8382-8730 subjects 8815-9155

WIN32: writing registry entries?
8569 [hadmut@da is] I am trying to change values in the Windows Registry.
8570 [jan.svitok@g] registry.open(regpath, Win32::Registry::KEY_WRITE) do |reg|

[ ruby-Bugs-5365 ] Ruby fails to setgid() properly on OS X
8572 [noreply@ru y] Bugs item #5365, was opened at 2006-08-09 08:58

[ ruby-Bugs-5366 ] actionpack-1.12.4 installation error
8573 [noreply@ru y] Bugs item #5366, was opened at 2006-08-09 20:23

1.8.5 final update candidates (was Re: RDoc picking up comments from function prototypes)
8577 [matz@ru y- a] Yep, it should be a few days earlier.  But I do want 1.8.5 to be the
8579 [steven.jenki] It's a two-line patch that fixes a fairly gross error in Rdoc.

Accessing the call stack in an OO way
8578 [jos@ca no k.] Why does Kernel#caller not return an Array of CallStackFrame objects? It

[ ruby-Bugs-5371 ] Can't launch Freeride
8583 [noreply@ru y] Bugs item #5371, was opened at 10/08/2006 10:06

[ ruby-Bugs-5372 ] Float#round sample code documentation error
8584 [noreply@ru y] Bugs item #5372, was opened at 2006-08-10 10:50

RDoc: extensions spread across multiple C files
8585 [tilman@co e-] In ruby-core:4326
8626 [tilman@co e-] Ping? Who's maintaining RDoc these days?
8974 [tilman@co e-] Ping 2, would someone _please_ realize that this is actually a problem
+ 8977 [steven.jenki] The tracker statistics on RubyForge are not very encouraging. Of the 108
| 8985 [tom@in oe he] I think also that perhaps some patches/bugs are posted and the poster
+ 8981 [drbrain@se m] This patch was applied on 8/27/2006.  Sorry if I forgot to send you
  8982 [tilman@co e-] Oh, that's awesome. Thanks :)

ri problem with the latest ruby_1_8
8593 [ksruby@gm il] Does anyone know why for some strange reason ri doesn't know about any
+ 8594 [drbrain@se m] It looks like the ri isn't being generated at all.  My /usr/local/
| 8595 [ksruby@gm il] ...
| 8596 [drbrain@se m] If ri data is generated for Module it will show up in the ri
+ 8597 [drbrain@se m] cvsroot=src&r1=1.2&r2=1.3
  8599 [ksruby@gm il] ...

[PATCH] String#split rdoc
8603 [jan.svitok@g] ...
8606 [drbrain@se m] Thank you.  I committed it.

x-compiling for armv4l (1.8.5-preview3)
8604 [kero@ch ll .] As usual, compilation is fine, Ruby runs fine. This time I spent a bit

Re: Welcome to our (ruby-core ML) You are added automatically
8605 [tom.gunderse] ...

[ ruby-Bugs-5401 ] problem with utf-8 string (chinese character)
8607 [noreply@ru y] Bugs item #5401, was opened at 2006-08-13 08:11

Another ri problem (ruby_1_8 branch)
8608 [ksruby@gm il] $ ri --system String -T | head -10
8634 [drbrain@se m] Fixed.
+ 8637 [ksruby@gm il] Thank you.
+ 8673 [ksruby@gm il] $ ri --system Dir | head
  8685 [drbrain@se m] I don't know if I can fix this one this close to 1.8 either.  I think
  8686 [ksruby@gm il] From my prospective, this is a showstopper.
  8688 [drbrain@se m] 1_8 is frozen.  I cannot fix it without matz' approval.
  8692 [ville.mattil] I think we should fix the RI in way or another. It is a vital tool

