6721-7089 subjects 7169-7394

^ Ruby Version numbering.
6966 [hgs dmu.ac.u] This is in response to Ruby-Core:06938, but see Ruby-Core:06956 for

^ Unable to remap
6971 [vanek acd.ne] If this is the wrong ML please point the way.

^ getpasswd
6973 [lists bertra] this is not actually the most beautiful piece of code I ever
+ 6975 [nobuyoshi.na] Hmmm, get{pw,gr}*_r() family aren't actually reentrant at all.
+ 6976 [meta pobox.c] I think the ideal thing would be to define an OS-independent
  6978 [djberg96 yah] See sys-admin, on the RAA.

^ Ruby on Windows
6974 [tmorgan99 gm] I am building Ruby on Windows, but I am running into an issue.

^ ruby 1.8.4 preview3
6979 [matz ruby-la] I have just put 1.8.4 preview3 on the server.
+ 6980 [kailden gmai] AIX alpha 3 5 ------------
| 6984 [kailden gmai] I was able to fix the remaining failures by changing sha2.c:70 to
| + 6985 [meta pobox.c] I noticed when writing my Rdoc for bigdecimal that the Init_ function in
| + 6989 [matz ruby-la] Thank you.  A patch will be applied.
| + 7030 [kailden gmai] It appears that having unistd.h included on my AIX system causes the
+ 6988 [chneukirchen] gcc version 3.3 20030304 (Apple Computer, Inc. build 1666)
  6990 [matz ruby-la] The latter one will be merged.  The former one (drb.rb) needs to be
  6993 [chneukirchen] Then at least you should ensure that even if the DRb test fails,

^ ruby 1.8.4 preview3
6981 [wwalker bybe] Anything I can do to help debug this?  I'm pretty new to ruby, but a
6982 [wwalker bybe] Ignore that.  I found that I'm running a known broken version of gdbm.

^ [ ruby-Bugs-3090 ] Test failure in 1.8.4preview3  [ test_shifting_size(TestLogDevice) ]
6987 [noreply ruby] Bugs item #3090, was opened at 23/12/2005 15:05

^ Namespace pollution due to YAML in 1.8.3/1.8.4
6994 [chneukirchen] I just moved to 1.8.4 and noticed it broke my main Ruby application,

^ 1.8.4 YAML bug?
6995 [vjoel path.b] This typo seems not to have been fixed since the discussion in

^ Problems building 1.8.4 with VS8 C++ Express Edition (cl 14.00)
6996 [halostatue g] Visual Studio C++ 2005 Express Edition (VS 8.0)
+ 6997 [usa garbagec] Does this patch help you?
| 6999 [halostatue g] Partially. Bizarrely, the Makefile patch that I gave only suppresses
| + 7000 [tmorgan99 gm] I have not found this extern. on my system.
| | 7004 [ocean m2.ccs] (2005/12/27 14:16)
| | 7021 [tmorgan99 gm] I have replaced the config/makefile setup for Ruby using C++ Express, and I
| | 7024 [halostatue g] I do not believe that this is necessary. To compile with Visual Studio
| | 7031 [tmorgan99 gm] I am using PSDK, (or so I believe.  I must confess, it is a chore puzzling
| | 7032 [halostatue g] @echo off
| + 7002 [usa garbagec] I have two versions of msvcr80.dll.
+ 6998 [nobuyoshi.na] If we can get rid of wrong optimization by -Oy- with former
  + 7001 [usa garbagec] Defining these macros in config.h is not good idea because
  | 7003 [nobuyoshi.na] Right, I found missing/file.h included from file.c causes a
  | 7005 [halostatue g] Unless -G6 causes problems with the "real" version of VS 2005, this is
  | 7006 [usa garbagec] Ah, I'm using comercial version of VS2005, and it reports same
  | 7007 [halostatue g] Fair enough. I'm still stuck, but it appears to be in openssl this
  | 7012 [usa garbagec] I've forgotten another patch which is already applied to HEAD.
  | 7023 [halostatue g] Thank you. This was sufficient to complete the build for Ruby.
  | 7025 [usa garbagec] Not yet, but I'll commit the patches for win32/win32.c and
  + 7040 [ocean m2.ccs] I have problem after OPTFLAGS was changed. (VisualC++6 on Win2000SP4)

