4054-4340 subjects 4501-4833

Re: Content of the Perl6 talk
4268 [neilh@sc nt ] A common runtime could at least produce a common library interface.
4270 [schneiker@ju] threading
4294 [schneik@au t] Conrad Schneiker
4295 [schneik@au t] While Guido pointed out some problems with a common runtime in
+ 4296 [schneik@au t] Conrad Schneiker
+ 4297 [claird@st rb] Yes.  It's happening.
  + 4312 [claird@st rb] Timbot points out in <URL:http://deja.com/=dnc/getdoc.xp?AN=652003755>,
  | 4315 [schneik@au t] People seem to be unaware of the original context of my post. Following my informal
  + 4320 [schneik@au t] Those of you who (1) hope/expect Ruby to become as popular as Perl may
    4321 [Dave@Th ma e] I wasn't keeping track before, but this one got through OK.

Crosswinds.net.  is the worst  8711
4271 [nlzvnv@ya oo] Hey everyone.  Public service announcement here.  LOL.  I wanted to spare you some heartache.  If you have a website at all DO NOT use Crosswinds.net.  I had a website there for a month and spent hours everyday working on it and have been getting a thousand hits a day and crosswinds.net just up and DELETED IT without even so much as a warning.  now all my hard work is gone.  I contacted them because I thought maybe they made a mistake and they informed me Oh well.  All because i had a song on my site.  I want everyone to know before anyone ever has to go thru this.  Thanks for listening.

OT: New Ruby Working Draft Published
4272 [schneiker@ju] Yes, this really is off topic, and yes, the title is accurate.

RAA-ANNounce: Ruby/Tk GUI builder -- specRuby
4273 [schneik@au t] The 0.5 version of "Ruby/Tk GUI builder -- specRuby" is available on RAA

[[:alnum:]] etc
4278 [Dave@th ma e] A quick question following a recent post on ruby-list: should we be
4281 [schneik@au t] IMHO, Yes! The more often you have to go scrounging for information
4282 [Dave@th ma e] I'm just wondering if there's a good reason _why_ they're not
4285 [matz@ne la .] It was because I've left it unfixed till I18N discussion.  For example,

Set manipulation module
4284 [hfulton@au t] ...

4286 [feldt@ce ch ] I'm debugging an application where the bug seems to be when objects are to
4287 [matz@ne la .] You have to invoke call_finalizer(obj) to arrange finalizer to be
4290 [feldt@ce ch ] IMHO, the example in the Lib reference on ObjectSpace.call_finalizer is
4291 [Dave@th ma e] Actually, that version is _totally_ misleading - it is 100%

ObjectSpace.add_finalizer more...
4288 [feldt@ce ch ] I didn't read the docs very well. I understand that I have to mark
+ 4289 [Dave@th ma e] The online reference is for Ruby 1.6, which has changed. The methods
+ 4298 [matz@ne la .] require 'final'      # built in after 1.6
  4306 [feldt@ce ch ] Yes, now I see it.

Small gotcha
4292 [aleksi.nieme] This took some time to find out while the logic behind is quite simple.

Ruby for Linux
4293 [DDouthitt@cu] It appears that the next version of Linux Mandrake will have Ruby in it; they already have ruby-1.4.5 (binaries, docs, and extensions) ready to go.  There is also mod_ruby (for Apache).

basic ops in some languages
4299 [faatdilac@my] $a = ($b = 2) * 3;
+ 4301 [hfulton@au t] That should work in C and C++.
| 4313 [c.hintze@gm ] It does not work in Python, as there assignment is a statement not an
+ 4303 [matz@ne la .] Because unlike other languages, assignment is not an expression in
+ 4304 [wscott@ic ip] In Python 'var = expr' is a statement.  It is NOT an expression and
+ 4309 [dcalvelo@ph ] In Python '=' is an assignment *statement*, not an operator. The fact that

