9947-10211 subjects 10325-15313

Another .document patch.
10135 [hgs@dm .a .u] and I wanted to see what else I could do, how they worked etc.

How to submit patch for RCR Integer#pred ?
10138 [robert.dober] ...
10142 [matz@ru y- a] It's already checked in for YARV.  In general, process of submitting
+ 10147 [rubyzbibd@ub] The current info on the Ruby website slightly differ from your comments
+ 10151 [robert.dober] ...

[ ruby-Bugs-8344 ] GC memory explosion
10140 [noreply@ru y] Bugs item #8344, was opened at 2007-01-31 18:11

Re: Including classes
10157 [pit@ca it in] Ola, why do you need it this way? If you have to exchange modules and
+ 10159 [pit@ca it in] OK. Could you tell us more about what you are trying to do?
| 10160 [james@gr yp ] I ran into a similar situation yesterday.  Here's my use case.
| 10161 [pit@ca it in] module NamespaceModule
+ 10162 [halostatue@g] Classes represent objects that contain state. Classes have
  + 10163 [why@ru y- an] Well, actually, the difference between classes and modules is very
  | 10165 [ mfp@ac .o g] This has been discussed before; what you are suggesting was championed by
  + 10168 [dan-ml@da 42] module M
    10173 [gwtmp01@ma .] I don't claim that this particular pattern makes sense but you

SVN revision corresponding to 1.8.5_p12?
10167 [charles.nutt] Simple question: what SVN revision corresponds to the 1.8.5_p12 release?
10169 [vincent.isam] Are you sure it's in SVN? If I remember well the 1.8 branch was still
10170 [vincent.isam] However the version designated with the tag v1_8_5_11 does not seem to
10171 [charles.nutt] I ended up just going with the most current code on the ruby_1_8_5

[ ruby-Bugs-8384 ] Discrepancy between GetoptLong.new and documentation
10172 [noreply@ru y] Bugs item #8384, was opened at 2007-02-02 10:06

math.c, acosh.h, domain_check, MS Windows - possible bug
10174 [Daniel.Berge] Ruby 1.8.5 p12
10178 [nobu@ru y- a] atanh(1) in missing/acosh.c tries log(2.0 / 0.0) / 2, but

Ruby 1.9: Why the change to the return values of #instance_variables?
10185 [halostatue@g] I have been preparing a release of Transaction::Simple 1.4 and want to
10187 [nobu@ru y- a] if rr.instance_eval {defined?(@__transaction_checkpoint__)}
10188 [nobu@ru y- a] You can use this code for previous versions.
10189 [halostatue@g] Thank you. I have already released Transaction::Simple 1.4 with a

Reusing Ruby's tests via svn:externals
10186 [charles.nutt] Would there be any objection to JRuby including a few SVN externals

[ ruby-Patches-8437 ] Added method ole_query_interface to WIN32OLE
10190 [noreply@ru y] Patches item #8437, was opened at 2007-02-05 11:35

