31110-41567 subjects 31568-34560

Tk and cygwin
31382 [Jean-Francoi] I develop an application under linux that works very well, but I would
31857 [nagai@ai ky ] Please try TkRoot/TkToplevel#protocol method (See Wm module).

Re: Swig Ruby documentation mods, and mkmf module
31391 [hgs@dm .a .u] OK, will do.
31394 [ballabio@ma ] "Use the source, Luke" :)
31397 [hgs@dm .a .u] Yes, I have done, but "I'm not sure how to" means: I don't have enough

attr_{reader|writer|accessor} for class methods
31401 [stephen.hill] how do I implement an atribute reader/writer/accessor for a class variable?
+ 31403 [paul@at es .] The reason this doesn't work is because :Foo.bar is the same as
| 31425 [chr_news@gm ] ml
| 31447 [paul@at es .] I was about to reply and say that it looked fine to me, but then I
| 31459 [chr_news@gm ] Sorry I always have problems with things like left/right, +/- etc.
+ 31406 [Dave@Pr gm t] class Dave

How to alias a Module method?
31407 [jfreeze@fr e] module MyModule
+ 31408 [jfreeze@fr e] Oops!
+ 31409 [decoux@mo lo] Same response than for attr_*
+ 31410 [dblack@ca dl] module MyModule
+ 31411 [Dave@Pr gm t] Heh - two answers in a row with the same technique!

irb ruby-script w/o exiting irb?
31423 [chris.morris] If I run irb and pass it a ruby script, how can I stay in irb after the
31429 [chris.morris] Nevermind. -r wasn't working for me, and I made a false assumption about its

A few words on threads
31424 [avdi@ya oo c] generalizations to follow.
+ 31432 [matz@ru y- a] For scripting language community, it's probably because implementing
| 31645 [comp.lang.ru] - - There are real problems with threads and unix in general.
| 31654 [cjack@cj ck ] On 1/18/02 9:18 AM, ";"
+ 31450 [gehlker@fa t] A good read, nonetheless.
+ 31454 [paul@pr sc d] I agree that Python's implementation is crude in an absolute sense (but
| + 31468 [cjack@cj ck ] It's a problem on single-processor machines too. If thread A wants resource
| | 31480 [paul@pr sc d] That's true in any threading model.
| + 31513 [ptkwt@sh ll ] Yes, there have been 2 or 3 different threading schemes in Perl.  About a
| + 31659 [tom.hurst@cl] That's correct; there's an extension which provides fork and some signal
+ 31483 [pehrens@no p] Not arcane, dangerous for programmers writing robust,
+ 31503 [martin@v. oe] One issue is that an interpreter has to make difficult
+ 31780 [matt@se ge n] This paragraph neglects one very important issue of fork(), certainly on

Re: any thoughts on creating jsp counterpart (rsp, ruby server pages) for dynamic web pages
31426 [ jimm@io co ] (http://www.ruby-lang.org/en/raa.html).
31515 [webmaster@ma] Yo!
31653 [bryan@te ra ] Well, then you might like to hear about my project!  It's a framework
31779 [matt@se ge n] Now this looks very interesting :-)

rubygarden temporarily hosed?
31430 [dempsejn@ge ] not to point out the obvious, but have others seen this?
31433 [Dave@Pr gm t] It should be back now - I had a name server issue.
31434 [dempsejn@ge ] Hey Dave,

Protecting a member hash
31435 [jfreeze@fr e] I have a class that stores data in a hash.
31438 [ptkwt@sh ll ] Why not just define a '[]' method, but not a '[]=' method in your class
+ 31443 [dempsejn@ge ] I don't understand how he was able to write to the hash with only an
| 31449 [jfreeze@fr e] The t.entity returns the hash. Once I have it in my possesion,
+ 31444 [jfreeze@fr e] Well. that would work, but the real story is that I have
  + 31445 [jfreeze@fr e] Well, after thinking about this a little longer and trying
  | + 31446 [chris.morris] attr_reader on the hash means code using the class could not replace the
  | + 31463 [alan@di ik t] Well you could declare []= private or protected in HashProtector.
  |   31467 [jfreeze@fr e] Not a bad idea, but I tend to shy away from such techniques since
  + 31452 [ptkwt@sh ll ] Jim,
    31457 [jfreeze@fr e] Hmmm..well, not being the XML expert, I am somewhat afraid

Re: A question on Ruby Threads(Ruby2.0 & event loops)
31437 [ptkwt@sh ll ] Will Ruby 2.0 have event loops?
+ 31498 [matz@ru y- a] I guess not in the beginning.
| 31501 [chris@da kr ] For ferite (see the signiture) I will be working on threading at somepoint
| 31521 [matz@ru y- a] I'm not sure how much we share the requirement, but if it's possible
| 31531 [chris@da kr ] Thread creation
| 31554 [matz@ru y- a] I guess it's easy to agree on this level.  The hard part is what you
+ 31821 [matju@sy pa ] Ruby 1.6 has an event loop if you bother to install MetaRuby 0.7.

