|
Prelude was borken (worked around now): sial.org/pbot/11497 | pugscode.org/ <Overview Journal Logs> | pugs.kwiki.org | paste: sial.org/pbot/perl6 | Chip needs help!!! www.geeksunite.net Set by autrijus on 4 July 2005. |
|||
| nothingmuch | ooh, those mushrooms seem to be effective | 00:25 | |
| dudley | There's something I haven't heard in a long, long time... :-) | 00:30 | |
| mugwump issues DROP ACID TAB command to dudley | 00:34 | ||
| dudley becomes engrossed in the backs of his hands and feels a peculiar urge to listen to Pink Floyd... | 00:38 | ||
| cdpruden | so, this is neat | 01:10 | |
| pasteling | "cdpruden" at 68.248.3.101 pasted "calling perl5 from perl6, inside-out hash" (21 lines, 417B) at sial.org/pbot/11583 | ||
| cdpruden | unless I'm doing something wrong -- it seems that keys and values are getting switched | 01:11 | |
|
01:16
Arathorn is now known as Aragone
|
|||
| cdpruden | crud, never mind -- I see theres already a test for it | 01:30 | |
| autrijus | greetings lambdacamels! | 02:15 | |
| autrijus is at $work and slowly recovering | 02:16 | ||
| f0rth_ | good morning | 02:24 | |
| autrijus | heya | ||
| Amnesiac | 'lo autarch | ||
| eerr autrijus :P | |||
| nothingmuch | blech | 03:16 | |
| svnbot6 | r5263, wolverian++ | S12.5: typo and grammar fixes | ||
| r5264, wolverian++ | S12.5: Further trivial fixes | |||
| nothingmuch is simply unable to fall asleep at normal people time | |||
| nothingmuch prepares for a fucked up day at work | 03:17 | ||
| seen luq | 03:23 | ||
| jabbot | nothingmuch: I havn't seen luq, nothingmuch | ||
| nothingmuch | seen luqui | ||
| jabbot | nothingmuch: luqui was seen 3 days 1 hours 6 minutes 26 seconds ago | ||
| nothingmuch | blech | ||
| Khisanth | i? :) | 03:24 | |
| nothingmuch: tried exercising? | |||
| nothingmuch | i do excercise | ||
| i ride my bicycle | |||
| it's not a persistent situation | 03:25 | ||
| it's just this jet lag | |||
| i can't fall asleep till 5 am | |||
| don't IRC as root, alexe | 03:26 | ||
| ImustDIE | does pugs have anything similar to data::dumper? | 04:36 | |
| mugwump | ?eval { 1=>2 }.perl | 04:37 | |
| evalbot6 | '{(\'1\' => 2)}' | ||
| nothingmuch | hola cwest | 05:06 | |
| nothingmuch is too lazy to rewrite his Kicker.xml hack | |||
| but the old disk is at home | |||
| *sob* | |||
| *sniff* | |||
| nothingmuch better rewrite it as a CPAN module | |||
| svnbot6 | r5265, Stevan++ | Perl6::MetaModel - Added docs to all files; added plans to all tests; added some expections for missing of bad arguments; some other general clean-up stuff too; | 05:29 | |
| gaal | morning! | 05:31 | |
| nothingmuch | nirbubg | 05:33 | |
| morning | |||
| nothingmuch is upset | |||
| magic mushrooms did not work as expected | 05:34 | ||
| gaal | were they tasty? | 05:37 | |
| nothingmuch | no, they tasted like mushroom soup | 05:38 | |
| gaal | what were they supposed to do? aren't shitake mushrooms just... edible like well mushroom soup? | ||
| nothingmuch | they were supposed to make me sleepy | ||
| gaal | i'm sure i've eaten them in restaurants with no this-will-make-you-sleepy warnings. | 05:39 | |
| nothingmuch | if you boil them in water | ||
| reduce 2 cups to 1, they are supposed to | |||
| at least so claims a friend | |||
| mugwump | sure, infusions and decoctions can have different effects | 05:41 | |
| gaal | yeah! 2 to 1 actually keeps you up for weeks! :-P | 05:43 | |
| nothingmuch | gaal: no, i'm sleepy now | ||
| just like i've been all week | |||
| gaal | eep, that sounds like the title of some spam | ||
| nothingmuch | i'm fighting it with good old deprevation today | 05:44 | |
| we'll see how that works | |||
| (usually it doesn't work too well for me). | |||
| it does sound like spam =) | |||
| gaal | it usually takes me about a week to get over this kind of jet-lag | ||
| and moving east is always harder for some reason. | 05:45 | ||
| nothingmuch | here too | ||
| i can stretch more easily | |||
| autrijus | gaal: so PCP is blocking on pugscc producing environment-ignorant code? | ||
| nothingmuch | staying up is easier than falling asleep | ||
| hola autrijus | |||
| autrijus | greetings yuval, gaal | 05:46 | |
| autrijus is finally feeling a wee bit alive | |||
| nothingmuch is not visually accustomed to seeing 'yuval' | |||
| gaal | autrijus (hi!!!), if you look at the actual code generated in PreludePC.hs, you'll see that the actual logic is missing | ||
| autrijus | right, I see that | ||
| gaal | like, function bodies | ||
| autrijus | yup. | ||
| gaal | i think otherwise it all works. | ||
| autrijus | 'kay. you++ | ||
| thanks for keep working on it while I'm semidead :) | 05:47 | ||
| svnbot6 | r5266, autrijus++ | * Finish the transition from @?CHECK and @?INIT to | ||
| r5266, autrijus++ | @*CHECK and @*INIT. | |||
| r5267, autrijus++ | * emulate this perl5 behaviour: | |||
| r5267, autrijus++ | perl -e 'print CGI->VERSION' -MCGI # works | |||
| r5267, autrijus++ | perl print_cgi_version.pl -MCGI # fails | |||
| r5268, autrijus++ | * fix a bogus test in return.t -- I seriously doubt the sanity | |||
| r5268, autrijus++ | of eval_is()... | |||
| r5269, autrijus++ | * two force_todo for later | |||
| gaal | LOL | ||
| autrijus | oh, hw2005 got rejected. :-/ I'll present Pugs at CUFP (commercial use of functional programming) track in the same ICFP conference | ||
| gaal | looks like you're alive now autrijus :) | ||
| nothingmuch | rejected? wtf? | ||
| gaal | ! :-( | ||
| autrijus | nothingmuch: my "presentation" is subpar it seems | ||
| i.e. too dense, not much background, prose sucks | |||
| nothingmuch | silliness | 05:48 | |
| autrijus | nah, the reviews is actually very helpful | ||
| nothingmuch | too "practical"? | ||
| autrijus | well, SyntaxNinja's cabal submission got two A and a B from reviewers | ||
| Amnesiac | hola autrijus | ||
| autrijus | but he's still rejected | ||
| so, extremely competitive it seems | |||
| nothingmuch | oh well | 05:49 | |
| better luck with CUFP. | |||
| autrijus | I got accepted at CUFP already | ||
| nothingmuch | huraah! | ||
| nothingmuch does the popularity dance | |||
| gaal | i like the "commercial" bit in association with Pugs. | ||
| autrijus | heh indeed | 05:50 | |
| nothingmuch | it'll probably provide lots of reassurance | ||
| autrijus | but yeah, mostly the reviewers think that the paper requires too much knowledge of the audience without enough preps into backgrounds, and hence not academic enough | 05:51 | |
| nothingmuch | don't people go to haskell confs to see what's been done with the language they know, instead of learn about the language? | ||
| gaal | oh, speaking of bits, can i have meta-? there were one or two people i could have added lately. | ||
| nothingmuch | gaal: i'll add | 05:52 | |
| (if I can) | |||
| gaal | that makes you a metametacommitter, no? | ||
| nothingmuch | well, since our metamodel is elegant enough, we don't need meta meta | ||
| autrijus | right. all metacommitter can add other metacommiters | ||
| so it's indeed elegant enough | 05:53 | ||
| nothingmuch | autrijus: any other candidates? | ||
| autrijus | nothingmuch: not at the moment | ||
| autrijus is so glad to be back to life | |||
| nothingmuch | no fair =( | 05:54 | |
| gaal: done | |||
| gaal | thank you! | ||
| nothingmuch | seen ingy | ||
| jabbot | nothingmuch: ingy was seen 13 hours 1 minutes 16 seconds ago | ||
| gaal | in a ploy to keep nothingmuch from normal life, /me announces Hackathon::TLV | 05:55 | |
| nothingmuch | lightstep, lumi - invited | ||
| gaal | cool. 21-23 July. details to follow. | 05:56 | |
| autrijus | TLV? | 05:57 | |
| gaal | Tel-Aviv | ||
| autrijus | ahh | ||
| nothingmuch | gaal: on the 21 - ~18:00? | 05:59 | |
| damn, spotlight rocks my tushy | |||
| gaal | i think that's reasonable, yes | 06:00 | |
| autrijus | mm, I wonder what will be Hacked at the Athon | 06:01 | |
| nothingmuch | pizza for one | 06:02 | |
| gaal | i have two things i want to advance, %*SIG and consolidated tests | ||
