14918-15184

14700-17040 subjects 15182-15607

^ usage of Ruby as an embedded interpreter on smartcards
14918 [dmitri.shori] Dear Ruby Community,
14921 [meinrad.rech] ruby's memory footprint might be too big. have you tried lua?

^ Re: Using values of mandantory arguments in default values
14929 [matz ruby-la] It's intended, and I don't think it can be changed.  I think under the

^ Re: Really strange GC behaviour: Was [BUG] memory leak?
14937 [ara.t.howard] for me it leaks up until '2' and then stops.  nice simplification.  i
14943 [ara.t.howard] cfp2:~ > cat leak.rb
14953 [rick.denatal] But is that the cart or the horse?
14958 [ara.t.howard] that's sure odd eh?
14986 [chris heathe] That is probably because the garbage collector looks on the C stack too,

^ SVN Revison 71
14939 [jflam micros] The major Ruby-visible change is the addition of super thanks to John Messe=
14940 [jflam micros] Whoops ... sent to wrong list. Sorry!
14949 [meinrad.rech] Nevermind. Thanks for the development status information on IronRuby ;)

^ [1.9] case when end
14944 [decoux moulo] when seems to have a little problem

^ 1.9 RI blowing up, not sure where to report it.
14959 [rick.denatal] It's unclear to me whether or not the tracker on Rubyforge is still
+ 14963 [rubys intert] For what it's worth, I'm not seeing this.  I have an automated job that
| 14964 [dave pragpro] Fails here too.
| 14987 [drbrain segm] Fixed.
| 14991 [rick.denatal] Verified, thanks Dr. Brain!
+ 14988 [drbrain segm] The RDoc project on Rubyforge is the correct place.
  15021 [ryand-ruby z] we should just merge over to ruby now that we've got real categories

^ [Bug?] Fiber in iterators
14962 [robert.dober] 625/126 > cat fiber2.rb && ruby1.9 fiber2.rb
+ 15043 [robert.dober] Well I kind of expected that somebody would call me stupid or
+ 15045 [gwtmp01 mac.] The fiber starts in a suspended mode.

^ Before I create a ticket
14965 [dave pragpro] Is it reasonable to expect the following to produce differing counts
+ 14966 [ara.t.howard] maybe throw a 'reverse' in there somewhere - right now it seems like
| 14969 [dave pragpro] Makes no difference :)
| 14972 [ara.t.howard] quite surprising then!
+ 14974 [matz ruby-la] * the default scheduling policy is SCHED_OTHER.  Ruby does not
  14975 [dave pragpro] If not a bug, perhaps we should deprecate Thread#priority[=], because

