pugscode.org/ | nopaste: sial.org/pbot/perl6 | ?eval [~] <m oo se> | We do Haskell, too | > reverse . show $ foldl1 (*) [1..4] | irclog: irc.pugscode.org/
Set by diakopter on 11 July 2007.
lichtkind PerlJam: hejho 00:01
meppl good night 02:10
dlocaus moritz_: you said you try and traslate some of the synopsises.. .what do you get stuck at? 03:03
moritz_ @tell dlocaus re translating synopsis: mostly lack of time and motivation 06:16
lambdabot Consider it noted.
agentz1 translating is tough :) 06:20
moritz_ aye 06:28
PessimistByNatur just popped in to say hello, and good luck ^_^ 06:36
[particle] thanks!
agentz1 grins.
PessimistByNatur I would attempt to help, but seeing as I don't know much perl at all, and that I'm still learning how to program in general, I'll leave it in your capable hands ^_^ 06:37
agentz1 heh
moritz_ PessimistByNatur: that attidue fits your nick name ;-) 06:38
PessimistByNatur I see no contradiction, as long as one can accept that one can't often control external circumstance, it's fairly simple to stay content even when you think everything will go wrong ^_^ 06:39
moritz_ mumbles something inaudible about "low standards", but generally agrees 06:41
PessimistByNatur keeps reading happily about perl 6, since he didn't hear anything ^_^ 06:42
moritz_ well, it's moving, at it's own, slow pace 06:44
PessimistByNatur no rush, it will be done when it is done ^_^ 06:45
moritz_ aye, that's our atitude as well ;)
[particle] luckily, otherwise we'd be *way* behind 06:47
akreal is away: Отошёл. 07:31
pugs_svn r19050 | fglock++ | [kp6-perl5] fixed array autovivification on ".push" 12:03
pugs_svn r19051 | fglock++ | [kp6-kp6] "Decl.emit_perl5" works 14:34
moritz_ what would æ match with the :basechar modifier in scope? an 'a', or 'ae'? or nothing of those? 14:42
pugs_svn r19052 | fglock++ | [kp6] Grammar - escape sequences inside single-quotes 15:13
pmurias fglock: got my messages? 15:14
fglock @messages 15:15
lambdabot pmurias said 17h 22m 47s ago: while attempting to remove eval uncovered a bug
pmurias said 17h 21m 9s ago: look at line 1586 compiled/perl5-kp6-kp6/lib/KindaPerl6/Grammar/Regex.pm inside eval
fglock looking
fglock pmurias: line 1586 is the end of file 15:18
pugs_svn r19053 | fglock++ | [kp6-kp6] fix Pad eval 15:22
moritz_ pugs: say 'a' ~~ m:b/a/ 15:25
exp_evalbot OUTPUT[*** Cannot parse regex: a␤*** Error: ␤␤]
pmurias fglock: sorry line 72 15:40
fglock k
pugs_svn r19054 | fglock++ | [kp6-kp6] Pad.add_lexicals work 15:44
rhr moritz_: it looks like æ doesn't have a unicode decomposition, so it wouldn't match any of those (not what I was expecting)
moritz_ rhr: thanks. I consider it a Unicode bug ;) 15:46
fglock less than 20 days to Christmas 15:51
moritz_ omg
I don't have a single present yet 15:52
[particle] you should have gone to nepal, presents are cheap there :)
moritz_ [particle]: and no family around, so need for so many presents anyway ;) 15:55
pugs_svn r19055 | fglock++ | [kp6-kp6] pointy block parsing generates a Sub node 15:58
pmurias fglock: seen the bug 16:00
? 16:01
fglock yes, fixing
thanks :)
pmurias fglock: remove the eval while forcerecompiling
fglock does it work without eval? 16:02
I wonder how circular dependencies are resolved
pmurias fglock: i can hand remove the bug and test
fglock and self-references
pmurias run_test.pl --backend=perl5rx works with kp6 16:03
except for a test which was broken before
transforming compiled code with vim is not nice,broke my image 16:07
:( 16:08
fglock: it should work the file has "no strict 'vars'"
fglock but the interpolation will generate a '' 16:09
doesn't it?
pmurias why? 16:10
fglock: test 1-10 in t-bootstrap work 16:11
* tests
fglock: the var in (??{$var}) dosn't get interpolated
fglock ok - I'll try it out after the fix 16:12
pmurias ok
moritz_ now I get a segfault in 02-grammar.p6 even with perl5.10.0 16:13
pmurias obra: i don't with perl5.8.8 16:16
obra: sorry meant moritz_
moritz_: i don't with perl5.8.8
moritz_ pmurias: on a different mashine with the same perl it works for me as well... 16:17
pmurias: the odd thing is: otherwise it works perfectly fine, and the perl5.10 copy that I tried passed all tests
pmurias moritz_: is it the segfaulty perl which gave a segfault before?
moritz_ pmurias: I reported the segfault before, yes ;) 16:18
with the same mashine
s/with/on/
oh wait, now I can reproduce the segfault on a second debian box 16:19
afk &
pugs_svn r19056 | fglock++ | [kp6-perl5rx] fixed rule name mangling - pmurias++ 16:25
fglock moritz_: does it segfault after test #11? (unfinished code after that) 16:26
hmm - the perl5.10 error message occurs in some "normal" code 16:40
it might be a code-inside-regex bug 16:41