Again Range=== bug
8609 [neleai@se na] Problem of discrete membership at Range#=== is that it returns unexpected
8610 [nobu@ru y- a] Really?
8611 [neleai@se na] Can someone simply cut my text and paste it???
8612 [dblack@wo bl] irb(main):010:0> (Time.new...Time.new+5)===Time.new
8613 [neleai@se na] My bad assumption that 1.9.0 is at sync with 1.8.4
8614 [kero@ch ll .] $ ruby1.9 -ve 't = Time.new; p((t...t+5)===t+1)'

[ ruby-Bugs-5415 ] invalid test in "sudo make install-doc"?
8616 [noreply@ru y] Bugs item #5415, was opened at 2006-08-14 12:01
8621 [nobu@ru y- a] What's the exact message?
8622 [ryand-ruby@z] I just found it...
8635 [nobu@ru y- a] Thank you.
+ 8654 [matz@ru y- a] Can you commit?
+ 9546 [werschlein@i] I think it is a bug, too. But it doesn't affect root only. File.stat doesn't
  9548 [nobu@ru y- a] Thank you.

[ ruby-Bugs-5416 ] invalid test in "sudo make install-doc"?
8617 [noreply@ru y] Bugs item #5416, was opened at 2006-08-14 12:02

8618 [ryand-ruby@z] func=detail&atid=1698&aid=5365&group_id=426

Process.gid= fails on OS X - need feedback
8619 [ryand-ruby@z] (sorry for repost, All I've added is a subject to attract eyes)
8620 [nobu@ru y- a] And [ruby-core:8555] and the thread started from it?

deadlock/segmentation fault when running barriers
8623 [sascha_hunol] When I was playing around with synchronization options in various
8708 [sascha_hunol] Anybody?

Another ri problem #2 (ruby_1_8 branch)
8653 [ksruby@gm il] $ ri --system -T IO | tail -10
8672 [ksruby@gm il] Anyone?
8684 [drbrain@se m] I think this is too close to 1.8.5 to fix.  private should work

char-actization Curses.getch
8656 [neleai@se na] Another issue at head is that Curses.getch still returns int.
8657 [nobu@ru y- a] It's a problem.
8658 [neleai@se na] Dont forget constants. All in Curses::Key are from keypad

[ ruby-Bugs-5450 ] String#split includes match groups in result
8659 [noreply@ru y] Bugs item #5450, was opened at 2006-08-17 23:06
8660 [matz@ru y- a] Yes.

Set ===
8661 [neleai@se na] class Set has include?0 and member? but not ===

[BUG] NODE_WHEN inside a case else body
8662 [dbatml@gm .d] case false
8663 [drbrain@se m] ...
8664 [dbatml@gm .d] Good idea, but it probably should be NEW_CASE(0, $3) instead of

[rake] patch to rule
8666 [nobu@ru y- a] The latest rake doesn't work with some patterns.
8691 [jim@we ri hh] Wow, you are really pushing the limits here.  Good!
+ 8693 [jim@we ri hh] BTW, I've made a beta of these recent changes (version,
+ 8694 [nobu@ru y- a] Fair enough.  How I use dot files sometimes is to "touch".

Building with -Wall => questions
8667 [hgs@dm .a .u] Having built the 1.8.5-preview4 ruby on Solaris9, I wondered if
+ 8668 [now@bi wi se] No!  Considering that a lot of code (basically any Ruby program)
| 8671 [llasram@gm i] Ditto.  I'm a fan of '-Wall -Wextra -Werror' [1] myself ;-).  In my
+ 8669 [Daniel.Berge] I doubt you'll get anywhere with this.  I've already tried. :)

Build of Preview4 on solaris 2.9
8670 [hgs@dm .a .u] ...

Ruby 1.8.5 preview5
8675 [matz@ru y- a] I have just submitted 1.8.5 preview5, which should be the final
+ 8676 [hgs@dm .a .u] ...
+ 8677 [emiel@re no ] Just one issue to report here.
+ 8682 [tilman@co e-] Something is wrong with ri (or maybe rdoc) in this one.
+ 8687 [drbrain@se m] $ make
+ 8689 [tom@in oe he] Yours,

[ ruby-Bugs-5489 ] Duplicate Win32API class declarations
8678 [noreply@ru y] Bugs item #5489, was opened at 2006-08-22 16:14
8680 [nobu@ru y- a] Thank you for the report.

