## 8611-9708

8388-9042 subjects 8778-9307

Re: k.Net, GTK.Net, and FOX.Net?
8611 [matt@ms .u s] The answer from Microsoft is ... yes, this is possible.  I don't know
8616 [ben_tilly@ho] Not impossible.  ActiveState's reputation in the Perl

Yet another SWIG question
8617 [Dave@Pr gm t] I'm running in to problems converting constant structures into Ruby
8743 [fukusima@go ] This is SWIG bug again :-(

Possible Ruby Suggestions?
8619 [mitch@ve ux ] I can't say enough good things about Ruby. I bought the book and read it
8624 [feldt@ce ch ] It is of course depends on what level of "security" you want from such an
8635 [mitch@ve ux ] True, very true.. Even making it inconvienent would be better than

mingw
8625 [dennisn@pe n] Can anyone point me to the source for the i386-mingw32 port of Ruby?
+ 8628 [feldt@ce ch ] There is one tarball for minw32 at http://www.os.rim.or.jp/~eban/ but
+ 8629 [zarathustra@] ...

fun with voicemodems, dtmf, soundcards & ruby
8627 [trk@hf .e st] I'm a non-programmer hoping to learn ruby by "scratching an itch"
8667 [sent@qu li y] I'm not sure how far we can go without getting

Interesting Language performance comparisons - Ruby, OCAML etc
8633 [g24ever@ho m] Interesting comparisons between languages for performance.
8650 [jmichel@sc u] The  individual memory/cpu  numbers are  quite interesting.  Contrary to
+ 8651 [g24ever@ho m] Good observations.
| + 8653 [Dave@Pr gm t] But by that logic, C wouldn't have replaced assembler. Java wouldn't
| + 8672 [decoux@mo lo] Well, in the ToDo there is
|   8684 [Dave@Pr gm t] Mais oui! As I said, performance is important, and we should continue
|   8685 [elderburn@mi] Actually, I believe a majority of the languages (and methodologies) considered
|   8687 [ben_tilly@ho] I hope not.
+ 8652 [Dave@Pr gm t] And the interesting thing about that is that Ruby was added to the
| 8670 [jstern@fo ha] The tests in this shootout seem to be not quite serious.
| + 8683 [Dave@Pr gm t] There's more to it than that. Languages such as Ruby that
| + 8696 [matz@ze ab t] Why don't you try and see by yourself?
| + 8710 [steve@de f. ] Oooh, a bloom filter. :)
+ 8663 [ptkwt@us r2 ] Actually, in the overall scoreboard on that page, Ruby is shown to be
8764 [charleshixsn] Well, OCaML is Object CaML where CaML is a relative of ML, and that's
+ 8768 [jstern@fo ha] Ocaml is strongly typed, and the creators put a lot of thought and effort

Any way to get enumerated values?
8642 [ljohnson@re ] I checked the FAQ but didn't see this one addressed. Is there any automatic
+ 8646 [Dave@Pr gm t] Ah, the old "names have been changed to protect the constants"
| 8655 [ljohnson@re ] Ahhh, I didn't specify enough constraints on the problem ;) And those would
| 8659 [c.hintze@gm ] ...
| + 8668 [sent@qu li y] I haven't looked at enumerate.rb yet, but my def
| | 8669 [feldt@ce ch ] If it's the only thread running... ;-)
| + 9013 [harryo@zi wo] I like this.  However, it got me wondering about how one might modify
|   9018 [Dave@Pr gm t] def enum(count, start=1)
|   9015 [harryo@zi wo] I'm guessing that #to_a returns an ASCII representation of the object.
|   9016 [harryo@zi wo] Since I love talking to myself :-) ...
+ 8648 [masa@st rs g] A,B,C,D,E = (0..4).to_a

Re: Inexplicable core dump
8656 [ntalbott@ro ] I had a problem...

Trouble with %x{} containing quotation marks
8657 [Joe.R.Swatos] ...
8706 [robertg@sw p] ...

FW:  Trouble with %x{} containing quotation marks
8658 [Joe.R.Swatos] Sorry about the formatting in the last message
8666 [eban@os ri .] It seems that the Cygwin DLL is old.  1.1.4?
8694 [Joe.R.Swatos] Great idea, thank you.  Updating helped other things (now dir works with

tk.after memory leak on w2k
8665 [danny@so er ] i'm new to the list and i'm sure this is old hat, but...
8676 [decoux@mo lo] Try with
+ 8707 [danny@so er ] GC
+ 8841 [decoux@mo lo] It's normal, you must know that commands send to tk are stored in an hash.

