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.
Eidolos so let's throw away the last seven years of implementation work and start over? >_> 00:01
wolverian they're not mutually exclusive..
perl 6 is a spec, not an implementation
diakopter Eidolos: again, :P ... there wouldn't be much that's not reusable, actually. 00:02
Eidolos Parrot?
diakopter the hard work is the find-grained specifying.
(of PGE, I'm thinking of)
fine-grained
Eidolos shrugs 00:03
diakopter nearly all of pugs would be reusable for #2, but yes, the REPL and unicode stuff could come directly from ghc. 00:04
and probably other stuff I don't have in the front of my brain at the moment.
diakopter frankly, I think the bootstrapping step would be nigh impossible anyway, due to Perl 6's (ever-further-pervading) requirements of laziness & parallelism, and the desired-but-as-yet-unspec'ed unified concurrency/event model/api. So much cross-platform stuff would need to be reimplemented... So many functional compiler optimizations would need to be ported from ghc's bootstrapped haskell compiler source... 00:13
based on my *vast* experience writing bootstrapping compilers LOL <eye-roll> 00:14
Zoffix diakopter, Regarding that typo in the href="". I never used `svn` before so I don't want to mess anything up. After I edit the files.. what do I do? 01:13
Never mind. 01:17
diakopter hm; how about you just nopaste the edits... sial.org/pbot/perl6 01:18
Zoffix What's there to nopaste? .... 01:19
heh
Whatever, nevermind.
pugs_svnbot r17142 | zoffix++ | fixed 404 01:23
r17143 | zoffix++ | fixed 404
diff: dev.pugscode.org/changeset/17143
lambdabot Title: Changeset 17143 - Pugs - Trac
Zoffix ewww.. trans >_< 01:25
pugs_svnbot r17144 | zoffix++ | fixed 404
diff: dev.pugscode.org/changeset/17144
lambdabot Title: Changeset 17144 - Pugs - Trac
Zoffix Done 01:27
pugs_svnbot r17145 | zoffix++ | fixed 404 on Perl6 link
r17146 | zoffix++ | fixed 404 on Perl6 link
diff: dev.pugscode.org/changeset/17146
lambdabot Title: Changeset 17146 - Pugs - Trac
Zoffix HTML code needs a major rewrite but I won't be doing it today ;)
Web "Edit Repos" is kinda broken BTW. I would submit the changes but it would say "ERROR: file is identical to the original" even though it isn't. I'd then hit F5 and it would commit :/ 01:32
diakopter Zoffix: thanks for reporting that bug with the "Edit Repos"... I'll email the vendor about it 02:35
Zoffix: also, thanks for fixing all those regressions. :) did you add yourself to /AUTHORS ? 02:38
Zoffix umm. I didn't do anything. But I am listed under "Authors" section somewhere on the site.. 02:39
RE: bug; No problem. It might be some caching issue on your server that I suspected causing IE6 not to scroll down to #fooBar links.
diakopter I meant, thanks for fixing all the regressions [that you did not cause :)]. 02:40
diakopter (the removed spaces) 02:41
Zoffix No problem. 02:42
Teratogen lol 03:50
masak it often seems to me that it would be possible to construct a CLI or editor tool that calculated that number after 'plan' in test files 10:08
the Entscheidungsproblem nonwithstanding 10:10
polettix in dev.perl.org/perl6/doc/design/exe/E...'s_a_Given it is said that this "ancient notation": 10:12
lambdabot Title: Exegesis 4: Syntax - perl6:
polettix when /^\w $/ { print "$name's on first\n" }
now would be written:
when /^ \w $/ { say "$name's on first" }
masak polettix: yes
polettix I couldn't help noticing that two spaces popped up in the regex 10:13
does this mean that the /x switch is implied?
masak polettix: yes
polettix: not sure if you can still leave out the spaces, though
polettix masak: do you mean that /^\w... would be an error? 10:14
pmurias fglock: hi ;) 10:25
fglock pmurias: hey
pmurias shouldn't binding just be assignment instead of calling 'BIND'? 10:28
pmurias is in a subversion free zone so dosn't have handy access to the pugs repo 10:40
fglock pmurias: it could be, after the optimization phase - but at this point we need to keep it as a real operation 10:54
meppl gugu 11:00
masak meppl: gugu 11:21
polettix: I mean I don't know whether it'd be an error or not
masak how come localtime() works in the DateTime test, but not in the module itself? 12:06
polettix masak: according to a bit of documentation I've read, not including spaces should not be a problem 12:49
masak polettix: good
polettix I would have been surprised of the contrary :) Thank you for answering, it was really something that the good ol' RTFM would have clarified in a snap 12:50
masak :) 12:51
polettix even tough I can justify myself stating that i RTFM then :)
*though 12:52
masak polettix: do you have time to help me investigate the localtime() strangeness? 12:56
you'll notice it when you run ext/DateTime/t/basic.t
cognominal_ babel sploder 18:00
oops 18:01
diakopter :) 18:04
spinclad perhaps it sploded the lack of babble, for a moment 18:23
dduncan opinion question: supposedly the identity value of the set intersection operator is a potentially infinite-sized set ... but if you're working in an environment where only finite sets are allowed, what's the best result for a reduce over set intersection? 19:15
eg, what should [U] return
for an empty list
or should it just fail() ? 19:16
TimToady maybe it should return * :) 19:17
dduncan whatever? 19:18
TimToady sounds kinda like the universal set
dduncan so basically you're saying it should return something, and not die?
dduncan that's the main thing I wanted to know, return something special, vs die 19:20
TimToady possibly it should return the Object protoobject 19:23
$x ~~ Object should always return true, so it is essentially the universal set 19:24
amnesiac ~~~~~ 19:29
meppl good night 21:20
masak meppl: 'night 21:22
meppl ;) 21:23
masak the Internet is fantastic. thousands of people are connected all over the world, and can wish each other good night with second-fast communication. this can actually bring smiles to people's faces, a fact which is communicated through a semicolon and a closing paranthesis 21:25
I think it's cool, really 21:26
thorat the internet has its moments ;) 21:27
masak thorat: ;) 21:28
dduncan opinion question: when speaking of a duration of time, does it make sense to have a negative duration, such as the result of subtracting a newer datetime from an older one? 23:30
spinclad sure 23:32
dduncan okay
spinclad time measured in seconds or days, at least, is affine (linear but without a 0 point), as i see it 23:34
spinclad time differences are linear 23:34
dduncan sure, well I see no reason not to support positive and negative durations, but I realized that I hadn't so far, so I'm just checking that it's good to switch ... seems that it is 23:45

Powered by Cro and the Raku® Programming Language.

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

Thank you!