77356-82423 subjects 77820-82320

77593 [ ted@ph .n t] Has anyone written/converted a pseudo-random number generator that IS
+ 77594 [jack_d_herri] Have you looked at LavaRnd?
| 77616 [Tim.Hunter@s] Here's a pair of classes that use www.random.org and
+ 77878 [feldt@ce ch ] I have a pure-Ruby implementation of Fortuna (the crypto-PRNG

Ruby-talk, NGW   C6J   K06VuMr   Mvy
77604 [fisaksen@ks ] <html>

77612 [simon_otedol] GULF BANK PLC,

OpenSSL & Ruby 1.8
77615 [nathaniel@NO] I'm trying to use the OpenSSL extension included in Ruby 1.8preview6 with

TCPSocket  & $SAFE=1
77617 [jolsen@ma l2] In a mod_ruby script I am trying to lookup an ip address with
+ 77643 [matz@ru y- a] How it fails?
+ 77685 [decoux@mo lo] You must untaint the String given to TCPSocket::gethostbyname : be
  77759 [jolsen@ma l2] Yes, that did the trick. Thanks for the hint.

error message for Shell class
77620 [internetlett] I trid to test Shell class
77635 [mwilson13@co] I tried this code using Ruby 1.8.0preview6 and it worked without any
77735 [internetlett] thanks

Extracting a parent class
77623 [mgarriss@ea ] Sorry about the newbie question but....
+ 77624 [hal9000@hy e] I don't know of a simple, direct way to do this.
| 77628 [mgarriss@ea ] It is VERY possible that I should be doing it in another way.  My
| 77630 [drbrain@se m] =20
| + 77632 [mgarriss@ea ] Well, not laugh., but I am using DRb.  I've had a difficult time
| + 77636 [mgarriss@ea ] I suppose that containment is a good idea but I do have methods that
|   77650 [aredridel@nb] Or query based on interface -- ask if it has the methods you expect, not
+ 77625 [dcarrera@ma ] a = b.class.superclass.new
  77626 [mgarriss@ea ] Sorry...I should have been more clear.  While this works it does not
  77627 [dcarrera@ma ] Oh, sorry.
  + 77653 [mgarriss@ea ] Just in case anyone is following this, I have a new solution to my
  | 77658 [drbrain@se m] =20
  | 77659 [mgarriss@ea ] Very cool, thank you.
  + 77688 [g@in er ec -] Unless you were to use the Struct class instead of a normal class, in

Knuthian vs Matsumoto era's, RDSL
77637 [john.carter@] The report on the RDSL project
77674 [mgarriss@ea ] This has something to do with computer power growing faster then brain

DBI-PG problems
77638 [culley@fa tm] I am having trouble with DBI-Pg on os X (ruby 1.6.7).  I installed it
+ 77641 [mgarriss@ea ] You need to install the postgres package from RAA first.  DBI sits on
| 77656 [culley@fa tm] this worked.  I had to upgrade ruby to 1.6.8-- OS X 10.2 comes with
| 77660 [mgarriss@ea ] no problem....I spent an hour staring at that 'load_driver' error
+ 77644 [alex@pr ss r] I had problems like this after installing DBD-Pg on Linux & on Cygwin
  77645 [mgarriss@ea ] I'm not sure but I don't think you would have made it through the
  77706 [B.Candler@po] The DBI package is pure ruby and doesn't depend on any C libraries, so it
  77741 [mgarriss@ea ] While installing DBI I got the error that it could not find the
  77745 [B.Candler@po] Strange, because that's not the same here. My machine has never been near
  77747 [mgarriss@ea ] I just tried it again and as you probably suspected you are right. :)

