4101-8816 subjects 4555-4797

[PATCH] RDoc: Show parsing progress for C files
4341 [tilman@co e-] the attached patch makes RDoc show parsing progress for C files like the

[BUG?] Ripper
4345 [k.shutemov@s] In attachment is a test. I think this test should but now it isn't. Is it

[BUG] Segmentation fault
4346 [andrew@wa ro] FYI, just got this random unexpected crash
+ 4347 [drbrain@se m] It would mean more with a backtrace, and if your Ruby was compiled with
| 4348 [matz@ru y- a] And if the error reproducing Ruby program (along with input data) is
| 4349 [andrew@wa ro] Sorry guys. I always try my best to provide a patch, a testcase etc (google me
| 4350 [drbrain@se m] You should be able to at least get a core-file you can examine with
| 4352 [andrew@wa ro] All useful advice .... If I could reproduce the bug ...  ;)
+ 4351 [lists@be tr ] Sample code to reproduce it?
  4354 [andrew@wa ro] Contrary to (it seems) popular belief, not all bugs can be reduced to a
  4355 [lists@be tr ] Yes, I know them and I like them ...
  4356 [ville.mattil] What optmization level you are using when compiling ruby? If it is O3, then
  + 4357 [akr@m1 n. rg] It fix immediate core dump with -O3 on my environment.  But Ruby is
  + 4358 [andrew@wa ro] I'm using -O2, so should be OK
    4361 [akr@m1 n. rg] -O0 may avoid some core dump I don't know.

problems with my skyos ruby port
4353 [Dannerbeck_D] i start a port of ruby 1.6.8 to skyos. the current version will follow!

Adding lastlog info to etc
4360 [Daniel.Berge] Any chance of adding lastlog information to the etc module?  I have a
4362 [nobu.nokada@] Do you think it should be included in Passwd struct?
4363 [djberg96@ya ] Yes.
4372 [nobu.nokada@] I can't agree.
4374 [djberg96@ya ] Ok, I can live with that.  Thank you. :)

windows socket connection freeze
4364 [wmorgan-ruby] Dear experts,
+ 4367 [jean-francoi] My patch resolves this issue. That patch was never integrated in the CVS
+ 4371 [usa@ga ba ec] It's fixed at CVS HEAD.
  4376 [wmorgan-ruby] Thank you. As things stand now, RubyTorrent behaves very poorly on
  4389 [ville.mattil] I have tested Nakamura's fix, and it seems to work ok, but the connect
  4400 [usa@ga ba ec] I found the bug, and fixed it on CVS HEAD.
  4401 [ville.mattil] I tested the latest fix and I got exactly same message  proper
  4427 [ville.mattil] Matz,
  + 4446 [wmorgan-ruby] +1. This is hurting RubyTorrent, Lektora, and basically anything that
  + 4447 [matz@ru y- a] Oops, I forgot to tell him OK.  Please commit if it's not done yet.
    4456 [usa@ga ba ec] Done.
    4459 [ville.mattil] Yes it is! Here is the test result. Works fine!

'when (cond):' causes SyntaxError
4368 [nakahiro@sa ] The following script causes SyntaxError under 1.9, only at 'when
+ 4369 [matz@ru y- a] A bug.
| 4379 [nobu.nokada@] A patch, to separate EXPR_BEG to 2 cases; beginning of
| 4381 [matz@ru y- a] Wow, huge patch.  I think a simple one like this would work.
| 4382 [nobu.nokada@] It simply prohibits labels.
| 4388 [matz@ru y- a] Oh, you are right.  Commit the fix, please.  But you don't have to
+ 4370 [nakahiro@sa ] Not 'after' but 'before'.

[ ruby-Bugs-1467 ] the latest HEAD gives excessive warnings whe compiling ext/dl  extension
4377 [noreply@ru y] Bugs item #1467, was opened at 2005-02-07 10:25

No stdout-flush before gets
4378 [lists@be tr ] user@host:~$ ruby -e "print '? ' ; gets"
4380 [akr@m1 n. rg] No.  1.9 flushes $stdout before reading stdin, now.
4384 [lists@be tr ] Ah, thank you.

