17739-18304

17500-18182 subjects 17944-18489

[Ruby 1.9 - Bug #256] (Open) defined?(Gem::RubyGemsVersion) behaves strange
17739 [redmine@ru y] Issue #256 has been reported by Kazuhiro NISHIYAMA.

Constant lookup within class methods
17740 [tilman@ap el] I am stuck in some constant lookup behavior and I would really
17741 [dblack@ru yp] It's not just class methods. This is (I think) the one difference
17742 [calamitates@] In case you (the OP that is, I'm sure David knows) are unsure where

[Ruby 1.9 - Bug #239] (Closed) Zlib::Deflate#params can corrupt output
17743 [redmine@ru y] Issue #239 has been updated by Yusuke Endoh.

[Ruby 1.8 - Bug #253] (Closed) Zlib::GzipWriter#mtime= sets wrong mtime for Time on 1.8
17744 [redmine@ru y] Issue #253 has been updated by Yusuke Endoh.

[Ruby 1.8 - Feature #259] (Open) String crypt method's salt argument should be optional
17746 [redmine@ru y] Issue #259 has been reported by Richard Laager.

[Ruby 1.8 - Feature #260] (Open) String should have a crypt! method
17747 [redmine@ru y] Issue #260 has been reported by Richard Laager.

[Ruby 1.8 - Feature #259] String crypt method's salt argument should be optional
17748 [redmine@ru y] Issue #259 has been updated by Richard Laager.

[Ruby 1.8 - Feature #260] String should have a crypt! method
17749 [redmine@ru y] Issue #260 has been updated by Richard Laager.

[Ruby 1.8 - Feature #260] (Rejected) String should have a crypt! method
17750 [redmine@ru y] Issue #260 has been updated by Yukihiro Matsumoto.

[Ruby 1.9 - Bug #268] (Open) "make benchmark" segfaults on amd64 Linux gcc 4.3.1
17751 [redmine@ru y] Issue #268 has been reported by Ed Borasky.

[Ruby 1.9 - Bug #268] "make benchmark" segfaults on amd64 Linux gcc 4.3.1
17752 [redmine@ru y] Issue #268 has been updated by Ed Borasky.

[Ruby 1.8 - Bug #212] Issues with Readline in Mac OS X
17754 [redmine@ru y] Issue #212 has been updated by Kouji Takao.

Ruby 1.9.1 Feature and 1.9.0-3 release plan
17759 [yugui@yu ui ] Thank you for your replies to [ruby-core:17644]. < all
17761 [znmeb@ce ma ] ...
+ 17766 [wilsonb@gm i] In order for DTrace to be optimally useful, the VM needs hooks that
| 17775 [znmeb@ce ma ] ...
+ 17777 [yugui@yu ui ] Joyent's dtrace patch is very useful. So I want to merge it into Ruby 1.9.
  17778 [znmeb@ce ma ] ...

ENV['PATH'] / Kernel#system behaviour
17760 [larsch@be un] When I set ENV['PATH'] to nil/empty or some path not including the ruby
17764 [vjoel@pa h. ] It seems that deleting the env var is different from setting it to an

Toplevel Issues
17762 [transfire@gm] The toplevel (ie. main) can be quite aggravating when meta-

[Ruby 1.8 - Bug #275] (Open) bug involving condition variables
17765 [redmine@ru y] Issue #275 has been reported by Anonymous.

[Ruby 1.8 - Bug #275] bug involving condition variables
17767 [redmine@ru y] Issue #275 has been updated by Nobuyoshi Nakada.

[Ruby 1.8.7 - Bug #276] (Open) take_while functions inconsistently
17768 [redmine@ru y] Issue #276 has been reported by Anonymous.

[Ruby 1.8.7 - Bug #276] take_while functions inconsistently
17770 [redmine@ru y] Issue #276 has been updated by Matthew Boeh.

1.9.0-2 doesn't have parallel makefile fixes
17771 [ryand-ruby@z] looks like it didn't get merged over?
17780 [nobu@ru y- a] =46rom what?  It seems to work now, can you elaborate?

[Ruby 1.9 - Bug #277] (Open) 1.9/trunk: build broken in ruby/ruby.h
17785 [redmine@ru y] Issue #277 has been reported by Ollivier Robert.