| nothingmuch | consolidated tests - nearly done | ||
| we need docs | |||
| gaal | %*SIG i keep not getting to finish :) | ||
| nothingmuch | nothingmuch.woobling.org/example.html | 06:03 | |
| i have a very concrete todo list if anyone is volunteering for TTH work | |||
| gaal | maybe port the test system to p6 then? :) | ||
| Aankhen`` | TTH? | ||
| nothingmuch | Test::TAP::HTMLMatrix | ||
| gaal | test-tap-htmlmatrix | ||
| Aankhen`` | Ah. | 06:04 | |
| What sort of work? | |||
| nothingmuch | Aankhen``: mixin Test::TAP::Model::Visual to Test::TAP::Model::Consolidated | 06:05 | |
| Aankhen`` | Oh, Perl work... nah. | ||
| nothingmuch | to avoid diamond inheritance crap, we need some sort of proper scheme | ||
| we need documentation of recent darcs repo changes | |||
| Aankhen`` | I thought maybe you were talking about HTML/JS/CSS stuff. | ||
| gaal | hmm, i need to get a whiteboard for the hackathon. | ||
| nothingmuch | oh no, that's finished =) | ||
| Aankhen`` | Congrats. :-D | ||
| nothingmuch | we need a "distinguishing string" sort of concept, for models | ||
| inherited to files | |||
| so that in the consolidated listing the blank space can contain relevant data | 06:06 | ||
| and this should be guessed | |||
| (platform, or other crap, from the extra field) | |||
| lastly we need the 'output_dir' method | |||
| which takes a directory | |||
| Aankhen`` | Ahh, I've got a lot of Perl stuff on my plate already. | ||
| nothingmuch | outputs 3 files - big summary, detail, independant file per model | ||
| err, N files | |||
| and CSS | 06:07 | ||
| trying to guess the link prefix to test files, if there is anything in the dir already | |||
| Aankhen``: take a look at example.html - maybe you have something to add to the JS | 06:09 | ||
| (btw, JS is mostly optional, even in the JS full version) | |||
| (there is also a completely js free version available) | 06:10 | ||
| Aankhen`` | Ah, good. | ||
| gaal | nm, so does the model handle aggregate data? or is there a model for a single smoke submission and a different model for consolidated data? | 06:11 | |
| Aankhen`` | You should add a <noscript><p>There is also a <a href="jsfree.html">script-free version available</a>.</p></noscript> | ||
| nothingmuch | Aankhen``: no, i don't know the user's URLs | 06:12 | |
| Aankhen`` | Hmm, that's not a good choice of words... | ||
| Ah. | |||
| gaal | another thing that's definitely worthy of hackage is finally a p6 getopt library | 06:23 | |
| more or less all of examples/ppt is blocking on that :-) | |||
| Aankhen`` | Well, iblech ported GetOpts::Std, but I believe he used Perl 6, not Pugs. :-P | ||
| gaal | here's a silly idea i had, tell me if i'm insane | 06:24 | |
| Aankhen`` | Want me to try hacking it to get it to work? | ||
| gaal | since p6 has pretty strong sub signatures, we could horribly exploit them and have the option spec look like a sub signature... | ||
| then eval it :) | |||
| modulo some munging. | 06:25 | ||
| Aankhen`` | O_O | ||
| Aankhen`` votes for "insane"> | |||
| gaal | i know this is evil :) but it has *some* hack value, no? | ||
| Aankhen`` | Yes. | ||
| gaal | of course not meant for production. | ||
| Aankhen`` | Suuure. :-P | ||
| nothingmuch | back | 06:26 | |
| Aankhen`` | How about we leave that for when Perl 6 is implemented fully? | ||
| BRB. | |||
| nothingmuch | eh, Aankhen``: you could add a 'noscript' link in the dir outputting method | ||
| gaal: i like it | 06:27 | ||
| gaal | my only problem is that when i think about opts, i keep saying "getopt long is all i need... except, well, more speccy control over validation.. and, well, automatic usage... and.. mutexes were a nice idea in getopt::declare.." etc. | ||
| Aankhen`` | Heh. | ||
| nothingmuch | sub flag is aliased 'f' (String $foo) { }; # invoked on --flag bar with "bar" | ||
| Aankhen`` | Mutexes? | ||
| nothingmuch | is that sort of what you meant? | ||
| gaal | getopt::std on ther other hand is a little too little for me | ||
| Aankhen``: i mean --pugs can't be used when you say --haskell, for example. | 06:28 | ||
| Aankhen`` | Ah. | ||
| gaal | in getopt::declare, you can actually *say* that in the spec | ||
| Aankhen`` | Hmm. | ||
| gaal | as for leaving this off for later, i think not! not having any wokring getopt is seriously putting of script hackage | 06:29 | |
| Aankhen`` | GetOptions(flag => sub (String $foo) { ... }) | ||
| How about that? | |||
| "Leaving it off" == Leaving off this particular idea, not GetOpts::*... | |||
| But that's moot anyway. :-P | |||
| gaal | ah :) | ||
| Aankhen`` | What about a hash of subs? | 06:30 | |
| gaal | dispatch table | ||
| Aankhen`` | Indeed. | ||
| gaal | for what, the spec? | ||
| or the implementation? | |||
| Aankhen`` | Spec. | ||
| GetOptions(flag => sub (String $foo) { ... }); | |||
| gaal | too much code? | ||
| Aankhen`` | People don't have to use it... | 06:31 | |
| autrijus | bbiab :) | ||
| Aankhen`` | See you autrijus. | ||
| gaal | nothingmuch, that's the idea, but the argument is not necessarily Str $foo | 06:32 | |
| Aankhen`` | They can use GetOpts::Std or ::Long if they feel it's too verbose. | ||
| Er, yeah, Str $foo, not String $foo. | |||
| Aankhen`` wonders why he never can get that right. | |||
| gaal | too much Java? | ||
| Aankhen`` | I've never done any. :-) | ||
| gaal | c++? :) | ||
| Aankhen`` | Nope. :-) | 06:33 | |
| gaal | CString *shudder* | ||
| Aankhen`` | JS, probably. | ||
| nothingmuch | i think it stuck because I tend to be verbose about my class names | ||
| and Junction is not Junc | |||
| Aankhen`` | Well, the dispatch table idea could go in GetOpts::Signature or something similar. | 06:34 | |
| nothingmuch | Aankhen``: any suggestions for TTH? | ||
| Aankhen`` | nothingmuch >> Not really, it looks fine to me. :-) | ||
| I love the way Indians keep inventing new words. | 06:36 | ||
| After all, who wants to be constrained by the boring "wireless"? "Wirefree" is where it's at! | |||
| nothingmuch | indians? | ||
| Aankhen`` | The non-American variety. :-) | ||
| nothingmuch | ah | 06:37 | |
| gaal | well, they'd rather have you care-free than careless. | ||
| Aankhen`` | LMAO. | ||
| nothingmuch | it's like autrijus's wireless thinkpad notebook | ||
| it's wireless in the same sense that this is: images.amazon.com/images/P/B0006MWR...ZZZZZ_.jpg | 06:38 | ||
| Aankhen`` | What the heck is that? | ||
| gaal | heh. | ||
| a 3.5kg book? :) | |||
| Aankhen`` | Ah | ||
| . | |||
| nothingmuch | obra was at some conf, and managed to get some promo material out of the IBM booth | 06:39 | |
| they gave away paper notebooks with 'ThinkPad' written on them, and they have no wire spiral | |||
| he gave one to autrijus | |||
| hence autrijus has a 'thinkpad wireless notebook' | |||
| Aankhen`` | LMAO. | ||
| s/^/ROTF/ | |||
| gaal | i never got the wireless pc-card nic i got with my laptop to work. i never actually took part in the wireless revolution! | 06:40 | |
| nothingmuch | wifi rocks so much | ||
| Aankhen`` | Yep. | ||
| nothingmuch | when you have places to go that provide it | ||
| gaal | also, the laptop does in fact weigh 3.5 kg so i don't like lugging it about anyway | ||
| Aankhen`` | (rocks for me since my sister leaves the desktop alone now :-P) | 06:41 | |
| nothingmuch | i also like working int he back yard | ||
| gaal | back yards++ | ||
| nothingmuch | it works better with the new laptop, since it appears to have a better antenna | ||
| reception is consistently better in every way | |||
| ooh, darwinports finished making me an irssi | 06:42 | ||
| switching to a proper version | |||
| Aankhen`` | stevan++ # Perl6::MetaModel | 06:44 | |
| cwest | jsan has mailing lists openjsan.org/community/lists.html | 06:48 | |
| That is all. | 06:49 | ||
| cwest & # sleep | |||
| Aankhen`` | Eek. | 06:50 | |
| I just messed up two subscriptions. :-\ | |||
| WB. | 07:00 | ||
| Uh oh. | |||
| That doesn't bode well. :-) | |||
| nothingmuch | =( | 07:01 | |
| irssi is misbehaving | |||
| Aankhen`` | BRB. | ||
