»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg camelia perl6: ... | irclog: irc.perl6.org | UTF-8 is our friend!
Set by sorear on 25 June 2013.
timotimo huh, strange 00:02
/home/siddhant/rakudo/install/bin/parrot - exists and is readable + executable?
00:11 stevan_ left, REPLeffect joined 00:12 Psyche^ joined 00:15 Psyche^_ left 00:18 REPLeffect left 00:22 hypolin joined 00:32 REPLeffect joined 00:40 REPLeffect left 00:51 REPLeffect joined 00:56 REPLeffect left, cognominal left 01:00 REPLeffect joined 01:04 REPLeffect left 01:14 btyler left 01:18 REPLeffect joined 01:27 REPLeffect left, REPLeffect joined 01:37 REPLeffect left 01:49 REPLeffect joined 01:56 FROGGS_ joined 01:57 preflex left, preflex_ joined, ChanServ sets mode: +v preflex_ 01:58 preflex_ is now known as preflex 02:00 FROGGS left 02:04 skids joined 02:08 btyler joined 02:11 lichtkind left 02:17 btyler left 02:40 odoacre joined 02:48 fridim__ joined 02:49 xinming left 02:51 xinming joined 03:00 tokuhirom left 03:03 tokuhirom joined 03:09 preflex left 03:11 preflex joined, ChanServ sets mode: +v preflex 03:21 geekosaur left, geekosaur joined 04:15 stevan_ joined 04:16 stevan_ left 04:18 stevan_ joined 04:23 popl joined
moritz \o 04:31
popl hi 04:32
moritz good morning popl 04:33
04:33 btyler joined
popl I'm running spec_smoldertest right now (I just finished building rakudo/parrot) so there will be some more qa data soon 04:34
I think I counted 918 tests
it's only through about 20 of them
(not that it's terribly interesting) 04:35
moritz: good morning 04:36
perl6 is not even released but it's very old. I have a lot to catch up on. :P 04:37
04:41 BenGoldberg left 04:44 btyler left
diakopter popl: well, there are regular releases by some peoples' definition of release, but not by others' 04:45
04:46 xenoterracide_ left
popl diakopter: yes, I understand it is nebulous 04:46
04:54 sftp left 04:57 sftp joined 04:59 Rotwang joined 05:37 popl left 05:41 SamuraiJack joined 05:47 raiph left 05:49 PacoAir joined, PacoAir left 05:50 PacoAir joined 06:06 popl joined 06:20 PacoAir left 06:37 denisboyun joined 06:41 denisboyun left 06:51 kaleem joined 06:57 darutoko joined 07:07 SamuraiJack_ joined, atroxaper joined 07:09 SamuraiJack left 07:13 zakharyas joined
atroxaper Hello #perl6 ! 07:14
07:15 kaleem left
atroxaper Does anybody know, does moritz has plans for making R* release for windows? 07:15
nwc10 atroxaper: I believe that usually Pm has done that 07:17
atroxaper Hope so... 07:18
Hello pmichaud. Will you build windows R* release? Please ;) 07:19
07:19 kaleem joined 07:28 kaleem left 07:29 kaleem joined
moritz just for the record, I don't have access to any windows machine with the appropriate development/compilation toolchain 07:29
if anybody wants to finance a sufficiently powerful laptop for dual boot and windows development, I might reconsider :-) 07:30
07:31 popl left
atroxaper moritz: ok) 07:31
07:36 daxim joined
FROGGS_ moritz: I have a windows7 + visual studio, if there is a guide I'll do my best 07:42
moritz FROGGS_: the Makefile has an 'msi' target 07:45
FROGGS_ hopefully it complains nicely about missing tools :o)
masak morning, #perl6 07:50
FROGGS_ o/
07:52 ssutch left 07:54 ssutch joined, odoacre left 07:59 dmol joined 08:05 hummeleB1 joined, cognominal joined 08:06 ssutch left 08:07 odoacre joined 08:08 fridim__ left
moritz .seen Mouq 08:10
yoleaux I saw Mouq 25 Sep 2013 04:42Z in #perl6: <Mouq> Oh, cool, I understand. Thank you TimToady
08:14 FROGGS_ left 08:16 hummeleB1 left 08:17 iSlug joined 08:27 daxim_ joined, daxim left
lizmat good *, #perl6! 08:30
moritz good morning, lizmat
masak lizmat! \o/ good morning! 08:31
lizmat jnthn: re jnthn: re irclog.perlgeek.de/perl6/2013-09-26#i_7637196 , the pop_phaser approach is only because there is currently no way (that I know of) to create a "remove_phaser" method that accepts the block to be removed
geekosaur: note that the initial aproach *did* have a once block insider an ENTER phaser 08:32
disadvantage of that, is that you will still run the ENTER phaser each time the sub/method is called
my approach would be that the ENTER phaser is only run once, to tell the world the sub/method has been deprecated
atroxaper lizmat o/ 08:33
lizmat so you could argue that this is a premature micro optimization
otoh, having the capability that phasers disable themselves, might be a good thing to have anyway 08:34
masak lizmat: sounds like you're sniffing very close to the "handle" model/solution.
i.e. at the point of adding the phaser, a handle object is returned, a kind of IDisposable that knows how to un-add the phaser. 08:35
we do that for function wrapping, which means a pancake can be removed from the middle of the pancake stack. it's quite pretty.
(it's also possible I'm completely off the mark here... but a quick skim of the backlog caused me to think of handles.) 08:39
08:40 sqirrel joined
atroxaper r: for ((1...3) X (3...1)) Z ^Inf -> $i, $j, $n {say "-$n|$i:$j-"} 08:41
camelia rakudo a79c94: OUTPUT«-3|1:0-␤-2|1:1-␤-1|2:3-␤-5|4:1-␤-3|2:6-␤-8|7:2-␤-2|2:9-␤-11|10:1-␤-3|3:12-␤-14|13:3-␤-3|2:15-␤-17|16:1-␤»
atroxaper Am i doing something wrong? 08:42
masak atroxaper: I think the tuple doesn't flatten like you think it will. 08:46
atroxaper masak: i just want to X two arrays and add index for that. 08:48
r: for ((1...3) X (3...1)) -> $i, $j {say "-$i:$j-"} 08:50
camelia rakudo a79c94: OUTPUT«-1:3-␤-1:2-␤-1:1-␤-2:3-␤-2:2-␤-2:1-␤-3:3-␤-3:2-␤-3:1-␤»
atroxaper But how can i and index for that ?! ) 08:51
masak r: say ((1..3 X reverse 1..3) Z ^Inf).perl
camelia rakudo a79c94: OUTPUT«((1, 0), (3, 1), (1, 2), (2, 3), (1, 4), (1, 5), (2, 6), (3, 7), (2, 8), (2, 9), (2, 10), (1, 11), (3, 12), (3, 13), (3, 14), (2, 15), (3, 16), (1, 17)).list␤»
masak there's your problem. 08:52
r: class Blorb { trusts ::Blorb::Handle }; class Blorb::Handle { has &!remove; submethod BUILD(:&!remove) {}; method remove { &!remove() } }; class BlorbHolder { method add($blorb) { say "$blorb added"; return Blorb::Handle.new(:remove({ self.remove($blorb) })) }; method remove($blorb) { say "$blorb removed" } }; my $handle = BlorbHolder.new.add(Blorb.new); $handle.remove
camelia rakudo a79c94: OUTPUT«Blorb<-25000658> added␤Blorb<-25000658> removed␤»
masak lizmat: ^^
08:53 BooK left, mtk left, FROGGS joined 08:54 BooK joined
atroxaper masak: i should think about that. Thank you for hint! 08:54
masak essentially, tuple parentheses are too "soft" and they "break" when you Z on them. 08:55
r: for ((1..3 X reverse 1..3).tree Z ^Inf) -> [$i, $j], $n { say "($i, $j) -- $n" } 08:56
camelia rakudo a79c94: OUTPUT«(1, 3) -- 0␤(1, 2) -- 1␤(1, 1) -- 2␤(2, 3) -- 3␤(2, 2) -- 4␤(2, 1) -- 5␤(3, 3) -- 6␤(3, 2) -- 7␤(3, 1) -- 8␤»
masak bows
.tree did le trick.
moritz it usually does :-) 08:57
.list in one way, .tree in the other
masak right. 08:58
that accords with my mental model.
and X and Z tend to .list if they have to.
just like many of the other primitives.