10193 [david@da id ] Does anyone (Matz?) know why the (new in 1.9) String.ord method is
+ 10194 [now@bi wi se] Because it only makes sense on a string of (character) length one.
| 10195 [david@da id ] I have to disagree.  It seems to me that String really needs a method
| 10202 [gavin@re in ] ...
+ 10197 [matz@ru y- a] I just followed Python convention here.
  10198 [david@da id ] Ah ha.  I don't know Python well enough, or maybe I would have realized
  + 10199 [djberg96@gm ] How would String#ord_at be a savings over String#[].ord?
  | 10200 [david@da id ] One method lookup and invocation versus two lookups and two invocations
  | + 10207 [dblack@wo bl] It's not going to be backward compatible in any case, since [] will
  | + 10208 [now@bi wi se] There is no 1.8 behavior to maintain, no backward compatibility to
  |   10213 [david@da id ] If you're working on binary data and want to read the raw byte string
  |   10215 [now@bi wi se] True.  But binary data is evil!  ;-)
  |   10216 [david@da id ] I think that these are the same thing.  And, if you're going to define a
  |   + 10218 [sroberts@un ] I found it strange at first that ruby didn't have a character class, and
  |   | 10226 [now@bi wi se] May I beg for an /important/ algorithm?
  |   | 10232 [david@da id ] Geez!  What does it take to satisfy you guys!  :-)
  |   | + 10234 [now@bi wi se] A lot, which is why most of us are using Ruby in the first place!
  |   | + 10235 [michael.seli] I have to agree with David here. All he is pointing out is that an
  |   | + 10237 [matchbo@gm i] FWIW, I think ord_at or a similarly named method sounds like a good idea.
  |   |   + 10238 [wyldeone@gm ] I'm not a contributor, but I am a user of ruby, and there is a
  |   |   | + 10241 [now@bi wi se] Well, #to_i only works for certain Strings.  I don't see how it goes
  |   |   | + 10242 [borg@uu .n t] I must disagree with you. This would make things even slower.
  |   |   |   10243 [now@bi wi se] I think it's time we see some actual use cases for String#ord.
  |   |   |   10248 [steven@lu os] In fact, that's as bad as--and equivalent to--String#unpack.
  |   |   |   + 10250 [borg@uu .n t] I dont belive you ever used String class (C++) for Bitmap
  |   |   |   + 10251 [now@bi wi se] Can we then have some example code that shows how String#ord_at would
  |   |   |   + 10275 [meta@po ox c] It shouldn't be something that a lot of time is spent optimizing, for sure.
  |   |   |     10277 [halostatue@g] Disagree. Different problem classes need different solutions. Ruby has
  |   |   + 10240 [gavin@re in ] I'll throw in my "me too" - I also think this functionality would be
  |   + 10225 [now@bi wi se] Well, depends on what angle you're looking at it from.
  |     + 10228 [now@bi wi se] What I meant was that if we're entering a realm where String#[] would
  |     + 10233 [david@da id ] I think that Nikolai and I are using "encoding" to mean "encoding +
  + 10201 [matz@ru y- a] Both having ordAt(index) or making String#ord to return codepoint of
    10209 [now@bi wi se] Like the fact that #ordAt isn't a very Rubyish name.  I really
    + 10212 [dblack@wo bl] There are quite a few multi-word methods (respond_to?, values_at,
    | 10214 [now@bi wi se] What I meant was that it seems that people have been clever enough to
    | 10217 [dblack@wo bl] irb(main):006:0> Object.new.methods.size
    | 10224 [now@bi wi se] Well, here's a another way of calculating the percentage (and gives a
    | 10229 [dblack@wo bl] I agree.  I don't think it's because people aren't willing to commit
    | 10231 [now@bi wi se] Precisely.
    + 10219 [matz@ru y- a] Of course, I have to choose more Rubyish name than ordAt for the

1.8.5 release on 3/3
10203 [shyouhei@ru ] I think it's nice we also release a 1.8.5 on this timing.  Please let me

Can we add a link to ViewVC in the ruby-cvs commit mails?
10220 [rubyzbibd@ub] Is it possible to bring back links to ViewVC in the Ruby-CVS commit
10221 [ko1@at ot ne] You can receive "diff" mail on ruby-changes ML.
10222 [rubyzbibd@ub] Thanks for the "diff" mail address this will act as a stop-gap

Ruby/YARV thread question
10223 [mneumann@nt ] In thread.c, three YARV thread models are described. Which models will be
10252 [ko1@at ot ne] Yes.  You must release Giant VM Lock before doing blocking task.  If you

[PATCH] Test::Unit::AutoRunner#parse_args bug, attributable to optparse documentation.
10230 [ mfp@ac .o g] holding tests to be executed). I found this when trying to use test/spec.
10239 [nobu@ru y- a] Sorry, it seems my fault.
10244 [ mfp@ac .o g] I guess this bug appeared behind you ;)... but it's just an easily

Build Problem
10236 [andy-ruby-co] Hoi,

[ ruby-Patches-8513 ] Define text color in rdoc html
10245 [noreply@ru y] Patches item #8513, was opened at 2007-02-08 10:35

Why are Symbol#object_id and Symbol#to_i different?
10246 [jflam@mi ro ] On my system, it appears that the result of to_i is simply object_id / 10 f=
10247 [matz@ru y- a] Under my implementation, Symbol#to_i value has some encoded
10249 [jflam@mi ro ] Great, thanks!

