106824-112265 subjects 107230-108485

VIM Extensions on OS X
107039 [nick@ac iv h] I'm getting the following errors with OS X 10.3 Panther preinstalled Vim
107041 [nick@ac iv h] More info...

sysread and buffered I/O
107051 [hal9000@hy e] I've been playing with telnet (and ssh) and I've been
107052 [akr@m1 n. rg] I proposed stdio friendly sysread like method: readpartial.
107053 [hal9000@hy e] Very interesting, I did not notice that. Thank you.
107054 [akr@m1 n. rg] An underscore is inconsistent with other IO read??? methods: readchar,
107055 [hal9000@hy e] readchars    # with an s
107057 [akr@m1 n. rg] Good name is necessary to incorporate a method to ruby.
+ 107058 [hal9000@hy e] Certainly, I agree fully.
| 107059 [akr@m1 n. rg] See [ruby-dev:23247] and [ruby-dev:23248].
+ 107062 [gsinclair@so] TANAKA Akira suggested a new method IO#readpartial, which is
| 107065 [akr@m1 n. rg] [ruby-dev:23247] contains a patch including RDoc comment in English.
| + 107083 [JohnFeezell@] Here are some other possible names that seems to fit with readchar,
| | 107158 [akr@m1 n. rg] "at most" doesn't represent the difference between IO#readpartial and
| | 107189 [JohnFeezell@] Thanks.  That makes the difference clearer.
| | 107254 [akr@m1 n. rg] Matz, is there a name good enough?
| + 107084 [ahoward@no a] why not
| | 107160 [akr@m1 n. rg] It doesn't represent the difference from IO#read.
| | 107196 [ahoward@no a] consider
| | 107255 [akr@m1 n. rg] I see.
| | 107257 [ahoward@no a] ...
| | 107277 [akr@m1 n. rg] Yes.  It is not surprised that BasicSocket#recv and IO#sysread is
| | 107290 [ahoward@no a] then perhaps using a short name, but it should be an extension to force
| | 107337 [akr@m1 n. rg] I don't like such small extension library.
| + 107094 [gsinclair@so] What about this?
| | + 107095 [mneumann@nt ] Regards,
| | + 107159 [akr@m1 n. rg] I don't see a difference between 4th and 5th.
| + 107287 [ysantoso-rub] How about:  STDIN.readposixly(4096) or STDIN.posix_read(4096)
|   107336 [akr@m1 n. rg] It is not wonder.  readpartial is a stdio friendly sysread.  sysread
+ 107260 [dooby@d1 .k ] I thought Hal's read_avail (perhaps readavail) was a fair description, but ...
  107276 [akr@m1 n. rg] readpartial blocks in such case.
  107299 [retnuh@ei co] read_next_avail?

ruby postgresql question
107060 [hash_bang@re] Does anyone know of a good way to basically seek through a large query
107078 [ahoward@no a] are you asking about cursors?
107082 [hash_bang@re] That is entirely possible (;
107088 [mneumann@nt ] Can you explain again, what exactly is your problem.
107091 [hash_bang@re] Well, I've written re-written an old BBS program.  Right now it holds it's
107092 [mneumann@nt ] can you describe the table layout in sql?
107101 [hash_bang@re] Sure.  Here is the sql statement for a message board table
+ 107103 [mneumann@nt ] aha, now I understand :-)
+ 107108 [angus@qu va ] [Mark Firestone <hash_bang@retrobbs.org>, 2004-07-21 17.45 CEST]

Little query
107068 [peter@se an ] Given that x is an Array how come I can do
+ 107069 [mneumann@nt ] class Array
| 107070 [peter@se an ] Thanks that was very useful.
| + 107073 [martindemell] +1
| + 107085 [matz@ru y- a] I'm not enthusiastic for them.  I feel like that they are not as
|   + 107086 [peter@se an ] For me it seems natural that if I can access the value from x.first I
|   + 107089 [mneumann@nt ] Me too. Then, there should be also Range#last= and Range#first=, but I
|     107090 [bob.news@gm ] "Michael Neumann" <mneumann@ntecs.de> schrieb im Newsbeitrag
+ 107072 [flgr@cc n. e] Because then x.last(5) = [1, 2, 3, 4, 5] would also have to work, but
  107188 [kristof@vl e] I'd like that.  I would use this to acces class variables as