08:58 fhelmberger joined
atroxaper Wow! 08:58
Cool!
masak so .tree becomes, semantically, "there is some interesting structure in there that I'd like to turn into []s"
like, "harden" the parentheses. 08:59
08:59 mtk joined
lizmat masak: it is indeed very close to the handle solution 08:59
however, the trait_mod doesn't have anywhere to return to 09:00
09:00 mtk left
lizmat the return value of adding the ENTER phaser would be needed *inside* that phaser 09:00
I guess we could do something like:
09:01 mtk joined
lizmat my $handle; $handle = $r.add_phaser('ENTER', { $handle.remove } ) ? 09:01
masak that looks... one step too superfluous, somehow. 09:05
specifically, having to manually specify the removal.
09:05 kaleem left
lizmat well, I want something that will not interfere with normal phaser execution 09:06
so checking a flag with every phaser execution, is a nono in my book
masak *nod* 09:08
lizmat adding some overhead when creating a self-annihilating phaser, is ok
masak "Remember that Rust is still a rapidly changing language, and releases have very little connotation of stability, support, or any other guarantees." -- news.ycombinator.com/item?id=6454443
see also the child comment, about "maturity #5 - production ready" 09:11
it would be interesting to map Perl 6 to those maturity levels. 09:12
nwc10 I don't think that it's possible to generate a URL for this page that defaults it to the "3 year" range filter: www.ohloh.net/p/parrot/commits/summary 09:16
I think that there is now more activity on www.ohloh.net/p/moarvm/commits/summary + www.ohloh.net/p/nqp/commits/summary + www.ohloh.net/p/rakudo/commits/summary than on www.ohloh.net/p/perl/commits/summary 09:18
09:18 kaleem joined
masak nwc10++ # digging up these 09:19
nwc10 I had been wondering for a couple of days if aggregate perl6 (core) activity is now more than perl5 (core)
(sorry, chromatic, I missed the spaces)
09:20 kaleem left
masak nwc10: that would be... very odd. 09:20
09:20 kaleem joined
nwc10 not necessarily 09:20
masak I can imagine that it's *true*, mind you.
but it still feels extremely strange. 09:21
nwc10 1) it's self hosted. It's more accessible
2) There is more obvious stuff to *do*
3) It's far more "progress" than "maint"
masak or, in other words: 1) easy onramps, 2) fun, 3) goal 09:22
it suddenly strikes me that we should milk 3) for all its worth. 09:23
jnthn o/
masak I mean, I've realized before that we need to set milestones and establish where we're going. (and we have, to an extent.)
but maybe it's the biggest unbound parameter at our disposal. 09:24
to *blast* out our current status and what needs to be done to reach the nice goal, to attract contributors.
nwc10 yes. I'd not thought of it like that
masak the message not being "hey, look how done we are" 09:25
but "hey, look how on the way to done we are -- care to give a hand?"
we blog till we're blue in the face trying to show how feature-complete and stable Perl 6 is.
maybe that's not what will attract people.
nwc10 maybe. But I think that a decade ago perl 5 sucked in contributors (at least partly) because enough of it worked to get them using it, but stuff they *also* wanted was broken and needed fixing 09:26
masak maybe the sense of being a settler, an early adopter, will.
right.
lizmat afk
masak the right mix of "I can use this" and "wouldn't it be nice if..." 09:27
nwc10 so I'm wondering if the road to getting more contributors is to suck them in from users
bonsaikitten are there any "frameworks" in perl6 ?
stuff that I'd casually use because it's awesome (and not stuff I need to write myself)
nwc10 is Rakudo on JVM now good enough to port ilbot to Perl 6? 09:28
jnthn .tell lizmat The remove the phaser thing is a premature opt. I've only been saying as much for the last week. :( 09:29
yoleaux jnthn: I'll pass your message to lizmat.
jnthn .tell lizmat Not to mention that working to make *deprecated* things faster feels odd. The point is for people to stop using them; "they're an iota slower 'cus of the deprecation warning" is a motivator to stop using 'em. 09:32
yoleaux jnthn: I'll pass your message to lizmat.
jnthn nwc10: It has the sockets for it... 09:34
masak if it's time to port ilbot to Rakudo/JVM, I'm willing to head or participate in a hackathon that does this. 09:44
some suitable weekend or other.
09:47 darutoko left
nwc10 beware that I'm saying this things from a somewhat outside/high level/unfamiliar with the code/probably not able to help do work 09:48
but I would hope that first ilbot and then later the website publishing atop Perl 6 would 09:49
a) generate useful reusable code
b) be a prominent example of dogfood
c) it would amuse me if the various Perl 5 IRC channels were being logged by Perl 6
(and the NSA, and GCHQ, and the Chinese, etc)
(and no, I'm not going to say anything stupid because the CIA are just down the hill from me, and I don't want the black segways to come visiting) 09:50
at least, I hope it would be that, and not predator drones 09:51
jnthn oh noes you just disclosed their secret location!
masak you never know these days.
nwc10 jnthn: the secret/"secret" has been an utter farce over the past month
apparently it ended up with 100 "friends of architecture" coming to photograph it: 09:53
derstandard.at/1378248419619/Ein-vi...r-NSAVilla
FROGGS hehe
nwc10 the building in question has a police cubical outside, a security guard inside, bristling with CCTV cameras, and *no name plate* 09:54
which immediately gives it away as interesting
so I asked, and a friend said "it's the CIA building. Everyone round here knows that"
the couldn't have made it more obvious if they tried
jnthn haha 09:55
fail
09:55 kivutar joined
FROGGS but that is really a nice building :o) 09:56
nwc10 goo.gl/maps/tptyE and it's not blurred out
10:02 hypolin left 10:07 sqirrel left
masak thinks that people in the west are too happy with their iPods, TV series and soda machines to ever rebel against their government, no matter how power-hungry, privacy-grabbing, or infringing on human rights 10:09
en.wikipedia.org/wiki/Bread_and_circuses has been refined to the point where power doesn't have to worry. 10:10
10:11 atroxaper left
masak I would be outraged about it, but I'm too busy watching "Boston Legal" right now. 10:12
jnthn tbh, I don't think it's "just the west". "We'll let you make money and have iPods if you leave us to the politics" is working reasonably well in China. :)
masak that's because China has becoming more western (in that sense) for the past 20 or so years :)
jnthn Aye 10:13
masak become*
jnthn Ah, you meant "west" as in "value wise" rather than "geography wise"...
masak also, having a populace leave you alone with your politics is especially easy in a country where for the past 40 years or so, *everything* was about politics, like it or not. 10:14
10:21 jnap joined 10:23 sqirrel joined 10:33 darutoko joined
dalek kudo/nom: 8182d13 | (Elizabeth Mattijsen)++ | src/core/ (2 files):
Just use a once block for is DEPRECATED and be done with it
 10:35
jnthn lizmat++
lizmat mind you, I think it would be good to have add_phaser return a Phaser object, that you could call .remove on
yoleaux 09:29Z <jnthn> lizmat: The remove the phaser thing is a premature opt. I've only been saying as much for the last week. :(
09:32Z <jnthn> lizmat: Not to mention that working to make *deprecated* things faster feels odd. The point is for people to stop using them; "they're an iota slower 'cus of the deprecation warning" is a motivator to stop using 'em.
10:46 jnap left
masak an iota of a motivation. 10:55
lizmat that much ?
sorry, can't be less handwavy than that 11:00
11:05 pmurias joined 11:13 sqirrel left 11:15 rindolf joined 11:17 kivutar left 11:18 jnap joined
Woodi hallo :) 11:29
lizmat Woodi o/
Woodi after opening that NSAvilla link my MS occupiet box frozen hard... maybe it was contacting "home" ? ;) even ctrlAltDel didn't worked :) 11:31
11:31 samri joined
samri helo, what is best implementation of perl6 to install and are there any book to buy about it? i checked oreilly i didnt find any.. 11:32
tadzik Rakudo is the most complete implementation, samri
at the moment, there are no books to buy. There is one free book, a draft actually
it might be a bit outdated, but you may find it to your liking. It's listed on perl6.org/documentation 11:33
samri so there is no up to date references on this language?
