»ö« | perl6.org/ | nopaste: paste.lisp.org/new/perl6 | evalbot usage: 'perl6: say 3;' or rakudo:, alpha:, pugs:, std:, or /msg p6eval perl6: ... | irclog: irc.pugscode.org/ | UTF-8 is our friend!
Set by lichtkind on 5 March 2010.
sorear jnthn: argh. Looks like I'll be needing invocation objects after all... 00:01
00:01 tylerni7 left, tylerni7 joined, tylerni7 left, tylerni7 joined
jnthn sorear: I suspected you may. 00:03
Eh well, at least there's one already written that you can get some inspiration by. :-)
sorear :)
At issue is that, sometimes, $foo->method means $main::AUTOLOAD="method"; UNIVERSAL::can($foo, "AUTOLOAD")->($foo); 00:04
jnthn oh, autoload
sorear now it would be so much simpler if, instead, P5 used UNIVERSAL::can($foo, "AUTOLOAD")->("method")->($foo)
jnthn ouch.
sorear but no.
00:08 meppl left 00:11 cdarroch left 00:18 payload joined
sorear jnthn: do you know anything about the semantics of 'provides'? When do I need to use it, and what do I need to do to use it? 00:20
jnthn sorear: I know you're likely not to need to worry about it for now. 00:21
sorear ok
jnthn sorear: It's some weird "interface" mechanism that essentially boils down to a list of strings, iirc
And occasionally Parrot maybe will look at them for...something...
I've got by with ignoring it anyway. :-)
sorear apparently, pmc2c calls die("") 00:23
lta error message
jnthn lovely.
:-/
sorear oh, no 00:25
it uses print Carp::longmess("") to generate a stack dump as if die("") was used
then it prints the error message
my years of P5 have trained me to look for error details at the top of two pages of spew, not the bottom 00:26
the actual bug is that I forgot to add the new pmc to the Makefile 00:27
also, this makefile isn't -j clean
jnthn: this Makefile.in looks suspiciously autogenerated 00:28
00:28 yinyin joined
sorear does it have a source I should be editing 00:28
jnthn Makefile.in is what Makefile is generated from
However, the original Makefile.in came from a "make a language shell" script that you use to start writing a new langauge 00:29
sorear is it, itself, generated?
oh
jnthn Only by a one-off "get you started" thingy.
sorear :(
sorear hates one-off code generators... it's like cargo culting, but worse
00:34 lichtkind left
jhuni perl6: ([\~] ^5).join(" ").say; 00:39
p6eval pugs: OUTPUT«0 01 012 0123 01234␤»
..rakudo fed468: OUTPUT«You can't backslash that at line 11, near "~] ^5).joi"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
..elf 30121: OUTPUT«pre without a prefix is unimplemented at ./elf_h line 2988␤»
jhuni perl6: ([\,] ^5).join(" ").say;
p6eval pugs: OUTPUT«*** No such subroutine: "&prefix:[\\,]"␤ at /tmp/BD08GfyJfQ line 1, column 2-9␤»
..rakudo fed468: OUTPUT«You can't backslash that at line 11, near ",] ^5).joi"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
..elf 30121: OUTPUT«pre without a prefix is unimplemented at ./elf_h line 2988␤»
jhuni Is [\,] a real operator? 00:40
00:42 lestrrat is now known as lest_away
jhuni pugs: ([,] (1,2,3).say; 00:43
p6eval pugs: OUTPUT«*** ␤ Unexpected ";"␤ expecting term postfix, operator, ":", "," or ")"␤ at /tmp/AxzTD8qPKj line 1, column 17␤»
sorear cannot find file 'p5sv.dump' in path '.',
I... am completely stuck.
jhuni pugs ([,] (1,2,3)).say;
sorear I've added p5sv to the build system in all the places p5scalar is mentioned
jnthn: do you have a clue why I can't compile blizkost? Or should I take this to #parrot? 00:44
jnthn #parrot may be more familiar with the tools and the errors...
You did re-configure to re-generate from Makefile.in?
sorear yes
jnthn Oh
There's another makefile in src/pmc/ too maybe?
sorear Yes.
I edited that one too. 00:45
jnthn OK, you twekaed the .in for that?
sorear Yes
jnthn (I actually dislike that it's split over two...)
00:45 mberends left
jnthn OK...well...those are the obvious reasons it'd not work. :-/ 00:45
sorear pastie.org/874714 is all the changes I've made
jnthn looking 00:46
sorear: Does p5sv.pmc certainly get compiled for it tries to do p5array.pmc? 00:47
Or anything that inherits from p5sv?