Is there a precompiled java ruby interface?
8678 [tammo.freese] I would like to use the java ruby interface with jdk1.3. I tried
9657 [Jan.Duennweb] where to get a "java ruby interface" ?
9708 [tammo.freese] java-module-0_5.tar.gz

tr_s!
8679 [jens@ir -n t] I have a problem with tr! ...
+ 8680 [decoux@mo lo] tr!("b-y", "a-z")
+ 8681 [dblack@ca dl] #tr takes strings as its arguments, not regular expressions.  So....

Comparing Ruby and Smalltalk
8682 [feldt@ce ch ] A friend of mine, who is a Smalltalk fan, asked me if there are any
8717 [c.hintze@gm ] I have used Smalltalk for some small projects and educational
8826 [wys@he bl ng] <<SNIP>>
8839 [feldt@ce ch ] Ok, and this might change in the future... 8-)
8845 [chadfowler@y] In most languages, IDE/toolset is orthogonal to the
8854 [wys@he bl ng] I agree, but Smalltalk seems to be slightly different in the sense that
8898 [patrickdloga] Maybe this has been discussed...
8933 [feldt@ce ch ] Thanks everyone who contributed to this thread. I summarized your points

"the 3 great virtues..." Ruby port
8686 [jens@ir -n t] c = "\sTif\s3\s", "hsfbu wj",
8688 [decoux@mo lo] pigeon% cat d.rb
8690 [aleksi.nieme] I'm virtually partially back. So let's hope I'm half as annoying as during
8695 [jens@ir -n t] That was not my problem... Perl is known for the fact
8697 [Dave@Pr gm t] cc  =  %w?Tif\ 3\ hsfbu\ wjsuvft\
8698 [hgs@dm .a .u] Consider the set of valid ruby programs such that
8699 [jens@ir -n t] ok, I'm a bloody Ruby newbie ;-))

Thanks Guys
8691 [ljohnson@re ] Thanks to everyone who responded! I think the quickest approach for my

Dev Ruby on Windows: How do YOU do that?
8700 [fgeiger@da e] I wonder how the Windows folks develop Ruby programs. Which editor do you
+ 8701 [ChrisM@SN LL] I use UltraEdit (www.ultraedit.com) -- not the best editor, but it handles a
| + 8703 [schneik@us i] # I wonder how the Windows folks develop Ruby programs. Which editor do
| + 8772 [fgleeson@my ] <2B92A3D0D399D311BA7E00A0C90F8FDD88BE50@mailgate.snellingcorp.com>,
+ 8705 [joeyGibson@m] I use EMACS or vi (Vim, actually), since they can both understand
+ 8765 [charleshixsn] I usually use RubyWin, though I dislike the way that it handles tabs.  I
8838 [philippe.luc] you

Braces and begin/end
8704 [robertg@sw p] I opened parse.y and made a serious try to understand why but my knowledge
+ 8708 [gotoken@ma h] You can consider {} is almost equal to do/end which is used for block
+ 8709 [c.hintze@gm ] Of course not! ;-)
+ 8719 [matz@ze ab t] It is my goal to build up friendly and respectful community as well as
8723 [sent@qu li y] As others have pointed out, { } is not intended
+ 8725 [Dave@Pr gm t] The style I've come to is to use do/end for multi-line blocks, and
| 8727 [nahi@ke na t] It's my turn.
+ 8730 [kom@ma l1 ac] Maybe you better take a closer look at the difference before deciding
+ 8732 [dblack@ca dl] Have you tried it?  :-)
| + 8734 [kom@ma l1 ac] Thank you for the correction.
| + 8761 [hal9000@hy e] This improvement came with 1.6, I believe.
+ 8735 [Dave@Pr gm t] That then invokes my second style rule. If there's any potential
8751 [sent@qu li y] Or my style rule, which is that ALL parameter
8752 [sent@qu li y] That's not bad. I rarely do single-line stuff,

Keyword arguments (was: Visions for 2001/1.7.x development?)
8711 [ntalbott@ro ] Referring to keyword arguments, matz said,
+ 8712 [avi@be a4 co] The TOM solution, if I remember correctly, was to always have keywords as
+ 8767 [charleshixsn] Yes.  It can make sense to have the parameters be a part of the method
8771 [steve@de f. ] def gigo(x, y, foo=10, bar=8, fred="hello", martha=[0, nil])

GC (was Re:  Re: Interesting Language performance comparisons - Ruby, OCAML etc)
8716 [aseltine@cs ] Is this already in the HEAD cvs branch (1.7)?
8736 [matz@ze ab t] No yet, but it will be merged in several weeks.