Woodi but on topic: Perl6 need GUI, IMO. like Java and C# - without GUI their instant-like market acceptation would be imposible...
FROGGS C:\Perl\bin\perl.exe -MExtUtils::Command -e chmod 755 modules/rakudo-debugger/perl6-debug.exe 11:34
Cannot chmod 755 modules/rakudo-debugger/perl6-debug.exe:No such file or directory at C:/Perl/lib/ExtUtils/Command.pm line 274.
moritz samri: there is reference at perlcabal.org/syn/
Woodi: then write one
FROGGS (when building on windows to make the msi)
tadzik Woodi: it should be quite possible to do a GUI these days, talking to JVM classes 11:35
well volunteered :)
FROGGS moritz: I could create the msi but the debugger would not be included 11:36
samri just got this when searched:
O'Reilly Media and the authors of the current Learning Perl are looking foward to a book on Perl 6 when it's stable. Honestly, we started a long time ago, twice actually, and everything we wrote is now obsolete or has disappeared from the language. It's our hope that someday they'll settle on the final language, allowing us time to make the book. When we start writing the book, we'll use this domain to let you participate in the proce
so i guess it is too early to learn it..
moritz FROGGS: well, fixing the build would be preferable, of course :-) 11:37
FROGGS: maybe you need 'make install' first?
FROGGS moritz: I guess the debugger is still broken 11:38
11:38 mkp19901 joined
FROGGS this was `nmake install` btw 11:38
11:38 jnap left, mkp19901 left
moritz :/ 11:38
11:39 samri left
FROGGS moritz: at least panda seems ok now, it doesn't copy the panda.bat by default to the bin dir 11:41
11:41 sqirrel joined
moritz FROGGS: anyway, I'd prefer a .msi without perl6-debugger to no .msi at all 11:44
FROGGS moritz: I am uploading a RC atm 11:45
11:45 BabsSeed left
moritz \o/ 11:46
FROGGS moritz: so if you want to buy another machine, install windows and try it you are welcome :o) 11:47
nwc10 an .msi without a debugger can be replaced with an .msi with a debugger
(if the problem is in the .msi build)
FROGGS: IIRC he said he was quite happy for *someone else* to buy him another
FROGGS nwc10: the debugger itself is broken on all platforms
nwc10 oh :-(
jnthn huh...
I only fixed it a few weeks back... 11:48
nwc10 although it's probably cheaper to buy beer to distract jnthn and use his machine while he's busy
jnthn What's happened to break it?
FROGGS jnthn: I'll give you a paste in a few mionutes
-o
lizmat probably broke it with method2sub branch ? 11:49
moritz lizmat: that was merged after the rakudo release, no?
lizmat yes
ok
moritz lizmat: then it wouldn't be broken in R*
lizmat then no :-)
nwc10 bisect! bisect! 11:50
jnthn fwiw, I just built it against a Rakudo and NQP built yesterday and it runs OK here.
nwc10 www.codinghorror.com/blog/2007/03/t...ogram.html
jnthn (using Makefile) 11:51
nwc10: :P
dalek kudo/nom: 575afc0 | (Elizabeth Mattijsen)++ | src/Perl6/Optimizer.nqp:
Make sure a problem in reporting, does not inhibit the report
 11:53
jnthn lizmat: You didn't need the curlies there, I think... ;) 11:54
lizmat: What did you find to trigger that, ooc? Something in setting building?
lizmat yes 11:55
jnthn lizmat: OK.
If it was triggerable in some other way it'd deserve an RT... :)
lizmat no, it was me being stupid in the setting 11:56
basically, optimizer telling me that DEPRECATED() should be called with 2 parameters
instead of one
jnthn Nice :)
Yeah, as it goes about doing inline analysis it can also spot things that'll never possibly work :)
lizmat and that erorr message bombed 11:57
before: Method 'Str<3477132080187517822>' not found for invocant of class 'Perl6::Metamodel::ClassHOW' 11:58
after: CHECK FAILED:
Calling 'DEPRECATED' will never work with argument types (str) (lines 1415, 1415, 6510, 6510, 7422, 7422, 8446, 8446, 8450, 8450, 8862, 8862, 9123, 9123, 14980, 14980, 15302, 15302)
Expected:
11:59 kaleem left
FROGGS jnthn: that looks more like a pandabug: gist.github.com/FROGGS/4cb5ec025cf7e2203908 11:59
jnthn Hm...why's it dupe...
the line nums
lizmat because it expected 2 params ? and I only gave it one ?
jnthn Feels odd
FROGGS even even 12:00
jnthn No, it should report once per call
lizmat running spectest now, will check in a bit
jnthn I mean, we can squish 'em but I wanna know where they're coming from. :)
oh neins...zugriff verweigert 12:01
But yeah, that's while installing Term::ANSIColor...
moritz jnthn: tray again please
*try
12:01 ksh left 12:02 skids left
jnthn ah, it means "access denied" 12:02
Odd...
I think I've seen that happen now and then on Windows with Panda... 12:03
dalek kudo/nom: 0327e30 | (Elizabeth Mattijsen)++ | src/Perl6/Optimizer.nqp:
jnthn++ tells me the {} were not needed
12:04 brrt joined
lizmat r: sub a ($a) {}; a("a","b","c","d") 12:04
camelia rakudo 8182d1: OUTPUT«===SORRY!===␤CHECK FAILED:␤Calling 'a' will never work with argument types (Str, Str, Str, Str) (lines 1, 1)␤ Expected: :($a)␤»
lizmat seems to always double, regardless of # params
jnthn yeah 12:05
12:05 iSlug left
jnthn Will look later... 12:05
FROGGS hmm, maybe its cwd is wrong and it can't find the .work? 12:08
jnthn That'd probably not give access denied, though?
FROGGS yeah, and it checks -d on that path before deleting it :/ 12:11
ahh, wait
dalek kudo/nom: 155487c | (Elizabeth Mattijsen)++ | src/core/ (20 files):
Mark "exists" as DEPRECATED, change all calls of .exists to .exists_key|pos
 12:12
lizmat errands&
12:14 denisboyun joined
JimmyZ wonders why exists is DEPRECATED 12:20
moritz JimmyZ: to emit deprecation warnings 12:21
JimmyZ I meant why 'exists' is not used 12:22
tadzik :)
moritz JimmyZ: because it's ambiguous 12:23
positional vs. associative
JimmyZ looks like it is 12:26
FROGGS jnthn: yes, doing `nmake` works... 12:28
[Coke] waves to the room before dayjobbing. 12:29
yoleaux 26 Sep 2013 23:48Z <sisar_> [Coke]: thanks for looking into it, but no luck. It fails with: sprunge.us/fVAH. I'm unfamiliar with the build process, but I thought that invoking '--gen-parrot' would automaticall build nqp too.
jnthn waves at [Coke] while dayjobbing
JimmyZ waves at [Coke] after dayjobbing 12:30
jnthn Timezones! \o/ 12:31
nwc10 merely waves
[Coke] .to sisar - hurm, maybe I was looking at the wrong line on the initial paste there. what version of rakudo are you using? what platform? have you tried removing the existing nqp directory and trying again? 12:32
yoleaux [Coke]: I'll pass your message to sisar.
masak .oO( if you wave at someone while dayjobbing, does the observer experience the Jobbler effect? ) 12:33
FROGGS iiiiiieeeeeeeeeeaaaaaaaaauuuuuuuuuu 12:39
12:39 kivutar joined
daxim_ aioeu 12:40
FROGGS no
well, maybe it is aioeu in austria :o) 12:41
12:49 xenoterracide_ joined 12:52 iSlug joined 13:04 jnap joined 13:05 raiph joined 13:08 ssqq joined 13:09 lowpro30 joined
ssqq #parrot channel no person 13:16
13:16 lowpro30 left
moritz ssqq: try #parrot on irc.perl.org 13:17
though last I looked there, not much was going on
ok, 3 days ago there was a bit of actual parrot discussion 13:18
13:19 sqirrel left
ssqq thanks moritz, If the project of parrot have some problem? 13:19
moritz ssqq: lack of active contributors and development 13:21
ssqq moritz: If the parrot have been stable enough? 13:22
moritz ssqq: stable enough for what?