Test::Unit leaving phanthom objects in ObjectSpace?
77647 [james_b@ne r] I have some code that examines the objects in ObjectSpace and returns a
+ 77686 [news@st d. t] I cannot reproduce it using ruby 1.8.0 (2003-07-24) [i686-linux], so
| 77776 [james_b@ne r] Your example code didn't quite mimick my situation (left-over objects
+ 77778 [nathaniel@NO] I don't think Test::Unit is doing anything funky; my guess is that it's just
  77781 [james_b@ne r] Hm.  That'a an idea ...
  77790 [nathaniel@NO] def my_object_space_code(object_space=ObjectSpace)

ANN: debugprint
77652 [ben@th ng ag] Growing out of the discussions about $DEBUG and $VERBOSE, and how
77661 [matz@ru y- a] How about direct messages to $stderr?
+ 77670 [ben@th ng ag] Good point, I knew it was a wee bit too simple.  I'll make the change
+ 77671 [ben@th ng ag] Actually, how about debug() goes to $stderr and info() goes to $stdout.
| 77717 [dblack@su er] See Brian's answer for something more in-depth :-)  But I'd say
+ 77709 [B.Candler@po] Why not $deferr ? Or has that distinction been lost now?
  + 77742 [mgarriss@ea ] Very nice Brian.  Can I use this code?
  | 77749 [B.Candler@po] Be my guest - I'm very happy to have someone else host it, maintain it,
  | 77779 [gsinclair@so] At which point you might consider using Log4r.
  + 77743 [matz@ru y- a] Yes, it's lost in the final 1.8.0.

A bundle of newbie queries
77662 [xlucid@us rs] I've finally overcome my newbie embarrassment enough to post about
+ 77666 [mgarriss@ea ] Wow..well my 2cents about #2.  Not sure but try using the block version
| 77668 [mgarriss@ea ] -I cut and paste multi line code into irb all the time.  Should work.
+ 77669 [wjl@ic ca er] aFile.rewind
+ 77672 [mwilson13@co] File.open("filename") { |f| f.each_line etc. ... }
+ 77715 [dblack@su er] Welcome, nuby!
| + 77719 [B.Candler@po] That example also reads in the entire file: that's what 'readlines' does :-)
| | 77722 [dblack@su er] Yikes.  I seem to have readlines disease.  Thanks for the corrections.
| + 77738 [xlucid@us rs] It was in IRB, so once I had the dictionary populated, I was also then
|   77739 [B.Candler@po] h = {}
|   + 77751 [xlucid@us rs] You're absolutely right - I pasted in the wrong bits of code!
|   + 77803 [dblack@su er] $ ruby/1.8.0/bin/ruby -ve 'h={};h[/b/]=1;h[/b/]=1;p h'
+ 77814 [xlucid@us rs] Thank you for all your help, and my apologies again for having given
  77830 [mwilson13@co] Thank you for all your help, and my apologies again for having given

Ruby-talk, ?? ? -  lFMRJXGuITJtpOk
77664 [Ollila@se dm] jLFUTvaxEKOLAZNtJtceOxmUcvMdDkGNXkTQGTojueXUeKdNUfTJCgE

Re: debugprint
77667 [ng@jo nw on ] Ben,
+ 77673 [ben@th ng ag] Yeah, only unless it's in a macro, those will always give the file and
| 77690 [g@in er ec -] caller(1).first - although it will return a string, so that may need
+ 77777 [flgr@cc n. e] Moin!

Re: Outlook
77676 [bdcms2@em il] hmm, okay so you want to save some money. take a look..

quick q
77677 [btodds@ya oo] holy pork bun! - you have to see this crazy site, I saved $3000...

static libraries with extconf.rb/mkmf
77680 [maillist@be ] How do I "have_library" with a static library in extconf.rb/mkmf?
77689 [decoux@mo lo] What is your problem with have_library ?