[RCR] #59 Add fsync method to IO class
31442 [hensleyl@pa ] Adding fsync and fdatasync methods to the IO class will allow Ruby to
31520 [matz@ru y- a] * how common are they?
+ 31523 [slumos@sl mo] I think fsync() is old POSIX and should be in every *nix.  fdatasync()
| + 31524 [matz@ru y- a] Thank you for useful information.  Does anybody know if they're
| | + 31549 [hensleyl@pa ] Windows has the function FlushFileBuffers
| | | 31556 [matz@ru y- a] Thank you for information.  Now I'm pretty positive about this RCR.
| | | 31599 [nobu.nokada@] VC runtime seems to use PBFlushFileSync() and PBFlushVolSync()
| | + 31567 [gehlker@fa t] Macs have fsync but not fdatasync
| + 31552 [hensleyl@pa ] I was following Linux's lead on this.  But now that you mention it,
+ 31647 [eric@ba tu .] Extremely common in any database system.
  31649 [matz@ru y- a] OK, fine.
  + 31662 [matt@li ke .] I vote 4 (2 is my second choice).  Why?
  | 31870 [tony@pe fo c] I agree.
  + 31665 [slumos@ya oo] I haven't used either of them, so my opinion is based on reading the
  | 31666 [cbroult@sa i] ...
  + 31681 [chris@da kr ] Personally I would go for 2 - that way you can write it to use them, and have
  + 31687 [mawatson@ag ] I probably missing something, but wouldn't closing and reopening the
    31696 [hensleyl@pa ] On Linux at least, closing a file does not guarantee that the data will

Re: A few words on threads (and introduction)
31453 [brettmcc@am ] I feel it is mostly because pthreads was the implementation widely

Jobs for Ruby Programmmers ?
31455 [ddevilliers@] I've read that Ruby is very popular in Japan (than Python/Perl) Any jobs

ruby docs
31458 [ahoward@fs .] are there any sources of docs other than 'the book' that are more
+ 31461 [brettmcc@am ] What is the definitive ruby book?  The oreilly book or the pragmatic
| + 31473 [harryo@zi wo] I would have said the O'Relly book that hasn't been released yet :-).
| + 31477 [Eirikur@Mi r] The O'Reilly "Ruby in a Nutshell" book is a typical O'Reilly nutshell guide.
+ 31476 [niklas@ka i.] Have you looked at the net/http.rb source file itself? There is some
+ 31485 [ cyo@gm .n t] Ara,
  32089 [ahoward@fs .] yes. that is what i missed!  thanks alot, problem solved in two lines ;)

[ANN] RDoc Alpha-8
31460 [Dave@Pr gm t] RDoc is a Javadoc-alike documentation tool for Ruby. See
31466 [jfreeze@fr e] Thanks for the great tool.
31484 [Dave@Pr gm t] I'm hoping that over time people will contribute both different .css

31462 [DHable@ph in] Thanks,
+ 31469 [alwagner@ua ] The only thing I have seen is rbwrap.  It is not maintained and I
| 31474 [brettmcc@am ] There's a ruby2c in RAA, but I doubt it works under windows.  Speaking
+ 31471 [ptkwt@sh ll ] I feel your pain ;-)

[ANN] rubycookbook.org opens new section for code
31478 [colin@co in ] Following discussion with Jack Dempsey, I added a new section on the

some problems with popen3
31479 [paul@at es .] 1) I need to know the pid of the child process, so I can kill it later.
31497 [matz@ru y- a] "open3" currently uses "double fork" technique, which does not tell

Re: ruby docs (and weird error)
31482 [brettmcc@am ] I've already got the pragmatic programmers guide, which is a great
31496 [matz@ru y- a] Because there's "no virtual class for Fixnums". ;-)
31502 [brettmcc@am ] Does that mean you can't create a virtual class for that type, and
31517 [matz@ru y- a] That means you cannot create a virtual class for some objects (Fixnum,

Theory and communities (was: Re: Small Methods - a ramble)
31488 [marick@vi ib] [Mostly off-topic to the list. Sorry.]
31504 [dblack@ca dl] I once saw -- literally saw -- part of a black-and-white film in color

Spotting code smells (was: Re: Small Methods - a ramble)
31489 [marick@vi ib] I am mainly talking about work habits.
+ 31495 [Dave@Pr gm t] No! Please! This is a tremendous thread.
+ 31500 [alwagner@tc ] In my criticism class, I personally was so disgusted with the pseudo-science
+ 31742 [dblack@ca dl] In principle I think it's all extremely relevant... but I've written a
  31743 [dblack@ca dl] 1200 words, that is.