13:22 ajr joined
ssqq moritz: stable enough for run Rakudo. 13:23
13:23 ajr is now known as Guest84621
ssqq The network name of irc.perl.org is? 13:23
moritz ssqq: it runs rakudo, yes. Though not as fast as we would like :-)
13:23 prevost joined
moritz ssqq: magnet 13:23
JimmyZ ssqq: PerlVim, 你好 13:24
13:24 rindolf left
ssqq JimmyZ: How are you 13:24
JimmyZ: Nice to see you here.
JimmyZ er
ssqq thanks moritz. 13:25
JimmyZ ssqq: 你的 makefile 实现怎么样了 13:27
13:29 Guest84621 is now known as ajr_ 13:33 skids joined 13:34 GibbaTheHutt joined, sqirrel joined 13:40 btyler joined 13:41 kaleem joined, kivutar left
ssqq JimmyZ: makefile 早就不写了。 13:41
FROGGS jnthn: star's Makefile is to blame about the perl-debug build. it prints the the statements but it does not execute them
jnthn huh? :) 13:42
FROGGS jnthn: I temporary prefixed them with cmd /c, and it works
jnthn ok :)
.oO( Makefail )
FROGGS yeah
no idea why it does it like that
k, creating another RC of the msi :o) 13:43
13:43 iSlug left
ssqq JimmyZ: 我原来以为一个内置的 make 会大大简化一个解释器的模型。 13:44
FROGGS moritz: SCHNARRENBLARKEN make SCHMETTERLING KAMPFPANZERWAGEN!! 13:45
:P
<--- coffee 13:46
ssqq JimmyZ: 现在发现影响一个语言解释器的最重要的东西是语法。
JimmyZ: 一个复杂语法的语言，最终将处理一个非常复杂的数据结构。 13:47
JimmyZ ssqq: 呵呵。我之前就讲过不可行了 13:48
13:50 cognominal left
JimmyZ FROGGS: try add 'SHELL = cmd' to Makefile 13:50
FROGGS too bad, windows is already shutting down :o)
I'll take a note and try later, thanks! 13:51
ssqq JimmyZ: Perl 6 在正则表达式和语言表达方面的能力越强大，Perl 6能参与进来的人就越少。因为太复杂了。
JimmyZ ssqq: 你多虑了 13:52
timotimo perl6 is not too complicated! ;)
JimmyZ .oO( timotimo knows about chinese ... )
timotimo does that concern perl6 the language (to users) or rakudo the implementation (to core developers)? 13:53
ssqq JimmyZ: 我最近看了非常精简的 Lua，非常简洁的 Ruby， 和扩展性很好的 Lisp。
timotimo 我不知道多少中国人。谷歌翻译帮助我了解。
翻译结果不包含“谷歌”或“翻译”。我很高兴。 13:54
没有人会了解我的秘密。
JimmyZ hehe
brrt can't really follow this discussion anymore
timotimo huehuehuehuehue
(from another asian country)
13:55 iSlug joined 13:56 ssqq left
dalek p: bc537f3 | Mouq++ | src/QRegex/P6Regex/Actions.nqp:
Fix bug with integer ** quantifiers

  / a ** 2_0 / previously matched "a" two times.
Now uses the .ast of <decint>.		 13:56
timotimo cool! :)
13:56 ssqq joined
jnthn oops :) 13:56
We probably need tests for that :)
tadzik wow :) 13:57
Mouq++
JimmyZ ssqq: 有空了解一下 MoarVM 13:58
FROGGS nqp: say("aaaa" ~~ / a ** 2_0 .. 4_0 /) 13:59
camelia nqp: OUTPUT«Spaces not allowed in bare range at line 2, near " 4_0 /)"␤current instr.: 'panic' pc 14693 (src/stage2/gen/NQPHLL.pir:5223) (src/stage2/gen/NQPHLL.nqp:279)␤»
FROGGS nqp: say("aaaa" ~~ / a ** 2_0..4_0 /)
camelia nqp: OUTPUT«aaaa␤»
FROGGS alright
13:59 kivutar joined
FROGGS now I see the min.ast in the diff too :o) 13:59
jnthn nqp: say("aaaa" ~~ / a ** 4_0..2_0 /) 14:00
camelia nqp: OUTPUT«Empty range at line 2, near " /)"␤current instr.: 'panic' pc 14693 (src/stage2/gen/NQPHLL.pir:5223) (src/stage2/gen/NQPHLL.nqp:279)␤»
jnthn That also needs a fix.
It's in Grammar.pm
no, wait, I meant
FROGGS LTA?
jnthn nqp: say("aaaa" ~~ / a ** 4..2_0 /)
camelia nqp: OUTPUT«Empty range at line 2, near " /)"␤current instr.: 'panic' pc 14693 (src/stage2/gen/NQPHLL.pir:5223) (src/stage2/gen/NQPHLL.nqp:279)␤»
jnthn That's the one. That should not be an error.
FROGGS ohh
14:02 ssqq left
FROGGS ohh, call for papers for the german perl workshop 2014 in my inbox 14:03
moritz: you are going to come to hannover in march? 14:05
timotimo: you? 14:06
I don't need to ask liz I guess :o)
timotimo hm, hannover is pretty far for me (i'm not an airplane person (though how cool would that be?)), so probably not 14:07
FROGGS the perl workshop in berlin was awesome!
sad that there are usually no hackathons 14:08
tadzik there was one in Erlangen, didn't attract many people
I think it attracted 3: me, moritz and jnthn
FROGGS 130ppl in berlin
*g*
tadzik add masak to that, and that's 4 people :)
FROGGS still LT "A".ord 14:09
jnthn Hm, I guess Hannover is a shortish train journey from Hamburg? 14:10
FROGGS jnthn: correct :o)
jnthn Hm, so it's trainable from here if I wanted to come.
FROGGS that would be nice 14:11
if you are not going to come I'll give a perl6-status talk too
... in germish
jnthn You could do that anyway :)
FROGGS yeah
jnthn And I could find other things to talk about
I'm pretty sure I won't talk about anything in germish :) 14:12
FROGGS I have time till mid december to decide what to say
(and how to say it)
timotimo hmm, when jnthn needs to "find other things to talk about" he usually comes up with something that leaves mouths agape :P 14:13
hoelzro: where do we have to look to find out the current progress on the perl6 syntax highlighting on github?
FROGGS yeah, and that means he will needs something extra-ordinary to hack on a few weeks before the conference 14:14
like... what if bytecode could hack for us? and then we need to call the hacking bytecode "backer" or so
timotimo: I guess first v1.7 of pygments must be releases: bitbucket.org/birkenfeld/pygments-...ip/CHANGES 14:17
timotimo github.com/tmm1/pygments.rb/pull/73 - this is the PR we were all looking at
14:18 PacoAir joined
PerlJam timotimo: Add a comment to the PR shaming github for lagging behind bitbucket? ;) 14:19
GlitchMr rn: say sort <1 10 2>
camelia niecza v24-95-ga6d4c5f: OUTPUT«1 2 10␤»
..rakudo 155487: OUTPUT«1 10 2␤»
TimToady I suppose out-of-sync highlighting would have to be called 色ばけ
GlitchMr rn: say <1 10 2>.perl
camelia rakudo 155487: OUTPUT«("1", "10", "2")␤»
..niecza v24-95-ga6d4c5f: OUTPUT«(val("1"), val("10"), val("2"))␤»
timotimo PerlJam: i'm not sure i understand
14:22 iSlug left
PerlJam timotimo: apparently bitbucket.org already does perl6 syntax highlighting 14:22
14:23 kaleem left
timotimo it ... does!? 14:24
that's neat!
do you have an example link?
PerlJam timotimo: no, just irclog.perlgeek.de/perl6/2013-09-17#i_7596584 14:25
timotimo bitbucket.org/birkenfeld/pygments-...at=default
14:25 sqirrel left, logie left
timotimo you can't spell properly without perl 14:26
14:27 kivutar left
jnthn nice :) 14:27
FROGGS absolutely
TimToady you can't spell nice without c 14:28
PerlJam timotimo: superlative ;) 14:29
FROGGS /o\ now I got it
diakopter you can't spell without the
TimToady sure you can, just use the other one instead 14:30
14:30 SamuraiJack_ left
timotimo FROGGS: not many of those work on two levels like this one 14:30
diakopter no, I meant *you* O_O
pmurias Woodi: re GUI, isn't most new GUI stuff done as web apps? ;)
timotimo we have SWT on jvm ;) 14:31
14:33 rindolf joined, fhelmberger left 14:34 prevost left
GlitchMr ($by.?count // 2) < 2 14:36
What does .? syntax mean?
