39799-40552

39584-40176 subjects 39920-41815

^ More erratum 2 in "Teach Yourself Ruby in 21 Days"
39799 [jzakiya mail] Chapter 12
+ 39802 [ms iastate.e] Yes, we had this one. (see http://www.ruby-lang.org/~slagell/misprints.html)
+ 39805 [philip dynas] I'm sorry to say, but this is correct :)

^ Ruby/GTK with no window movement updates
39808 [fjm_maillist] I'm just starting to play around with Ruby, especially with it's
39825 [jeremy chaos] Frank,
39926 [fjm_maillist] I figured something was going on like that, but don't know enough about how
39932 [nobu.nokada ] Oops, it's X's or kernel's issue, perhaps.

^ 'Ultimate' FreeRIDE ?
39809 [xlucid users] [Note:  This is a post to the FreeRIDE developers mailing list, which I have
+ 39832 [Steve.Merric] Euan Eye, surely? ;-)
| 39851 [hal9000 hype] Ha ha... that depends on context, Steve...
+ 39871 [llothar web.] Its the same goals as ArachnoRuby*, which will be released in the
  39875 [alan digikat] Will ArachnoRuby be Open Source?
  39880 [rich infoeth] And free?
  39888 [llothar web.] No the price will be the same as for the other Scripting/Python IDEs.
  39908 [tobiasreif p] It could be released as Free Software (GPL), and you still could charge
  + 39925 [llothar web.] Sorry most companies showed that this is not a good idea.
  + 39953 [xlucid users] I doubt this in practice.
    + 39959 [pate eylerfa] This only seems to work where there is a perceived benefit to the user in
    + 39969 [dossy panopt] You don't make a living selling development tools, GPL'ed or otherwise.

^ raf update [ruby applications finder]
39813 [lafor arraki] while Raa.succ arrives, I remember you the existence of raf
39929 [stewartNO SP] I notice that under Mozilla 0.98 in Linux, you can follow links
39933 [cameron cwru] Yeah, I am pretty sure its due to the way JavaScript is calling things.

^ Execute .rb file in Ruby path
39820 [home clabs.o] If I install myscript in a sitever sub-directory called mystuff, I can
39823 [matz ruby-la] ruby -r mystuff/myscript.rb -e 0

^ [Q] non-alphabetic character in symbol
39821 [kwatch lycos] I have a question.
39824 [matz ruby-la] Yes.  Minero Aoki once proposed % notation for symbols, but consensus
+ 39866 [kwatch lycos] Thanks matz,
| 39883 [matz ruby-la] If you call intern everytime, hash['+symbol'] (1 hash ref) is faster
| 39924 [kwatch lycos] Oh, I understand clearly.
| 39956 [matz ruby-la] Yes.  It's defined in st.c
+ 39890 [nobu.nokada ] Have these been proposed already?
  39893 [aamine mx.ed] It's not me but Hiroshi Igarashi.
  39910 [nobu.nokada ] I guessed it equals "aa\#{str}bb".intern, it accepts only
  39915 [jean_hugues_] either :A_WORD[:]
  39918 [nobu.nokada ] I guess it's not good for ::WhatEver:, because of parser's
  39922 [jean_hugues_] Well... :\:Whatever: (and also :\\Whatever: for that matter). Seems
  39927 [nobu.nokada ] Also it may change meanings of foo?:Foo:bar, not?
  39935 [jean_hugues_] Yes, that construct could generate an error because foo ? :Foo: bar is

^ RE: [FR-devel] 'Ultimate' FreeRIDE ?
39822 [curt hibbs.c] [snip, lots of good stuff]

^ How-to Cc when using Net/SMTP?
39829 [herman_graal] How can I CC when sending emails with Net/SMTP?
39842 [decoux moulo] You need to add the address 'some_adress@some_host.com' when you call