[ ruby-Bugs-8537 ] Module:constants returns unexpected result
10253 [noreply@ru y] Bugs item #8537, was opened at 2007-02-09 17:23
10264 [sroberts@un ] Accessible, yes.

[ ruby-Bugs-8538 ] uninitialized variable in function rb_syswait()
10254 [noreply@ru y] Bugs item #8538, was opened at 2007-02-09 17:25
10265 [sroberts@un ] In C, static variables are default initialized to 0.
10272 [borg@uu .n t] Since when? Usualy passing debug flag do compiler does that..
10274 [sylvain.joye] It is not a compiler feature, but a norm. Period. Search google for 'C

[ ruby-Bugs-8539 ] String:upto loops forever if argument is modified inside block
10255 [noreply@ru y] Bugs item #8539, was opened at 2007-02-09 17:27
10260 [neleai@se na] And what did you expect?
10261 [Daniel.Berge] One suggestion that came up was that receivers could be temporarily

[ ruby-Bugs-8540 ] Unexpected behaviour if IO:write will be overwritten
10256 [noreply@ru y] Bugs item #8540, was opened at 2007-02-09 17:29
10263 [shyouhei@ru ] OK tell us exactly what you expected in this.

[ ruby-Bugs-8541 ] coredump when invoking Kernel:syscall
10257 [noreply@ru y] Bugs item #8541, was opened at 2007-02-09 17:31
10267 [sroberts@un ] If you provide more arguments than rb_f_syscall can use on your system,
+ 10278 [shiba@ma l2 ] I see.  It must be a Ruby side's (and typical) bug.
| 10282 [shyouhei@ru ] It should be included.  I'll merge the patch into ruby_1_8_5 later (the
+ 10285 [knu@iD em ns] Thanks, Sam and Tadashi.  I've just committed a fix to trunk.

[ ruby-Bugs-8542 ] Unexpected exception when invoking Kernel:syscall
10258 [noreply@ru y] Bugs item #8542, was opened at 2007-02-09 17:35

Segmentation fault: Ruby 1.8.5 Under VC++ express 2005
10259 [zhimin.wen@g] ...
10266 [usa@ga ba ec] Hell,
10268 [masaki.suket] No. I checked that `nmake test' works fine.
10270 [zhimin.wen@g] ...
10317 [usa@ga ba ec] Thank you, Suketa-san and wen-san.
+ 10318 [usa@ga ba ec] ...
+ 10319 [masaki.suket] Here is the result.
  10320 [usa@ga ba ec] Thank you, this is expected result.

Limiting ?iso-8859-1?Q?=B4?= or popen subprocess time
10262 [hadmut@da is] I need to read the output from a subprocess (binary, not ruby) from within a
10271 [hadmut@da is] The popen command does not use the unix popen library call under Linux, instead it does all of the job itself and stores a pid in the handle popen returns.

[ ruby-Bugs-8548 ] YAML::Omap loses entries, replacing them with random references
10269 [noreply@ru y] Bugs item #8548, was opened at 2007-02-09 20:23

Re: Limiting  or popen subprocess time
10273 [neoneye@gm i] How about 'select' ?

[ANN] fastthread now default in ruby_1_8
10276 [knu@iD em ns] I have just finished integrating MenTaLguY's fastthread mutex
10308 [mental@ry ia] The version of fastthread which was merged doesn't handle interrupted
10311 [knu@iD em ns] Thanks, just committed the patch.
10315 [mental@ry ia] Thank you very much,

Singleton Inheritance?
10279 [ceo@ha th rn] ...
10281 [sroberts@un ] The .jpg files you attached aren't JPEG, they are PNG. Anybody having

Mistake in Singleton Image
10280 [ceo@ha th rn] ...