I suspect it may be that p5array.pmc's compilation needs p5sv.dump 00:48
00:49 dalek left, dalek joined
sorear jnthn: p5array inherits from p5sv, yes 00:49
I couldn't parse your first sentence
jnthn oh 00:51
that's 'cus it wasn't very well formed English. :_)
Does p5sv.pmc get compiled before all of the other PMCs that inherit from it? 00:52
sorear I added p5sv to the beginning of all the lists.
However, at the time of the error there are no generated files.
except the makefiles and init_with_xs.h 00:53
jnthn You may need some more dependencies if doing inheritance.
sorear how do I add dependencies? 00:54
jnthn AFAIK, p5sv.pmc needs to be fed to pmc2c, which will generate p5sv.dump, before you can generate any of the PMCs that inherit from it.
sorear ...wow
jnthn Just the usual makefile way
sorear two-phase compilers without automatic dependency handling, fail
jnthn Yeah. :-/
00:56 brushie_ left
sorear wants to write something to automate this 00:57
can you look at build/src/pmc/Makefile.in and tell me where to add a dependency? 00:58
jnthn I think there used to be some scripty thing that handled the process overall somewhat more
sorear I don't think I can without a significant restructuring of the makefile 00:59
jnthn And then it was decided makefiles were better. :-/
oh, hmm 01:00
sorear but but distutils
jnthn the generate stage comes before the compile stage
And $(PMC_SOURCES) has p5sv first, no?
jnthn boggles over what could be wrong 01:01
We do handle an inherited PMC in the Rakudo makefile
But Pm didn't like the way the Parrot ones were done so did it differently.
...maybe you're discovering why he disliked it now...
jnthn looks at the diff again 01:02
sorear my ideas for this basically boil down to "autotools without the suck" 01:04
jnthn nod 01:05
Does p5sv.pmc build if you just have that one in there?
snarkyboojum phenny: tell masak, check out the 'tardis' script on the github repo - basic shell and integration with Tardis::Debugger 01:10
phenny snarkyboojum: I'll pass that on when masak is around.
01:11 nihiliad joined
snarkyboojum phenny: tell masak, github.com/masak/tardis/commit/aefa...d1e303d236 01:11
phenny snarkyboojum: I'll pass that on when masak is around.
snarkyboojum needs to still rid himself of certain perl5-isms :) 01:13
01:16 lest_away is now known as lestrrat 01:19 meppl joined 01:22 TiMBuS left 01:25 arthur-_ left
jnthn > our multi infix:<wtf>($a, $b) { $a ~ 'WTF' ~ $b } 01:26
> say "OMG" wtf "BBQ"
OMGWTFBBQ
> say ["OMG", "LOL" Zwtf "BBQ", "masak"]
OMGWTFBBQ LOLWTFmasak
(our just for spreading it out over lines in the repl; it works fine lexically too)
sorear jnthn: Yes, it does 01:27
01:27 daemon left 01:32 envi^home joined
jnthn sorear: OK, I suspect an ordering issue then 01:32
01:32 snarkyboojum left
jnthn It shoulda produced a .dump file - does it match the name of the one that was being looked for? 01:33
sorear yes 01:40
sorear studies the Rakudo PMC builder 01:41
01:50 nihiliad left 01:52 arthur-_ joined 01:54 nihiliad joined, snarkyboojum joined 01:58 meppl left
dalek kudo: 2189449 | jonathan++ | src/Perl6/Grammar.pm:
First cut of user defined operators. Works for infixes just fine, and the meta-op forms then also work, modulo some little tweaks we'll need on the lookup. postfixes don't work, but I'm too tired to work out why tonight.