| On second thoughts, I'mma go waste some time. | 07:03 | ||
| UT2K4. & | |||
|
07:39
davidra is now known as jp-autark
|
|||
| nothingmuch needs a new quote to paste on his monitor | 08:17 | ||
|
09:05
scook0_ is now known as scook0
|
|||
| Aragone | hrm, some fucker's trying to blow up the tube :/ | 09:28 | |
|
09:29
Aragone is now known as Arathorn
|
|||
| gaal | london tube? | 09:29 | |
| clkao | yes | ||
| osfameron | best take it to MAGnet/#london.pm prob'ly | 09:30 | |
| Arathorn shifts appropriately | 09:31 | ||
| svnbot6 | r5270, autrijus++ | * `[1] <=> [2]` was misparsed as `[1]{'='}[2]`. Fixed. | 10:06 | |
| r5271, autrijus++ | * repair the (now mostly passing) namespaces.t | |||
| nothingmuch | wikipedia is so effective | 10:14 | |
| en.wikipedia.org/wiki/2005_London_t...explosions | |||
| autrijus | cwest++ # adopting the chronicle format | 10:20 | |
| nothingmuch | chronicle? | 10:22 | |
| autrijus? | 10:25 | ||
| autrijus | nothingmuch: use.perl.org/~cwest/journal/ | ||
| nothingmuch | oh, for jsan logging | 10:26 | |
| autrijus | yeah | ||
| nothingmuch falls asleep waiting for qt guy to help him out | 10:27 | ||
| QtPlatypus blinks. | 10:28 | ||
| autrijus | hm, still no kwid2pod in sight? | 10:30 | |
| ingy: how's life in perldoc land? | |||
| cafe_kioto | hi all | 10:44 | |
| i'm a newbie | |||
| can i get any info about perl6 | |||
| QtPlatypus | Sure cafe_kioto what do you wish to know. | 10:45 | |
| cafe_kioto | Parrot | 10:48 | |
| QtPlatypus | What about parrot do you wish to know? | 10:49 | |
| cafe_kioto | the difference with a simple parser and the advantage | ||
| autrijus | cafe_kioto: www.parrotcode.org/ has some links. irc.perl.org #parrot is the channel. feather.perl6.nl/~chip/Chip_APW.pdf has some overview; leo's journal is use.perl.org/~leo/journal | ||
| cafe_kioto | of vm | ||
| ths | 10:50 | ||
| and when it's possible to use official perl6 | |||
| autrijus | I have no idea :) | 10:55 | |
| I wonder if anyone does. | |||
| I've covered some of these topics in pugscode.org/talks/yapc/slide1.html | 10:56 | ||
| but I need to run for dinner now. bbiab :) | |||
| castaway | Thursday. cafe_kioto | 11:13 | |
| cafe_kioto | castaway humour ? | 11:16 | |
| masak | i read that as "now" | 11:17 | |
| castaway | I didnt say which Thursday | 11:21 | |
| yes, cafe_kioto | |||
| cafe_kioto | thanks :) | 11:23 | |
| nothingmuch | autrijus: thanks for posting hackathon notes | 11:41 | |
|
11:44
Aankh|Clone is now known as Aankhen``
|
|||
| Aankhen`` | ?eval module Foo { module Bar { my $baz = "quux"; } }; $Foo::Bar::baz; | 11:49 | |
| evalbot6 | \undef | ||
| Aankhen`` | o_O | ||
| nothingmuch | oh shit | 11:50 | |
| mmmd blah blah blah | |||
| "loop that accepts the first alternative." | |||
| that is not very good | |||
| are there any hooks to reorder the MMD list? | |||
| and maybe a way to implement type distance checking by using a catchall in the first slot? | 11:51 | ||
| or at least apply type distance dispatch using the metamodel? | |||
| Aankhen`` | Where is that? | 11:52 | |
| nothingmuch | autrijus's hackathonn summary | ||
| cafe_kioto | hei people can you think about python | 11:53 | |
| nothingmuch | cafe_kioto: . o O ( big snake ) | ||
| done! | |||
| cafe_kioto | yes done | ||
| nothingmuch | seriously though - what did you mean? | 11:54 | |
| cafe_kioto | ok | ||
| if i want learn one of the two language between perl and python | 11:55 | ||
| nothingmuch | python - strict, strictifying, growing smaller | ||
| functional paradgims are bad | 11:56 | ||
| perl - loose | |||
| very loose | |||
| sometimes dangerous | |||
| depending on your experience and self discipline you might get more out of one than the other | |||
| the definate winner for perl is CPAN | |||
| python just doesn't have that | |||
| there are poor imitations that adamk will be happy to tell you about | 11:57 | ||
| cafe_kioto | i'm php developer | ||
| php when work without apache it's no good choice | |||
| nothingmuch | php is basically perl | 11:58 | |
| but you take away: | |||
| cpan | |||
| good paradgims | |||
| flexibility | |||
| mugwump | closures | ||
| nothingmuch | polymorphism of librariries | ||
| mugwump | use strict | ||
| nothingmuch | (there is a different function name for every SQL "do" per DB... how yucky is that?) | ||
| you are psychologically bound to a certain structure - template oriented | 11:59 | ||
| did I mention no CPAN? | |||
| cafe_kioto | pear Db | ||
| sorry | |||
| nothingmuch | i remember someone mentioning something like 9 sort functions | ||
| where perl gets by with one, which accepts an optional closure | 12:00 | ||
| i think that says a lot | |||
| cafe_kioto | i'm interesting to create | ||
| windowed application | 12:01 | ||
| nothingmuch | tnx.nl/php | ||
| cafe_kioto: in perl you have Qt, Tk, Gtk | |||
| WxWindows? | |||
| kungfuftr | xul | ||
| nothingmuch | in python you should have the same offerrings | ||
| Win32::GUI | |||
| CamelBones | |||
| kungfuftr | X11::Protocol | ||
| cafe_kioto | i think that the language get new live with | 12:03 | |
| parrot engine | |||
| nothingmuch | that's not ready yet | ||
| don't count on it to make your life easy in the next year or two | |||
| cafe_kioto | but you programming with perl for your customer | 12:04 | |
| in my country more language like perl, php, python is not knowed | 12:05 | ||
| QtPlatypus | cafe_kioto: Where do you live? | 12:06 | |
| cafe_kioto | italy | ||
| larsen | me too | 12:11 | |
| cafe_kioto | very nice | ||
| larsen | cafe_kioto: are you italian ? | 12:15 | |
| cafe_kioto | yes i'm | ||
| larsen | if you want, there are italian perl mongers in #nordest.pm, irc.slashnet.org | 12:16 | |
| cafe_kioto | thanks larsen | 12:18 | |
| for the advice | |||
| but the official site of perl project is www.perl.com on o'reilly site | 12:21 | ||
| sorry perl.org | 12:24 | ||
| Aankhen`` | Anyone got enough knowhow to implemented File::seek? | ||
| gaal | i can look into it | 12:26 | |
| we can only offer basic file io for now though, no posix semantics. is that enough for what you need? | 12:27 | ||
| btw: new smoke coming up at perlcabal.org/~gaal/ | 12:28 | ||
| Limbic_Region | gaal - it looks like autrijus is back in the land of the living so some of the pugsbugs I have been waiting for might be fixed | 12:30 | |
| Limbic_Region was just checking the recent commits and see some things for the spaceship parsing bugs :-D | 12:31 | ||
| gaal | :) | ||
| i have to go now for ~2 hrs | 12:39 | ||
| i'm nopasting the prim stuff, feel free to complete it | 12:40 | ||
| (it's pretty trivial, just add an import of hSeek and write up the prelude side) | |||
| pasteling | "gaal" at 192.115.25.249 pasted "seek WIP" (21 lines, 635B) at sial.org/pbot/11597 | 12:42 | |
| gaal | hack away or wait two hours :) | ||
| gaal & | 12:43 | ||
| Aankhen`` | Oops, sorry, I got distracted and didn't reply... | ||
| Aankhen`` thinks to himself, "POSIX semantics? Huh?" | |||
| I just need to be able to open a file, seek to a particular position, write to it, close it, et al. | 12:45 | ||
| cafe_kioto | thanks to all | ||
| bye | |||
| wolverian | I have to get x working again, console is terrible :) | 12:52 | |
| Nattfodd | tss tss | 13:02 | |
| as long as you don't want to watch a movie, console is great | |||
| wolverian | I can't get utf8 to work quite right - elinks still displays some things wrong. and the font I want to use is too large. | 13:03 | |
| also, I don't really want to mess with fb :) | 13:04 | ||
| Aankhen`` | seen cwest | 13:31 | |
| jabbot | Aankhen``: cwest was seen 6 hours 42 minutes 21 seconds ago | ||
| autrijus | gaal: cool work | 13:35 | |
| however I'm too sleepy to complete it now | |||
| but journal up (for a change) | |||
| autrijus waves and sinks back to blissful sleep... & | 13:36 | ||
| wolverian | cheers :) | 13:37 | |
| autrijus | :) | ||
| autrijus ponders what next to do, now the 6-month sabbatical is less than a month away. | |||
| autrijus decides to sleep on it | |||