Mozilla and Tk; Komodo and Ruby|Tcl
4300 [schneik@au t] Does anyone have a reasonable idea of (or actual experience with) how

Ruby .NET
4302 [neilh@sc nt ] If anyone is interested in working on a version of Ruby for Microsoft's
4305 [schneik@au t] Thanks for the pointer.

isnan, nan? and ruby versions
4307 [feldt@ce ch ] I was about to post a question on other ways (compared to: f.to_s ==
+ 4308 [Dave@th ma e] The functions are in the current 1.5.5 and seem to work fine. The 1.5
+ 4310 [gnhurst@hu s] I like this idea.
  4311 [Dave@th ma e] I just realized you meant the entire thing, not just each entry. We

REBOL books
4314 [faatdilac@my] While waiting anxiously for any Ruby book in english to appear, I

Re: GUI creation tool for windows NT??
4317 [schneik@au t] Conrad Schneiker

4318 [Dave@th ma e] Is anyone here planing on attending OOPSLA?

Thirty-seven Reasons [Hal Fulton] Love[s] Ruby (long)
4322 [schneik@au t] The item that follows below is on
4324 [hfulton@au t] Greetings all,

Re: Phython and Microsoft's IL -- C# & .Net
4323 [schneiker@ju] I have some further follow-up comments.

Re: Thirty-seven Reasons [Hal Fulton] Love[s] Ruby (long) [CORRECTED  REPOST]
4325 [schneik@au t] My apologies to Hal and those who got the original version of this by

RUBY NEWSGROUP FAQ -- Welcome to comp.lang.ruby!  (Revised 2000-08-07)
4326 [schneiker@ju] RUBY NEWSGROUP FAQ -- Welcome to comp.lang.ruby!  (Revised 2000-08-07)

DRY and embedded docs.
4341 [hgs@dm .a .u] print <<"END"
4344 [Dave@th ma e] And a darn fine rule it is too.. ;-)
+ 4345 [hgs@dm .a .u] [Example of reading DATA trimmed]
+ 4351 [toshirok@yb ] What does it mean?
  4353 [Dave@th ma e] I think the latter would be nice. Perhaps it should look for an entry
  4354 [toshirok@yb ] Printing Usage message is one of user interface, so a program
  + 4355 [Dave@th ma e] That's a good point. I was also think we could probably benefit from
  + 4360 [hfulton@au t] Your premises are correct, but I disagree with your
    4370 [schneik@au t] I'm not familiar with all the formatting features, but is included files

Emacs font-lock colors
4342 [mulperi@ik .] How do I enable font-lock colors for my ruby files. Currently only
4343 [mulperi@ik .] I just found, that ruby-mode was evaluated before font-lock was loaded

DATA and rewind.
4346 [hgs@dm .a .u] The DATA file refers to information after __END__ in a ruby program.
4347 [decoux@mo lo] pigeon% cat b.rb
4348 [hgs@dm .a .u] OK, principle of "least surprise for Perlists" is a good idea. I have not
4356 [schneik@au t] Well, to reassure the rabid Perl-haters lurking in the shadows, even I
4361 [hfulton@au t] I'll agree with that. It's good to copy useful language features,
+ 4363 [hfulton@au t] bumos == bumps
| 4364 [Dave@th ma e] Ah! I thought it was the plural of bummer.
+ 4369 [hgs@dm .a .u] This is basically what I had in mind.  I thought I had not put it quite
  4373 [schneik@au t] On the topic of how Ruby does DATA.rewind,

