|
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. |
|||
| svnbot6 | r14612 | Darren_Duncan++ | removed my in-progress docs/talks/rdbms_tutorial_draft.txt ... something analagous may or may not be added later | 00:19 | |
|
00:40
Steve_PH_Pythuby joined
00:53
eggzeck[laptop] joined
00:54
eggzeck[laptop] is now known as eggzecl
00:55
eggzecl is now known as eggzeck,
qmole joined
01:07
digitalice joined
|
|||
| digitalice | hey | 01:07 | |
| !listo | |||
| > cos 9 | |||
| lambdabot | -0.9111302618846769 | ||
| digitalice | :D | ||
| > www.google.com | |||
| lambdabot | Not in scope: `com' | 01:08 | |
| digitalice | > @google www.google.com | ||
| lambdabot | Parse error | ||
| digitalice | |||
| lambdabot | Parse error | ||
| digitalice | |||
| lambdabot | Empty search. | ||
| digitalice | @google mvd shuffle | ||
| lambdabot | www.mvdshuffle.com/ | ||
| Title: Cargando... | |||
| digitalice | @google foro mvd shuffle | ||
| lambdabot | www.mvdshuffle.com/ | ||
| Title: Cargando... | |||
| digitalice | @google foro.mvd shuffle | ||
| lambdabot | dodeca.org.uy/foro/viewtopic.php?pid=1580 | ||
| Title: DodecĆ” // Foro / MVD Shuffle! | |||
| digitalice | @google foro.mvdshuffle.com | ||
| lambdabot | www.teatromagico.net/phpbb2/viewtopic.php?t=8102 | 01:09 | |
|
01:09
digitalice left
01:21
dmq joined
01:25
jferrero joined
01:35
mjk joined
01:46
lichtkind joined
|
|||
| lichtkind | when filling an capture it is possible to do my |car_cap = \("Model T", doors => 2); | 01:47 | |
| audreyt | there's no | sigil anymore | 01:49 | |
| it existed for exactly one day and one revision | |||
| Limbic_Region | for some definitions of "day" | ||
| TreyHarris | urgh.... i'm so confused when i try to do Perl 5 now.... sometimes I confuse it with Perl 6, sometimes I confuse Perl 6 with Perl 5, and sometimes I just go off in some wild direction that is neither... | 01:50 | |
| in Perl 5, should "if (my $value = bar() and $value ne 'bad thing') { ... }" do what I expect? i.e., go into the block iff both: a) bar() returned something true, and b) that same call to bar() did not return 'bad thing'? | |||
| lichtkind | audreyt but i read it still in svn.openfoundry.org/pugs/docs/Perl6...apture.pod | 01:51 | |
| lambdabot | tinyurl.com/y5x48u | ||
| lichtkind | so now captures are just skalars? | 01:52 | |
| audreyt | hm? | ||
| the Capture.pod has no car_cap | |||
| no |car_cap that is | |||
|
01:53
loumz joined
|
|||
| audreyt | my $car_cap = \("Model T", doors => 2); | 01:53 | |
| my $car = make_car(|$car_cap); | |||
| lichtkind | yes | ||
| audreyt | you can also say | ||
| my $cap_cap := \(...) | |||
| lichtkind | and thatswhy it seemed to me that $car_cap is reference to capture | ||
| audreyt | but yeah, there's no special sigil anymore | 01:54 | |
| well, it's a Scalar containing a Capture object | |||
| if you say | |||
| my $c_c = \(...) | |||
| hm, c_c is a cute face | |||
| TimToady | Trey_Harris: yes, it should do what you expect. | ||
| audreyt | to interpolate the capture into the arguments | ||
| you need to |$c_c | 01:55 | ||
| because otherwise it's passing in the Capture itself | |||
| not interpolating its contents | |||
| TreyHarris | TimToady: thanks for the sanity check :) | ||
| TimToady | er, wait, no. | ||
| P5 doesn't introduce the new variable that soon. | |||
| lichtkind | thanks audreyt makes sense so is this only meaning of | ? | ||
| TimToady | now you're confusing me... | ||
| TreyHarris snaps his fingers | |||
| *that's* what I thought | |||
| :) | |||
| you see? | 01:56 | ||
| audreyt | with MP6 emitter looking good, tomorrow will be PCR-port-to-MiniPerl6, which will finish MP6 bootstrap | ||
| and then if we have a emitter to js, then you can eval"" MP6 in your browser | |||
| TimToady | been in P6-land too long, and not long enough... | ||
| audreyt | if we have an emitter to C, then we have libp6re | ||
| and so on and so forth. | |||
| but p5 and hs will still be the first two emitters we'll work on. | 01:57 | ||