[Solved] Re: Can we add a link to ViewVC in the ruby-cvs commit mails?
10283 [rubyzbibd@ub] I noticed that the ruby-cvs mail now has links to ViewVC.

Can't seem to run tests?
10284 [farrel.lifso] I'm not sure if this is the correct list to ask this (let me know and
+ 10287 [neleai@se na] make test. But takes forever to complete. And you probably get ton of
+ 10292 [steven@lu os] Did you disable extensions by chance? The targets I (think I) know
  10293 [farrel.lifso] I actually don't know if I disabled extensions I just downloaded the
  10328 [steven@lu os] Does "./ruby -C test runner.rb ruby/test_array.rb" from the top-level

[ ruby-Bugs-8592 ] gsub to replace small amount of content with a large amount causes original trailing text to repeat
10286 [noreply@ru y] Bugs item #8592, was opened at 2007-02-12 22:26

[ ruby-Bugs-8597 ] Socket library should support abstract unix sockets
10288 [noreply@ru y] Bugs item #8597, was opened at 2007-02-13 16:10
+ 10294 [sroberts@un ] Yes, problem is pack_sockaddr_un() doesn't allow a leading NUL, though
| 10296 [matz@ru y- a] Is this OK?  I mean I though sun_path requires NUL termination, so
| 10303 [sroberts@un ] Since the sockaddr length is passed/returned explicitly in system calls,
| 10310 [sroberts@un ] For Socket, this allows abstract sockets by allowing first character to
| 10314 [matz@ru y- a] I need more info to merge this in.  What should NUL in sun_path should
| 10327 [sroberts@un ] - If you semi-manually pack the sockaddr, you can use abstract sockets
+ 10360 [bram@lu n. e] % svn diff
  + 10361 [nobu@ru y- a] Equivalent to offsetof()?
  | 10362 [bram@lu n. e] Yes, that was my intention... I didn't know about the existence of the
  + 10370 [sroberts@un ] Why allow abstract sockets with UnixServer and UnixSocket, but not with
    10407 [sroberts@un ] *ping*

[OT] Communicating Ruby programs and processor choice
10289 [hgs@dm .a .u] ...

[PATCH] URI::Generic#userinfo
10290 [zimbatm@or e] ...
+ 10295 [meta@po ox c] Those are not errors. Username and password are not allowed in HTTP
| 10297 [zimbatm@or e] Right, but since it seems handled by ruby's URI, why not make it
| 10322 [meta@po ox c] I feel like there's some strange new usage of the term "right" that
| 10333 [zimbatm@or e] ...
+ 10298 [rubyzbibd@ub] In case you are wondering, it appears that Matz has commited your
  + 10299 [zimbatm@or e] ...
  + 10312 [knu@iD em ns] Done.  1.8's lib/uri has been synchronized with trunk.
    10316 [knu@iD em ns] I slightly re-fixed it so that an empty user name or password is

debug.rb does not work in some cases?
10291 [Martin.Kraus] $ cat test.rb

rb_class_real Question?
10300 [ceo@ha th rn] OK - Stupid Question: What (in english prose) is this line actually doing???
10309 [neleai@se na] Best look at source

10301 [ceo@ha th rn] OK - I look at again --- I seem to remember it was confusing the
10304 [matz@ru y- a] 'real superclass', yes.

RHG Translation project
10302 [ceo@ha th rn] Sorry --- But this code is driving me a little crazy

character codes in regular expressions?
10305 [hadmut@da is] is there any way to include characters by code in a regular expression?
10306 [murphy@ru yc] you can use the default string escape sequences, like /[\x80-\xff]/.