RE: Windows Automation
31491 [david.naseby] require 'win32ole'

oh, and Avdi...
31493 [pehrens@no p] Your Apache is way bad misconfigured...

Weekly RCR Summary
31494 [RubyGarden@t] This is an automatically generated list of Ruby Change Requests.
+ 31510 [matt@li ke .] Nice!  I hope this list gets much shorter.
+ 31511 [matt@li ke .] Nice!  I hope this list gets much shorter.

FXRuby: How to print on Windows
31506 [david.naseby] I'm currently having a problem making Fox print. Has anyone got some sample
31540 [LJohnson@re ] As far as I know, FXDCPrint isn't finished (i.e. doesn't work yet) but you

31508 [ptkwt@sh ll ] Some of the descriptions of Ruby I've seen mention that Ruby has
31597 [chr_news@gm ] To make a long story short meta classes are responsible for the lookup
31614 [decoux@mo lo] Well, I know nothing in python but I think that metaclass are based on

Hello! Array sub classing?
31512 [markt@go ir ] ...
31514 [harryo@zi wo] Welcome aboard!
31516 [markt@go ir ] ...
31518 [dsafari@xt a] If you are reading in a string why not use String#split ? If I understand you
31519 [markt@go ir ] ...
31522 [harryo@zi wo] For starters, you could take a look at ...
+ 31620 [markt@go ir ] ...
| 31622 [dblack@ca dl] Can you post some of your code?  I'm not sure where it's going wrong,
| 31667 [markt@go ir ] ...
| 31671 [dsafari@xt a] Markt, I'm not sure this is what you want, but I'll take a stab at it.
| 31694 [markt@go ir ] ...
| + 31697 [dsafari@xt a] Markt, perhaps you want to go through some sort of checking afterwards with
| | 31705 [dblack@ca dl] Don't forget #compact :-)
| + 31706 [dblack@ca dl] I'm not sure how that would happen exactly.  If you've got tabs separating
+ 31828 [markt@go ir ] ...
  31840 [list@ch om t] Sorry, I must be missing something.  Does Array have unmodifiable
  31848 [markt@go ir ] ...
  + 31851 [matju@sy pa ] special values: nil, false and true.
  + 31855 [hal9000@hy e] That's laggin behind the release of the book,
  | 31856 [markt@go ir ] ...
  + 31867 [dblack@ca dl] But it's still a string :-)  (not an array with commas between its
    31928 [markt@go ir ] ...
    31936 [dblack@ca dl] I think we're talking at cross purposes -- I'm talking about splitting

Yet another ruby book listet at amazon.com
31525 [mjais@we .d ] I found another book: "Making use of Ruby", to be published in May 2002
+ 31526 [Erik.Bagfors] Looking at the list of authors and what they have written, I found out
| + 31527 [erik@ba fo s] Sorry about that mail..By misstake I hit ctrl-enter which is "send the
| + 31530 [feldt@ce ch ] Seems that none of the book in this series has been published yet; they
+ 31532 [tobiasreif@p] It's sufficient for amazon links to look like

[ANN] Log4r 1.0.0 - Logging library for Ruby update
31529 [leon@ug s. a] Log4r is an extremely flexible logging library for Ruby that was inspired

Possible bug in Mac version?
31533 [Dave@Pr gm t] I was sent the following by Philippe de Rochambeau, but not having a
+ 31534 [decoux@mo lo] This is not a problem of stack size ?
| + 31535 [Dave@Pr gm t] Most likely, although a SEGV seems somewhat harse :)
| | 31548 [matju@sy pa ] i remember a stack size of 1/2 kB was common on DOS, Windows 3, etc.
| + 31541 [chr_news@gm ] which is a bug in my opinion.
+ 31550 [gehlker@fa t] I'm seeing that this works fine though n = 51. At n = 52 it just hangs. Then
+ 31566 [gehlker@fa t] I'll try to work on this in my copious spare time. It seems like a good
| 31569 [huber@al m. ] ulimit [-SHacdflmnpstuv [limit]]
| + 31575 [gehlker@fa t] First place I looked. No ulimit.
| | 31578 [cjack@cj ck ] cpu time (seconds)         unlimited
| | 31583 [gehlker@fa t] Man zsh | grep ulimit
| | 31588 [harryo@zi wo] Are you running as the same user?  Ie, were you perhaps root the
| | 31601 [gehlker@fa t] Thanks for responding. I finally figured it out. I wasn't changing who I
| + 31579 [gehlker@fa t] Ignore previous post. It's 'man ulimit' returns nothing but 'man limit'
+ 31585 [gehlker@fa t] Now that I finally figured out, with lots of help here, that you *can* set
+ 31644 [ rtan@vt ed ] I think this is related to the problem I posted in