^ nil encoding as synonym for binary encoding
14976 [david davidf] The following just appeared in the ChangeLog
14977 [matz ruby-la] I don't think there's no string in Ruby without encoding.  Every
+ 14978 [dave pragpro] I'm certainly not an expert on encoding, so take this with a grain of
| + 14979 [david davidf] Yes, but keep in mind that Encoding::BINARY is a synonym for
| | 14993 [dave pragpro] It is now. I'm suggesting that is a mistake: Encoding::Binary should
| | + 15001 [hramrach cen] Well, ascii-8bit is the way the strings are mostly used. If you
| | | 15002 [dave pragpro] But what would you return? An ASCII character? Or an integer? I'd
| | | 15035 [duerst it.ao] Well, a binary byte sequence of length one is what would be returned
| | + 15003 [ed.odanow wo] I agree, except for details of the last remark.
| | | 15005 [dave pragpro] But how to you know it's ascii? It might be EBCDIC. Or, if the top bit =20=
| | | + 15006 [Tomas.Matous] I guess the returned single-byte string has binary encoding as well.
| | | + 15019 [ed.odanow wo] Do you mean the '# => "a"'? - For binary encoding an integer value
| | |   15023 [gwtmp01 mac.] ruby1.8:  string[x]      returns byte at offset x as integer
| | |   + 15027 [ed.odanow wo] Sorry - a language problem for me, I mean "put it onto a device which needs
| | |   + 15033 [duerst it.ao] Please try bin[0..4].unpack('C*').     Thanks,   Martin.
| | |     15039 [gwtmp01 mac.] Sure that will work but it is ugly and requires an
| | + 15007 [akr fsij.org] Why the result is "cat"?
| |   15008 [dave pragpro] It's "cat" because that's what the input was...  The fact that we
| + 14994 [halostatue g] I have to agree, here. Part of that is because I personally see all
+ 14980 [gwtmp01 mac.] If I grab an Ethernet packet off the wire, part of it is going to be
  14981 [matz ruby-la] As Akira Tanaka stated in [ruby-core:14617] and [ruby-core:14657], we
  14995 [david davidf] I actually composed and sent a message to the list about that, but it has
  + 14996 [matz ruby-la] Hmm, in that case, we have to seek something instead of nil for
  | 15009 [david davidf] I don't understand.  Are you saying that in the current implementation
  | 15012 [matz ruby-la] set_encoding(nil) - default
  | + 15015 [david davidf] 0) Is no check on write really different than checking for
  | | 15036 [duerst it.ao] I'm currently not thinking to support transcoding either from or to
  | | 15037 [vincent.isam] Instead (or in addition to) this third argument, what about having the
  | | + 15038 [rubys intert] +1
  | | + 15046 [duerst it.ao] That's also planned. But a third argument can be much faster for
  | + 15016 [rubys intert] That makes it much clearer, at least to me.
  + 14997 [rubys intert] Since nil seems to imply different things to different people in this
    15000 [rubys intert] Eek!  I hadn't yet received Matz's email indicating that nil meant
    15010 [matz ruby-la] If keeping consistency weigh with us.  But I am not sure.  At least I
    15011 [dave pragpro] Seems like you're overloading methods. Maybe split into two, and have
    15013 [matz ruby-la] I don't get it.  Am I overloading what methods?  Can you elaborate?
    15014 [dave pragpro] does it mean don't change, change to default, change to binary?  So I

^ Stack memory leaks on i386
14985 [chris heathe] While debugging a Ruby 1.9 crash, I noticed some machine stack memory

^ Re: [Ruby 1.9] NKF
15018 [naruse airem] This is bug, sorry.  I fixed it at r14997.
+ 15031 [naruse airem] This is spec following ruby19 -Ks -e'p Encoding.default_external'.
+ 15077 [drbrain segm] You may need to rm .ext/rdoc as RDoc is not good at making updates to
  + 15079 [naruse airem] NARUSE, Yui  <naruse@airemix.com>
  + 15091 [decoux moulo] not enough :-)

^ base64
15020 [funkaster gm] Is anyone working on base64? is it obsolete now or will be replaced
15024 [rubys intert] Just use ['foo'].pack('m') and "Zm9v\n".unpack('m').first

^ Re: Next Ruby 1.9 release (1.9.1)?
15028 [ed.odanow wo] As I understood the next Release will be Ruby 1.9.0-1 end of January.
15030 [matz ruby-la] Yesterday, I talked with Koichi and he wanted more time to prepare for

^ [1.9] rb_rescue2 crashes 1.9.0-0
15040 [decoux moulo] This is in reference to [ruby-talk:287175], there is a problem when
15071 [matt upillar] unsubscribe

^ String methods and internal regexp usage
15041 [lopx gazeta.] Some String methods (notably split, scan, sub and gsub) use regexp

^ Maintainers needed ?
15042 [laurent mold] I think I may have some spare cycles in the coming weeks/months and I

^ Build failures 15007-15013
15044 [rubys intert] Just FYI, starting with 15007, I am not able to do a successful build.
15048 [decoux moulo] This is that it don't like
15054 [radek.bulat ] Rm9yIHJldmlzaW9uIDE1MDI4LCBkdXJpbmcgY29tcGlsaW5nIEkgZ2V0OgoKcnVieSAuL2VuYy9t
15059 [duerst it.ao] The order of the arguments was wrong in one case.
15140 [radek.bulat ] Rm9yIHJldmlzaW9uIDE1MTIyLCBkdXJpbmcgY29tcGlsYXRpb24gSSBnZXQ6CgptYWtlWzFdOiBX

^ Does Symbol#next work unter Ruby 1.9 with intention or by accident?
15049 [ed.odanow wo] Dear language designers and developers!

