»ö« | perl6-projects.org/ | nopaste: paste.lisp.org/new/perl6 | evalbot usage: 'perl6: say 3;' | irclog: irc.pugscode.org/ | UTF-8 is our friend!
Set by moritz_ on 30 July 2009.
00:14 yahooooo left 00:19 justatheory joined
KyleHa Back. 00:19
00:20 broquaint left 00:24 justatheory left, ihrd left 00:25 carlin left
pmichaud oops 00:34
I mis-judged the utc time
I'll be back around 01:30, actually
KyleHa Oh, OK.
Thanks for letting me know. 8-) 00:35
00:41 SmokeMachine joined
jnthn -> sleep, night all 00:51
KyleHa G'night jnthn.
jnthn KyleHa++: Best of luck on the release!
KyleHa Thank you! 00:52
00:55 __ash__ joined 01:17 Limbic_Region joined 01:19 drbean joined 01:33 jferrero left 01:37 cdarroch left
dalek kudo: 929a700 | (Kyle Hasselbacher)++ | docs/announce/2009-08:
announce/2009-08: passing 82% of spectests
 01:37
01:47 meppl left
pmichaud 82%? I suspect no. 01:52
KyleHa Hmmm. 01:53
pmichaud checking.
KyleHa I think you're right. It's 82% of the ones in spectest.data
01:55 ruoso left
KyleHa What counts the tests for spectest-progress.csv? 01:57
pmichaud a special script I have does it.
It's just an estimate
I just pushed an updated commit 01:58
do you plan to release within the next 3 hours, or after that?
KyleHa It should be all ready to go in the next three hours.
pmichaud actually I should say 3.5 hours :-)
okay.
We'll go with the current numbers then.
12369 passing, 17675 in suite == 71.5%
KyleHa I've already made the tarball and checked that it works, spectested and all.
pmichaud I'm about to push the updated announcement 01:59
KyleHa I get 69.9%
Pretty much 70%, really.
pmichaud oh, I typoed my number earlier. 02:00
KyleHa rakudo: say 12369 / 17675 * 100;
Oh, OK.
p6eval rakudo 0d4fe0: OUTPUT«69.980198019802␤»
pmichaud I did 12639 (oops!)
we go with 70%.
eh, 69.9% 02:01
I'd rather round down than up.
KyleHa So that spectest-progress.csv is updated by a cron job somewhere of yours?
pmichaud it's not a cron job -- it's not completely automated
dalek kudo: 7685eb3 | pmichaud++ | docs/spectest-progress.csv:
spectest-progress.csv update: 430 files, 12339 passing, 0 failing
kudo: 8d8ffc7 | pmichaud++ | :
Merge branch 'master' of [email@hidden.address]
pmichaud it's a script I run, yes.
but sometimes "things go wrong" so it's not entirely trustable to a script
KyleHa I see. 02:02
pmichaud although it's been a while since anything has gone drastically wrong
at any rate, I wouldn't want it automatically pushing commits
KyleHa Nah. I'd have it commit locally and push it whenever. 02:03
pmichaud even then, I don't necessarily want to commit locally
at any rate, it's super-simple to handle it manually right now 02:04
KyleHa Yup, looks it.
pmichaud it only takes a few minutes (except for the time needed to run the tests)
KyleHa Looks like the announce needs the PDX flavor text, then a new tarball, and it's all set.
pmichaud I also want some notes in there about handling of the perl6 executable
KyleHa The problem as I recall is... 02:06
You can build it as usual, but you have to execute it from the directory you built it in.
The fix is 'make install'
After a 'make install', the installed perl6 will run from anywhere. 02:07
Is that right?
02:12 Whiteknight left
pmichaud yes. I'm writing the announcement text for that now. 02:14
KyleHa Oh, thank you!
02:19 molaf joined 02:22 Limbic_Region left 02:23 ruoso joined, justatheory joined
pmichaud just pushed a few announcement updates -- reviews/suggestions/patches welcomed. (I'm still adding more notes.) 02:32
KyleHa Looking.
dalek kudo: 82ef264 | pmichaud++ | docs/announce/2009-08:
More 2009-08 announcements.
 02:37
KyleHa Is there a guideline on the announcement text width? I reformed a paragraph, and it's wider than the others. 02:42
dalek kudo: 3d5359d | pmichaud++ | docs/ (2 files):
More text updates in preparation for release.
KyleHa I edited the two paragraphs about installation. I think it reads better, but you should check it for accuracy. 02:47
dalek kudo: c88002f | (Kyle Hasselbacher)++ | docs/announce/2009-08:
announce text edits
 02:48
kudo: bffd1ca | (Kyle Hasselbacher)++ | :
Merge branch 'master' of [email@hidden.address]
KyleHa pmichaud: Thanks so much for being here to help me with this. I appreciate you making yourself available on my schedule. 02:49
pmichaud my pleasure 02:50
actually, it's not so much that I'm adopting your schedule, as that this was the only time I'd have available to do it before the release anyway. :-)
but yes, I'm glad to be around to assist as needed :) 02:51
KyleHa Heh. OK, then I'm just lucky. 8-)
At this point, I think I'm waiting for your stamp of approval on the announce text before I make the tarball and test it. 02:53
pmichaud okay. I'm doing a README update at the moment to better explain "make install"
(I started to put the explanation in announce, then realized it belongs in README)
KyleHa No rush.
wayland76 if I have code that says "class Foo does Str {...} ; my Array of Foo @arr = func();", and func() returns a list of Str, will it create the Foo objects from the Strs, assuming that Foo.operator<=> works? 03:03
(ie. should it in Perl 6, even if it's NYI?) 03:04
pmichaud it will not.
KyleHa I think Foo.Str would have to work, but I'm half guessing.
pmichaud first, "my Array of Foo @arr" doesn't mean what you think it means :-) 03:05
it means that each element of @arr must be an Array
(in fact, it has to be an Array of Foo)
so you probably meant "my Foo @arr" if you want @arr to be an array of Foo objects 03:06
wayland76 That's what I meant, thanks :)
pmichaud second, "my Foo @arr" means that each element of @arr has to be an instance of Foo
since Str is not a Foo, you'll get type check exceptions with my Foo @arr = func();
the "Foo" in "my Foo @arr" is a type constraint, not a type coercion 03:07
03:07 justatheory left
wayland76 Ah. So is there some way to do a coercion? 03:07
pmichaud my guess would be my Foo @arr = (Foo($_) for func()); 03:08
wayland76 Or do I have to do my Foo @arr = map { new Foo($_) } func();
ok 03:09
So if we call a type name as a function, does that call "new" then?
s/new/constructors/ ?
pmichaud no, but there's a way to define a coercion
i.e., in class Foo you'd define how to build a Foo from a Str
wayland76 Ok, I'll look in the doco, thanks :) 03:10
pmichaud I'm not sure what that would look like.
Running "make install" will install Rakudo into the appropriate 03:17
locations, and those executables can be run from any directory.
(from the announce text)
I'm not sure that's quite specific enough.
KyleHa OK. 03:18
pmichaud fixing.
how about...
Running "make install" will install Rakudo and its libraries into the Parrot installation that was used to build it. The executables can then be run from any directory.
dalek kudo: 42d8ebd | pmichaud++ | README:
Some README refactorings.
kudo: 0e5e157 | pmichaud++ | :
Merge branch 'master' of [email@hidden.address]
KyleHa I'd make it one sentence. 03:19
(more)
Running "make install" will install Rakudo and its libraries into the Parrot installation that was used to build it, and then the executables will run from any directory. 03:20
pmichaud wfm
KyleHa Ultimately, though, I think this is turning to bikeshedding. 8-)
wayland76 pmichaud: Actually, if you don't define the conversion, it defaults to running "new"
pmichaud well, a phrase like "appropriate directories" is just asking for questions :-)
KyleHa I often overwork my sentences too. 03:21
pmichaud wayland76: that's in the spec? wow
03:21 frew__ joined
wayland76 S13:205 or so 03:22
Unfortunately, it call it as Dog.new($spot) instead of Dog.new(Parameter => $spot) 03:24
pmichaud that does seem to be an interesting discontinuity
dalek kudo: fd8fc8a | pmichaud++ | docs/announce/2009-08:
More announcement bikesh^H^H^H^H^H^Hupdates.
wayland76 Since I'm keen to join the bikeshedding, let me point out that ^W would've been better than ^H^H^H^H^H^H :) 03:25
pmichaud (sigh) 03:26
git commit --amend .....
:-)
KyleHa So there's notes about the install in the announce and the README. 03:28
pmichaud yes.
but the announce mainly points to the README
KyleHa Yep. Good good, I say.
pmichaud the README has notes before; I just rewrote them a bit.
*had
wayland76 README haz cheezeburgerz? :)
KyleHa Announce has "69.98%" because nines look better than zeros. 03:29
pmichaud Because we're >this< close to 70%. :-) 03:30
KyleHa *grin*
03:31 hercynium left
KyleHa OK, I'm going to make the release tarball out of this, test it, and do the trumpet work tomorrow. 03:31
pmichaud 70% is rounding up, so I didn't want that. 69.9 is honest, but makes people think "gee, they're almost at 70%". 69.98 says "okay, it really is 70%"
69.98 will also probably produce a few smarmy comments about how we're being overprecise or something like that, too. :-)
okay, assuming that we're still getting clean spectest runs, I can't think of anything else that has to be done before the release. It's all yours, KyleHa++ 03:33
and thanks a bunch for doing this release
KyleHa Thanks again for all your yelp.
s/yelp/help/
pmichaud glad to yelp anytime :-)
KyleHa Heh.
03:39 molaf left
KyleHa Builds OK, 'make test' works. make spectest running... 03:42
03:43 leedo left 03:44 leedo joined 03:51 TimToady joined
KyleHa spectests passed. Yay! 03:58
03:59 alester left
pugs_svn r28039 | ash++ | Removed a rakudo specific test skip in S12-class/inheritence.t 04:09
__ash__ yay, now make spectest passes 2 more tests
04:26 zloyrusskiy joined 04:38 mee joined 04:39 Zloy_Russkiy joined
PerlJam pmichaud: awake? 04:39
pmichaud awake. 04:40
just noticing that rakudo no longer builds against Parrot trunk.
PerlJam So ... I was just looking at the release instructions ... is there any reason it's not just "make release" instead of "make release VERSION=YYYY-MM" ? 04:41
or, rather than "instead of", how about "in addition to"
(i.e. VERSION=... would be optional)
pmichaud if you have an easy way to make that happen, I'd be okay with a patch to do it :-) 04:42
the main reason is that I didn't want to enforce VERSION=, and I wasn't sure how to make it optional in a makefile
KyleHa Release is done. 04:43
pmichaud checks
KyleHa I went ahead and announced and uploaded and all anyway. I'll be a zombie tomorrow.
I created a use.perl.org account just for this. 8-) 04:44
pmichaud good :) 04:45
looks awesome to me
KyleHa++
KyleHa Thank goodness. And thank you, pmichaud++ 04:46
G'night! 04:47
pmichaud Good night!
04:47 KyleHa left 04:54 Zloy_Russkiy left 04:56 __ash__ left 05:00 unitxt left
wayland76 Btw, does Rakudo * mean that we stop recommending people use github? 05:04
05:04 LionMadeOfLions left
PerlJam wayland76: er, why? 05:05
wayland76 Just wondering :)
Oh, and KyleHa++ even if he's left :) 05:06
05:09 lionWrk joined
pugs_svn r28040 | wayland++ | [S32/IO] and [S16] Changes include: 05:12
r28040 | wayland++ | * Removed spurious references to "Array of", now that pmichaud++ has explained things to
r28040 | wayland++ | me
r28040 | wayland++ | * Added a number of things to better cope with the Encoding
r28040 | wayland++ | * Other minor cleanups
05:15 nihiliad left 05:16 wayland76 left 05:17 dnukem joined, justatheory joined 05:23 frew__ left 05:25 justatheory left
PerlJam pmichaud: Try this patch gist.github.com/170857 05:26
05:27 ruoso left 05:41 orafu left, orafu joined 05:45 ihrd joined, wayland76 joined, tak11 left 05:48 tak11 joined 05:54 abra joined 05:56 solarion joined 05:59 finanalyst joined 06:00 youwin left 06:02 youwin joined 06:04 szabgab left 06:05 cognominal left
moritz_ rakudo: multi a ($x, $y?) { say "a" }; multi a ($x) { say "b" }; a(3); 06:25
p6eval rakudo 0d4fe0: OUTPUT«Ambiguous dispatch to multi 'a'. Ambiguous candidates had signatures:␤:(Any $x, Any $y?)␤:(Any $x)␤in Main (/tmp/RtdgIucQeV:2)␤»
06:28 cognominal joined
moritz_ I CAN HAZ RELEAZ! 06:28
KyleHa++ 06:29
06:37 youwin left
ihrd hi 06:39
06:39 youwin joined
ihrd I compile last Rakuso 06:39
argh, Rakudo, and find out I can run perl6 from rakudo dir only 06:40
from any other place it said
moritz_ ihrd: you have to install it
ihrd Null PMC access in isa()
current instr.: 'parrot;P6metaclass;register' pc 633 (runtime/parrot/library/P6object.pir:430)
called from Sub 'perl6;Perl6Role;onload' pc 5091 ((unknown file):-1)
called from Sub '_block138' pc -1 ((unknown file):-1)
moritz_ it's known and documented (see README)
ihrd ok, thank you 06:41
moritz_ (nobody reads READMEs anyway, we should call them READMENOTORYOULLDIEAVIOLENTDEATH) 06:42
ihrd I read it carefuly first time
dalek kudo: 48f6de1 | moritz++ | README:
remove trailing ws in README
ihrd but for now I just compile rakudo as usual 06:43
moritz_ yes, I understand
ihrd ok, I read it. Can we replace "You can now use 'make' to build Rakudo Perl." by "You can now use 'make' and 'make install' to build and install Rakudo Perl." instead of "You can now use 'make' to build Rakudo Perl." 06:48
moritz_ will do later on 06:49
07:03 rfordinal joined 07:07 rfordinal left
moritz_ rakudo: Perl6::Grammar.parse('1') && say $/.perl 07:08
p6eval rakudo 0d4fe0: OUTPUT«Method 'perl' not found for invocant of class 'Perl6;Grammar'␤»
moritz_ rakudo: Perl6::Grammar.parse('1') && say ~$/
p6eval rakudo 0d4fe0: OUTPUT«1␤»
07:09 barney joined 07:12 sarthak joined, sarthak left 07:14 Chillance joined 07:21 zamolxes left 07:30 eiro left 07:31 KatrinaTheLamia left, Patterner left, moritz_ left, jan__ left, gbacon left, moritz_ joined, KatrinaTheLamia joined, Patterner joined, jan__ joined, gbacon joined, irc.freenode.net sets mode: +o moritz_ 07:32 eiro joined 07:38 abra left 07:50 ihrd left 07:52 dduncan joined 07:59 zloyrusskiy left 08:00 zloyrusskiy joined
lisppaste3 dduncan pasted "Rakudo build problem - no HLLCompiler" at paste.lisp.org/display/85717 08:01
dduncan any idea why this is happening or how to get Rakudo to build? 08:03
pmichaud was Parrot installed with "make install-dev" ?
moritz_ was just about to ask that :-)
dduncan just make install, as the README says 08:04
moritz_ dduncan: the README from which century? :-)
pmichaud README says "make install-dev"
line 64
We should've put a check for that into Configure.pl. Oops. 08:05
it's on my todo list now.
(if anyone else beats me to it, great :-)
dduncan I see it now, in Rakudo's README, in the section mentioning --parrot-config, which I did use ... but Parrot's README makes no mention of install-dev, and it should, considering people who go and install Parrot first without looking at Rakudo 08:09
I will try re-installing Parrot using make install-dev now 08:10
pmichaud that's a very valid point as well.
