pugscode.org/ | nopaste: sial.org/pbot/perl6 | pugs: [~] <m oo se> (or rakudo:, kp6:, elf: etc.) (or perl6: for all) | irclog: irc.pugscode.org/
Set by Tene on 29 July 2008.
davidm123 why are so many links on www.pugscode.org broken? 00:33
lambdabot Title: Pugs - pugscode
Auzon Do they point to other sites? That's one probable cause. 00:34
davidm123 no, just broken. examples: Download, Try Pugs Now, Smoke test status 00:36
well, they are on a different domain
and the link mentioned to Perl6-Pugs on cpan has a distribution dated Oct 2006. 00:37
Auzon perl6: sub foo {...}; say foo.perl 01:25
p6eval elf 21621: OUTPUT[Parse error in: /tmp/0RIeqnpdZo␤panic at line 1 column 9 (pos 9): Missing right brace␤WHERE: sub foo {...}; say foo.perl␤WHERE: /\<-- HERE␤ STD_red/prelude.rb:99:in `panic'␤ STD_red/std.rb:255:in `_block_rest'␤ STD_red/std.rb:242:in `block in block'␤
..STD_red/prelude.r...
..pugs: OUTPUT[*** Unimplemented stub called␤ at /tmp/QK2yyI9RyE line 1, column 20 - line 2, column 1␤ /tmp/QK2yyI9RyE line 1, column 10-13␤]
..rakudo 29834: OUTPUT[Missing '}' at line 1, near "...}; say "␤current instr.: 'parrot;PGE::Util;die' pc 120 (runtime/parrot/library/PGE/Util.pir:82)␤]
Auzon Is /<[]]>/ a valid regex? STD.pm doesn't think so, charset.t seems to think so, and S05 doesn't specify if ] or ]> closes the character class. 01:32
pugs_svn r21622 | ruoso++ | [smop] finishing rename of s1p_code to s1p_ccode 01:42
Auzon tryfile++ for catching my syntax errors :) 01:43
pugs_svn r21623 | Auzon++ | [gsoc_spectest] Added more angle bracket syntax tests. Almost done with it. Moved charset.t and combchar.t to spec. (added 17 tests)
pugs_svn r21624 | ruoso++ | [smop] starting the test for s1p_attribute 02:00
ruoso @tell pmurias I've finished the code => ccode renaming. about postcircumfix:<()>, we should fix it, I don't know where I took .call() from... I'm starting Attribute but we should probably refactor Method into using Code/CCode instead of how it's done now... 02:01
lambdabot Consider it noted.
s1n anyone know why 'make localtest' is hanging on a wait() without actually running the test?
ruoso sleep & 02:02
bacek pugs: say -1.i 02:22
p6eval pugs: OUTPUT[*** No such method in class Int: "&i"␤ at /tmp/IZJhWroPwf line 1, column 5 - line 2, column 1␤]
pugs_svn r21625 | s1n++ | [spec] all but one of the 'e' tests pass, there's still a parse error on the last test 03:56
pugs_svn r21626 | s1n++ | [spec] trig tests pass except for atan2. 04:01
meppl good morning 04:37
pugs_svn r21627 | Auzon++ | [evalbot] STD.pm on evalbot, first attempt. We'll see how this goes. 04:50
Auzon evalbot control restart 04:52
Auzon p6eval control restart 04:52
hm
Auzon evalbot control restart 05:13
Auzon (one more try... 05:13
Auzon yay, it works. 05:14
std: 1+1
p6eval std 21627: OUTPUT[parse ok]
Auzon std: /<[]]>/
p6eval std 21627: OUTPUT[regex assertion not terminated by angle bracket at line 1:]
pugs_svn r21628 | Auzon++ | [evalbot] made STD work in a basic sense 05:17
Auzon @tell moritz_ I added STD.pm to evalbot. It seems to work well enough. I want to make it a bit more descriptive when things fail, but that'll wait til later. 05:18
lambdabot Consider it noted.
moritz_ Auzon++ 06:29
lambdabot moritz_: You have 1 new message. '/msg lambdabot @messages' to read it.
