6966-7294 subjects 7383-8402

[ ruby-Bugs-3264 ] YAML produces many warnings
7169 [noreply@ru y] Bugs item #3264, was opened at 2006-01-16 19:27

[ ruby-Bugs-3267 ] Problem with Ruby/DL and double arguments (non-x86)
7170 [noreply@ru y] Bugs item #3267, was opened at 2006-01-17 13:37

yaml and Bignum
7171 [vjoel@pa h. ] The above fix seems to still be needed in ruby-1.8.4, to fix the
7172 [vjoel@pa h. ] irb(main):007:0> class Bignum
7173 [ocean@m2 cc ] Can you try latest ruby_1_8 branch?
7176 [ryand-ruby@z] % ./ruby ./runruby.rb -v -ryaml -e 'p YAML.load(YAML.dump

Still $SAFE
7177 [lists@be tr ] as many of you probably noticed, I didn't get an answer to
7258 [matz@ru y- a] Sorry somehow I missed your posts.

[PATCH] Add XHTML 1.0 Output Support to Ruby CGI
7178 [pabs@pa lo r] The attached patch against Ruby 1.8.4 adds XHTML 1.0 output support to
7180 [ruby@ja es r] If I understand the XML prolog production rule correctly, the (optional)
7192 [wilsonb@gm i] I've always simply left the XML declaration off, since it melts
7193 [pabs@pa lo r] on,

[PATCH] Improved XHTML 1.0 Output (was Re: [PATCH] Add XHTML 1.0 Output Support to Ruby CGI)
7182 [pabs@pa lo r] [snipped]

[ ruby-Bugs-3302 ] irb cannot parse /\^/
7183 [noreply@ru y] Bugs item #3302, was opened at 2006-01-23 02:43

BigMath::log speedup patch
7184 [erlercw@gm i] Someone else submitted a BigMath::log speedup at

Wilderness: NODE Obj and Class Obj --- Can they be ID'ed by inspection?
7185 [ceo@ha th rn] I think I know the answer, but...
7194 [cmills@fr es] Yes.

[PATCH] Ruby 1.9 and FHS
7186 [k.shutemov@s] - pure-ruby files install separete from ruby extentions(binary).
7187 [halostatue@g] I don't think that this is a useful modification. If a particular
7188 [k.shutemov@s] I don't think that scrap-heap of header, rb and binary files is very
+ 7189 [halostatue@g] That's your opinion. The current way it works is, however, very easy to
+ 7190 [now@bi wi se] I agree.  I keep going into /usr/include to look for the Ruby headers.
  7191 [halostatue@g] Except that Ruby is pre-FHS, and FHS is a provincial standard not

[ ruby-Bugs-3317 ] trouble due ruby redefining posix function eaccess
7195 [noreply@ru y] Bugs item #3317, was opened at 2006-01-24 15:33
7204 [nobu@ru y- a] Why `throw' in C?
7206 [halostatue@g] It may be compiled with a C++ compiler and "throw" is somehow defined
7207 [mrueckert@su] yes. as mentioned on #ruby-lang we triggered it with compiling a koffice
7214 [nobu@ru y- a] Okey, anyway I committed a fix to check the presense of

[ ruby-Bugs-3318 ] net/https does not submit RFC compliant requests
7196 [noreply@ru y] Bugs item #3318, was opened at 2006-01-24 16:52

SSL-enabled DRb fds on SSLError?
7197 [ctm@ar i. om] Howdy,
7198 [drbrain@se m] ...
+ 7199 [ctm@ar i. om] Adding soc.close there works fine under 1.8.4 from the Fedora Core 4
+ 7200 [ctm@ar i. om] Patch worked fine against HEAD.
  7202 [drbrain@se m] Now you just need somebody to check it in.
  7213 [matz@ru y- a] Hmm, Eric, don't you have an account?
  7221 [drbrain@se m] I do not!
  7225 [matz@ru y- a] OK, I will apply.

bcc32's memory manager bug
7203 [ocean@m2 cc ] I found serious bug of bcc32. Following code crashes.