Still, Rakudo's Configure.pl ought to catch this, so we should make that happen.
dduncan both are a good idea 08:11
pmichaud I'm a bit surprised that PCT/HLLCompiler.pbc isn't installed by "make install" even for Parrot, though. Almost nothing will run (much less build) without it.
according to Parrot's MANIFEST.generated file, it should've been installed. 08:12
dduncan now that I know install-dev exists, I can understand why that separation might be done ... perhaps analogous to Catalyst's separation into Catalyst-Runtime and Catalyst-Devel ?
pmichaud oh, wait, perhaps now. 08:13
*not
moritz_ will patch parrot's README
dduncan but that implies maybe pre-compiled HLL
pmichaud hmmm?
oh, yes, it would imply pre-compiled HLL. That's Parrot's working model of distribution. 08:14
It expects most people to obtain precompiled components, not to build them directly.
dduncan right, well, good to design in the separation early as possible, as it is done
just a documentation omission then
moritz_ documentation enhanced in (parrot) r40670 08:15
pmichaud very interesting... "make install" doesn't install the PCT components. That's bad.
moritz_ so even a precompiled rakudo needs install-dev? 08:16
pmichaud more to the point, parrot's "make install" should install the PCT components 08:17
on a different topic, I have no idea what to think about use.perl.org/comments.pl?sid=43556&cid=70186 . I think I'm just going to ignore it. 08:19
08:19 M_o_C joined
dduncan okay, using Parrot's "make install-dev" appears to have fixed my reported problem with Rakudo make 08:19
moritz_ pmichaud: ignoring it seems like a good option. 08:21
08:21 larsen joined
Matt-W pmichaud: It displays quite clearly that he didn't understand a word of what you were trying to say 08:22
moritz_ and it contains factual errors
"Perl 5 was a complete rewrite of Perl 4" -- wtf? that's the first time I read that
Matt-W If it was, it was only the interpreter 08:23
it wasn't a rewrite of the language itself
Matt-W ponders some sort of reply... but would need to get a use.perl account first, so is suffering a lack of enthusiasm 08:25
moritz_ Matt-W: I think it's not worth replying 08:26
it looks like a semi-informed flame bait with no intention to get any corrections, just food for flames
Matt-W I'm tempted to try
I can always stop
Shall think about it for a bit 08:27
pmichaud Actually, I think I have a reply. 08:30
08:30 ejs joined 08:31 broquaint joined 08:32 PZt left
dduncan good night 08:37
08:37 dduncan left 08:40 DanielC joined
Matt-W has a fairly rambling sort of reply he should probably try to conciseify 08:43
08:43 Chillance_ joined
pmichaud I just wrote mine: use.perl.org/comments.pl?sid=43556&cid=70194 08:43
DanielC Hi 08:44
Matt-W good one pmichaud
DanielC Reply to what? Did someone say something stupid?
Matt-W I was going to say some things along the same lines
along with some other stuff
I might just say the other stuff now
pmichaud DanielC: use.perl.org/comments.pl?sid=43556&cid=70186
DanielC *click*
cognominal explodes PGE : regex { <[x10000..xEFFFF]> } 08:47
moritz_ cognominal: maybe <-[\x0..\x10000]> is less explody? 08:48
pmichaud PGE doesn't know how to do ranges in enumerated character lists short of building the entire set of characters at the moment 08:49
cognominal I just wanted to match xml tags
pmichaud so the range \x10000..\xEFFFF is a little... large
cognominal indeed
:) 08:50
moritz_ wonders how hard it would be to use inversion lists for such ranges
pmichaud the tricky part is breaking up the character enumeration into smaller pieces 08:51
we also have to deal with things like \w, \s, \W, etc.
moritz_ well, I thought about doing it just for the special case of enumerated char classes 08:52
cognominal what is an inversion list?
DanielC pmichaud: That was a good response.
pmichaud DanielC: thanks. 08:53
cognominal www.ibm.com/developerworks/linux/li...cpinv.html
pmichaud I meant \w, \s, \W in enumerated lists
e.g., <-[\W]> 08:54
or
<-[\s 0..9]> # exclude whitespace and digits
or 08:55
<-[\s a e i 0..9 o u]> # exclude whitespace, digits, and lowercase vowels
moritz_ pmichaud: are \W etc. implemented by querying the unicode character database? 08:56
pmichaud at the moment PGE maintains a table of whitespace chars (more)
but for \d and \w we ask Parrot, and in the case of Unicode Parrot asks ICU 08:57
there's not a way in Parrot to say "enumerate all of the \w chars", there's just a way to say "is this character a \w". (Yes, we could loop through all possible codepoints asking if each one is a \w...) 08:58
moritz_ would you be interested in a perl based prototype which uses inversion lists + direct questions to a different implementation? 08:59
Matt-W cuts off his urge to ramble forever and posts a reply
moritz_ ie that would make <[$range]-\w> working by first doing an inversion list lookup, and then asking parrot if the character matches \w?
pmichaud moritz_: that sounds too special case-y to me 09:00
09:00 Chillance left
pmichaud it's not the case of <[$range] - \w> that is tricky on its own -- in fact, that's not even valid. 09:00
the tricky part of enumeration lists is that we have at least three things that can appear there: (1) individual characters, (2) ranges, and (3) certain character properties such as \w, \d, \s, etc. 09:01
moritz_ well, my idea might cover that 09:02
pmichaud so PGE needs to come up with a data structure that can represent the various combinations of these, including their negations
moritz_ inversion lists can represent (1), (2) and negations
pmichaud I think I can more efficiently represent (1) and (2) directly, without the inversion list 09:03
I can certainly evaluate a direct representation more efficiently. 09:04
moritz_ well, you certainly have more experience in that area than I do :-)
pmichaud :-)
thanks for the tip on inversion lists, though, I hadn't really seen that before
09:10 Chillance__ joined 09:12 tak11 left 09:27 Chillance_ left 09:30 meppl joined
frettled I get spectest failure on S12-attributes/class.rakudo and S14-roles/basic.rakudo, but I can't recall whether these were known or not. I've searched RT, but haven't found matching entries. New stuff or known? 09:33
09:37 ispy_ joined 09:51 Chillance__ left, Chillance joined 09:58 guest_007 joined 10:02 Whiteknight joined 10:08 Su-Shee joined
Su-Shee good day! :) 10:08
DanielC o/ 10:12
10:13 Whiteknight left
jnthn morning 10:14
Su-Shee started rakudo lobbying in the office today. 10:15
DanielC Su-Shee: Are you lobbying that your office start using Rakudo?
10:17 azawawi joined
azawawi std: "helllooo"; 10:17
p6eval std 28040: OUTPUT«ok 00:04 37m␤»
szbalint ;)
azawawi moritz_: hi 10:18
colomon rakudo.org/how-to-get-rakudo needs to be modified to add the "make install" bit.
Su-Shee DanielC: yes. 10:19
DanielC Su-Shee: What do you want to use it for?
azawawi moritz_: is rakudo going to be released today?
Su-Shee DanielC: in 2010, the entire system is scheduled for "throw away, write new" ;) 10:20
DanielC oh
Su-Shee: Do you think Rakudo is ready for that?
Su-Shee DanielC: essentially, web plus dbi plus heavy web service apis plus document converting.
DanielC Does Rakudo have those things already? (e.g. dbi) 10:21
Su-Shee DanielC: we'll see. even if we exchange minor modules with rakudo, it would be a nice start. we have some perfect candiate for grammars.
(for example)
azawawi 2009-08-20 Rakudo #20 "PDX" (kyle) 10:22
10:22 masak joined
masak hideloho, perl6ers. 10:22
DanielC o/
Su-Shee: I didn't think that Rakudo was ready for production use like that (heavy web service), it would be very exciting if you could make it work for that. 10:23
Su-Shee DanielC: well it's going to be a) a process and b) - we're planning for something which at least should run the next decade with perl. so, at least smaller things with rakudo doesn't seem to be impossible and then slowly migrating. 10:25
DanielC I see.
masak Su-Shee++ 10:26
DanielC Su-Shee++
Su-Shee well let's see how much of a success my lobbying really will be. ;)
DanielC What other alternatives is your office considering? 10:27
Su-Shee perl 5
DanielC Hmmm...
moritz_ oh hai
Su-Shee no different language, though.
moritz_ azawawi: in case you haven't noticed, it's already released
DanielC Difficult choice. Perl 5 has better features right now, but who knows where things will be in 10 years? 10:28
moritz_ different.
Su-Shee DanielC: well my crystal ball, my female intuition and the tealeaves earlier this morning said clearly "perl 6".
DanielC :-) 10:29
azawawi moritz_: cool thanks 10:31
Su-Shee wow. some people really hate perl 6.
DanielC Su-Shee: Q: Are you located in Germany?
Su-Shee: Your website URL is in German.
Su-Shee DanielC: yes
masak wow, this furry_marmot fellow sure sounds angry. 10:32
colomon Su-Shee: Is that a reference to people at your work, or furry-marmot?
Su-Shee colomon: neither. discussion on #perl right now.
colomon Oh.
DanielC What did they say? 10:33
Su-Shee "vapor6" "silimar name"
moritz_ I updated rakudo.org/how-to-get-rakudo to mention 'make install'
colomon moritz_++
KatrinaTheLamia yeah, I've been reading that discussion...
DanielC moritz_++
Su-Shee moritz_: ui, there's a make install now? cool.
KatrinaTheLamia how is libSDL support in Perl6? I have a project that I want to do that won't reach maturity anytime soon... figured I'd get it started in Perl6. 10:34
moritz_ it's actually mandatory now - a non-installed rakudo will only work from the root of the build directory
KatrinaTheLamia: I know there is some support in parrot for it, which might be usable from within rakudo 10:36
KatrinaTheLamia: actually there's a small Perl 6 example in the parrot repository, in examples/sdl/blue_rect.pl 10:37
KatrinaTheLamia moritz_, I'll look into it. I am really new to current Perl standards--I programmed for it long ago, but left for a while. From what I've been reading on various Enlightened Perl docs and blogs, it has came a long way... I really like some of the stuff some people are doing, and I want to be a part of it... so, you'll have to excuse me if I seem silly every now and then... it is purely ignorance on the matters at hand. 10:39
moritz_ KatrinaTheLamia: note that most "modern perl" blogs talk about perl 5
KatrinaTheLamia: both perl 5 and perl 6 has come a long way though :-)
Su-Shee: in case you haven't see it: perlgeek.de/blog-en/perl-6/svg-adve....writeback - does that count as "beautiful perl"? :-) 10:40
Su-Shee moritz_: pretty perl, yes. :) 10:41
moritz_ pretty, ok :-)
KatrinaTheLamia moritz_, well, I also have a few other projects I want to get done in Perl 5... including a few video games... however the Perl 6 game, I do not envision being near complete for some time, so I figured I'd do it in Perl 5
Perl 6*
Su-Shee KatrinaTheLamia: well you see the different views here and in #perl over there perfectly...
DanielC I find Perl 6 as a language far more interesting than Perl 5. 10:43
It has borrowed a lot of cool things from Haskell (functional language features) as well as Ruby/Smalltalk ("everything is an object")
yath but not .inspect :( 10:44
moritz_ KatrinaTheLamia: anyway, you're welcome to join our large quest/mmorpg :-)
DanielC My favorite languages were Perl 5, Haskell and Ruby. In some ways, Perl 6 is a combination of all of the above.
Matt-W Yup
That's how I see it
KatrinaTheLamia moritz_, nice. Thanks ^.^ 10:45
jnthn yath: .inspect?
moritz_ DanielC: would you be interesting in porting a cool ruby project to perl 6?
DanielC moritz_: Is there a specific one you have in mind?
yath jnthn: in smalltalk you can do "object inspect!" and see all doc-strings, methods and attributed of the object
(or the class)
moritz_ DanielC: yes, a plotting/charting module that produces nice SVG... let me find the link
yath s/attributed/attributes/
KatrinaTheLamia heh, the Perl 5 game I want to do is, ironically based on a forgotten web comic known as "Kid Radd". I think having a forgotten web comic be done well and get going in what is often (incorrectly) regarded as a forgotten language would be a good satire. 10:46
yath jnthn: or, to talk smalltalk-ish: every message an object instanciated from that class would recongnize ;-)
KatrinaTheLamia I'd make sure to make fun of the people like those in #perl regularly in the Kid Radd games I want to do... though, it would be rather tongue in cheek
reqamst rakudo source looks dirty for me... :( different filetypes... i can't find myself 10:47
DanielC moritz_: I haven't touched Ruby in years (I have to work with PHP - ugh) but I'm willing to give it a shot.
Matt-W yath: Well we've got metaclass methods for finding methods and attributes and things...
yath Matt-W: yeah, but inspect was really easy
Matt-W $a.^methods isn't? 10:48
yath Matt-W: just typed foo inspect! into the REPL and got a list of everything the object knows about
Matt-W: well, inspect also had the doc-strings along with the method :)
reqamst i think pugs is easier to recognize, but.. is it dead currently?
moritz_ DanielC: scruffy.rubyforge.org/ this one
DanielC *click*
moritz_ DanielC: I'd really like to help, but I don't know ruby at all 10:49
Matt-W yath: we'll have those according to Damian's latest doc ides, but they won't all come out of one method that's all
reqamst moritz_: how could i help with this project? 10:50
moritz_ reqamst: by translating ruby to perl 6, or explaining some ruby constructs to me, setting up a repo, thinking of a good name etc. 10:51
DanielC moritz_: I'm only a Ruby beginner, but I'll give it a shot. This library looks really neat.
reqamst: Do you know Ruby? 10:52
moritz_ DanielC: I looked at it, it's about 3k lines of code including blank lines and comments
and most classes are < 100 lines
DanielC ok
reqamst DanielC: yes, I know
DanielC Doesn't sound too bad.
reqamst: You are probably better at Ruby than I am. Do you want to try to port this library to Perl 6 with me? 10:53
masak I'm willing to help too, if someone else sets up the repo.
moritz_ setting up a repo is easy. Coming up a with good name is hard.
masak yeah... 'puffy' is not an immediately good candidate. :P 10:54
reqamst DanielC: we can try... I am discover source
KatrinaTheLamia I am not too bad with Ruby... even though I cannot stand RoR.
DanielC I'll try to think of a good name...
moritz_ once we've got a name (if only a provisional) I'll set up a repo
reqamst discovering
masak perhaps something to do with 'lickable (plots)'?
Su-Shee KatrinaTheLamia: why is that? isn't that what everybody wants today?
moritz_ plitable? 10:55
DanielC KatrinaTheLamia: You don't like RoR? Me neither. What I don't like about it is that it feels very restrictive.
moritz_ pletable?
masak moritz_: sounds like a table module.
KatrinaTheLamia Su-Shee, I tried to learn RoR, and found that the documentation on it, like most Ruby projects, is deplorable at best
DanielC RoR feels like wearing a straight jacket.
jnthn yath: While I think Perl 6 makes all that information available, I concede there isn't one nice method to get you all of it in one go.
KatrinaTheLamia you'd think with Ruby being highly associated with rdoc, the projects would be very well documented when released
DanielC It is really easy to do the very specific things that the makers intended, but really hard to do anything different and novel. 10:56
Matt-W yes that's the problem I had with it
masak 'graphsome'? :)
KatrinaTheLamia jnthn, what are some good Perl 6 resources? Maybe I should see about gathering as much as I can onto Nimh Labs.
Matt-W Catalyst seems more flexible, although at the expensive of initial complexity
DanielC I was surprised that RoR became so popular. I thought the Ruby language was supposed to be about flexibility.
moritz_ KatrinaTheLamia: we try to have links to nearly everything on perl6-projects.org/
KatrinaTheLamia moritz_, thankies ^.^ 10:57
Su-Shee Matt-W: I'm a Mojo fan, I have to admit.
KatrinaTheLamia moritz_, next time a trout to the face and telling me /topic will do ~.^
DanielC Su-Shee: What is Mojo?
masak KatrinaTheLamia: we're not big on slapping people's faces here. :)
moritz_ KatrinaTheLamia: we try to be friendly here, but I'll remember :-) 10:58
Su-Shee DanielC: CPAN -> install Mojo and mojolicious.org
masak we can make exceptions :)
DanielC *click*
jnthn KatrinaTheLamia: perl6-projects is, as moritz said, a good place. For keeping up to date plus seeing lots of good blog posts, planetsix.perl.org/ is worth watching.
10:58 dbrock left
Su-Shee DanielC: very tiny-tidy sleek mvc framework 10:58
DanielC Su-Shee: Sounds good.
reqamst DanielC: source looks simple 10:59
DanielC Mojo -> bookmarked
reqamst: That's good news.
I haven't looked at the source - I'm at work right now, I should work at least a little :) 11:00
moritz_ aye, it did look simple to me when I looked at it
moritz_ wonders if cardinal runs it 11:01
DanielC simple == good
KatrinaTheLamia thank you jnthn 11:02
masak all this _why nostalgia has made me want to check out Camping. :)
KatrinaTheLamia masak, moritz_ yeah, I mostly said that, as I realised how silly I was for not checking the topic 11:03
anyways, I'd love to join this MMORPG and be a well known blogger
masak KatrinaTheLamia: asking to get slapped with a trout is about the only way to get slapped with a trout around here.
KatrinaTheLamia: oh, you've read something. *blushes* 11:04
reqamst DanielC: and it's well documented
KatrinaTheLamia heh... though, it should be noted, when playing games, I tend to take the role of "loonie" of the four major arch types
DanielC reqamst: Yes. It looks well documented.