 02:03
jnthn And now I sleep. Night! 02:05
o/
02:05 wknight8111 left 02:11 xomas left
colomon jnthn++ # didn't expect to come home to that! 02:15
rakudo: sub infix:<ç>($a, $b) { $a / $b.Num }; say 4, 4 Zç 5, 3 02:35
02:35 rv2733 joined
p6eval rakudo fed468: OUTPUT«Confused at line 11, near "say 4, 4 Z"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤» 02:35
colomon ah, not updated yet. 02:36
02:50 hercynium joined, k23z__ left
colomon just pushed with one spectest failure, in S03-smartmatch/array-array.t. Can't see how it could be related to what I am pushing. 02:56
rakudo: sub infix:<ç>($a, $b) { $a / $b.Num }; say 4, 4 Zç 5, 3
p6eval rakudo 218944: OUTPUT«Null PMC access in find_method('HOW')␤current instr.: 'perl6;Perl6Role;ACCEPTS' pc -1 ((unknown file):-1)␤»
colomon rakudo: sub infix:<ç>($a, $b) { $a / $b.Num }; say 4 ç 5 02:57
p6eval rakudo 218944: OUTPUT«0.8␤»
dalek kudo: c52d193 | (David Romano)++ | t/spectest.data:
Merge remote branch 'upstream'
 03:00
kudo: d52d50d | (David Romano)++ | (3 files):
Merge remote branch 'upstream'
kudo: 27ff4c7 | (David Romano)++ | src/Perl6/Grammar.pm:
[Grammar] Comment for infix_postfix_meta_operator
kudo: a223dc0 | (David Romano)++ | docs/ROADMAP:
Merge remote branch 'upstream'
kudo: 2abcdcd | (Solomon Foster)++ | src/Perl6/Grammar.pm:
Merge branch 'master' of [email@hidden.address]

  [docs/ChangeLog] partial draft of the 2010.03 release
03:04 Jedai left 03:12 snarkyboojum left 03:19 Jedai joined, am0c joined 03:20 justatheory joined 03:22 snarkyboojum joined 03:24 ShaneC left 03:30 justatheory left 03:31 justatheory joined 03:52 snarkyboojum_ joined, snarkyboojum left, snarkyboojum_ is now known as snarkyboojum 04:00 Guest95631 left 04:04 nihiliad left, petdance joined 04:05 ned joined 04:06 ned is now known as Guest80479 04:14 jaldhar joined 04:23 stephenlb left 04:31 Chillance left
lue ...hello? 04:34
sorear no 04:35
lue darn. :) 04:38
04:38 vamped joined
lue I'm currently on the planet Hyperop, took a while to find a signal. 04:38
goodnight 04:57
05:04 jhuni left 05:19 snarkyboojum left, astrojp left, PacoLinux left, cbk left, nsh left, hejki left, akl left, borup left, broquaint left, _ilbot2 left, jnthn left, colomon left, fda314925 left, charsbar left, silug left, Helios left, buubot left, petdance left, envi^home left, diakopter left, cotto left, zoi left, REPLeffect left, kst left, EvanCarroll left, Khisanth left, mtve left, obra left, allbery_b left 05:24 Bzek left 05:25 Bzek joined 05:43 mberends joined
spinclad a serious and lasting split, here 05:58
sorear they happen occasionally
05:58 quietfanatic joined
lisppaste3 [email@hidden.address] pasted "log of #rakudosketch 2010-03-17" at paste.lisp.org/display/96546 06:04
06:19 mberends_webchat joined
mberends_webchat arghh, this irc server (bear.freenode.net) is not on the same side of the netsplit as the irclogger, jnthn and pmichaud :-( 06:23
06:23 mberends_webchat left 06:26 justatheory left 06:28 rv2733 left 06:35 mberends_ joined
mberends_ hmm, tunneling via Netherlands doesn't get to feather either :( 06:38
06:40 mberends_ left
TimToady well, I'm here... 06:43
06:46 TimToady sets mode: +vv dalek hugme, TimToady sets mode: +vv ilogger2 IRSeekBot, TimToady sets mode: +vvv lisppaste3 p6eval phenny
mberends :-) I emailed pmichaud the #rakudosketch log, I hope that reaches him 06:46
TimToady I'm surprised it's taken so long to rejoin
mberends yes, the irclog is missing almost an hour 06:47
probably it will all spring to life when a $human toggles the $Big-Red-Switch 06:51
TimToady while I'm asleep tonight, I hope... 06:55
zzz &
07:01 kaare joined 07:07 Su-Shee joined
mberends hi Su-Shee 07:08
Su-Shee mr berends!
mberends !
half the #perl6 channel has been disconnected from us for over an hour 07:10
Su-Shee the silent masses. ;) 07:11
07:11 uniejo joined
mathw Morning 07:11
mberends morning mathw 07:13
07:14 snarkyboojum joined, diakopter joined, jnthn joined, colomon joined, astrojp joined, PacoLinux joined, fda314925 joined, nsh joined, cotto joined, charsbar joined, zoi joined, silug joined, EvanCarroll joined, REPLeffect joined, Helios joined, buubot joined, kst joined, hejki joined, akl joined, borup joined, broquaint joined, _ilbot2 joined, Khisanth joined, avuserow joined, yves joined, jrockway joined, pmichaud joined, mtve joined, allbery_b joined, obra joined, farmer.freenode.net sets mode: +ooo diakopter jnthn pmichaud
mberends hey! join! 07:14
07:14 allbery_b left, buubot left, quietfanatic left, quietfanatic joined
mathw relocates -> $work 07:15
07:15 kaare is now known as Guest4091, buubot_ joined
mberends phenny, tell pmichaud the #rakudosketch log is in paste.lisp.org/display/96546 07:16
phenny mberends: I'll pass that on when pmichaud is around.
