pugs.blogs.com/dist/Perl6-Pugs-6.2.13.tar.gz released! | run.pugscode.org | spec.pugscode.org | paste: sial.org/pbot/perl6 | pugs.blogs.com
Set by audreyt on 17 October 2006.
eviltwin_b remmebers why the file-time things ara a pain in the butt in Haskell... 00:03
eviltwin_b ...yep, looks like they should return Integers. yug 00:32
svnbot6 r14541 | audreyt++ | * Pugs.Parser.Operator: Sqaush the annoying 'term or "\("' 00:51
r14541 | audreyt++ | message on parse-errors waiting on a term.
audreyt (at houston airport; boarding in 30min) 00:52
cmarcelo audreyt: ping 01:07
audreyt cmarcelo: pong 01:08
(but only for 20min)
ingy: I'll probably not make it to Seattle now that neither stevan nor lwall is going to make it afaics
cmarcelo audreyt: "darcs pull moh" before getting in the plane :o) 01:10
audreyt already did 01:11
:)
first thing I did, actually 01:12
cmarcelo what's the estimated time of arrival? 01:17
I will arrive in São Paulo by Wed or Thu, you folks will be hacking at the hotel or somewhere else or dunno?
audreyt ooh, nice, I look forward to two days of intense MO hacking with you 01:19
I'll stay in conference hotel
probably will hack there
will arrive tomorrow morning
boarding now... see you in brasil!
cmarcelo see you :) 01:20
eviltwin_b hm, filetimes need $*BASETIME. mmm, yak shaving 01:46
eviltwin_b well, this ought to be interesting :) 02:15
eviltwin_b hm, whoops, forgot to deal with the EpochTime issue setting up $*BASETIME. guess I'll deal when it errors out :) 02:22
hm, smells like I've committed an encapsulation violation. feh. 02:30
eviltwin_b but I did introduce a dependency loop. feeeeeh. 02:33
wish S29 didn't commit to keeping perl5's semantics for -M/-C/-A 02:34
eviltwin_b woo, think I got it (and fixed the encapsulation issue at the same time) 02:47
guess I meant S16 instead of S29 earlier. 03:02
Eidolos audreyt: I'm reading through your recent Pugs slides. Very nice work. :) 06:47
eviltwin_b so if I have implemented a new operator (which the parser already knows about), jhow do I hook it in? 06:54
I thought just adding the appropriate `op1 ...' foo (in this case `op1 "-M"') would do it due to pattern matching, but pugs complains about prefix:-M not being defined 06:55
hm, ok, so that error implies it found it but thought it mismatched in types. think I'm too tired to untangle tis now 07:03
Haskell types are (obviously) correct, not sure how the internal Perl types work although I cloned the file size stuff to set it up
lumi eviltwin_b: I think it needs to be added to the prec table? 07:05
eviltwin_b hm, looked to me like all the file tests werer already there, just without implementations 07:10
(but with tests, somewhat incorrectly)
lumi Which I think is initSyms in Pugs.Prims
See, the other -X are in there
eviltwin_b ah, yes, missed that somehow. rebuilding now. lumi++ 07:12
eviltwin_b should probably not have bitten off quite so much the first time :)
lumi Nonsense, always leap before you think 07:13
eviltwin_b huh, didn't realize perl5 returned fractional days. okay, get to tweak a bit 07:18
otherwise, I appear to have success :)
palomer will I have to put @ before hash tables in perl 6? 07:22
Eidolos The grades test case (in junctions) is kind of weird. if (one(@grade) < $threshold) { you may go to the movies } else { you may not } 07:36
But if all your @grades are above threshold, you still may not go to the movies.
It should probably use all(@grade) >= $threshold 07:37
(but then it might just be a test for one in which case the odd output is probably excusable) 07:38
Er, it's not a test case, sorry, it's an example. 07:42
eviltwin_b palomer: a hash variable would always be prefixed by % --- you never use @ (like in perl5 for hash slices) or $ (for individual lookups) 07:43
woo. 07:45
only problem with this code (aside from someone probably disliking the style :) is silliness with EpochTime 07:46
afaict the only way to get it into a useful numeric form is read $ show
eviltwin_b hmph. think I confused svn somehow; svn diff (empty) and svn status (various generated files) don't do useful things 08:16
Eidolos svn status is probably reporting mysteriously new files, and svn diff isn't showing them because they aren't svn added. 08:18
eviltwin_b er? the created files are things like Makefiles
I added no files otherwise
Eidolos What does svn status look like?
eviltwin_b but I'd expect to see the changes I made to e.g. src/Pugs/Prim.hs
Eidolos oh
eviltwin_b ktrace suggested svn diff wasn't doing much of anything, actually 08:19
eviltwin_b huuuuh. "make test" blocks svn somehow? now that it's done, svn status etc. work 10:27
Debolaz Can a perl 6 module inherit from a perl 5 module with pugs? I see it's possible to use a perl 5 module, but inheriting didn't work as smoothly. 11:06
pmurias hi, 12:54
i'm coding a symmetry detection algorithm
Juerd Images? 12:56
pmurias how do i determine the angle formed by edges at each vertex if i know the coordinates of the three points?
Juerd: polygons