8681 [tad.bochan@b] I have been making a foray into the world of SOAP/WSDL etc and having

a ruby-core primer
8690 [ruby-core@wh] I'd like this page to be our English primer to this mailing list and to core
+ 8696 [jgbailey@gm ] First, Thanks for taking the time to put this page toghether, it gives
| 8697 [ruby-core@wh] The only two relevant tags right now are RUBY_1_8 and HEAD, but I will
| 8704 [nobu@ru y- a] Why not `cvs diff'?
| 8714 [ruby-core@wh] Yes, okay, let's do that.  Thanks!
+ 8699 [mental@ry ia] Does the K&R guideline still apply?  Even eval.c was recently
| 8703 [nobu@ru y- a] Protoized sources are in 1.9, not only eval.c.  eval.c in 1.8
+ 8700 [jmg3000@gm i] page.sub!( /persistant/, "persistent" )
+ 8701 [hgs@dm .a .u] Thank you for creating this.  Firstly a couple of off-topic [more
| + 8702 [james@gr yp ] I have forwarded your comments to the VIT Core discussion list for
| + 8715 [ruby-core@wh] Thankyou, Hugh!  Very apt feedback.  I was able to incorporate most of it.
+ 8705 [nobu@ru y- a] Though this is my personal preference, I'd like the coding
+ 8706 [2006@ho el .] Outstanding. :)
  + 8707 [znmeb@ce ma ] I don't know how this is handled on MacOS, but on "most modern Linux
  + 8713 [meta@po ox c] ...
    8716 [2006@ho el .] Yay. :)
    8718 [ruby-core@wh] I'm really glad for your lengthy input, Dave, and I'm working on a second page
    + 8734 [surrender_it] possibly one about --program-suffix too?
    | 8745 [drbrain@se m] make test
    | 8746 [nobu@ru y- a] `make check' is same as them.
    + 8751 [2006@ho el .] You're most welcome.

More ri-problems (ruby_1_8 branch again)
8709 [holmberg@ia ] I have noticed the previous threads about problems with ri in the
+ 8711 [nobu@ru y- a] I guess it is the feature of rdoc, not a bug.
+ 8712 [ksruby@gm il] I agree the situation with ri is quite bad. Besides several apparent
  + 8717 [2006@ho el .] I'm a bit confused here. These sound like problems from rdoc, not ri.
  | 8720 [drbrain@se m] RDoc doesn't record a method's file of origin, so you can't tell that
  + 8719 [drbrain@se m] I'm working on it, but haven't figure out how to do it right yet.

Incorrect return of the TkFont configure instance method when no value
8710 [lionel.maiau] Thanks to let me know it is not the right place for such topic :-)

Work syscall at 64bits?
8721 [neleai@se na] In io.c is
8722 [nobu@ru y- a] Use ext/dl instead.

[PATCH] division by zero in rational.rb
8723 [calamitates@] ...

[PATCH] typo in rdoc/parsers/parse_rb.rb
8724 [calamitates@] ...
8731 [drbrain@se m] Thank you, I committed it.

[ ruby-Bugs-5529 ] error in test_s (File.size?) documentation in file.c
8725 [noreply@ru y] Bugs item #5529, was opened at 2006-08-25 16:41
8729 [drbrain@se m] On Aug 25, 2006, at 4:41 PM, <noreply@rubyforge.org>

killing ensure (was Re: sandbox timers & block scopes)
8726 [ruby-core@wh] I will try to approach this differently.  Is there a way to circumvent ensure?