mberends shower &
07:17 allbery_b joined 07:20 xomas joined, xomas left, xomas joined, eternaleye left 07:22 riffraff joined 07:31 am0c left 07:39 eternaleye joined, eternaleye left 07:47 buubot_ left, buu left 07:51 barney joined 08:04 mberends left 08:06 payload left 08:30 rgrau joined 08:31 payload joined 08:37 payload left 08:48 payload joined 08:51 vamped left 08:58 hanekomu_ joined 08:59 hanekom__ joined 09:01 payload left 09:03 hanekomu_ left 09:34 payload joined 10:05 smash_ joined
smash_ hello everyone 10:05
sorear hello
10:06 masak joined
masak oh hai, #perl6! 10:06
phenny masak: 01:10Z <snarkyboojum> tell masak check out the 'tardis' script on the github repo - basic shell and integration with Tardis::Debugger
masak: 01:11Z <snarkyboojum> tell masak github.com/masak/tardis/commit/aefa...d1e303d236
smash_ masak: mornin' 10:07
masak czechs
smash_: hi there!
smash_: when's the release today?
smash_ masak: starting working on it now
sorear has XS working sanely in Blizkost *thud* 10:08
masak snarkyboojum: wow, that is awesome!
sorear: whoa.
smash_ masak: where's my ChangeLog update ? ;) 10:10
masak smash_: that's why I asked about the release. :) how much time do I have?
I have 50 minutes now before lunch. I could do something with that.
smash_ masak: don't worry, you have time.. still ned to run the spectest, it won't go out before late in the afternoon 10:11
masak ok. good.
masak gets to it
smash_ masak: you can do it after lunch, no worries 10:12
masak I might need both time slots, so I'd better start now. 10:13
maybe I can co-opt jnthn++ for some last-minute help with named enums after lunch. :)
(though cowboy patching just before release may or may not be a wise idea...) 10:14
smash_ masak: if you patch it later, i'll need to rn the spectest again 10:15
say and build and all
masak yeah :/
I'll try to get named enums for next release :)
smash_ it's fine with me, as long you don't break anything that can't be fixed
(fixed before release)
masak I don't really know how much is left. I'll see what jnthn++ says too. 10:16
smash_ roger that
masak does the CHANGELOG 10:17
there's an impressive number of commits this month 10:18
smash_ i'm guessing this month's release is named Copenhagen because of the hackathon 10:20
masak aye.
but it might as well have been named 'colomon', too :)
smash_ is there a colomon.pm ? :)
masak not yet :)
smash_ hehe
smash_ make spectest 10:24
masak use.perl.org/~jhuni/journal/40252 # jhuni++ 10:25
10:28 lestrrat is now known as lest_away 10:32 riffraff left 10:35 riffraff joined 10:36 guest_ joined
masak rakudo: say (1/0).WHAT 10:40
p6eval rakudo 2abcdc: OUTPUT«Rat()␤»
masak rakudo: say +(1/0)
p6eval rakudo 2abcdc: OUTPUT«Inf␤»
masak is there any support in some well-known ISO standard for this strange numification of a zero-denominator Rat? 10:41
masak wants NaN
m6locks aye, lim 1/x is Inf, but division by zero itself is not a number 10:42
masak ok.
Java seems to give 'Infinity' too, so I yield. 10:43
time for lunch, it seems. I'm about halfway through the ChangeLog. 10:44
smash_: should I commit what I have so far?
smash_ masak: you can commit later when you're done, either way is fine by me
masak I pushed what I have so far. I'll do the rest after lunch, if no-one beats me to it. 10:45
smash_ masak: ok, thank you
10:47 flip214 joined
flip214 hello p6! 10:47
10:48 xinming_ joined
10:50 molaf joined
jnthn morning 10:50
10:50 yinyin left
jnthn Oh yay, ChangeLog updates 10:50
masak++
10:51 lest_away is now known as lestrrat, xinming left
smash_ jnthn: mornin' 10:54
guest_ Unimportant and slightly OT, but I don't know where else to post this: I think there is an accidentally mistake in the definition of pi at perl6-general-examples/Examples/nums.pl (jhuni's examples) line 3: my $pi ::= 3.1415_1_926_358979323846; shouldn't it be 3.14159265358979...? That would give a more correct result. (Closer to the first statement.) 10:55
11:04 meppl joined
jnthn guest_: Not sure that it should define pi anyway, I think it comes built-in... 11:21
rakudo: say pi
p6eval rakudo a8d808: OUTPUT«3.14159265358979␤»
guest_ cool^^
smash_ rakudo: say []~~[] 11:22
p6eval rakudo a8d808: OUTPUT«1␤»
smash_ rakudo: say []~~[1]
p6eval rakudo a8d808: OUTPUT«1␤»
jnthn ...huh? :-/
smash_ rakudo: say [1]~~[1]
p6eval rakudo a8d808: OUTPUT«1␤»
smash_ rakudo: say [1]~~[]
p6eval rakudo a8d808: OUTPUT«0␤»
smash_ yeap, that test is failling
rakudo: say !([]~~[1]) 11:23
p6eval rakudo a8d808: OUTPUT«0␤»
smash_ rakudo: say !([1]~~[])
p6eval rakudo a8d808: OUTPUT«1␤»
jnthn That's...odd. 11:27
masak I'm pretty sure there's a rakudobug in there somewhere. 11:29
rakudo: say []~~[1]
p6eval rakudo a8d808: OUTPUT«1␤»
masak that one, say.