Self printing programs.
4349 [hgs@dm .a .u] #!/usr/local/bin/ruby

Re: Thirty-seven Reasons [Hal Fulton] Love[s] Ruby
4350 [DDouthitt@cu] I like these 37 reasons!  Anyway, I agreed with almost (almost) all the reasons.....
+ 4352 [hgs@dm .a .u] This one jarred me as well.  tclsh has the property that undefined
| 4358 [hfulton@au t] That is an excellent point.
+ 4357 [schneik@au t] Well, it's certainly nice for search purposes, general user
| 4362 [hfulton@au t] Matz commented on this once; I won't speak for him,
| + 4365 [aleksi.nieme] Well, I have to say I'm standing with Hal here. What isn't restricted does
| + 4368 [schneik@au t] Hmm. I'm not sure even Perl would stoop to dependency on what would look
+ 4359 [hfulton@au t] #20 - Probably a matter of taste. I do like being able to grasp a little

Looking for input on a 'links' page
4366 [Dave@th ma e] I'm putting together a 'Links' page on RubyCentral, and I'd appreciate
4371 [aleksi.nieme] I'd like to note there are loads of significant code on the Japanese web

Anyone up to doing a Perl6 RFC to use Ruby's extension scheme?
4367 [schneik@au t] (1) Is there anything about Perl5 that would have precluded it from
4414 [schneik@au t] Maybe part 2 scared everyone off. Can anyone answer this part (or at
4415 [Dave@th ma e] I guess I was confused by the question. What part of Ruby's extension
4416 [schneik@au t] Do you mean internals insofar as basic datatype representations are
4417 [Dave@th ma e] It's pretty much everything - when you write a Ruby extension, you're

TC Reply Manifesto: "Quoting Strategies and the Jeopardy Game"
4372 [schneik@au t] Last year, Tom Christiansen used to post the following item on

Re: Scope surprise
4393 [decoux@mo lo] `while' is a keyword and it will not create a new scope (this is perhaps
4394 [gotoken@ma h] Yes, it little confuse people occasionally.  However, one has found a
4397 [matz@ze ab t] Yes. The internal scope by blocks is the biggest Ruby pitfall remains.

Re: RAA development ideas (was: RE:  Looking for inp ut on a 'links' page)
4395 [hgs@dm .a .u] Explanation of how to submit things is lacking on the page
+ 4404 [charleshixsn] ...
+ 4405 [Dave@th ma e] I think it would be a good idea if every submission had it's own
| 4408 [aleksi.nieme] Yup, agreed. This was actually what I had in mind as a detail enabling easy
| 4409 [aleksi.nieme] Yep, that's a good idea. I might have misunderstood you, but I regard this
+ 4410 [schneik@au t] Good point. How about E=Expanding or E=Extensive.
  4411 [aleksi.nieme] Peer review would be really good of course. I think, however, that it's more
  + 4412 [aleksi.nieme] This is the best candidate so far if we're going to change the name. I like
  + 4422 [hgs@dm .a .u] I suggested this because of you remarks about freshmeat.net filling up
    4444 [schneik@au t] Given that we seem to have a problem with *too-few* submissions of
    + 4445 [Dave@th ma e] We could do that reasonably objectively too, by setting certain
    + 4462 [hgs@dm .a .u] This meets my aim of providing quality code.

Re: New Require (was: RAA development ideas (was: RE:  Looking for inp ut on a 'links' page))
4396 [hgs@dm .a .u] [...new require operation...]
4400 [Dave@th ma e] A while back we had a thread where I suggested a kind of super-require
+ 4401 [hgs@dm .a .u] I remember that.
+ 4406 [schneik@au t] I'm not sure how general I am, generally speaking, but I didn't dislike

diags from eval.
4398 [hgs@dm .a .u] Can I get eval to be more specific when errors occur?
4399 [decoux@mo lo] Give it a name (and line number if you want)
4402 [hgs@dm .a .u] neelix hgs 98 %> ./!$
4403 [decoux@mo lo] Don't try to give the name "(eval)" :-)