CGI multipart 1.8
77687 [ gm@na el gr] I noticed a strange change in cgi.rb in version 1.8. The mulipart

77691 [sinsa_sokkou] ޤǤ

cgi error 1.8.p4
77692 [ gm@na el gr] /usr/local/lib/ruby/1.8/cgi.rb:804:in `initialize': superclass method
77693 [decoux@mo lo] Well, 804 is
77899 [ gm@na el gr] gmosx@sith:~/dev/n1> ruby -vrstringio -e 'p StringIO.new'
77937 [decoux@mo lo] Well, the error was given when it tried to create a new StringIO : I don't
77978 [drak@na el g] Guy, thanx for your hint! Problem solved - mod_ruby had to be

77697 [sotedola@em ] GULF BANK PLC,

is rubyforge down ?
77698 [0bz63fz3m1qt] PING rubyforge.org ( 56 data bytes
77699 [0bz63fz3m1qt] OK..10 minutes later, RubyForge is back again.
77707 [tom@in oe he] ==============
77720 [0bz63fz3m1qt] Looks good.
77726 [tom@in oe he] Could be fallout from our ISP's recent network maintenance downtime....
77799 [0bz63fz3m1qt] Right now I cannot get in touch with Rubyforge.. ISP problems again :-)
77804 [tom@in oe he] Back up again now.... sigh... very odd.
+ 77822 [hal9000@hy e] I'll bet it was the Cisco flaw. ISPs everywhere,
| 77828 [tom@in oe he] Rich and I set up a ping-o-matic to keep an eye on RubyForge; if the
| 77860 [0bz63fz3m1qt] What must not happen, happens.
| + 77865 [tom@in oe he] Hm.  I'm most familiar with CVS, and there seem to be lots of tools
| + 77866 [gsinclair@so] BitKeeper is probably the only VCS that handles offline commits (I'd
+ 77872 [james_b@ne r] The site appears to be up and stable, but I registered for an account
  + 77874 [dooby@d1 .k ] I registered 2.5 hours ago and the receipt arrived within 5 mins.
  + 77876 [dooby@d1 .k ] That might not be a question for Tom because if you mailed
  | 77877 [james_b@ne r] Do you know this to be true, or are you just guessing?
  + 77886 [tom@in oe he] [root@rubyforge gforge]# cat /var/log/maillog* | grep britt
    + 77889 [tom@in oe he] Thinking about this some more; it all seems pretty automatable.  If the
    + 77893 [james_b@ne r] <snip/>

ruby 1.8.0 preview7
77701 [matz@ru y- a] According to our roadmap, this must be the last preview.
+ 77729 [tom@in oe he] Yours,
+ 77754 [ruby-talk@wh] I'd actually like to mirror the whole the distribution tree on a server I have
| 77821 [matz@ru y- a] I am no longer primary site maintainer.
+ 77758 [airboss@no e] readline.c:545: `HISTORY_STATE' undeclared (first use in this function)
  77792 [kero@ch ll .] You'll probably need
  77819 [airboss@no e] If it were as simple as that, I wouldn't have mentioned it. But there
  77825 [kero@ch ll .] Did you throw away ext/readline/Makefile before running configure?
  77904 [ajm@nb sy pa] Lyle, I tried to build 1.0.24 against the Ruby 1.8.0 pre 7 and I get a marco

immediate assitance
77703 [acc_anhammed] I know this mail may come to you as a surprise, I got your email

BigDecimal from CVS
77705 [dsafari@pa a] I just got the CVS version of Ruby installed and was pleasantly surprised to
77718 [decoux@mo lo] svg% ruby -rbigdecimal -rbigdecimal/util -e 'p "111.333333".to_d'
77772 [dsafari@pa a] Ok, thank you for pointing out my user error. What about the Float#to_d, is
77801 [decoux@mo lo] Well selt is probably a typo, now for BigFloat
77810 [dooby@d1 .k ] selt may be a fypo ... like BigTloaf

77710 [blasl@fz .d ] ...