jnthn Yes 11:30
masak Array Array arrays are comparable $_ «===» X (dwims * wildcards!) 11:31
masak submits rakudobug
jnthn It worked when I put Array.ACCEPTS back in.
So it's somehow regressed.
11:31 bkeeler left
smash_ well, i guess it was wotking in the last release.. the test passed back then 11:34
11:35 masak left, masak joined
jnthn masak: Worketh thee on the changelog? 11:35
phenny: tell pmichaud I also emailed you a summary of what happened at #rakudosketch - dunno if you saw that 11:36
phenny jnthn: I'll pass that on when pmichaud is around.
11:37 quester joined, JimmyZ joined
JimmyZ Hello, Rakudo * will be late? 11:37
quester Good morning #perl6!
11:41 guest_ left
masak jnthn: I work, but distractions are plenty :/ 11:42
11:44 JimmyZ left 11:45 prism joined 11:47 flip214 left
smash_ 9 subtests UNEXPECTEDLY SUCCEEDED 11:47
ok, 2 fails: 1 syntax error in a test file, and that smartmatch issue 11:48
masak I'm only on Feb 26, and now I'm being pulled into a meeting here. :/
it will not last more than an hour, I'm sure. 11:49
smash_ masak: plenty of time, don't worry
masak but maybe I should defer the ChangeLog writing to someone more available?
ok.
good.
jnthn masak: I can do a bit too if you like? 11:50
masak jnthn: yes, please do. just pushed my last additions.
smash_ i need to sort out the failling tests first
jnthn smash_: Which one has a syntax error? 11:51
smash_ the test is: for "par 1", "par 2" -> $p { ... } 11:52
this works: for ("par 1", "par 2") -> $p { ... }
dalek kudo: 5418e06 | masak++ | docs/ChangeLog:
[docs/ChangeLog] a bit more of ChangeLog draft progress
jnthn rakudo: for "par 1", "par 2" -> $p { ... } 11:53
p6eval rakudo a8d808: OUTPUT«No exception handler and no message␤current instr.: '&fail' pc 17032 (src/builtins/Junction.pir:450)␤»
jnthn rakudo: for "par 1", "par 2" -> $p { say $p }
p6eval rakudo a8d808: OUTPUT«par 1␤par 2␤»
smash_ grr, no syntax error now for me either :-\ 11:55
smash_ make specttest again
jnthn smash_: Just got a small patch to tidy up some lose ends from a patch I put in last night. 11:58
smash_ jnthn: commit away, i'll re-run test && spectest again later.. no worries
colomon \o/ 11:59
BTW, world;
jnthn yayitscolomon!
12:00 hanekom__ left
colomon rakudo: sub infix:<op>($a, $b) { -> $op { $op($a, $b) } }; my $a = 3 op 4; say $a(*+*); say $a(&infix<*>); 12:00
p6eval rakudo a8d808: OUTPUT«7␤Could not find non-existent sub &infix␤current instr.: '_block21' pc 29 (EVAL_2:0)␤»
colomon rakudo: sub infix:<op>($a, $b) { -> $op { $op($a, $b) } }; my $a = 3 op 4; say $a(*+*); say $a(&infix:<*>); 12:01
p6eval rakudo a8d808: OUTPUT«7␤12␤»
moritz_ oh hai
colomon o/
jnthn++
jnthn colomon: Yeah, I'm just patching up my bleary-eyed screw-ups in that patch at the moemnt too. Or trying to anyway. 12:02
rakudo: sub infix:<wtf>($a, $b) { "{$a}WTF{$b}" }; say "OMG" wtf "BBQ"
p6eval rakudo a8d808: OUTPUT«OMGWTFBBQ␤»
jnthn rakudo: sub infix:<wtf>($a, $b) { "{$a}WTF{$b}" }; say "OMG" Zwtf "BBQ"
p6eval rakudo a8d808: OUTPUT«Null PMC access in find_method('HOW')␤current instr.: 'perl6;Perl6Role;ACCEPTS' pc -1 ((unknown file):-1)␤»
jnthn fejl!
