8815-9155

8569-9548 subjects 8999-9356

^ Segfault in libc strlen, via rb_str_new2
8815 [ser germane-] charset="utf-8"
8817 [drbrain segm] xmlTextReaderPtr	xmlNewTextReaderFilename	(const char * URI)
+ 8818 [drbrain segm] Eric Hodel - drbrain@segment7.net - http://blog.segment7.net
+ 8821 [ser germane-] charset="iso-8859-1"
  8830 [drbrain segm] I don't have one of those (PPC G4).  It is very likely that you'll
  + 8837 [haisenko com] Yes, for example AMD64: sizeof(int) == 4, sizeof(void*) == 8
  | + 8841 [ser germane-] charset="iso-8859-1"
  | | 8848 [drbrain segm] You did ask a question about a non-working example in public...
  | + 8880 [vincent.four] Unless maybe in when compiling some extension for ruby ?
  |   8881 [haisenko com] The fact that people do this stuff doesn't mean it's correct to do so.
  |   8883 [steven.jenki] Right. To be strictly pedantic, however, the check should be
  + 8840 [ser germane-] charset="iso-8859-1"

^ [PATCH] OptionParser < Hash
8826 [eegreg gmail] From the example in the code of optparse.rb, It seems to me that we
+ 8827 [eegreg gmail] Think I forgot to attach the patch.
+ 8831 [nobu ruby-la] Thank you, but sorry, it's not acceptable.  The idea that OptionParser
  8834 [eegreg gmail] This is my first patch submission, so I hope I am not being rude, and
  + 8835 [nobu ruby-la] Don't mind.
  | 8856 [eegreg gmail] This is great, Nobu!
  | 8858 [nobu ruby-la] Though I forgot to mention the usage, I'm glad you to see it.
  + 8846 [drbrain segm] No.

^ [ ruby-Bugs-5711 ] REXML fails to parse UTF-16 XML.
8828 [noreply ruby] Bugs item #5711, was opened at 2006-09-11 01:25
8833 [matz ruby-la] REXML is converting body twice, once from initialize, one more from
8842 [ser germane-] charset="iso-8859-1"
8843 [matz ruby-la] Shall I merge the patch into 1.8?  Or will you?
8844 [ser germane-] If I do it, it will not get done until this weekend.  If it needs to go
8845 [matz ruby-la] There's no reason to hurry.  Probably it's better for you to commit.

^ [ ruby-Bugs-5712 ] require 'rdoc/usage' fails on ppc
8829 [noreply ruby] Bugs item #5712, was opened at 2006-09-10 18:06
8832 [matz ruby-la] We need information about the version of Ruby, and how did you install
8853 [tom infoethe] So I've closed #5712 and left #5563 open.

^ YAML errors with indentation indicator
8836 [funai.akira ] 1) malformed block scalar emitted with to_yaml()

^ [PATCH] rdoc with optparse
8838 [eegreg gmail] This is a patch to allow rdoc to read the help message from optparse,
8847 [drbrain segm] This is not a patch, please use diff to make a patch.
8849 [eegreg gmail] sorry, diff is attached.

^ constants (rcrs 178, 343,345)
8850 [neleai sezna] I looked at my rcr 343 and 345. Could you give me some feedback?

^ README.EXT -- English adjustments.
8851 [hgs dmu.ac.u] I've been looking at README.EXT and thought I could make the English
8852 [Daniel.Berge] <snip good stuff>
8855 [hgs dmu.ac.u] s/Ptr()./Ptr() to get a char* from a VALUE./ ?  This sentence doesn't

^ [ ruby-Bugs-5740 ] RDOC doesn't show string  "#" or "\#"
8857 [noreply ruby] Bugs item #5740, was opened at 2006-09-13 11:55

^ [ ruby-Patches-5748 ] (test/unit) assert_raise without parameters should check if any exception is being raised
8859 [noreply ruby] Patches item #5748, was opened at 2006-09-13 19:34

^ [ ruby-Bugs-5752 ] Date._parse doesn't understand hour only time, e.g. 2pm
8860 [noreply ruby] Bugs item #5752, was opened at 2006-09-14 10:40

^ new changes in strings+symbols
8861 [matju arteng] This message is in MIME format.  The first part should be readable text,
8863 [matz ruby-la] The security check was too strict.  I just fixed in the HEAD.
8872 [mental rydia] =20
8879 [matz ruby-la] OK, let me consider.  Perhaps moving unification process after intern
8886 [chneukirchen] Or even drop it?  What is it good for?