Memory allocation strategy (was: Visions for 2001/1.7.x development?)
8724 [kom@ma l1 ac] I can't speak for matz, but I guess the logic behind it is something

Re: Praise for Ruby/GTK
8726 [schneiker@ju] =================================================
8728 [schneiker@ju] Peter Mattis and the GTK+ team
8740 [edwin@lo al ] If you mean "are these features unique to GTK?", I'd say no. For
8742 [pixel_@ma dr] AFAIK the ugly GtkCList and GtkCTree are deprecated in the new Gtk+ (v1.4)

Python to Ruby translation
8729 [jtc@di en io] I'm considering porting some scripts I've written in Python to Ruby.
8731 [c.hintze@gm ] Not that I am aware of one, but perhaps somebody has already done such

Re: overriding Hash#[] (was: Re: bug??)
8733 [dblack@ca dl] You can get around that, I think, with #has_key?

abstract  classes
8737 [maurice@eu o] is there any way to declare abstract classes/methods in ruby?
8754 [sent@qu li y] It depends on what you mean by abstract classes.
8756 [chadfowler@y] It seems like a lot of the rationale for having
8831 [maurice.szmu] ...
+ 8835 [hipster@xs a] Have a look at
+ 8891 [charleshixsn] It sounds like what you might want to do is
8896 [sent@qu li y] The private_class_method method (in Module) can

SystemExit exception in RubyWin
8738 [0317025435@t] I use "ruby 1.6.1 (2000-10-05) [i386-cygwin]"

FloatDomainError exception, bug in Ruby?
8739 [0317025435@t] Still using: "ruby 1.6.1 (2000-10-05) [i386-cygwin]"
8753 [masa@st rs g] snip

Druby and backquotes (bug or just me?)
8741 [ptkwt@us r2 ] #drb_srvr.rb

Backquotes somecommand don't work on Windows (bug)
8744 [ptkwt@us r2 ] Using backquotes to run a system command and collect it's output does not
8748 [0317025435@t] That is probably because you are starting command.com (MS-DOS emulating
8770 [ben_tilly@ho] cmd.exe /c ver

(start..end) where start > end
8745 [0317025435@t] Is there any reason why Ruby can't handle a "backwards" range?
8746 [dblack@ca dl] Oh thou of little faith! :-)
+ 8749 [0317025435@t] Do you agree with me when I say that (7..0) is more beautiful? And it would
8750 [dblack@ca dl] It's certainly more beautiful than the to_a.reverse version.  I don't
8758 [gnhurst@hu s] This leads into the discussion (again ;-) about #succ and
8759 [matju@ca .o ] class Numeric; def pred

subscribe Jonghyouk Yun
8755 [guru13@ne sg] subscribe Jonghyouk Yun

Backward Range implementation
8757 [sent@qu li y] Here's what I came up with that seems to work. I
8788 [dblack@ca dl] Using yours as a starting point, I tried to generalize it so that it
8799 [crippel@pr m] There is probably some code out there which relies on
8802 [dblack@ca dl] No doubt.  The code Kevin and I were writing is not

ruby under dos 6.22?
8760 [rubinubi@cy ] Is it possible for me to get ruby to run on an old 486-dx laptop which is
+ 8763 [chadfowler@y] 1.  Install Linux
+ 8776 [gotoken@ma h] It is compiled with djgpp.

>actually bug in backquotes on Windows
8773 [ptkwt@us r2 ] Actually, I figured out that the problem is with using backquotes on the