Real world performance problems
4407 [aleksi.nieme] I thought to share my problem in case you might have major improvement ideas
4413 [Dave@th ma e] I changed your code slightly, and you're correct-the performance is
4424 [aleksi.nieme] I meant to say temporary objects.
+ 4425 [aleksi.nieme] I might add that maybe we would like to be heading to incremental GC world,
| 4435 [matz@ze ab t] Masato Kiyama at Hiroshima City University implemented generational GC
| + 4436 [aleksi.nieme] Have you changed company or something :?
| | + 4448 [matz@ze ab t] My company changed its name after the merger.  My position unchanged.
| | + 4449 [masato@cs s.] Sorry.
| + 4438 [hal9000@hy e] Very interesting.
|   + 4439 [aleksi.nieme] Me neither. So I'm not answering here, much at least.
|   | 4440 [Dave@th ma e] I've read about hierarchical garbage collection, where objects start
|   | 4441 [aleksi.nieme] Very interesting. I actually started to browse a little... :)
|   + 4447 [matz@ze ab t] I doubt it.  Switching GC is very hard.  I have no idea how to
+ 4429 [Dave@th ma e] But what you might try is to disable GC for each group of (say) 1000
  4430 [decoux@mo lo] GC.disable
  4431 [aleksi.nieme] Dave and Guy proposed the same thing with different words almost
  4432 [decoux@mo lo] perhaps I'm wrong but you need to enable the GC before a call to rb_gc()
  4433 [aleksi.nieme] Perhaps you're not wrong, and that's the reason for core dump. I was lead
  4434 [decoux@mo lo] void
  + 4437 [Dave@th ma e] Hmm. I wonder when this changed. I think the 1.4 behavior was also true
  + 4446 [decoux@mo lo] I don't know but I like the modifications in 1.5.*, for example the

Array#hash ruby rewrite
4418 [aleksi.nieme] I guess Array#hash does not work very well. Here's an example and Ruby
4419 [decoux@mo lo] There is a *big* difference between 1.4.5 and 1.5.x for your example.
4420 [aleksi.nieme] Great, good to know. I'll look it when I'll have some free time. Of course I
4421 [decoux@mo lo] Just to give you an example (with -pg), with 24001.times do end

Getting started with RubyUnit?
4423 [hgs@dm .a .u] Has anyone written any English docs on getting started with RubyUnit?
+ 4426 [aleksi.nieme] I'd say the distributed package should contain all necessary instructions
| 4427 [hgs@dm .a .u] It was not a criticism, people learn in different ways.  There is
+ 4428 [Dave@th ma e] Sorry - we don't have a chapter on it (umm... maybe we should have)
  4461 [hgs@dm .a .u] OK

Ruby GC
4442 [aleksi.nieme] I thought we might be talking about Ruby GC more, so I started new thread.
4443 [Dave@th ma e] I've seen no obvious evidence of any in the (admittedly limited) tests

Few random notes
4450 [aleksi.nieme] Here're few random notes I've written down for myself. Maybe some of them
+ 4451 [decoux@mo lo] pigeon% ruby -e 'a = {1 => 2}; a[1] = nil; p a'
| 4452 [decoux@mo lo] pigeon% cat b.rb
+ 4453 [Dave@th ma e] See http://dev.rubycentral.com/articles/re_obj.html
+ 4454 [Dave@th ma e] Firstly, I appreciate the 'on request' part of this--I personally
  4455 [aleksi.nieme] Great! They work probably differently on 1.4 series.
  4456 [Dave@th ma e] Bug! I'll fix it.
  4457 [aleksi.nieme] Thanks for nice article.
  + 4458 [decoux@mo lo] Unlike perl, ruby don't have the notion of "last successful match"
  + 4459 [Dave@th ma e] Well, the article _does_ say that $~ is thread-specific. Elsewhere in
    4460 [aleksi.nieme] Yup, my mistake.
    4463 [decoux@mo lo] pigeon% cat b.rb
    4464 [aleksi.nieme] So I just shut up :).