pugs_svn r21629 | pmurias++ | [elfX] sub {...}, lack of support from smop yet 06:40
pmurias ruoso: we should be able to bootstrap the object model with Code and and a way to be able to set the .dispatcher 06:46
(and LexPad to do it cleanly)
pugs_svn r21630 | pmurias++ | [smop] a stub smop_s1p_code, test 21_code.p6 dosn't pass yet 06:48
viklund shouldn't %s=%*ENV work? 06:54
rakudo: my %s=%*ENV 06:55
p6eval rakudo 29834: OUTPUT[Parrot VM: Can't turn to a singleton type!␤current instr.: 'parrot;Perl6Hash;infix:=' pc 4544 (src/gen_builtins.pir:3083)␤]
pmurias pugs_svn: my %s=%*ENV;
pugs: my %s=%*ENV;
p6eval pugs: RESULT[\()]
pmurias ruoso: i'll finish code later (likely tommorow), but you can feel free to hack on it if you want to 06:57
cycling trip&
moritz_ Auzon: I'm pretty sure <[]]> isn't a a valid regex because you can have multiple [] groups in a <...> rule like this: <[ ] + [...]>, so in <[]]> the first ] actually closes the char class 06:57
viklund oh, well, I'll submit a bug 06:59
moritz_ viklund: plese test first locally, sometimes p6eval behaves strange 07:00
viklund moritz_: I've done that, that's how I found it 07:01
moritz_ viklund: ok, just wanted to check
viklund np
ruoso @tell pmurias you forgot to add smop_s1p_code... 09:31
lambdabot Consider it noted.
ruoso work & 09:35
pasteling "masak" at 130.243.201.126 pasted "Tried to run make in src/perl6/, got this" (7 lines, 217B) at sial.org/pbot/31718 09:49
masak what am I doing wrong?
moritz_ tries 09:50
moritz_ masak: works for me. Did you try to rm -rf lex/? 09:50
masak moritz_: nope. trying... 09:51
moritz_ or just 'make clean'
masak I didn't have a lex/
same result.
moritz_ do you have one afterwards? 09:52
masak no.
I'm running perl5.10.0 as ordained
moritz_ try to rm STD.pmc and then run make again 09:53
masak `perl STD.pmc`?
moritz_ rm, not perl ;-) 09:53
masak ah, 'rm', not 'run' 09:54
:)
moritz_ btw the makefile is slightly... weird. For the perl -c STD.pmc step you need 'perl' to be a perl 5.10.0 09:55
masak ah
that might be the thing
I'm not running 5.10.0 as the default
moritz_ I changed it to read perl5.10.0 in my local makefile 09:56
masak I would, but I haven't figured out a good way to usurp 5.8.8 that comes with Mac OS X
now YAML/XS.pm is missing. I suppose 5.10.0 has different @INC paths... 10:01
yep.
how do I start CPAN using 5.10.0?
moritz_ perl5.10.0 -MCPAN -eshell 10:02
masak dz 10:03
pugs_svn r21631 | moritz++ | [spec] removed illegal regex from S05-metasyntax/charset.t 10:06
masak shit... the test Speech/t/Speech in Mac::Carbon always scares me witless :/ 10:13
moritz_ ;) 10:14
masak I got strange compilation errors during the installation of re::engine::TRE... 10:21
something about 'undefined symbols' 10:22
what does one do about such a thing?
moritz_ "undefined symbols" *could* be a missing requirement
bbbike.radzeit.de/~slaven/cpanteste...e-TRE+0.04 doesn't look too bright :/ 10:23
lambdabot Title: CPAN Testers Matrix: re-engine-TRE 0.04, tinyurl.com/5hspfx
moritz_ masak: maybe bug avar, perhaps he can help 10:23
masak moritz_: hm, ok 10:24
so, basically, those who got this try5 script running are on Linux
moritz_ yes ;) 10:25
masak curse you and your superior OS! ;)
moritz_ though the failure reports about reT are different than yours
masak nonetheless, that probably means that behind my failures are more failures
pugs_svn r21632 | fglock++ | [v6.pm] more operators 12:32
Rashad01 hi to all 15:21
moritz_ hi 15:22
zamanfou is now away: In the rain my tears are forever lost. 15:24
pugs_svn r21633 | fglock++ | [v6.pm] 0.031 15:25
pugs_svn r21634 | pmurias++ | [smop] added smop_s1p_code (forgot to in the previous commit) 16:10
Auzon moritz_: thanks :) 17:12
moritz_ Auzon: no problem 17:13
STD.pm in evalbot is really nice ;-)
did you also add a cron job that rebuilds it from time to time?