[ ruby-Bugs-8663 ] fastthread: rb_bug() due to stale wait queue entries
10307 [noreply@ru y] Bugs item #8663, was opened at 2007-02-14 17:25

[ANN] Code frozen for Ruby 1.8.6
10313 [knu@iD em ns] Soft code freeze for Ruby 1.8.6 has just begun.  I'll make a

[ ruby-Bugs-8676 ] File.basename fails on Windows root paths
10321 [noreply@ru y] Bugs item #8676, was opened at 2007-02-15 10:09
11160 [nobu@ru y- a] Drive letter is part of base name?  It feels very strange to me.
11161 [halostatue@g] It's right, though.
+ 11162 [nobu@ru y- a] What's the reason?  Do you consider a drive letter to be a part
| + 11164 [hramrach@ce ] When you include some sort of networking into file path you usually
| + 11182 [djberg96@gm ] File.split would have to be redefined as well, based on the new
+ 11163 [now@bi wi se] Fs.GetBaseName("C:/") = ""
  11181 [djberg96@gm ] Try it with backslashes. I don't think we want an empty string. :)
  11183 [nobu@ru y- a] They are not equivalent, and the example PathStripPath() in
  11184 [djberg96@gm ] Without a suffix it looks equivalent to me. Why do you think they aren't
  11186 [gus@pr gr ss] 1) Ruby works the same as Windows scripts on Windows and the same as UNIX

Trouble with xmlrpc
10323 [james@gr yp ] Some of the Ruby code used by TextMate makes use of xmlrpc/
+ 10324 [Daniel.Berge] elsif expected.to_i != data.size and resp["Transfer-Encoding"].nil?
| 10326 [james@gr yp ] That makes a lot more sense to me, since the code goes through the
| 10337 [james@gr yp ] ...
| + 10339 [zdennis@mk e] According to HTTP1.1 RFC
| | 10340 [james@gr yp ] This quote is actually referring to clients sending messages to the
| + 10375 [james@gr yp ] Was there anything wrong with this patch?  If not, can we please
|   10376 [knu@iD em ns] What does the maintainer say about it?  Is he irresponsive these days?
|   10378 [james@gr yp ] Do I need to try emailing him off this list?  Can anyone tell me who
|   10379 [knu@iD em ns] Note the To: field in my mails. :)
|   + 10380 [james@gr yp ] Thank you for all your help.
|   + 10391 [mneumann@nt ] I think it's now more than 48 hours later ;-)
|     + 10392 [james@gr yp ] Alex Young talked me out of that one.  The spec pretty much requires
|     + 10400 [knu@iD em ns] Thanks for taking the time.  Both have been committed.
|       10403 [james@gr yp ] Thanks to everyone for all the help in fixing these.
+ 10342 [james@gr yp ] While I am complaining about xmlrpc, we have another issue.  It's
  + 10343 [alex@bl ck e] Why is it a problem that the time isn't converted to UTC?
  | 10344 [james@gr yp ] Fair warning:  this issue is certainly debatable.
  | + 10346 [sroberts@un ] Why do you say that? Isn't the format of the dateTime.iso8601 field that
  | | 10348 [james@gr yp ] I'm sorry.  I wasn't very clear on that point.  The XML-RPC spec
  | | 10355 [sroberts@un ] Tag                Type      Example
  | + 10347 [alex@bl ck e] Really?  I'd have expected the reverse - given the freedom in the spec,
  |   10349 [james@gr yp ] That doesn't make any sense to me (though the spec *does* seem to say
  |   10352 [james@gr yp ] Ah, I think I just got what you are saying Alex.
  |   10356 [alex@bl ck e] No worries :-)
  |   10377 [james@gr yp ] ...
  |   10382 [sroberts@un ] Which DateTime are you trying to support?
  |   10383 [james@gr yp ] ...
  |   10401 [knu@iD em ns] OK, revised patch committed.
  + 10350 [drbrain@se m] Please stop giving links to things you can paste in-line.  External
    10351 [james@gr yp ] Sorry.  I've been using the links from our TextMate maintainer's