^ how to "borrow" the RDoc::RubyParser and HTMLGenerator
15050 [phlip2005 gm] I feel like syntax-highlighting some Ruby code, just the same way as RDoc
+ 15053 [phlip2005 gm] I forgot one point: It needs to match.
+ 15060 [drbrain segm] To have RDoc parse a chunk of ruby, call RDoc::RubyParser#scan.  This
  15062 [phlip2005 gm] Notice how each feature (such as Throttle) has an exemplary test case. Then
  15073 [drbrain segm] You shouldn't need to use the HTML generator at all, look in
  + 15082 [phlip2005 gm] Thanks - that's where I'l start back on this next.
  + 15089 [phlip2005 gm] Here ya go!
    15103 [drbrain segm] If these are in the ruby core, fine.  If this requires external
    15105 [phlip2005 gm] Didn't see it in 1.8.6.111 - should I get Ruby 2 ?
    15117 [drbrain segm] Use trunk.
    15129 [phlip2005 gm] This is a multi-part message in MIME format.
    15139 [drbrain segm] This test has too many dependencies.  If your goal is to test
    15141 [phlip2005 gm] I didn't know how to invoke the formatter without warming up every object that
    15142 [drbrain segm] It also involves around 4x more code than assert_equal, which is the
    15144 [phlip2005 gm] This isn't the forum for this topic, but assert_equal will break over irrelevant
    15145 [drbrain segm] White-space changes are completely irrelevant to testing HTML that is
    15152 [phlip2005 gm] That's what I meant; you are agreeing with me here.

^ [BUG]Very strange bug which causes segmentation fault.
15055 [radek.bulat ] SSBmb3VuZCB2ZXJ5LCB2ZXJ5IHN0cmFuZ2UgYnVnIGluIHJ1YnkgMS45LiBJJ3ZlIGRpc2NvdmVy
15067 [radek.bulat ] SGVyZSdzIG15IGFkZHIybGluZSBvdXRwdXQgYmFzZWQgb24gcHJldml1dXMgZHVtcDoKCi9ob21l

^ How to use "addr2line"
15056 [ko1 atdot.ne] Sorry for slow response.  Today, I'll debug VM.
+ 15057 [radek.bulat ] VGhhbmtzIGZvciBleHBsYW5hdGlvbi4gSSBoYXZlIG9uZSBxdWVzdGlvbiByZWdhcmRpbmcgdG8g
+ 15061 [rubys intert] Below is an stack trace based on a clean checkout of revision 15032.
| 15064 [decoux moulo] For this test, the problem is with
+ 15063 [ville.mattil] This is a multipart message in MIME format.

^ warning messages of little consequence
15058 [rubys intert] ruby/1.9.0/net/https.rb:127: warning: `*' interpreted as argument prefix

^ Monotonic scheduler
15065 [zimbatm oree] I have changed the ruby scheduler to use a monotonic clock. This is
15066 [zimbatm oree] And here is the current patch.

^ Ruby magic local-var leak
15068 [zimbatm oree] The participating people have been quite far but at this stage, we

^ [BUG] native_mutex_destroy return non-zero: 16
15069 [rubys intert] This is a multi-part message in MIME format.
15070 [nobu ruby-la] Errno 16 is EBUSY?
15072 [rubys intert] I believe so, yes.
15075 [nobu ruby-la] OK, it seems too strictly.
15076 [rubys intert] Yes!  :-)
15090 [rubys intert] Would it be possible for this patch to get committed?  It would be nice

^ Encoding alias odditities
15074 [rubys intert] I'm giving up building a smaller test case for the moment.  The symptoms