Launching Ruby scripts and the future of MVM
7205 [headius@gm i] It has come to our attention that there may need to be a standard way

bug - Ruby 1.8.4 Stack problem
7208 [bryant.doug@] ...
7209 [pav@Fr eB D.] sting.
7210 [bryant.doug@] Thanks  Pav.  Running against the build without pthreads fixed the

Some troubles with an embedded ruby interpreter
7211 [matt.mower@g] I posted this message to ruby-talk yesterday but didn't receive any
+ 7212 [matt.mower@g] Hrmm...
+ 7215 [wvdbos@go da] ...
  7219 [matt.mower@g] Ah. I was working from the information in Programming Ruby (2nd ed)

[ ruby-Bugs-3329 ] String#scan loops forefever if scanned string is modified inside block.
7216 [noreply@ru y] Bugs item #3329, was opened at 2006-01-26 10:55

[ ruby-Bugs-3331 ] YAML segfault when loading unknown subclass of builtin class
7217 [noreply@ru y] Bugs item #3331, was opened at 2006-01-26 09:42

[ ruby-Bugs-3332 ] YAML cannot restore instance variable of subclass of Hash/Array/String
7218 [noreply@ru y] Bugs item #3332, was opened at 2006-01-26 09:55

[PATCH] Clarify String#scan Documentation (was Re: [ ruby-Bugs-3329 ] String#scan loops forefever if scanned string is modified inside block.)
7220 [pabs@pa lo r] block.
7222 [nobu@ru y- a] * Scan frozen shared string.

[ ruby-Bugs-3344 ] RDoc.usage doesn't parse caller correctly on windows
7223 [noreply@ru y] Bugs item #3344, was opened at 2006-01-27 14:26
7224 [Daniel.Berge] A better solution is this: main_program_file = caller.last.split(":")[-2]

Fwd: Re: Question about massive API changes
7226 [ser@ge ma e-] charset="iso-8859-1"
+ 7232 [dblack@wo bl] What's a good antonym for "agile"? :-)
| + 7235 [ser@ge ma e-] charset="iso-8859-1"
| + 7236 [matju@ar en ] What's a good way to say "big kicking upfront a dead whale down the
+ 7248 [ceo@ha th rn] 1) I personally like total duplication,  the code is all in one place
  7253 [lukfugl@gm i] Amen. It's already been determined that 2.0 will be a non-compatible
  7255 [benschumache] Sean-

Re: Question about massive API changes
7227 [gsinclair@gm] My instinct: new library "REXML2".  It's a hack of a version-migration
+ 7228 [caleb@ae -t ] I'm not a Dave, but I agree with Matz' #4.
| + 7229 [ser@ge ma e-] charset="iso-8859-1"
| | 7230 [caleb@ae -t ] True, but I'll venture a guess that very few apps are going to
| + 7233 [wilsonb@gm i] My opinion is rather extreme: if Ruby is already at a point where
|   + 7234 [ser@ge ma e-] charset="iso-8859-1"
|   | + 7238 [djberg96@ya ] I understand it completely.  We cannot continue to be
|   | + 7239 [wilsonb@gm i] That's what I'm saying.  The majority of C code compiles perfectly
|   | + 7242 [meta@po ox c] Is there some reason why a thin API facade is impossible? You wouldn't
|   |   7252 [ser@ge ma e-] charset="iso-8859-1"
|   |   7267 [meta@po ox c] Something that accepts calls via the old API, and translates them into
|   |   7269 [ser@ge ma e-] charset="iso-8859-1"
|   |   7299 [james@gr yp ] What about using lazy evaluation for the old style methods?  The
|   + 7237 [caleb@ae -t ] How is this any different than picking Ruby 2.0 as your new language ?
|   + 7241 [meta@po ox c] This is exactly the issue I was trying to discuss in the thread about
+ 7231 [dblack@wo bl] Please not a '2' library.  There are already so many ruined names of

[ ruby-Bugs-3361 ] Ruby 1.8.4 parse errors
7240 [noreply@ru y] Bugs item #3361, was opened at 2006-01-29 17:09

waking up a Ruby thread asynchronously?
7243 [joshua@re er] I am writing a C extension for Ruby that wraps a networking API.  I

Ruby IPC
7244 [marcantonios] ...
+ 7245 [drbrain@se m] ruby-talk is the correct forum for this.  I answered your email over
+ 7246 [vjoel@pa h. ] Eric's suggestions are probably the best place to start. Ask your

