAA that will
|   + 95750 [jamesUNDERBA] I (sort of) understand the value of programming with training wheels,
|   | 95753 [neoneye@ad l] Do you have a url ?
|   | 95767 [jamesUNDERBA] Interesting perspective.   I would think that such explanations would
|   + 95764 [discord@ma .] behavior!
|   + 95766 [djd15@po cw ] In matters of religion, isn't it best to let people choose their own?
+ 95695 [drbrain@se m] class No_not_really
+ 95696 [harryo@qi so] Imagine a workplace where the dominant language is Java and hence people are used to using camelCase ... and it's probably specified in a coding standard, if one exists.
+ 95700 [cc1@ce .w st] I would probably prefer most library code to have similar conventions,
| 95707 [bob.news@gm ] "Charles Comstock" <cc1@cec.wustl.edu> schrieb im Newsbeitrag
| 95710 [spam@de k. k] ruby-lint is a much better idea. I'd put my vote on a ruby-lint in
| 95720 [neoneye@ad l] Is Ruby-Lint, a static analysis tool ?
+ 95706 [charleshixsn] I would consider this a very bad idea.
+ 95716 [timsuth@ih g] There are libraries which provide a bridge between Ruby and e.g. Java, .NET
| 95718 [spam@de k. k] there's also a .NET bridge, the .net libraries capitalize their method
| + 95719 [neoneye@ad l] Interesting idea. That could be useful. Unfortunatly as Anders points out,
| + 95721 [flgr@cc n. e] irb(main):001:0> def FooBar(arg = :none); arg; end
| + 95725 [batsman.geo@] => 1
|   95773 [me@ib tt .n ] this is a constructor. something which is named as the class it belongs to.
|   95781 [flgr@cc n. e] Wrong.
|   95786 [me@ib tt .n ] yep, you are right..
+ 95733 [dblack@wo bl] (posting through my Usenet provider, hoping the gateway will
+ 95747 [bg-rubytalk@] Well, it enforces a leading capital, CamelCase is just a convention.
| 95748 [jfh@ci e. fl] It seems a lot of people find Ruby so much closer to perfect than any other
| 95762 [dblack@wo bl] Or rush out to introduce non-traditional variable and/or method names
+ 95765 [dagbrown@LA ] Only for the initial letter, and the reason for that is that

Gateway problems again?
95737 [djberg96@ho ] It appears there's some sort of gateway issue again.  The ruby-talk

optparse and required options
95739 [ jimm@io co ] I'm just starting to use optparse. I want to force the user to supply an
+ 95743 [nobu.nokada@] optparse itself doesn't supply options semantics.
+ 95744 [ahoward@fa t] opt.on('-e', '--email=EMAIL', EMAIL address') { | e | email = e }
+ 95770 [gsinclair@so] def show_help(parser, code=0, io=STDOUT)

RDoc and the one file option
95746 [mdavis@se as] /usr/local/lib/ruby/1.8/rdoc/generators/html_generator.rb:1335:in `generate_xml': uninitialized constant RDoc::Page::ONE_PAGE (NameError)
95751 [dave@pr gp o] Hmm.. this has been fixed for a while now. Are you using the latest CVS
95812 [mdavis@se as] It works great with the latest version of Ruby.  Thanks:)  I was working with an older snapshot.
95815 [dave@pr gp o] Doesn't --template work? You'll need to define the ONE_PAGE template in

File.stat.size errors with file sizes over 2Gigs.
95769 [walter@mw ew] File.stat('file_name').size returns erroneous file sizes when the
+ 95772 [hal9000@hy e] I'd like to think it's a bug. Not sure.
| 95774 [walter@mw ew] unfortunately, I am looking at some files over 14 Gigs.
+ 95775 [discord@ma .] It looks like a bug (that is, it doesn't happen that way on my box).
| 95777 [walter@mw ew] box).
| 95783 [discord@ma .] It's a mac, so it's a *nix box.
| 95792 [walter@mw ew] I tried that but just got zero, so I tried
+ 95776 [Ephaeton@gm ] I'm not sure about windows, but on unix[1] the file size usually is an
  + 95794 [walter@mw ew] an
  | 95826 [NOSPAM@ke py] It's not a ruby issue. You need to compile it (and your system as
  | 95829 [nobu.nokada@] The main problem is that MSVCRT doesn't provide it.
  | 95832 [NOSPAM@ke py] I have no idea what MSVCRT is, but since it starts with MS, it must be
  | 95833 [Ephaeton@gm ] It smells like microsoft visual c runtime.
  | + 95845 [NOSPAM@ke py] I didn't know, thanks.
  | + 95995 [nobu.nokada@] Unfortunately, it is not all of large file support.  Of course,
  + 95846 [djberg96@ho ] If you're on a 64 bit platform you can build Ruby as a 64 bit app, and