| masak | sleep well | ||
| autrijus | s/away/left/ | 13:38 | |
| svnbot6 | r5272, autrijus++ | * correct a test typo: `is (...)` should be `is(...)` | 13:52 | |
| r5273, autrijus++ | * Autovivify variables and globals even as rvalues. | |||
| r5274, autrijus++ | * unTODO, cleanup, etc collect tests. | |||
| r5275, autrijus++ | * todo+fail erratic destruction tests for release. | 14:05 | ||
| Aankhen`` | Hiya autrijus. :-) | 14:07 | |
| nothingmuch | hola. | 14:12 | |
| hola | |||
| baah | |||
| hola theorbtwo | 14:14 | ||
| theorbtwo | Hola, Yuval. | 14:15 | |
| nothingmuch | "Yuval" doesn't pop up a floater =) | 14:16 | |
| use nothingmuch, it's more reliable | |||
| the moment I get a working irssi i'll have growl set up properly | |||
| theorbtwo | Heh. | 14:17 | |
| In that case, hola, nothingmuch. | |||
| nothingmuch | hola! | ||
| irssi is lyingf | 14:18 | ||
| it says "connecting" | |||
| but it makes no such syscall | |||
| nothingm1ch | hmm... this is a proper irssi, but not on the box I want it to run on | 14:22 | |
| eric256 | i gotta ask, what is an irssi ? | 14:23 | |
| nothingmuch | irc client | ||
| the problem is that I'm not able to get a proper working build. | |||
| either it's too old, or badly packaged, or doesn't actually connect, or it's on a machine I don't want to use it on | |||
| wolverian | nothingmuch: get it from svn and build yourself? | ||
| nothingmuch | wolverian: it's got many ugly deps | 14:24 | |
| notably glib | |||
| wolverian | it just worked here :) | ||
| nothingmuch | which I think is the actual problem with it not working on my darwin | ||
| wolverian | oh, darwin | ||
| nothingmuch | i ktraced through it | ||
| it forks a child, the child resolves the host | |||
| prints back on the pipe, and exits | |||
| the parent just sits there | |||
| doing nothing | |||
| no wait(), no sigchld handler, no pipe read, no connect | 14:25 | ||
| wolverian | weird. | ||
| nothingmuch | since supposedly glib is supposed to handle the lower level mishmash, i suspect it's at fault | ||
|
14:26
nothingm1ch is now known as nothingmuch_past,
nothingmuch_past is now known as nothingmuchpasta
14:27
nothingmuchpasta is now known as nothingmuch_,
nothingmuch_ is now known as nothingmuchpasta
|
|||
| nothingmuch | there, new nicks registered | 14:28 | |
| stevan | morning nothingmuch | 14:32 | |
| gaal | rehi | 14:35 | |
| nothinmuch is a kind of pasta fresca. | 14:36 | ||
| nothingmuch | gaal: pasta is my home box hostname | 14:37 | |
|
14:37
nothingmuchpasta is now known as nothing_pasta
|
|||
| stevan | hey gaal | 14:37 | |
| nothingmuch | i think I prefer this | ||
| gaal just came back from late lunch at an Italian with friends | |||
| s/s$// | |||
| heya stevan! | 14:38 | ||
| gaal resumes hSEek | |||
| nothingmuch | hola stevan | ||
| i finally got my algorithms book | |||
| stevan | nothingmuch: nice :) | ||
| nothingmuch | and my m. ward CD | ||
| stevan | what was that ... 24-36 weeks for delivery | ||
| gaal | what's m.ward? | 14:40 | |
| nothingmuch | stevan: approx | 14:41 | |
| gaal: i'll rip | |||
|
14:43
nothing_pasta is now known as addnet
14:44
addnet is now known as help,
help is now known as delnet
14:45
delnet is now known as listnick
|
|||
| nothingmuch | oops, dumb renames | 14:46 | |
| nothingmuch suddenly realizes his mother will be in london tomorrow | 14:47 | ||
| good thing twas not today | |||
| hah, fixed it | 15:26 | ||
| finally | |||
| gaal | hmm, i wonder why the seek is failing. | 15:47 | |
| how do i make System.IO errors raise exceptions? | 15:48 | ||
| Limbic_Region | salutations fglock | 16:04 | |
| how goes the work on the DateTime project? | |||
| fglock | hi - I can't login through svn. | 16:06 | |
| I registered in rt.openfoundry.org - is this the right place? | 16:08 | ||
| hcchien | fglock: yes | 16:12 | |
| ingy | hi hcchien | 16:13 | |
| hcchien | ingy: hola | 16:14 | |
| Limbic_Region | fglock - define can't login? | 16:17 | |
| sorry was afk earlier | |||
| I had an issue of not staying logged in when I was trying to give you the commit bit | |||
| the problem happened to be with my firewall/FF configuration | |||
| fglock | Limbic_Region: it keeps asking me for the pass. I'm trying again with https. | 16:18 | |
| Limbic_Region | but you are able to commit via svn or svk? | 16:22 | |
| svnbot6 | r5276, iblech++ | * t/builtins/lists/reduce.t -- Fixed "is(...), ..." to "is((...), ...)". | ||
| r5276, iblech++ | * Usual svn props added to new files in lib/Perl6-MetaModel/. | |||
| fglock | Limbic_Region: no, I can't commit yet. I'll try again. | 16:23 | |
| Limbic_Region | let me login and see what it says your status is | ||
| flgock - dunno but your username may be case sensitive | 16:25 | ||
| gaal | i'm logged in now, l~r. what's your username, fglock? | ||
| Limbic_Region | and by default, svn/svk will attempt to use you username from the environment variable | ||
| fglock | "fglock" | ||
| gaal | i can't see "fglock" in the member list. | ||
| Limbic_Region | you will need to use a command line option to change that | ||
| if it isn't the same that is | |||
| gaal | can you /msg me your email, i'll re-add you? l~r, i hope that's okay since you're having cookie problems :) | 16:26 | |
| Limbic_Region | gaal - cookie problems are from home | ||
| I am not having any problems here | |||
| I am looking now myself | |||
| gaal | oh. okay then, go ahead :) | 16:27 | |
| Limbic_Region | fglock - will use same email as before | ||
| oh - I think it is my fault | |||
| it should work now btw | 16:28 | ||
| sending the invite wasn't enough - had to mark you as "member" in step 2 | |||
| that has been done now | |||
| I see flgock as a member now | |||
| fglock | ok - I got the email this time - thanks! | ||
| Limbic_Region | Flavio S. Glock | ||
| Limbic_Region should have paid more attention but he was fighting with those pesky cookies before | 16:29 | ||
| fglock | it worked! | 16:31 | |
| Limbic_Region | well, welcome aboard (again) | 16:32 | |
| svnbot6 | r5277, fglock++ | Set::Infinite initial commit - just a skeleton | 16:33 | |
| fglock | datetime-set is almost empty - should i upload anyway? | 16:35 | |
| Limbic_Region | fglock - how else is anyone else going to contribute | 16:43 | |
| Aankhen`` pokes gaal. | 16:51 | ||
| svnbot6 | r5278, fglock++ | fixed the object variables | ||
| r5279, fglock++ | initial commit - just a skeleton | |||
| r5280, fglock++ | README fix | |||
| gaal | hey :) | ||
| Aankhen`` | Hiya. :-) | ||
| Any progress on `seek`? | |||
| gaal | i have the code, but it fails. | ||
| shall i commit it? | |||
| i haven't figured out how to trace it in the haskell level | |||
| Aankhen`` | Hmm. | ||
| No, it's okay. I can wait. | |||
| gaal | why wold a seek fail, anyway? | 16:52 | |
| on a file. | |||
| Aankhen`` | The file didn't open? | ||
| gaal | it did | ||
| Aankhen`` | Hmm. | ||
| You don't have read permissions? :-P | |||
| wolverian | why not commit it and let everyone take a look? | ||
| Aankhen`` | Maybe you're seeking too far ahead. | ||
| (i.e. the file is too small for the quantity) | 16:53 | ||
| gaal | of course i do :) i did my $f = File.open(<AUTHORS>); $f.seek(50, 0); say =$f | ||
| Aankhen`` | Ah | ||
| . | |||
| gaal | nope on all of the above :) | ||
| committing. | |||
| r5281 | 16:54 | ||
| have at it :) | |||
| svnbot6 | r5281, gaal++ | seek builtin. failsforme, silently. | 16:59 | |
| wolverian | wow, saying <AUTHORS> as a way to quote it is so confusing to a perl5 coder :) | 17:00 | |
| Aankhen`` | gaal++ # for working on it | ||
| ?eval (<foo>, <bar baz>) | |||
| evalbot6 | ('foo', 'bar', 'baz') | ||
| Aankhen`` | ?eval <foo> | ||
| wolverian | it looks like a filehandle read | ||
| gaal | heh :) i do it because i switch between shells a lot | ||
| evalbot6 | 'foo' | ||
| wolverian | gaal: heh. | ||
| Aankhen`` | ?eval <foo bar> | 17:01 | |
| evalbot6 | ('foo', 'bar') | ||
| gaal | including the windows "shell" (my ass) which has horrible quoting problems | ||