Auzon Not yet
Auzon I need to add that and also make the failed output more informative 17:14
[particle]1 is it part of 'perl6:'? 17:14
moritz_ perl6: say "dunno" 17:15
p6eval elf 21634: OUTPUT[dunno␤]
..pugs: OUTPUT[dunno␤]
..rakudo 29834: OUTPUT[dunno␤]
moritz_ [particle]: does that answer your question? ;-)
[particle] yes, however *i* didn't spam the channel when asking :)
Auzon :P
I didn't think it'd go well, since most code directed to perl6: parses 17:16
moritz_ [particle]: hehe ;)
what we could do is to summarize identical output
[particle] has anybody run make test in pugs recently? 17:17
i wonder how it fares with the spectests
moritz_ so the above you just be `elf 21634, pugs , rakudo 29834: OUTPUT[dunno␤]'
[particle] it'd be nice to compare rakudo to pugs that way
interesting that pugs doesn't have a revision 17:18
masak [particle]: trying...
no, I can't build Pugs
I've got it through Cabal 17:19
moritz_ same here
maybe we could run test without 'make'?
[particle] prove t/spec
masak good idea
[particle] or prove6
moritz_ [particle]: pugs can't parse new-style POD 17:20
[particle] nice.
so, rakudo is passing more spectests then pugs :)
moritz_ let's see, there are still some files wihtout POD 17:21
masak moritz_: maybe one could fudge away the new-style POD in pugs? 17:22
moritz_ masak: I think most of the POD can simply be removed, and substituted with plain old comments
masak good.
[particle] what's wrong with pod? it's so easy to parse 17:23
moritz_ [particle]: yes. But it has to be done ;)
[particle]: feel free to patch pugs' parser... oh, wait... ;)
pmurias moritz_: what does he have to wait for? 17:24
lambdabot pmurias: You have 1 new message. '/msg lambdabot @messages' to read it.
[particle] iirc you can build pugs with a sufficiently old rev, and ghc 6.6.x
moritz_ pmurias: haskell fu?
[particle] it seems silly to me to take pod out of the official test suite because of a broken implementation
moritz_ yes 17:25
but is there a good reason to keep it?
I mean the coresponding synopsis is stalled and not being discussed, and in such a state rather worthless 17:26
[particle] um, as documentation?
moritz_ and what does it offer us over plain comments, in the form it is currently used?
[particle] damn, i should have asked damian about the pod spec last week
moritz_ aye 17:27
Files=277, Tests=5012, 420 wallclock secs ( 7.00 usr 2.20 sys + 386.05 cusr 9.94 csys = 405.19 CPU)
Result: FAIL
Failed 182/277 test programs. 531/5012 subtests failed.
seems like pugs still wins ;)
[particle] ok, then. time to add more pod :) 17:28
pmurias the ghc in lenny compiles pugs?
moritz_ pmurias: yes, if you use the cabal route
a simple "make" doesn't work 17:29
[particle] iiuc, the problems with pod6 are entirely outside the bounds of its current usage
i wonder...
moritz_ [particle]: not entirely 17:30
pmurias moritz_: the README is wrong than?
moritz_ [particle]: if we acknowledge that we need to have stuff that is both POD and program, such as sub declarations, POD syntax might change
[particle] does the 'sublanguages' concept as applied to regexes, quoting, et cetera in the perl 6 grammar allow the tools necessary to alleviate the concerns regarding pod
moritz_ pmurias: the README just says "see INSTALL" 17:31
pmurias: and the INSTALL describes several methods, one of which actually works
[particle]: the point has always been that you shouldn't need the Perl 6 grammar to parse POD 17:32
pmurias moritz_: the methods that don't work should be documented as such 17:32
[particle] moritz_: what if perl 6 is a sublanguage of pod? 17:33
moritz_ [particle]: why should it be? doesn't sound logical to me 17:34
moritz_ pmurias: I agree 17:34
masak "Perl 6: a sublanguage of POD... with side effects"
[particle] :)
moritz_ do you think I can send a POD6 proposal to p6l without being killed by Damian and the rest of the crew? 17:35
masak: ;)
[particle] yes
masak moritz_: yes
moritz_++
moritz_ I'll have to brood a bit more over it, so it probably won't be until 20th of august or something because I have vacations inbetween 17:37
pmurias did the ghc core represetation change much from the one describe in "An External Representation for the GHC Core Language." 17:46
www.haskell.org/ghc/docs/papers/core.ps.gz
?