[Ruby 1.9 - Bug #277] 1.9/trunk: build broken in ruby/ruby.h
17786 [redmine@ru y] Issue #277 has been updated by Ollivier Robert.
17787 [usa@ga ba ec] It seems that re-configure didn't run.

[Ruby 1.9 - Bug #277] 1.9/trunk: build broken in ruby/ruby.h
17792 [redmine@ru y] Issue #277 has been updated by Ollivier Robert.

[Ruby 1.8.7 - Bug #278] (Open) segmentation fault on regexp
17799 [redmine@ru y] Issue #278 has been reported by Anonymous.
17826 [nobu@ru y- a] Didn't any other exceptions occur before the crash?

[Ruby 1.8 - Bug #281] (Open) Time.parse does not throw ArgumentError on all invalid arguments
17810 [redmine@ru y] Issue #281 has been reported by Anonymous.

IRHG -- MISSING IN ACTION
17811 [admin@ha th ] Core,

Tracing versus coverage (was Re:  Re: Features to be included in Ruby 1.9.1)
17812 [rocky.bernst] Sorry for not noticing sooner. It occurs to me that the built-in
17816 [shyouhei@ru ] It seems to me what you need is not a coverage system but a general hook
17818 [rocky.bernst] I just looked at the code to set the coverage hash and it seems to
17819 [mame@ts .n .] No, trace instructions for coverage event are installed
17820 [rocky.bernst] My mistake I was looking at a very old version of this file. But
17821 [rocky.bernst] I don't see any difference between a line event and a "instruction

Call for Talks - 8th. Jornadas Regionales de Software Libre
17813 [matiasbrutti] ...

[Ruby 1.8 - Bug #281] (Rejected) Time.parse does not throw ArgumentError on all invalid arguments
17814 [redmine@ru y] Issue #281 has been updated by Shyouhei Urabe.

rdoc defines Hash#method_missing
17822 [mame@ts .n .] rdoc defines Hash#method_missing in lib/rdoc/ri/driver.rb, which
17823 [drbrain@se m] I will fix it this week.
17830 [mame@ts .n .] I confirmed that rdoc is upgraded, but unfortunately its tests have
17844 [drbrain@se m] Ok, sorry.
17856 [mame@ts .n .] Aha.  I think that `make test-all TESTS="rdoc"' may help.

[Ruby 1.8 - Bug #286] (Open) open-uri leaking file handles
17824 [redmine@ru y] Issue #286 has been reported by Anonymous.
17825 [nobu@ru y- a] Which version?

[Ruby 1.8.7 - Bug #213] Different ERB behavior across versions
17827 [redmine@ru y] Issue #213 has been updated by Roger Pack.

[Ruby 1.8 - Bug #255] CGI element generation methods should convert keys/values to Strings before escaping.
17828 [redmine@ru y] Issue #255 has been updated by Arthur Schreiber.

FAILURE of  "expand_path"
17829 [admin@ha th ] ...
17838 [shyouhei@ru ] Sorry if you've already noticed this, but have you read the
17841 [admin@ha th ] When the path ~name it by definition IS NOT A VALID  pathname.
17847 [mboeh@de pe ] test"
17848 [admin@ha th ] Matthew,
17849 [mboeh@de pe ] Why are you calling File.expand_path on these? The two uses of expand_path =
+ 17850 [admin@ha th ] OK, maybe I am going about things wrong.
+ 17859 [rune@sn sk t] '/Users/rune'
  17863 [admin@ha th ] Thanks, both you and Matthew Boeh have pointed

Object allocation tracking
17833 [cthompson@ne] Please excuse the blog spam.
+ 17862 [vjoel@pa h. ] There's this, too, which tells you _how_ each object is reachable (does
| 17932 [cthompson@ne] My patch does not incorporate the reachability patch, though it does
+ 17925 [sylvain.joye] The patch in itself does nothing special: it simply adds a fast way to know how
  17933 [cthompson@ne] Yes, other people at my place of employment have used ruby-prof and

Tested expand_path failure with 1.8.7 --- Same Results
17834 [admin@ha th ] Core,

1.9 (2008-02-09) also fails expand_path() test!
17835 [admin@ha th ] TEST 1

[Ruby 1.8 - Bug #286] open-uri leaking file handles
17836 [redmine@ru y] Issue #286 has been updated by Anonymous.