| wolverian | gaal: nah, it quotes fine, as long as you remember ' is not a quote ;) | ||
| Aankhen`` | Aye aye! | ||
| wolverian >> And *shudder* double up your double quotes to escape them? | |||
| gaal | oh yeah? why does find *require* quotes then? | 17:02 | |
| wolverian | gaal: hm? | ||
| gaal | dir | find moose # doesn't work | ||
| dir | find "moose" # does | |||
| wolverian | gaal: haha, really? that's pathetic | ||
| I thought the shell at least did the quoting | |||
| when it comes to " | |||
| are you saying it just passes the quotes in to the programs? | |||
| gaal | sometimes :) | 17:03 | |
| fglock | I can say Bool.true - but Bool.false does not work? | ||
| gaal | i don't know how it works really. w32 is all voodoo | ||
| Aankhen`` | fglock >> What about Bool::true and Bool::false? | ||
| Er. | |||
| gaal | say bool::true and bool::flase | ||
| Aankhen`` | `bool::true` and `bool::false`. | 17:04 | |
| gaal | lowercase | ||
| Aankhen`` | ?eval 1 == 1 | ||
| evalbot6 | bool::true | ||
| Aankhen`` | ?eval 1 == 0 | ||
| evalbot6 | bool::false | ||
| wolverian | :: is so ugly. :/ | ||
| fglock | thanks | ||
| wolverian | I wish we used # or something to delimit packages | ||
| Aankhen`` | wolverian >> Hey! I didn't raise you to talk bad about punctuation! | ||
| wolverian | (or just . and put them into the same namespace... although that's severely limiting) | 17:05 | |
| gaal | by the way! i presume it's a known problem that this doesn't worl? | ||
| work? | |||
| ?eval class Foo; our $class_var = 8; | |||
| evalbot6 | \8 | ||
| gaal | huh!? | ||
| it didn't workforme :) | |||
| Aankhen`` | I like ::, personally. Atleast, when it's not overused. "Bool::false" is so-so, "bool::False" is strange, but "Bool::False" looks okay. :-) | ||
| gaal | ?eval class Foo { our $x = 8; } | 17:06 | |
| evalbot6 | Error: Undeclared variable: "$x" | ||
| gaal | heh | ||
| clkao | w/in21 | ||
| wolverian | no, you won't! you're staying here! | ||
| gaal | scoped our doesn't work it seems. known problem? | ||
| wolverian | gaal: well, scoped our isn't even completely specced yet | ||
| gaal: in roles anyway | |||
| gaal | hmmm | ||
| eric256 | why would yo uuse our anyway? | ||
| gaal | public constant | 17:07 | |
| Aankhen`` | eric256 >> Class variables. | ||
| gaal | and class variables too :) | ||
| Aankhen`` | class Foo { our $.bar = 10; }; say $Foo::bar; | ||
| `our $bar`, even. | |||
| gaal | File::SEEK_START | ||
| Aankhen`` | class Foo { our $bar = 10; }; say $Foo::bar; | ||
| gaal | $File::SEEK_START that is | ||
| wolverian | gaal: what's SEEK_START? | ||
| eric256 | ahh as opposed to instance variables. makes a certain kind of since (a bunch of little variables all say "hey those are 'our' variables." | ||
| Aankhen`` | Heh. | 17:08 | |
| I like the way you think. =) | |||
| eric256 | s/variables/objects/ | ||
| wolverian | I mean, it looks like a pretty self-explanatory name, but I don't understand what it would mean as a class variable | ||
| gaal | wolverian, $fh.seek(50, SEEK_START) | ||
| $fh.seek(-1024, SEEK_END) | 17:09 | ||
| well, $File::SEEK_END | |||
| wolverian | gaal: hmm. that doesn't seem like a very good interface | ||
| we could generalise the issue with a builtin Delta class | 17:11 | ||
| when seek() is given a delta, it uses the current position, otherwise absolute. hmm, there'd need to be some way to get at the end of the file too. | |||
| oh well. :from<end>? | |||
| gaal | that's an oop version of the p5 interface. | ||
| i'll also add tell. | |||
| wolverian | gaal: yeah, and it sucks, I think | 17:12 | |
| (no offense - you're probably just porting it, and that's fine) | |||
| gaal | i'd be happy to add more interfaces, but you get to petition them on p6-l :-) | ||
| wolverian | right, sure. | ||
| I'll mail about it on p6l if you want? | |||
| or do you want to? :) | |||
| gaal | please do. | ||
| wolverian | sure. | ||
| gaal | i don't feel strongly about this, and where i don't feel strongly and there is no spec yet i tend to just port. | 17:13 | |
| wolverian | hmm | 17:14 | |
| where do SEEK_END and such come from in perl5? | |||
| POSIX? | |||
| Fcntl? | |||
| my perldoc is so slow :) | |||
| gaal | Fcntl. | 17:15 | |
| wolverian | thanks | ||
| gaal: sent mail | 17:18 | ||
| gaal | wolverain++ | 17:19 | |
| Aankhen`` | wolverian++ | 17:20 | |
| (nick completion)++ | |||
| wolverian | wolverain-- # it's raining wolverines | ||
| gaal | that said, i think :from<end> is nice. $fh.end - 10 won't work well i think. | ||
| wolverian | performance? | 17:21 | |
| gaal | there are three modes to seek, traditionally. absolute, relative to current, and offset at end. i expect them all to be one syscall. | 17:22 | |
| (not sure though.) | |||
| wolverian | right. I have a hazy concept of such things | ||
| translate that to real english if you want | 17:23 | ||
| perl5 is surprisingly high level. :) | |||
| gaal | are you on linux? | ||
| wolverian | not right now, since I broke my X (I'm running ubuntu breezy, the development branch) so I'm booted to windows currently | 17:24 | |
| my server is running linux, though | |||
| Aankhen`` | Hiya iblech. :-) | ||
| wolverian | (the one that irssi runs on :) | ||
| gaal | as an illustration look at the manpage for fseek: eg www.die.net/doc/linux/man/man3/fseek.3.html | 17:25 | |
| masak | Aankhen``++ # for giving me the idea to try nick completion, this is great :) | ||
| gaal | you'll see that perl didn't invent any of these names | ||
| Aankhen`` pokes cwest for signs of life. | |||
| gaal | even "whence" is from the c interface | ||
| wolverian | right. | ||
| Aankhen`` | masak >> I'm glad /someone/ learned something. ;-) | 17:26 | |
| masak | ;) | ||
| wolverian | the point being that C is much lower level than perl - so as long as we can keep the performance up we should fix the interface | ||
| because we _can_ | |||
| gaal | since this is so deep in the tradition, i bet underliying os implementations have a single system call that does any of the three seeks | 17:27 | |
| iblech | Aankhen``: Hi :) Unfortunately I'll be off again soon (school trip to Paris, coming back Sunday night)... | ||
| gaal | of course we can fix the interface; but doing this: | ||
| Aankhen`` | iblech >> Awww. :-( | ||
| wolverian | right. then a :from<end start current> mapping to those is probably the best idea | ||
| gaal | $f.seek($fh.end - 50) | ||
| iblech | Aankhen``: Yes... :( | ||
| Aankhen`` | iblech >> How much longer do you have in school/ | 17:28 | |
| ? | |||
| wolverian | (although <end start current> aren't very well connected as properties of the file, which is why $fh.end would be better as an abstraction and on the human language level) | ||
| gaal | will actually work with "my" interface (i made ?$position = start) | ||
| but is needlessly inefficient. | |||
| wolverian | right. | ||
| I just hate needless globals | |||
| gaal++ # thanks for coding things! | 17:29 | ||
| gaal | i'm all for abstraction and good interface, but if the system wants to help you work cleanly, why not let it? :) | ||
| iblech | Aankhen``: In approx. three weeks, we'll have six weeks vacation :) and then two more years | ||
| Aankhen`` | iblech >> Gak! Two years? | ||
| How many years of schooling do you have in total? | |||
| wolverian | gaal: sure - I'm fine with :from<foo>, which is equivalent to the SEEK_* globals, except with a nicer interface | ||
| gaal: both should map to the same syscalls | |||
| svnbot6 | r5282, fglock++ | added a few tests | ||
| wolverian | s,should,can, | 17:30 | |
| gaal | yup. i'm happy with that - but i'll let others spec it :-) | ||
| wolverian | yeah. :) | ||
| there might be better ideas, or larry might be harbouring some insane idea again | |||
| not that those are mutually exclusive... | 17:31 | ||
| gaal | also, adding tests is liable to get it coded your way (hint, hint :-)) | ||
| wolverian | heh. but tests are spec and I'm not larry! | ||
| gaal | i love larry's insane idea except for when i'm 80% done implementing something the way i thought it shold be. :-) | ||