ooh, my compile is nearly done... 12:03
12:03 payload left
pugssvn r30122 | moritz++ | [t/spec] fix more smartlinks; down to 33 broken links 12:03
colomon jnthn: right, all the meta-ops seem to fail. 12:04
12:05 envi^home joined
jnthn colomon: For custom ops? 12:07
Yes
colomon yes 12:08
jnthn I know why.
colomon ooooooo
jnthn The base sub is assumed to be package scoped
We need to check if it's lexical
I've got a patch that I'm compiling now to try and do that.
colomon so the metaop code is whats wrong?
\o/
jnthn Yes 12:09
dalek kudo: d82c83e | moritz++ | t/spectest.data:
we pass S02-literals/subscript.t
jnthn Also, the reason postfixes didn't work was...embarassing.
Anyway, locally I just did:
> sub postfix:<!>($n) { my $r = 1; for 1..$n { $r *= $_ }; $r }; say 5!
120
colomon do tell :)
jnthn colomon: Heh. I was apparently concious enough last night to write the code to determine the default precedence level based on the operator category, but not concious enough to then use the result I'd computed in place of a hard-coded '%additive' :-) 12:10
So it tried to parse a postfix as an infix and got horribly confused. :-) 12:11
colomon I should have noticed that when I reviewed the code last night.
jnthn Meaning the entire patch is: 12:12
- PAST::Val.new( :value('%additive') )
+ PAST::Val.new( :value($prec) )
:-)
colomon :)
jnthn Oh well
At leat it was *easy* to fix. :-)
Also: yay, our cool factorial examples can has a work again.
colomon \o/
though it will be cooler when [*] works again. 12:13
12:14 takadonet joined
takadonet morning all 12:14
colomon \o
takadonet how are you doing colomon?
colomon okay
pmichaud good morning, #perl6 12:15
phenny pmichaud: 07:16Z <mberends> tell pmichaud the #rakudosketch log is in paste.lisp.org/display/96546
pmichaud: 11:36Z <jnthn> tell pmichaud I also emailed you a summary of what happened at #rakudosketch - dunno if you saw that
colomon \o
jnthn hi, pmichaud
pmichaud I saw the summary, yes, but I'd also like to see the discussion itself :)
jnthn pmichaud: OK - that's fine. Just didn't want you to miss it. :-)
pmichaud (I saw that you sent a summary. I have not read it in detail yet. I'll likely do so this morning.) 12:18
the 96546 looks very partial 12:19
colomon pmichaud: agreed. :(
wonder if nopaste has a length limit.
smash_ pmichaud: mornin'
pmichaud as in, it looks like a lot was said prior to the nopaste
smash_: good morning
colomon pmichaud: practically everything was said prior to the nopaste. 12:20
jnthn Oh.
:-/
pmichaud :-(
jnthn wonders if that's the same one he was emailed
pmichaud: What does it start with?
pmichaud <jnthn> On the upside, when I did fall asleep, I didn't ahve nightmares.
<jnthn> Anyway, I think this issue is in hand.
<spinclad> s/pieces of that in/part of/
<jnthn> Just needs more hack hack hack.
* chromatic (~chromatic@c-67-168-199-244.hsd1.or.comcast.net) has joined #rakudosketch
<jnthn> Any other comments on the last two? :-)
colomon jnthn: chromatic is there near the top, he didn't arrive until very late as I recall.
jnthn oh, ouch
:-(
pmichaud so, no log? 12:21
jnthn pmichaud: I have the lot in my irssi backscroll, I can extract it.
I just thought somebody else had some more conveient way
pmichaud Please do. I do want to see the conversation details themselves, even if I can only respond well-after-the-fact. 12:22
.oO( The more convenient way is to hold the discussion on #perl6? ;-)
 12:25
12:26 k23z__ joined, k23z__ left, k23z__ joined 12:32 ruoso joined
jnthn pmichaud: mailed it to you 12:34
pmichaud jnthn++ # thanks
colomon jnthn++
12:34 mberends joined
jnthn Feel free to post it up somewhere too 12:34
I've got too much to do for the number of hours in the day today. 12:35
colomon ditto++
jnthn colomon: Hmm...postfixes work and prefixes too. The Obvious Patch for the meta-ops with them didn't work. I think I may have to leave that until I have a bit more time. 12:36
pmichaud understood. I'm not sure what I'll be doing today... yesterday was a Really Bad Day and I suspect I'll either be trying to (a) distance myself entirely from it or (b) try dealing with it directly. If (a), then Rakudo stuff is a welcome distraction :-)
jnthn It does work if you declare it "our"
colomon jnthn: if our makes it work then I'd say we're good for now. 12:37
jnthn colomon: Yeah. I don't know why the fix I tried didn't work, which makes me think it'll need Deeper Digging, and I don't want to get bogged down in that too.
s/too/now/ 12:38
hugme: hug pmichaud
hugme hugs pmichaud
12:39 payload joined
mberends is it possible to create an object using a class whose name is specified in a string variable? (for FakeDBD) 12:41
pmichaud perhaps ::($foo).new ?