^ ruby-1.8 CVS make check fails
8862 [ville.mattil] uname -a

^ documentation of ruby internals
8864 [denigeorge c] Can I get some documentation regarding the internal working of Ruby.
+ 8865 [asbradbury t] I'm not sure what else is available, but the work that's been doing on
| + 8867 [denigeorge c] Thanx Alex .
| | 8873 [drbrain segm] parse_tree_show from the ParseTree gem will show you the internal
| + 8870 [mike rubywiz] You can use Google translation to read the original. It produces
|   8871 [denigeorge c] thank you
+ 8866 [hgs dmu.ac.u] I have suggested that it might be useful if RDoc were extended to support
  8868 [denigeorge c] yea .. actually i am working on a project for doing some optmizations on
  + 8869 [hgs dmu.ac.u] You may find useful information in the virtual machine development activities
  + 8877 [matju arteng] This message is in MIME format.  The first part should be readable text,
    8878 [ mfp acm.org] 1.9/HEAD still uses the good old rb_eval AST walker, as it hasn't been merged

^ [ ruby-Bugs-5763 ] Problem with generated soap requests / classes using the WSDL Tools
8874 [noreply ruby] Bugs item #5763, was opened at 2006-09-14 17:52

^ Ruby Hackers Guide Translations
8875 [ceo hawthorn] I have a full machine translation of Ruby Hackers Guide
8876 [dreamhead.cn] Maybe this message should be sent to rhg (the project that translates

^ Curses problem
8884 [sam.overdorf] charset="US-ASCII"

^ [ ruby-Patches-5774 ] numeric.c fails to build on 64-bit platforms (Fedora Core 5 x86_64 gcc 4.1.1)
8885 [noreply ruby] Patches item #5774, was opened at 2006-09-16 12:19
8887 [matz ruby-la] The patch has been merged.  Thanks.
8893 [shiba mail2.] Typo?
8895 [nobu ruby-la] Indeed.  orz

^ 1.9 build failure.
8888 [hgs dmu.ac.u] Testing a patch (nowhere near digest) against the latest 1.9 on Sun
+ 8889 [nobu ruby-la] What's digest_new.c?
| 8890 [hgs dmu.ac.u] OOPS! that was something I was doing with previous patch.
+ 8891 [matz ruby-la] How did you get this 'digest_new.c' file?  It's not in the

^ Getting docs out of README.EXT
8892 [hgs dmu.ac.u] I've been trying to find the easiest way to get README.EXT into
8894 [drbrain segm] It would need to be attached to a class or module somewhere.

^ [ ruby-Bugs-5821 ] FXRI/IRB - Shift + Backspace crashes help viewer
8896 [noreply ruby] Bugs item #5821, was opened at 2006-09-20 11:59

^ [ ruby-Bugs-5822 ] Ruby's 'etc' module cannot handle the UID of OS X 'nobody' properly
8897 [noreply ruby] Bugs item #5822, was opened at 2006-09-20 11:13
8900 [matz ruby-la] Would anybody report what happens when we run the above script on OSX?
+ 8901 [wilsonb gmai] Did something related to BigNum change in 1.8.5?
+ 8902 [jimfreeze gm] charset=US-ASCII;
  8903 [matz ruby-la] Thank you guys.  I fixed it on the CVS.  Would somebody check the fix?

^ Typo in example comment for recent string.c v 1.274 commit
8898 [rubyzbibd ub] In the CVS commit for ruby/string.c v 1.274, I think there is a typo
8899 [matz ruby-la] Indeed.  I fixed it.  In addition, I fixed one serious bug in

^ patch bignums
8904 [neleai sezna] First refactoring are num_op where first argument can be
8905 [nobu ruby-la] Thank you, commited.
+ 8906 [Daniel.Berge] Oh, cool - are we getting .odd? and .even? for 1.8.x? Or just 1.9?
| 8909 [matz ruby-la] Just 1.9
+ 8907 [sroberts uni] Did the attachment get stripped or forgotten?
| 8908 [nobu ruby-la] Forgotten, sorry.
+ 8910 [neleai sezna] Why? This is used when you are not sure if its fixnum or bignum.