[ ruby-Patches-1469 ]  [doc patch->object.c] instance_variable_set will create an instance var
4383 [noreply@ru y] Patches item #1469, was opened at 2005-02-07 11:40

[PATCH] add color_set support to curses.c
4385 [pabs@pa lo r] I'm not sure why this is missing from the Curses binding, but the
4387 [matz@ru y- a] Do you know whether wcolor_set is available when USE_COLOR is defined?
4393 [pabs@pa lo r] Err, USE_COLOR is defined in curses.c.  If you're asking whether or not
4396 [matz@ru y- a] That's exactly what I wanted to know.  Thank you.
4398 [usa@ga ba ec] Ah, I know the environment which have USE_COLOR definition

mknod, mkfifo
4390 [lists@be tr ] the functions `mknod' and `mkfifo' are still missing.
4394 [nobu.nokada@] You forget a patch to configure.in.  See also [ruby-talk:91735].
4395 [lists@be tr ] Yes.

Vim syntax highlighting for Ri
4391 [lists@be tr ] I started to write a Vim syntax highlighting for Ri output.

HTTP Basic authentication for open_uri
4392 [ksibilev@be ] Can somebody apply the following patch for open_uri in order to enable
4397 [akr@m1 n. rg] Since the format is deprecated, open-uri doesn't support the format.
4399 [ksibilev@be ] OK, I didn't know about it. In this case, should it still provide a
4416 [akr@m1 n. rg] Implemented.
4417 [ksibilev@be ] Much appreciated!

BUG: Struct.new(:a?).instance_methods
4402 [csaba-ml@be ] be the proper reaction upon doing what's in the subject.
4404 [matz@ru y- a] Right.  "a?" should not be allowed for a struct member name, since
4405 [hgs@dm .a .u] Matz, would it be possible to change the message to guide people to
4406 [matju@sy pa ] Also, attr_accessor and its relatives should be subject to the same thing.
+ 4407 [george.marro] And if there's any chance of it, I'd really love to see x? local variables
+ 4408 [chneukirchen] += 1

Re: Unknown OS X 10.2 Socket constants (+script to generate)
4403 [sroberts@un ] Here are all the OS X 10.2 consts from sys/socket.h that it looks like
4409 [sroberts@un ] Thanks for putting in SO_REUSEPORT.

Re: problems with ./config.guess: uanble to guess system type
4411 [Dannerbeck_D] Thank you so much for your contact:) Ok. The ./configure Script from Version 1.6x works now by simple typing
4413 [Dannerbeck_D] Yea. It use the PE format. It's a makro that we implement in the make file. It makes  the  *.app  files.

[PATCH] RDoc - show protected methods by default
4412 [tilman@co e-] atm, RDoc will only show/document protected methods if the "-all" switch
4420 [matz@ru y- a] I know Dave Thomas is too busy to maintain RDoc, and it seems

[PATCH] Doc patch.
4414 [g_ogata@op u] FileUtils and String.
4419 [matz@ru y- a] Thank you.

[PATCH] ruby-mode.el heredoc indentation
4415 [g_ogata@op u] def foo
4418 [matz@ru y- a] Thanks.  I will merge it.

Passing a block to methods in mkmf.rb
4421 [Daniel.Berge] I'm curious as to what the optional block is for in the mkmf.rb methods
+ 4423 [decoux@mo lo] try it with
+ 4425 [nobu.nokada@] The block must return new source code, not nil.

[ ruby-Patches-1486 ] have_enum_member for mkmf.rb
4422 [noreply@ru y] Patches item #1486, was opened at 2005-02-11 09:47
4426 [nobu.nokada@] It doesn't check enum membership -- rather, it's impossible
4428 [matz@ru y- a] Is const always int?  If so, I'd like to apply the patch.
4430 [Daniel.Berge] I just noticed - you'll want to replace "type" with "const". :)

page reference document
4424 [akr@m1 n. rg] I found 2 page reference in io.c and process.c.