TimToady lizmat: actually, we should pessimize the deprecated function, not be trying to optimize it; maybe throw in a sleep 1 into the function we're deprecating :)
GlitchMr $by is subroutine.
timotimo if the call doesn't succeed (no method found or something) return Nil instead
r: say "foo".?heeeyooooo() 14:37
camelia rakudo 155487: OUTPUT«Nil␤»
timotimo r: say "foo".heeeyooooo()
camelia rakudo 155487: OUTPUT«No such method 'heeeyooooo' for invocant of type 'Str'␤ in block at /tmp/4AOZFBYVix:1␤␤»
GlitchMr oh,ok
So it's allowed to not exist.
timotimo probably also turns exceptions into nil
PerlJam GlitchMr: aye, it doesn't die when the method doesn't exist
timotimo r: class A { method test() { die "oh noes!" } }; say A.test()
camelia rakudo 155487: OUTPUT«oh noes!␤ in method test at /tmp/PMT_nyFXyS:1␤ in block at /tmp/PMT_nyFXyS:1␤␤»
timotimo r: class A { method test() { die "oh noes!" } }; say A.?test() 14:38
camelia rakudo 155487: OUTPUT«oh noes!␤ in method test at /tmp/x5_PScbt8O:1␤ in block at /tmp/x5_PScbt8O:1␤␤»
timotimo ah. nope.
14:38 daxim_ left
jnthn No, it's just about if a method with the name exists 14:38
FROGGS I guess it is like: A.can('test') ?? A.test() !! Nil
14:38 daxim_ joined
jnthn Yes, but a tad more efficient. :) 14:39
FROGGS sure
lizmat TimToady: first make sure the spectest is not calling any deprecated functions, then I'll add the sleep 1 :-)
TimToady r: try "foo".heeeyoooo()
camelia ( no output )
lizmat $ perl6 -e 'my %h; say %h.exists("a")' 14:40
Method 'EnumMap.exists' has been deprecated, please use the :exists adverb instead in block at -e:1
instead of mentioning exists_key
14:41 cognominal joined
dalek kudo/nom: 384a0fa | (Elizabeth Mattijsen)++ | src/core/ (8 files):
Mention the :exists adverb, instead of 'exists_key'
 14:41
14:42 thou joined
TimToady I'm sure the user felt they were calling Hash.exists, not EnumMap.exists :) 14:44
so maybe the message should look at the actual type of the invocant, not the static type 14:45
or maybe it should put both if they differ
lizmat will fix after cycling& 14:46
dalek p: 8bcdb14 | Mouq++ | src/QRegex/P6Regex/ (2 files):
Fix empty range error in ** quantifier
14:52 jnap1 joined 14:54 jnap left 15:09 jnap1 left 15:10 sqirrel joined 15:12 jnap joined
moritz FROGGS: I'm not sure I'll make it to GPW2014, due to the dates 15:12
FROGGS moritz: some day we will make it to the same conference ;o) 15:13
moritz FROGGS: or we just visit each other :-)
FROGGS: where are you based?
FROGGS moritz: near Berlin 15:14
.tell BenGoldberg This issue might need to go to RT (via [email@hidden.address] github.com/rakudo/star/issues/31 15:15
yoleaux FROGGS: I'll pass your message to BenGoldberg.
FROGGS brb 15:16
15:16 FROGGS left 15:17 FROGGS joined
FROGGS btw: froggs.de/perl6/rakudo-star-2013.09-RC1.msi 15:17
spread the world! o/
15:17 FROGGS left
moritz should I upload that to rakudo.org? 15:17
15:20 rindolf left 15:21 kaare_ joined 15:23 krokite joined 15:24 brrt left 15:32 zakharyas left 15:34 benabik joined 15:41 raiph left 15:47 kivutar joined 15:50 sisar joined
sisar so I deleted my rakudo directory, started with a fresh clone, but I'm still unable to build nqp: sprunge.us/AdSb . Any pointers as to what may be wonrg ? I'm using Cygwin (32-bit) on Windows 7 (64-bit). 15:52
yoleaux 12:32Z <[Coke]> sisar: - hurm, maybe I was looking at the wrong line on the initial paste there. what version of rakudo are you using? what platform? have you tried removing the existing nqp directory and trying again?
15:54 GibbaTheHutt left 15:55 xenoterracide_ left
sisar I'm behind a HTTP proxy, maybe that is the problem ? However, I've set git to use the proxy, and it works. git was able to fetch the rakudo repo over https, and was able to fetch the parrot repo too. 15:55
I also used the command "git config --system url."".insteadOf git://" beacuse the git:// protocol is unable to go through my institute's proxy. 15:57
benabik It doesn't seem to be related to network issues. 15:58
It's complaining that /home/siddhant/rakudo/nqp/install/ doesn't exist. 15:59
16:00 sqirrel left
benabik But parrot seems to have installed itself into /home/siddhant/rakudo/install/, so I don't know why NQP is trying a different directory. I'm guessing a --prefix when missing somewhere? 16:00
16:01 kivutar left
sisar I'm sorry, I don't understand. How would I fix that ? 16:01
benabik sisar: I think it's a bug in the Configure.pl script. NQP is trying to install itself somewhere different than parrot. Hopefully someone who knows the scripts a bit better can try to fix it. (I don't have time right now.) 16:03
sisar benabik: Ah. Ok. Thank you :)
16:07 krokite left
moritz or maybe interference with a previously installed nqp or parrot 16:12
sisar moritz: where do I look for that ? There's nothing in /bin 16:17
moritz sisar: do a find ~ -name parrot -type f for example 16:22
sisar: also, which parrot
16:22 dmol left
TimToady supposes we'll have to define a FatNum to pick the largest supported floater, since num80 and num128 are arguably overspecific to be portable 16:26
sisar moritz: find ~ -name parrot -type f has no output, and output of which parrot is "/home/siddhant/rakudo/install/bin/parrot" 16:29
[Coke] sisar: please try: 16:30
perl Configure.pl --gen-parrot --gen-nqp
16:30 sqirrel joined
sisar gives it a try 16:30
[Coke] (you might also need to rm -rf nqp install parrot if you have a botched install in progress) 16:31
16:36 xenoterracide_ joined 16:40 bluescreen10 joined
[Coke] diakopter: can you think of anything happening on host06 that would cause the rakudo.jvm run (using t/spec/test_summary.pl) to abort every test after N tests are run? 16:53
the evalserver is still running, but everything aborts. no diagnostic output anywhere.
diakopter no..
[Coke] hokay. 16:54
diakopter :(
[Coke] been going on for about a week. hopefully will be able to dig into it this weekend.
16:54 sisar left
diakopter [Coke]: maybe try on host07? 16:55
[Coke] eh. I can try on my laptop, too.
-> 16:56
16:59 FROGGS joined
FROGGS o/ 16:59
moritz: would be nice if someone could test it before uploading it... 17:01
17:18 btyler left 17:20 ajr_ left 17:21 ajr joined 17:22 ajr is now known as Guest85219, Guest85219 is now known as ajr_
jnthn FROGGS: It just needs installing? 17:25
FROGGS: I can download and run the MSI here if you like...
17:28 ajr_ left 17:29 popl joined
jnthn FROGGS: Hm, it comaplains "Could not load oplib `nqp_ops'" here 17:30
dalek rl6-Acme-Addslashes: b1e401e | (Konrad Borowski)++ | lib/Acme/Addslashes.pm:
Use `s///` notation instead of `subst`
 17:31
jnthn FROGGS: And moving my existing installed one out the way didn't help...
FROGGS jnthn: :/ 17:32
jnthn: I dunno, I did nothing special 17:33
jnthn FROGGS: I see the issue, I think 17:35
libdir => 'C:/rakudo-star-2013.09/install/lib'
If you run parrot_config.exe that is in there
But it installs it to c:\rakudo\bin\lib
oops
c:\rakudo\lib 17:36
FROGGS ohh
hmmm
jnthn I think you may have to build it in the "right palce"?
*palace
*place
FROGGS yeah, looks like
jnthn grr, dinner, bbi20 :)
FROGGS k, time for RC2 :o)
popl smolder.parrot.org/app/projects/rep...ails/43699 # \o/
FROGGS ohh, we have smolder reports again 17:37
popl I just submitted that one.
