7585-12269 subjects 7981-8420

Another class of problem
7775 [rubypate@gm ] (Well, it looks like someone cleaned up a possible defect in the

possible defect in ext/openssl/ossl_ans1.c (with fix)
7785 [rubypate@gm ] 768  		    VALUE klass = *ossl_asn1_info[tag].klass;
7786 [ville.mattil] I think this is false positive.
+ 7787 [rubypate@gm ] [reasonable looking explanation deleted]
+ 7789 [lukfugl@gm i] But lines 778 and 779 are never executed if the condition in line 770
  7790 [ville.mattil] if(!rb_obj_is_kind_of(value, rb_cArray) && tag ==
  7793 [steve@wa ts ] Regardless, I'd just like to throw in that initialization is never a
  7798 [gotoyuzo@no ] Yes. I think flag should be initialized, but more checks are

[PATCH] use -shared to gcc for shared link Solaris
7791 [andrew.c.mor] ...

Another unitialized variable defect
7792 [rubypate@gm ] (I hope I'm finally getting the hang of this.)

Right way to handle memory allocation failures (related to fix for array.c issue)?
7795 [dominique.br] To fix the integer wrap in rb_array_fill, there is a simple check that

[ ruby-Bugs-4284 ] Float near Float::MIN
7796 [noreply@ru y] Bugs item #4284, was opened at 2006-04-27 21:14
7797 [matz@ru y- a] The built-in strtod() function cut off exponent digits too early.

Patch: code-cleanup (k&r style)
7799 [stefan@hu hn] ...
7801 [matz@ru y- a] Thank you I will apply the patch.
7802 [decoux@mo lo] ruby-core:[07764]
7803 [matz@ru y- a] Oh, that one.  I applied it but not yet committed.
7804 [stefan@hu hn] Any timeframe to see these one or both of the two patches commited to cvs?
7805 [matz@ru y- a] I am in the middle of huge modification of CVS trunk, so that it would

Net::IMAP#disconnect + SSL hangs
7800 [drbrain@se m] When calling Net::IMAP#disconnect when using an SSL connection the

[ ruby-Bugs-4336 ] Ruby fails to parse valid wsdl
7806 [noreply@ru y] Bugs item #4336, was opened at 2006-05-02 12:55

[ ruby-Bugs-4339 ] Ruby fails float 80 test in stable tarball (2 May 2006) when compiled with Mingw
7807 [noreply@ru y] Bugs item #4339, was opened at 2006-05-02 19:56
7817 [matz@ru y- a] I think I fixed it.  Try it again with the latest stable snapshot,

one bug and one false positive defect in yaml2byte.c (we think)
7808 [rubypate@gm ] We've been looking at coverity defects during our hacking night. there are two

uninit bug in yaml/emitter.c
7809 [rubypate@gm ] During our hacking night, we also looked at an UNINIT bug in yaml/emitter.c
7816 [matz@ru y- a] I think you're write.  I will merge the patch.  I hope _why could
7826 [ruby-core@wh] I will do now.  Thankyou, matz & pat.
7830 [vjoel@pa h. ] Are the bugs/patches at http://code.whytheluckystiff.net/syck/report/1

patch for UNINIT error in array.c
7810 [rubypate@gm ] Along with the unitialized beg, we found a len which doesn't appear to be
7812 [warrenbrown@] If you're going to initialize beg and len, you might as well

[ ruby-Bugs-4341 ] Sather-like iterators
7811 [noreply@ru y] Bugs item #4341, was opened at 2006-05-03 05:28

[ ruby-Bugs-4344 ] :!~ not a symbol
7813 [noreply@ru y] Bugs item #4344, was opened at 2006-05-03 17:41
7815 [matz@ru y- a] Yes, and it's because !~ (along with &&, ||, etc.) is a part of
7827 [ruby-core@wh] Hey, matz.  What is the use of the unary !@ operator?  Can it actually
7828 [matz@ru y- a] No. Hmm, since it's not callable, nor re-definable, it shouldn't be

[ ruby-Bugs-4345 ] parsedate doesn't work with GMT dates
7814 [noreply@ru y] Bugs item #4345, was opened at 2006-05-03 08:26

(security-related) patch to ALLOC macros to prevent integer overflow bugs
7818 [dominique.br] While fixing the integer overflow in rb_ary_fill(), it occurred to me
+ 7819 [hgs@dm .a .u] (lines got folded in this quote)
+ 7820 [matz@ru y- a] 1.9.0 has code against integer overflow attack.  If someone could
  7821 [dominique.br] Great! I will try to look it over when I get a chance. As an aside, I