Dynamically define method
95771 [rohitlodha@h] I need something like this
+ 95778 [vjoel@PA H. ] I think that's just a limitation of define_method--you can define a
+ 95779 [discord@ma .] It doesn't look like you can use #define_method, since blocks can't do

Quick Introduction to RubyGems
95782 [gsinclair@so] If any of you are not aware of the power and ease of use of RubyGems,
+ 95784 [neoneye@ad l] Nice work Gavin, provides a good insight how to use rubygems.
| 95787 [nick.faiz@ce] I'm looking forward to trying it out.
+ 95795 [paul@vu ma k] I'm running WinXP 1.8 ruby and get this error.
| 95796 [jim@we ri hh] Did you get the CVS version of RubyGems?  The latest CVS version has been
| + 95797 [paul@vu ma k] Okay...I did not get the cvs - have to find out how on windows
| | + 95803 [sdate@ev re ] HTH,
| | + 95809 [jim@we ri hh] I did not see the error you got when you tried
| | | + 95824 [ jimm@io co ] It looks like Paul is having a different problem than I am. BTW, I'm using Mac
| | | | 95834 [gsinclair@so] ~>> gem --remote-list
| | | | + 95840 [paul@vu ma k] %>gem --list
| | | | + 95847 [ jimm@io co ] The released version.
| | | + 95839 [paul@vu ma k] Here ya go with both..--remote-list and --remote --list, i think they
| | |   95843 [pit@ca it in] try "--remote-list" (no space between "--remote" and "-list").
| | |   95855 [paul@vu ma k] The last gives me an invalid option error.
| | + 96137 [charleshixsn] gem --search="[0-z]"
| + 95801 [ jimm@io co ] /usr/local/lib/ruby/site_ruby/1.9/rubygems/remote_installer.rb:78:in `find_latest_valid_package_in_caches': Could not find --list > 0.0.0 (RuntimeError)
+ 95857 [gavin@re in ] Fantastic! Thanks for this, Gavin S. This was exactly the sort of thing

RubyGem and Rake
95788 [ jimm@io co ] I'm switching over to using RubyGems. I like it!
+ 95789 [ jimm@io co ] Jim Menard, jimm@io.com, http://www.io.com/~jimm/
| 95791 [hal9000@hy e] My understanding is that gems currently are good for libs, not
| 95807 [jamesUNDERBA] Ask the user when he or she installs a gem?
| 95818 [rich@in oe h] We are working through these issues.  The gem developer will be able to
+ 95790 [discord@ma .] Or symlinks. Then, when uninstalling, look for symlinks pointing to
+ 95793 [jim@we ri hh] The GEMS team is debating how to handle applications at the moment.

two-pass tree traversal via callcc (was: Re: Create a directories list with sub-directories)
95800 [vadimn@re ha] I needed a variation of the above script that would print out, for

ruby-dev suumary 23175-23213
95802 [maki@ru yc l] This is a summary of ruby-dev ML in these days.

Ruby Install request
95805 [mdavis@se as] During the make install of Ruby 1.9, I noticed that all the RDoc documentation was built.  Could the building of the RDoc documentation be moved to make rather than make install?  This would make changing between versions of ruby a bit faster.  Generally I keep each version of Ruby on my hard drive, I cd to the version I want to run with and type make install.  Keeping make install as streamlined as possible would be great.

thread/fork/tk nightmares
95810 [Ara.T.Howard] i'm working on an tk ap that drives alot of complex code, some of which is not
95825 [NOSPAM@ke py] Nice idea IMHO.
95848 [ahoward@fa t] i also saw that the ruby-dev suumary 23175-23213 by Masayoshi Takahashi spoke
95879 [akr@m1 n. rg] Thread.new {
95881 [ahoward@fa t] i keen observation - with that change the script runs fine even using open3.