No :<, :>, etc. methods for Array
8774 [green@Fr eB ] So, why not include Comparable in Array by default?  It shouldn't have any
8779 [matz@ze ab t] Because arrays are not always comparable.  sort', min', and max' in
8780 [green@Fr eB ] So it's done to "discourage" people from doing it?  E.g. currently the best
+ 8781 [gotoken@ma h] Enumerable#inject' is introduced there.  That may be adopted if an
| 8782 [green@Fr eB ] Except that I don't really ever see the need for the inject functionality
| + 8783 [dblack@ca dl] You're assuming a lot about what people might or might not do.
| | 8785 [green@Fr eB ] I simply mean that even with a lack of an inject method due to there not
| | 8789 [gotoken@ma h] Does anyone know the name of operator family which consists of $\sum$,
| | 8790 [dblack@ca dl] I don't know it, but I doubt it's "Jean".
| | 8793 [gotoken@ma h] Ouch, I shoud not omit verb until I understand English more... :-(
| | + 8794 [avi@be a4 co] Actually, you were using a perfectly normal and correct, if casual,
| | + 8795 [dblack@ca dl] Anyone have a pen I could borrow?  David?
| | | 8798 [gotoken@ma h] Oops, I might make serious mistake; I was so boorish that I fourced
| | + 8857 [hal9000@hy e] Ah, Gotoken, he is only kidding you. There is nothing really ambiguous
| + 8784 [gotoken@ma h] def product
| + 8786 [jmichel@sc u] It can also be used to create product (also very important), and, if the
| + 8797 [crippel@pr m] Obviously,
| | 8800 [gotoken@ma h] I think such collections can be implemented as extension libraries if
| | + 8804 [sent@qu li y] I think I agree, and not for performance reasons.
| | + 8820 [crippel@pr m] I see your point ... . I guess my idea was to be less greedy
| |   + 8822 [ben_tilly@ho] I have mentioned before, perhaps I should mention again.
| |   | + 8824 [crippel@pr m] Since I am pretty about ignorant about Perl (among many
| |   | | 8827 [ben_tilly@ho] If you have Perl installed the following commands will
| |   | | 8828 [crippel@pr m] Thanks
| |   | + 8851 [dblack@ca dl] True, but when you write STORE() and FETCH() and all that in Perl,
| |   |   8855 [ben_tilly@ho] I would not advocate using the names used in Perl.
| |   |   + 8856 [decoux@mo lo] pigeon% wc -l bdb.c
| |   |   + 8859 [dblack@ca dl] Ummmm, not my book.  (Which is called "Law in Film" -- hey, I never
| |   |     8867 [ben_tilly@ho] Ask yourself the following questions first.
| |   |     8869 [decoux@mo lo] True perhaps for DB_File but not for BerkeleyDB, verify it (no
| |   |     8870 [ben_tilly@ho] Erk.  Oops.
| |   |     8887 [dblack@ca dl] [me:]
| |   |     8890 [ben_tilly@ho] Say modules called ArrayInterface, HashInterface and
| |   |     8918 [aleksi.nieme] At least David Alan Black and Ben Tilly have been conversating, the latter's
| |   |     8924 [dblack@ca dl] I think the whole Perl tying part of this is not all that relevant.  In
| |   |     8926 [ben_tilly@ho] It is a simple magic. :-)
| |   |     + 8927 [dblack@ca dl] Hmmm....  Why is that good advertising?  Well, we can agree
| |   |     + 8963 [crippel@pr m] and probably the real reason behind the current
| |   + 8836 [crippel@pr m] I guess the type::''  prefix is totally unnecessary ... . This makes only
| + 8849 [jweirich@on ] I've used inject (or "reduce") in the following to build a list from
+ 8805 [Dave@Pr gm t] When we get around to adding versioning, I hope we'll have a version
8811 [schneik@us i] it,

Modules and mixins
8775 [dsafari@pa a] I find myself having a really hard time understanding what is really going on
+ 8777 [chadfowler@y] Wouldn't it be nice to see something like "Ruby Best
+ 8801 [c.hintze@gm ] do not feel to sad! It is not such a big issue. What makes it a bit
8803 [ben_tilly@ho] It supports more of course. :-)
+ 8806 [sent@qu li y] Personally, I don't go looking for a reason to
| 8899 [dsafari@pa a] Just to clarify, I wasn't looking for a reason to use them, merely for a way
+ 8810 [c.hintze@gm ] Of course. But because the original post was about modules, I have only
| 8815 [ben_tilly@ho] Ah, yes. :-)
| 8858 [aleksi.nieme] I once tried the same, in particular to create keyboard short cuts easily. I
| 8879 [yashi@ya hi ] good to see you back, Aleksi.
| 8897 [ben_tilly@ho] def bind_loops(fn, my_range, *more_args)
| 8901 [crippel@pr m] The length was not the issue on the python ml. The original author of the MRange
| 8902 [crippel@pr m] Ops
| 8917 [ben_tilly@ho] Say no more.  Explicit calculations open up room for all
| 8962 [crippel@pr m] they are also very inefficient and are either (virtually)
| 8967 [crippel@pr m] actually
| 8975 [sent@qu li y] Bingo. Typical aggregation would allow you to
| + 8977 [Dave@Pr gm t] Just to back up what Kevin says: Mixins are also not just passive.
| | + 8979 [ben_tilly@ho] While the point is understood, in this case I suspect that
| | | 8984 [crippel@pr m] I am not quite sure what the this case'' is refering to?
| | + 9030 [crippel@pr m] I agree with you that Ruby facilitates AOP techniques (note that
| |   9032 [hipster@xs a] is the site for general information about AOP (FAQ, working groups,
| |   9033 [hipster@xs a] I forgot to mention
| |   9034 [crippel@pr m] Thanks
| + 9004 [dsafari@pa a] Thank you very much for every contributing to my question. This has been a
+ 8861 [crippel@pr m] Yes, this is a very need solution!
`