Problem with setreuid under Mac OS X
8727 [vincent.isam] I think there is a little problem in the modification for setregid of
8728 [nobu@ru y- a] setreuid() in the system library is broken.
8733 [vincent.isam] Yes, I understand that but even if ruby defines a better one, it

[ ruby-Bugs-5551 ] Error Message: "Unhandled Exception Caught"  in Dialog Box "FreeRIDE went south"
8732 [noreply@ru y] Bugs item #5551, was opened at 2006-08-27 23:46

Legal operator symbols
8735 [now@bi wi se] Why are :>, :>=, :<=, :< fine as symbols, while := isn't?
+ 8736 [dblack@wo bl] The only thing that strikes me is that the </> ones are all methods,
| 8737 [now@bi wi se] Ah, true.  :== is fine, while := isn't.  Oh well.
| 8738 [vincent.isam] I think that's because you can redefine all comparison operators, but
| 8742 [vjoel@pa h. ] irb(main):020:0> x = []
| 8744 [headius@he d] ...
+ 8740 [vjoel@pa h. ] I don't know why (maybe a parser issue, or maybe := has been reserved
| 8741 [headius@he d] ...
+ 8754 [matju@ar en ] ...

ruby from inetd/openssl: Socket in/out ?
8739 [hadmut@da is] I want to write a network service in Ruby, which is started by inetd

bug? continuation called across trap
8743 [youngh@ca da] I posted this on ruby-talk but didn't get any replies.  Hopefully
8747 [nobu@ru y- a] No error on Linux.  Maybe the issue around libedit on Mac OS X.

[ ruby-Patches-5555 ] Readline.completion_with_context_proc = returns current text and full line
8748 [noreply@ru y] Patches item #5555, was opened at 2006-08-29 03:55

ruby-lang.org services are unstable
8749 [shugo@ru y- ] Services of ruby-lang.org are unstable because of hardware trouble.

[ ruby-Bugs-5563 ] rdoc/usage uses an uninitalized constant
8750 [noreply@ru y] Bugs item #5563, was opened at 2006-08-29 13:18

sandboxes and rb_ensure
8752 [mental@ry ia] Been thinking about the ensure issue -- the thing is, right now, the

Server maintenance
8753 [shugo@ru y- ] Services except mailinglists will be down for a server maintenance on
8757 [shugo@ru y- ] Services are restarted. Thank you.

sandbox r47, exposing top_cref v. emulating require
8755 [ruby-core@wh] Finally, some big developments for the sandbox.  I've got it running RubyGems!
8756 [ruby-core@wh] Oh, I mean shared libs.  Sorry bout that.

sandbox r50, here we go, loading conflicting gems
8758 [ruby-core@wh] Checky.
8760 [nobu@ru y- a] I guess it has to be prefixed to be public.
8761 [ruby-core@wh] ...
8762 [matz@ru y- a] Rename top_cref to ruby_top_cref, and apply the patch to ruby_1_8 (and
8763 [ruby-core@wh] Thankyou, done.  And sandbox r58 reflects.[1]

cooperating with other GCs
8759 [mental@ry ia] C++ application which uses its own conservative garbage collector,
8769 [headius@he d] ...

[Q] How would you do method extensions (sometimes called class extensions) in Ruby?
8764 [gaelli@em rg] (1) The Problem: Bridge Gap between Tests and Documentation with
+ 8765 [nicolas.desp] I think you better ask this question on the ruby-talk ML.
+ 8766 [gaelli@em rg] Sam and Nicolas,

Footnote to Re: [Q] How would you do method extensions (sometimes called class extensions) in Ruby? (was: Fwd: Thoughts from an outsider)
8767 [gaelli@em rg] ...

Thread#kill! patch
8768 [ruby-core@wh] Wow, MenTaLguY has worked up a patch[1] to ruby-1.8.5 and ruby_1_8 branch for
+ 8772 [mental@ry ia] for
+ 8809 [matz@ru y- a] Is it possible for you to patch both HEAD and ruby_1_8?