| *ideas | |||
| wolverian | I think autrijus feels your pain too :) | ||
| Aankhen`` | LMAO. | ||
| iblech | Aankhen``: 13, but with the beginning of the next year, new pupils will only have to go 12 years to school | 17:32 | |
| Aankhen`` | Ah. | ||
| iblech | Aankhen``: Germany, you know... | 17:33 | |
| Aankhen`` | Heh. | ||
| Very thorough. | |||
| iblech | Indeed | ||
| Aankhen`` | How are recursive patterns specified? | 17:36 | |
| For example, a pattern to match an unescaped backslash, i.e. a backslash not preceded by a backslash not preceded by a backslash... | 17:37 | ||
| wolverian | egh, 21mb of sid updates | ||
| things are slow on a computer with a /tmp of 35 mb and 48 mb RAM :) | 17:38 | ||
| Aankhen`` | wolverian >> BTW, you were asking about whether modules are global by default, right? | 17:40 | |
| wolverian | Aankhen``: yeah, I think. | ||
| Aankhen`` | @Larry: > They are "our" by default, not "my". Just as subs are "our" by | ||
| > default. And indeed, that is probably what package/module/class | |||
| > names are going to default to as well. | |||
| wolverian | ah, right. and what does "our" do exactly in Perl 6? | 17:41 | |
| Aankhen`` | Um... package/class/module-level var? | ||
| svnbot6 | r5283, iblech++ | Usual svn props added to fglock++'s new ext/DateTime-Set and ext/Set-Infinite, | ||
| r5283, iblech++ | include the appropriate svn:ignore to hide autogenerated files like Makefile, etc. | |||
| Aankhen`` | class Foo { our $bar = 10; } say $Foo::bar; # prints "10\n" | ||
| module Foo { our $bar = 10; } say $Foo::bar; # prints "10\n" | 17:42 | ||
| wolverian | right. thanks. | ||
| Aankhen`` | NP. :-) | 17:43 | |
| wolverian | module Foo { module Bar { } } Foo::Bar; # the way to look up Bar now? | ||
| Aankhen`` | Yes. | ||
| wolverian | thanks. | ||
| hmm | |||
| Aankhen`` | redo; | ||
| wolverian | I can't find seek() in ruby | ||
| oh, here we are | |||
| IO instead of File :) | |||
| ah. it uses IO::SEEK_* globals | 17:44 | ||
| (well. package globals..) | |||
| Aankhen`` | Heh. | 17:45 | |
| BRB. | |||
| wolverian | I'm not sure if one needs seek() at all really | 17:47 | |
| maybe perl6 can map $fh[$start .. $end] intelligently enough (and lazily enough) to be efficient and easy to ue :) | |||
| gaal | you want perl to be an OS? :) | 17:48 | |
| wolverian | hmm. no. I want seek() to be available as well, and the [] to use that to seek to $start automatically | 17:49 | |
| e.g. for =$fh[-10...] -> $line { ... } # iterate efficiently over the last ten lines | 17:50 | ||
| gaal | <3 Tie::File | 17:51 | |
| wolverian | right. I forgot Perl 5 does it already :) | ||
| gaal | but that's a nifty module to port! | 17:52 | |
| wolverian | we need the iterators/generators thing first, I think, for it to be usable as above | ||
| svnbot6 | r5284, iblech++ | * Continued chromatic++'s work of s/Perl6/Perl 6/ in lib/, ext/, and docs/. | 17:56 | |
| r5284, iblech++ | * Fixed alignment in READTHEM. | |||
| Aankhen`` | ?eval class Foo { method bar (::?CLASS $self: ) { .ref } }; Foo.new.bar | 18:03 | |
| evalbot6 | Error: unexpected ":" expecting formal parameter, end of input or ")" | ||
| svnbot6 | r5285, iblech++ | Test fixes: | ||
| r5285, iblech++ | * EOLs at EOFs (also in ext/). | |||
| r5285, iblech++ | * "#!/usr/bin/pugs" added to test missing that line. | |||
| r5285, iblech++ | * eval_is(...) -> is(eval(...)). | |||
| Aankhen`` pokes iblech. | |||
| Shouldn't ::?CLASS and ::?ROLE parse? | 18:04 | ||
| iblech | Aankhen``: It should, doesn't it? | ||
| ?eval ::?CLASS | |||
| evalbot6 | \::main | ||
| iblech | ?eval ::?ROLE | ||
| evalbot6 | \::main | ||
| Aankhen`` | Hmm. | ||
| Doesn't appear to parse in signatures. | 18:05 | ||
| ?eval class Foo; method bar (::?CLASS $self: ) { "bar" }; | |||
| evalbot6 | Error: unexpected ":" expecting formal parameter, end of input or ")" | ||
| iblech | Hm... | ||
| wolverian | hmm. does the ::?CLASS there DWYM when the class 'trusts Bar' and you call Bar.new.bar? | 18:06 | |
| Aankhen`` | ::?CLASS is the class the method is declared in, AFAIK. | ||
| wolverian | then that method is unusable from a trusted class | 18:07 | |
| Aankhen`` | `trusts` only affects access to instance variables | ||
| . | |||
| Yes. | |||
| wolverian | hmm, I read autrijus's message differently | ||
| Aankhen`` | Well, I've been known to be wrong. ;-) | ||
| wolverian | upon rereading it seems to only pertain to accessors, yes | ||
| but who's saying bar isn't an accessor? :) | 18:08 | ||
| Aankhen`` | I dunno, but I /am/ saying it's not an instance variable. :-P | ||
| wolverian | okay, time to install breezy colony 2 over the broken breezy partition and see if it works :) | 18:10 | |
| brb | |||
| Aankhen`` | ?eval gather { for (a => "b", c => "d").kv -> $k, $v is copy { $v .= uc; take $v } } | 18:12 | |
| evalbot6 | Error: Insufficient number of required parameters: 0 actual, 1 expected | ||
| Aankhen`` | ?eval gather { for (a => "b", c => "d").kv -> $k, $v is copy { $v = $v.uc; take $v } } | ||
| evalbot6 | Error: Insufficient number of required parameters: 0 actual, 1 expected | ||
| Aankhen`` | ?eval gather { my %foo = (a => "b", c => "d"); for %foo.kv -> $k, $v is copy { $v = $v.uc; take $v } } | ||
| evalbot6 | ('B', 'D') | ||
| Aankhen`` | ?eval sub foo () { "I know kung foo" }; &::("foo").() | 18:14 | |
| svnbot6 | r5286, fglock++ | more tests | ||
| r5287, fglock++ | oops - missed some files | |||
| evalbot6 | 'I know kung foo' | ||
| Aankhen`` | ?eval sub foo () { "I know kung foo" }; my $bar = "o"; &::("f" ~ ($bar x 2)).() | 18:15 | |
| evalbot6 | 'I know kung foo' | ||
|
18:16
Arathorn is now known as Aragone
|
|||
| iblech | Aankhen``: Fixed the ::?FOO parsing bug, checking in in a moment :) | 18:16 | |
| Aankhen`` | Yey! | 18:17 | |
| iblech++ | |||
| Woo. | 18:20 | ||
| fglock | I've got a problem in ext/Set-Infinite/t/set-functional-span.t - "use_ok" doesn't actually "use" the module? | ||
| I have to "use" it again | |||
| Aankhen`` | I'm finding some strange parsing problems. :-D | ||
| iblech | fglock: It does. But probably, the subs to be exported are exported into Test.pm's namespace, not into the namespace of your test file | 18:21 | |
| Aankhen`` | No, it only `require`s the module. | 18:22 | |
| fglock | it doesn't recognize the class name - I get 'No compatible subroutine found: "&Set::Functional::Span"' | ||
| Aankhen`` | eval "package $caller; require $module"; | ||
| fglock >> Is the test ci'd? | |||
| iblech | Aankhen``: Oh, I retract my comment then :) | 18:23 | |
| fglock | yes - it works (with the extra "use") | ||
| Aankhen`` | OK, I'm going to fiddle around a bit locally. | ||
| gaal | if I add a haskellish Prim with a fully qualified name, woudl it be dispatched correctly on objects? | 18:25 | |
| svnbot6 | r5288, iblech++ | Pugs.Lexer, Pugs.Parser -- Made ::?CLASS in subroutine signatures parse again, | ||
| r5288, iblech++ | thanks to Aankhen++ for noticing. | |||
| r5289, iblech++ | * Usual svn props added to fglock++'s new files in ext/Set-Infinite. | |||
| r5289, iblech++ | * Documented some more conventions (svn:executable, shebang lines, naming | |||
| r5289, iblech++ | conventions) in pugs::hack. | |||
| iblech | Ok, time to go... See you monday! :) | 18:28 | |
| Aankhen`` | See you! | ||
| gaal | bye iblech! :) | ||
| Aankhen`` | Have fun! | ||
| iblech | Thanks :) | 18:29 | |
| off & | |||
| Aankhen`` | Hmmm. | 18:36 | |
| The relevant line from Test.pm is this: eval "package $caller; require $module"; | 18:37 | ||
| I want to replace that with: &::($module).import.goto(...); # it needs to pass the relevant parameters, of course | |||
| But apparently I can only do that by using ::, not ., like so: &::($module)::import.goto(...); | 18:38 | ||
| gaal | looks good to me. (don't know about the ::/. stuff though) | ||
| Aankhen`` | Do you know where Sub::goto (or whatever the full name is) and the like are specced? | 18:39 | |