[ ruby-Bugs-4361 ] StringScanner#scan() fails on zero-length matches at eos
7822 [noreply@ru y] Bugs item #4361, was opened at 2006-05-04 21:52

Coverity: Patch/Fix CID 17 OVERRUN_STATIC File: ruby/signal.c
7823 [tewk@te k. o] This is a simple out of bounds array access error.
+ 7824 [ville.mattil] I agree that this is a valid bug. When I checked this, I couldn't find
+ 7825 [matz@ru y- a] I did.  Thank you.

[ ruby-Bugs-4365 ] Ruby fails float 80 test in stable tarball (4 May 2006), on multiple platforms
7829 [noreply@ru y] Bugs item #4365, was opened at 2006-05-05 18:51

Resolv::DNS and IPv6 nameservers
7831 [stephan@sp c] UDPSocket requires address family given on initialization, so it's not
7836 [akr@m1 n. rg] Thank you for the patch.  I'll incorporate it.
7838 [stephan@sp c] How about... "Use AF_INET6 for nameservers containing colons"

Re: [python_tutor] Welcome to our (ruby-core ML)         You are added automatically
7832 [fcoimbra@vo ] help

[BUG] segfault on Proc#call after setting a trace_func
7833 [ mfp@ac .o g] $ cat bug2.rb
7834 [decoux@mo lo] Not sure, but it seems to be in proc_invoke()
7835 [matz@ru y- a] That's it.  Thank you.  Here's the patch to fix this.
7911 [nobu@ru y- a] what state is this patch?
7912 [matz@ru y- a] Applied to the local copy of the repository.  Waiting to be
7927 [nobu@ru y- a] OK, committed, with volatile modifier as older versions.

Coverity CID 14 15 Patch
7837 [tewk@te k. o] ...
7840 [ville.mattil] how about

Coverity CID 6 7  Appear to be false positives.
7839 [tewk@te k. o] CID 6 appears to be a false positive due to the sequence of assignments
7841 [ville.mattil] I think you analysis is correct.  Could you put some comment to these CIDs

[ ruby-Bugs-4382 ] XMLRPC support extended iso8601 DateTime Format
7842 [noreply@ru y] Bugs item #4382, was opened at 2006-05-07 21:51

Possible YAMl bug in 1.8.4
7843 [damphyr@fr e] OK, while parsing the td2 data from the ruby-lang website we stumbled on
7846 [ml.chibbs@gm] ...
+ 7849 [damphyr@fr e] Sorry for not quoting, but my webmail spits Base64 :)
| 7850 [james@gr yp ] I use Mac OS X.
+ 7852 [mental@ry ia] Hmm.  Assuming that's the case, are fixed snapshots available anywhere else, or is this the kind of thing I have to jump in and pull out the relevent bits from a cvs diff?
| 7853 [ml.chibbs@gm] ...
| 7854 [mental@ry ia] Ah ... no ... see, I need to do Mongrel + YAML-y things using Ruby 1.8.4 on PPC Linux.  Win32 binaries aren't terribly helpful.  Is there a more specific timestamp available for the snapshot it was built from?
+ 7855 [ mfp@ac .o g] $ ruby -v yaml-bug.rb
  7857 [ml.chibbs@gm] ...

[ ruby-Bugs-4391 ] buggy tr_s in jcode
7844 [noreply@ru y] Bugs item #4391, was opened at 2006-05-08 22:34
7848 [matz@ru y- a] It's fixed in the latest stable snapshot.

Method call syntax
7845 [meta@po ox c] "Omission of parentheses around method arguments may lead to unexpected
7847 [matz@ru y- a] Omission of parentheses around method arguments for method calls in
+ 7851 [james@gr yp ] But you are still planning to allow them to be dropped in the non-
+ 7856 [lists@be tr ] I wouldn't mind to take account of a precedence rule, at

Ruby threads working with native threads
7858 [garbagecat10] ...
7859 [drbrain@se m] Why can't you call rb_thread_select?
7860 [garbagecat10] ...

Confirmed: YAML bug in 1.8.4
7861 [ruby-core@wh] All Ruby HEAD patches have been backported to Syck CVS yesterday.  Also,

650M process, huge stack trace
7862 [evan@fa li g] ...
7863 [tjw@om ig ou] What OS is this on?  If it is OS X, you could try using vmmap to
7864 [evan@fa li g] It's on linux 2.6. A modified debian box.