^ [ ruby-Patches-5869 ] Typo in documentation of Kernel#rand
8911 [noreply ruby] Patches item #5869, was opened at 2006-09-23 14:25

^ NODE_POSTEXE / gc
8912 [decoux moulo] ruby crash with extconf.rb for extensions
+ 8918 [matz ruby-la] Thank you.  Your patch has been merged.
+ 8931 [matju arteng] This message is in MIME format.  The first part should be readable text,

^ [ ruby-Bugs-5884 ] Dir.chroot and '/'
8913 [noreply ruby] Bugs item #5884, was opened at 2006-09-24 18:54
8919 [nobu ruby-la] Thank you for the report.  Dir.mkdir("/") also raises ENOENT instead of
8921 [matz ruby-la] Can you commit the fix?

^ Behavior change for Hash#default in 1.8.5?
8914 [djberg96 gma] It appears that something has changed with Hash#default between 1.8.4
+ 8915 [wilsonb gmai] irb(main):001:0> hash = Hash.new {|h,k| h[k] = []}
+ 8916 [matz ruby-la] Yes.

^ [PATCH] rdoc capture output (help message)
8920 [eegreg gmail] attached is patch for rdoc to capture execution output of the file it
+ 8937 [eegreg gmail] there was a bug in original implementation.  attached is a new rdoc.rb diff
+ 8940 [drbrain segm] This command line option looks too complicated for me to accept,
  8941 [eegreg gmail] The simplest command line would be
  8943 [vjoel path.b] I think what Eric was suggesting was something like this, using rake (do
  8953 [eegreg gmail] It looks like you could seperate this out into a rake task, but then
  8955 [drbrain segm] If a dependency already exists between the rdoc command and your
  8965 [eegreg gmail] Thanks for pushing back on this as a good maintainer should.   You
  8966 [drbrain segm] No. Its best if just left to external code.  It is too special-

^ Problem with latest array.c changes
8922 [ksruby gmail] This change
8924 [matz ruby-la] Ah, thank you for finding and fixing the bug.
8934 [ksruby gmail] Shouldn't it be applied to ruby_1_8 branch too? Thanks.

^ make test 1.9 fail
8926 [neleai sezna] Print nil prints now nothing and not 'nil'. So following test fails

^ Re: RDoc patch, possible bug in socket.c for TCPSocket.new
8928 [Daniel.Berge] Here's the patch again, without the bad formatting from gmail, but with the
8929 [gwtmp01 mac.] This is the normal behavior for a TCP session.  After the server
+ 8930 [Daniel.Berge] Ok, thanks.
| 8932 [gwtmp01 mac.] I was a bit confused.  I was thinking your code implemented a server
+ 8933 [garbagecat10] A little more: TIME_WAIT is only incurred by the side of the connection that
  + 8935 [djberg96 gma] The platform is Solaris 10.
  + 8936 [gwtmp01 mac.] On Mac OS I can't see the session sitting in TIME_WAIT via netstat but I

^ [ ruby-Bugs-5921 ] "ri Array.new" doesn't show assignment values
8938 [noreply ruby] Bugs item #5921, was opened at 2006-09-27 01:52

^ [RDoc] Patch for object.c and Class.inherited
8942 [djberg96 gma] Below is an rdoc patch that adds documentation for the Class.inherited

^ [ ruby-Bugs-5940 ] "nundefined method `to_str' for nil:NilClass" exception from mkmf.rb when doing "make distclean"
8944 [noreply ruby] Bugs item #5940, was opened at 2006-09-28 10:59
8947 [nobu ruby-la] Thank you.  Though it is incorrect since $extout may contain

^ [ ruby-Bugs-5941 ] "nundefined method `to_str' for nil:NilClass" exception from mkmf.rb when doing "make distclean"
8945 [noreply ruby] Bugs item #5941, was opened at 2006-09-28 10:59