| gaal | there was precious little speccage of that when i wrote goto.t | ||
| Aankhen`` | They don't seem to be in S06... | ||
| gaal | but that was a while ago | ||
| i just grepped S* | |||
| Aankhen`` | Is there anything about overriding `import`? | 18:42 | |
| gaal | sorry, dunno. | ||
| Aankhen`` | &func.goto(@args) | 18:45 | |
| (groups-beta.google.com/group/perl.p...43be3915f) | |||
| svnbot6 | r5290, Aankhen++ | * Test.pm: added (currently commented out) alternative implementation of &use_ok. | 18:58 | |
| Limbic_Region yawns and wonders what is new in the exciting world of Pugs | 19:27 | ||
| svnbot6 | r5291, gaal++ | * WIP: new [internals] builtins: IO::tell, Pugs::Internals::hIsOpen, | 19:51 | |
| r5291, gaal++ | Pugs::Internals::hIsClosed, Pugs::Internals::hIsReadable, | |||
| r5291, gaal++ | Pugs::Internals::hIsWritable, Pugs::Internals::hIsSeekable. | |||
| r5291, gaal++ | For some reason, most of these lie (eg, hIsReadable retrurns true on :r | |||
| r5291, gaal++ | files) so this needs some investigation still. | |||
| gaal | err | ||
| the lie is that hIsWritable says true on :r :-) | 19:52 | ||
| svnbot6 | r5292, fglock++ | cleanup | 20:11 | |
| elmex | what is about the $self->method () syntax, is thre anything decided? | 20:27 | |
| svnbot6 | r5293, fglock++ | added DateTime::Span | ||
| Aankhen`` | G'night. | 20:43 | |
| wolverian | wow, ruby's templating systems all suck, or are too heavy (like rails) | 20:54 | |
| masak | how is rails too heavy? (I haven't used it so, I don't know) | 21:01 | |
| nothingmuch collapses | 21:06 | ||
| wolverian | masak: it does too much for this projec | 21:08 | |
| t | |||
| dudley | Who exactly is @larry? | 21:12 | |
| integral | (TINC) | 21:14 | |
| cwest | someone in .cz | 21:15 | |
| or someone connecting from there anyway | |||
| masak thinks that is a funny answer | |||
| dudley doesn't get it | 21:17 | ||
| masak | dudley: @larry is the core team designing perl6 | ||
| larry wall and a few others | |||
| dudley | masak: Yeah, I meant for @larry -> {say $_;} | 21:18 | |
| I've been following along for a while and I just realized that $larry was the only one of @larry that I knew... | 21:19 | ||
| svnbot6 | r5294, fglock++ | "complement" | ||
| dudley gets it now... | 21:34 | ||
| masak doesn't | 21:43 | ||
| svnbot6 | r5295, fglock++ | half-implemented "union" | ||
| masak | explain, please | ||
| autrijus | rehi. | 21:47 | |
| "Introduction to Pugs" and "Perl 6 in the Real World" are both rejected from Euro OSCON :-/ | 21:48 | ||
| wolverian | :/ | ||
| PerlJam | autrijus: why? | ||
| autrijus | PerlJam: no idea... "topics was too narrow, covered by other presentations, or too vendor specific" | 21:49 | |
| PerlJam | If they already have a full billet, I can understand it. | ||
| autarch | what vendor is that? | ||
| autrijus | however, my "Introduction to Haskell" and "Perl and Internationalization" got accepted (a while ago) | ||
| PerlJam | It was probably the "topic too narrow" part that got them | ||
| autrijus | indeed | 21:50 | |
| PerlJam | autrijus: just use pugs in your introduction to haskell ;-) | 21:51 | |
| autrijus | PerlJam: I very much doubt it's as entry level as I billed it :) | ||
| I'll perhaps use Harrorth instead | |||
| nothingmuch++ # superb introduction material | 21:52 | ||
| masak | nothingmuch++ # agreed | ||
| PerlJam | autrijus: well, clearly perl6 is relevant to perl and internationalization. :-) | ||
| autrijus | PerlJam: yeah :) | 21:54 | |
| Limbic_Region | welcome back to the land of the living autrijus | 22:00 | |
| clkao | hello autrijus | 22:01 | |
| you recovered? | |||
| autrijus: are you also offered 25% discount for euro oscon? | |||
| oh you got something accepted | |||
| gaal | i figured out why seek and tell were failing | 22:02 | |
| autrijus | clkao: they don't want svk? :-/ | ||
| gaal | on windows theyt only work in binary mode | ||
| autrijus | Limbic_Region: heya, glad to be back | ||
| gaal: ! | |||
| gaal | #haskell++ | ||
| clkao | autrijus: no. | ||
| autrijus | sad. | ||
| clkao | autrijus: it's probably too vendor specific | ||
| though i am sure there will be subvesrion tutorial again | 22:03 | ||
| gaal | yes. well. | ||
| autrijus | clkao: so you won't be @ amsterdam? | ||
| clkao | i probably will. | ||
| autrijus | or is fotango going to pay your 75%? | ||
| cool! | |||
| Limbic_Region | autrijus - when you are sufficiently rested and want to here the bizzare story of why Prelude.hs failed to compile on my home machine Win32 - let me know | ||
| clkao | i guess so. we have conference budgets | ||
| Limbic_Region | I also journaled it, but skimmed over the details | ||
| clkao | autrijus: how's it going? no new pugs release? you suck | ||
| autrijus | clkao: I suck, I was totally dead | 22:04 | |
| autarch | clkao: How long have you been at Fotango? | ||
| gaal | autrijus, what's the Right Way to return in a Void Prim? I did `return undef` because `return ()` didn't compile, but that looks wrong. | ||
| clkao | autarch: since nov. but physically since apr. | ||
| autarch | ah | ||
| gaal | wow, what a barrage of questions :) | ||
| autrijus | Limbic_Region: I've read your journal; the .dll searchpath problem is annoying | ||
| clkao | autrijus: what's blocking? topic? | ||
| autrijus | gaal: return; | ||
| clkao | autrijus: does svk suck for you lately? | 22:05 | |
| gaal | in haskell | ||
| autrijus | gaal: wait, prim? you don't return in prim | ||
| oh. | |||
| retEmpty | |||
| gaal | thanks! | ||
| autrijus | np | ||
| clkao: yeah, just the topic | |||
| Limbic_Region | I would be fine with GHC having a hard coded path to its own perl.exe if it at least spit out useful error messages when it couldn't find it | ||
| autrijus | clkao: but I need to work on some p5 (chaldea) now | ||
| clkao | autrijus: i am going to release 1.01... so let me know if something sucks | 22:06 | |
| or you don't use svk anymore? | |||
| autrijus | Limbic_Region: indeed. post to GHC's mailing list or sf.net tracking system? | ||
| clkao: er, I use svk so much that I forgot its existence | |||
| Limbic_Region thinks it is kinda silly to force a hardcoded path anyway since there is nothing special about the actual executable | |||
| I replaced it with a different one and the problem went away | |||
| gaal | okay, finally open($file, :layer<:raw>) is useful for something | ||
| clkao | autrijus: same as i do, so it gets weird to hack on it | ||
| autarch: how about you? | 22:07 | ||
| Limbic_Region | autrijus - does the list allow for non-members to post? I am not going to subscribe just to say I think it is silly | ||
| autrijus | Limbic_Region: I have no idea; probably you can subscribe, post, and unsubscribe; otoh, they do look at their sf.net bugtracking | ||
| which requires no subscription | |||
| autarch | clkao: you mean work? I'm at Socialtext since September | ||
| clkao | autarch: i know, i mean svk :) | 22:08 | |
| autarch | oh, sorry | ||
| Limbic_Region | ok - will look into it | ||
| autarch | I haven't used it recently | ||
| still needing that emacs mode ;) | |||
| clkao | autrijus: what are you hacking on chaldea? | ||
| autarch: i thought there's a emacs mode on the wiki | |||
| autarch | it's busted ;) | ||
| and I have no lisp-fu | |||
| clkao | i only know it requires development release of emacs | ||
| autarch | ah | 22:09 | |
| I have what's in debian unstable | |||
| autrijus | clkao: validation etc | 22:10 | |
| clkao | ah, forward port | ||
| svnbot6 | r5296, gaal++ | make seek and tell raise exceptions on failure (GHC requires binary mode for | 22:11 | |
| r5296, gaal++ | these operations on win32) | |||
| clkao | autrijus: have you listened to gugod's new song? | 22:12 | |
| gaal | Aankhen`` (hoping you backlog): you can now use seek. | ||
| who wants to help me figure out why hIsWritable lies to me in pugs, although a ghci test case shows it works correctly by itself? | 22:13 | ||
| did I define it correctly? | 22:14 | ||
| # op1 "Pugs::Internals::hIsWritable" = boolIO hIsWritable | |||