FROGGS popl: awesome!
popl It took a long time to test. 17:38
FROGGS popl++
popl I was more than halfway finished with the testing yesterday but then the adapter plug fell out of my laptop and my machine lost power so I had to start over.
FROGGS ohh, hehe 17:39
popl some of the tests that failed have TODOs so I guess failure is not unexpected there. 17:45
17:49 rindolf joined 17:50 sqirrel left 17:56 btyler joined 17:57 bbkr joined, raiph joined 18:00 daxim_ left
timotimo is that the same thing as the results page that is hosted on some ip-only web address? 18:10
18:12 SamuraiJack_ joined, dmol joined
TimToady r/j seems to be testing out fine on my supercomputer at home; the ulimits are the same as host06, same arch, only difference is host06 is running 3.5.0-17-generic, while mine is running 3.8.0-19-generic 18:12
[Coke]: ^^ 18:13
18:15 rindolf left 18:22 logie joined 18:23 darutoko left
dalek kudo-js: bed6635 | (Pawel Murias)++ | runtime/.gitignore:
Add .gitignore for the generated unicode char name to codepoint database.
 18:25
kudo-js: aa0b73a | (Pawel Murias)++ | run (7 files):
Refactor ops that depend on node modules to seperate files.
kudo-js: 6d60cca | (Pawel Murias)++ | / (46 files):
Refactor the runtime in nqp-runtime-core (the platform indepdent part) and nqp-runtime-node (the node.js specific part).

The nqp-runtime loads the right runtime (currently always nqp-runtime-core). Fails some bugs due to a newly surfaced bug.
jnthn
.oO( Isn't faling bugs good? :) )
 18:26
18:26 odoacre left
diakopter lolz 18:26
pmurias heh
* fails some tests
GlitchMr perl6: say sort <file1 file2 file10 file20>
camelia rakudo 384a0f, niecza v24-95-ga6d4c5f: OUTPUT«file1 file10 file2 file20␤»
GlitchMr It makes sense, but well, if you want sort by ASCII. 18:27
18:27 btyler left 18:29 logie left, odoacre joined
FROGGS C:\MoarVM>perl6 -e "say 42" 18:41
PARROT VM: Could not load bytecode
Could not load oplib `nqp_ops'
jnthn: you do you check which path it is looking for?
jnthn FROGGS: parse fail 18:42
...oh, "how do you..."? 18:43
Run parrot_config --dump
FROGGS k
jnthn Installed in the bin folder
diakopter bring newspapers
FROGGS hmmm, C:/rakudo/... everywhere 18:44
18:45 odoacre left
jnthn FROGGS: Does running it from within the install dir work? 18:47
nwc10 FROGGS++ # SCHMETTERLING KAMPFPANZERWAGEN (even if it was a bit cheeky)
FROGGS there is no install dir, but it doesnt work from within C:\rakudo\bin 18:48
nwc10 unicode-- # no butterfly icon (yet)
FROGGS *g* 18:49
TimToady I guess they figure the CJK character is good enough
FROGGS jnthn: nqp_ops.dll is in C:\rakudo\lib\parrot\5.5.0\dynext though
TimToady well, at least they've got a bactrian camel
nwc10 I don't think that Ronja (sp?) cares for camels 18:50
FROGGS jnthn: okay, I think I move these to the bin folder now
TimToady so that gives 'em, what, another ten fifteen years or so to come up with a butterfly
FROGGS damn 18:51
C:\rakudo>perl6 -e "say 42"
"load_bytecode" couldn't find file 'ModuleLoader.pbc'
TimToady some of the cuneiforms look kinda like bugs 18:53
𒋧
18:54 dmol left
nwc10 TimToady: on some platforms C's long double is only 64 bytes (even if the CPU can do (or emulate) something bigger) 18:54
at least some ARM ABIs, and PPC OS X
lue nwc10: surely you mean 64 bits?
nwc10 not that PPC OS X will really be around when Christmas arrives
lue: yes, I do
naughty fingers
TimToady though mostly they look like fish
18:57 iSlug joined 18:58 odoacre joined
jnthn FROGGS: Oddness 19:01
What's the parrot_config --dump look like for you now?
FROGGS jnthn: what I said, all paths start with C:\rakudo\ 19:02
jnthn OK, then it..."should work" now :)
FROGGS yeah, but it does not 19:03
jnthn :S
FROGGS it is like if the parrot lib dir is wrong
let me check that
19:03 stevan_ left
TimToady oh, hey, today is Google's birthday :) 19:06
FROGGS hmmm
jnthn: prefix => 'C:/rakudo/install', but there is not install dir
PerlJam when's Perl 6's birthday? Is it still in gestation, or does the cup throwing incident count as "birth" ?
TimToady hah, the mouseover 19:08
FROGGS jnthn: there is an install dir when I create the msi, but after installing it it has no install dir 19:09
clearly I am doing something wrong
jnthn FROGGS: yeah, what --prefix are you passing when building?
Just C:\rakudo ?
FROGGS none
just perl Configure.pl --gen-parrot
jnthn Oh, you checked out into c:\rakudo and then build there?
Yeah...that won't work :)
FROGGS meh 19:10
I'd like to ha a guide
have*
jnthn Get a checkout somewhere, and then
perl Configure.pl --prefix=C:\rakudo --gen-parrot # or similar
And then it will use C:\rakudo as the install directory, not create a c:\rakudo\install 19:11
FROGGS hmmm
jnthn And yes, document what this ;-)
iSlug What are you coding actually?
I've no idea this night :/
FROGGS and the `nmake msi` target will find it and bundle it correctly?
jnthn FROGGS: Believe so
lizmat back from cycling but too tired to do any reliable programming tonight 19:12
jnthn looks like from makefile
lizmat going to do some r&r and then sleep&
jnthn
.oO( rum and rakudo )
 19:13
masak .oO( ribosomes and resistors ) 19:22
19:22 thundergnat joined, grondilu joined 19:24 telex left
thundergnat Good morning (UGT) #perl6 19:25
19:25 raiph left 19:26 telex joined
thundergnat I updated my rakudo install and ran into a bunch of problems with NQP and Rakudo installing in wrong directories. 19:26
FROGGS thundergnat: on what os are you on? 19:27
19:27 stevan_ joined
thundergnat FROGGS: Linux, ubuntu 12,04 19:28
Seems that if you have a DESTDIR set in your environment and DON"T specify a destination directory, it installs in odd ways.
Or, at least, that's what it seems like to me. 19:29
FROGGS hmmm, so you have destdir set to an existing path, but dont specify --prefix? 19:30
thundergnat Yes. At least that's what I did first.
FROGGS k
19:31 iSlug left
thundergnat DESTDIR is set for panda, but the Configure script grabs it too for SOME of the parrot / nqp / rakudo build... but not all of it. 19:32
19:33 SamuraiJack_ left
notjack quit 19:37
19:37 notjack left
jnthn .u 2205 19:38
yoleaux U+2205 EMPTY SET [Sm] (∅)
thundergnat To be more specific, with a env DESTDIR set , --gen-parrot and no --prefix, parrot builds and installs in {cwd}/install/; then nqp builds and installs in {DESTDIR}/install/; then the build process fails because it can't find NQP. (On my machine) 19:41
Like I said, I was able to work around it.. but it was certainly LTA. 19:42
masak std: ∅ ** ∅ 19:47
camelia std 7c17586: OUTPUT«===SORRY!===␤Bogus statement at /tmp/QriLe68jrf line 1:␤------> <BOL>⏏∅ ** ∅␤Parse failed␤FAILED 00:00 42m␤»
masak bah. not even STD knows about ∅ :)
std: set(<b>) ⊂ set(<a b c>) 19:49
camelia std 7c17586: OUTPUT«===SORRY!===␤Confused at /tmp/uQKhk5oMgH line 1:␤------> set(<b>) ⏏⊂ set(<a b c>)␤ expecting any of:␤ feed_separator␤ infix or meta-infix␤ infixed function␤ statement modifier loop␤Parse failed␤FAILED 00:00 43m␤…»
masak things are not in sync here. 19:50
flussence std: set(<b>) (<) set(<a b c>)
camelia std 7c17586: OUTPUT«===SORRY!===␤Confused at /tmp/z2tQKTHyEf line 1:␤------> set(<b>) ⏏(<) set(<a b c>)␤ expecting any of:␤ feed_separator␤ infix or meta-infix␤ infixed function␤ statement modifier loop␤Parse failed␤FAILED 00:00 44m␤…»
flussence ENOSETOPS
TimToady weird, it's in CORE.setting 19:52
ah well, STD was always a massive kludge, and has probably seen most of its usefulness already 19:53
masak STD++ 19:54
best kludge ever.