^ [PATCH] SCRIPT_LINES__ issue when loading a file more than once
15078 [rocky.bernst] I while ago on ruby-core there was some discussion about SCRIPT_LINES__ and

^ Sort of a memory leak for Ruby 1.8.6?
15080 [joaopedrosa ] dewd@rubynho:~/loose/find_kill_leak$ /opt/ruby_test/bin/ruby -v found.rb

^ Ruby 1.9 Tinderbox
15081 [rubys intert] ...  Ruby   : 833 tests, 0 failures, 7 known bugs

^ Why @hash ||= Hash.new in Set#initialize?
15083 [dblack rubyp] def initialize(enum = nil, &block)
15088 [dblack rubyp] I guess it would have that effect, but I've seen lots of instance
15116 [ara.t.howard] there is one in that very file ;-)
15119 [dblack rubyp] I meant I don't remember ever seeing @var ||= value in an initialize

^ Incorrect Date.to_s behavior
15084 [vsizikov gma] There was a bug report against JRuby, which turned out against MRI as well.

^ [Bug?] Module/Class definitions cannot take non literal blocks
15092 [robert.dober] I was just about to rewrite pure-traits to be 1.9 compatible just to
15093 [dblack rubyp] => "1.9.0"
15094 [robert.dober] Literally passing a block to ClassOrModule::new works as in 1.8
15095 [decoux moulo] This is a known problem.
15096 [chuyeow gmai] Chu Yeow
15097 [robert.dober] Thank you all, just another question, is there a better way to browse

^ Bug in Date::Infinity#<=>
15098 [d.bussink gm] charset=US-ASCII;
15099 [Daniel.Berge] Is there any sort of mathematical standard that explains what should
+ 15100 [d.bussink gm] Well, I like the current implementation (at least, without the bug).
+ 15112 [brabuhr gmai] I have, since first posting this question discussed the issue with both

^ Rdoc (1.8 and 1.9): Wrong processing if using do blocks in an expression
15101 [ed.odanow wo] This is a multi-part message in MIME format.

^ REXML::Element.write is deprecated.  See REXML::Formatters
15102 [phlip2005 gm] okay, I saw them. Now what?
15104 [rubys intert] That's a bug.  It will be fixed in the next release of REXML, which
15106 [phlip2005 gm] Thanks but...
+ 15110 [rubys intert] I guess I'll let Sean take it from here, other than to note that .to_s
| 15111 [phlip2005 gm] I understand you only mean that's a workaround - it is indeed already deployed.
| 15135 [ser germane-] Your rationale, or mine?
| 15136 [ser germane-] Sorry about that.  If I'm not careful about my SMTP host, my company
+ 15146 [ser germane-] charset="iso-8859-1"

^ gets is blocking print in simple example
15107 [dblack rubyp] print "Hello. Please enter your name: "
+ 15126 [dblack rubyp] It's certainly possible to do that, but it seems awfully heavy for
+ 15127 [matz ruby-la] Fixed today.  Thanks for pointing it out.

^ ruby-1.8.6-p111/lib/rdoc/markup/test fails
15108 [phlip2005 gm] just out of curiosity...

^ RDoc tests Was: Re: how to "borrow" the RDoc::RubyParser and HTMLGenerator
15109 [jan.svitok g] the others at http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/branches/ruby_1_8/lib/rdoc/markup/tests
15118 [drbrain segm] Code in the ruby_1_8 branch will be left as it is.  Look in trunk,

^ Ruby 1.8/1.9 performance tests?
15113 [ks kurtsteph] I'm looking for general performance tests that will run in both ruby 1.8
15115 [znmeb cesmai] I think ActiveRecord hits the GC fairly hard, but I don't know if it's
15137 [ks kurtsteph] Considering the ActiveRecord will introduce dependency on a external

^ Ruby 1.9: PATCH: Fixnum#gcd (Reprise)
15114 [ks kurtsteph] This is a multi-part message in MIME format.