[Ruby 1.8 - Bug #286] open-uri leaking file handles
17837 [redmine@ru y] Issue #286 has been updated by Anonymous.

[Ruby 1.8.7 - Bug #337] (Open) xmlrpc/client.rb is missing webrick library
17839 [redmine@ru y] Issue #337 has been reported by Anonymous.
17846 [james@gr yp ] This issue can be closed.

Patch for expand_path fix ---
17842 [admin@ha th ] ...

Exapand_path Patch good as stands.
17843 [admin@ha th ] ...

[Ruby 1.8.7 - Bug #337] xmlrpc/client.rb is missing webrick library
17845 [redmine@ru y] Issue #337 has been updated by James Gray.

rdoc test uses hardcoded /tmp.
17854 [akr@fs j. rg] test/rdoc/test_rdoc_info_formatting.rb uses /tmp/rdoc-$$.
17907 [drbrain@se m] I've updated all these tests to use Dir.tmpdir.

fyi: ruby-debug broken by svn rev 17702 in p252
17861 [stephen.bann] I get this error in latest patchlevel 289 of ruby 1.8.6
17864 [rocky.bernst] I believe Kent Sibilev has fixed this in the current SVN sources for
17867 [stephen.bann] Thanks,

Expand_Path:  New Patch - Modified Processing
17865 [admin@ha th ] ...
17870 [nobu@ru y- a] This is my patch, though still I'm not sure if it should ignore
17874 [admin@ha th ] Your patch is much cleaner than mine, good job.
17929 [matz@ru y- a] Interestingly, zsh raises error for non-existing user name.
17934 [mboeh@de pe ] Processing"
17935 [admin@ha th ] OK,
17936 [matz@ru y- a] Just because specifying non-existing user in expand_path is most
17937 [admin@ha th ] OK,  I give in.  There is a number of ways to get around it.

Expand_path:   Can someone point me a doc on regression test Ruby!
17866 [admin@ha th ] I would like to find documentation on running

command files
17868 [likon@li on ] Can somebody tell me how can I write a new command to ruby?
17876 [vjoel@pa h. ] You want to write a C extension for ruby? There are two good starting

duping the NilClass
17871 [rubylearner@] ...
17872 [shyouhei@ru ] I don't think so. Nil is _the_ nil, one and only instance of NilClass.
17873 [meinrad.rech] ...
+ 17877 [shyouhei@ru ] Marshal.load(Marshal.dump(obj)) and you get a deep copy of obj.  This is
| 17879 [ks@ku ts ep ] Marshal.load(Marshal.dump(obj)) is far more expensive than a well-crafted #dup
| 17880 [rubylearner@] ...
| 17881 [dblack@ru yp] I understand not wanting true.dup etc. to break things, but it's hard
| + 17882 [shyouhei@ru ] +1 to David.  A convenient way to do Marshal idiom should be a new
| | + 17884 [duerst@it ao] So you agree that the current behavior just breaks things?
| | | 17886 [dblack@ru yp] I'm taking people's word for it. I've never encountered it myself.
| | | 17890 [jim.weirich@] Interesting.  Would you wan [].reverse to fail because it can't
| | + 17885 [robert.dober] I strongly disagree with this, it looks like an ugly workaround to me.
| |   17887 [dblack@ru yp] It's a question of the accuracy of the method name. If dup doesn't
| |   + 17888 [billk@ct .c ] Interesting.  So it would appear that for you (and ruby,
| |   + 17889 [robert.dober] Hmm I am not sure about that, I believe that true can be seen as a
| |     + 17891 [shyouhei@ru ] irb(main):001:0> a = true
| |     | + 17902 [billk@ct .c ] Oh.  Forgot about that...  (I see it works with symbols, too.)
| |     | + 17911 [duerst@it ao] Is this an actual example shunk down, or is it just made up to
| |     |   17912 [rick.denatal] ...
| |     + 17894 [dblack@ru yp] I'm afraid I don't follow that at all. What does it have to do with
| |       17895 [robert.dober] dup is a bad name for Symbol or NilClass as it does not really create
| |       + 17896 [robert.dober] ERRATUM
| |       + 17897 [dblack@ru yp] Yes. There's no general "___ does not really ___" rule. The only
| |         17904 [robert.dober] I do not want to bother this list with my limited ability of
| |         + 17905 [dblack@ru yp] There would be, though. dup! would be entirely additional. That was my
| |         + 17908 [jim.weirich@] Actually, there is a NilClass#dup.  That's why you get a type error
| |           17909 [robert.dober] *>
| + 17893 [jim.weirich@] Funny, I've occasionally come across this need in cases where I was
|   + 17898 [jim.weirich@] Heh, this thought just came to me.  We can explain :dup on nil (and
|   | 17900 [vjoel@pa h. ] irb(main):001:0> s = "foo"
|   | 17901 [jim.weirich@] Ahh, you see, that is the part that is optimized away. :)
|   + 17903 [brent@mb ri ] I believe the current behavior of the dup method is correct given that
+ 17878 [ks@ku ts ep ] I agree: I end up using "foo.dup rescue foo" in far too many places because