[ ruby-Bugs-3367 ] More relaxed checking for XML RPC Content-Type value
7247 [noreply@ru y] Bugs item #3367, was opened at 2006-01-30 12:51

PATCH: append option to sysread
7249 [ysantoso-rub] ...
+ 7250 [akr@m1 n. rg] It seems possible core dump.
| 7251 [ysantoso-rub] rb_str_locktmp(str);
| 7263 [akr@m1 n. rg] There is check code.
+ 7257 [rubyfan@gm i] Maybe instead of the patch you send, this should be made into an

Struct#select behavior vs rdoc
7254 [ryand-ruby@Z] % ri Struct#select
7256 [matz@ru y- a] Yes.

TCP/UDP server weird lags on 1.8.4 linux
7259 [billk@ct .c ] I have been debugging this for a few days, and finally ran strace on
+ 7261 [billk@ct .c ] Note, when I remove the UDP socket I/O entirely, the problem seems
+ 7262 [caleb@ae -t ] Bill,
  7264 [billk@ct .c ] Interesting, thanks.  I added that, but the strange delay still
  7265 [caleb@ae -t ] The other thing I did was to add the host I was connecting to to
  7266 [billk@ct .c ] Thanks Caleb!!!

possible bug - does Socket.gethostbyname behave correctly?
7260 [sroberts@un ] Socket.gethostbyname gethostbyname(3), which is surprising. It can

Re: PATCH: append option to sysread -- retracted
7268 [ysantoso-rub] if (RSTRING(str)->len != initial_len + ilen) {

[ ruby-Bugs-3399 ] [PATCH] OS X core dumps when $0 is changed and then loads shared libraries
7272 [noreply@ru y] Bugs item #3399, was opened at 2006-01-31 22:25
7285 [ocean@m2 cc ] Interesting, but Starndard C gurantees argc and argv should be modifiable,
7291 [blair@or aw ] ...

[patch] Symble syntax support of RDoc parser
7281 [usa@ga ba ec] I found that the process of generation of RDoc documentation
7284 [matz@ru y- a] Could you commit this?

Re: ruby-dev summary 28206-28273
7286 [ara.t.howard] # predicate methods to avoid double evaluation of expr
+ 7287 [Daniel.Berge] seconds before I saw this.
| 7288 [ara.t.howard] too many computer science classes out there i guess.  ;-)
+ 7289 [Daniel.Berge] Nice.  If only "if" were a Kernel method that we could tie it to back to
+ 7296 [cboos@bc -t ] if (elt = a[1]) and elt.strip.empty?

[RCR] meaningful RUBY_VERSION#<=>
7290 [ara.t.howard] i'm sure something much more elegant could be made, but this summarizes desired

ANDCALL / iff? / &? (was Re: ruby-dev summary 28206-28273)
7292 [meta@po ox c] I have to say, I don't like any of these. Personally, I write
7293 [meta@po ox c] Sorry, that should of course have been
7298 [ruby@ja es r] I'm not too keen on terse syntax; clarity for the read takes precedence
7310 [evanwebb@gm ] I'm not sure we even need the 'with' syntax. Even if we do, it breaks
+ 7311 [ruby-ml@ma i] A minor nitpick, you need
| + 7313 [evanwebb@gm ] Umm, on what version are you seeing a warning there? I don't and never
| | 7315 [ruby-ml@ma i] Only when dealing with literals, sorry.
| | 7316 [evanwebb@gm ] I'd by far prefer it never emit a warning. The warning is assumes you
| | 7317 [meta@po ox c] I'd like -w to continue to give me as many warnings as possible. If
| + 7314 [drbrain@se m] $ ruby -vwe 'a = ""; if t = a[0] and t.strip.empty? then end'
+ 7319 [ruby@ja es r] VB users never seemed to have a problem with the concept of 'with', and
  + 7320 [ara.t.howard] can you explain that?
  + 7323 [evanwebb@gm ] Having never used VB, I had no idea it had a with. Well, then thats at

possible bug - does Socket.gethostbyname behave correctly?
7295 [sroberts@un ] Socket.gethostbyname gethostbyname(3), which is surprising. It can
7309 [matz@ru y- a] It is pretty easy to apply your patch.  But I'd like to hear from
7312 [marcantonios] I've had this issue creep up into other projects as well.  I usually

YAML::quick_emit api change
7297 [ara.t.howard] i had a method like this in my OrderedHash class

[ ruby-Bugs-3438 ] (2**32).to_s(2) fails with exception on Redhat Opteron with gcc 3.4.3 compile
7300 [noreply@ru y] Bugs item #3438, was opened at 2006-02-03 18:41
7301 [ville.mattil] 3.4.3
7302 [ocean@m2 cc ] I have fixed this problem on CVS. Please try latest ruby_1_8 branch.
7303 [ville.mattil] The CVS works as expected.