^ [ ruby-Bugs-5942 ] "nundefined method `to_str' for nil:NilClass" exception from ext/extmk.rb when doing "make distclean"
8946 [noreply ruby] Bugs item #5942, was opened at 2006-09-28 11:07

^ socket (and many others) not building on osx?
8948 [ryand-ruby z] I'm stumped. A brand new clean build doesn't build socket.
8949 [nobu ruby-la] Can you show mkmf.log file under ext/socket?
8950 [ryand-ruby z] ...
8951 [nobu ruby-la] Sorry, I had committed the patch to fix it to the local repository.
8952 [ryand-ruby z] Thank you sir, that seems to have fixed everything.

^ The %? .. ? Operator
8954 [james graypr] I'm needing to know the full list of characters that can (or cannot)
8957 [drbrain segm] /%[QqWwxrs]?(.).*\1/
8959 [james graypr] This is interesting.  The operators do differ slightly when you omit
8961 [drbrain segm] Yes, for a bare string ' ' will work, but not when % is a method call.
8962 [james graypr] Wow.  As always you have taught me new things Eric.  Thank you.
9001 [matju arteng] This message is in MIME format.  The first part should be readable text,

^ [BUG] 1.8 CVS regressions
8956 [ville.mattil] The 1.8 cvs is currently suffering some "make check" regressions.
8960 [ksruby gmail] Can you try to apply this patch?
8967 [ville.mattil] Yes, this patch fixes the

^ rb_protect in rb_f_fork?
8958 [ctm ardi.com] Howdy,
8980 [matz ruby-la] Good question.  But it needs a few spare days to inspect.  Sorry for
8983 [ctm ardi.com] Cool.  It's definitely super low-priority.

^ abort trap on pure ruby (1.8.x--yesterday's cvs)
8963 [ryand-ruby z] I get the following error from the code below. Doesn't repro in gdb
8964 [ryand-ruby z] It seems to be this test specifically (or in combo with something else).

^ [ ruby-Bugs-5972 ] No configure option for non-standard library locations (e.g., TCL/TK)
8968 [noreply ruby] Bugs item #5972, was opened at 2006-09-30 02:18
8970 [matz ruby-la] It already does.  For example configure accepts --with-tcl-dir option.

^ [ ruby-Bugs-5973 ] Build summary would be helpful
8969 [noreply ruby] Bugs item #5973, was opened at 2006-09-30 02:22

^ fileutils.rb
8971 [ryand-ruby z] 1.36.2.10    (aamine   18-Sep-05):   def fu_check_options(options,
8972 [aamine lover] I don't now about Rake, but why Rake calls fu_check_options?
8973 [ryand-ruby z] Yes. In particular: #sh, #ruby, and #safe_ln.
8975 [jim weirichh] It looks like rake's dependency on fu_check_options and
8987 [jim weirichh] Latest beta version (0.7.1.5) of Rake does not depend upon fu_xxxx.

^ Insecure warnings on sticky-bit directories
8976 [laurent.sans] Currently, Ruby prints a warning when a directory that is part of
8986 [nobu ruby-la] It is definitely unsafe to include a world-writable directory
+ 8993 [laurent.sans] You're right, but what if the world-writable sticky-bit directory is
| + 8994 [laurent.sans] err, I meant /tmp/ro. Same result though.
| + 8995 [nobu ruby-la] That is what has been fixed already.
|   8996 [laurent.sans] Thank you, sorry for the noise.
+ 9000 [matju arteng] This message is in MIME format.  The first part should be readable text,

^ [ ruby-Bugs-5990 ] Inheritance and Autorunner: Default_test causes a problem
8978 [noreply ruby] Bugs item #5990, was opened at 2006-10-02 10:05
8979 [matz ruby-la] I think I understand the situation, but not for sure.  Could you show
8990 [dblack wobbl] Since this went to the Ruby bug site, it may have been posted by
8991 [nobu ruby-la] It occurs even by test-all, and fixed now.
8992 [tom infoethe] I'm not sure of the best way to solve this without opening the door to