| lichtkind: correct | |||
| lichtkind | audreyt thanks your very kind | ||
| TimToady | only meaning of unary | anyway | ||
| infix | is of course junctional OR | 01:58 | ||
| audreyt | unless it's in type option, where it denotes a type union | ||
| s/option/position/ | |||
| TimToady | It's a junctional OR that gets coerced to a set. :) | ||
| audreyt | well, it's really a different syntax there, much like its use in grammars | 01:59 | |
| to denote a choice | |||
| TimToady | yah | ||
| so what's the js emitter gonna do with 'use v5' :) | 02:01 | ||
| audreyt | we take out "use v5" already :) | ||
| in the MP6 boot chain | |||
| TreyHarris | so it just gets run optimistically as if it were Perl 6 code? :) | ||
| audreyt | or maybe it assumes it's running inside Inline::SpiderMonkey... | 02:02 | |
| er JavaScripdt::SpiderMonkey | |||
| rgs: how far away (by plane or car) is your currennt meatspace position from Nice, France? | |||
| POPL'07 people (the SIGPLAN conference for lambdafolk language designers) just agreed to fly me there for opening keynote. | 02:03 | ||
| January 2007 | |||
| rgs: so I wonder if we can have a couple days of minihackathon on unified perl runtime | 02:04 | ||
| lichtkind | audreyt and due the module of Perl6::Doc ....im now member of perldoc2.0 project, its all starting just and we sorting ideas and aim no will smash out a prototype app but i gladly will help adopt this system for perl6 if it will become standart | ||
| audreyt | "aim no will smash out"? | ||
| lichtkind | im mean now smash out hm a german term im mean throw together | 02:05 | |
| audreyt | but cool, we need a perl6doc viewer very much | ||
| cool! | |||
| lichtkind | but hold on its more complicated :) | ||
| im about to write article for perl.com and the example code of this lession will be the building of a multi tab pod viewer | 02:06 | ||
| that can easily be extended for perl6 needs i think | 02:07 | ||
| audreyt | nice | ||
| audreyt nods happily | |||
|
02:07
kisu joined
|
|||
| lichtkind | the app for the perldoc2 project is more a tool for translators because perldoc2 is about to have centralized translated perldocs | 02:07 | |
| audreyt ponders the curoius fact that there's both marsneedswomen.pm.org and women.pm | 02:09 | ||
| lichtkind | audreyt so throw me just some lines with what you think you need for features to [email@hidden.address] or right here in channel so that i can copy the text in my dev dir as todo list | ||
| audreyt | sure. I'd like to have "perldoc -f keyword Some::Module" | 02:11 | |
| where keyword searches into sections and indices | |||
| lichtkind | yeah but would be command line tool better that gui app? | 02:12 | |
| audreyt | I think it can invoke a gui app... | ||
| I like gui apps that can be invoked by smart commandline options | 02:13 | ||
| lichtkind | yes but i think many like vi because they can work with it text based | 02:14 | |
| audreyt | I have to admit my preferred gui interface is html | ||
| but I'm biased :) | |||
| loumz | hello, small question about filetest operators... if I have sub w {} and I write - w("some-arg") with a space between "-" and "w" then it still executes the -w filetest | ||
| is this behavior correct? | |||
| audreyt | no that's utterly wrong. please write a test | 02:15 | |
| TimToady | no | ||
| audreyt | in filetest.t or other places | ||
| do you have a commit bit? | |||
| if not, what's your email address? :) | |||
| loumz | no, I don't | ||
| william99 at myrealbox.com | |||
| eviltwin_b prefers command line for stuff like this: a properly written command line app can be guified easily enough, plus I can run it in emacs :) | 02:16 | ||
| lichtkind | so we make first command line version | ||
| pasteling | "loumz" at 125.164.185.231 pasted "filetest operators question" (17 lines, 476B) at sial.org/pbot/20869 | ||
| loumz | there's some code that demonstrates it | 02:17 | |
| audreyt | loumz: welcome aboard, please add yourself to AUTHORS :) | ||