[ ruby-Bugs-4457 ] Strange interactions between Struct and 'pp'
7865 [noreply@ru y] Bugs item #4457, was opened at 2006-05-12 17:13
+ 7866 [drbrain@se m] $ cvs diff lib/pp.rb
| 7871 [florgro@gm i] irb(main):001:0> Struct.new(:class).new.class
+ 7869 [akr@m1 n. rg] Thank you for the report.
  7907 [tom@in oe he] I also received an additional note from Gustav Munkby, who reported the

[ ruby-Patches-4463 ] performance enhancement for OpenSSL::Cipher::Cipher.update
7867 [noreply@ru y] Patches item #4463, was opened at 2006-05-13 06:55

[ ruby-Patches-4464 ] Better error messages in OpenSSL::SSL.[accept, connect]
7868 [noreply@ru y] Patches item #4464, was opened at 2006-05-13 06:58

[ ruby-Bugs-4467 ] RDoc looks for doc headers at function prototypes
7870 [noreply@ru y] Bugs item #4467, was opened at 2006-05-13 13:50

Nonblocking socket-connect
7872 [garbagecat10] ...
7873 [akr@m1 n. rg] I think it's good idea to introduce new nonblocking methods.
7874 [garbagecat10] ...
7875 [matz@ru y- a] require 'io/nonblock' would do what you proposed.  But I'm not
7876 [garbagecat10] Well, it's ok then. I'm comfortable adding in the nonblocking
7877 [matz@ru y- a] Note that I'm not against for non-blocking connect.  I just oppose to
+ 7878 [Daniel.Berge] A separate AIO class as Python does?
| 7879 [garbagecat10] I have to admit, I seriously dislike AIO. In my experience, it only
+ 7880 [garbagecat10] How about Socket#nbconnect and Socket#nbaccept?
| 7890 [vladgalu@gm ] I don't know the usual meaning of the exclamation mark, but I think
| + 7893 [garbagecat10] ...
| | 7894 [halostatue@g] #accept_nonblock, #connect_nonblock, and #recvfrom_nonblock would be
| + 7898 [drbrain@se m] In ruby the ! usually means the method modifies the receiver where
|   7900 [ara.t.howard] exit!
+ 7913 [akr@m1 n. rg] Do you find an acceptable name?
  7914 [matz@ru y- a] *_nonblock sound better than others from my POV.  Any opinion?
  + 7915 [sroberts@un ] I like the _nonblock best of the 4.
  | 7916 [garbagecat10] ...
  + 7917 [akr@m1 n. rg] I think [ruby-talk:113813] should be reverted: accept method
    7918 [matz@ru y- a] OK, with the nonblock methods, it is useless now.  Can you apply these

[ ruby-Bugs-4491 ] Segfault on x86_64 when built with -O0 in CFLAGS
7881 [noreply@ru y] Bugs item #4491, was opened at 2006-05-16 12:46
7886 [decoux@mo lo] Well not really a problem with -O0 (i.e. it can possible to have a version
7887 [matz@ru y- a] Thank you, Guy.  Does this patch could solve the problem?
7888 [matz@ru y- a] Oops, here's the patch.
7889 [decoux@mo lo] yes, I think.

reproducible bug in DRb on OSX
7882 [cremes.devli] I've been tearing my hair out the last few days trying to track down
7883 [drbrain@se m] IPv6 can be the bane of your existence.  Setting up proper name
7884 [cremes.devli] Adding "allow ::1" also makes this work. So, it appears to be a ipv6
7885 [drbrain@se m] is not a valid IPv6 address.  This was a bug in your code
7897 [sroberts@un ] It might be useful if DRb could be told what IP versions to use with

[ ruby-Bugs-4498 ] raise can create custom exception objects which aren't properly initialized
7891 [noreply@ru y] Bugs item #4498, was opened at 2006-05-17 11:36
7892 [decoux@mo lo] I've not understood

[ ruby-Bugs-4500 ] Webrick: Missing -c parameter forces utf-8 even if with code page specified in meta tag
7895 [noreply@ru y] Bugs item #4500, was opened at 2006-05-17 16:44
7899 [drbrain@se m] WEBrick doesn't parse what it serves to figure out content types, nor

Fwd: Bug: Time#-(1e-6) doesn't substract one microsecond.
7896 [dave@bu t. d] ...