[Ruby 1.9 - Bug #340] (Open) 1.9/trunk does not work when compiled with llvm-gcc4 2.3 (gcc 4.2.1)
17883 [redmine@ru y] Issue #340 has been reported by Ollivier Robert.

Downloading Latest Ruby's
17899 [admin@ha th ] Core,
17906 [nobu@ru y- a] Nobu Nakada

[Ruby 1.8 - Bug #342] (Open) [PATCH] REXML Pretty Formatter's private method wrap breaks if no space is found
17913 [redmine@ru y] Issue #342 has been reported by Anonymous.


17914 [liangyin@ga ] ...

select returning an enumerator
17915 [dblack@ru yp] I notice that select without a block returns an enumerator in 1.9. I'm
17916 [david@da id ] David,
+ 17917 [meinrad.rech] ...
| 17920 [david@da id ] You need to write a.each.with_index.select or a.each_with_index.select.
+ 17919 [dblack@ru yp] You're right that I knew that, though I hadn't factored it in,
  17921 [david@da id ] The proposed with_memo function would presumably work in place of, or in

String methods on subclasses return "uninitialized" objects
17918 [vsizikov@gm ] During the work on RubySpec tests, we had an interesting discussion on
17924 [gdefty@at gl] Vladimir,

[Ruby 1.9 - Bug #345] (Open) 1.9 racc appears to seg fault
17922 [redmine@ru y] Issue #345 has been reported by Roger Pack.

[Ruby 1.9 - Bug #161] Profile library seems broken in 1.9 15427cat t.rv
17923 [redmine@ru y] Issue #161 has been updated by Roger Pack.

Expand_path Bug:  HOW to report a bug and a patch to fix it!
17926 [admin@ha th ] Is there a document to explain submitting a bug report
17927 [vsizikov@gm ] You could file bugs there, and attach patches. You could even
17928 [admin@ha th ] Vladimir,

[Ruby 1.9 - Bug #143] (Closed) rdoc and magic comment
17930 [redmine@ru y] Issue #143 has been updated by Yui NARUSE.

[Ruby 1.9 - Feature #235] default charset of rdoc
17931 [redmine@ru y] Issue #235 has been updated by Yui NARUSE.

[Ruby 1.9 - Feature #235] default charset of rdoc
17938 [redmine@ru y] Issue #235 has been updated by Martin Dürst.

