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.
adorablepuppy hello 02:52
Tene Hi! 02:54
adorablepuppy I need regex help. :) Though I am currently asking in #regex. 02:57
Tene I can help you with Perl 6 Rules, which are the replacement for regexes. 02:58
adorablepuppy ok
Tene If you want regex help, though, #regex or #perl are better places. 02:59
adorablepuppy thanks 03:00
pugs_svnbot r18252 | Darren_Duncan++ | ext/Muldis-DB/ : the Muldis DB API now has users specify the AST language of interaction 04:03
diff: dev.pugscode.org/changeset/18252
lambdabot Title: Changeset 18252 - Pugs - Trac
pugs_svnbot r18253 | Aankhen++ | [kp6-lisp] 09:47
r18253 | Aankhen++ | * don't explicitly specify package of MAIN function in generated code.
r18253 | Aankhen++ | * implemented `our` variables.
diff: dev.pugscode.org/changeset/18253
lambdabot Title: Changeset 18253 - Pugs - Trac
pugs_svnbot r18254 | Aankhen++ | * [kp6-lisp] implemented truth for hashes. 09:56
diff: dev.pugscode.org/changeset/18254
lambdabot Title: Changeset 18254 - Pugs - Trac
pugs_svnbot r18255 | Aankhen++ | * [kp6-lisp] more robust (DECLARE (IGNORABLE ...)) handling in WITH-KP6-{PAD,PACKAGE}. 09:59
diff: dev.pugscode.org/changeset/18255
lambdabot Title: Changeset 18255 - Pugs - Trac
fglock I was reading about efficient threads, and I thought how about using BerkeleyDB (and C) as a backend 11:57
avar Aankhen``: boo 12:20
fglock: why use bdb for threads?
Aankhen`` avar: Eek. 12:26
avar echo '%hash.elems' | perl kp6-mp6-perl5.pl --lisp 12:37
I guess that's object methods for ya:)
Aankhen`` Indeed.
I'm working on signatures right now.
Then I'll turn to subs. 12:38
avar :)
Aankhen`` Why did the metamodel-in-Perl 6 die out? 12:39
Aankhen`` wanders off for a while.
allbery_b andrea vanished 12:40
er, audrey. /me thwaps brain
Aankhen`` allbery_b: Ah, I didn't realize audrey was also working on it. Looks like you're not the only one whose brain is malfunctioning. ;-) 12:48
allbery_b just isn't fully awake yet 12:49
avar also? who else? 12:50
fglock avar: re bdb, because it seems to do it right 12:56
fglock or maybe not - www.oracle.com/technology/documenta...am/mt.html 13:02
lambdabot Title: Berkeley DB Reference Guide: Multithreaded applications, tinyurl.com/2ekxq3
allbery_b bdb seems to crap itself a lot, in my experience 13:03
fglock goes read about erlang again 13:20
Juerd lwn.net/Articles/252073/ 13:23
lambdabot Title: To Sir, with Love: How To Get More Women Involved in Open Source (O'ReillyNet) [ ...
Juerd Oh my...
avar fglock: What does bdb have to do with /threads/? 13:38
Use it to store something related to them
?
allbery_b several packages use bdb's locking specifically for thread synchronization, ignoring the actual database foo
fglock avar: I don't know yet, I'm reading about it 13:39
fglock i'm not particularly interested on bdb, just fast threads 13:41
ferreira hi folks! I invite you to take a look at the next article on Perl 6 operators: feather.perl6.nl/~ferreira/perl6-op...pare2.html 13:43
lambdabot Title: compare2.pod6
fglock ferreira: @values.sort( { $^b cmp $^a } ) # don't need the arrow 13:48
@values.sort: { $^b cmp $^a } # don't need paren 13:49
ferreira fglock: ok. This is much prettier ;-) That means placeholders are not only for -> { } things ? 13:57
rhr in fact I think placeholder vars are illegal in pointy blocks by S06:1402
ferreira oops
Gothmog_ But you could write -> $a, $b { $b cmp $a } iirc 13:59
That of course isn't as nice as { $^b cmp $^a }. :)
ferreira now fixed.
rhr TimToady: should S04:293 have the -> removed because of the $^thing ?
ferreira Gothmog_: that's exactly what I thought ;-) 14:00
Juerd ferreira: Comments like "0 before 1" are vague. Do they describe the code? the return value?
ferreira hum, not really, I just thought of that as anterior to introducing Order::Increase, etc. 14:02
masak ferreira: nice article 14:04
ferreira maybe some extra words would convey the intented meaning, without having to repeat Order::Increase, ::Same, ::Decrease thrice. What do you think of "answers 0 before 1" instead of just "0 before 1" and so on?