11:04 synth left
KatrinaTheLamia masak, well, I saw that article in my RSS reader some time ago ^.^ 11:04
11:04 synth joined
masak KatrinaTheLamia: I'm no expert. what's a "loonie"? 11:05
KatrinaTheLamia masak, I'll look for the old USENET article ^.^
frettled masak: seiyuu.com/okamoto/gaming/realmen.htm
KatrinaTheLamia: see above :)
KatrinaTheLamia oh thanks frettled ^.^ 11:06
KatrinaTheLamia clicks
it has been too long since I read that ^.^
masak KatrinaTheLamia: "The Loonie -- The type who will do anything for a cheap laugh, including casting a fireball at ground zero." I think you'll be a good Rakudo tester/bug submitter. :) 11:07
KatrinaTheLamia naw... I've got a _better_ project to be a tester/bug submitter... JPerl6 ~.^ 11:12
KatrinaTheLamia is shot
masak would that be Perl 6 for the JVM? 11:13
rakudo: /\ /
p6eval rakudo 0d4fe0: ( no output )
Matt-W that'd be cool... but hard
masak rakudo: /\ /; say "alive" 11:14
Matt-W: as opposed to other compiler projects around here? :P
p6eval rakudo 0d4fe0: OUTPUT«alive␤»
11:15 SmokeMachine left
KatrinaTheLamia I dunno, I'd mostly imagine it would largely be seen as an abomination of Perl6... though, now that I think of it, it would succeed at putting Perl6 into the enterprise world... the only issue I can foresee on my part is a large amount of lack of experience... also while I have read a few books on compiler design, my readings still have neglected the infamous Dragon Book... 11:19
which you know what? I have money... time to buy it ^.^
Matt-W KatrinaTheLamia: it wouldn't be an abomination. Perl 6 is supposed to be implemented in as many ways as people want!
And having that foot into the Java world would be nice
Someone was looking at doing it on .NET
Which would also be cool 11:20
masak very much so.
Matt-W It's all very well having Rakudo on Parrot, but running on an existing VM as well would be a huge boost 11:21
&
masak just to note, bleeding-edge Parrot/Rakudo produces segmentation faults and bus errors here on my box.
KatrinaTheLamia alright then... I'll look into it... however, it will likely be _really_ bad at first... as while I knew Perl5 during the Perl 5.6 and prior era... I don't really know current knowledge that well. Maybe I can use it to learn Perl6? I find implementing systems helps me best understand the very nature of how such a system works, as oppose to using it.
masak not matter what I run.
moritz_ masak: that's known, bacek will take care later today
DanielC It would be cool to have Perl running on the Davlik machine so it runs on Android. 11:22
masak moritz_: oh, good.
KatrinaTheLamia any suggestions on books to buy and places to get them? I've been neglecting getting that damn Dragon Book for far too long... is Amazon a good place to get that?
moritz_ KatrinaTheLamia: look at isbn.nu/ it has nice overviews of where to buy things 11:23
KatrinaTheLamia I cannot admit to being an expert in CS... but then, when I went to NAIT (the local standard on CS), I found the systems standards are exceedingly low... with me often hijacking the class from the teacher. But what I've read some _really_ nice things with this ^.^
moritz_, thankies ^.^
11:26 omega_ joined
masak omega_: o/ 11:26
11:27 omega left 11:28 rjh joined
reqamst can you recommend me any module as an example of good perl6 practices? 11:29
masak reqamst: anything in particular? module organisation? OO? comments? 11:31
reqamst especially OO 11:32
masak reqamst: on a general level, you should check out (the Perl 6 part of) proto. it's fairly nice. github.com/masak/proto/
for OO especially, check out Druid. github.com/masak/druid/
it even uses the Listener pattern, and is heavily commented with the newest S26 commenting praxis. 11:33
KatrinaTheLamia quietly notes both of those are masak own projects :P 11:35
11:35 ihrd joined
masak KatrinaTheLamia: I'd feel guilty about it, but try coming up with other examples of the same size -- I dare you. :) 11:35
KatrinaTheLamia masak, give me enough time to learn Perl6 better ~.^ 11:36
masak KatrinaTheLamia: you know, I like your attitude. :)
jnthn Heh. masak++ probably has written some of the most extensive Perl 6 code out there so far. :-)
And in the process weeded out the most bugs. :-)
masak I know STD.pm is pretty large, and massively impressive.
jnthn should actually try writing some Perl 6 code.
masak it requires its own compiler just to run. :) 11:37
Elf is also pretty large. I don't know much about its status.
the spectests, if one were to view them as some sort of application, constitute the largest Perl 6 application in the wild.
KatrinaTheLamia yeah... jnthn don't feel bad... I am still stuck in the tutorial mode... then it is off to do my mass genocide of Slimes in the world of Perl6 ^.^ 11:38
masak and then there's all the secrent government projects running on Perl 6 next-gen codebases, but which... I'm not at liberty to talk about, unfortunately. :P
noo! what have the Slimes done to you? all they do is sit around!
reqamst masak: thanks 11:39
KatrinaTheLamia heh... well, I did technically case a series of paradoxes that caused Perl 6 to come into reality after mentioning to the wrong people that Eris makes extensive use of Perl 6 in her works. So it doesn't surprise me that the Illuminatus make use of Perl6... I couldn't see any other ue ^.^
jnthn masak: If you consider Rakudo's setting...
masak: That's a non-small example too.
masak jnthn: also true.
KatrinaTheLamia cause*
masak KatrinaTheLamia: also, while Druid is completely written by me, I'll note that proto is written by a mosaic of Perl 6 authors. 11:40
ihrd hi there 11:41
masak ihrd: o/
ihrd jnthn: do you saw me emails?
KatrinaTheLamia masak, I am just teasing you on that... you have every right to be proud of your code
ihrd masak: 8) 11:42
masak KatrinaTheLamia: oh, I am. just trying to be fair about the attribution.
finanalyst Matt-W: how long did it take to get an account on use-perl?
reqamst hmm... it could be difficult to recognize when I made bug, and when it's rakudo bug... especially when I am not advanced in perl6
masak reqamst: yes. that's what #perl6 is for.
reqamst :)
jnthn ihrd: Yes, I saw 'em.
masak rakudo: ?/is this a bug/; say "nope" 11:43
p6eval rakudo 0d4fe0: OUTPUT«Use of uninitialized value␤nope␤»
masak oh.
jnthn lol
masak implicit comparison with $_
KatrinaTheLamia hands reqamst a kitten, "perl6 is dangerous to go alone, have this ^.^"
anyways, back to reading... so that maybe one day, I can be learned and smart ^.^ 11:44
masak that reminds me, I still haven't called a project of mine "Kittens"
reqamst KatrinaTheLamia: ^^
KatrinaTheLamia masak, if you do, I would love to see it as an integral part of Perl6. Mostly just so it could be advertised via that Legend of Zelda Meme. 11:46
masak :)
11:47 ruoso joined
masak I think I know which project to name that, then. coincidentally, I had trouble naming that project, so it suits me fine. 11:51
jnthn masak: Everytime I break it though, I'd feel bad about having killed a kitten. 11:54
moritz_ so, how do we name the Perl 6 port of scruffy? 11:57
plotty? 11:58
Su-Shee plotticilous ;)
jnthn PlotHead
Su-Shee *hihi* 11:59
DanielC But this is not a plotting library.
It is a generic SVG library, no?
Or is it really just for plotting?
moritz_ no, it's charting/plotting
DanielC ah
charty
Su-Shee pieplott
reqamst charty looks nice
moritz_ +1 for charty 12:00
DanielC just wrote the first word that crossed his mind...
Su-Shee chartyplott 12:01
plottychart
moritz_ plarty :-)
Su-Shee chotty! ;)
DanielC You could just forget about having the name mean anything and just name it after your favourite animal (e.g. lizzard).
reqamst cockatoo!
Su-Shee perlplott 12:02
reqamst perlchart
DanielC Well, that's certainly a unique name (cockatoo).
phart ;-) 12:03
masak I like 'charty' or 'chartie'.
commute &
moritz_ www.cockatoosoftware.com.au/
12:03 masak left
moritz_ okay, anybody against 'charty'? 12:03
DanielC no objections 12:04
moritz_ 3
2
1
Su-Shee moritz_: Open source Web Charting and Graphing using only CSS and Javascript.
moritz_ sold
Su-Shee already exists.
moritz_ ouch
Su-Shee charty -> sourcefroge
DanielC wallaby
platypus
there are a lot of animal names.
reqamst so chotty? 12:05
DanielC Is there any animal that is known for drawing?
(besides humans)
jnthn Humans?
Oh, damm.
DanielC heh
reqamst :)
Su-Shee DanielC: purple
I mean the snail which delivers purple.
and *hihi* the red lice :) 12:06
reqamst purple is part of pidgin
jnthn Squids have ink.
But Squid is already well know open source project.
reqamst yes
it's pigdin main library
DanielC purple sounds ok, even if pigdin uses it. 12:07
azawawi rakudo build failure on win32 xp gist.github.com/171023
DanielC there are many things called purple
moritz_ plurple :-)
DanielC graphene ? 12:08
12:08 takadonet joined
takadonet morning all 12:08
azawawi moritz_: any idea why rakudo build is failing gist.github.com/171023 ?
DanielC grapheme would be sort of a reference to Perl's linguistic roots. 12:09
KatrinaTheLamia moritz_, you could name it after something from lovecraftian fiction ^.^
moritz_ likes graphene
KatrinaTheLamia: haven't read any lovecraft, iirc
KatrinaTheLamia moritz_, heh, you should, it gets referenced a lot by geeks ^.^ 12:10
12:10 abra joined
DanielC Note: graphene and grapheme (n vs m) are different things. I spelled it wrong the first time. With an "n" it is something to do with chemistry. 12:10
moritz_ DanielC: actually my working group does lots of transport calculations on graphene :)
KatrinaTheLamia DanielC, hmm... are you suggesting we name it after various designer drugs then? 12:11
KatrinaTheLamia is trying hard to make a joke about that... but everything she can think of just ends up being extremely tasteless >.>
moritz_ there doesn't seem to be any software named 'graphene' yet
DanielC KatrinaTheLamia: I was actually thinking about grapheme, the smallest unit of written language.
A reference to Perl's linguistic roots. But I'm happy with graphene too. 12:12
KatrinaTheLamia DanielC, since it would be for graphing, that would actually be pretty good, to be honest... something to do with Escartes geography, or something like that
I dunno, I am being silly I guess
DanielC Yeah, I was thinking along those lines... grapheme/graphene sound like graphing.
12:15 nErVe joined
reqamst maybe look for name amongst famous artists? 12:16
moritz_ picassocharts?
DanielC I like graphene / grapheme
Su-Shee will surely get sued by paloma picasso.
12:16 ihrd left
KatrinaTheLamia naw... if we were to do an artist, we'd clearly need to go with something obscure, yet completely fitting for the subject 12:17
reqamst pastel?
moritz_ would be happy with graphene too
KatrinaTheLamia yeah, I like graphene too as well ^.^
Su-Shee moritz_: like the graphene transistor clocks? ;) 12:18
reqamst :D
12:18 cj left 12:19 cj joined
Su-Shee (and now we know why companies invent totally bogus and stupid product names ;) 12:19
12:19 cj left
reqamst scribble? 12:19
12:20 ilogger2 joined
Su-Shee reqamst: exist, is a text editor ;) 12:21
reqamst Tufte, cool!
azawawi tries to build rakudo on win32 after mingw32-make realclean 12:22
reqamst it's short, unique and have something common with charts ;) 12:23
DanielC Yeah.
moritz_ +1 12:24
12:24 huf left, mattp left, PerlJam left, arthur-- left, pjcj left, breinbaas left, hatseflats left, frettled left, colomon left, Lorn left, ilbot2 left, jisom left, mkelly32 left, ascent left, dalek left, ingy left 12:25 nicomen left, lucs left, andreasg_ left, spinclad left, he_ left, cxreg left, tarbo2_ left, p6eval left, edenc left, zetta left, c9s left, cosimo left, integral left, bigpresh_ left, krunen_ left, rafl left, shachaf left, Tene left, mberends left, Aisling left, sbp left, nsh left, Maddingue left, tomaw_ left 12:26 moritz_ joined, jrockway joined, obra_ joined, literal_ joined, nErVe joined, ruoso joined, rjh joined, omega_ joined, azawawi joined, Su-Shee joined, guest_007 joined, ispy_ joined, meppl joined, DanielC joined, broquaint joined, ejs joined, larsen joined, M_o_C joined, zloyrusskiy joined, eiro joined, gbacon joined, jan__ joined, Patterner joined, KatrinaTheLamia joined, barney joined, youwin joined, cognominal joined, solarion joined, wayland76 joined, dnukem joined, lionWrk joined, mee joined, TimToady joined, leedo joined, drbean joined, diakopter joined, gfldex joined, wolverian joined, BinGOs joined, mikehh joined, sri_kraih joined, simcop2387 joined, xomas_ joined, baest joined, cotto joined, reqamst joined, athomason joined, rhr_ joined, [particle] joined, irc.freenode.net sets mode: +o moritz_, japhb joined, frew joined, RonOreck joined, jantore joined, kent\n joined, yath joined, Gothmog_ joined, ewilhelm_ joined, kolibrie_ joined, gabiruh joined, cxreg joined, mattp joined, nsh joined, sbp joined, ascent joined, breinbaas joined, Aisling joined, Lorn joined, zetta joined, huf joined, lucs joined, arthur-- joined, hatseflats joined, spinclad joined, jisom joined, PerlJam joined, Tene joined, nicomen joined, shachaf joined, rafl joined, ingy joined, colomon joined, pjcj joined, andreasg_ joined, dalek joined, Maddingue joined, tomaw_ joined, tarbo2_ joined, ilbot2 joined, p6eval joined, mberends joined, frettled joined, irc.freenode.net sets mode: +oooo PerlJam Tene mberends frettled, he_ joined, integral joined, krunen_ joined, mkelly32 joined, c9s joined, cosimo joined, edenc joined, bigpresh_ joined, bloonix_ joined, sunnavy joined, zol joined, kst joined, cookys joined, c1sung joined, eldragon2 joined, hcchien joined, adhoc joined, Helios joined, Trey joined, lambdabot joined, clkao joined, awwaiid joined, charsbar joined, buubot joined, pnu joined, felipe joined, Grrrr joined, szbalint joined, Ingmar joined, pmichaud joined, jeekobu joined, IRSeekBot joined, cls_bsd joined, chipdude joined, sjohnson joined, skids joined, Woody2143 joined, phenny joined, michaelr joined, avar joined, allbery_b joined, silug joined, christine joined, patmat joined, irc.freenode.net sets mode: +oo pmichaud sjohnson
reqamst dufte tufte 12:26
Su-Shee in the north there's also "toefte" ;)
moritz_ (the u in 'dufte' is pronounce the the oo in cool)
anyway, tufte it is then
12:26 adhoc left, mattp left, [particle] left, mikehh left, rjh left, nErVe left, youwin left, huf left, cotto left, Helios left, Trey left, solarion left, hcchien left, reqamst left, lambdabot left, PerlJam left, arthur-- left
DanielC +1 12:26
12:27 PerlJam joined, nErVe joined, rjh joined, youwin joined, solarion joined, mikehh joined, cotto joined, reqamst joined, [particle] joined, mattp joined, huf joined, arthur-- joined, hcchien joined, adhoc joined, lambdabot joined, Trey joined, Helios joined, irc.freenode.net sets mode: +o PerlJam, yath left, diakopter left, pjcj left, chipdude left, hatseflats left, TimToady left, breinbaas left, IRSeekBot left
azawawi moritz_: it nows builds on win32...cool... it was a corrupted parrot directory... 12:27
12:28 TimToady joined, diakopter joined, yath joined, breinbaas joined, hatseflats joined, pjcj joined, IRSeekBot joined, chipdude joined, skids left, sjohnson left, jeekobu left, szbalint left, frettled left, colomon left, rhr_ left, Su-Shee left, buubot left, eldragon2 left, michaelr left, Lorn left, allbery_b left, larsen left, phenny left, lionWrk left, charsbar left, ruoso left, avar left, c1sung left, pnu left, clkao left, ilbot2 left, ewilhelm_ left 12:29 frettled joined, ruoso joined, Su-Shee joined, larsen joined, lionWrk joined, rhr_ joined, ewilhelm_ joined, Lorn joined, colomon joined, ilbot2 joined, c1sung joined, eldragon2 joined, clkao joined, charsbar joined, buubot joined, pnu joined, szbalint joined, jeekobu joined, sjohnson joined, skids joined, allbery_b joined, avar joined, michaelr joined, phenny joined, irc.freenode.net sets mode: +oo frettled sjohnson, Grrrr left, RonOreck left, athomason left, sri_kraih left, barney left, jisom left, mkelly32 left, pmichaud left, jrockway left, silug left, Woody2143 left, gfldex left, cls_bsd left, simcop2387 left, pmichaud joined, jrockway joined, barney joined, gfldex joined, sri_kraih joined, simcop2387 joined, athomason joined, RonOreck joined, jisom joined, mkelly32 joined, Grrrr joined, cls_bsd joined, Woody2143 joined, silug joined, irc.freenode.net sets mode: +o pmichaud
moritz_ github.com/moritz/tufte/tree/master 12:31
DanielC *click*
moritz_ anybody who wants commit access please give me your github ID
12:31 tomaw_ left
DanielC danielc 12:32
moritz_ reqamst: do you also have a github account?