moritz_ fglock: no, earlier... 16:45
fglock :( 16:46
moritz_ on a server inside a chroot after # ** now testing: <val_int> 16:47
on the same server outside the chroot after ok 11 - term\n...\nadd pad
on my laptop with 5.8.8: same as inside chroot 16:48
on my laptop with 5.10.0: same as the server oustide the chroot 16:49
fglock I see a fatal error after "ok 11" - Bizarre copy of ARRAY in sassign
no idea why
with 5.10.0-RC2 16:50
hmm - there are several reports of this bug - maybe it's not my fault 16:52
"See previous threads on this" - looking 16:53
moritz_ the "previous thread" had some discussion on Carp::Heavy iirc 16:54
though I lost interest in it because I didn't understand it at all 16:55
fglock I can't find anything useful yet 16:59
pmurias fglock: i get *** Syntax Error in sub: missing closing curly bracket here 17:00
fglock pmurias: that's ok, it's a kp6 error message 17:01
but "Bizarre" is a Perl 5 error
pmurias fglock: should i forcerecompile without eval? 17:02
fglock pmurias: I think it's too early to test this, because we don't have much working code to test it on 17:05
otoh, if it fixes kp6-kp6 please do it!
coffee & 17:09
pmurias i asked if you weren't forerecompiling anything so we don't get conflicts 17:10
fglock pmurias: I'm not, feel free 17:12
pmurias if i don't have much time to spare, i can at least contribiute my CPU cycles ;) 17:13
moritz_ I built 5.10.0 with debugging symbols and know I have a stack trace of the segfault 17:18
moritz_ but I don't know perls internals good enough to do anything usefull with it 17:18
pmurias moritz_: valgrind can fire up the debuger in the place where the segfault happened 17:20
moritz_ sial.org/pbot/29055 17:20
lambdabot Title: Paste #29055 from "moritz_" at 89.12.225.191
moritz_ Invalid read of size 4 at 0x806FB52: Perl_cv_clone (pad.c:1495) 17:23
pmurias moritz_: valgrind can fire the debugger when an invalid memory read happens 17:26
moritz_: although i don't know perl5 internals at all so i can't help you 17:27
pugs_svn r19057 | fglock++ | [kp6] added new target KP6-BOOT-NOREGEX 17:41
pugs_svn r19058 | pmurias++ | [kp6-perl5rx] does not emit superflous eval's 17:43
fglock_ pmurias: the tests still pass; still get the "Bizarre copy" too 17:53
moritz_ I pasted the backtrace and valgrind's output to #p5p, Nicholas had a look at it 17:56
no result so far, but at least they know about the segfault now 17:57
fglock_ moritz_++ 17:59
I'm compiling a no-regex version 18:00
in no-regex mode, Grammar.pm compiles to... 129841 lines 18:06
moritz_ loves such small an concise programs
pmurias we are a large project now,hundreds of thousands lines of code ;) 18:08
[particle] urk
i bet that's *fast*
moritz_ [particle]: sure, it's a fast way to occupy your CPU ;) 18:09
fglock_ perl5 loads incredibly fast 18:10
pugs_svn r19059 | fglock++ | [kp6-kp6-noregex] added a simple Match.perl; 3 tests pass 18:19
fglock_ not bad - but it must be much slower than the regex backend
pmurias fglock_: what's the -I i should use? 18:27
fglock_ compiled/perl5-kp6-kp6-noregex/lib 18:29
you need to create the directories before running make 18:30
pmurias fglock_: can't you svk add them? ;)
fglock_ k
pugs_svn r19060 | fglock++ | [kp6-kp6-noregex] added compiled/ directories (empty) 18:36
pmurias fglock_: some files are missing 18:39
and some are present
fglock_ I didn't add all files 18:40
$ du -h compiled/perl5-kp6-kp6-noregex/ # 42M
pmurias fglock_: that's not good 18:41
fglock_: is the result fast enough
fglock_ it doesn't do much yet, but it's probably slow 18:42
alternately, we can compile Perl 6 regexes directly to Perl 5, like PCR does 18:44
it would take some time to implement this :( 18:45
TimToady <time passes> ...so, is it done yet? :) 19:22
moritz_ as a child when my parents said "soon", I used to answer "when is 'soon' over?" or "now 'soon' is over!" ;-) 19:23
fglock_ TimToady: :) 19:27
I got distracted
home & 19:28
pugs_svn r19061 | dlo++ | [kp6] Updated documentation, did minor reformatting. New documentation needs review. 19:45
pugs_svn r19062 | dlo++ | [kp6] cleaned up documentation a bit more. 19:57
r19063 | dlo++ | [kp6] Minor formating correction. 20:01
pugs_svn r19064 | dlo++ | [kp6] I have added the big picture to this end of this document. If someone could review it, that would be great. dlocaus@ #perl6 irc.freenode.net. 20:17
dlocaus lunch and dog walk & 20:18
lambdabot dlocaus: You have 1 new message. '/msg lambdabot @messages' to read it.
obra fglock_ ping 21:05
Tene pugs: my @a := (1,2,3); say @a.perl; 23:49
exp_evalbot OUTPUT[\1␤]

Powered by Cro and the Raku® Programming Language.

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

Thank you!