| autrijus | clkao: eh no. url? | ||
| gaal: boolIO is a try that returns false on failure | |||
| clkao | 03:33 <@gugod> osx.gugod.org/ftp/3AM.mp3 # ååč· lukhnos åēę°ę | ||
| 03:35 <@gugod> osx.gugod.org/index.cgi?gugod # ęč© | |||
| gaal | autrijus: my problem is that i open a file for read only, and isWritable returns true :-) | 22:15 | |
| autrijus | gaal: you want something else entirely :) | ||
| gaal | simply liftIO it? | ||
| autrijus | op1 "Pugs::Internals::hIsWritable" = do | ||
| gaal | ..and VBool the resutl? | ||
| autrijus | op1 "Pugs::Internals::hIsWritable" = \x -> do | ||
| fh <- fromVal x | |||
| fmap VBool (liftIO $ hIsWritable fh) | 22:16 | ||
| something like that | |||
| clkao | still need to use gcc3 on macosx right? | ||
| autrijus | clkao: 6.4.1. snapshots are out | 22:17 | |
| gaal | i get to use op1cast for that, no? :-) | ||
| autrijus | clkao: which should've fixed it, not sure | ||
| gaal: whatever works :) | |||
| gaal | some lifting is definitely in order! i have five functions that should all be oneliners. | 22:18 | |
| clkao | ēę°ć§ćć | ||
| autrijus | gaal: welcome to the pointless club | ||
| Limbic_Region | autrijus - 1234458 is the tracker ID for the GHC bug report FYI | 22:19 | |
| autrijus | er, pointfree club. | ||
| Limbic_Region: wonderful. thanks | |||
| gaal | yeeha! | ||
| Limbic_Region | autrijus++ # thanks for fixing the spaceship parsing bugs | 22:20 | |
| autrijus | :) | ||
| clkao | ghc-pkg: cannot find package plugins | ||
| because mine is ghc-6.4-pkg | |||
| Limbic_Region still has no idea what made him think to try adding back in the perl.exe since he removed it a long time ago | 22:21 | ||
| svnbot6 | r5297, fglock++ | more "TODO" added | 22:22 | |
| Darren_Duncan | just so you guys know, I have a series (2 parts at least) of large updates to make to my database modules before I think of porting them to perl 6 ... however, I *may* start said port prior to OSCON | ||
| also, I'll want to remove all the workarounds in LKT first | 22:23 | ||
| my current impression is that the ObjectModel stuff being worked on now is a prerequisite to some native Pugs support for object features, though I could be wrong | |||
| I mean, MetaModel | 22:24 | ||
| autrijus | that is. | ||
| pasteling | "gaal" at 192.115.25.249 pasted "haskell style Q" (10 lines, 327B) at sial.org/pbot/11605 | ||
| Darren_Duncan | for example: will MetaModel be a prerequisite to doing any of these things ... | ||
| 1. declaring classes in any order and they can still invoke each other naturally | 22:25 | ||
| 2. being able to use can() and functions like that | |||
| 3. having private attributes etc | |||
| autrijus | 1) needs the PIL evaluator -- i.e. separate the parsing from compiling | ||
| 2/3 depends on porting the metamodel to haskell | |||
| Darren_Duncan | right, so I'll have to wait for those before I can do my SQL::Routine etc port | 22:26 | |
| while Locale-KeyedText could be re-ordered to work around it, that won't work with my other modules | |||
| since classes within them each refer to the other | 22:27 | ||
| autrijus | you can port ahead :) LKT and T::Builder both drove a lot of pugs OO features | ||
| Darren_Duncan | I have considered that | ||
| autrijus | so S::R shouldn't be different -- just abandon all workarounds so we can code up to it | ||
| Darren_Duncan | so then, I just do what I did with the last SQL::Routine port ... don't expect it to run? | 22:28 | |
| not quite actually | |||
| so I won't try to work-around issues when I do those modules ... I'll report bugs and wait for fixes before continuing | 22:29 | ||
| autrijus | yup. | ||
| I think many of the feature you require are already in t/oo/ | |||
| but then maybe some are not. I'd be very grateful if we can improve t/oo/'s coverage further | |||
| Darren_Duncan | I consider them to be fairly basic ... mainly just what Perl 5 already supports | ||
| right, so that's the plan | 22:30 | ||
| autrijus | yup | ||
| Darren_Duncan | still, I plan to do said major updates before the port, so I don't have to merge those updates into it later | ||
| autrijus | k. | ||
| Darren_Duncan | fyi, today I will commit a 'Rosetta' and '...Generic' major update ... they're done except the changelogs and I tend to be detailed in those | 22:31 | |
| after that, all my distro versions in CPAN will work with each other | |||
| following that, there's at least 2 large updates across the line ... then the porting begins | 22:32 | ||
| autrijus | k. | ||
|
22:32
Maddingue__ is now known as Maddingue
|
|||
| svnbot6 | r5298, clkao++ | Don't force gcc_select while people can just use $ENV{CC}. | 22:32 | |
| obra | nice | 22:33 | |
| Limbic_Region | autrijus - of your hackathon notes - what % is represented by current pugs? | 22:39 | |
| svnbot6 | r5299, autrijus++ | * improve the CC env message a little bit. | ||
| autrijus | Limbic_Region: what %? | ||
| you mean pugs's implementation of the notes? | 22:40 | ||
| I think most of them are covered by new tests | |||
| not many of them are implemented | |||
| Limbic_Region | gotcha | 22:42 | |
| clkao | ghc-6.4: | 22:48 | |
| src/syck/syck.o: unknown symbol `_printf$LDBLStub' | |||
| is this caused by gcc4 or ? | |||
| gaal | how can i write code that is polymorphic over VBool, VInt, and other types, but which receives one of these types as an argument? sorry if this isn't very clear | 22:52 | |
| i want to reuse autrijus' code from above, which did | |||
| fmap VBool (liftIO $ hIsWritable fh) | 22:53 | ||
| i want to make both VBool and hIsWritable parametric | |||
| i experimented with one or two things... btu what's the type of VBool when thought of as a function? | 22:54 | ||
| (brb) | |||
| autrijus | clkao: yeah, you need clean first | 22:59 | |
| clkao | i did | ||
| maybe ghc doesn't respect CC | |||
| autrijus | that's gcc4 | ||
| clkao | hate | ||
| gaal | golfing haskell is wholly different from golfing perl. | 23:00 | |
| autrijus | gaal: the type of VBool is (Bool -> Val) | ||
| but your are looking at existentials | |||
| (forall a. Value a => a -> Val) | |||
| see vop1x in src/Emit/PIR.hs | |||
| gaal | ah, so *thats* how you get the class in there | 23:01 | |
| I knew i wanted to make VBool more generic but Value didn't work there | 23:02 | ||
| autrijus | study vop1* | ||
| and you'll eventually grok sexy types :) | |||
| gaal | oo | ||
| gaal studies | 23:03 | ||
| autrijus plugs for pugs on lambda the ultimate | 23:04 | ||
| gaal | some funky syntax to pick up there. <:=, <--, <== | ||
| autrijus | # lambda-the-ultimate.org/node/view/827 | ||
| gaal: er, nvm them, they are just userdefined infix | |||
| you know, like perl6 | |||
| gaal | uh huh | ||
| autrijus | see line 203 | ||
| gaal | okay, yup | 23:05 | |
| so there's a fundamental difference beween the commented out sig and the existential version. what is it? | 23:06 | ||
| --vop1x :: (RegClass a, RegClass b) => SubName -> PrimName -> a -> b -> Decl | 23:07 | ||
| or are they equivalent? | |||
| i think the Gentle tutorial doens't cover this. is it not in haskell 98? | 23:08 | ||
| wolverian | autrijus: s/it's/its/ # just being pedantic :) | 23:10 | |
| autrijus: (re: ltu) | |||
| oh, that wasn't your article. duh. | |||
| wolverian hides under a rock | |||
| gaal runs out of wakies | 23:12 | ||
| i'll get these supporting functions in tomorrow - they're not blocking anyone, anyway | |||
| if you see him tell Aankhen`` he has seek now. :) | |||
| gaal waves. zzZZ & | 23:13 | ||
| autrijus | gaal: they diff. the forall's are in different place. | 23:14 | |
| gaal: cool :) next time we're both up I'd gladly show you the wonder of sexy typing | |||
| clkao | autrijus: so has svk reached the 'what's the big deal' stage? | 23:27 | |
| autrijus | clkao: I'd say so. | ||
| clkao | autrijus: ok, good luck carry on with pugs :) | ||
| autrijus | clkao: on the other hand... there's one phase after that stage | 23:28 | |
| it's called "world domination" ;) | |||
| clkao grins | |||
| meppl | :D | 23:41 | |