| loumz | audreyt: thanks | ||
| lichtkind | audreyt what you mean with "perldoc -f keyword Some::Module" are there current modules in perl6? | ||
| audreyt | t/operators/filetest.t is a fine place for it, as is making a new file under t/syntax/parsing/ | ||
| lichtkind: sure, there's plenty of modules under ext/ | 02:18 | ||
| but I mean the -f keyword would be nice to work with modules not only perlfunc | |||
|
02:18
Khisanth joined
|
|||
| lichtkind | ok but i thougt that module ist firstly for the synopses | 02:18 | |
| audreyt | synopses are modules too | ||
| lichtkind | ah :) | 02:19 | |
| audreyt | Perl6::Spec::Operator | ||
| TreyHarris | i'm a module, you're a module, wouldn't you like to be a module too? | 02:20 | |
| lichtkind | no | 02:21 | |
| eviltwin_b | he's a module... | ||
| lichtkind | im special :) | ||
| eviltwin_b | so you're a pragma? | ||
| lichtkind | hihi no even stranger, stranger than fiction | 02:22 | |
| ok just wanted show my goodwill need some sleep before down, keep up the good work @ all here | |||
| audreyt | goodwill is appreciated :) | 02:23 | |
| have fun hunting | |||
| I mean sleeping | |||
| [particle] | audreyt: see parrot's shiny new src/pmc/capture.pmc | 02:24 | |
| (it's missing an invocant slot, but not for long) | |||
| audreyt | cool! | ||
| [particle] | soon we'll have subclasses for match, past, and maybe signature | 02:25 | |
| lichtkind | gnight | 02:26 | |
| audreyt | sure, that's how it's supposed to be used. nice! | ||
| [particle] | yes, indeed. this should speed up pge a bit, too :) | ||
| audreyt | will be nice to benchmark the capture-enabled pge with the native-code p6re thing fglock++ and I am working on :) | 02:27 | |
| [particle] | then we need to convert the emitter from pir to c... that'll make it really fast | ||
| yes, it will. hopefully soon i'll be implementing your miniperl6 grammar in pir | 02:28 | ||
| audreyt | implement in pir? | ||
| I thought PGE already handles it? | |||
| [particle] | it doesn't handle everything there 'does', 'of' | 02:29 | |
| audreyt | I mean, if you ignore the first paragraph | ||
| [particle] | 'where' | ||
| audreyt | svn.openfoundry.org/pugs/v6/v6-Mini...Grammar.pm | ||
| lambdabot | tinyurl.com/y6j9yc | ||
| audreyt | should be totally PGE friendly | ||
| so all you need is to write a MP6->PIR emitter | |||
| and then you have MP6 working inside PGE as transformation language | |||
| all bootstrapped for free. | |||
| i.e. change the perl5 bits in svn.openfoundry.org/pugs/v6/v6-Mini...Emitter.pm to pir bits. | 02:30 | ||
| lambdabot | tinyurl.com/wrq2p | ||
| [particle] | hrmm | ||
| audreyt | the idea is the emitter is first run with pugs | 02:31 | |
| then you have entire MP6 chain emitted to PIR | |||
| afterwards you save those .pir inside parrot treee | |||
| and then you don't need pugs anymore, since you effectively have a MP6 language in native PIR | |||
| [particle] | right, but we can't bootstrap ourselves | 02:32 | |
| audreyt | hm? | ||
| you can | |||
| after that you just use the MP6-in-PIR to run modified perl6 code | |||
| it's not handwritte PIR | |||
| [particle] | ah, ok | ||
| audreyt | but why should you care? the master source is in perl6. | ||
| [particle] | so we use pugs once-and-once-only | ||
| audreyt | which I hope is still the goal for PGE :) | ||
| yes. | |||
| [particle] | yes, it is the goal | ||
| then we get shared spec, shared tests, and (mostly) shared implementation :) | 02:33 | ||
| whee! | |||
| audreyt | the only bit not shared is the emitter | ||
| but that's as it should be | |||
| [particle] | right, totally appropriate | ||
| however, we may have an ast transformation before emitting | 02:34 | ||
| audreyt | if you look in Grammar.pm | ||
| there's AST transformation right there | |||
| written in MP6 :) | |||
| which is a tree transformation language. | |||
| [particle] | right. we'll likely have to write one for perl6 ast to parrot ast | ||
| audreyt | that again is just the emitter. | ||