^ Bug with ruby1.9
15120 [dohzya gmail] I've found a bug in ruby1.9 with a semantic error.
15121 [decoux moulo] Thanks, it's a known bug and was already reported in [ruby-core:14641]

^ IRHG - What is the second parameter used for in gc_mark(ptr, lev)?
15124 [ceo hawthorn] The lev parameter has been added in the last several major versions-

^ test_m17n segmentation fault
15125 [rubys intert] I don't know if anybody else is seeing this, but on a build of 15098 I'm
15128 [matz ruby-la] I fixed similar problem today.  Try r15104 or later.
15130 [rubys intert] That fixed it!

^ test/rdoc/test_simple_markup_attribute_manager.rb:2:in `require': no such file to load -- rdoc/markup/simple_markup/inline (LoadError)
15131 [akr fsij.org] test-all failed as follows.
15132 [phlip2005 gm] Somebody partied with the RDoc modules and files without (>ahem<)
15133 [drbrain segm] Yes, the tests are not very helpful for finding problems since they
15134 [phlip2005 gm] patches

^ Build error, revisions 15119-15126
15143 [rubys intert] STDERR.reopen(STDOUT)

^ String initialziation
15147 [mailing.mr g] Mayby strange question but what happens when someone for example types
15148 [lopx gazeta.] As a rule 'new' (which in turn calls initialize) is not called when
15149 [gwtmp01 mac.] Just to connect the dots, in the irb scenario, the line of text is =20
15150 [mailing.mr g] That is what i wanted to know exactly.
+ 15151 [lopx gazeta.] Not without hacking the interpreter, (take a look at rb_newobj and
+ 15154 [duerst it.ao] What's your use case? Just curious.
+ 15156 [ed.odanow wo] I don't think so, even "set_trace_func" says nothing :-(
  15157 [mailing.mr g] I thought that it might be possible to make translating of Ruby programs
  + 15158 [mailing.mr g] It's not - at least not everywhere, looks like in some places to_s is
  + 15159 [phlip2005 gm] How to distinguish user-oriented strings from internal strings?
    15161 [mailing.mr g] touche, there's few solutions but you're right
    15162 [phlip2005 gm] I had guessed it already did!
    15163 [phlip2005 gm] uh, OLPC?

^ [Fwd: IRHG - GC_MARK Question]
15153 [ceo hawthorn] This is a multi-part message in MIME format.

^ an example of performance improvements
15155 [duerst it.ao] For those not reading ruby-dev, I just wanted to point to
+ 15165 [nobu ruby-la] The example code in [ruby-dev:33189] is one of worst cases.  It
+ 15168 [pbrannan atd] AFAIK, the cause of the non-linear performance I mentioned in
| 15171 [pbrannan atd] Oops, that should be [ruby-core:13246].
+ 15175 [vincent.isam] I think this example can be shorten much more to see the real cause of
  15179 [pbrannan atd] I wonder if this is related to the problem I'm seeing, which seems to
  15180 [charles.nutt] Shot in the dark: more GIL lock contention since threads can actually
  15184 [pbrannan atd] * vm.h: remove dangerous assembler sentence.

^ SimpleDelegator.class changed?
15160 [rubys intert] Should the following program produces different results with Ruby 1.8.6

^ convert rubynode transforms back into source code?
15164 [phlip2005 gm] If this sounds like the stupidest _advanced_ question asked in the history of
+ 15166 [luislavena g] ParseTree + Ruby2Ruby?
| 15169 [phlip2005 gm] The great thing about rubynode is you can reach out and squeeze your own
| 15178 [ryand-ruby z] You can do that with ParseTree as well. Take a look at pt_testcase.rb
+ 15167 [pbrannan atd] Both ruby2ruby and nodewrap support this.
  15170 [phlip2005 gm] Sorry - gotta use rubynode.
  15176 [dbatml gmx.d] I am the author of rubynode. As far as I know there currently is no code
  15181 [phlip2005 gm] Tx. I will also look at the 1.9 issue.

^ Vectored Exception Handling?
15172 [djberg96 gma] I hadn't heard of Vectored Exception Handling before, but I came across
threads.html
top