^ [ ruby-Bugs-3115 ] Install fails to create directories
7008 [noreply ruby] Bugs item #3115, was opened at 2005-12-28 05:00
7010 [nobuyoshi.na] Seems `ruby' didn't get built.  Have you done `make' before
7014 [meta pobox.c] Surely the build target should be listed as a dependency of the install
7015 [matz ruby-la] It might make you compile the whole things, which may include nasty
7016 [matju arteng] How is the current "make install" any more secure than that?
7017 [nobuyoshi.na] I don't think it's nice idea too.  Instead, what about other

^ [ ruby-Bugs-3116 ] Install failed: fileutils.rb:832:in `stat': No such file or directory - ruby (Errno::ENOENT)
7009 [noreply ruby] Bugs item #3116, was opened at 2005-12-28 05:10

^ [ ruby-Bugs-3119 ] RDoc HTML generator doesnt close img tags
7011 [noreply ruby] Bugs item #3119, was opened at 2005-12-28 13:32
7027 [pabs pablotr] See forthcoming ruby-core patch.

^ [ ruby-Bugs-3132 ] no download or update avilable
7013 [noreply ruby] Bugs item #3132, was opened at 28/12/2005 18:08

^ [PATCH] More robust content-type check: lib/xmlrpc/client.rb
7018 [andy.loafoe ] Trivial patch which dowcases the content-type before compare. This

^ Math.sqrt(-1) inconsistancies
7019 [jeffrey.dik ] I've noticed some inconsistancies with Math.sqrt(-1).  Here's what the
7026 [djberg96 yah] Documentation bug.  It calls rb_sys_fail() internally,

^ [ ruby-Bugs-3146 ] Rails installation complete?
7020 [noreply ruby] Bugs item #3146, was opened at 2005-12-29 12:43
7022 [tmorgan99 gm] id=3D426

^ [PATCH] Ruby 1.8.4 RDoc HTML Cleanups and HTML Language Support
7028 [pabs pablotr] Attached is a patch against Ruby 1.8.4 which cleans up the HTML output
7091 [ocean m2.ccs] I have reviewed your patch, and have noticed two things.
7132 [pabs pablotr] I think most browsers ignore attributes that they don't understand.  So,
7134 [ocean m2.ccs] Yes, and most browsers can parse <img ....> without end tag.
7141 [pabs pablotr] Invalid elements are not the anywhere near the same thing as invalid
7143 [ocean m2.ccs] I recomend you to check html with W3C validater. Upload html and click [validate]

^ [ ruby-Bugs-3149 ] deadlock on Generator#next?
7029 [noreply ruby] Bugs item #3149, was opened at 2005-12-30 11:24

^ ruby 1.8.3+, configure and install-sh
7033 [djberg96 yah] Based on ruby-talk:172798 and some offline

^ [ ruby-Bugs-3155 ] Dir#seek segfaults on Windows XP
7035 [noreply ruby] Bugs item #3155, was opened at 2005-12-30 20:23
7037 [ocean m2.ccs] You should pass postion which Dir#tell returns. Otherwise, behavior is unspecified.
7038 [ocean m2.ccs] This is patch. This is still hot, so I'll commit after my brain
7041 [djberg96 yah] Excellent, thank you!

^ YAML bug - Ruby 1.8.3
7036 [jlsysinc all] $ cat testyaml.rb
7092 [ocean m2.ccs] (2005/12/31 13:35)
7096 [jlsysinc all] Correct.  I did some research after posting this and concluded it's not a

^ [ ruby-Bugs-3160 ] Call to NoMethodError#message hangs ruby
7039 [noreply ruby] Bugs item #3160, was opened at 2006-01-01 09:53
7042 [ocean m2.ccs] (2006/01/01 19:04)

^ RUBYOPT versioning?
7043 [caleb aei-te] Is there any chance we could get a RUBYOPT variable that's specific
+ 7044 [halostatue g] I like this idea, but ...
| 7045 [caleb aei-te] Not sure - I'm open to suggestion.  At first glance, I think having
+ 7046 [drbrain segm] alias ruby19='ruby19 -rubygems'

^ RDoc patches for BigDecimal in Ruby CVS
7050 [meta pobox.c] Now that 1.8.4 is out and the initial flurry of problem reports has died
7131 [gsinclair gm] More than a week and no response, so... "bump".
7136 [ocean m2.ccs] (2006/01/13 13:14)
7201 [meta pobox.c] I've tried three times. His mail server bounces all my e-mail as spam.