^ require variants
39830 [tsiivola cc.] Regarding the recent dicussion of having require "dir" recursively require
39840 [pbrannan atd] Handling these would be nontrivial.  I think a better solution is to

^ Why is this a SyntaxError ?
39834 [David.Morian] class A
39836 [dsafari xtra] I would gather that the << is not being bound to the object a. The following

^ Is the FXGIFCursor available on FXRuby ?
39841 [dmg tragsate] I am trying to use the FXGIFCursor primitive, but I receive
+ 39844 [jlj cfdrc.co] FXGIFCursor is available to FXRuby (I don't know why you'd get the message
| 39891 [nobu.nokada ] Perhaps, David might write like FXGIFCursor(...) instead of
| + 39897 [daniel zeped] He's right. I've done that many times.
| + 39936 [jlj cfdrc.co] Yes, I didn't think of that. As a recovering Python programmer I sometimes
+ 39847 [jlj cfdrc.co] OK, more information after some debugging. I did find that it's important to

^ [ANN] QuantLib-Ruby 0.3.0
39843 [ballabio mac] QuantLib-Ruby 0.3.0

^ ruby in cocoon's XSP
39850 [fewayne facs] i just googled this idea with little result, but wanted to ping y'all

^ begin ... rescue WinSuxx; trow RubySuxx; end
39852 [kgergely mla] I have again and again problems with sockets and select.

^ RegExp: using $& in 2nd gsub argument
39853 [chris.morris] "test\n (he\nllo) \nhey".gsub(/\((.*\n.*)\)/, $&.gsub("\n", ''))
39854 [gotoken notw] Move the second argument to inside of block, i.e.,

^ gets and eof?
39856 [pbrannan atd] I have a program that reads input from a socket and displays output on
+ 39857 [wconrad yagn] I can't answer your actual question, but I know that gets seems to
+ 39859 [dossy panopt] In the code shown, gets will read up to the line seperator which comes
+ 39930 [kwatch lycos] .============================ prog1.rb ==
  39937 [decoux moulo] ruby use an active model (is EOF waiting at the stream) rather than a
  39996 [kwatch lycos] I tried another program.
  40006 [decoux moulo] To know if the stream is at EOF ruby try to read from this stream.

^ [ANN] JTTui 0.11.0 - textmode user interface
39865 [j.travnik sh] new version of JTTui has been released.

^ Ruby in Mac OS X 10.2
39867 [gehlker fast] I just heard that Apple has announced that ruby will be pre-installed in the
+ 39872 [rich infoeth] Article link
| 39884 [chris cmb-en] Womder if there's any chance of Ruby coming with the 10.2 Dev Tools.
+ 39904 [ptkwt shell1] Where did you hear this?
  39941 [gehlker fast] Email from a friend at the conference. Now there is an article link.

^ Want application to read file or from pipe
39868 [jim freeze.o] Does anyone have a sample application that can read
+ 39869 [wconrad yagn] wconrad@pluto:~/foo$ cat foo.rb
+ 39870 [tim vegeta.a] $ cat filter.rb
  39873 [jim freeze.o] Close, but I am looking for something without the
  + 39874 [wconrad yagn] Jim, I'm confused.  I thought my little example did it all.  Let me
  | 39877 [ian caliban.] The only change I'd make here is to substitute ARGF for $<, since it's
  | 39879 [wconrad yagn] Agreed.
  + 39876 [dossy panopt] $ ruby -pe '' filename.txt
  + 39878 [tim vegeta.a] Very well. To slurp an entire file into memory all at once, just use the

^ print to cgi error log
39882 [brian www.co] I'm stuck on a silly question.  How do I print to the httpd error log
+ 39885 [ian caliban.] Use '$stderr.puts' to get a new-line at the end of your string. You
| 39892 [brian www.co] I tried $stderr.flush() with no change.  This is a bit frustrating!  Do
| + 39894 [Dave Pragmat] $stderr.puts "Wombat"
| + 39913 [mfp cs.mu.OZ] "The Ruby Way" has a fair chunk on CGI (pp. 452-473 approx.) It deals
+ 39895 [sean chitten] Are you under mod_ruby?  -sc