12:32 jnthn joined
takadonet graphs..... no thanks moritz_ 12:33
DanielC masak sure does.
jnthn irc server fail
reqamst moritz_: yes, i have
moritz_: reqamst
moritz_ ok, masak, DanielC and reqamst are committers 12:34
DanielC Well, I'm off for lunch. 12:35
12:35 DanielC is now known as DanielC_away 12:37 tomaw_- joined 12:39 literal_ is now known as Hinrik, Hinrik is now known as literal 12:40 PZt joined 13:07 masak joined 13:08 Psyche^ joined 13:10 zloyrusskiy left
wayland76 KatrinaTheLamia: It looks like I'll have to fight you for the title of "channel clown" :) 13:13
13:14 unitxt joined
masak no need to make it a singleton role. :) 13:14
wayland76 swings at KatrinaTheLamia with a bladder full of whitewash on a string attached to a stick, missed, and the bladder his wayland76 in the back of the head.
s/his/hit 13:15
KatrinaTheLamia well wayland76 ... with every channel clown there is always two, no more, no less. Master and apprentice... 13:16
KatrinaTheLamia pulls out a rubber chicken with a pulley in the middle
in this case, I'll make an exception...
KatrinaTheLamia has quit on error exception, "apparently she fights like a cow" 13:17
whoops >.>
masak moritz_: great name. kudos.
wayland76 wonders out loud if he get seniority, completely unaware of any rubber chicken shenanigans :)
KatrinaTheLamia depends... do you speak Swahili? 13:19
wayland76 has not read any Lovecraft either, neither HP or the Compaq variety, but has played the Call of Cthulhu roleplaying game
wayland76 doesn't know whether he can speak Swahili or not, having never tried :)
KatrinaTheLamia well, we are in the same boat then 13:20
meh, we could always do something similar to DC... I'll be the channel's Harley Quinn and you can be Mr. J ~.^
wayland76 I had a book recommended to be personally by Larry Wall. Does that count? :)
moritz_ masak: not my idea, but thanks anyway
wayland76 discovers, after a quick Google, that not owning a television makes you miss references occasionally 13:21
KatrinaTheLamia: How about we take turns being the straight man :) 13:22
@karma KatrinaTheLamia
lambdabot KatrinaTheLamia has a karma of 0
wayland76 KatrinaTheLamia++
@karma KatrinaTheLamia
lambdabot KatrinaTheLamia has a karma of 1
KatrinaTheLamia I find the concept of being a "straight man" very hard to understand ^.^
wayland76 "The first hit is free, kid" :)
13:22 nErVe left
KatrinaTheLamia oh... does this mean I got to dress in horrid fashion and hang around in seedy neighbourhoods now, Mr. W? 13:23
Su-Shee popcorn anyone? ;)
13:23 Patterner left, Psyche^ is now known as Patterner, frederico joined
wayland76 KatrinaTheLamia: Metaphorically. And having been brought up sheltered/bookish, I'm afraid I speak pre-1960s English, so that when I say "man", you have to guess whether I mean "man", or "person" 13:23
The one neighbourhood that we expect you to frequent is #perl6 13:24
13:25 KyleHa joined
wayland76 But that's certainly not seedy -- I'm not the only person who has said something like "I like being on #perl6 because I'm the stupidest person there" :) 13:25
KatrinaTheLamia well... that "not being subjectated to the horrors of Tv" kind of puts you ahead of others in my books... though some of the cartoons in the 1990s were not too bad
wayland76 I saw Astro Boy a few times :)
Su-Shee who needs a tv to be up2date? ;)
wayland76 Su-Shee: Who needs to be up2date? :) 13:26
KatrinaTheLamia wayland76, which version of the Mighty Atom? The 1960s series? The 1980s series? or the 2003 series?
wayland76 KatrinaTheLamia: Whatever was on television in the 1980s :)
Su-Shee wayland76: tv meme sluts like me ;)
KatrinaTheLamia I dunno, the reason Mighty Atom was build kind of disturbs me... kind of a work around of the actual greaving process of the person who build him. 13:27
13:27 bpetering joined, bpetering is now known as missingthepoint
KatrinaTheLamia (the Mighty Atom, Astro Boy, was built to be modeled after his creator's son who had died in a car accident) 13:28
wayland76 only knows about Mighty Atom because he once read an article about how middle-aged Japanese car company executives really want to build the Mighty Atom = Astro Boy :)
13:28 araujo joined
KatrinaTheLamia wayland76, well, for proper clowning, look for episodes of the EWJ cartoon, the Sam and Max cartoon, Freakazoid, and oh shoot... I am forgetting another >.> 13:29
KyleHa "proper clowning"
KatrinaTheLamia also Reboot is fun to watch... but Season 3 and 4 got rather serious. However Season 4 is fun to parody. Who wouldn't want to be a Mainframe Neoviral?
wayland76 I was thinking more of "The Goon Show", a 1950s BBC Radio comedy that sometimes even had a script beforehand.
KatrinaTheLamia wayland76, see, and you have just out crazied me ^.^ 13:30
wayland76 Although I have only listened to the reruns :)
KatrinaTheLamia I love the Goon Show... why didn't it occur to me ^.^
mkelly32 freakazoid... heh. the entire internet from 1998 in his head
wayland76 KatrinaTheLamia: What country? 13:31
KatrinaTheLamia mkelly32, it could be worse. He could have Something Awful and 4chan in his head... but that wouldn't make for a very child friendly cartoon... I don't think even Heavy Metal magazine would dare run such a thing
masak are List and Array iterable in some sense? 13:32
moritz_ masak: yes
mkelly32 and ricardo montalbán as his nemesis...
moritz_ you do that every day, no?
masak moritz_: whiche sense is that?
moritz_ for @array { .. }
that iterates over @array
see S07 for the proposed low-level API
masak moritz_: so they have a .get method, yes?
moritz_: is there a role involved too? 13:33
and does List and Array do that role themselves, or is it Positional that does it for them?
wayland76 Definitely a role
KatrinaTheLamia Mr. W, I live in a Pirate Outpost in the Communist Terrorist Stronghold in the Frozen Arctic Tundra North of the 49th parallel... or at least that is what I gather my country is from American news... though they usually leave out the 49th Parallel bit... it helps when they don't give an actual location on the map.
mkelly32 they had a few star trek referennces ("ooey goey worms in your ear"), but i don't think they ever quite yelled "KAAAAHN!"
wayland76 It was designed by ruoso so it has to have a role :)
moritz_ masak: ruoso is the one to ask, if S07 isn't enlightening 13:34
masak tries S07 first, then ruoso
13:34 KyleHa left
wayland76 I did some S07 hacking, but most of the ideas came from ruoso 13:34
ruoso might be here... might not be here... who knows...
KatrinaTheLamia schroedinger's ruoso? 13:35
wayland76 KatrinaTheLamia: I just remember hearing something about John Lennon trying to explain the Goon Show to Americans, so I wondered (I'm in .au, in case anyone cares) :)
13:35 frew__ joined
wayland76 He may or may not have been, but he can't be any more -- we've observed him :) 13:36
13:36 jauaor joined
ruoso hopes nobody tries to observe me 13:37
KatrinaTheLamia heh... well, my mom is Orange Irish, and my dad spent most of his youth in Briton when grandpa was stationed there (technically Canada was still part of UK until like 1984, 1983)... so my parents had a lot of British influence. It just entered into me as I grew up.
wayland76 apologises to ruoso for dashing his hopes
KatrinaTheLamia well... now that ruoso has been observed... care to swing him around my his tail? 13:38
masak I don't quite get the answers I want from S07.
ruoso: OH HAI.
wayland76 KatrinaTheLamia: I have American parents, but grew up in .au, and picked up some British related culture somehow
ruoso X| 13:39
masak ruoso: if I promise not to observe you, can I ask some questions?
ruoso masak, you can ask'em... and I'll try to answer... but I'm quite in a hurry
masak ruoso: is the idea 'Array does Iterator' in some sense?
wayland76 does not swing ruoso, due to respecting people who are actually working on an implementation of Perl 6 :)
masak ruoso: would this work: my @array; my $first-item = $array.Iterator.get; 13:40
KatrinaTheLamia wayland76, well, from what I've seen of Aussies here in Canada, most of what Aussies say and do makes a lot more sense than what a lot of Americans suggest as being normal
ruoso yes... it would...
masak, but @array.get should work as well
except that @array.get will consume the @array
masak ruoso: what simple check can I do on an object to find out whether it'll behave like an iterator?
ruoso while @array.Iterator.get won't
KatrinaTheLamia wayland76, also, I think I am going to need to call you Bruce. Just to ease confusion around here.
masak $obj.can('get')?
13:41 wayland76 is now known as BruceLogicalPosi
ruoso masak, $obj ~~ Iterator 13:41
KatrinaTheLamia lmao
masak ruoso: thanks. that's all.
BruceLogicalPosi tivismBradman
There, now I got it all in
13:41 BruceLogicalPosi is now known as wayland76
KatrinaTheLamia Mr. W, care to remind #Perl6 of the faculty rules? 13:42
wayland76 KatrinaTheLamia: I would consider it, but that would probably contravene the #perl6 culture :) 13:43
KatrinaTheLamia lol 13:44
missingthepoint can I be Barry?
wayland76 Although strangely enough, there's an editor that (IIRC) does Perl 6 syntax highlighting called "Padre"
KatrinaTheLamia missingthepoint, well, as long as you don't try to claim you are the Tick
anyways... time to eat something... though, I didn't roll a 20 on kitchen appliances, so I cannot has waffles of +9/+9 delicious... so... I am think blueberry pancakes with maple syrup. 13:46
wayland76 wayland76 hates it when his server crashes due to being too cold (it's a weird bit of hardware)
missingthepoint hates it when BOSS's server crashes for no good reason
oh, wait. there is a good reason. IT'S RUNNING WINDOWS 13:47
moritz_ usually electronics run *better* when cool :-)
wayland76 moritz_: Not this one. Unless it's too warm to wear my jacket, the thing won't even boot. It's a reproducible problem :)
That's why I'm in the process of replacing it. Should be done in a few weeks 13:48
13:48 missingthepoint is now known as BarryCategorical
BarryCategorical ImperativeAblett 13:48
wayland76, we are similar, you and I 13:49
Su-Shee understanding other people's code is really tiresome after a few hours.
KatrinaTheLamia BarryCategorical, hmm... clearly it needs to run a real OS. I suggest until he learns better how to handle a computer, he runs Contiki OS on a good solid C=64
BarryCategorical, IMO, it would be a vast improvement
BarryCategorical the nick? :D 13:50
wayland76 is reading up on Barry now :)
BarryCategorical: No ContikiOS on C64
You've heard of C4, the explosive?
moritz_
.oO( my favourite :-)
 13:51
wayland76 Well, C-64 is 16 times as explosive
BarryCategorical is suddenly vastly scarder of moritz_ ...
wayland76 That's why they used it on the moon mission :)
KatrinaTheLamia I've had people scared I'd make C4 in High School... I will note, school life got a lot better for me after columbine ^.^
wayland76 somehow managed not to have too bad a school experience. Not sure how that happened, though 13:52
BarryCategorical ^.^
KatrinaTheLamia just realised that she is still an extremely young snake
moritz_ BarryCategorical: bombing over IRC is not yet ripe for a wide market
so don't worry
KatrinaTheLamia moritz_, yeah, that is why envelopes and anthrax were invented ^.^. They make great gifts ^.^
wayland76 lies on the floor, looking for carpet bombs, just in case :)
moritz_ KatrinaTheLamia: but they are also hard to deliver via IRC... 13:53
anyway, I should point out that this channel is publicly logged :-)
KatrinaTheLamia moritz_, meh, I've already long since been put on the no fly list
though, I do find that kind of retarded... there is one terrorist on a plane, and over fifty other people... I am sure the others can easily take him. 13:54
mkelly32 i used to be on that, too. at least, some M* Kelly was...
because, clearly, no terrorist would ever think to fly under an alias and use fake identification... 13:55
szbalint some people were joking at YAPC::EU that americans wouldn't name buildings C1 to C8
because someone might ask "where is C4?"
wayland76 No, that's N* Kelly. 13:56
BarryCategorical "it was liberated"
mkelly32 heh. that reminds me of a prank i heard of. someone got 4 pigs, labeled them "1", "2", "3", "5"... set them loose on a school campus. people spent all day trying to find pig #4.
KatrinaTheLamia I lived in an apartment C5... everybody kept going to apartment C4 to find me for some odd reason though >.>
moritz_ somehow that reminds me of the disk world where the rooms where numberd 5, 6, 7, 7b, 9, ...
wayland76 They don't let him on planes because he won't take off his metal armour to go through security :)
13:57 __ash__ joined
KatrinaTheLamia wayland76, I would have figured it was R. Kelly... but then that joke was old and stupid the first time it was made. 13:57
jauaor lives in C1 apt
KatrinaTheLamia didn't really get the whole R. Kelly case in the states... it really made no sense to her what so ever
jauaor :P
wayland76 missed the R. Kelly and M. Kelly references, but posts images.google.com.au/images?hl=en&a...mp;start=0 for people looking for the N Kelly reference 13:58
BarryCategorical Barry and Bruce inflict massive Australian culture on #perl6... 13:59
mkelly32 is getting hilighted too many times!
jauaor fame has a price
wayland76 remembers the difference between Australia and yoghurt
...
moritz_ just imagine how many times _why got hilighted :-)
wayland76 yoghurt has live culture :)
14:00 azawawi left
masak _why's disappearance has certainly made people appreciate the guy more. 14:01
moritz_ also liked his brother better when he was across the ocean 14:02
BarryCategorical wow, _why really disappeared. 14:03
14:03 frew__ left
wayland76 suddenly realises where he's seen the word "Lamia" before -- AD&D Mostrous Manual :) 14:04
14:05 sri_kraih_ joined
KatrinaTheLamia I am not that much of a monster... Hera just is jealous cause Zeus likes us better than her... hence a rather long slander campaign of the ages 14:05
part of why it is so easy for me to be a Discordian ^.^ 14:06
wayland76 Ah, yeah, you get that. Those Greek gods, I dunno. You and Arachne... :)
KatrinaTheLamia well, the biggest issue was that the Greeks, great blokes, couldn't really be trusted with history. They suffered from indigestion you see ^.^ 14:07
wayland76 wonders if he's missed the joke, unless it's that a bad past gives you bad history, and a bad repast gives you indigestion 14:08
KatrinaTheLamia wayland76, ! Ah, in which case, it is time you were enlightened, Eris style.
KatrinaTheLamia grabs the link
principiadiscordia.com/ << here we go wayland76 ... a treasure trove of enlightenment ^.^ 14:10
Tene KatrinaTheLamia: →← 14:12
14:12 ejs left 14:17 rfordinal joined 14:21 sri_kraih left 14:24 hercynium joined
KatrinaTheLamia cannot make out what Tene wrote... 14:26
Tene KatrinaTheLamia: in ascii, -><- 14:27
five-fingered hand of eris
KatrinaTheLamia ah ^.^
Tene brb, teaching.
KatrinaTheLamia yes, I am quite madly in love with Eris... she took my heart some time ago, and replaced it with a golden apple. I took that to mean I am very beautiful ^.^
Tene A very reasonable interpretation. 14:28
KatrinaTheLamia then all is lost ^.^
14:29 arthur-- left, arthur-- joined 14:30 arthur-- is now known as arthur-_
jnthn back 14:30
KatrinaTheLamia wb jnthn 14:31
wayland76 jnthn: From the time I challenge KatrinaTheLamia for the title of channel clown, there's nothing worth backlogging
14:31 Psyche^ joined
wayland76 (unless you count that ruoso said that Array does Iterator) 14:31
KatrinaTheLamia wayland76, such is the nature of challenging somebody for channel clown ^.^ 14:32
jnthn I already knew that. ;-)
jnthn skips backlogging
KatrinaTheLamia jnthn, also ruoso has been observed, that may be of note too ^.^
14:32 BarryCategorical is now known as missingthepoint
wayland76 KatrinaTheLamia: Yeah, I know. But I wanted jnthn to have some free time to do coding, since he's one of the two core developers of Rakudo Perl 6 14:32
masak despite specifically hoping not to be observed. 14:33
wayland76 (unless I've missed another core developer)
__ash__ i have a question about class bodies that i am wondering if it some people may, are they supposed to be executable in the sense of how ruby's class bodies are executable?
s/may/may know/
moritz_ __ash__: class bodies are executed at compile time 14:34
__ash__: and during execution they construct the class
KatrinaTheLamia hmm... I should prolly head on down to my therapists... he may be worried about me... 14:35
szbalint (short circuiting backlogging)++
KatrinaTheLamia hmmm 14:36
(-><-)++
wayland76 Now for a demo:
@karma (-><-) 14:37
KatrinaTheLamia @karma -><-
lambdabot (-><-) has a karma of 1
-><- has a karma of 0
wayland76 @karma wayland
lambdabot wayland has a karma of 125
KatrinaTheLamia hmm... well atleast it is not 1006... prolly about the level of half a Radita.