^ [ ruby-Bugs-3175 ] enum.find with ifnone calls ifnone process even if find succeeds
7051 [noreply ruby] Bugs item #3175, was opened at 2006-01-03 19:49
7052 [nobu ruby-la] Yes.

^ More on VC++ 2005
7055 [halostatue g] Okay. I've got Ruby compiling. I'm attempting to get everything in
7058 [nobuyoshi.na] Where's that mt command?  I've installed VC++ 2005 Express
+ 7060 [usa garbagec] Regards.
| 7062 [nobuyoshi.na] I see.  Is it true for .so files?
| 7063 [usa garbagec] It seems that DLL does not need .manifest if base .exe uses
| 7064 [nobuyoshi.na] OK, Makefile.sub only.  Since the format of config.status
+ 7066 [halostatue g] Hm. How would I do that, assuming that I'm starting from win32/configure.bat ?
  7069 [halostatue g] The patches look good on 1.8.4.
  7085 [nobuyoshi.na] Thank you.
  7086 [halostatue g] Mmmm. I think embedders who want to use VS2005 or later to compile
  + 7087 [usa garbagec] Hmm, you can see my scripts at
  + 7088 [ml.chibbs gm] This probably isn't relevant to your current line of inquiry, but I'm
  | 7090 [halostatue g] You're right; it isn't directly relevant. However, it *will*
  | 7095 [ml.chibbs gm] Its entirely possible that I may pick up on your fine work in the
  + 7094 [nobuyoshi.na] Still I'm not sure what to be documented.  .manifest files won't
    7174 [halostatue g] [patch elided]
    7175 [nobuyoshi.na] 1.9.  The patch for 1.8 is attached.

^ Newbie question
7056 [mtscolony gm] I just started learning Ruby. I have good experience in Java and Perl and
7059 [james graypr] You will probably have a lot more luck with answers if you post this
7061 [mtscolony gm] thanks, i will post it there.

^ 64-bit Solaris READ_DATA_PENDING Revisited
7057 [steven lumos] I have to start off with an apology.  Over a year ago, I submitted a
7065 [akr m17n.org] It works for me too on Solaris 8 on SPARC with gcc 4.1.0
7067 [ville.mattil] This patch is quite clever! I've been struggling with 64 bit issue on
7068 [Daniel.Berge] You could probably get it into OpenSolaris.  Dunno about Solaris proper,
7106 [ville.mattil] here is the updated patch. With ChangeLog entry. I've tested this with
7107 [steven lumos] I was thinking something similar in [ruby-core:6983] although I was on
7116 [ville.mattil] word

^ [ ruby-Bugs-3195 ] ruby interpreter crashes with invalid input
7070 [noreply ruby] Bugs item #3195, was opened at 2006-01-06 20:22
7071 [nobu ruby-la] Thank you for the report, fixed now.

^ YAML and '---' doc separatore
7072 [ara.t.howard] it seems this came up and nothing was resolved...
7073 [james graypr] Just the opposite actually.  A late patch to 1.8.4 was allowed by

^ Wilderness: Version 2.1 of IDBG - Fixes for Multiple File Debugging
7074 [ruby-core ha] I will post the latest 'tar' (idbg_2.1_tar.tgz) on

^ [ ruby-Bugs-3211 ] typo in webrick
7075 [noreply ruby] Bugs item #3211, was opened at 2006-01-08 21:21

^ [ ruby-Bugs-3212 ] Readline does not provide enough context to the completion_proc
7076 [noreply ruby] Bugs item #3212, was opened at 2006-01-09 17:44

^ Fast Ruby Debugger?
7077 [joe.moore gm] After developing in Java for years in IDEs I'm starving for a decent ruby
7081 [ryand-ruby z] grab ZenHacks and check out zendebug. You can install it and all of