Re: DBUnit for Ruby? (Or how databases play in Rails)
107080 [david@lo dt ] Indeed. I'd dare say that AR's Fixtures (ARF) goes beyond the DBUnit on

mod_ruby or fastcgi+ruby?
107093 [nospam@no pa] What are the pros & cons of mod_ruby vs fastcgi+ruby?  Ruby will most
+ 107096 [khaines@en g] + Nice access to Apache internals.
| + 107129 [nospam@no pa] Thanks for the great info!  It was very helpful.
| | + 107133 [khaines@en g] Each seperate Apache process has its own interpreter.  However, each of the
| | + 107207 [mortonda@dg ] That's true, as each apache instance gets its own interpreter.  To keep
| |   107208 [khaines@en g] There are three issues.  First, namespaces don't protect you if another
| + 107165 [lists@za a. ] Depends on the definition of "complex". I myself see fastcgi as far less
|   107194 [khaines@en g] I think the use of complex here refers to the overall setup and config.
+ 107105 [carl.youngbl] Neither is very hard to set up in a run-of-the-mill unix environment.

XML line numbers
107100 [martindemell] I'm using the REXML stream parser to read in an xml file and check the
107190 [ser@ge ma e-] Martin,
107212 [martindemell] Thanks - I'm going out of town for a few days, but I'll take a look at

tempfile iterator?
107107 [snowzone5@ho] what's the easiest way to iterate through a Tempfile?
107113 [ahoward@no a] t = Tempfile.new $$
107115 [snowzone5@ho] tempfile iterator?
107116 [msparshatt@y] if @lg is a variable holding the filename then that should be
107135 [snowzone5@ho] yup.  i figured that out when i applied some common sense...
107167 [bob.news@gm ] "tony summerfelt" <snowzone5@hotmail.com> schrieb im Newsbeitrag

substring by range parameter (bug?)
107114 [email55555@h] s="0123456789"
107117 [ahoward@no a] for ranges, start must be <= end and -1 is not <= -2
+ 107119 [email55555@h] Thanks. I know that.
| 107126 [ahoward@no a] sorry - did not read your post closely enough.
+ 107120 [ssk@ch l. os] I understand what you mean.
  107123 [email55555@h] OK. You got my point.
  + 107124 [email55555@h] OK. You got my point.
  | 107125 [hal9000@hy e] This is the most logical analysis of this issue that I remember
  + 107127 [ssk@ch l. os] I think getting a substring using range is somewhat confusing.
    107128 [email55555@h] Correct to my previous conclusion.

[ANN] Ruby Installer for Windows 1.8.2-14_RC5 (from Ruby 1.8.2 preview1)
107118 [curt@hi bs c] Today Matz released the official preview1 for Ruby 1.8.2. This release
+ 107132 [lymans@gm il] Curt,
| 107136 [curt@hi bs c] No, Oniguruma is not in the Ruby Installer for Windows. The installer is
+ 107162 [davinciSINSP] Do it includes ruby-postgres precompiled?
| 107184 [curt@hi bs c] No, ruby-postgres is not included, so you would have to install it yourself.
| 107185 [mailinglists] But this does still not help on binary extensions. And until the
| + 107186 [lyle.johnson] Sure it would, if the package authors provide both source and
| | 107209 [rich@in oe h] And, we are adding to Gems the ability to utilize the platform attribute
| | + 107211 [lymans@gm il] Perfect!!
| | | 107221 [rich@in oe h] Yeah...we will work that out.  It will likely be listed as two gems (one
| | + 107222 [lyle.johnson] As long as we're on the subject, are there any plans to include (and
| | + 107284 [surrender_it] just one thing: I'm not sure if this is just a sample or is closely
| |   107285 [lyle.johnson] Excellent point. I think it's true that binaries compiled with
| |   107303 [chadfowler@g] This is why this feature isn't done yet. :)  It's going to take some
| + 107195 [curt@hi bs c] RubyGems *does* support platform-specific binary gems. This was an issue
|   + 107197 [lymans@gm il] Curt,
|   | + 107200 [curt@hi bs c] Yes, please do send it along to [curt at hibbs dot com].
|   | + 107201 [mailinglists] I have done some testing for Arachno Ruby. But i will wait until
|   + 107231 [quixoticsyco] RubyGems supports binary gems in the sense that you can put any file
|     107265 [cc1@ce .w st] I really think it would make more sense to have all that info stored as
+ 107191 [laurent.sans] Thanks for the package.  I guess it's very convenient for Microsoft users.
| 107204 [surrender_it] on the lines of the hread, I think the best thing would be to package
| + 107205 [surrender_it] I mean, when the one click gets installed, pop up a gui with "select
| + 107206 [curt@hi bs c] That's the approach I want to take with almost all extensions -- provide
| + 107266 [laurent.sans] Mmh good idea.  I will have a look at it after the next release.  I
+ 107271 [unet01@ra io] Thanx for the intaller. Is there a special reason, why the windows
| 107465 [curt@hi bs c] No particular reason other than we haven't taken the time to update to a
+ 107272 [zakaria@su r] Does it fix ri bug related to PAGER in WinMe [ruby-talk:103662]?
| 107288 [curt@hi bs c] I have fixed the general problem with how I was building the RI database in
| 107289 [dave@pr gp o] I must have missed this one. But unfortunately I don't have an ME box
| 107484 [zakaria@su r] Here's what I do to track down the problem.
+ 107328 [jfontan@gm i] Will the final version have the opengl bindings? I was so surprised
| 107331 [curt@hi bs c] Hmmm.... I think it should have been there. I'll have to look into what
| 107340 [jos@ca no k.] I commented out the OpenGL bindings because they were causing the builder.rb
| 107342 [curt@hi bs c] No, that's not the problem. Lyle Johnson sent me a new binary build which I
| + 107343 [curt@hi bs c] Yeah, I messed this one up... I just figure out what I did. Opengl support
| + 107345 [jos@ca no k.] Okay, glad to hear that's working again, Curt. It wasn't at the time I checked
+ 107362 [usenet2@no p] Am I right that this package does not contain OpenSSL?
  + 107365 [surrender_it] with the older preview it seemed to work
  | 107385 [dooby@d1 .k ] I'm sure Gabriele meant to attribute NAKAMURA Usaku (usa).
  | 107386 [surrender_it] sorry sorry sorry and thanks for the precisation :)
  + 107464 [curt@hi bs c] Yes, you are correct.