pmurias Juerd: it probably would be simpler to do it with images as there seems to be more info on it 12:57
:) 12:58
cs.haifa.ac.il/hagit/courses/semin...tching.ppt 13:00
if you are intrested in the problem
pmurias trigonomery skills are lames, so he has to ask to people for help with the mathy part :( 13:02
s/second to//
Juerd My trigonometry skills are near to non-existent 13:07
I know Pythagoras, and that's about it :)
(Have been using it last week, though)
svnbot6 r14542 | kudra++ | This week's summary 13:20
pmurias Juerd: thanks for trying to be helpfull, i'll code the KMP and ask somebody at school tomorrow 13:40
pasteling "rindolf" at 88.152.243.90 pasted "Perl 6 Building Problems" (22 lines, 1K) at sial.org/pbot/20731 14:58
rindolf So what do I do to resolve sial.org/pbot/20731 ? 15:01
lambdabot Title: Paste #20731 from "rindolf" at 88.152.243.90
integral looks 15:07
you haven't installed enough GHC
did you install from source?
If you installed from source, you need to follow what the last paragraph of the error says: you need both -src and -src-extralibs (untar them both from the same dir, they both untar into the ghc-6.6 subdirectory) 15:08
rindolf integral: I compiled the ghc-6.6 SRPM that I adapted myself. 15:11
integral: I didn't unpack src-extralibs.
integral That's the problem then.
rindolf integral: so I'm going to do that.
integral If you understand how to build cabal modules indidually, you could build extralibs as its own srpm(s)
*individually
svnbot6 r14543 | audreyt++ | * stm-contend.pl: Sleep 0.1 instead of 1 sec each step. 16:06
r14544 | audreyt++ | * Pugs.Prim: async{...} should use user-level threads by 16:09
r14544 | audreyt++ | default; system-level threads are to be created via some
r14544 | audreyt++ | other mechanism.
r14544 | audreyt++ | * Pugs.Prim: Allow sleep for fractional seconds.
r14545 | audreyt++ | * util/build_pugs.pl: "make prof" doesn't work on threaded
r14545 | audreyt++ | runtime, so disable -threaded when building profiled.
rindolf Man! It takes a helluva long time to build ghc. 16:21
Debolaz Can perl 5 modules be used for inheritance, or are you merely able to 'use' them? 16:54
audreyt Debolaz: they are not yet integrated into the metaobject model 16:58
so currently no, but I'll keep that req in mind this weekend working with cmarcelo on MOH
there's nothing preventing them from being inherited, it's just SMOP :)
Debolaz Ah. 16:59
Debolaz tried using DBIx::Class. :-) 16:59
rindolf audreyt: what's SMOP?
Debolaz It's probably a good thing (tm) if that works at some point.
rindolf audreyt: hi, BTW. 17:00
Debolaz rindolf: Some Matter of Programming.
s/of/Of/
audreyt *Simple
Debolaz Yes, that. 17:01
:)
Debolaz Still trying to find a project to use perl 6 for. It can't be something big and important, but it should be something I need. :/ 17:06
audreyt nodnod 17:10
to be honest 90% of my use of pugs at this point is one-off scripting
that involves a lot of =$=DATA and .comb and the sorts
nothingmuch .hairgel 17:11
rindolf Hi nothingmuch 17:15
I can't believe it! I built ghc for the third time and Pugs still isn't happy.
audreyt goes catch up sleeping... 17:18
nothingmuch hi rindolf 17:22
integral what's it this time? 17:24
Debolaz ponders if O'Reilly even bothered to test their Safari Books system with Opera. 17:25
I take that back, I can't get it to work properly in any browser. 17:31
Patterner lynx? 17:33
rindolf integral: the %setup macro unpacked -extra-libs into the wrong directory. (ghc-6.6/ghc-6.6/ instead of ghc-6.6/). I fixed it now. 17:37
integral ah 17:39
TSa HaloO 19:39
rindolf Hi TSa 19:44
integral hmm, anyone had any luck with profiling pugs? I've just run a "pugs-prof examples/mandel.pl" but the pugs-prof.prof file is 0 length :-/ 19:46
TSa HaloO rindolf 19:49
rindolf Yay! I finished building ghc. 20:08
xie Hi, I'm wondering if it's possible to use Perl 6 rules from Pugs without Parrot 20:20
integral afaik know the current version uses the perl5 implementation by default 20:23
xie Hmm, I'm running with #!/usr/local/bin/pugs 20:28
use v6;
and I get the error *** Cannot parse PGE: ^\-?\d?\.?\d+$ 20:29
*** Error: Cannot find the parrot executable in PATH
integral oh, hum. 20:30
integral hrm, what's the difference between the Val type from Pugs.AST.Internals and Pugs.Val? Is the latter a new one that needs transitioned to still? 20:46
integral wonders why op2Numeric uses pattern guards instead of normal pattern matches 20:48
eviltwin_b d'oh! missed fromEnum somehow. code is much cleaner suddenly. 21:54
but still have a bit of a problem. need to wait for this test run to finish (*sigh* svn oddness) to put diffs up somewhere for reference 21:58
pugs uses ClockTime, stat times are EpochTime, I'm not sure they're guaranteed to be related. so $*BASETIME is an EpochTime instead of a ClockTime 22:00
eviltwin_b (obviously on unixlike systems they're related, one is gettimeofday() and the other is time(). no clue about e.g. win32) 22:10

Powered by Cro and the Raku® Programming Language.

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

Thank you!