String#nstrip ?
7901 [Daniel.Berge] When using the Win32API package, I often have to resort to this idiom to get a
+ 7902 [Daniel.Berge] # nullbench.rb
+ 7903 [vjoel@pa h. ] Why won't slice do the trick?
+ 7904 [ara.t.howard] regards.
  7905 [Daniel.Berge] Thanks all.  My benchmarks show the best solutions at about ~4x faster.
  + 7906 [brabuhr@gm i] In the case that someone else may have this same need, I added a link
  + 7908 [angus@qu va ] STRING.unpack("Z*")[0]
    7910 [Daniel.Berge] A little slower than the regex solutions, but not much.  Good to know,

[PATCH] SCRIPT_LINES__ issue when loading a file more than once
7909 [ mfp@ac .o g] SCRIPT_LINES__ is an obscure feature very few people care about, but I happen
+ 7947 [ mfp@ac .o g] Does it work that way by design (WONTFIX) or is it just that nobody cares
| 7950 [matz@ru y- a] I care.  But I couldn't have time to investigate the patch.  Could you
| 7951 [ mfp@ac .o g] Of course, there's no urgency here and there are indeed more important things.
+ 7952 [nobu@ru y- a] I vote to (b) for HEAD too.
  7953 [matz@ru y- a] Do you?  But redundant lines in SCRIPT_LINES__ are not useful at all,
  7956 [nobu@ru y- a] I agree that they are unuseful, and (b) overwrites previous
  7957 [matz@ru y- a] I guess so.  Can it be a problem?
  7958 [nobu@ru y- a] I don't guess it would cause a problem.  By analogy with an

7919 [garbagecat10] ...

[ ruby-Bugs-4555 ] Range "max" method does not handle Infinity
7920 [noreply@ru y] Bugs item #4555, was opened at 2006-05-24 04:37
7921 [matz@ru y- a] It is supported by 1.9.  I'm not sure it should be backported to 1.8
7924 [erlercw@gm i] ...

loading rubygems at startup without RUBYOPT
7922 [nobu@ru y- a] By linking following extension library statically, rubygems.rb will be
8055 [nobu@ru y- a] This can achieve other hacks.

Nonblocking accept
7923 [garbagecat10] ...
7925 [akr@m1 n. rg] Since accept is for connection oriented sockets, UDP doesn't need it.
7926 [garbagecat10] ...

[Bug?] set_trace_func: binding has wrong self value for return events
7928 [florgro@gm i] ...
7963 [florgro@gm i] From the response to Mauricio's patch I understand that right now
7964 [matz@ru y- a] It is a bug, and expected output should be
7968 [florgro@gm i] Ah, hm. Any chance of having binding(n) where n=0 is the current binding
7969 [matz@ru y- a] I want to make sure that this is a totally different story from
7970 [florgro@gm i] Yup, that makes a lot of sense.

RDoc --force flag
7929 [vincent.four] ...

[ ruby-Bugs-4601 ] FOR SALE:nokia 8800,6600,n92,91,90,80,70 and Plas EDTV samsung at just $200
7930 [noreply@ru y] Bugs item #4601, was opened at 2006-05-29 17:35
7932 [tom@in oe he] I've suspended this user's RubyForge account, sorry for the spam.

[ ruby-Bugs-4605 ] Critical 1.9 mathn bug (require 'mathn' failure)
7931 [noreply@ru y] Bugs item #4605, was opened at 2006-05-29 18:40
7933 [matz@ru y- a] I will remove it soon.  Thank you.

[BUG] round trip failure with YAML; both ruby_1_8 and HEAD
7934 [ mfp@ac .o g] $ cat yamlbug.rb
7935 [mental@ry ia] orig = "hoge\n\nhoge"

Possible YAML error?
7936 [Ola.Bini@ki ] While trying to process the source index Gemspec with RbYAML I found
7937 [ruby-core@wh] Egads.  You're right.  For some reason the ampersand was listed twice in Syck's
7938 [Ola.Bini@ki ] No worries!

Including Ruby's .rb libraries in JRuby distribution
7939 [headius@he d] ...
7940 [matz@ru y- a] I don't think of any problem from library inclusion.  Thank you always
7941 [headius@he d] ...

[ ruby-Bugs-4634 ] faculty : segmentation violation
7942 [noreply@ru y] Bugs item #4634, was opened at 2006-06-02 10:37

typo in forwardable docs
7943 [jan.svitok@g] ...
7944 [matz@ru y- a] Thank you.  I will apply the patch.