^ [ ruby-Bugs-5995 ] OpenSSL::X509::Certificate.new is broken in the 1.8.5 libraries
8984 [noreply ruby] Bugs item #5995, was opened at 2006-10-02 15:16
9003 [matz ruby-la] Perhaps, this is too few information.  Does anybody (including

^ Business Intelligence Framework
8988 [mukembo gmai] I have developed several libraries that I think we can work with to develop
8989 [haisenko com] If you like a new language whose /syntax/ is a mix of the above: good luck

^ Re: [ruby-cvs:18323] ruby: * eval.c (splat_value): use "to_splat" instead of "to_ary" to
8997 [matju arteng] This message is in MIME format.  The first part should be readable text,
8998 [matz ruby-la] Hmm, even if to_splat remind you list context in Perl, it does not
+ 9029 [matju arteng] This message is in MIME format.  The first part should be readable text,
| 9030 [matz ruby-la] Every language grows in more recent version, as Ruby does, and PL/I
| + 9031 [dblack wobbl] I think you should trust yourself more :-)  Growing doesn't have to
| | + 9033 [jmg3000 gmai] "Perfection is achieved, not when there is nothing more to add, but
| | | 9063 [matju arteng] This message is in MIME format.  The first part should be readable text,
| | + 9036 [matju arteng] This message is in MIME format.  The first part should be readable text,
| |   9037 [znmeb cesmai] A very interesting point of view. I guess I think of the language, the
| + 9034 [dave burt.id] Matz, I don't understand why "to_a was too general." What's wrong with
| | 9041 [matz ruby-la] It should be x={1 => 2, 3 => 4}, y=nil, I think, and 1.9 behave so.
| | 9042 [dblack wobbl] The name "to_splat" is a bit problematic.  Since "splat" is itself a
| | + 9043 [matz ruby-la] Um.  For "to_" + verb, we already have "to_open".  I am open for the
| | | + 9044 [dblack wobbl] I can't find to_open (is it in an extension?) but I don't think
| | | | + 9045 [matz ruby-la] It's only in 1.9.
| | | | | + 9046 [hgs dmu.ac.u] I'm probably getting confused here, but isn't this operator doing
| | | | | | 9048 [dblack wobbl] That's why I call it the unary unarray operator :-)
| | | | | + 9047 [dblack wobbl] I actually meant just "starred", without the to_.  I'm not sure the
| | | | |   + 9049 [matz ruby-la] Hmm, you are a native English speaker, so your feeling must be
| | | | |   | 9052 [dblack wobbl] A list.
| | | | |   + 9050 [james graypr] to_params
| | | | |   | 9053 [ruby-ml kitt] =2E..
| | | | |   | + 9054 [dblack wobbl] I like to_list.  It's specific enough that misuse/overuse will be
| | | | |   | | 9066 [matz ruby-la] Well, that name assumes we are converting an object into a list which
| | | | |   | | + 9067 [james graypr] Is there any where they can be used that aren't arguments lists?  I
| | | | |   | | | 9071 [matz ruby-la] Currently splat is used for every unary * except for * in right hand
| | | | |   | | + 9070 [joshua rever] Does it have to be a to_*?  What about just #expand?  Whenever I see
| | | | |   | | + 9072 [dblack wobbl] Hash[1,2,3,4]     # a list of four numbers -- no array involved
| | | | |   | | | + 9074 [rick.denatal] That *a in to_values doesn't seem to be syntax known to either ruby
| | | | |   | | | + 9083 [matz ruby-la] Understandable.  But some may feel different from the word List, for
| | | | |   | | |   + 9084 [dblack wobbl] I see what you mean.
| | | | |   | | |   + 9085 [rick.denatal] Yes,  but the proposal is not to_value but to_values  which might be
| | | | |   | | |     9086 [matz ruby-la] to_values converts a value to values (actually, an array represents
| | | | |   | | |     + 9087 [james graypr] I believe I understand where Rick is coming from.
| | | | |   | | |     | + 9089 [dblack wobbl] I've been thinking about the whole thing, and I tend to agree.  I'm
| | | | |   | | |     | | 9098 [ruby-ml kitt] =20