I still remember the feeling back when Rakudo and Pugs were so underpowered, grammar-wise, that what STD did seemed like pure magic. 19:55
we've caught up since then. gradually.
TimToady well, rakudo still hasn't quite caught up in Q sublanguages, but yeah, it's a lot closer than it was 19:56
GlitchMr Isn't STD needed to modify Perl 6 grammar?
TimToady we're talking about the prototype standard grammar that runs by a very ad hoc translation to Perl 5 19:57
though not as ad hoc as it used to be
pmichaud It's Google's birthday also?
TimToady but P5 really doesn't have the chops to do STD correctly
19:58 SmokeMachine joined
TimToady also STD is written largely in terms of hashes-as-objects, and we'll be wanting to move away from that mindset 19:58
pmichaud knocks 126 candies out of the Google pinata.
TimToady pmichaud: apparently the birthday of their first search page, anyway
lue wonders if std will run on rakudo or if that's still a ways away. 19:59
GlitchMr I guess it won't.
Even Niecza needs modified version of STD.
TimToady can't get to the piñata because google is overlaying it with more personalized greetings :)
pmichaud :)
GlitchMr ===SORRY!=== 20:00
Could not find DEBUG in any of: /usr/lib/parrot/5.6.0/languages/perl6/site/lib, /usr/lib/parrot/5.6.0/languages/perl6/vendor/lib, /usr/lib/parrot/5.6.0/languages/perl6/lib, /home/glitchmr/.perl6/2013.08/lib
it won't.
TimToady is scared to ask google's new search engine whether it knows my SSN
pmichaud is not scared. 20:01
GlitchMr STD.pm depends on Perl 5 code.
I assume .pmc is Perl 5 code.
lue Really? I could swear it was written in P6
FROGGS hi pmichaud
TimToady .pm6 is written in Perl 6
popl scared of a hummingbird? ;P
20:02 REPLeffect left
TimToady here in Mountain View, Google is the 800-lb hummingbird... 20:02
GlitchMr lue, yes, but it uses modules that are written in Perl 5.
TimToady the .pmc is translated from the .pm6
popl that is kind of scary
TimToady by viv, which is a STD-based parser :) 20:03
pmichaud Google doesn't give me a scary answer for "what is my ssn?"
TimToady just because it's bootstrapped that way doesn't mean it isn't a kludge
it does things like substitute local variables for dynamic variables
pmichaud It's a slightly more scary answer for "What is Larry Wall's social security number?"
TimToady what does it say? 20:04
pmichaud and after typing "What is Larry Wall", Google autocompletes with " doing now?" :-P
TimToady typing
strongly
pmichaud TimToady: it pulls up some obituary records. :-/
TimToady oops, my secret is out
lue O.o
popl TimToady is undead.
pmichaud TimToady: well, not surprising I guess, since TheDamian claims you're immortal. 20:05
TimToady was hoping the new gray hairs would fool everone
jnthn TimToady: Where is Q in Rakudo not quite caught up, ooc? 20:06
jnthn did a bunch of aligning with STD there a while back... :)
o/ pmichaud
TimToady oh, does it work by mixins now?
pmichaud jnthn: somewhere in the Continuum, I suspect. 20:07
jnthn TimToady: yes, as of Nov last year, when I also was putting in heredocs, etc. :)
pmichaud (sorry, somewhat obscure)
jnthn pmichaud: Too obscure for me ;)
pmichaud www.google.com/url?sa=t&rct=j&...bmCg7eowFQ
arrgh!
en.wikipedia.org/wiki/Q_Continuum
better
benabik Arg, yeah, the google links. :-( 20:08
GlitchMr Google links will kill us all.
FROGGS jnthn: this should work: froggs.de/perl6/rakudo-star-2013.09-RC2.msi
GlitchMr And after they will kill, they will use die to throw an exception. 20:09
FROGGS what? one has to explain what the Q Continuum is?? 20:10
jnthn pmichaud: Ah. :)
FROGGS well, at least tadzik++ knows
jnthn FROGGS: trying
FROGGS jnthn: thank you
jnthn aww, my frist crack at termdef explodes...
...maybe 'cus I typed something really great...
FROGGS jnthn: I even made .bat files for perl6-debug, p6doc and panda 20:11
jnthn perl6-debug is already a .exe, no? :)
FROGGS hmmm, yeah
then not for perl6-debug :P
'explodes' is a term that is not known in the real world in that context btw
20:12 SmokeMachine left
FROGGS my colleagues are always shocked when I use it 20:12
or at least amused
tadzik tadzik knows!
knows what
FROGGS hehe, my man! 20:13
tadzik ach, Q :)
FROGGS yeah, Q-who, deja-Q, and all these things
20:14 silug_ left, REPLeffect joined
tadzik I liked the one about Picard re-living his live _very_ much 20:14
possibly the second best TNG imho, after The Measure of a Man
FROGGS nah, the inner light is the best one
but I think you mean the one where he gets a heart-surgery? 20:15
lue
.oO(I imagine it also frightens normal people when you talk about killing programs.)
 20:22
tadzik yes, that one
inner light, hmm
oh, that's very good to
too
nothing beats The Measure of a Man imho 20:23
I cried watching that one
a little bit, but still :)
arnsholt Speaking of weird usages, I used hysterical raisins at work one time. Only one colleague caught it =) 20:24
thundergnat Sigh. Is there a reason why .subst on an undefined value is now a fatal error? It didn't used to be. 20:26
r: my @a = <a1 b2>, Any, <d4 e5>; .subst(/\d/,'',:g).say for @a;
camelia rakudo 384a0f: OUTPUT«a␤b␤No such method 'subst' for invocant of type 'Any'␤ in block at /tmp/wENJTYbBID:1␤␤» 20:27
FROGGS tadzik: really? hmmm, seems like what this episode is to you, is the inner light to me
masak arnsholt: "hysteriska russin"? :)
20:28 silug_ joined 20:29 FROGGS left, benabik left, FROGGS joined
GlitchMr Perl 6 debug is annoying to use with white terminal background. 20:29
Seriously, yellow?
jnthn You were menat to use it with a black one :P
Seriously, patches welcome to support custom color schems and config. :) Or wait for me to eventually get to it. :) 20:30
FROGGS r: my @a = <a1 b2>, Any, <d4 e5>; .Str.subst(/\d/,'',:g).say for @a # not fatal
camelia rakudo 384a0f: OUTPUT«a␤b␤use of uninitialized value of type Any in string context in block at /tmp/Q3zehi4oL5:1␤␤␤d␤e␤»
FROGGS jnthn: and we need to disable colors when these are not supported
TimToady there are people who program with a white background?!? 20:31
jnthn FROGGS: When's that happen, ooc? :)
FROGGS dunno how to check for it though
jnthn: standard cmd.exe prints the escape codes, which makes the debugger useless
TimToady: there is always more that one way to ... :o) 20:32
arnsholt masak: Nah, in English. Some of my colleagues don't speak Norwegian
jnthn FROGGS: Oh...we need a Term::WindowsColor that knows how to do what cmd.exe wnats :)
20:33 silug_ left
arnsholt Also, the pun doesn't really work outside of English =) 20:33
jnthn FROGGS: And then a Term::AnyColor that frontends the two :)
FROGGS O.o
20:33 silug_ joined
FROGGS ++jnthn :P 20:33
masak arnsholt: in Esperanto it becomes "por histeraj koaloj" -- the word for "reason" being "kialo". 20:35
tadzik and HTML colours
masak hm, pro*
FROGGS esperanto?? are you serious? 20:37
lue masak's very fluent, from what I've seen (and reminded me I should get back to it sometime). 20:38
20:39 xinming left
grondilu paroletas esperanton 20:40
(longe vivu la verda stelo de esperanto :) )
20:41 xinming joined 20:43 REPLeffect left 20:44 ajr joined 20:45 ajr is now known as Guest91954
masak FROGGS: I'm serious, to the extent I'm ever serious. 20:45
FROGGS: I learned it back in 2000-ish, and it's one of my three fluent languages.
lue
.oO(What are the other two, you english-speaking Swede?)