| instead of emitting pir you emit past calls | 02:35 | ||
| [particle] | this is true. | ||
| audreyt | hopefully this can let us conveniently obsolete PGE, Parsec and PCR. | ||
| which is #1 diverging part so far. | |||
| the idea is the new p6re is written in MP6 | 02:36 | ||
| compiles rules to MP6 | |||
| which is emitted via whatever emitter to efficient native code that matches stuff. | 02:37 | ||
| it's currently living in svn.openfoundry.org/pugs/perl5/Pugs...Ratchet.pm | |||
| lambdabot | tinyurl.com/ydzvqn | ||
| audreyt | will be updated to MP6 and moved to v6/ space tomorrow. | ||
| [particle] | great, i look forward to it | 02:38 | |
| audreyt | :) | 02:39 | |
|
02:59
SubStack joined
|
|||
| pasteling | "loumz" at 125.164.185.231 pasted "proposed t/syntax/parsing/filetest.t, does this look right?" (19 lines, 361B) at sial.org/pbot/20870 | 02:59 | |
| TimToady | looks pretty good to me. | 03:01 | |
| go ahead and check it in. we operator on the forgiveness-rather-than-permission principle around here. | 03:02 | ||
| loumz | TimToady: thanks for the review | ||
| TimToady | *operate | 03:03 | |
| eviltwin_b | operator:<forgiveness> > operator:<permission> | ||
| loumz | all I have to do for the test to work in "make test" is create a new .t file right? | 03:04 | |
| TimToady | I believe so | 03:05 | |
| might run util/add-svn-props.sh on it too | |||
| loumz | TimToady: OK, I guess I should read the Pugs/Doc/Hack.pod first | 03:08 | |
| TimToady | that would probably be more than most of us have done. :) | 03:09 | |
| we just don't use the word "should" around here terribly often... | 03:10 | ||
| svnbot6 | r14613 | loumz++ | Add my name to the AUTHORS file. | 03:14 | |
| r14613 | loumz++ | Add filetest parsing tests for "- f(...)". | |||
| loumz | hopefully that will work :) | 03:15 | |
|
03:25
seano joined
|
|||
| audreyt | didn't quite finish the MP6-bootstrap-katamari journal, so I posted something from earlier draft instead | 03:51 | |
| commitbit ACL transition is done, so we jut need a flag day or something. sleep, see you tomorrow :) | |||
|
03:54
diakopter left
|
|||
| audreyt | s/or something/to relocate master repo to svn.pugscode.org on feather/ | 03:54 | |
|
04:35
esstone joined
|
|||
| gnuvince | Good morning, afternoon and evening depending on your timezone | 04:53 | |
|
05:33
bennymack joined
05:45
toni_ joined
06:15
BooK joined
|
|||
| wolverian | whoa - nice news about popl07. congratulations, audreyt :) | 06:20 | |
| clkao | , | 06:34 | |
| nothingmuch | hola homies | 06:43 | |
|
06:53
esstone joined
06:54
chaoslawful joined
|
|||
| seano | um... word, g. | 06:59 | |
| nothingmuch | you know what i'm sayin | 07:00 | |
| seano | true dat. | ||
| I mean, fo sheezy. | |||
| nothingmuch | maybe you'd like to participate | ||
| avva.livejournal.com/1673636.html | |||
| lambdabot | Title: avva: geek rap contest (Š°Š½Š³Š».) | ||
| seano | verily, ma neezy | 07:02 | |
|
07:17
eggzeck[laptop] joined
07:39
codermattie joined
07:47
codermattie left
|
|||
| rgs | audreyt: Paris <-> Nice is 6 or 7 h by train, 1 h by plane, lots by car. | 08:04 | |
|
08:06
loumz joined
08:16
gene9 joined
08:22
kisu joined
08:24
crem_ joined
08:39
larsen_ joined
08:41
gene9 joined
08:46
iblechbot joined
09:12
_fab joined,
Excedrin joined
09:23
elmex joined,
theorbtwo joined
09:25
marmic joined
09:43
elmex joined
09:51
spinclad joined
10:06
loumz joined
10:11
polettix joined
10:13
Excedrin joined
11:08
rodi joined
11:11
TreyHarris joined
11:12
the_dormant joined
11:16
kanru joined
11:35
kane-xs joined
11:40
TreyHarris joined
12:04
TSa joined,
TimToady joined
12:06
Excedrin joined
12:29
elmex joined
12:30
nusgnaf joined
12:31
Psyche^ joined
12:47
Psyche^ is now known as Patterner
12:55
the_dormant joined
13:11
Excedrin joined
13:21
Debolaz2 is now known as Debolaz
13:23
anatolyv joined
13:29
rodi joined
13:33
mayweed joined
13:35
the_dormant joined,
Excedrin_ joined
13:44
fglock joined
13:46
polettix joined
|
|||