Newbie Q: Lowercase CGI output
95813 [sven_schott@] I know this might be simple but I haven't found a decent(simple Ruby
+ 95814 [neoneye@ad l] You will have to do downcase ala this (untested).. perhaps more downcasing
+ 95841 [me@ib tt .n ] yes, it is prefered. xhtml requires lower-case tags.

ftp/socket problem on OSX
95816 [dave@pr gp o] The following code
+ 95827 [ jimm@io co ] No ideas, but here is another data point. I ran the code on my Mac OS X
+ 95830 [peter@se an ] From my OSX box I get
+ 95916 [david@lo dt ] -000A958E6254%40loudthinking.com

Ruby,XUL and toolkit agnostic GUIs
95817 [ptkwt@ar cn ] Looking back in ruby-talk for 'Ruby' and 'XUL' shows a discussion from
95862 [ser@ge ma e-] Yeah, I've got some basic code working using FOX as a backend, but it
+ 95863 [curt@hi bs c] Very cool! Please consider wxRuby as a backend.
+ 95868 [ptkwt@ar cn ] I'm thinking that we would just use Ruby to generate Mozilla XUL.  We
  95943 [ser@ge ma e-] It doesn't do anything for me if I can't write my application in Ruby.

95820 [sabymus@re i] ...

About type inference for ruby
95821 [surrender_it] I'm going to put some memes toghether running away in few minutes, to
95877 [discord@ma .] a, b, c, d = ARGV

rubygem: progressbar is invalid
95822 [nvivek@nc t.] I'd gone over to the wiki on rubyforge and was experimenting with
+ 95878 [discord@ma .] % gem --verify progressbar
| 95886 [chad@ch df w] Thanks for the reports.  I've just returned from a very tiring trip and
+ 95956 [chad@ch df w] Now I'm back and rested, and I realize that I could have answered part

module and alias_method
95823 [raphael.baud] is it possible, in a module, to create an alias for a method of the object including it?
95872 [vjoel@PA H. ] module CKHTMLTableUser

rubygem goes too fuzzy
95828 [surrender_it] I believe I spot a misbehaviour in gem
+ 95836 [rich@in oe h] will work out the search algorithm...thanks!
+ 95880 [flgr@cc n. e] Moin!
  + 95885 [gsinclair@so] Good comments.  It has been discussed without objection on the
  + 95898 [surrender_it] mh.. mandrake's urpmi does some fuzzy match,
    95900 [discord@ma .] hmmm, I like that idea. If it could be installed the same way as the
    95952 [gsinclair@so] I'm not sure it's appropriate.  Gems are for code, not for

Detecting hardware with Ruby
95835 [jgb3@em il b] Here's a rather off-the-wall question: how would I go about detecting
95837 [neoneye@ad l] This should be possible with the DDC protocol
95842 [jason@jv eg ] I solved this problem by using init scripts.  I have two init scripts
95858 [jgb3@em il b] Ah, much nicer.  I've got a custom runlevel already for booting without
95897 [im_not_givin] You could always start X in -probeonly mode (or something), then parse

DBI, Oracle, output params and dbms_output
95844 [djberg96@ho ] Ruby 1.8.1
95989 [kubo@ji ba .] dbh.do("BEGIN DBMS_OUTPUT.ENABLE; END;")

[ANN] session-2.1.5
95852 [Ara.T.Howard] - added Session.use_spawn=, AbstractSession.use_spawn=, and an :use_session=>
95853 [rich@in oe h] This totally kicks butt!
95861 [ahoward@fa t] thanks.  it's been really useful thus far.