More RubyUnit questions.
4465 [hgs@dm .a .u] I am beginning to get a feel for this, but I still have a few more
+ 4466 [Dave@th ma e] You are right, it _is_ a singleton method, but of class
| 5996 [hgs@dm .a .u] class TestCase
| 6001 [Dave@th ma e] As far as I know, the only different is that methods defined within
+ 4505 [CQN02273@ni ] Sorry for being late to reply.
  4523 [hgs@dm .a .u] [trimmed.]

e-commerce solution in Ruby
4467 [jilanik@ti .] is there, somewhere on the web, an e-commerce solution, possibly free and open

ruby cgi questions
4468 [philip_mikal] Where can I find any English language documentation on
4469 [Dave@th ma e] Let me see if I can get permission to publish our chapter.

__name__ == "__main__" equivalent?
4470 [graham@cc .u] if __name__ == "__main__": main ()
+ 4471 [matz@ze ab t] if $0 == __FILE__
| 4483 [graham@cc .u] Thanks!
+ 4472 [Dave@Th ma e] if __FILE__ == $0

Ruby 1.4.6 repacked
4473 [matz@ru y- a] md5sum is
4474 [matz@ze ab t] <snip all>

RubyUnit.  Warnings to be expected?
4475 [hgs@dm .a .u] With ruby 1.4.4 (2000-04-14) [sparc-solaris2.7]
4478 [decoux@mo lo] You are probably running ruby with -w
4484 [hgs@dm .a .u] That is interesting.  My script has
+ 4485 [decoux@mo lo] Yes, like a P language.
+ 4486 [Dave@th ma e] t.rb:  #!/usr/bin/perl
| 4487 [hgs@dm .a .u] /bin/sh doesn't do this -- Hey! it never used to, but it does on
+ 4499 [decoux@mo lo] There is a problem with this, no ?
+ 4500 [decoux@mo lo] What I've said  is totally stupid,

Getting started
4476 [sp00fD@ya oo] So, I'd have to say that (so far) Ruby looks like exactly what I need.
4477 [Dave@th ma e] You might want to have a look at

this is cool!
4479 [sp00fD@ya oo] I just found this and I'm sure other languages must have this
4480 [Dave@th ma e] If you're into writing control structures, you could also have a look
4482 [jweirich@on ] I don't know if this counts as cool ... but here is an example using
+ 4488 [sp00fD@ya oo] So, continuations are kind of a pipe between methods?  I've never used
+ 4489 [sp00fD@ya oo] So, continuations are kind of a pipe between methods?  I've never used
  4490 [Dave@th ma e] Basically, continuations are like C's setjmp/longjmp on steroids.

Invoking an extension after compilation
4481 [Dave@th ma e] I want to write a C extension module which runs automatically after a
+ 4491 [matz@ze ab t] I'm not sure what did you want to do, but I guess
| + 4492 [Dave@th ma e] The problem is that at that point the rest of the program hasn't been
| + 4493 [decoux@mo lo] Have you tried with set_trace_func and intercept the first call to "line" ?
|   4494 [Dave@th ma e] Clever! I'll try it.
+ 4646 [matju@ca .o ] (general question)
  4654 [matz@ze ab t] I think it requires standard internal structure (like S-expression for
  4763 [matju@ca .o ] I think the goal for a 2.0 version could be this. what do you think?
  4765 [matz@ze ab t] Possible.  Among the bunch of goals of 2.0.  Stay tuned. ;-)
  4830 [matju@ca .o ] What other goals would you see for 2.0 ? I bet you have some place in
  4889 [matz@ze ab t] Check ToDo file in the distribution.  Everything prefixed by star (*)

RubyUnit -- Thanks.
4495 [hgs@dm .a .u] The idea of Unit Testing is that you should write the tests first.  Well,

4496 [rubikitch@ru] I am interested in Rubicon.
4497 [Dave@th ma e] We haven't yet packaged it up - we'll be releasing it, probably via
4498 [rubikitch@ru] I'm looking forward to releasing it!!