Problem with process
77711 [blasl@fz .d ] can anyone give a hint, why ruby skips the input, when i run this
+ 77714 [decoux@mo lo] search = $stdin.gets # otherwise ruby read a line from the
| 77724 [pbrannan@at ] This confused me the first time I saw it, and now I always use
| 77730 [bob.news@gm ] "Paul Brannan" <pbrannan@atdesk.com> schrieb im Newsbeitrag
+ 77716 [bob.news@gm ] "Markus Blasl" <blasl@fzi.de> schrieb im Newsbeitrag

OT: Nice
77721 [martindemell] Very neat language I discovered recently - similar in feel and spirit if
77723 [pbrannan@at ] Interesting.  I think this is the right approach to take when creating a
+ 77725 [dcarrera@ma ] I think it just "declares" p as an instance of Person.  So it's not
+ 77728 [decoux@mo lo] to specialize the argument (it implement multi-dispatch)
| 77732 [martindemell] Well, a 'default' implementation that the multimethod dispatcher falls
| 77733 [decoux@mo lo] Apparently he make a difference between function and methods
| 77736 [martindemell] Oops, sorry - I misunderstood your last post.
+ 77731 [martindemell] display(p) -> default implementation

Partnership / Investment
77737 [charlee_why@] Dear Sir,

error compiling ruby 1.8.0 preview 7 on i386 OpenBSD -current, diff attached
77746 [deicher@sa d] ...
77761 [nagai@ai ky ] Thank you for your report. I fixed it.

Creating sessions with a specified cookie path
77748 [david@lo dt ] Is there any way to specify the path in the cookie that gets created on
77750 [decoux@mo lo] Have you tried with
77752 [david@lo dt ] Excuse my ignorance. My bad for believing that it wouldn't exist just

[MEETING] CORE Meeting August 9th
77753 [djberg96@ya ] CORE (COlorado Ruby Enthusiasts) will be having a

RDoc now does constants
77757 [dave@pr gp o] I've added experimental support for documenting constants to RDoc. Right
77765 [hal9000@hy e] Does this mean that Structs will now be
77769 [dave@pr gp o] # A Person is used to hold information on
+ 77770 [hal9000@hy e] Well, call me crazy, but I sometimes create a Struct
| 77775 [dave@pr gp o] Me too - but Structs and classes are different enough to make that
+ 77802 [wew@wi li mw] I was the one discussing documenting structs on irc, and also who
  77812 [dave@pr gp o] Umm.. it is a class :)

$stdout.sync = true in Ruby1.8
77760 [cedric.foll@] I'm using the debian package of Ruby 1.8 found on deb.ruby-lan.org.
77785 [matz@ru y- a] Could you describe how it differ, preferably with problem reproducing

rubynet-announce Digest, Vol 6, Issue 1
77762 [rubynet-anno] Send rubynet-announce mailing list submissions to

rb_raise with dynamic strings and GC
77771 [danj@3s el c] char *message;
77774 [nobu.nokada@] Yes, it is expected so.

1.8.0-preview7 (bccwin32 build)  make install problem
77782 [dooby@d1 .k ] Borland build on Win98se.
77786 [eban@os ri .] I cannot reproduce your problem on Windows 2000 Pro.
77793 [dooby@d1 .k ] I'm getting the error with every bcc32 Ruby
77800 [eban@os ri .] ===================================================================
77808 [dooby@d1 .k ] Works here :-)))

Fwd: Great Site
77784 [bclaudia@ya ] I refinanced my mortgage and this site got me the best financer available

Re: 1.8.0-preview6/7 (bccwin32 build)  make install problem
77787 [dooby@d1 .k ] ... same for -p6 ..

2 seconds
77788 [bredd@em il ] only the banks know about this, but it will save you a fortune

Anonymous, single instance class creation
77791 [nathaniel@NO] some_external_object = Object.new
77795 [B.Candler@po] module MyModule
77811 [nathaniel@NO] But of course ;-)
+ 77813 [B.Candler@po] I was hallucinating, it's Class.new which can take a block. But it looks
+ 77815 [B.Candler@po] ...
| 77829 [nathaniel@NO] Oh, I've created that class many times in the past... but I guess after
+ 77882 [ged@Fa ri MU] require 'test/unit'