aop in ruby
107121 [csshsh@st uc] im currently working on a aop library for ruby. most of the things like
+ 107134 [jan.sabbe@st] Well, I have no idea what aop is, but I showed the following code to a
+ 107172 [surrender_it] mh.. two things.
  107176 [csshsh@st uc] yes. i checked out aspectr. its quite great. there are a couple of

Referencing objects and program design
107122 [Stephan.Kaem] suppose I have some kind of object which holds 'object ids' referring to
+ 107150 [assaph@av ya] to other objects,
+ 107168 [bob.news@gm ] "Stephan K?mper" <Stephan.Kaemper@Schleswig-Holstein.de> schrieb im
  107213 [Stephan.Kaem] Thanks for that answer (this goes to Assaph too, of course).
  + 107250 [DocBoobenste] Traversing object networks... that seems like an application for
  | 107279 [bob.news@gm ] "Claus Spitzer" <DocBoobenstein@gmail.com> schrieb im Newsbeitrag
  | 107295 [DocBoobenste] I see your point. While ObjectSpace.each_object can be limited to
  | 107298 [bob.news@gm ] "Claus Spitzer" <DocBoobenstein@gmail.com> schrieb im Newsbeitrag
  + 107280 [bob.news@gm ] "Stephan K?mper" <Stephan.Kaemper@Schleswig-Holstein.de> schrieb im

[PATCH] was - Re: substring by range parameter (bug?)
107130 [ahoward@no a] if you look at the source, it is clear that rb_range_beg_len intends to fail
+ 107131 [ahoward@no a] @@ -22,6 +22,7 @@
+ 107146 [matz@ru y- a] I think I understand you, but not sure.  Can you show me an example of
  107149 [ahoward@no a] ~ > irb
  + 107164 [discord@ma .] This seems right to me... aren't these two expressions equivalent?
  | + 107193 [email55555@h] When I found that on the first time, I thought that it is a bug.
  | + 107198 [ahoward@no a] yes - both are broken by exactly the same bug!  ;-)
  + 107270 [matz@ru y- a] Now I fully understand.  I will commit your fix.  Can you prepare
    + 107273 [akr@m1 n. rg] Do you notice that the patch causes many errors in the test?
    | 107286 [ahoward@no a] i have too.  but what to do - it should be fixed?
    | 107338 [akr@m1 n. rg] I think the patch should be rejected.
    + 107291 [ahoward@no a] tanaka has pointed out that this patch causes test failures.  however, it does

nesting #{} problem
107139 [crf@sb x. u-] irb(main):040:0> width = nil
+ 107140 [assaph@av ya] "door and width=99"
| 107144 [matz@ru y- a] I'd call it IRB "limitation".  IRB do not parse nested interpolation
| 107147 [assaph@av ya] well.
| 107148 [hal9000@hy e] I don't think so. Ruby is based on yacc, and I think irb is
| 107155 [mailinglists] Right. The problem is that yacc can't generate a state based
+ 107141 [halostatue@g] The problem is partially in irb. Your code works if you put it in a Ruby file
  107199 [crf@sb x. u-] works - thanks!