Raditz*
14:38 rfordinal left
wayland76 (afk, back in 5) 14:38
14:39 Patterner left, Psyche^ is now known as Patterner
pmichaud Good morning, #perl6 14:39
jnthn lolitspmichaud
14:40 nErVe joined 14:41 Su-Shee left, azawawi joined
moritz_ oh hai 14:43
azawawi moritz_: any idea what PERL6HERE is doing in Cursor.pmc and STD.pm ? 14:44
14:44 molaf joined
moritz_ azawawi: no. 14:44
azawawi moritz_: the default value is a strange utf-8 that is showing as a block here
14:45 KyleHa joined
moritz_ then you know more than I do 14:45
azawawi :)
14:46 azawawi left 14:52 DanielC_away is now known as DanielC 14:53 mikehh_ joined 14:54 payload joined
__ash__ jnthn: if I were to try to get $a.Role::foo(); to work, would that be a daunting task you think? 14:55
14:56 nihiliad joined
jnthn __ash__: I was pondering hacking the punner. 14:57
Though was wondering if I might be able to think of something cleaner.
__ash__ whats the punner? 14:58
wayland76 'night all. 1am means it's after my bedtime 14:59
14:59 cotto left
wayland76 __ash__: When a role is made to work like a class, we say it's "punned" as a class. Does that help? 14:59
__ash__ okay, is that when it makes a blank object and applies the role to it? kinda like calling new on a role? 15:00
15:05 zloyrusskiy joined
jnthn __ash__: Yes, pretty much 15:06
Well, .new on the role puns it and instantiates the pun.
Basically, find_method on a role hands back an invokable that puns and re-dispatches. 15:07
We may be able to handle this by detecting in that invokable (I think it's defined in src/classes/Role.pir) that we haven't actually got a role as the invocant...
And then do something from there.
It'll probably take a little experimenting. 15:08
15:08 zloyrusskiy left 15:09 mikehh left 15:10 zloyrusskiy joined, cotto joined
__ash__ in the !select method? 15:11
jnthn __ash__: No 15:12
Right at the bottom
See P6opaque.pmc for the name...I forget it off hand (sorry, a bit distracted dealing with some other bits...) 15:13
dalek kudo: e0820fa | pmichaud++ | build/Makefile.in:
Patch to allow Windows to link with ICU.
__ash__ no worries, i am not trying to keep you from getting your work done, i might play around in the Role.pir file and see if i can get it to explode, or work, probably explode though 15:14
15:14 zetta left 15:20 rfordinal joined 15:21 justatheory joined 15:22 payload left 15:27 eldragon2_ joined 15:28 eldragon2 left, cj joined
cj diakopter: any news on ironperl? :) 15:29
15:29 eldragon2_ is now known as eldragon2
jnthn __ash__: OK, I can concentrate on Rakudo again now. :-) 15:33
__ash__: It's .sub '!pun_helper' that I was meaning.
__ash__ kk, i saw that one 15:34
jnthn Hmm. Looking it now I'm not quite so sure though. 15:35
15:37 nErVe left
__ash__ i think right now its the !select that throws the error 15:38
jnthn What is the error again?
is it that it can't call !select?
__ash__ Method '!select' not found for invocant of class 'Foo'
jnthn Right.
That's because normally you do TheRole.meth 15:39
And so the first parameter is TheRole
Whereas here it's not.
__ash__ i am not sure if this is related, or something to worry about right now, but indirectly calling a role method (like how you can with a class) doesn't work either 15:40
it gives invoke() not implemented in class 'Undef' as its error 15:41
jnthn __ash__: Code example?
moritz_ rakudo: role A { method foo { 'bar' } }; class B does A { }; B.new.'foo'.say
p6eval rakudo 0d4fe0: OUTPUT«bar␤»
moritz_ seems to work.
moritz_ wonders if we have tests for that
jnthn moritz_: I dunno. We don't have a test for that glorious hyper + .+ + candidate list dispatch I demonstrated yesterday either though ;-) 15:42
__ash__ rakudo: role A { method foo { 'bar' } }; class B does A { }; my B $b .= new; A::foo $b;
p6eval rakudo 0d4fe0: OUTPUT«Null PMC access in invoke()␤in Main (/tmp/BPUPLKpzHd:2)␤»
jnthn __ash__: Oh, that.
I'm not convinced that should work..
What if it's a parametric role, and all that... 15:43
15:43 carlin_ joined
moritz_ uhm, that's trying to call a method as a sub 15:43
don't think that's supported at all
__ash__ oh, okay, i wasn't sure if it was in the spec or not, i just assumed it worked for classes, and since a role method is essenctially a class method i figured it would function similiarly
jnthn moritz_: It is, and it's fine in a class. 15:44
Well the thing is that in a sense roles are always parametric...
__ash__ rakudo: role A { method foo { 'bar' } }; class B does A { }; my B $b .= new; B::foo $b;
p6eval rakudo 0d4fe0: OUTPUT«Null PMC access in invoke()␤in Main (/tmp/KDenz5Ocas:2)␤»
jnthn __ash__: Right, but B::foo does not exist. 15:45
(We shouldn't Null PMC Access, but we certainly should not find anything there.)
Anyway, back to the task at hand...
I'm tempted as a first cut to, in P6opaque, attach the role to the clone'd !pun_helper just as we attach the name now. 15:46
That'd get us some of the way. 15:47
Then after that it can be handled in !pun_helper. 15:48
15:48 moritz_ sets mode: +oo jnthn __ash__, jnthn is now known as wesome_i_haz_op
wesome_i_haz_op woo 15:48
15:48 wesome_i_haz_op is now known as jnthn
moritz_ prepared for a *really* bad programming joke? ircz.de/static/pics/2009_06_29_19_4...ore017.gif 15:49
TimToady i can haz op 2?
phenny TimToady: 19 Aug 10:21Z <wayland76> tell TimToady Next time I mess up the documentation, feel free to tell me to fix it. If I'm in documentation mode, I can usually find the time
15:50 moritz_ sets mode: +o TimToady
moritz_ TEHR U GO 15:50
TimToady kthnx
jnthn moritz_: That's...terrible. 15:51
__ash__ i didn't see it coming, then it happened... i just died a little inside... *now to forward to all my programming buddies* 15:52
15:52 zloyrusskiy left, jferrero joined
jnthn OK, so if I'm in a branch in git and I want to pull the latest changes in from master 15:54
I just do git rebase master ? 15:55
masak yup.
jnthn++
moritz_ and cross your fingers :-)
pmichaud TimToady: ...did you learn any new advanced hieroglyphics that we can stick into Perl 6? ;-) 15:56
jnthn masak: omg it worked!
;-)
15:56 dnukem left, cdarroch joined
masak jnthn: I know! rebase is 21th century technology, I tell you! 15:56
jnthn continues happily breaking stuff in his branch.
TimToady pmichaud: I want opening and closing cartouche characters now for brackets 15:58
15:58 payload joined
masak adds a (hopefully) motivating SYNOPSIS to the Tufte README 16:00
KyleHa What's a cartouche? Something to do with automobiles? 16:01
moritz_ masak++ # Just Ducking Foo It
masak :>
DanielC Can I get the Tufte URL again please? I forgot to bookmark it. 16:02
KyleHa I'm thinking "back bumper".
moritz_ DanielC: github.com/moritz/tufte/tree/master
jnthn Tufte? 16:03
DanielC *click*
*bookmark*
16:03 hercynium left
jnthn oh, that's what it ended up getting called 16:03
moritz_ right.
jnthn Well, it's better than PlotHead. ;-)
moritz_ funny thing is that Mr. Tufte seems to criticize the way normal charts/plots are done
like we want to produce them :-) 16:04
DanielC jnthn: en.wikipedia.org/wiki/Edward_Tufte
masak moritz_: well, let's try to make him at least a little proud. :)
moritz_ masak: aye
KyleHa I like PlotHead, but I might have gone with PlotTwist.
moritz_ heh, SVG supports twisting coordinate transformations :-) 16:05
or something like that
DanielC If we ever find another name we like better we can change name.
KyleHa Maybe you could get some US Govt attention by calling it TerroristPlot. 16:06
reqamst yes, name isin't the most important
DanielC KyleHa: I once heard about an airport announcement looking for a guy called Al Kyda. 16:07
moritz_ right, beautiful charts are
KyleHa DanielC: I'd laugh if it weren't plausible. 16:08
DanielC KyleHa: I don't think the guy was in trouble. I got the impression that it was just a funny coincidence. But I don't know. 16:09
16:10 M_o_C left
DanielC masak: Looking at the README... what do the colons mean? As in :width => 300 16:10
masak oops, did I miss those? :/
masak checks
reqamst ?
masak yup, thanks.
DanielC Ahh... they are a Rubyism?
16:11 payload left
mkelly32 does => not auto-quote the LHS anymore? 16:11
masak DanielC: sort of... I took the Ruby code and translated it. missed those colons.
fixed.
DanielC++
DanielC k
@karma 16:12
lambdabot You have a karma of 21
masak the previous code would parse, but probably NDWYM.
16:12 masak left
__ash__ jnthn: oh, i found where i got the impression you should be able to indirectly call role methods, in S12, methods, indirect notation, but i just realized its a different syntax than what i was referring to, it would be more like Role::method $obj: ; (forgot about the : in the indirect notation) 16:15
well, theoretically, if you can do $obj.Role::method(); then according to that Role::method $obj: ; should work also
jnthn __ash__: Ah, that is different 'cus I think we essentially emit the same code. 16:16
That is, Role::method $obj: ; should probably result in the same PAST as $obj.Role::method
__ash__ yeah, i see that now, i forgot about the : making it indirect
KyleHa @karma 16:17
lambdabot You have a karma of 40
KyleHa @karma kyle
lambdabot kyle has a karma of 256
KyleHa Nyuck nyuck nyuck.
16:21 alester joined, DanielC left
__ash__ well, i have to head to class, jnthn i am going to poke around in the Role.pir to see if i can learn anything more about how role's methods work and how to get $a.Role::foo() to work 16:35
jnthn __ash__++: great! 16:37
Have fun. ;-)
missingthepoint DanielC: TV show, here... "Chaser's War on Everything" :) 16:38
16:38 __ash__ left
missingthepoint Rubyism: :foo is a symbol 16:39
that is, ":foo" refers to a Symbol object 16:40
16:40 szabgab joined
missingthepoint and every instance of :foo in your code refers to the same symbol object 16:40
rationale: less memory usage, apparently
(not sure i buy that)
mkelly32 well, each :foo is just a reference to some global :foo. whereas each "foo" is a reference to a different "foo", as i understand it 16:43
missingthepoint masak++ # naming, i can haz tufte book amazon?
TimToady better to just keep track of which things are mutable/immutable and let the system decide whether to coalesce immutables 16:44
missingthepoint TimToady: why let the system decide as opposed to the user? 16:45
TimToady because it's really "make", not let 16:46
and the system is the slave, not the user
mkelly32 well, i think "foo" is mutable in ruby, isn't it? don't think they have separate String and StringBuffer or whatever, like Java
missingthepoint well, amen to the second point... is that all? i'm interested in the semantics of similar stuff in perl 6, seeing i'm supposed to be helping masak with his scruffy port :) 16:47
TimToady S02 takes a first crack at documenting which types are mutable and which aren't 16:48
16:51 lionWrk is now known as LionMadeOfLions
Tene Sup dawg, I heard you like lions. 16:51
missingthepoint Tene: ? :) 16:53
16:54 szabgab left
Tene missingthepoint: "10:51 -!- lionWrk is now known as LionMadeOfLions" 16:57
I made a reference to an INTERNETS MEME. The implied continuation of my statement was "So I put a lion in your lion." 16:58
missingthepoint oh i missed that one. where's the meme school?
(waiting for someone to say 'here')
moritz_ it's called 4chan 16:59
jnthn How is meme formed? How man get knowledge?
missingthepoint moritz NEWLY updated and jnthn. you have both nearly made me spit coke over my keyboard.
(moritz_: java joke.) 17:00
Tene lol'd.
17:08 nihiliad left, nihiliad joined
mberends missingthepoint missed a point. typical. 17:09
missingthepoint mberends doesn't know how right he is... :P
17:10 nihiliad left, nihiliad joined
reqamst I first time compiled rakudo and it's sloooow. :( 17:11
moritz_ scruffy makes heavy use of delegation, including lvalues
moritz_ wonders if rakudo supports that
jnthn moritz_: Delegation as in, handles? 17:12
moritz_ jnthn: right
jnthn Rakudo support for handles is quite featureful.
And tested.
reqamst there will be some hard optimisations?
moritz_ even for lvalues?
reqamst: there will.
jnthn But I'm not sure it's had enough abuse^W yet.
moritz_: I'm not sure if that's any different that lvalue routines in general? 17:13
moritz_ tests
17:13 nErVe joined
[particle] reqamst: yes, someday 17:13
moritz_ Null PMC access in find_method() 17:14
jnthn reqamst: There are various bottlenecks. One of them is calling / invocation.
takadonet reqamst: They are mostly working on features first
17:14 nErVe left
jnthn reqamst: That particular area will improve in the not too distant future. 17:14
But there's still plenty of others.
17:14 zloyrusskiy joined
jnthn moritz_: Heh, file rakudobug ;-) 17:14
moritz_ jnthn: yes, but I'll commit tests first 17:15
reqamst Good to know. :) Today I need 10 seconds to run 'say "hello"'
moritz_ reqamst: that *is* slow - 1.6s on my machine 17:16
missingthepoint reqamst: just curious, what's your hardward like? :)
reqamst yes, my testmachine is old (celeron 900mhz, 256ddr ram), but...
jnthn < 1s on mine...
moritz_ reqamst: you can also try to configure parrot with --optimize
reqamst moritz_: oh, I will try
moritz_ tries that too 17:17
reqamst it's pure arch linux, without X
jnthn almost gets the container traits refactor working... 17:24
moritz_ reqamst: with optimized parrot it's "only" 1.1s here (down from 1.6) 17:28
missingthepoint rakudo: say 10 * (1.1 / 1.6) 17:30
p6eval rakudo 0d4fe0: OUTPUT«6.875␤»
reqamst this could mean 5sec faster for me 17:35
and getting rakudo usable 17:36
moritz_ jnthn: it turns out lvalue method delegation *does* work... 17:37
jnthn: however there's something fishy with initializing the attribute on which the 'handles' is defined
actually it's nto related to 'handles' at all ;-) 17:38
jnthn moritz_: Ah, OK. 17:39
jnthn wonders if this is good or bad. :-)
moritz_ rakudo: class A { }; class B { has A $.x; submethod BUILD { $!x .= new() } }; say B.new.x.WHAT
p6eval rakudo 0d4fe0: OUTPUT«Failure()␤»
moritz_ jnthn: that should be A() 17:40
rakudo: class A { }; class B { has A $.x; submethod BUILD { $!x .= new() } }; say B.new.x.defined
p6eval rakudo 0d4fe0: OUTPUT«0␤»
jnthn moritz_: Ah, that bug. 17:41
moritz_ so it's know=
?
and ticketed?
jnthn I'm pretty sure I saw a ticket related to it. 17:42
moritz_ ok
17:42 pugs_svn joined
pugs_svn r28042 | moritz++ | [t/spec] simplify and unfudge lvalue delegation tests 17:45
reqamst o, pugs is alive 17:46
frew reqamst: that's the spectest
not really pugs itself
reqamst mhm
moritz_ it just lives in the pugs repo
reqamst why is it died? 17:47
people don't like haskell?
moritz_ because audreyt++ got ill and had to re-focus after her recovery
missingthepoint how is she doing? 17:52
moritz_ I think she's doing fine 17:53
but she doesn't do so much publicly exposed hacking anymore (which caused lots of stress, I suppose)
jnthn rakudo: my Int $x = 4.2; 17:55
rakudo: my Int $x = 4; $x = 4.2;
p6eval rakudo 0d4fe0: OUTPUT«Assignment type check failed; expected Int, but got Num␤in Main (/tmp/gxAZpxEcZj:2)␤» 17:56
rakudo 0d4fe0: OUTPUT«Assignment type check failed; expected Int, but got Num␤in Main (/tmp/IqZiKWkP2z:2)␤»
missingthepoint ok, good to hear. tis a shame when a community loses someone of that skill level... look what just happened with _why
17:56 alester left
moritz_ aye 17:57
she does continue to contribute to perl - just look at her CPAN dir
pmichaud back from lunch 17:59
missingthepoint aye.