XMLRPC::FaultException directly inherits Exception
4429 [sheepman@tc ] In lib/xmlrpc/parser.rb, XMLRPC::FaultException directly inherits Exception.

SecurityError on Method#call? (might be DRb specific)
4431 [florgro@gm i] Moin.
4433 [pit@ca it in] I've no idea either, but what are the differences between your environment and

[bug?] curses + threads = non-blocking getch
4432 [wmorgan-ruby] Using the curses library, it looks like non-blocking getch (i.e. when
4435 [akr@m1 n. rg] It is not a non-blocking I/O.  It checks an available input before
4436 [wmorgan-ruby] Thank you, that works.
4437 [akr@m1 n. rg] I don't say it's a bug or not.  I leave the decision for the

[ ruby-Bugs-1498 ] win32ole: problem with unicode
4434 [noreply@ru y] Bugs item #1498, was opened at 2005-02-15 11:29

[ ruby-Bugs-1502 ] FilUtils.remove_dir doesn't change the rights if it fails
4438 [noreply@ru y] Bugs item #1502, was opened at 2005-02-16 10:41

Thread-safe Ruby Status?
4439 [vincent.isam] I saw in the ChangeLog in the CVS trunk some recent modifications
+ 4440 [matz@ru y- a] _THREAD_SAFE means to avoid crashing when pthread enabled library is
| 4442 [vincent.isam] And regarding the thread-safeness of the interpreter itself - by that
| 4443 [matz@ru y- a] Not quite.  Removing global variables itself is not that difficult,
+ 4441 [nobu.nokada@] No.  It's just to load extension libraries linked with pthread

[ ruby-Bugs-1504 ] Open3.popen() doesn't set $?
4444 [noreply@ru y] Bugs item #1504, was opened at 2005-02-16 11:40
4445 [matz@ru y- a] open3 uses the "double fork" technique, so that you cannot take the
4467 [zn@mb .n ft ] ...

[PATCH] add persistent history to irb
4448 [dblack@wo bl] I've often wanted the up-arrow in irb to give me my history from the
+ 4449 [drbrain@se m] How will this conflict with using the Readline powered history?
| 4450 [dblack@wo bl] Cool -- I hadn't seen that one before.
| 4451 [drbrain@se m] Ah!  I thought your version was Readline-free, because it made no
| 4452 [dblack@wo bl] The checks have already been performed by the time you get to the part
+ 4454 [uehli@bl ew ] There is already a history file support in 1.9. Just set SAVE_HISTORY=true in
  4455 [dblack@wo bl] Oh, cool.  Thanks for letting me know.

bug in IRB with $_ matching a range of regexps
4453 [ryand-ruby@z] ryand-ruby@zenspider.com - Seattle.rb -
4457 [nobu.nokada@] This shorthand is only for one liners, i.e., scripts given by
4458 [ryand-ruby@z] I'll update my quickref.
4460 [nobu.nokada@] $ echo 'while gets; print if /^b/../^c/;end' > foo.rb
4462 [matju@sy pa ] How old? I recall it's in the Pickaxe 1 (Ruby 1.6.2) and it was not even

[ ruby-Bugs-1531 ] rdoc/usage fails if file name includes colon (e.g. 'C:/dir/prog.rb')
4461 [noreply@ru y] Bugs item #1531, was opened at 2005-02-22 10:05