| | | | |   | | |     | + 9114 [rick.denatal] Yes that pretty much sums up my feelings.
| | | | |   | | |     + 9115 [rick.denatal] Matz, sorry, I guess I misread your position. I thought that you were
| | | | |   | | |       9118 [dblack wobbl] It's the * that really converts it, though.  The pre-star method is
| | | | |   | | + 9092 [angus quovad] Why not *@ ?
| | | | |   | |   9102 [chneukirchen] +1, for consistency.
| | | | |   | + 9055 [james graypr] to_args is probably a bit more Rubyish, since *args is an oft-seen
| | | | |   | | 9056 [dblack wobbl] I think, though, that the goal is to have a way to define *anything
| | | | |   | | 9057 [Thomas.Enebo] One issue with naming is if you choose something too synonymy it
| | | | |   | | + 9058 [dblack wobbl] I'd rather not see things added to the language that are in the
| | | | |   | | + 9061 [matju arteng] This message is in MIME format.  The first part should be readable text,
| | | | |   | | + 9088 [Charles.O.Nu] Of those proposed so far I like to_values best. There's no list/array
| | | | |   | |   9090 [dblack wobbl] Am I the only one who thinks that "splat" is a goofy, babyish word,
| | | | |   | |   + 9091 [Charles.O.Nu] I like the word; it's very descriptive of slamming an array against a
| | | | |   | |   | 9096 [dblack wobbl] I see what you mean.  I think there's been ambiguity in some of the
| | | | |   | |   + 9093 [steven.jenki] Not the only one. Even a good joke tires with retelling; this one isn't
| | | | |   | |   + 9094 [hgs dmu.ac.u] I see your point.  To put this more mildly (for those who'd prefer...)
| | | | |   | |   + 9095 [billk cts.co] I've started two replies on this general notion over the past two
| | | | |   | |     9097 [dblack wobbl] This isn't an attack on fun by non-fun.  I've found Ruby fun for six
| | | | |   | |     + 9099 [billk cts.co] OK, but that was sort of my point, too.  I've found Ruby fun for
| | | | |   | |     + 9100 [ruby-core wh] Because visual terms can be more powerful, in terms of associating with a
| | | | |   | |     | 9101 [dblack wobbl] It's definitely not original to Rubyists; I remember it from
| | | | |   | |     | 9103 [billk cts.co] Any studies on why some languages seem less fun than Ruby?  :-)
| | | | |   | |     + 9148 [steven.jenki] 'burst'? At least it connotes the operation (and loosely, the result).
| | | | |   | + 9073 [rick.denatal] to_parms and to_parameterlist don't feel right to me because the *
| | | | |   + 9051 [gwtmp01 mac.] I remember reading some notes about Ruby 2.0 and the desire to
| | | | + 9060 [matju arteng] This message is in MIME format.  The first part should be readable text,
| | | |   9075 [znmeb cesmai] Yes, lambda notation is much friendlier :).
| | | + 9069 [steven lumos] Not general enough, I know: to_arguments.
| | + 9059 [matju arteng] This message is in MIME format.  The first part should be readable text,
| |   9076 [znmeb cesmai] Yes ... calling it "disarray" gruntles me. :)
| + 9035 [matju arteng] This message is in MIME format.  The first part should be readable text,
| | 9040 [matz ruby-la] Yes, languages are mortal.  I'd rather see Ruby being a black hole,
| + 9068 [steven lumos] Just to add the voice of my group, we appreciate things getting better
+ 9117 [warrenbrown ] Without sounding too flippant, why not "to_ar"?  It would be "in
  9124 [vjoel path.b] I thought that method was used to render strings into Pirate-speak.
  + 9125 [Daniel.Berge] - Dan
  | 9140 [vjoel path.b] self.to_rofl!
  | 9144 [dblack wobbl] The "dangerous" version?  Too near the stairs? :-)
  | + 9149 [vjoel path.b] Or while drinking milk.
  | + 9150 [steven lumos] Maybe he meant (self.to_rofl)!.
  + 9155 [znmeb cesmai] No ... no ... it's "Talk Like A **Parrot** Day!" Parrot!
threads.html
top