^ Thread#join doesn't accept a timeout?
39887 [dossy panopt] In trying to port JUnitPerf (which supplies nice test
+ 39889 [rick rickbra] Yes.
| 39947 [dossy panopt] Rick,
| 39949 [pate eylerfa] I think too many people wait for code to be perfect before sending it out.
| 39967 [dossy panopt] Thanks for the positive encouragement!  It's motivated me to
+ 39900 [matz ruby-la] Interesting.  Should it be an RCR?
| 39903 [ptkwt shell1] Yes.  It looks good, though I have to admit that I didn't see the original
| 39912 [jean_hugues_] I guess timeout should be in seconds, as in sleep(), shouldn't it ?
| 39917 [nobu.nokada ] Agreed.  And also in IO.select.
| 39923 [dossy panopt] But, it still gives millisecond precision, right?
| 39942 [nobu.nokada ] Yes, with Float.
| + 39945 [dossy panopt] Hmm.  Still doesn't help me much with regard to mswin32 unless
| | 39946 [nobu.nokada ] Or you may try timeout.rb.
| | 39948 [dossy panopt] Excellent suggestion.  I was just trying to avoid creating
| | 39976 [nobu.nokada ] I added thread.rb to Ruby Shim, so you can use it in 1.6 too.
| + 39960 [matz ruby-la] OK, please commit.
+ 39961 [nathaniel ta] Are the threads critical, or could you use the timeout extension?
  39968 [dossy panopt] I did try using the timeout extension, but this just doesn't
  39970 [nathaniel ta] The timeout call should block until the send either finishes, or passes

^ It's an apache issue (Re: print to cgi error log)
39896 [brian www.co] I just tried the same in a test file.  No luck.  Out of curiosity, I