| fglock | audreyt: PCR-in-Mp6 is under way - now writing a few tests | 13:49 | |
| audreyt | fglock: where are you? and when is lightning talks | 13:50 | |
| fglock | I'm at the Perl booth - don't know about the l-t yet | 13:54 | |
|
13:57
turrepurre joined
14:17
fglock joined
|
|||
| audreyt | fglock: coming to Perl booth in 15mins | 14:29 | |
| clkao | zz | 14:30 | |
| audreyt | greetings clkao. is there a way to make hiveminder and comitbit share auth? | 14:31 | |
| or do you think separate auth makes more sense? | |||
|
14:31
weinig joined
|
|||
| integral | a bitcard Jifty plugin would be handy... | 14:32 | |
| fglock | audreyt: ok | ||
|
14:41
ruoso joined
|
|||
| fglock | ruoso: hi | 14:45 | |
| ruoso | hi fglock | ||
| ruoso just sad for not being in CONISLI | |||
| fglock | CONISLI++ | 14:46 | |
| ruoso | how is it going? | ||
| fglock | the main talks are going pretty well, and we have a nice booth thanks to Segula :) | 14:47 | |
| ruoso | :) | ||
| are there video/audio streamings? | 14:49 | ||
| the_dormant | Hi, is there any tutorial to Perl6 rules ? | 14:51 | |
| fglock | ruoso: no, but the video will be available later | 14:53 | |
| ruoso | fglock, at least :) | 14:54 | |
|
15:09
polettix joined
15:11
Limbic_Region joined
15:17
nperez joined
15:26
jferrero joined
15:32
diakopter joined
15:39
TSa joined
16:18
miyagawa_ joined
16:19
the_dorm1nt joined
16:48
eggzeck joined
|
|||
| audreyt | visiolization: pugs.blogs.com/photos/visiolization...tstrap.png | 16:52 | |
| lambdabot | tinyurl.com/yzjxzo | ||
| audreyt | bold-black-lined boxes are the components written in mp6 that we produce in this hackathon | 16:53 | |
| [particle] | you produced the regex source too, no? | 16:54 | |
| audreyt | hm? | ||
| it goes like this | |||
| MP6 source containing both sub{...} and token{...} gets parsed | 16:55 | ||
| for the token{...} part the content goes into a subcompilation cycle | |||
| and then spliced back so it becomes just another sub{...} | |||
| so one MP6 source produces one MP6 AST | |||
| TimToady | I think the regex part was mostly already there due to :ratchet emitter | ||
| audreyt | which is then emitted | ||
| actually the only vapourware part now is emitter | 16:56 | ||
| and emitter-specific prelude | |||
| which we're working on now :) | |||
| note the vapour cloud around "backends" | |||
| TimToady | I was wondering last night whether we need to partition out t/ into MP6 and non-MP6 | ||
| audreyt | otoh MP6 uses no assignment | ||
| so it warrants its own t/03- or something | 16:57 | ||
| audreyt is also working on a formal operational semantics for MP6 | |||
| [particle] | is Test.pm valid MP6? | ||
| audreyt | not at all | ||
| not by far! | |||
| TimToady | not without running it through the P6-to-MP6 translator. :) | 16:58 | |
| audreyt | but that is like, slow :) | ||
| emulating P6 on MP6 is prolly not wise | 16:59 | ||
| the plan is to make MP6 AST a proper subset of P6 AST | |||
| not only the surface syntax | |||
| once the MP6 boot completes | |||
| and so we can reuse the emitters | |||
| and gradually add P6 nodes to it | |||
| I think that's the plan, anyway. still sketching madly with fglock | |||
| ENOPAPER at the moment, so going to hunt for some | |||
| bbiab | |||
| [particle] | i didn't realize mp6 ast wasn't a subset of p6 ast | 17:01 | |
| TimToady | more like p6 ast isn't yet a proper superset of mp6 ast. :) | ||
| I predict someone will eventually write P6-to-MP6 just so that they can run P6 on their nanobots. Who cares if it's "slow" if there's quadrillions of 'em running at molecular speeds.... | 17:05 | ||
| It's possible the gray goo catastrophe will be programmed in Perl... | |||
| talk about disruptive technologies... | 17:06 | ||
|
17:06
fglock joined
|
|||
| audreyt | fglock: should we perhaps go back checking for lightning talk schedule first so we don't miss it? | 17:07 | |
| fglock | audreyt: sure | 17:08 | |
| Limbic_Region | pugs is like a box of chocolates - you never know what you're gonna get | 17:09 | |