[ ruby-Bugs-3453 ] PStore#fetch does not reture default value.
7304 [noreply@ru y] Bugs item #3453, was opened at 2006-02-05 12:40

Re: Problem with weak references on OS X 10.3
7305 [ mfp@ac .o g] 0x1d421c.to_s(2)                                   # => "111010100001000011100"

Fwd: Is the SQLite3 gem known to have memory corruption?
7306 [tjw@om ig ou] ...

crash with ruby cvs head
7307 [matju@ar en ] Ruby, I get various errors such as segfaults or unknown-nodes or
7308 [matz@ru y- a] ftools.rb comes with the distribution never has a END{} block.

[RCR] ruby --quiet; was Re: ANDCALL / iff? / &? (was Re: ruby-dev summary 28206-28273)
7318 [ara.t.howard] i agree.  however -w should not be changed.  simply do
+ 7321 [tjw@om ig ou] These two statements seem contradictory to me.  If -w is useful,
| 7322 [ara.t.howard] array bounds checking is useful - but it's often desirable to turn it off
| 7324 [evanwebb@gm ] I think you've hit the point exactly, -w is only really useful at
| 7327 [ara.t.howard] agreed.  but -w is not 'off' be default.  it's as in -W1.
+ 7325 [eban@os ri .] % ruby -h | grep -e -W
  7326 [ara.t.howard] indeed - i had just noticed that - my mistake.  i guess my question is why -W1

DRbObject.new(nil, nil)
7328 [vjoel@pa h. ] If the uri argument to DRbObject.new is nil, then the error raised is
7329 [vjoel@pa h. ] vjoel : Joel VanderWerf : path berkeley edu : 510 665 3407

[ ruby-Patches-3503 ] Correction to string-to-float conversion of string with trailing 'e'
7330 [noreply@ru y] Patches item #3503, was opened at 2006-02-08 12:31

[ ruby-Bugs-3506 ] Set containing duplicates
7331 [noreply@ru y] Bugs item #3506, was opened at 2006-02-08 22:52
7332 [Daniel.Berge] require "set"
7333 [vjoel@pa h. ] irb(main):020:0> s = Set[]
7334 [vjoel@pa h. ] I don't think it's really a bug. If you have a set with members (like
7335 [ruby-ml@ma i] Might it be appropriate to add a similar method to Set, then?

SOAP errors for boolean types
7336 [panzalone@gm] I have generated a SOAP client using wsdl2ruby.rb and have established

Parse error within Regexp
7337 [lists@be tr ] user@host:~$ ruby -e '"".scan( / ,/)'
7338 [matz@ru y- a] Ruby interpreted '"".scan / ,/' as
+ 7339 [lists@be tr ] Ah, that gives me insight and I agree this interpretation
+ 7340 [ mfp@ac .o g] Would something like this make sense?
  7341 [dblack@wo bl] I'm not sure the "unterminated string" makes sense as the error
  7342 [ mfp@ac .o g] I think the second '/' marks the start of the regexp, as seen by
  7343 [dblack@wo bl] Perhaps, but I don't think "unterminated regexp" describes the
  7346 [ mfp@ac .o g] The point is that both cases (actual, and spurious unterminated regexp

Ruby 1.8.4 on Mac OS X 10.4 Intel
7344 [daesan@gm il] I tried compiling Ruby 1.8.4 on Mac OS X 10.4 Intel.  While it
7345 [langstefan@g] zlib.c:440: warning: pointer targets in assignment differ in