mod_ruby/eruby with IIS6?
95860 [NoSpamPlease] Is it possible to create a dynamic website with mod_ruby/eruby using IIS6
+ 95864 [usenet@an re] Not with mod_ruby, you would have to write a isapi dll. But I would try
| 95865 [NoSpamPlease] How do you configure FastCGI for Ruby with IIS?
| + 95867 [ksibilev@be ] Try this: http://www.caraveo.com/fastcgi/
| + 95869 [rasputnik@he] I don't use IIS, but the general way is to have a plugin for the webserver
+ 95892 [hatespyware@] I like to use the ActiveRuby distribution
  95925 [NoSpamPlease] It works! Is there any good example using Ruby to manipulate some MS Office
  95940 [hatespyware@] Sure - that is just win32ole.  OLE automation works the same whether
  96036 [rubytuzdayz@] Is there any good Ruby example that manipulate any Office Docoments,

Catching DRb exceptions
95871 [sdmitry@lr .] I'm wondering is there a way to catch exceptions raised in the DRb server
95903 [guillaume.do] You can not catch a exception raised by the server in your client it's a
95942 [sdmitry@lr .] Yes, you're right.

Errors with "make" step, while compiling ruby under AIX 5.2
95875 [ruby4lover@y] ...
95883 [nobu.nokada@] Seems to be used by assert() macro.  This patch will disable it.
+ 95888 [ruby4lover@y] ...
| 95895 [nobu.nokada@] You miss to apply the patch.
+ 95918 [ruby4lover@y] ...
  95957 [nobu.nokada@] The patch is in the format called "unified diff".
  96043 [ruby4lover@y] ...
  96047 [nobu.nokada@] Seems your mail client issue.  If you use a web mailer, save
  96057 [ruby4lover@y] ...
  96083 [nobu.nokada@] Since the patch is to change compile flags, so you have to
  + 96140 [ruby4lover@y] ...
  | 96144 [nobu.nokada@] Sorry, the patch has no effect until Makefile is re-made, you
  | + 96146 [ruby4lover@y] ...
  | + 96152 [ruby4lover@y] ...
  + 96142 [ruby4lover@y] ...

Re: tk missing (was Re: )
95876 [discord@ma .] *that* looks awfully familiar... :) It's complaining that you don't

Getting up arrow to work in irb on OSX
95884 [ptkwt@ar cn ] How do you get the up arrow to give prior commands in irb on OSX?
+ 95889 [ahoward@fa t] you need to compile with readline.  probably readline is not installed or was
| 95899 [surrender_it] support for up-arrow even without readline (but no completion). It
| 95934 [ptkwt@ar cn ] Generally it does work.  On my Linux box it works fine.
+ 95890 [sroberts@un ] There's a way you can do without recompiling (for 10.2, anyhow).
+ 95939 [sera@fh an .] ...

Ruby/DL and C++ libraries
95887 [google@re he] I'm trying to use a C++ shared library from Ruby and wanted to use
+ 95894 [tietew-ml-ru] [Ruby/DL and C++ libraries]
| 95936 [ptkwt@ar cn ] Swig is great when you've got the source code for the library available.
+ 96238 [pbrannan@at ] - C++ function names are mangled, but there is no standard for how

SOAP question
95891 [john@di io .] I ran this code[1] and got these results[2]. As you can see, the resuting
95923 [nahi@ke na t] CVS version of soap4r can handle SOAPHeader better.

mod_ruby/DBD:Mysql segfaults with php loaded
95893 [spectra@de i] I've just got a really weird segfault. It happened when accessing Mysql

non blocking read and thread
95896 [yannick@da z] I noticed yesterday then any non blocking socket return in blocking mode
95906 [decoux@mo lo] client.close
95910 [yannick@da z] Thanks you for your answer, but that's not really what I meant. I know
95911 [decoux@mo lo] Re-read your script, you have 2 threads : the main thread waiting that
95921 [yannick@da z] You are right, after further tests; it looks like a thread scheduling
95929 [akr@m1 n. rg] I recommend sysread instead of IO#read with nonblocking mode.
+ 95932 [ahoward@fa t] can you please elaborate on this?  why exactly?
| 95953 [akr@m1 n. rg] Because nonblocking sucks.
+ 95946 [yannick@da z] Thank you for the tip, using sysread does fix this problem. But after
  95951 [akr@m1 n. rg] The problem is not Kernel#select but IO#eof?.