| audreyt | fglock: also do you know if there's a better place to get your new laptop other than shopping centre norde? | ||
| Limbic_Region | I thought we were going to get 6.28 (objects) but instead we look like we are gonna be getting bootstrapping | 17:10 | |
| woot | |||
| fglock | audreyt: I'll ask | ||
| anatolyv | S02:2327 says, "in case of a "tie" the grammatical category that is notionally "first" in the grammar wins." But is the ordering of the categoris specified? (in the list that follows, statement_control comes *after* prefix, so it doesn't seem to be it). | ||
| audreyt | Limbic_Region: we're getting 6.28. from the Haskell side this means we're getting a Rules engine _without_ requiring embed perl5 | ||
| and that runs on native haskell speed | |||
| which will be like much faster. | |||
| anatolyv | unless the list in S02 is meant to specify categories from "last" to "first". | ||
| fglock | Limbic_Region: 6.28 might happen at both Haskell and Perl, who knows :) | ||
| Perl5, that is | 17:11 | ||
| audreyt | that requires another visiolization picture | ||
| to include MO and fullperl6 AST's places | |||
| TimToady | anatolyv: the ordering in question is supposed to be how the magical hashes are ordered within the <%foo|%bar|%baz> construct | 17:13 | |
| Limbic_Region | audreyt - neato | 17:14 | |
| TimToady | S02 is obviously not terribly clear on that point. | ||
| Limbic_Region | fglock - super neato | ||
| anatolyv | TimToady: oh, I see. | 17:15 | |
| fglock | Limbic_Region: we are accelerating so time slows down - audreyt++ only has a few days left here | ||
| anatolyv | TimToady: is the way they're ordered going to be constant - that is, category X will always come before category Y, no matter which context they both appear inside the category list for the lexer - or are there cases when one category is more important than the other in one context and vice versa in another context? | 17:16 | |
| Limbic_Region | fglock - but the distortion in the space-time continuum is localized. The few of us in IRC land are witnessing what is supposed to be impossible | ||
| audreyt | we need a singularity emitter | 17:17 | |
| Limbic_Region just watched a StarGate SG1 episode in which the gate was connected to a world being sucked into a black hole and this very same phenomina was being experienced | |||
| fglock | Limbic_Region: we are avoiding to observe it so that it doesn't collapse | 17:18 | |
| Limbic_Region wanders off to go grocery shopping | |||
| fglock | the vapourware seems to be densifying however | ||
|
17:19
TSa joined,
seano joined
|
|||
| fglock | audreyt: are you getting to the lobby? | 17:20 | |
| audreyt | yes, give me another 5min or so | 17:21 | |
|
17:24
pen1 joined
|
|||
| svnbot6 | r14614 | fglock++ | MiniPerl6 - added initial Regex->MP6AST parser, Token->MP6 emitter | 17:24 | |
| TimToady | anatolyv: that will be up to whoever writes the grammar. the specified mechanism is local, so global ordering would be by policy, not by technological requirement. | 17:28 | |
|
17:30
Excedrin joined
|
|||
| audreyt | back with more paper and design | 17:39 | |
|
17:45
mr_ank joined
17:51
mr_ank joined
|
|||
| audreyt | full-p6 boot plan sketched | 18:09 | |
| with run- and compile-time MO protocols | |||
| more details later | |||
| but we'll go laptop shopping now :) bbl | |||
|
18:14
mr_ank joined,
beppu_ joined
18:33
weinig is now known as weinig_
18:34
weinig_ is now known as weinig
18:35
mr_ank joined
18:42
gnuvince joined
|
|||
| eggzeck | Limbic_Region: Audrey just spoke in #perl6, heehee | 18:57 | |
|
18:58
Excedrin joined
19:02
buetow joined
19:04
elmex joined
19:05
mauke joined
19:08
cmarcelo joined
19:16
Excedrin joined
|
|||
| cmarcelo | @tell audreyt audrey and flavio: will you be at the lobby around 18h? im watching mongers talks. the lightinings are scheduled for tomorrow. | 19:16 | |
| lambdabot | Consider it noted. | ||
| cmarcelo | lambdabot: tks | 19:17 | |
|
19:21