Auzon moritz_: STD.pm should be rebuilt every 5 minutes now 17:52
moritz_ Auzon: very good 17:53
[particle] hrmm, i may have to get all MakeMaker on src/perl6/ :/ 17:56
pmurias [particle]: can't you just grab a make? 17:57
[particle] i'm on windows 17:59
'./gimme5' doesn't work
Auzon Some part of it doesn't work on Mac, either, from what I hear
[particle] if i convert to 'perl gimme5' then will it respect the #! in gimme5?
moritz_ [particle]: only the arguments (like -w, if any) 18:01
[particle] figured as much. i suppose that's ok, since there's 'use 5.010' in the files
moritz_ [particle]: if you want to change Makefile (and commit the changes), rather make it 'perl5.10.0 gimme5' 18:02
pugs_svn r21635 | moritz++ | [evalbot] group identical results in output of 'perl6:' 18:07
moritz_ std: 1
p6eval std 21634: OUTPUT[parse ok] 18:07
moritz_ Auzon: did you modify evalbot locally? 18:09
Auzon Yes, but the changes are in SVN as well 18:10
moritz_ because I got a conflict on updating
on timtowtdi.org
Auzon Sorry about that :-/
moritz_ I simply reverted it
Auzon: no problem, I just wanted to know if it had changes worth preserving 18:11
Auzon Nope. I just wanted to edit it there, then realized it was easier to commit on my computer, and I didn't get the exact same changes 18:11
moritz_ rakudo: say "basic sanity" 18:11
p6eval rakudo 29834: OUTPUT[basic sanity␤] 18:12
[particle] moritz_: there is no perl5.10.0 on windows. there is only 'perl'
moritz_ perl6: say "basic sanity"
p6eval elf 21635, pugs, rakudo 29834: OUTPUT[basic sanity␤]
moritz_ [particle]: I didn't know that. How inconvenient
[particle] oh, oops, i do have 5.10.0 (i typed it wrong)
moritz_ YaY, less line noise from p6eval
perl6: print 2.perl 18:13
p6eval elf 21635, pugs: OUTPUT[2]rakudo 29834: RESULT[21]
moritz_ oops, missing a newline
pugs_svn r21636 | moritz++ | [evalbot separate different outputs from 'perl6:' with newlines 18:15
[particle] what's the diff between OUTPUT[] and RESULT[] ? 18:15
moritz_ [particle]: RESULT is when something didn't produce output 18:15
perl6: print 2.perl 18:16
p6eval elf 21636, pugs, rakudo 29834: OUTPUT[2]
[particle] odd
moritz_ I got a segfault from parrot during the first eval
pugs: print 2.perl
[particle] perl6: $_ = 'foo'; .say;
p6eval pugs: OUTPUT[2]
pugs, rakudo 29834: OUTPUT[foo␤]
..elf 21636: OUTPUT[Use of uninitialized value in concatenation (.) or string at ./elf_f line 3408.␤syntax error at (eval 115) line 4, near ";␤->say"␤ at ./elf_f line 3861␤]
viklund` uhm, bug in rakudo I think: 18:44
rakudo: my $p='monkey'; say substr($p,0,0) 18:45
p6eval rakudo 29834: OUTPUT[monkey␤]
masak ah, that's probably because the second 0 is treated as nonexistent 18:47
moritz_ which is wrong, of course
is there a test for that?
masak aye
...it's wrong, I mean 18:48
don't know about the test
moritz_ doesn't seem like 18:49
I'll add one
pugs: my $p='monkey'; say substr($p,0,0)
p6eval pugs: OUTPUT[␤]
viklund` this works 18:50
masak ah, it treats -1 as last char minus 1
viklund` rakudo: my $p='monkey'; say substr($p,0,-$p.chars)
p6eval rakudo 29834: OUTPUT[␤]
masak and 0 as last char
pugs_svn r21637 | moritz++ | [spec] added substr test with 0 as third arg, masak++
moritz_ but the last arg should be length, not position, right? 18:51
masak moritz_: that too 18:51
moritz_: unless negative
moritz_ masak: care to open a ticket? 18:53
masak moritz_: oki 18:54
moritz_ #perl6-soc in 25 19:09
Auzon Yep, :) 19:10
pugs_svn r21638 | fglock++ | [v6.pm] added docs for the YAML backend 19:32
r21639 | fglock++ | fixed last commit 19:34
pugs_svn r21640 | fglock++ | [v6.pm] operator fixes 20:49
pugs_svn r21641 | moritz++ | [spec] another substr test 20:52
Limbic_Region salutations all 22:49

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!