(nyi in rakduo afaik) 12:42
mberends thanks pm, will workaround temporarily
jnthn Easy workaround is eval($name).foo 12:43
er
eval($name).new
mberends cool :-) work() &
12:43 mberends left
dalek kudo: 240e2f0 | smashz++ | docs/release_guide.pod:
Testing commit.
 12:44
kudo: c63e4b2 | smashz++ | (2 files):
Merge branch 'master' of github.com:rakudo/rakudo
smash_ that second push wasn't intended, sorry
pmichaud smash_: you mean the second commit? np -- I do it all the time. 12:45
smash_ yeap, i meant the second commit
pmichaud jnthn: confirmed receipt of #rakudosketch log, thanks
smash_ jnthn: pushed everything already ?
everyone: last call for pushing changes 12:46
need to re run tests all over again
last call for commiting test changes too
colomon go smash_ go! 12:47
smash_ i'm off to lunch, will run everything again when i'm back.. so push anything you want to make to release before i'm back :-)
jnthn smash_: one coming now!
smash_: Also, working on ChangeLog
smash_ jnthn: cool, thks
ok, be right back
smash_ off & 12:48
dalek kudo: 3be6916 | jonathan++ | src/Perl6/Grammar.pm:
Fix thinko, tweak comments. Custom prefixes and postfixes should now work.
 12:49
12:57 M_o_C joined 12:58 payload left
jnthn This change log is epic. 12:59
13:06 meppl left
dalek kudo: d537cea | jonathan++ | docs/ChangeLog:
Fill out ChangeLog - should cover most things.
 13:07
takadonet I bet you the next one will be even longer 13:08
mathw likes 13:11
colomon jnthn: I'm confused, it looks like you undid masak's last patch to the changelog almost completely? 13:12
jnthn huh? 13:14
not intentionally
huh...wtf
jnthn smacks git 13:15
colomon Also, I think you forgot the ! metaop, and none of the metaops are quite "proper" yet. 13:16
(in particular, precedence is wrong on ! and R, and X and Z don't generate the operators properly, or handle cases with more than two list arguments.) 13:17
13:18 quester left
jnthn colomon: Patch it as you wish. 13:18
dalek kudo: 6dd31df | jonathan++ | docs/ChangeLog:
Re-instate masak++ ChangeLog entries that I somehow managed to lose.
colomon jnthn: 'k 13:19
done 13:23
13:23 xinming_ is now known as xinming
dalek kudo: 855ecb5 | (Solomon Foster)++ | docs/ChangeLog:
Tweak changelog on metaops.
 13:24
jnthn Yay. 13:25
Do we have a release announcement drafted?
smash_ back
jnthn welcome back, smash_++ :-)
masak man, that was a soul-sucking meeting :(
...and now they're proposing another one... :( 13:26
smash_ make realclean ; make ; make test ; make spectest
jnthn recommends a Configure somewhere in there ;-)
masak: I think ChangeLog is muchly done, but please do review. :-)
And tweak as you wish.
masak jnthn++
jnthn Ah, I see no 2010.03 yet 13:27
masak will review
mathw Is there anywhere an intention to at some point make Parrot and Rakudo not need Perl 5 in order to build?
masak I think so, but it's very far ahead. 13:28
jnthn Not in the near future. It's probably a long-term ideal though. 13:29
oops, I forgot about lunch.
jnthn fixes that
masak ChangeLog looks good. 13:32
jnthn: we seem fundamentally unable to agree on how to spell 'deferral' :)
arnsholt mathw: That probably means making them use autotools. You volunteering? =)
smash_ thks for the ChangeLog updates, creating the annouce file now
mathw arnsholt: umm. no.
jnthn masak: I deferred learning the correct spelling. 13:33
mathw Of course, rakudo could probably configure itself in NQP, but that might be tricky for Parrot
jnthn Or maybe defered.