kane-xs joined
19:28
idiotmax joined
19:45
Aankhen`` joined
|
|||
| xinming | audreyt: where to see the full plan of bootstraping perl 6 please? :-) | 20:23 | |
| Limbic_Region | xinming - in her head | 20:28 | |
| wolverian | audreyt, yay! | 20:29 | |
| audreyt | @tell cmarcelo we are in the lobby and I'm rsync'ing my entire /Applications/ to fglock. come join us! | 20:38 | |
| lambdabot | Consider it noted. | ||
| gnuvince | audreyt: have you had hardware problems with your MacBook? | 20:40 | |
| audreyt | no... | 20:41 | |
| lambdabot | audreyt: You have 8 new messages. '/msg lambdabot @messages' to read them. | ||
| wolverian | heh | 20:42 | |
| someone should fix that bug | |||
| where lambdabot doesn't realise @tell means the person is present | |||
| gnuvince | audreyt: my iBook died, I want to buy a MacBook, but I keep hearing horror stories | ||
| audreyt | it all works for me | 20:43 | |
| gnuvince | Great | ||
| Thanks for the info | |||
|
20:54
eggzeck joined
20:56
wilx joined
21:14
fglock joined
|
|||
| audreyt | fglock: hi :) | 21:15 | |
| fglock | hey | ||
| audreyt | fglock: "svk pull" instead of "svn up" | 21:16 | |
| otherwise all commands are same | |||
|
21:19
esstone joined
|
|||
| clkao | audreyt: so commitbit-p? | 21:20 | |
| audreyt | clkao: a sec. setting up things for fglock's all new macbook | 21:23 | |
| you'll find him running tests before commit now :) | |||
| clkao | yay | ||
| audreyt | and also svk won't take 1min to just startup | ||
| fglock | :) | 21:24 | |
|
21:28
SubStack joined
21:34
cmarcelo joined
|
|||
| svnbot6 | r14615 | fglock++ | MP6 - added a Regex parser test | 21:44 | |
|
21:47
_alex joined
21:48
_alex left
21:55
lisppaste3 joined
21:57
ingy joined
22:13
larsen_ joined
22:19
jrockway joined
|
|||
| clkao | is this comit from new laptop? :) | 22:21 | |
|
22:32
brent[] joined
|
|||
| audreyt | yes :) | 22:32 | |
| trying to find him an editor | |||
| surprisingly hard to find one for ultraedit addict | |||
| (you remember how you took two weeks to brainwash me into vim?) | |||
| so TextMate for now | |||
| Limbic_Region | TextPad perhaps | 22:33 | |
| audreyt | textpad is for mac? | ||
| Limbic_Region | oh | ||
| audreyt | (context: I just helped fglock buying a new macbook) | ||
| Limbic_Region | forgot that part | ||
| audreyt | well there is CrossOver | ||
| obra | TextMate is likely the most right thing | 22:34 | |
| audreyt | worst case I can get fglock working with UltraEdit on Wine | ||
| but that's like insane | |||
| yeah. it even has svk support | |||
| wolverian | does textmate have vim keybindings? | ||
| (or even vi) | |||
| obra | possibly jedit | ||
| obra has been finally getting around to actually configuring vim himself | 22:36 | ||
| clkao | audrey-vim is insanely slow | 22:38 | |
| audreyt | doesn't feel slow in my macbook here... | 22:39 | |
| wolverian | what's audrey-vim? :) | ||
| eviltwin_b | remember that brainwashing she mentioned? :) | 22:40 | |
| obra | audreyt: it does on my mbookpro. I think it had to do with the buffer-scanning code | ||
| jrockway | doesn't emacs have the best haskell and perl6 modes ATM? that sounds like the right editor choice :) | 22:41 | |
| audreyt | obra: aw. disable them then... | ||
| gnuvince | hmmm... | ||
| wolverian | hm, what buffer-scanning code? now I'm intrigued | ||
| gnuvince | ?eval 1538 / 60 | 22:42 | |
| evalbot_r14615 | 769/30 | ||
| gnuvince | ?eval (1538 / 60).fmt("%f") | ||
| evalbot_r14615 | \"25.633333333333333" | ||
| obra | audreyt: I have. it's behaving better. | ||
| I'm starting to play with minibufexplorer and taglist | 22:43 | ||
| gnuvince | fib(28) (recursive) on my PC with Pugs took 25 minutes :-) | ||
| gnuvince considers using a loop | |||
| wolverian | ghc is a bit faster there... | 22:44 | |
| obra | but now -> hackerkitchen | 22:45 | |
| Limbic_Region | gnuvince - with or without memoization? | ||
| wolverian | the lazy list method doesn't work yet :/ (flattening is eager) | 22:49 | |