Latest change to eval.c
7347 [ksruby@gm il] It seems that the latest change to eval.c (1.616.2.154) has broken irb.
7349 [matz@ru y- a] Sorry, fixed.
7354 [ksruby@gm il] Thanks, Matz.
+ 7355 [drbrain@se m] Is your ri up-to-date?
| 7356 [ksruby@gm il] Yep, I've just rebuilt ri database with
+ 7357 [matz@ru y- a] Hmm, I saw this error during debugging.  Could you remove all object
  7358 [ksruby@gm il] Yeah, I should have done it in the first place. I've check out it
  7359 [matz@ru y- a] Forgive me.  I checked in the fix right now.
  7360 [ksruby@gm il] Perfect. Thank you very much, it fixed compile problem and this one

Socket.getaddrinfo & Bonjour (Was: Re: Ruby 1.8.4 on Mac OS X 10.4 Intel)
7348 [drbrain@se m] => [["AF_INET6", 80, "kaa.local", "fe80::211:24ff:fe1f:37b6%en1", 30,
7350 [daesan@gm il] Eric, are you also using Ruby 1.8.4 on Apple's new iMac Core Duo?  If
7352 [drbrain@se m] Yes

[ ruby-Bugs-3539 ] typo in net/ftp.rb documentation
7351 [noreply@ru y] Bugs item #3539, was opened at 2006-02-13 11:17

more restrictive options for test/unit/autorunner.rb ?
7353 [dev@sa iz at] is there a way, or interest in creating a way, to add switches to

7361 [ara.t.howard] i was thinking about patching io.c and ext/fcntl/fcntl.c so one could do

undefining Class.inherited makes class not subclassable anymore
7362 [tilman@co e-] 1.) cFoo =3D rb_define_class ("Foo", rb_cObject);
7363 [dblack@wo bl] A semi-guess, but I think you could use rb_remove_method instead of
7387 [tilman@co e-] Thanks, rb_remove_method() does the trick :)

[ ruby-Bugs-3565 ] Method object used as Object#instance_eval block doesn't work (as expected)
7364 [noreply@ru y] Bugs item #3565, was opened at 2006-02-15 02:32
7365 [matz@ru y- a] The code says the receiver was passed as an argument by intention, but
7371 [ocean@m2 cc ] Fowarding.... (Ozgur Murat Homurlu)
7374 [matz@ru y- a] A method object is tightly bound to its receiver.  Even instance_eval
7375 [ozgurmurath@] I see. This is the the answer I was expecting. I was just experimenting
+ 7380 [ozgurmurath@] Here's the patch which documents the behaviour of instance_eval when a
+ 7381 [ozgurmurath@] Sorry, resending the patch because its header is previously corrupted...
  7388 [matz@ru y- a] Merged.  Thank you.

OP_ASGN1 and OP_ASGN2 allow calling of private operators and problem with splat
7366 [dbatml@gm .d] irb(main):001:0> class Fixnum;private :*;end

Ruby 1.8 CVS miniruby core dumps
7367 [ville.mattil] On
7369 [matz@ru y- a] Couldn't reproduce your problem on my machine.  Probably Solaris
7370 [ville.mattil] I didn't found frame, did you mean ruby_frame?
7372 [matz@ru y- a] The compiler might optimized out the local variable.  But it's OK.
7373 [ville.mattil] Yes it works great. All tests passed! Thanks for quick fixing.

Ruby 1.8 CVS miniruby core dumps
7368 [ville.mattil] I resend this because my mailer set incorrect Date.

Minor tracer.rb patch
7376 [Daniel.Berge] Here's a patch that gets rid of the unnecessary (and occasionally problematic)
7379 [matz@ru y- a] The small code at the bottom should not be removed, since it's for
7382 [Daniel.Berge] Who would do this other than the author?  I don't think too many folks are
7384 [matz@ru y- a] Whatever.
7389 [akr@m1 n. rg] After tracer.rb changed, make install-doc generates many output.
7391 [matz@ru y- a] My mistake. Fixed. Thank you.

Re: patch for rational.rb (gcd)
7378 [matju@ar en ] The version of gcd that uses >>=, which is often called "binary gcd", may
7392 [matju@ar en ] What's differing in versions of Ruby so that the difference in execution
7393 [dbatml@gm .d] class Integer
7394 [matju@ar en ] Wow! that must be the slowest implementation of GCD I've seen since...