^ cgi params api
39898 [patrick-may ] One thing that I don't like is the way cgi forces you to treat every
+ 39901 [brian www.co] Since I'm bumping my head against that multiple stuff as I type this,
| 39911 [jean_hugues_] That would be nice. However I don't understand how it works.
| 39943 [patrick-may ] You stick the multiple method on the object stored in the params hash.
+ 39902 [djkea2 mugca] Although I'm only a very casual user of cgi.rb ... me too.
| 39951 [patrick-may ] Well, I'd first maintain a patched version of cgi.rb.  If it ever
+ 39954 [dblack candl] I'm wondering whether there's a nicer name than "multiple" for this --
+ 39962 [wakou fsinet] cgi.rb has undocumented method "CGI#param".
  39985 [patrick-may ] This looks better than my suggestion.  I like that you make the single
  + 39987 [sean chitten] Before these cgi param threads get too far down whatever path they're
  | 39999 [patrick-may ] the single arg is the common case -- instead of cgi['key'].to_s all
  + 40032 [patrick-may ] I just noticed that :to_a conflicts with Enumerable.to_a, which String
    40037 [jean_hugues_] cgi["name"].values ?
    40049 [wakou fsinet] cgi["name"]         # ==> "value"
    40052 [jean_hugues_] Yes ! That works. I learned it recently ! You basically
    40073 [wakou fsinet] cgi["name"].values
    40077 [jean_hugues_] Well... I was comparing .values() versus .multiple()...
    40086 [wakou fsinet] yes,
    40137 [patrick-may ] I feel like I like values better, since it emphasizes the plural more
    40139 [sean chitten] Let me bounce this over from IRC.  This isn't exactly the nicest
    + 40140 [patrick-may ] First of all, sorry I posted my reply to the list in a wierd spot.  These
    | 40145 [dossy panopt] cgi['key'].to_a.each { | value | ... }
    | 40175 [patrick-may ] True, except that :to_a conflicts with the :to_a that String mixes in
    + 40144 [dossy panopt] My two cents?
      + 40148 [dblack candl] "to_a", to_me, suggests something being converted to an array.  That's
      | + 40153 [wakou fsinet] $/ = "\0"
      | | 40155 [dblack candl] An array seems to me like a more natural container for multiple values
      | | + 40161 [wakou fsinet] and,
      | | | + 40169 [wakou fsinet] ???  size is unnecessary.
      | | | + 40178 [patrick-may ] cgi["name"].values is better than cgi["name"].to_a, because of the
      | | |   40190 [wakou fsinet] Enumerable depends on each(). to_a() is defined with each().
      | | + 40177 [patrick-may ] Array is the more natural container for multiple values than a string.
      | |   40193 [dblack candl] My point, though, was that although "val1\0val2\0val3" is a single
      | + 40159 [dossy panopt] I'm saying that 'val1' == 'val1' and that 'val1'.to_a == ['val1']
      |   + 40162 [wakou fsinet] if cgi["name"] return "val1"       ... maybe I don't notice "val2"
      |   | 40191 [dossy panopt] True, but in the Ruby idiom, cgi["name"] ought to return a
      |   | 40194 [dblack candl] Do you mean the CGI idiom?  In Ruby, a hash value can be (and often
      |   | + 40195 [dblack candl] Well, that was stated very narrowly, since it really isn't a Hash
      |   | + 40197 [dossy panopt] Sorry, it's 7:50 AM and I'm still in the "playing with my two
      |   |   40198 [dblack candl] I'm in "about-to-drive-50-minutes-in-the-rain-to-graduation" mode,
      |   |   40203 [dossy panopt] Heh.  I just realized that you're at SHU -- another New Jerseyean!
      |   |   + 40207 [wakou fsinet] Oh, yes! It is important what a class of cgi['foo'] is.
      |   |   + 40209 [Dave Pragmat] How can this be? You show cgi['foo'] as a String in your first case.
      |   |   | 40222 [dossy panopt] I was wrong.  :-)  I didn't mean for it to be a String.  I'm a big
      |   |   | 40228 [wakou fsinet] not good?
      |   |   | 40229 [dossy panopt] Good point.
      |   |   | 40250 [wakou fsinet] I thought so, too. but, is CGI::Param necessary?
      |   |   | 40291 [dossy panopt] Yes.
      |   |   | + 40305 [dblack candl] But see above...
      |   |   | | 40309 [dossy panopt] cgi["key"] could only returns a String if there's only one value
      |   |   | | 40310 [dblack candl] But according to Wakou's proposal, cgi["name"] would always return the
      |   |   | | 40312 [dossy panopt] Ah, but the problem is what if the "name" param has multiple
      |   |   | | 40378 [patrick-may ] I agree that multiple forks would give us a chance to sort the more
      |   |   | + 40322 [wakou fsinet] class CGI
      |   |   |   + 40325 [dossy panopt] Wakou,
      |   |   |   | 40331 [wakou fsinet] Oh, self.replace is not so good, maybe super() is better.