 20:46
masak :) 20:47
masak .oO( 我不知道...荷兰语？ ) :P
hm, that one should probably end in a 吗... 20:48
20:48 REPLeffect joined
jnthn
.oO( There's no question about it... )
 20:49
masak jnthn: unless your mom's horse does weed. :P 20:50
20:53 Guest91954 is now known as ajr_ 20:54 REPLeffect left 20:55 Chat0979 joined
Chat0979 hi 20:55
20:55 Chat0979 left
jnthn bye 20:55
dalek kudo/nom: b1d6759 | jnthn++ | src/Perl6/ (2 files):
Start getting defterm handling in place.

This enables things like:
   constant term:<inator> = "I'll be back";
popl Does that happen often?
masak popl: jnthn commits? yes. :)
jnthn My mom's horse doing weed? All the time, man... 20:56
popl I meant drive-by greetings
jnthn oh :)
lue what exactly *is* defterm, anyway? is it being able to define these:<things> ?
jnthn yeah
popl whenever someone ends a sentence with "doing weed" or "on weed" it makes me think of Jon Stewart's character from the film Half-Baked.
lue (as something other than subs and similar that is) 20:57
20:57 skids left
jnthn TimToady: Should things like this work: 20:58
sub foo(\infix:<whatevs>) { say 1 whatevs 2 }; foo(*+*)
?
20:58 REPLeffect joined
jnthn If so, does it magic up an & on the start of the name? 20:58
Or should we for now conservatively constrain it to really only let you do \term:<x>? 20:59
21:05 SmokeMachine joined
jnthn r: bar(); sub foo() { my $x = 42; our sub bar() { say $x; } }; foo(); bar(); 21:10
camelia rakudo 384a0f: OUTPUT«===SORRY!=== Error while compiling /tmp/57oiivIJUi␤Undeclared routine:␤ bar used at line 1. Did you mean '&bag'?␤␤»
21:10 REPLeffect left
jnthn r: GLOBAL::bar(); sub foo() { my $x = 42; our sub bar() { say $x; } }; foo(); GLOBAL::bar(); 21:11
camelia rakudo 384a0f: OUTPUT«Could not find symbol 'bar'␤ in method <anon> at src/gen/CORE.setting:11803␤ in any at src/gen/Metamodel.nqp:2671␤ in any find_method_fallback at src/gen/Metamodel.nqp:2659␤ in any find_method at src/gen/Metamodel.nqp:946␤ in block at /tmp/v1bYovewov:1␤␤…»
jnthn r: sub foo() { my $x = 42; our sub bar() { say $x; } }; foo(); GLOBAL::bar();
camelia rakudo 384a0f: OUTPUT«Could not find symbol 'bar'␤ in method <anon> at src/gen/CORE.setting:11803␤ in any at src/gen/Metamodel.nqp:2671␤ in any find_method_fallback at src/gen/Metamodel.nqp:2659␤ in any find_method at src/gen/Metamodel.nqp:946␤ in block at /tmp/fEzWIxGlya:1␤␤…»
jnthn r: sub foo() { my $x = 42; our sub bar() { say $x; } }; foo(); GLOBAL::<&bar>();
camelia rakudo 384a0f: OUTPUT«42␤»
21:12 REPLeffect joined
masak r: sub foo() { my $x = 42; our sub bar() { say $x; } }; GLOBAL::<&bar>() 21:12
camelia rakudo 384a0f: OUTPUT«(Any)␤»
21:16 REPLeffect left, TgMts joined 21:19 stevan_ left
masak ponders the slogan "scopes are objects, instantiated on entry" 21:19
21:39 thundergnat left, Mouq joined
masak actually, I guess that slogan explains pretty well why scopes can be used for encapsulation (like in JavaScript). 21:44
Mouq .tell moritz I've been around + check the IR Clogs, I'm just busy with school and cross country 21:45
yoleaux Mouq: I'll pass your message to moritz.
21:55 moritz left, moritz joined, _ilbot left, bluescreen10 left, _ilbot joined 21:57 popl left, prammer left 21:58 Bucciarati joined 22:00 mls left, mls joined, _ilbot left 22:02 moritz left 22:03 kaare_ left, moritz joined 22:07 denisboyun left, stevan_ joined 22:08 prammer joined 22:09 stevan_ left, stevan_ joined, _ilbot joined 22:19 xinming left 22:20 grondilu left 22:21 xinming joined 22:23 sftp left, pmurias left 22:24 sftp joined, PacoAir left 22:29 TgMts left 22:44 mls left, mls joined 22:46 benabik joined
TimToady I'd say we can conservatively ignore \infix for now, since we'd probably get the wrong precedence on it anyway... 22:49
jnthn: ^^
in any case, we can't really change the language based on a run-time binding; that's more the province of a macro, and we haven't defined mixed-arg macros yet 22:51
jnthn r: sub foo(\term:<lol>) { say lol }; foo(42) 22:54
camelia rakudo b1d675: OUTPUT«===SORRY!===␤No such method 'orig' for invocant of type 'NQPMu'␤»
jnthn hmmm
r: constant \term:<lol> = 42; say lol 22:55
camelia rakudo b1d675: OUTPUT«42␤»
jnthn r: constant term:<lol> = 42; say lol
camelia rakudo b1d675: OUTPUT«42␤»
jnthn r: constant lol = 42; say lol
camelia rakudo b1d675: OUTPUT«42␤»
jnthn Well, that lot works now at least...
Will have to look at the other one
r: constant \term:["lol"] = 42; say lol 22:56
camelia rakudo b1d675: OUTPUT«===SORRY!===␤No such method 'ast' for invocant of type 'NQPMu'␤»
jnthn grrr
TimToady r: constant \term:<lol> = 42; say lol + lol 22:57
camelia rakudo b1d675: OUTPUT«84␤»
Mouq I don't think thing:[anything()] is implemented yet in rakudo 22:59
jnthn r: sub infix:["lol"]($a, $b) { "$a lol $b" }; 'omg' lol 'toaster' 23:01
camelia rakudo b1d675: OUTPUT«===SORRY!===␤No such method 'ast' for invocant of type 'NQPMu'␤»
jnthn I thought that one was...
jnthn wishes we didn't have a gazzilion forms for these things... :) 23:02
23:07 benabik left
TimToady n: sub infix:["lol"]($a, $b) { "$a lol $b" }; 'omg' lol 'toaster' 23:07
camelia ( no output )
TimToady n: sub infix:["lol"]($a, $b) { "$a lol $b" }; say 'omg' lol 'toaster'
camelia niecza v24-95-ga6d4c5f: OUTPUT«omg lol toaster␤» 23:08
jnthn n: sub getlol() { "lol" }; sub infix:[getlol]($a, $b) { "$a lol $b" }; 'omg' lol 'toaster' # curious
camelia ( no output )
jnthn n: sub getlol() { "lol" }; sub infix:[getlol]($a, $b) { "$a lol $b" }; say 'omg' lol 'toaster' # curious
camelia niecza v24-95-ga6d4c5f: OUTPUT«omg lol toaster␤»
TimToady a gazillion forms is what we invented syntax for to let us treat them as a few semantic constructs :)
jnthn
.oO( just BEGIN it... )
 23:09
TimToady yeah, those have to resolve at compile time 23:10
one of those little semantic constructs I mentioned :P
23:12 PacoAir joined
TimToady but a lot of things are obvious to sorear++ that are not obvious to the rest of us :) 23:14
Mouq sent in a CLA but hasn't heard anything about it… he just makes a pull request 23:18
23:19 Rotwang left 23:20 BenGoldberg joined
Mouq …On the topic of my PR, STD.pm6/grammar Regex both has a token that disallows unspace, and a backslash:<unspace> that matches it. 23:34
TimToady the latter is likely a fossil 23:46
jnthn 'night, #perl6 23:47
TimToady sleep pretty
colomon \o
23:54 ilogger2 joined, ChanServ sets mode: +v ilogger2

Powered by Cro and the Raku® Programming Language.

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

Thank you!