^ CRC - a proof-of-concept Ruby compiler
7078 [chucky dtek.] During the last few months me and Andreas Farre have been doing our
7079 [ncannasse mo] Some people are interested in doing the same thing for Neko
7080 [chucky dtek.] Nicolas,
+ 7082 [chneukirchen] While we are at it, did you consider llvm?  llvm.cs.uiuc.edu
| 7083 [d00farre dte] Honestly no. We were more concerned about evaluating how well C-- would
+ 7093 [ncannasse mo] Yes I knew about C-- way of doing things, I just mistake my questions by

^ mathn: ugly warnings
7084 [hadmut danis] I was just looking for the origin of ugly warnings in a ruby
7097 [Daniel.Berge] I agree.  It looks like mathn, rational, complex and matrix could all use some

^ Design contracts and refactoring (was Re: mathn: ugly warnings)
7098 [meta pobox.c] Actually, it's worse than that. There's no documentation of what
+ 7099 [deanwampler ] Let me suggest an XP-style alternative; make thorough unit tests
+ 7118 [meta pobox.c] Unit tests are not an alternative. They are an additional requirement.
| + 7119 [deanwampler ] Sorry, but I think you don't understand the purpose of Test Driven
| + 7121 [matju arteng] I find unit-tests to be often decomposable like this. Start with something
| + 7123 [matju arteng] Here's another way to factor unit-tests that I haven't mentioned in the
| | 7124 [ara.t.howard] first of all, included no docs - only tests and a samples/ dir.  the samples
| | 7125 [ruby jamesbr] But what will the user see from ri and rdoc?
| | 7126 [ara.t.howard] nothing.  i've found myself using them less and less - at this point i use
| | 7127 [ruby jamesbr] For a certain subset of Ruby developers.
| | 7133 [matju arteng] In order to entrench the tests-as-documentation habit firmly in the Ruby
| | 7140 [ara.t.howard] oh yeah - i forgot to mention that the README is auto generated using the
| + 7128 [matju arteng] Let's follow the argument of both of you to the end.
|   7130 [deanwampler ] I'm not sure I follow you exactly, but I think I more or less agree ;)
+ 7142 [meta pobox.c] a) In the real world, the requirements do not constitute the entirety of
  + 7144 [deanwampler ] The XP philosophy is that if you implement more functionality than is
  + 7145 [hgs dmu.ac.u] You are not testing number theory, you are testing software.  All
    + 7147 [djberg96 yah] While this conversation is interesting, I don't feel
    | 7148 [matju arteng] hehehe, if you want to play that card, would you please tell me how the
    + 7149 [matju arteng] You are testing "implementation" in general, not software in particular.
      7160 [meta pobox.c] *Dean Wampler *<deanwampler gmail.com>
      7161 [meta pobox.c] Yes, and how do you do that?

^ core dump with ruby 1.9.0 (2006-01-10) and bdb-0.5.8
7100 [akr m17n.org] I found following test script dumps core.
7101 [decoux moulo] I always thought that it's just a madness to run a test suite in the
7102 [akr m17n.org] I see.  I reduced the test script.  Now it doesn't use
7103 [decoux moulo] This is this that it don't like. I'll see what I can do.
+ 7104 [ocean m2.ccs] (2006/01/11 18:53)
+ 7108 [akr m17n.org] "GC.stress = true" makes GC related problem more

^ [ ruby-Bugs-3237 ] win32ole bug  'method missing': DisableAutoMacros (WIN32OLERuntimeError)
7105 [noreply ruby] Bugs item #3237, was opened at 2006-01-11 13:13

^ Calling flock with block?
7109 [lists bertra] File.open "somefile", "a" do |f|
7111 [matju arteng] class File
+ 7112 [vjoel path.b] It seems likely to me that a lot of people have implemented exactly that
| 7113 [ara.t.howard] i sure have.  on the otherhand it is generally redundant since closing the
| 7115 [lists bertra] As there's no use of nested locking it should be inhibited
| 7117 [ara.t.howard] well...  i'm not sure that's quite true.  the example above could be quite
+ 7114 [lists bertra] I _did_ use `rb_ensure' in my proposal. This was the tricky