18:00 KyleHa1 joined, KyleHa left
missingthepoint oh hai pmichaud. 18:02
18:04 arthur-_ left, araujo left, unitxt left, ascent left, obra_ left, omega_ left, broquaint left, dalek left, ingy left, arthur-_ joined, araujo joined, unitxt joined, obra_ joined, omega_ joined, broquaint joined, ascent joined, ingy joined, dalek joined 18:05 rfordinal left, he_ left, cxreg left, wayland76 left, literal left, tarbo2_ left, guest_007 left, p6eval left 18:06 rfordinal joined, literal joined, guest_007 joined, wayland76 joined, cxreg joined, tarbo2_ joined, p6eval joined, he_ joined
jnthn gah, so close and yet so far 18:07
ok, dinner break
18:13 simcop2387 left
nicomen Solved the remaining issues with perl6 here, but forgot to write it down! :-) my.opera.com/Wargamez/albums/showpi...bum=859407 18:23
missingthepoint pmichaud: did you see the progress on csv? 18:31
18:32 simcop2387 joined
mikehh_ rakudo (e0820fa) builds on parrot r40676 - make test PASS, make spectest (up to 28042) 3 tests fail - Ubuntu 9.04 amd64 (gcc) 18:33
t/spec/S06-advanced_subroutine_features/wrap.rakudo fails with Lexical '@log' not found after ok 6 - wrapper after
t/spec/S04-declarations/my.rakudo - Failed test: 48 18:34
t/spec/S29-context/eval.rakudo - Failed test: 1
18:36 pmurias joined
pmurias hi 18:36
lambdabot pmurias: You have 1 new message. '/msg lambdabot @messages' to read it.
18:38 szabgab joined
pmurias ruoso: re including smop test in p5 build i haven't tried it yet, but i'll look into it 18:38
18:42 M_o_C joined 18:45 masak joined, zloyrusskiy left
missingthepoint pmichaud: github.com/bpetering/Text--CSV/blob...rogress.pl 18:45
moritz_ missingthepoint: we should plug that into SVG::Plot somehow to generate the graphics in Perl 6 18:46
which means I have to patch it to support multiple data sets ;-) 18:47
missingthepoint ;) 18:48
PerlJam is sleepy today
moritz_ missingthepoint: each step closer to self-hosting is a success 18:49
missingthepoint moritz_: indeed.
18:50 [particle] left
missingthepoint dunno what you think of ray kurzweil, but i just watched a talk of his - "15 years and 31 days" 18:50
18:50 zloyrusskiy joined
moritz_ doesn't know him at all 18:50
18:51 [particle] joined
missingthepoint "the singularity is near"... ? 18:51
masak and Perl 6 is the language which will get us there.
missingthepoint masak++, a man after my own heart :P 18:52
masak missingthepoint: do you happen to have a URL?
moritz_ missingthepoint++ # nice work with Text::CSV
masak waitwait, there's a Text::CSV now? :)
moritz_ github.com/bpetering/Text--CSV 18:53
masak missingthepoint: we should probably merge our projects. :)
missingthepoint www.ted.com/index.php/talks/ray_ku...rm_us.html
masak thanks.
missingthepoint obvious parallels to perl 6. :) 18:54
masak ah. yes, I saw that talk not so long ago. 18:55
moritz_ missingthepoint: it would be nice if you placed the modules into a lib/ directory, which is quasi standard for perl modules
missingthepoint masak: key phrase for your blogging: "15 years and 31 days"
masak I think I missed when he said that, though. 18:56
missingthepoint moritz_: true... i intend a bit of work on T::CSV soon :)
masak: about sequencing HIV versus SARS 18:57
masak missingthepoint: oh, aha.
I recall.
18:57 rfordinal left
missingthepoint masak: you have an amazing memory. how long ago did you watch that? o 18:57
masak yes, there are obvious parallels to Perl 6 there, indeed.
missingthepoint ... _O
masak missingthepoint: perhaps two weeks ago. 18:58
missingthepoint _O
masak "dear diary. today we invented the continuation-based smiley on #perl6..." 18:59
18:59 payload joined
missingthepoint nono, you're making me laugh, and ppl are sleeping! :) 18:59
masak missingthepoint: now you've just made it a challenge instead. :P 19:00
moritz_ still admires the VVLLSS LOLSPK
masak isn't that 'VWLLSS'?
and 'LLSPK'? :)
moritz_ you know that lolcats can't spell ;-)
masak oh, right.
LOLICANTSPEL
missingthepoint lolcats spell excellently in a non-standard english orthography. 19:01
masak I wish I could believe that. 19:02
moritz_ if you transform the coordinate system all lines are straight ;-)
missingthepoint moritz_: NOT YOU TOO!?!?!
masak there are many findings about Lolspeak that I've made reading the Lolcat Bible, that I haven't blogged about or otherwise shared.
one is that (surprise, surprise) the Lolcat Bible is not a quality work. 19:03
dalek kudo: 9716d3f | pmichaud++ | docs/spectest-progress.csv:
spectest-progress.csv update: 431 files, 12371 passing, 0 failing
masak and that's understating things a bit.
missingthepoint masak: i used to work in a christian bookstore. there's an "Aussie bible", with, well.... predictable content :D
masak oh my. 19:04
doesn't the original bible contain enough predictable content?
missingthepoint breaks uncomfortable silence... 19:07
... so you were making me laugh? :)
masak yeah, but I forgot how. 19:08
let's talk about our duplicated efforts instead.
missingthepoint done.
so i have some thoughts about Perl 6 csv parsing. 19:09
masak I'll have a look at your Text::CSV, and you can look at my CSV, and then we can merge them, preferably in your repository.
missingthepoint why mine?
yours is more "central"
masak I doubt that.
moritz_ on a surface of a sphere nothiing is more central ;-) 19:10
missingthepoint oi!
masak I did things in the wrong order, I implemented CSV first, and then started studying how CSV worked.
jnthn I've done that with Perl 6 occasionally.
;-) 19:11
missingthepoint well, i went extra-dimensional: started porting Text::CSV, then learned how it worked, then learned enough perl6 to do so :S
masak missingthepoint: wow, you already have more code than I. :)
missingthepoint moritz_: "oi" is Australian for "hey you, stop that" :) 19:13
pronounced like "oy" in "oy vey"
PerlJam likes the methaphors clothed_field and nude_field 19:14
good names
moritz_ missingthepoint: telling me to stop shilling semi-scientific bullshit won't do you any good ;-)
masak for some reason, I associate 'oi' with the Internet Oracle. as in 'oi, Zadoc!'.
moritz_: keep on shilling semi-scientific bullshit! 19:15
PerlJam masak: you're more connected, thus seemingly more "central"
missingthepoint PerlJam: exactly.
masak PerlJam: how do you count conntections? module dependencies? 19:16
moritz_ PerlJam: proto makes everything a distance "1" from itself
missingthepoint so to prevent code duplication, (ie multiple Text::CSVs), i vote for masak's github.
PerlJam masak: *you* are more connected :)
masak missingthepoint: our codebases are already very different. I don't see how we can easily merge them. 19:17
19:17 hercynium joined
masak but I'm already willing to change the name of mine s/CSV/Text::CSV/, in order to... uh... increase confusion. never mind. 19:17
[particle] make the module name the same, and change the authority :)
masak NYI, but good idea. 19:18
moritz_ swap module name and authorities
[particle] that's the easiest merge, let CPAN fight it out :)
moritz_ use masak :auth<Text::CSV>;
masak :)
missingthepoint actually, i vote for getting rid of 'C' entirely.
masak ENOCPAN.
moritz_: :P
PerlJam missingthepoint: me too.
missingthepoint: xSV ?
[particle] IP6AN
incomprehensible perl 6 archive network
missingthepoint PerlJam: too p5-internalsy 19:19
masak missingthepoint: Text::SV?
moritz_ text is certainly a scalar values, yes ;-)
missingthepoint masak: ok... still in Text::? 19:20
PerlJam heh
missingthepoint moritz_: my thoughts exactly :D
PerlJam context is king. Only a minority of people are going to associated SV with perl 5
moritz_ PerlJam: I know
PerlJam (it just so happens that some of that minority is probably a majority of the active #perl6 crowd right now, but still :) 19:21
masak remember, we want something that's just like CPAN, preferably _is_ CPAN, but it should also contain all manner of Parrot-based language modules, except CPAN only accepts Perl things, and it should be distributed like github, and it should have a cool name, except there's no consensus at all about the name. and someone needs to implement the thing, too.
missingthepoint but still, i concur
TimToady I think I may have discovered the coercion declaration syntax
masak oh?
TimToady my Foo() $x; # short for my Foo(Any) $x;
std: my Int() $x;
p6eval std 28042: OUTPUT«===SORRY!===␤Confused (two terms in a row?) at /tmp/qiwNlfCSqn line 1:␤------> my Int() ⏏$x;␤ expecting any of:␤ bracketed infix␤ infix stopper␤ standard stopper␤ statement modifier loop␤ terminator␤ trait␤FAILED 00:02 38m␤»
TimToady currently illegal 19:22
masak STD says no.
jnthn TimToady: That means it coerces on each assignment?
TimToady I haven't taught it yet :)
19:22 ispy_ left
moritz_ TimToady: should that be the same as 'my $x as Foo'? 19:22
TimToady my Foo() @foo = "x" xx *
don't think so, if as is returning an rvalue 19:23
anyway, it looks like in declarative context we have () available after typenames
jnthn my Foo() @foo = "x" xx *; # what does that boil down to?
TimToady my Foo @foo = map { Foo($_) }, "x" xx 8; 19:24
PerlJam Hmm. Why not? we already have [] and {} after typenames
TimToady and Foo($x) as rvalue is a coercion
jnthn TimToady: Hmm? I meant more, my Foo @foo creates an Array[Foo]
TimToady it's still an array of Foo
I think 19:25
storage-wise
jnthn That's awkward.
TimToady but it's an array of Foo() that knows it accepts Any
PerlJam and if I think of coercion as a transformation function, the my Foo() $x; makes good mnemonic sense too
TimToady so my Foo() @x; @x[42] = $nonfoo; should still try to do Foo($nonfoo) coercion 19:26
it's advertised as Any in the sig
moritz_ as Any(Foo)? ;-)
jnthn TimToady: What do we parameterize a role with?
PerlJam TimToady: but what else goes in the ()? would we have things like my Foo(Array of Int) @foo; ?
jnthn In order to ask fro this? 19:27
TimToady: I can imagine this may work:
TimToady any type you want to advertise it as accepting goes inside
[particle] my SomeClass() $obj = $some_sig; # would this run .new or .BUILDALL or just fail to coerce?
TimToady but default is Any
PerlJam okay
19:28 synth^2 joined
PerlJam and multiple coercions? 19:28
TimToady it would depend on whether anyone defined coercion to SomeClass
jnthn my Foo() @x; # really means: subset TEMP of Any where -> $x is rw { $x = Foo($x) }; Array[TEMP]
TimToady it still depends on SomeClass dispatching to &.()
jnthn However, if youdo that
19:28 synth^3 joined
jnthn Then you can't use multiple dispatch to distinguish @x and @y in my Foo() @x and my Bar() @x. 19:29
That may be surprising.
TimToady I don't see the problem offhand
what you just said doesn't make sense to me 19:30
jnthn ...
moritz_ well, we want Turing complete multi dispatch at one point
jnthn You didn't actually answer my original question.
moritz_ so we should be able to dispatch on everything ;-)
PerlJam moritz_: aren't we already there?
;)
jnthn Which was
moritz_ PerlJam: only with where-blocks (which are kinda cheating)
jnthn my Foo @x # it's an Array[Foo]
my Foo() @x # it's an Array[???] # what goes in the ???
It's kinda unimplementable without answering that. 19:31
19:31 pyrimidine joined
PerlJam It's still Array[Foo], but assignments get passed through Foo($thingy) to coerce as I read things 19:32
TimToady well, but the generic code probably needs to know both types
the internal and the external
jnthn PerlJam: What tells assignment that? 19:33
TimToady so perhaps Array[Foo(Any)] can be made to work
jnthn Hmm.
19:33 tak11 joined
jnthn That could work somehow. 19:33
TimToady but we have to be careful about maing Type(Type) work even in rvalue context, in that case
jnthn I think.
TimToady *making
jnthn Yeah 19:34
It's one of those "ouch, the type system was already complex enough" things.
TimToady yes, well, the question is whether coercion is entirely operational or wants to be more declarative in spots 19:35
jnthn I guess really though
TimToady and we're typically biasing things a bit toward the declarative in p6
19:35 barney left
PerlJam I don't think I've every thought of coercion as declarational. 19:35
s/y//
jnthn If you have a ::T that you set as a "coercion type", and you use it in a signature, we could somehow try to be smart enough about it to make it as if you'd written an Any $x as Foo
But if we're doing that 19:36
We maybe want to toss as
And just write Any(Foo) $x
oh wait
Foo(Any) $x
PerlJam So ... what problem does this solve?
TimToady yes, I'm looking for something that generalizes better than as
it can be treated as a single type until you want to differentiate the inside and the outside 19:37
jnthn PerlJam: The problem is that we need to make specification of a coercion kind of "first class".
Or a single type that you can talk about.
PerlJam jnthn: why?
jnthn Because otherwise we can't actually implement my Foo() @x
Or at least
Not by just having it fall out of a parametric role.
Which it really needs to. 19:38
That is, we want to be able to write a role Array[::T] and not have to worry about whether T jsut implies a constraint or a constraint plus a coercion.
My feeling is that Foo(Any) will be a kind of type constructor. 19:39
And when building a signature out of it we'd probably deconstruct it somewhat.
And it has an ACCEPTS that coerces.
TimToady: We might be able to sneak this in with a bit less pain that I first thought... 19:40
TimToady almost more of a storage issue, so the "int" type is more like int(Int)
19:40 synth^3 left, synth^3 joined
TimToady I'm officially Int, but whenever you store me, turn me into an int 19:40
so possibly related to boxing/unboxing 19:41
jnthn Hmm.
TimToady though it's funny that "unboxing" is associated with the tighter constraint
jnthn That's interesting too. 19:42
TimToady arguably int is a smaller box than Int :)
jnthn Yeah
PerlJam I haven't used C++ in a good while, but I remember there was something like, if you defined a constructor for your object that took a type X, then anywhere you specified a variable to be a type of your class, you could use type X and it would just magically work. I remember this magic being ultimately evil. This "coercion type" feels similar to me right now.
moritz_ implicit usage of such coercions would be evil, yes 19:43
jnthn Which is what my Foo() $x is kinda asking for. ;-)
PerlJam exactly.
19:44 jferrero left
PerlJam I guess perl has much been scoping control though, so that may mitigate any problems. 19:44
TimToady and we have a community to point out the evil of one's ways
PerlJam i.e. with "my Foo() $x", the coercion is scoped the same as $x, right? 19:45
jnthn "That's SO evil! I like it!!"
PerlJam jnthn: heh
jnthn PerlJam: Correct
PerlJam: Which makes it feel a bit less awful.
moritz_ PerlJam: it's only only scoped like $x, it's also tied to the variable $x
jnthn Aye
PerlJam moritz_: right, good poitn
er, point
jnthn It's very much like dialects.
There's always a use statement to tell you what language you're in. 19:46
TimToady anyway, my point was primarily about the syntax, not the pragmatics :)
19:47 synth^2 left
PerlJam TimToady: well, I was getting to that :) parens seem a small indicator of such magic 19:47
(perhaps I'm just too curmudgeonly today)
TimToady twice as big as it was gonna be in my earlier syntax :) 19:48
which was something like
my >Foo $x
or
my Any>Foo $x
moritz_ granted, Foo() is bettter. 19:49
TimToady multi infix:<+> (Num() $x, Num() $y) { #`<primitive addition here> } # the old Any,Any version 19:54
moritz_ does it still lift()?
or did it never?
TimToady no redispatch in this form 19:55
jnthn TimToady: Ah, it coerces while binding?