| svnbot6 | r14616 | audreyt++ | * MP6::Grammar - Parse for escaped chars inside quoted strings. | 22:50 | |
| r14617 | fglock++ | MP6 - Grammar::Regex compiles | 22:53 | ||
| Limbic_Region | gnuvince - first perl 6 I have written in months | 23:00 | |
| recursive but with memoization | |||
| svnbot6 | r14618 | audreyt++ | * v6.pm - change env knob V6NOTIDY to V6TIDY -- that is to say we | ||
| r14618 | audreyt++ | don't run tidy by default. it slows down compilation 4x. | |||
| Limbic_Region | good stuff | ||
| perlbot nopaste | |||
| perlbot | Paste your code at sial.org/pbot/perl and #perl will be able to view it | ||
| lambdabot | Title: sial.org Pastebot - pasteling | ||
| pasteling | "Limbic_Region" at 24.35.57.240 pasted "recursive fibonacci with memoization via state variables" (25 lines, 444B) at sial.org/pbot/20890 | ||
| Limbic_Region | fib(28) is pretty much instantaneous (minus pugs overhead) | 23:01 | |
| (state variables)++ | 23:02 | ||
| pasteling | "Limbic_Region" at 24.35.57.240 pasted "recursive fibonacci with memoization version 2 (more succinct)" (15 lines, 281B) at sial.org/pbot/20891 | 23:05 | |
| audreyt | Limbic_Region: examples/algorithms please :) | 23:06 | |
| I mean, add to | 23:07 | ||
| pasteling | "Limbic_Region" at 24.35.57.240 pasted "recursive fibonacci with memoization (even more succinct but still too perl 5ish)" (10 lines, 187B) at sial.org/pbot/20892 | 23:08 | |
| Limbic_Region | audreyt - will do | ||
| would you mind suggesting how to make it more Perl 6ish? | |||
| audreyt | Int $n | 23:09 | |
| not Integer | |||
| don't need $fib1 $fib2 | |||
| Limbic_Region | yeah | ||
| audreyt | %seen{$n - $_} //= | ||
| Limbic_Region | look at 20892 | ||
| gnuvince is back from food | 23:10 | ||
| audreyt | that's it | ||
| Limbic_Region | ok | ||
| gnuvince | Limbic_Region: and no memoization; I took the Perl 5 code from shootout.alioth.debian.org and I am converting some of those programs to Perl 6, and the recursive fibonacci took a long, long, long time | ||
| audreyt | yes, becasue subroutine call is extremely expensive in our evaluator. | 23:12 | |
| gnuvince | audreyt: gotcha | ||
| If I get enough examples ported, I might commit them in examples/ | 23:13 | ||
| Limbic_Region | gnuvince - the memoization returns fib(42) almost instantaneously as well | 23:15 | |
| ;-) | |||
| waiting for the commit to show up now | |||
| felt good to write some Perl 6 again but can't get sucked in | 23:16 | ||
| gnuvince | Limbic_Region: I'm sure it does, however the "rules" of the alioth shootout is that all examples must be implemented the same way | ||
| svnbot6 | r14619 | Limbic_Region++ | Recursive implementation of fibonacci using state %seen for memoization. Inspired by gnuvince++ | ||
| Limbic_Region | gnuvince - sure | 23:17 | |
| gnuvince - as long as a shootout has a comprehensive set of benchmarks to fully test the strengths and weaknesses of the language it is fine | 23:18 | ||
| the trouble begins when all the examples favor the strengths of 1 language and everyone else has to follow suit | |||
| gnuvince | Limbic_Region: the shootout has about 15 "problems" to solve | 23:19 | |
| svnbot6 | r14620 | audreyt++ | * MiniPerl6::Emitter::Token - Change __END__ to =begin END. | 23:28 | |
| r14620 | audreyt++ | * MiniPerl6::Grammar::Regex - Change assignment (=) to binding (:=). | |||
| pasteling | "anatolyv" at 89.0.235.26 pasted "even more succint and less perl5-ish" (9 lines, 145B) at sial.org/pbot/20894 | 23:36 | |
| anatolyv | Limbic_Region: see paste | ||
| anatolyv: not sure it's clearer though! | |||
| Limbic_Region | yeah, I am looking now | 23:42 | |
| anatolyv - feel free to update the example in examples but I find my version easier to understand | 23:44 | ||
| anatolyv | Limbic_Region: it probably is. | ||
| Limbic_Region | it is odd watching what happens when you type fib(1000) since ghc offers arbitrary precision | 23:45 | |
| anatolyv | it's nice, indeed. | 23:47 | |
|
23:58
GeJ joined
|
|||