[PATCH] socks_init in ruby/ext/socket/socket.c
8770 [sven@c3 2. e] This is a multi-part message in MIME format.
8771 [matz@ru y- a] I merged it into HEAD.  Thank you.

suprising benchmark
8774 [neleai@se na] I wondered what gives you optimalization
8775 [drbrain@se m] Something is wrong with your benchmark.
8781 [neleai@se na] No just statistical error.

Literal inconsistency
8787 [calamitates@] irb(main):014:0> 2.times { puts "hello".object_id }
+ 8788 [matz@ru y- a] Since Ruby strings are mutable, string literals creates new string
+ 8789 [chneukirchen] I don't see the inconsistency.  Regexps are immutable anyway, so this
  8790 [calamitates@] 2.times do
  8791 [shyouhei@ic ] The term "immutable" is not clearly defined in Ruby.  You may surprise
  8792 [calamitates@] I know, I've tried lots of these things when I was learning Ruby, like
  + 8793 [now@bi wi se] Like String#<method>!s.
  + 8800 [matju@ar en ] ...

8794 [neleai@se na] I want ask how look integration of faster bignums.
+ 8795 [matz@ru y- a] Right.  I will merge your patch.  Thank you!
+ 8801 [matju@ar en ] ...
  8819 [neleai@se na] I thought that assembler has something like MULC which like ADC handles
  + 8822 [nobu@ru y- a] You should use SIGNED_VALUE.
  | 8825 [neleai@se na] Assembler IMUL usualy result in twice large number stored at two registers. But it's not crossplatform. GAS manual doesnt mention amd64 only ia-64. Don't know how with other platforms.
  + 8882 [matju@ar en ] ...

[ ruby-Bugs-5627 ] gem_server fails with getnameinfo: ai_family not supported
8796 [noreply@ru y] Bugs item #5627, was opened at 2006-09-05 00:48
8797 [drbrain@se m] On Sep 4, 2006, at 10:48 PM, <noreply@rubyforge.org>

okay, threading & sandbox r70 -- the latest patch
8798 [ruby-core@wh] ...
+ 8808 [matz@ru y- a] Commit it both HEAD and 1.8, please.
| 8811 [ruby-core@wh] I'll be offline for a few days, I will commit patches on Monday at the latest.
| 8812 [matz@ru y- a] It's unavoidable when we expose thread_t.
| 8839 [ruby-core@wh] Very good.  I am still away, in a few more days then.
+ 8917 [nobu@ru y- a] Sorry, I was aware of the problem but have forgotten to mention this.
  8923 [ruby-core@wh] Should this change in node.h?  Or should I move to ruby.h or, perhaps, a new
  8925 [nobu@ru y- a] Moving the separated include after the macro is defined?
  8939 [usa@ga ba ec] Some people are complaining that they cannot compile ruby on

[PATCH] WEBrick::Cookie support for multiple cookies per set-cookie
8802 [aaron_patter] WEBrick's cookie class has a method for parsing Set-Cookie headers, but
8805 [gotoyuzo@no ] Thanks. I've applied the patch, with some modifications.
8824 [aaron_patter] Cool.  Thanks!

Fix: no black boxes in rdoc diagrams anymore
8803 [flori@ni e. ] ...

[PATCH] Emacs ruby-mode: here-doc indentation & more
8804 [llasram@gm i] ...
8854 [matz@ru y- a] Thank you for the patch.

loginfo during check-in
8806 [ser@ge ma e-] charset="utf-8"
8807 [matz@ru y- a] It's just a message, just ignore it.

Server maintenance
8810 [shugo@ru y- ] Services except mailinglists will be down for a server maintenance on

[ ruby-Bugs-5701 ] ruby-1.8.5 loads fox16.so more than once
8813 [noreply@ru y] Bugs item #5701, was opened at 2006-09-08 20:59
8814 [drbrain@se m] What do we need to do to reproduce this?
8820 [lyle.johnson] Since FXRuby is so difficult to build, I put together a dead simple
8823 [nobu@ru y- a] I guess that I happened to fix it yesterday.