1.8.0-previewX  rb_sys_fail() on socket instead of an Exception.
77794 [kero@ch ll .] Find some code below to reliably cause a rb_sys_fail() for server.rb,
+ 77798 [B.Candler@po] rb_sys_fail(mesg)
| 77824 [kero@ch ll .] yup.
+ 77805 [decoux@mo lo] The problem is that appendline() use ferror() when rb_sys_fail() use
  77807 [B.Candler@po] Interesting. The FreeBSD manpage for ferror says exactly the same, although
  77809 [decoux@mo lo] No, ruby has cleared errno (see the source of server.rb). The problem is
  77827 [matz@ru y- a] I want to fix this.  But I don't still understand the problem.
  77832 [decoux@mo lo] Well, server.rb is
  77839 [matz@ru y- a] Aha, I understand.  If I move clearerr() before rb_sys_fail(), will it
  + 77851 [kero@ch ll .] Interestingly, the program now never stops with printing 'nil' (as if the
  | 77880 [decoux@mo lo] Well, I hope that you have understood that there is a problem in your
  | 77883 [kero@ch ll .] Yes, I understand :)
  | 77885 [decoux@mo lo] For me, it's scheduled, here an example
  + 77879 [decoux@mo lo] Well, the problem is that there is many call to ferror() in io.c (for

[BUG] 1.8.previews smtp.rb auth
77796 [invalid@in a] I think that smtp auth in the 1.8 series rubies has a small bug. smtp.rb
77797 [aamine@lo er] Thank you.  Checked in now.

Indentation Style
77806 [lone-star@ho] I've been meaning to ask this for quite some time.  Why is and
+ 77816 [matz@ru y- a] Just because I felt too much indentation is like wasting horizontal
+ 77817 [tom.hurst@cl] This is a common style for GNU projects too.  Actually, GNU projects
| + 77838 [seth@cq .c m] That is true of any tab size, providing you don't mix spaces in (as noted above).  In fact I tend to use two, but I expand it to four or eight temporarily is I want to see the block structure more clearly.
| | 77840 [ben@th ng ag] You tab-users do make a good argument for tabs, so what happens when you want
| | + 77841 [seth@cq .c m] There is no need for any additional code or info for emacs to strip
| | | 77843 [mgarriss@ea ] Vim can do this too of course.  In a large development group (like open
| | | 77916 [harryo@qi so] I tried to avoid jumping in, because this often turns into a religious debate, but I thought I'd make just one point ...
| | | + 77924 [aredridel@nb] The pager issue I can see, but there's the beauty of expand(1) or
| | | + 77926 [seth@cq .c m] All UNIX printer drivers have a "convert tabs to spaces" option.
| | + 77842 [mgarriss@ea ] Tab until you get to the 'right level' (in this case, in line with the
| | | + 77906 [fxn@ha hr f.] The problem here is that the alignment depends on your local settings
| | | + 77923 [lone-star@ho] yeah.  this is what works for any setting, and allows you to still use
| | |   78554 [dagbrown@LA ] Yes it does; it's done it for ages.
| | |   78586 [lone-star@ho] OMG...read the manual, try it out.  see what happens?  after you indent
| | + 77844 [kent@ze os h] $ man 9 style
| | | + 77845 [seth@cq .c m] That identation style, with which I have no disagreement, has
| | | + 77864 [xlucid@us rs] It sounds to me like what they actually want is a '1/2 tab'  fo
| | + 77933 [wsdng@on in ] I hate this indentation style. I only use it because it seems to be a
| |   78628 [d.borodaenko] That one is better, but doesn't save you much vertical space anyway, how
| + 77852 [tokikenshi@f] Aren't two spaces the unspoken standard for shell-scripts?
+ 77859 [jweirich@on ] Plaenty of others have weighed in on the style issue and tabs vs
+ 77863 [ ted@ph .n t] Mee, too.  I set tabstop=4 in vim.  2 spaces makes it hard for me to
+ 78654 [sera@fh an .] One thing that hasn't been brought up in this discussion is that a lot