RUBY_ENGINE?
17943 [vsizikov@gm ] It was discussed in the past but as far as I can tell, there were no
+ 17949 [charles.nutt] I concur...it would be nice to get it out there and finally have an
+ 17950 [akr@fs j. rg] What the usage of RUBY_ENGINE?
| + 17951 [ryand-ruby@z] not command name, but ruby "flavor": (mri) ruby, rubinius, jruby, etc.
| | + 17952 [laurent.sans] What would be the value if this constant, for 1.8 vs 1.9? Also, String
| | | 17957 [vsizikov@gm ] It could just be "mri'. For version numbers, you could then inspect
| | | 17962 [akr@fs j. rg] def self.default_dir
| | | + 17963 [hramrach@ce ] I guess this is bogus since there is no guarantee that RUBY_ENGINE is
| | | + 17967 [vsizikov@gm ] Ah, good point.
| | | + 17976 [drbrain@se m] I expect the MRI value to be "ruby" as it is the reference
| | + 17961 [billk@ct .c ] RUBY_ENGINE or similar constant to indicate implementation
| |   + 17966 [vsizikov@gm ] Ah, I forgot about that discussion...
| |   | 17997 [charles.nutt] ruby_install_name could be ruby1.8 if the install name is different than
| |   + 17975 [drbrain@se m] ruby = File.join RbConfig::CONFIG['bindir'],
| |     17977 [billk@ct .c ] (Note, I don't have any vested interest in this, I am
| |     18000 [charles.nutt] Say you have a library written in C for MRI, modified C or Ruby with
| |     18019 [billk@ct .c ] I've been hesitant to respond; I don't want to clutter up the thread
| + 17958 [vsizikov@gm ] 1. RubyGems uses it to invoke #{RUBY_ENGINE}.rb file with
|   17981 [akr@fs j. rg] I'm not sure what the customizations to be done.
|   + 17986 [luislavena@g] One example is the gem stub files generated by RubyGems that need to
|   | + 17989 [brixen@gm il] RUBY_ENGINE is needed to definitively identify the implementation on
|   | | 17994 [hramrach@ce ] However, there are at least two interpretations of of "the thing that
|   | | 17998 [charles.nutt] JRuby is JRuby no matter how you want to slice it. But there's no
|   | | 18018 [hramrach@ce ] Nobody is denying JRuby is JRuby, and it would be useful to know one
|   | + 17990 [brixen@gm il] RUBY_ENGINE is needed to definitively identify the implementation on
|   | + 17991 [brixen@gm il] RUBY_ENGINE is needed to definitively identify the implementation on
|   + 17999 [charles.nutt] It's not quite that simple, since just having specific libraries doesn't
|   | 18021 [akr@fs j. rg] Agreed.  It should be possible to test a method is really
|   | 18025 [charles.nutt] Not a method...a feature. A method is not a feature. JRuby has an entire
|   + 18002 [vsizikov@gm ] I agree, that in this case there is need for RUBY_ENGINE, but there
|     18020 [akr@fs j. rg] What's problem of rescue LoadError?
|     18022 [luislavena@g] gem install activerecord-jdbc-adapter --platform=java even java is not
|     18026 [vsizikov@gm ] Also, what if the application to support more implementations? How
|     18027 [charles.nutt] And another reason: ideally people could eventually be able to share gem
+ 18001 [hongli@pl n9] ...
| 18003 [matz@ru y- a] Don't be so quick.  You guys haven't made it clear that
| 18004 [hongli@pl n9] They are both MRI. But it's easy to distinguish between them by checking
| + 18005 [hongli@pl n9] By the way, Ruby Enterprise Edition's goal is to be eventually merged
| | 18006 [luislavena@g] Before considering merging of REE into Ruby you should also take a
| + 18009 [laurent.sans] Mmh, Ruby Enterprise Edition is compatible with MRI, but the engine
|   + 18011 [charles.nutt] Rubinius is intended to be compatible with MRI 1.8. JRuby is intended to
|   | 18023 [jeremymcanal] Exactly.  This constant would've been infinitely helpful to a project
|   + 18014 [duerst@it ao] And not to mention the differences with respect to character encodings
+ 18107 [charles.nutt] So where do we stand on this? I'm still convinced RUBY_ENGINE is needed,
+ 18108 [charles.nutt] ...
+ 18284 [charles.nutt] Could RUBY_ENGINE please be added to 1.8.x and 1.9.x so users can rely
  18287 [matz@ru y- a] In that case, what should the value for each implementation be?
  + 18289 [gregory.t.br] I think both of these should be "ruby", as you can find the version by
  | 18295 [matz@ru y- a] The point is we have different core for 1.8 and 1.9, but anyway, I
  | + 18296 [laurent.sans] FYI, I added RUBY_ENGINE as "macruby" in my private branch.
  | + 18302 [charles.nutt] Excellent Matz, thank you very much!
  + 18290 [curth@mi ro ] IronRuby already returns "ironruby" for RUBY_ENGINE.
  + 18304 [ryand-ruby@z] symbols? better for case stmts / hashes etc.
threads.html
top