Strange argc check in stable snapshot
4463 [Daniel.Berge] I'm just curious why this type of thing has appeared in the latest
+ 4464 [matz@ru y- a] Right.  It is unnecessary check.  Thank you.
| 4466 [steven.jenki] Maybe a good idea nonetheless. It might catch a bug someday. Nothing
| 4468 [Daniel.Berge] I'll pledge $100 to RubyCentral immediately if someone can actually
| + 4469 [sroberts@un ] If you don't trust your runtime, nothing is guaranteed... and you are
| | 4471 [steven.jenki] The argument value in question (in array.c) is not set by the runtime.
| | + 4472 [Daniel.Berge] Not exactly what I meant, but I'll donate $100 anyway.  I would consider
| | | 4473 [steven.jenki] Of course it's a bug. That's what the test is for.
| | + 4474 [sroberts@un ] ^^^^                     ^^^^^^^^^^^^^^^
| |   4475 [Daniel.Berge] Actually, this is all the more reason to use rb_scan_args() over
| |   4476 [dblack@wo bl] This all goes to Ruby Central, Inc., right? :-)
| |   + 4477 [Daniel.Berge] Yep.  The check is in the mail - for real. :)
| |   + 4478 [steven.jenki] No, I meant Ruby Central dot uh, uh, cat. Yeah, that's it--the charity I
| |     4481 [dblack@wo bl] Being a space cadet, I hadn't even caught the original Ruby Central
| + 4470 [steven.jenki] int main(int argc, char *argv[])
|   4552 [pbrannan@at ] Program startup
+ 4465 [matju@sy pa ] According to the gospel, demonic possession is a rather common thing, and

Requesting addition to IRB (configurable standard output)
4479 [se@di it le ] I was talking to Florian Gross on IRC about his Breakpoint library
+ 4480 [drbrain@se m] PrettyPrint is probably your best, least intrusive solution.
+ 4482 [sroberts@un ] How about irb uses #to_irb if it exists, then #inspect if it doesn't.
  4483 [drbrain@se m] Object#prettyprint
  4488 [sroberts@un ] $ irb -r pp
  4489 [drbrain@se m] Oh that's right its pretty_print and it takes an arg, making it less
  4499 [sroberts@un ] They want a way to override how irb displays stuff.

Function request: rb_raisef()
4484 [Daniel.Berge] /* 'x' and 'dev' already have values somewhere */
+ 4485 [matju@sy pa ] extern "C"{
+ 4486 [dooby@d1 .k ] IIUC, rb_raise already handles this.
  4487 [Daniel.Berge] Well, dangit, I thought I tried that a while back and it didn't work.
  4492 [matju@sy pa ] have formats, but i emulated that feature anyway; it was to insert one

implementing #to_pp, and pp.rb doc diffs (Re: Requesting addition to IRB (configurable standard output))
4490 [sroberts@un ] Did you try?
+ 4493 [drbrain@se m] It only works for builtins.  If you want fancy printing for your
| 4495 [sroberts@un ] Your (series) of suggestions was to the effect that I didn't have to
| 4498 [drbrain@se m] Let me step back here,
+ 4520 [akr@m1 n. rg] Thank you for the rdocification.  It is merged.
  4528 [sroberts@un ] Great, thanks.

Re: Unknown OS X 10.2 Socket constants
4491 [sroberts@un ] Ping.
4504 [matz@ru y- a] Thank you for the ping.  Somehow I missed the original mail.  I will

Ripper patch
4494 [Peter.Vanbro] ...
4496 [aamine@lo er] Checked in.  Thank you very much.

Consistency patch for exception messages
4497 [ian@ca ib n.] Here is a patch against current CVS to render Ruby's exception messages
+ 4500 [ian@ca ib n.] ...
| 4501 [ian@ca ib n.] Sigh. I accidentally got some other crap caught up in that patch. Let's
+ 4502 [dooby@d1 .k ] Done.
+ 4505 [matz@ru y- a] Thank you.  I will merge.

ruby, solaris, and high cpu load relating to SIGVTALRM
4503 [perlguy@gm i] Apologies if this shows up twice - I wasn't sure from the autoreply if
4506 [hgs@dm .a .u] I think I've seen this, ruby processing appearing high up with top.

[PATCH] callable argument for PTY.getpty (aka PTY.spawn)
4507 [csaba-ml@be ] The patch attached enables you to pass a callable object to PTY.spawn,
+ 4508 [matz@ru y- a] Hmm, I myself do not use tty often, so I'd like to hear from others
+ 4509 [csaba-ml@be ] Umm, a small correction... it should be checked whether is there any

[ ruby-Bugs-1567 ] Test Failure for test_gc(Rinda::TupleSpaceTest)
4510 [noreply@ru y] Bugs item #1567, was opened at 02-03-2005 09:59

Minor doc fix
4511 [emiel@il fo ] This prevents rdoc from complaining.
4517 [matz@ru y- a] Thank you for the report.