masak: thanks, you're very kind
masak in a real-life app, @values.sort: { $^b cmp $^a } should probably be just @values.sort.reverse or similar, because of higher readability 14:05
but as a proof-of-concept, your example code is just fine
Juerd ferreira: Operators don't "answer"... They return or evaluate. A full sentence may be in order if you want to avoid the classes 14:06
ferreira masak: got that. I prefered the redundancy that causes 'cmp' to be used. The last example as well would be better written with KeyExtractor's, but I don't even know how to write them, but I think it would be something like: 14:10
ferreira @values.sort: ( { $^h<name> }, {$^h<id> } ) but that would be too much (I mean, explaining all these in a light article on <=>, leg and cmp) 14:11
lambdabot Unknown command, try @list
Juerd ferreira: Maybe "returns that 0 is before 1" 14:17
ferreira: Maybe explain cmp more, like that it does numbers, strings, and objects 14:18
pmurias hi 14:22
fglock pmurias: hi 15:06
avar fglock: isn't there some abstraction for {_value}->{_array} and {_hash} in the p5 runtime? 15:08
fglock avar: not sure what you mean, that's the low-level representation 15:09
avar just wondering why it wasn't say ::DISPATCH($foo, "array") 15:11
avar Just wondering how best to implement stuff like ' ( @{ ' ~ $cond.emit_perl5 ~ '->{_value}{_array} } )' in CL 15:11
pmurias avar you want to get to p5landish value
fglock ::DISPATCH($foo, "array") works, but it returns an Array object; you need {_array} (or p5landish) to get to the p5 data 15:12
pmurias i get it its perl->cl
fglock: what tests are supposed to fail? 15:13
avar so stuff like ' ( ' ( @{ ' ~ $cond.emit_perl5 ~ '->{_value}{_array} } )'@{ ' ~ $cond.emit_perl5 ~ '->{_value}{_array} } )' can be abstracted but the EmitPerl5 emitter is just peeking at internals since it's old code?:) 15:14
fglock avar: maybe, can you get an iterator object from lisp Array
fglock i think 'for' will end up being implemented as a method on Array/List/Range 15:14
pmurias: re tests - none :) 15:15
Aankhen`` fglock: No iterators in CL, AFAIK. 16:12
avar no, but you can roll your own in any language 16:13
Aankhen`` Sure,
s/,/./
pinskian xml is arghhh 17:06
pmurias pinskian: Captures are a sort of xml 17:09
pinskian i want to put digital sig of a file inside xml file, a field will be file name, another will be file signature (for example) 17:13
how i do not know of
pmurias <file sig="" filename=""/> 17:15
what's the problem?
pinskian BUT 17:16
newline-free text string
pmurias pinskian: it it ASCII?
pinskian no
pmurias uuencode it 17:17
or base64 it
pinskian exactly base 64 encoding 17:18
thought xml supported newline
pmurias "exactly base 64 encoding" means? 17:19
pinskian there is no work aorund than having to base64 encode t
it
like within xml it self a newline-free txt string 17:20
Aankhen`` didn't understand the problem. 17:27
pmurias how to store binary data in a xml file 17:28
Aankhen`` Why use XML for binary data? 17:28
(ignoring "why use XML at all?" for now.)
/\(i/(I/ 17:29
s/^/s/, too. :-P
pmurias the binary data is a checksum
Aankhen`` Ah.
pmurias a digital signature
pmurias a script to check in which of the last revision in kp6 a test work would be helpfull 17:35
pmurias * revisions,worked 17:37
pugs_svnbot r18256 | pmurias++ | [kp6] Match isa Capture
diff: dev.pugscode.org/changeset/18256
lambdabot Title: Changeset 18256 - Pugs - Trac
pmurias hates typoes 17:38
fglock: test 50 fails 17:40
fglock pmurias: looking 17:43
ludan hi 17:44
fglock home & 17:52
/home - not yet 17:57
pmurias: it's broken for some time
now home &
pmurias fglock: autothreading on method calls broke it 18:03
the $.type should be specified in the constructor rather the by calling an attribiute 18:04
TimToady fglock: fwiw the semantics of for are currently defined to be equivalent to map. 18:11
ludan goodnight 23:01

Powered by Cro and the Raku® Programming Language.

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

Thank you!