^ 1.9 dl w/o cdecl/stdcall
7110 [steven lumos] What needs to be done to support a compiler that doesn't have cdecl,

^ [ ruby-Bugs-3245 ] irb attempts to escape '\c' inside single quotes
7122 [noreply ruby] Bugs item #3245, was opened at 2006-01-12 13:41

^ YAML.load({[]=>""}.to_yaml)
7129 [akr m17n.org] I found that current YAML doesn't round trip {[]=>""}.
7135 [ocean m2.ccs] ruby 1.8.4 (2005-12-24) [i386-bccwin32]
7154 [ocean m2.ccs] How about this? Function name is a bit long though.
7155 [akr m17n.org] Although I'm not test it yet, I recommend the test script in
7156 [ocean m2.ccs] After changing test script a bit, ("self.send m" => "self.funcall m") looks working.
7157 [akr m17n.org] with GC.stress = true ?
7158 [ocean m2.ccs] No, I didn't try it. But anway, 100 times was not enough. After increased it to million times,
7159 [ocean m2.ccs] This is because YAML is not handling CR and LF properly.
7163 [ocean m2.ccs] I have commited but

^ [ ruby-Bugs-3247 ] RDoc: cannot create diagram if only dir.c is specified
7137 [noreply ruby] Bugs item #3247, was opened at 2006-01-13 10:47

^ [PATCH]  solaris 10  isinf and ruby_setenv fixes
7138 [ville.mattil] Here is a patch that fixes two issues for me.
+ 7151 [akr m17n.org] I found a description about isinf in autoconf.texi.
| 7181 [ville.mattil] I haven't seen these fixes in CVS.  Is there problem with these, I can
+ 7270 [ocean m2.ccs] You experienced actual memory corruption? Or just assumed so?
  7271 [ville.mattil] test_bracket(TestEnv): ./ruby/test_env.rb:19:in `[]=': failed to allocate
  7273 [ocean m2.ccs] Hmm, I also think it's better to use setenv(3). But for curiousty,
  7274 [ville.mattil] With this patch it fails similarily. E.g either no memory exception or
  7275 [ocean m2.ccs] charset="ISO-8859-1"
  7276 [ocean m2.ccs] charset="ISO-8859-1"
  7277 [ville.mattil] This patch works correctly for me. However there is still the isinf trouble
  7278 [ocean m2.ccs] Well, should I commit your patch [ruby-core:7273],
  7279 [matju arteng] hey, i'm really curious, what's @{ and @}. I didn't know that @ was a
  + 7280 [ville.mattil] trouble
  | 7283 [ocean m2.ccs] I'm not sure I understand [ruby-core:7151] well, but when I tried following patch
  | 7294 [ville.mattil] following
  + 7282 [ocean m2.ccs] (2006/02/02 00:43)

^ Building fat binaries on Mac OS X
7139 [neil hakubi.] I'm not sure if this is the right list for this, but I had some trouble

^ [PATCH] Additional RDoc HTML Cleanups (was "Re: [PATCH] Ruby 1.8.4 RDoc HTML Cleanups ...")
7146 [pabs pablotr] [snipped]
7179 [pabs pablotr] This is the official six-day bump on this follow-up patch.

^ Array#[] with float indices
7150 [matju arteng] /usr/bin/ruby -e 'p RUBY_VERSION,[42][0.0]'
7152 [matju arteng] nevermind, fixed in a more recent 1.9.0.

^ Regular Expression Support, Defined by Unicode UTS#18 Requirements
7153 [wonado donne] I have a question regarding future Ruby versions, that is the versions 2.x
7164 [sndgk393 ybb] **> I have a question regarding future Ruby versions, that is the
7165 [wonado donne] Thank you for the information. It is very helpful to know the properties of

^ FileUtils.mv does not unlink source file when moving over filesystem boundary
7162 [pav FreeBSD.] I found a bug in ruby 1.8.4 in FileUtils.mv() function. When moving a
7166 [ara.t.howard] eeks!  this is terrible even with the fix.  emulating mv with copy will
7167 [lukfugl gmai] Am I mistaken, or isn't that how the OS handles move across filesystem
7168 [ara.t.howard] if you mean that programs like 'mv' work this way then you are quite right.