      |   |   |   | 40335 [dossy panopt] Oh, duh.  I forgot String mixes in Enumerable.  Thanks.
      |   |   |   | 40338 [sean BUHBYES] Random idea.  Why not push all the CGI values into a REXML Document or
      |   |   |   | 40403 [wakou fsinet] one of reason, Ruby has no bundled XML libraries.
      |   |   |   + 40328 [dblack candl] Personally I don't like it: it still feels to me like #to_a (on what
      |   |   |   | + 40332 [wakou fsinet] yes, This is a very difficult problem.
      |   |   |   | | 40336 [dossy panopt] This is my preferred idiom, personally.  I don't know about
      |   |   |   | | 40377 [patrick-may ] even if you aren't looking for multiple params?
      |   |   |   | | 40387 [dossy panopt] Yes.  Because this way the code works if there's only one value
      |   |   |   | | 40389 [dblack candl] You could define your #[] to return an empty array instead of nil
      |   |   |   | | 40390 [dossy panopt] Given the parameters:  foo=bar&quux=
      |   |   |   | | 40394 [dblack candl] I was responding to the [0] point quoted above, specifically the
      |   |   |   | | + 40398 [dossy panopt] Right, and as I said, defining #[] to return [] instead of nil
      |   |   |   | | | 40401 [michael_s_ca] Is that important?  If so, conceptually or pragmatically?
      |   |   |   | | | 40404 [dossy panopt] Conceptually, so far.  I haven't come across any good examples
      |   |   |   | | | 40405 [michael_s_ca] You see where I'm going with this train of thought, I'm sure.
      |   |   |   | | | 40425 [dossy panopt] Cool!  Are you lucky enough to get paid to do Ruby, or just here
      |   |   |   | | | 40487 [patrick-may ] I am paid to write software in Ruby.  My suggestions for the cgi api
      |   |   |   | | | 40488 [dossy panopt] Writing web apps in Ruby?  What's the size of the application, would
      |   |   |   | | | 40552 [patrick-may ] My current project is ~ 87k of ruby sources, though I'd love to see
      |   |   |   | | + 40456 [patrick-may ] + cgi["key"] returns multiple values, which is not the most useful
      |   |   |   | | | 40510 [wakou fsinet] ok, I think that it is good.
      |   |   |   | | + 40518 [jonathan ine] Comments at the end...
      |   |   |   | |   40523 [dblack candl] I'm not sure what you mean about thinking of them "as objects".  As
      |   |   |   | + 40334 [dossy panopt] You're not dealing with a string.  You're dealing with a Param,
      |   |   |   |   + 40341 [dblack candl] Right, but what I'm saying is that this Param on which I'm remarking
      |   |   |   |   + 40379 [patrick-may ] Mixin CGI::Param instead of subclassing String, don't walk over any
      |   |   |   |     40386 [dossy panopt] Please explain, using executable Ruby code, how you would mix-in
      |   |   |   + 40344 [patrick-may ] This way, the api still works for file uploads / multipart forms.
      |   |   |   + 40351 [matz ruby-la] Unless you really really seek compatibility, faking something behave
      |   |   |     40402 [wakou fsinet] I like, values is an Array object. because,
      |   |   |     40489 [patrick-may ] This makes the most sense.  It solves the problem from my initial
      |   |   + 40252 [sean ruby-la] Perl is the wantarray() function.
      |   |     40290 [dossy panopt] Yes!  This is definitely wantarray envy.  I've just spent the last
      |   + 40164 [dblack candl] So, if p['name'] returns 'val1', then the object on which you're
      |     40196 [dossy panopt] Ah, I see the problem now.  p['name'] would return some kind
      + 40151 [jean_hugues_] p['name']               >> 'valA'

^ Does the FXGIFCursor draw only in B&W ?
39906 [dmg tragsate] I am trying to use the setDragCursor method on a FXCanvas, but the
39938 [jlj cfdrc.co] [I'm going to also try to post this message to the

^ xemacs & mode-ruby.el, broken ?
39914 [jean_hugues_] I added an autoload in my .xemacs/custom.el and that apparently works.

^ xemacs & mode-ruby.el, broken ?
39916 [jean_hugues_] Apparently there is a problem with both " and ' inside
39919 [nobu.nokada ] You mean font-lock-mode issue?  Then it's known but hard to
39921 [jean_hugues_] Thanks. I am still having difficulties with syntax highlighting
39944 [nobu.nokada ] Concretely, in what case?
40011 [jean_hugues_] When I open a Ruby file, it is displayed with some coloring, very little.
40013 [Dave Pragmat] I found that with XEmacs I have to add a (load "font-lock") to the
40014 [nobu.nokada ] Fontifying codes are surrounded by (cond ((featurep 'font-lock))).
threads.html
top