TimToady without coercion, Any/Any redispatches to Num/Num, but this is conceptually simpler 19:56
jnthn TimToady: Any reason not to have it, if it's primitive, as
TimToady yes, that's the idea
jnthn :(num() $x, num() $y)
I guess it's impl. dependent...
TimToady well, if you want to coerce to the native type, sure
moritz_ what about Inf?
jnthn I'd figure you may for a primitive.
moritz_ 'Inf' + 3 19:57
jnthn moritz_: num can represent inf
moritz_ jnthn: ok
jnthn *infinity
TimToady presumably a coercion failure would result in binding failure though, and dispatch elsewhere more general
moritz_ sounds sane.
TimToady maybe
moritz_ (this part, at least :) 19:58
TimToady something to think on for a while, anyway
jnthn note to self: doing ^C while viewing a git diff has a high chance of screwing up the terminal
TimToady let people's subconsciouses stew for a while to see what bubbles up
19:59 japhb left
PerlJam Is the dispatch order of Num vs. Num() such that Num comes first in the candidate list? 20:00
moritz_ PerlJam: since Num() would be like Any, Num is more specific
TimToady well, my brane needs to go inoperative for a bit, in an old geezery kind of way. & # zzz 20:01
PerlJam moritz_: that's what I would think. Just checking. :)
jnthn Num would be narrower than Num(), yes 20:02
PerlJam X(Y) is treated like Y
jnthn Right.
The multi-dispatcher will probably be ignorant of the existence of the X
If we can get away with that anyway...
moritz_ at least for building the candidate list 20:03
it will be aware of it during the bindability check
jnthn At least.
Oh, hmm
We need to be careful there...
If you make your infix:<+> need a bindability check, it means we can't do type based caching, which means slowness 20:04
Oh
Maybe we could in that we know which types can coerce
But that still may be too much of an assumption.
moritz_ and it would make user defined types slower
jnthn Wow, my throat is bad enough today that I can actually sing like the Opeth guy... 20:06
pmichaud while on this topic...
jnthn ...Opeth?
...my throat?
pmichaud it seems a bit bizarre that Foo($x) defaults to Foo.new($x), given that the default .new doesn't know how to handle a single positional (iiuc)
jnthn heh, I didn't know it did.. 20:07
I thought it did $x.Foo
moritz_ that's what I thought too
pmichaud this came up in discussions yesterday
looking for the reference(s)
20:07 ruoso left
pmichaud S13:204 20:08
As a fallback, if no
method responds to a coercion request, the class will be asked to attempt to
do C<Dog.new($spot)> instead.
jnthn spec fail.
moritz_ aye.
pmichaud I also was wondering if the default 'new' method is a multi 20:09
jnthn pmichaud: Not afaik.
Though that's only an issue if you monkey-patch Object, no?
pmichaud for example, in the thread "Custom object constructors" on p6l, ruoso++ suggests
method new(Str $timestamp) { self.SUPER::new(ts => strptime('...',$timestamp));
}
and I was wondering if that's in fact more like 20:10
multi method new(Str $timestamp) { nextwith(ts => strptime('...', $timestamp)); }
or, if .new is multi
jnthn Why isn't it method new(Str $timestamp) { nextwith(ts => strptime('...', $timestamp)); } 20:11
?
pmichaud multi method new(Str $timestamp) { .new(ts => strptime('...', $timestamp)); }
jnthn But yes, nextwith is the nice thing to use here.
YOu don't have to be a multi to defer though.
Deferal goes through multis and then up the inheritance hierarchy.
pmichaud right
jnthn So no need for it to be a multi for this to work.
pmichaud but with multi, one doesn't have to use nextwith at all
masak why is nextwith better than naming the method?
jnthn The default constructor is not generated though, afaiu?
pmichaud one can just redelegate to .new 20:12
jnthn It's inherited?
True
pmichaud the original question was how to make something like Foo.new($x) dwim
moritz_ masak: it's not the same
jnthn That's more about people declaring their own .new a multi though, so dispatch can skip over it and keep looking?
masak and when you guys say .new, you do mean self.new, right? </nitpicky> 20:13
jnthn masak: yes
pmichaud masak: yes -- sorry about that
masak oh, good.
jnthn Just being lazy fingered. ;-0
moritz_ $.new for short ;-)
masak don't want to have that bikeshed discussion again. :)
(it was in 2006)
PerlJam masak: wait ... $_ should be aliased to the invocant in methods ... ;)
jnthn I might just post the three word message "SHUT UP ALREDY" if that one happens again...
masak PerlJam: argh! 20:14
jnthn That way people will discuss my attitude problem instead of having the discussion again. ;-)
mkelly32 heh, you're on lists that got that guy, too?
pmichaud <bikeshed>I don't think it's an attitude problem, it's more like a character flaw</bikeshed> :-) :-) 20:15
jnthn pmichaud: I'm not quite sure how it'd dwim with positionals.
Thanks pm. :-P
pmichaud jnthn: I'm not talking at all about auto-creating new in any form, if that's what you're thinking.
jnthn pmichaud: OK, good, I just hadn't yet worked out why you wanted to write a multi new. 20:16
pmichaud given that I'm creating a class Foo, and I want to be able to define a .new for Foo that works based on positional arguments, what's the canonical way to do that?
jnthn I see it now (because it's a trick to dispatch up the tree).
masak Rakudo works again! bacek++, I guess.
jnthn It's Perl, there's many ways to do it...
pmichaud jnthn: sure, but the versions given in the thread all looked somewhat bletcherous -- they were basically overriding BUILD and new both (and claiming that both had to be done) 20:17
jnthn I guess if you're subclassing, you care a bit about the constructors of the parent.
I don't get why both have to be done.
pmichaud I don't think they do
jnthn If you take positionals, then you just map them to names when calling .bless
pmichaud so your code looks like... ?
jnthn method new($foo, $bar) { self.bless(*, :$foo, :$bar) } 20:18
well, and a return in there if you want to be explicit.
PerlJam hadforgotten about the :$foo shortcut
perl++
jnthn The other nice thing about naming your parameters in the signature
As in, giving them the names of the attributes
Is that as well as calling that $x.new(4, 2) 20:19
you can also call it $x.new(foo => 4, bar => 2)
;-)
(Well, not in Rakudo yet.)
pmichaud what happens if Foo itself is subclassed, though? 20:20
it still works out?
jnthn It'd inherit that constructor.
The subclass would need to define a constructor that it liked more.
Or Foo could declare it's candidate a multi so failure to bind there isn't fatal. 20:21
pmichaud right, which gets back to my "is <new> a multi" question
jnthn Depends on the semantics a class prefers to have.
If it wants to say "thou shalt construct me this way and only this way", then only.
If it wants to say "here's another way", then multi. 20:22
pmichaud I meant the default <new>
jnthn The default new is in Object.
pmichaud correct
is it a multi, or no? or does it not matter?
jnthn It doesn't matter.
There's nowhere to look beyond it.
pmichaud so if I have class Foo defined with
multi method new($foo) { self.bless(*, :$foo); }
jnthn multi method dispatch in Perl 6 does not consider all multis from all classes together.
It's a class by class consideration. 20:23
Up the inheritance hierarchy.
pmichaud I can still do Foo.new() and have it get to Object's .new ?
jnthn Correct.
pmichaud even if Object.new isn't defined as multi ?
jnthn If a class provides no viable multi candidates, we continue looking up.
Correct.
pmichaud okay. 20:24
jnthn We consider the options provided by each class in isolation.
Failing to bind to an only is always an error.
But failing to bind to any of the available multis just means we consider the parents.
pmichaud got it.
PerlJam but, if Foo.new($x) is to do something useful as per S13, then Object's new will have to be multi :)
jnthn PerlJam: Heh, true. 20:25
I'm a little distrubed about that though.
pmichaud Yes, that issue still exists independent on this one.
moritz_ isn't it multi anyway?
pmichaud I'd like to see the self.bless(*, ...) answer added to the thread on p6l, though.
jnthn I can't think of a good default.
moritz_ pmichaud: I just mentioned it
pmichaud okay, good.
jnthn oh arse...so much for my patch being good. 20:26
jnthn failed a sanity test
pmichaud moritz++
jnthn pmichaud: *almost* got trait application on containers into shape. 20:27
pmichaud jnthn: nice!
jnthn pmichaud: trait_mod:<of> is now written entirely in Perl 6. ;-)
reqamst someone vandalised pugs article on wikipedia, I fixed it :(
pmichaud that's simultaneously scary and cool.
moritz_ reqamst++ 20:28
jnthn pmichaud: Yeah, though annoyingly I've managed to cause some weird breakage...hopefully nothing too hard to track down. 20:31
Doesn't help that the PIR line number seems to be kinda off...
Tene jnthn: that's been the case for ages. I vaguely remember suspecting that imcc isn't counting annotations when counting lines. 20:32
But I don't know that I ever confirmed it.
20:36 japhb joined 20:49 Caelum joined 20:58 tak11 left 21:01 tak11 joined 21:06 zloyrusskiy left 21:08 nihiliad left 21:18 tak__ joined, tak11 left 21:19 alester joined
KatrinaTheLamia hmm... I may look into signing up for git hub... though, do I need to be a completely a total git to use it? Or is it something I can work on? Like I mean I am a little daft--does that help? 21:22
KyleHa1 The name 'git' refers to its creator, not the users. Sorry. 21:23
21:23 tak__ is now known as tak11
moritz_ it generally doesn't help to be daft when dealing with technology ;-) 21:23
KyleHa1 deft, another story.
KatrinaTheLamia yeah, I recall the quote Linus made about him naming all his projects after himself... I chuckled 21:24
pyrimidine git?
KatrinaTheLamia moritz_, I dunno, with windows, php and so many other things being so popular with technology, I wouldn't think it would hurt ~.^
moritz_ KatrinaTheLamia: well, maybe I'm wrong. 21:25
KatrinaTheLamia pyrimidine, git is a slang term for somebody who is useless... or something similar to that. Generally if you call somebody a "git" it generally is not complement.
pyrimidine yes, I know (mum's english)
KatrinaTheLamia hooray! 21:26
masak KatrinaTheLamia: if it's a nudge you need to sign up with github: do it! you won't regret it.
github is better than Facebook and Twitter combined.
friending is called 'forking' on github.
KatrinaTheLamia masak, alright--I was going to in a bit... I figured I'd make a joke...
and oh dear... I am not entire sure I want to be on github now ~.^
masak KatrinaTheLamia: I'm glad you did.
KatrinaTheLamia: because of the 'forking' part? it doesn't hurt or anything. 21:27
pyrimidine just so everyone knows: www.oreillynet.com/pub/e/1394
masak wonders if he's too risqué
pyrimidine allison pointed this out the other day
KatrinaTheLamia I mean facebook? I generally regard that site as lower than the various chans when it comes to intelligent discussion on the interweb
masak KatrinaTheLamia: tomayko.com/writings/github-is-mysp...or-hackers 21:28
pyrimidine (that's the link to the 'Git in an Hour' webcast)
masak KatrinaTheLamia: oh, me too. I don't go near it with a wet towel.
KatrinaTheLamia oh dear... comparing it to myspace now... makes me not even want to go there more
masak KatrinaTheLamia: but... what that URL says.
KatrinaTheLamia: I'm afraid you misunderestimate. :) 21:29
KatrinaTheLamia masak, well, you should generally be going everywhere with your towel ~.^
KatrinaTheLamia stops complaining and learns to click the links.
masak github is great at letting you click links.
21:30 zamolxes joined
pyrimidine the git webcast I linked to above was a very good introduction 21:30
also: progit.org/book/ 21:31
21:31 molaf left 21:34 Limbic_Region joined
jnthn pmichaud: Have happiness, the bacekator has fixed the regression, apparently... 21:42
masak is teh happy about that
jnthn I on the other hand have managed to write a patch that skillfully passes everything we passed before...apart from 1 test in assign.t. Gah. 21:43
moritz_ jnthn: review the test for correctness before worrying about it
jnthn rakudo: my $x //= 42; say $x;
moritz_ jnthn: assign.t is full of only half-correct tests, I fear
p6eval rakudo 0d4fe0: OUTPUT«Use of uninitialized value␤␤»
21:44 Whiteknight joined
masak oh, I remember that one. 21:45
jnthn oh wtf
pyrimidine rakudo: my $x \\= 42; say $x;
jnthn not ok 77 - //= also works in declaration
# got: undef
# expected: 5
p6eval rakudo 0d4fe0: OUTPUT«Statement not terminated properly at line 2, near "\\\\= 42; sa"␤in Main (src/gen_setting.pm:3390)␤»
jnthn OK, how on *earth* was this test passing at all before?
masak jnthn: :D
21:46 synth^3 left, synth joined, KyleHa1 left
moritz_ lol 21:47
masak or, as the Brits say, loul.
jnthn switches back to master to see if it works there. 21:48
masak oh, and by the way, missingthepoint and I had a long privmsg discussion, in which we totally connected and decided to mind-meld our two Text::CSV modules in a few days' time. 21:50
synergy++
jnthn rakudo: use Test; my $f //= 5; is $f, 5, '//= also works in declaration';
p6eval rakudo 0d4fe0: ( no output )
jnthn oh wtf 21:51
rakudo: my $f //= 5; say $f eq 5
p6eval rakudo 0d4fe0: OUTPUT«Use of uninitialized value␤0␤»
masak the evalbot borders on being useless when it doesn't output stuff.
jnthn OK, color me confused. 21:52
The test does pretty much that.
And...passes.
epic wtf.
wayland76 masak: I was worried for a minute that the two of you were going to become a borg :)
masak wayland76: oh, we haven't ruled that out. 21:53
jnthn OK, so it really does work in the context of the test. 21:54
:'-(
masak jnthn: that's an amazing discovery!
jnthn But only in master 21:55
Not in my branch.
I wonder if I'm allowed to call that out as a bogus pass.
Given
That it clearly doesn't really work... :-|
rakudo: my $x //= 5; say $x;
rakudo: { my $x //= 5; say $x; }
p6eval rakudo 0d4fe0: OUTPUT«Use of uninitialized value␤␤»
rakudo 0d4fe0: OUTPUT«5␤»
jnthn ...erm...
moritz_ rakudo: my @a = <a b c>; for ^10 { print @a[$_ % *] }
p6eval rakudo 0d4fe0: OUTPUT«abcabcabca» 21:56
moritz_ rakudo++
masak wow.
that is beautiful.
moritz_ it is.
jnthn wow...in theory I knew that works, but I'd never thought to try it :-)
moritz++
That is, I can see exactly why it would work.
But yeah, very neat idea. :-) 21:57
moritz_ it's amazing what you can do with whatever-stars
jnthn I'm not entirely sure I want to know why //= only works in a block...
...and why my changes made it just consistently not work.
masak that's the big mystery, it seems. 21:59
moritz_ just hacked multiple bars into SVG::Plot 22:00
masak moritz_++
moritz_ next step: stacked bars 22:01
then we have all we need to re-generate the spectest charts with Perl 6.
masak \o/
missingthepoint pmichaud: ^^ 22:02
moritz_ well, let's wait until it's actually done ;-)
22:02 pyrimidine left
missingthepoint moritz_: ah, i just provided extra motivation... ;) 22:03
jnthn moritz_: Will they be prettier too? ;-) 22:04
missingthepoint off to do Text::CSV .t work. 22:05
jnthn moritz_: And maybe we can incorporate a picture of a pony? ;-)
moritz_ jnthn: I hope so.
22:05 MoC` joined
moritz_ a pony and a butterfly. 22:05
wayland76 What does that % operator do up above in moritz_'s rakudo code?
moritz_ wayland76: modulo 22:06
masak wayland76: it gives you the remainder of an integer division.
22:07 payload left
masak that's the fifth arithmetic operation, one that they sadly tend to under-emphasize in the basic math curriculum. 22:07
wayland76 That's what I thought. I still don't understand the code. Is it some sort of iterator thingy?
masak wayland76: no, but I had to look at it for a while too to grok it.
wayland76: the secret is that * means .elems in array indices.
moritz_ wayland76: replay * by the number of items in the array 22:08
so it's just <a b c>[ $_ % 3 ]
wayland76 Ah, I see the magic :)
moritz_ but written for an arbitrary number of items
wayland76 thanks :)
moritz_ yw
wayland76 rakudo: my @a = <a b c>; print @a[map { $_ % @a } 1..10] 22:10
p6eval rakudo 0d4fe0: OUTPUT«Statement not terminated properly at line 2, near "[map { $_ "␤in Main (src/gen_setting.pm:3390)␤»
wayland76 rakudo: my @a = <a b c>; print @a[map { $_ % @a } 1..10];
p6eval rakudo 0d4fe0: OUTPUT«Statement not terminated properly at line 2, near "[map { $_ "␤in Main (src/gen_setting.pm:3390)␤»
moritz_ wayland76: comma after }
wayland76 rakudo: my @a = <a b c>; print @a[map { $_ % @a }, 1..10];
p6eval rakudo 0d4fe0: OUTPUT«bcabcabcab» 22:11
jnthn Anyone else finding rakudo.org a little bit failed at the moment?
wayland76 rakudo: my @a = <a b c>; print @a[map { $_ % @a }, 0..9];
moritz_ well, ^10 is is 0..9
p6eval rakudo 0d4fe0: OUTPUT«abcabcabca»
moritz_ jnthn: why?
jnthn moritz_: When I try to load it I get a blank page... :-/
moritz_ works fine here. 22:12
jnthn huh
:-/
moritz_ oh dammit, I shouldn't try to be too clever. 22:13
jnthn loliblogged: use.perl.org/~JonathanWorthington/journal/39499 22:15
moritz_ lolib0rkedmysvgplot 22:19
masak jnthn++ # nice, meaty post 22:21
jnthn moritz_: That sounds oddly unlolish.