Strange change of behaviour
31536 [hutch@xa pl ] I'm using Ruby 1.7.2 (Dec 27 or so).
31539 [decoux@mo lo] Use #eql?  and #hash

[possibly a bit OT] RubyGarden -> org and com?
31538 [Stephan.Kaem] I just noticed the little popup when moving the mouse pointer over the head

Segfault for define_method
31542 [chr_news@gm ] Talking about bugs, on cygwin 1.7.2 ...
+ 31544 [decoux@mo lo] Try this patch
| + 31547 [decoux@mo lo] This one is better, otherwise it don't do a fixpos if the node is NULL
| + 31553 [chr_news@gm ] Thank you very much - working fine on my machine as a well
+ 31555 [matz@ru y- a] It's a bug (of course).  I will fix it soon.

delete complete content of a dir
31546 [tobiasreif@p] how could I delete the complete content of a directory containing filled
+ 31557 [matz@ru y- a] system("rm -rf dir")
| 31559 [chris.morris] As far as a Ruby solution, this just came up last week -- I posted a routine
| 31562 [dempsejn@ge ] Not sure if you've seen the past e-mails, but this exactly the type of thing
+ 31563 [mike@le to .] filled > dirs which are filled etc.?

Threading and the ACE Libraries
31551 [avdi@ya oo c] I have to say, all this discussion on threading has been

Problem: Installing Ruby for Windows !
31560 [ddevilliers9] I'm having problems installing Ruby for Windows
+ 31565 [andy@to ls e] Lennie,
| 31676 [ddevilliers@] Andrew Hunt,
| 31737 [mkelly2002NO] Have you tried the one from the site above?  This is the download
+ 31594 [mkelly2002NO] I had a similar problem using the one from the Practical Programming
| 31754 [ddevilliers@] Michael Kelly,
| 31806 [mkelly2002NO] Hmmmm, on looking mine installed the samples.  Oh well, the installer
+ 31738 [mkelly2002NO] Oh yeah, I forgot to mention, if you do get it going you'd

The first alternative RDoc template
31564 [Dave@Pr gm t] Rich Kilmer wins with a rather attractive RDoc template that moves the
+ 31606 [nahi@ke na t] Samples are at http://rrr.jin.gr.jp/doc/ now.
| + 31613 [rich@in oe h] Samples are also here: http://www.infoether.com/ruby/jabber4r/doc/index.html
| + 31616 [tobiasreif@p] The Kilmer layout put a smile on my face; It works better for me. Will
|   + 31621 [jfreeze@fr e] I liked it too, but if it is included, please fix the Netscape problem first.
|   | 31627 [niklas@ka i.] Another issue with the layout and browser twirks. In Konqueror (2.2.1),
|   + 31630 [Dave@Pr gm t] rdoc -T kilmer ...
+ 32191 [mdavis@se ai] I am unable to connect via cvs to get the rdoc new template.  Any
| 32193 [Dave@Pr gm t] What are you typing? What happens when you try?
| 32203 [mdavis@se ai] cvs -d:pserver:anonymous@cvs.rdoc.sourceforge.net:/cvsroot/rdoc login
| 32204 [Dave@Pr gm t] And you're pressing just ENTER when prompted for a password (ie there
| + 32206 [tobiasreif@p] Anyone has a tip for a simple way to get stuff from CVS repositories, on
| | + 32210 [vruz@ru y- s] you can get the standard CVS tool, compiled for Win32 and many
| | + 32213 [curt@hi bs c] Try WinCVS -- http://www.wincvs.org/
| + 32212 [mdavis@se ai] Works great! Thanks.  Sorry for not catching that.
+ 32240 [mdavis@se ai] Thank you for rdoc :)  It has inspired my to start documenting my code a
  32242 [Dave@Pr gm t] Hmm. that's a workaround that Rich added - apparently his browser
  + 32252 [rich@in oe h] Dave,
  | 32261 [Dave@Pr gm t] Sorry - my mistake, and will do.
  + 32265 [mdavis@se ai] Worked like a charm.  Thanks.
  | 32268 [Dave@Pr gm t] The latest CVS auto-generates that file if needed.
  + 32269 [dsafari@xt a] Sorry I can't remeber the name of the person that did it, but I think the