Problem with ActiveRecord and Cerise
107151 [k@v2 tu io c] I'm developing a website using the cerise app server and the
107178 [david@lo dt ] Active Record is not thread safe. So if Cerise uses threads to handle
+ 107202 [mneumann@nt ] Should be easy to fix by using thread-local variables instead of a class
| 107218 [k@v2 tu io c] Where retrieve_connection calls establish_connection with the approriate
+ 107219 [k@v2 tu io c] The appplication persists through requests, so how about keeping the

Incident Analysis of the intrusion on helium.ruby-lang.org
107154 [shugo@ru y- ] As already reported, helium.ruby-lang.org, which is one of the servers

Bug? String interpolation and continuations
107156 [tunah.usenet] (Typed into IRB. The results are similar in ruby but because of the
+ 107177 [angus@qu va ] [Sam McCall <tunah.usenet@tunah.net>, 2004-07-22 08.52 CEST]
| 107234 [gltewalt@ya ] puts "[#{callcc &store_cont}" + "]"
| 107274 [decoux@mo lo] Well, this is probably this that he want
| 107341 [tunah.usenet] Yes :)
+ 107344 [matz@ru y- a] It's not a bug.  It's a feature.
  107360 [tunah.usenet] Just to be clear, I did expect the infinite loop :-) But I assumed that

[OT] FreeBSD <-> Debian
107161 [r_mueller@im] what do you think about FreeBSD in general? I am about using FreeBSD too, but
107163 [muellerix@gm] To make it short: I love it :-)
+ 107166 [r_mueller@im] sounds great,
| + 107170 [mneumann@nt ] Yes, you can run Linux binaries on FreeBSD. The linux syscalls are
| | 107180 [nospam@no pa] Good point.  This isn't emulation and precisely why Linux binaries can
| + 107171 [hramrach_l@c] The ports system benefits from the focus on source packages. Afret
| + 107174 [surrender_it] that's cross compiling IMO is a different thing
| + 107179 [muellerix@gm] The other already mentioned, that some linux-progs might mork...
+ 107187 [drossruby@ya] Ok, this seems pretty outdated to me. FreeBSD 5.2.1 is
  107203 [mneumann@nt ] I don't feel that's even as near as stable as 4.10. If you want 4.10++,
  + 107215 [muellerix@gm] I think I'll stick with 4.X unless I need stuff like USB 2.0 etc... Probably
  | 107216 [drossruby@ya] Yes, DragonFly uses the FreeBSD port system plus its
  | 107220 [mneumann@nt ] Do you mean, the ACPI implementation on FreeBSD in general is not that
  | 107225 [drossruby@ya] Yes, I do. It is BIOS manufacturer dependant. There
  + 107282 [rasputnik@he] <aol>
    107294 [drossruby@ya] OK. please do not answer again, you are trolling. It
    107297 [rasputnik@he] wtf is wrong with you?
    107300 [drossruby@ya] Good god dude, grow up. Thats why I said not to

Osama Found Hanged
107192 [frankjones@h] Osama Bin Ladin was found hanged by two CNN journalists early Wedensday evening.  As evidence they took several photos, some of which i have included here.  As yet, this information has not hit the headlines due to Bush wanting confirmation of his identity but the journalists have released some early photos over the internet..

rb_load_file takes char *
107210 [kristof@vl e] Is there any reason why rb_load_file takes a
107268 [matz@ru y- a] No reason but historical one.  Thank you for the suggestion.

[ANN] Ruby-GetText-Package-0.6.1
107214 [mutoh@hi hw ] Ruby-GetText-Package-0.6.1 is now available.

File.new("file.ext", "w").write data
107223 [pan@er kv en] Consider...
+ 107226 [rcoder@gm il] Erik,
+ 107227 [alex@ve k. n] Garbage collector will undoubtedly close it, but it is a very unhealthy
+ 107237 [k@v2 tu io c] If being short and readable is your concern, you can always extend the
+ 107245 [vjoel@PA H. ] I don't think it can be collected during the #write method, because the
+ 107281 [bob.news@gm ] "Erik Veenstra" <pan@erikveen.dds.nl> schrieb im Newsbeitrag

[NUBY] Reading stderr, stdout and the exit status of a process
107224 [alex@ve k. n] and stderr as Strings, as well as the exit status.
107228 [ahoward@no a] see http://raa.ruby-lang.org/project/session/
107229 [alex@ve k. n] Thank you very much!
107233 [ahoward@no a] oh... ;-(
107238 [alex@ve k. n] As far as I remember, NT and 2000 support POSIX calls, but XP doesn't
107242 [ahoward@no a] google for 'create process' and window - i do think there is someway to do