4512 [lists@be tr ] I'm quite sure some diligent programmer proposed this before
4515 [Daniel.Berge] If it were *just* for that, yes.  I have this sneaking suspicion there
4516 [brent@mb ri ] Well, I'm an old assembly language geek, so...

Patch for denial of service vulnerability in WEBrick
4518 [dom@bl ck at] We built a service using WEBrick with SSL enabled, and in the process
4519 [gotoyuzo@no ] Thanks.
4521 [dom@bl ck at] The patch looks good. It catches both error cases we noticed. Thanks for

Undefined Errno::EPROTO and the like raises NameError
4522 [flori@ni e. ] I just run into the problem, that certain System Call Errors aren't
4524 [matz@ru y- a] Yeah, it's on my to-do list for a long time, but without any idea to
+ 4525 [matz@ru y- a] I got an idea right after sending above mail.  Than you Florian for
| + 4526 [djberg96@ya ] How about Errno::UNKNOWNERROR?  Or Errno::UNDEFERROR?
| | 4529 [matz@ru y- a] would never be raised from any system call.  An exception class which
| + 4530 [flori@ni e. ] I like it -- it solves both problems very well.
+ 4527 [sroberts@un ] I used to think like Florian, but now I think it is better the way it is

[ ruby-Bugs-1590 ] Ruby's YAML parser doesn't ignore trailing whitespace
4523 [noreply@ru y] Bugs item #1590, was opened at 2005-03-07 15:01

[DOC-PATCH] Regex.new
4531 [g_ogata@op u] diff -Naur ruby.orig/re.c ruby/re.c
4532 [matz@ru y- a] Thank you.

[PATCH] giving acces readline to rl_line_buffer
4533 [csaba-ml@be ] The patch attached gives acces ruby's readline to rl_line_buffer.

Win32 Non-ASCII Filename Access
4534 [halostatue@g] Matz,
4535 [vincent.isam] MacOS X uses filenames encoded in UTF-8, AFAIK.
4536 [halostatue@g] According to what I've read, that is dependent upon the filesystem
4537 [Daniel.Berge] #ifdef UNICODE
4538 [halostatue@g] I know.
4539 [Daniel.Berge] That isn't my understanding, though perhaps I'm not "getting it".  From
4540 [halostatue@g] Okay -- let's try again. Ruby isn't written in Microsoft's dialect
4541 [Daniel.Berge] Why?  Who's to say MS got it wrong?  Types like TCHAR can easily be
4543 [halostatue@g] MS got it wrong. wchar_t isn't portable. UTF-8 is.
4547 [Daniel.Berge] Ok, I think I see your point now.  I also noticed that the -U option
4549 [halostatue@g] ;) Still,
4550 [halostatue@g] One other thing: I would also do the work to make UNC paths work in

[patch] logger.rb - allowing logger to be innitialized in a single line
4542 [sroberts@un ] SomeClass.logger = Logger.new(name, shift_age = 'weekly') { |l| l.level = Logger::ERROR }
4544 [Daniel.Berge] Hm...how about this modification?
4545 [sroberts@un ] Sure, how about some docs? :-)
4546 [Daniel.Berge] Ok.  Also, I guess there is an attr_accessor for 'level' already, so we

[ ruby-Bugs-1613 ] Ruby 1.8.2 array of hash entries functions incorrectly
4548 [noreply@ru y] Bugs item #1613, was opened at 2005-03-09 19:49
4551 [matz@ru y- a] I'm sorry but I don't get the above description.  Can somebody
4553 [dblack@wo bl] I don't quite understand it either, but I think it's possible that
4554 [tom@in oe he] to ruby-core.  So occasionally someone needs to sort of do some