Windows XP SP2 socket issues
7945 [dan.hatfield] ...
7965 [usa@ga ba ec] Can you explain it more?
7966 [dan.hatfield] ...

7946 [nitayj@gm il] ...
7948 [florgro@gm i] Hm, am I understanding you correctly in that you want a method that will
7949 [nitayj@gm il] ...
7954 [florgro@gm i] It's not added in 1.9 yet. I'm not even sure if I posted a RCR or RFE

Failing tests in ruby since 1.8.2
7955 [caleb@ae -t ] The following tests have been failing in Ruby for a long time, including
7959 [ara.t.howard] are you compiling as root?  if so be aware that the behaviour of the linker on
7960 [ville.mattil] thelinker on
7961 [ara.t.howard] from some time ago.  i've seen similar issues since then - sorry i cannot give

posixlock in Debian (fwd)
7962 [ara.t.howard] i've got quite a few emails from people who rely on the posixlock extension
7974 [meta@po ox c] For what it's worth, I second this. I needed a file lock the other

YAML booleans.
7967 [Ola.Bini@ki ] irb(main):017:0> YAML.load("a: n")
7971 [ruby-core@wh] This should be brought up on Yaml-core.  (And has been before.)  The spec is
7972 [Ola.Bini@ki ] Ah!

YAML options hash.
7973 [Ola.Bini@ki ] I found some people talking about this, but no answer. What's up with
7976 [ruby-core@wh] Would you mind putting these issues in the Syck tracker rather than on

[ ruby-Patches-4704 ] OpenSSL Elliptic Curve patch (please review)
7975 [noreply@ru y] Patches item #4704, was opened at 2006-06-08 04:04

[ ruby-Bugs-4722 ] Lighttpd/webrick binding errors
7977 [noreply@ru y] Bugs item #4722, was opened at 2006-06-09 18:02
7979 [drbrain@se m] On Jun 9, 2006, at 11:02 AM, <noreply@rubyforge.org>

Patch for Unix socket peer credentials
7978 [jfh@ci e. fl] ...
+ 7980 [sroberts@un ] I tried to do this on OS X recently using DGRAM AF_LOCAL sockets based
| 7988 [sroberts@un ] Yes, but credentials are (supposed) to be attached to each message with
+ 8004 [akr@m1 n. rg] I think it's good feature.
  8005 [jfh@ci e. fl] Ok -- I think I'm going to take Sam's advice and turn it into one
  + 8007 [sroberts@un ] It would be nice to get API feedback from more than 2 people though!
  + 8010 [akr@m1 n. rg] I like a hash: {:euid => euid, :egid => egid }
    8191 [jfh@ci e. fl] I'm ready to get back to this now.
    8193 [akr@fs j. rg] I think it's good enough.
    8212 [jfh@ci e. fl] ...
    + 8214 [drbrain@se m] Are you sure this is right?
    + 8216 [akr@fs j. rg] I think a struct is not appropriate representation in this
    + 8217 [nobu@ru y- a] have_library will append that macro automatically, if
      8257 [jfh@ci e. fl] ...
      + 8258 [drbrain@se m] Negative UID and GID are valid on some operating systems.
      | + 8259 [drbrain@se m] $ ruby -r etc -e 'p Etc.getpwnam("neguid")'
      | + 8260 [jfh@ci e. fl] Would it be better then to initialize the hash with nils at the beginning
      | | 8261 [drbrain@se m] I think so.
      | + 8295 [jfh@ci e. fl] ...
      |   + 8296 [drbrain@se m] On FreeBSD uid_t and gid_t are unsigned integers.  Searching google
      |   | 8297 [sroberts@un ] Other __UNIX__ operating systems? Are you sure?
      |   | 8298 [drbrain@se m] Yes, typo, I meant to write "signed integers".  But not pid_t, uid_t.
      |   + 8355 [akr@fs j. rg] An example of negative UID is on NFS: "nobody" is -2.  [RFC 1094]
      |     + 8384 [sroberts@un ] Whether allowed or not by the OS, returning -1 in the hash as :uid and
      |     + 8407 [jfh@ci e. fl] I'll work on another patch that addresses your issues when I can (I'm
      + 8263 [nobu@ru y- a] The wrong point is using have_library in order to check if the
        8292 [jfh@ci e. fl] Got it. Thanks,