.oO( post not suitable for vegitarians? )
masak jnthn: uhm, quorn-y post?
jnthn eww 22:22
masak tofu-y post?
doesn't quite have the same ring to it.
jnthn "I will now proceed to eat a tofu."
masak I'm sure the vegetarians will understand.
22:22 payload joined 22:23 M_o_C left
jnthn (tofu quote from www.weebls-stuff.com/wab/hams/ - warning, awful pun at the end) 22:25
masak jnthn: I couldn't help but note that your PublicReportRenderer doesn't explicitly inherit from your ReportRenderer...
jnthn masak: oh, fail
masak: corrected, thanks like 22:26
masak np.
so, wow, nextsame et al. actually does both inheritance-hierarchy and multimethod deferral?
jnthn Correct. 22:27
masak which one does it do when faced with a choice of both?
jnthn always multi-methods, and if the last applicable one of those defers too then it goes up the inheritance chain.
masak ok. 22:28
jnthn masak: Imagine making a list of every multi-variant in every class in the hierarchy that you could call.
wayland76 masak: You left my character classes out of your blog post
jnthn That's what it's walking through.
masak wayland76: oh noes! which post, and which character classes?
22:28 jan__ left
wayland76 The one about Eve Online (Aug 11) 22:29
22:29 pmurias left
wayland76 On the other hand, we don't need newbies volunteering to be spec re-drafters :) 22:30
masak wayland76: I think you might fall under 'priest', actually. but consider my classes an open set.
wayland76 Yeah, I figured I was closest to "priest", with attempts at healing also involved 22:31
masak we're all mixes, I think. 22:32
real-world RPGs tend not to be so clear-cut.
jnthn figures he's probably something of a Mage. 22:33
moritz_ has the branedumb
I have an array @a which contains arrays of numbers (of same lenght)
and I want to calculate the maximal row sum
wayland76 masak: I've played shadowrun, which lets you cut across categories :)
moritz_ hwo do I do that?
masak rakudo: @a = [1,2,3], [4,5,6], [7,8,9]; say [max] @a>>.sum 22:34
22:34 tak11 left
p6eval rakudo 0d4fe0: OUTPUT«Symbol '@a' not predeclared in <anonymous> (/tmp/25HpXCMmZ2:2)␤in Main (src/gen_setting.pm:3390)␤» 22:34
masak rakudo: my @a = [1,2,3], [4,5,6], [7,8,9]; say [max] @a>>.sum
p6eval rakudo 0d4fe0: OUTPUT«Method 'sum' not found for invocant of class 'Perl6Array'␤»
masak O RLY?
wayland76 max(for @a -> @b { [+] @b })
rakudo: @a = [1,2,3], [4,5,6], [7,8,9]; say max(for @a -> @b { [+] @b }); 22:35
p6eval rakudo 0d4fe0: OUTPUT«Statement not terminated properly at line 2, near "(for @a ->"␤in Main (src/gen_setting.pm:3390)␤»
moritz_ I'd want to get 3 + 6 + 9 = 18 in masak's example
masak rakudo: sub sum(@a) { [+] @a }; my @a = [1,2,3], [4,5,6], [7,8,9]; say [max] map { sum($_) }, @a
p6eval rakudo 0d4fe0: OUTPUT«24␤»
masak moritz_: oh.
moritz_: is that what you call 'rows'? :)
moritz_ maybe I formulated it a bit weirdly
masak aye.
moritz_ masak: for some values of transposition :/
jnthn rakudo: my @a = [1,2,3], [4,5,6], [7,8,9]; say [+] @a.map( .max } 22:36
p6eval rakudo 0d4fe0: OUTPUT«Statement not terminated properly at line 2, near "( .max }"␤in Main (src/gen_setting.pm:3390)␤»
jnthn rakudo: my @a = [1,2,3], [4,5,6], [7,8,9]; say [+] @a.map({ .max })
gah
:-)
p6eval rakudo 0d4fe0: OUTPUT«18␤»
jnthn rakudo: my @a = [1,2,3], [4,5,6], [7,8,9]; say [+] @a.map: *.max
p6eval rakudo 0d4fe0: OUTPUT«18␤»
jnthn Less brackety. 22:37
moritz_ rakudo: my @a = [3, 2, 1], [4, 5, 6], [7, 8, 9]; say [+] @a.map: *.max
p6eval rakudo 0d4fe0: OUTPUT«18␤»
moritz_ not quite what I want
masak thought so.
moritz_ should be 16 in that case
jnthn rakudo: my @a = [1,2,3], [4,5,6], [7,8,9]; say [+] @a.map: *.end
masak jnthn: the sum of the column with the largest sum.
jnthn oh no
p6eval rakudo 0d4fe0: OUTPUT«6␤»
jnthn Oh! 22:38
wait, no...really?
moritz_ rakudo: @a = [1,2,3], [4,5,6], [7,8,9]; say ([Z] @a).perl
p6eval rakudo 0d4fe0: OUTPUT«Statement not terminated properly at line 2, near "@a).perl"␤in Main (src/gen_setting.pm:3390)␤»
jnthn Oh, columm as in...1 + 4 + 7
OK
wayland76 wonders if we can make people saints (in protestantism, every Christian is a saint), and whether he can be a presbyter or not :)
masak rakudo: sub col-sum(@a, $col) { [+] map { $_[$col] }, @a }; my @a = [3, 2, 1], [4, 5, 6], [7, 8, 9]; my $max-col = sort { col-sum(@a, $_ }, ^@a[0]; say $max-col; say col-sum(@a, $max-col) 22:40
p6eval rakudo 0d4fe0: OUTPUT«Statement not terminated properly at line 2, near "(@a, $_ },"␤in Main (src/gen_setting.pm:3390)␤»
masak rakudo: sub col-sum(@a, $col) { [+] map { $_[$col] }, @a }; my @a = [3, 2, 1], [4, 5, 6], [7, 8, 9]; my $max-col = sort { col-sum(@a, $_) }, ^@a[0]; say $max-col; say col-sum(@a, $max-col)
p6eval rakudo 0d4fe0: OUTPUT«0 1 2␤Use of uninitialized value␤Use of uninitialized value␤Use of uninitialized value␤0␤»
masak hrm.
moritz_ rakudo: my @a = [3, 2, 1], [4, 5, 6], [7, 8, 9]: say [max] @a.keys.map: { [+] @.values[*][$_] } 22:41
p6eval rakudo 0d4fe0: OUTPUT«Statement not terminated properly at line 2, near ": say [max"␤in Main (src/gen_setting.pm:3390)␤»
moritz_ rakudo: my @a = [3, 2, 1], [4, 5, 6], [7, 8, 9]; say [max] @a.keys.map: { [+] @.values[*][$_] }
p6eval rakudo 0d4fe0: OUTPUT«Lexical 'self' not found␤»
masak rakudo: sub col-sum(@a, $col) { [+] map { $_[$col] }, @a }; my @a = [3, 2, 1], [4, 5, 6], [7, 8, 9]; my $max-col = (sort { col-sum(@a, $_) }, 0..@a[0].elems)[0]; say $max-col; say col-sum(@a, $max-col)
p6eval rakudo 0d4fe0: OUTPUT«Use of uninitialized value␤Use of uninitialized value␤Use of uninitialized value␤3␤Use of uninitialized value␤Use of uninitialized value␤Use of uninitialized value␤0␤»
masak back to the drawing board.
moritz_ got it. 22:42
rakudo: my @a = [3, 2, 1], [4, 5, 6], [7, 8, 9]; say [max] @a.values[0].keys.map: { [+] @a.map: -> $a { $a[$_] } } }; 22:43
p6eval rakudo 0d4fe0: OUTPUT«Syntax error at line 2, near "};"␤in Main (src/gen_setting.pm:3390)␤»
wayland76 Just so I understand the problem, we add up each "column", then pick the largest?
moritz_ rakudo: my @a = [3, 2, 1], [4, 5, 6], [7, 8, 9]; say [max] @a.values[0].keys.map: { [+] @a.map: -> $a { $a[$_] } };
p6eval rakudo 0d4fe0: OUTPUT«16␤»
moritz_ wayland76: correct.
masak rakudo: sub col-sum(@a, $col) { [+] map { .[$col] }, @a }; my @a = [3,2,1], [4,5,6], [7,8,9]; say [max] map { col-sum(@a, $_) }, ^@a[0] 22:44
p6eval rakudo 0d4fe0: OUTPUT«16␤»
masak moritz_: yep, our solutions are equivalent.
moritz_ right; I just inline col-sum as another map 22:45
which means I have stacked bar charts working.
masak yay!
I, on the other hand, completely ran into a wall with the CSV module. 22:46
moritz_ with yours? or with missingthepoint's?
masak with mine.
time to push what I have and sleep on it.
missingthepoint masak: i have further thoughts for you. 22:47
(whenever you want them)
masak missingthepoint: preferably tomorrow, if that's alright.
missingthepoint masak: no problem :) 22:48
masak missingthepoint: I just pushed the progress for the evening.
I got many of the early TODO points working.
moritz_ has to fix lots of docs after his stacked bars 22:49
masak missingthepoint: I'm currently stuck on newlines inside of quotes, so I'm going to sleep on that.
moritz_ moritz.faui2k3.org/tmp/out.svg
22:49 jonathanturner joined
moritz_ that's how the stacked bars can look like 22:49
missingthepoint masak: ok. 22:50
masak moritz_++
sleep &
22:50 masak left
jnthn night, mas...oh, to late. 22:50
moritz_: Ooh, nice.
reqamst moritz_: could you pull it to the github? 22:56
moritz_ reqamst: it's already on github (but it's not Tufte :/ ) 22:57
github.com/moritz/svg-plot/
it's nice for now, but not very extensible.
reqamst oh, I was thinking it's Tufte 22:58
moritz_ that would be great, but it requires more initial work 22:59
23:02 simcop2387 left
jnthn moritz_: warn "stacked: $stacked"; # leftover debugging output? 23:03
moritz_ jnthn: yes
jnthn: removed, thanks 23:04
23:04 jonathanturner left
sjohnson perl 6 is a big deal 23:08
jnthn orly? :-) 23:09
sjohnson =)
i should subscribe to the rakudo blog
but i am not quite sure what i can offer to the cause, though i would love to
my main strength about Perl 6 development is getting excited and optimistic about it 23:11
moritz_ you can start writing stuff in perl 6 and blog about it
that's basically what masak++ does ;-)
sjohnson definitely a good start 23:12
reqamst hmm.. I can run rakudo's perl6 only from that same directory where is perl6, otherwise I receive paste.lisp.org/display/85764
is it common? 23:13
moritz_ reqamst: you need to run 'make install', the installed version should run from everywhere
missingthepoint sjohnson: i have a few ideas about how you could help if you're interested... :)
sjohnson what did you have in mind?
missingthepoint sjohnson: well, you have web skills? i suspect rakudo.org could use some updating and TLC... (but you might need to run this by pmichaud or alester) 23:15
sjohnson wouldn't be a bad start 23:16
i do have said skills
missingthepoint sjohnson: also, hanging around here and being friendly is always useful... you actually pick up a fair bit just idling, and you can pass that on to visitors :)
reqamst moritz_: I made
missingthepoint sjohnson: i have half a "Not-Implemented-Yet List" for Rakudo... I could use some help with that, and I think it could go on rakudo.org/status... 23:18
sjohnson missingthepoint: the occasional pat on the back to others and telling a silly joke go a long way too :]
alester sjohnson: We are glad to get any kind of help on rakudo.org
whether visually w/tmeplates, or with creating pages, or whatever. 23:19
reqamst moritz_: but now I think it's not related to rakudo
nvm
sjohnson missingthepoint: is it online?
alester: i'll see what i can do
missingthepoint sjohnson: soon... :) 23:20
could some sort of admin privileges be organized for sjohnson on rakudo.org? :) 23:22
(better not hamstrung, methinks...) 23:23
sjohnson *puppy dog eyes* 23:25
alester i'm glad to hand out privs once there's some sort of idea of what you'll be doin'
Not that I don't trust what you'll be doin'. Mostly the fewer privs that go out, the fewer privs I need to wrangle. 23:27
sjohnson do you need special privs to post on the main blog?
wayland76 sjohnson: Are you planning to mostly edit content?
redesign css?
alester you need privs to do anything.
sjohnson ehh before I get too ahead of myself, i won't make any claims. i have hardly enough time as it is, but i was thinking maybe posting some relevant paragraphs on the main site after giving rakudo a thorough testing 23:28
wayland76 (when I said "edit content", I mean "edit/write content"
alester Although excitement and optomistic is free. :-)
sjohnson by writing cute programs, like puzzle solvers, flashcard programs, etc
alester That's great. I"d love to see what you can come up with.
sjohnson: You know what else I need help with? 23:29
wayland76 Sounds like "content creation, etc"
missingthepoint sjohnson++, good ideas :)
sjohnson alester: what's that
alester perlbuzz.com/2009/08/help-perl101or...omers.html I really want to make perl101.org be dual-languaged.
wayland76 alester: post it in the channel; maybe someone will volunteer :)
alester For example, I put some Perl 6 here: perl101.org/arrays.html
wayland76 rakudo is in Drupal 23:30
sorry, rakudo.org is
alester This will help newbies to both Perl 5 AND Perl 6 AND for people transitioning to Perl 6.
sjohnson i can speak french
cute bees btw 23:31
:)
23:31 MoC` left
missingthepoint sjohnson: can you write french? 23:33
sjohnson yes 23:34
missingthepoint sjohnson: excellent! why didn't you say so sooner? :)
23:34 payload left
sjohnson too obsessed with learning Japanese now :) 23:35
alester I don't believe it. He should have said oui.
missingthepoint sjohnson: have you seen u4x? (svn.pugscode.org/docs/u4x/README)
sjohnson link won't work for me
alester: ... oui bien sur
missingthepoint oh hm, oops, svn.pugscode.org/pugs/docs/u4x/README 23:36
sjohnson if i didn't know "oui" for real, it would be a sad day in Perl history
23:36 simcop2387 joined
sjohnson someone will have to put me to the translator test 23:39
and have faith that I won't use babelfish Google language tools 23:40
（　｀ー´）
alester We wouldn't really know if you did. :-)
Anyone who has not yet read the Zompist Phrasebook needs to right now. www.zompist.com/phrases.html 23:41
missingthepoint I think there are a few people from .fr here... if you have tuits they might be able to do some quick checks...
sjohnson places his hand on the Camel Book and recites the Perl oath
heheh
23:41 synth^2 joined
wayland76 "I do solemnly swear to promote laziness, be impatient, and share the hubris at every opportunity" :) 23:42
sjohnson "The bellboy won't score me any coke. "
i love how that's slipped in
alester oh, keep going
23:46 frew__ joined
sjohnson that was a good read 23:47
i showed my brother
he also enjoyed it :) 23:48
alester "I know I'm naked, can you please just tell me how to get back to the hotel" is the capper for m.e
wayland76 I love the translation of "Where is the Cathedral?" :)
alester I know many are not real.
What is it?
wayland76 "Où est le bordel ?" 23:49
alester hahah
ok going home
wayland76 (for backloggers who don't speak french, look up the English word "bordello")
23:49 alester left
sjohnson where's the brothel i think it translaets to 23:50
wayland76 sjohnson: Yes, I know. I was giving people a chance to improve their English at the same time :)
sjohnson that's my wayland! 23:51
jnthn Gotta love the ones for "I understand your language perfectly."
sjohnson :)
23:52 nihiliad joined
sjohnson anyone around here around 27 years old? 23:52
and played NES?
hopefully this isn't too off-topic 23:53
23:53 synth left
sjohnson this perl101 site is great 23:58
i'd love to contribute some tricks i've learned over the years to this site
23:59 jan_ joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!