.oO( I should learn English properly some day )
masak jnthn: oh! there's a duplicate. 13:34
13:34 rv2733 joined
masak jnthn: first, I thought you changed my spelling, but it seems you added your own entry. 13:34
jnthn :-O
oops!
masak jnthn: I'll change the spelling and remove my entry.
jnthn OK :-)
masak pushed. 13:35
dalek kudo: 4c251d4 | masak++ | docs/ChangeLog:
[docs/ChangeLog] fixed typo, removed dup
13:35 rv2733 left
jnthn rakudo: sub postfix:<!>($n) { my $r = 1; for 1..$n { $r *= $_ }; $r }; say 5! 13:35
p6eval rakudo 3be691: OUTPUT«120␤»
jnthn \o/
smash_ make spectest 13:36
mathw jnthn++ 13:37
masak jnthn: we can haz user-defined operators? woot! 13:39
13:39 riffraff left
jnthn masak: Yeah, I did them last night. 13:39
masak jnthn++
jnthn ...until about 3am. Then I slept until 11!
masak :) 13:40
hanekomu now we can haz APerL6 13:45
jnthn \o/ 13:46
masak bbiab & 13:48
13:48 masak left
colomon rakudo: sub postfix:<!>($n) { (^$n).reduce( &infix:<*> ) }; say 5! 13:49
p6eval rakudo 855ecb: OUTPUT«0␤»
colomon doh!
smash_ rakudo: say []~~[1] 13:50
colomon rakudo: sub postfix:<!>($n) { (1..$n).reduce( &infix:<*> ) }; say 5!
p6eval rakudo 855ecb: OUTPUT«1␤»
rakudo 855ecb: OUTPUT«120␤»
jnthn ^$n fail? 13:52
colomon jnthn: ^$n is 0..($n-1)
so colomon fail
jnthn oh 13:53
good point :-)
0 *...yes ;_)
of course there's...
rakudo: multi postfix:<!>(0) { 1 }; multi postfix:<!>($n) { $n * ($n - 1)! }; say 5! 13:54
p6eval rakudo 855ecb: OUTPUT«120␤»
jnthn That one is kinda pretty :-)
13:55 nihiliad joined
smash_ that []~~[1] test will probably fail, any suggestions for fixing it in time for release ? or skip the test ? 13:56
jnthn smash_: Let's skip the test for the release. 13:57
smash_ that means i need to commit the skip test in the pugs repo correct ?
jnthn Even just # out the appropriate line in t/spectest.data
You can do it that way too. 13:58
I think we can fix it soon, just not in time for the release.
moritz_ when will the release be?
smash_ moritz_: in a couple of hours we hope, unless something cames up 13:59
13:59 M_o_C left
moritz_ smash_++ # release-managing 13:59
smash_ jnthn: but in the t/spectest.data i need to comment out the entire test file, let me try to skip just that one test 14:00
moritz_ can't find an Array.ACCEPTS in rakudo anywhere
jnthn smash_: Yes, true. :-)
moritz_: It's in Seq akshually
Which Array inherits from.
moritz_ smash_: you can add a #?rakudo todo 'some regression' line before the test to the test file
14:02 payload joined
smash_ moritz_: that can work, thks 14:03
14:06 uniejo left 14:07 pmurias joined 14:09 payload left
pugssvn r30123 | smash++ | Todo test for rakudo release #27. 14:09
smash_ make spectest 14:10
14:14 Bzek left 14:15 k23z__ left
[particle] jnthn: i have rakudo compiling, and failing a handful of spectests 14:19
smash_ [particle]: on linux ? 14:21
[particle] w7-msvc 14:22
14:22 clintongormley joined
smash_ can you paste somewhere the tests that are failling please ? 14:22
[particle] sure, give me 10 minutes or so 14:23
jnthn [particle]: Compiling is good. Rest depends on how big your hand is.
smash_ [particle]: sure, thks.. and also make sure your git is up-to-date 14:24
jnthn: no harm in looking at the tests that are failing, maybe there's something fixable or skipable there
jnthn smash_: Aye. 14:25
14:28 k23z__ joined, k23z__ left, k23z__ joined, REPLeffect left
[particle] ok, it'll take me longer to rebuild/rerun spectest, but i'll nopaste results 14:29
building src/gen/core.pir, i'm peaking at 670MB so far 14:30
smash_ make spectest ..... All tests successful, 1174 subtests skipped. 14:33
14:33 rgrau` joined
smash_ LAST CALL, for late pushes to the repo.. speak now if you want to make it before today's release 14:34
mathw \o/ 14:35
rakudo: my Str $a = "Holy camel, onion!"; $a ~~ s/camel/rakudo/; $a.say; 14:36
p6eval rakudo 4c251d: OUTPUT«Holy rakudo, onion!␤»
mathw \o/
smash_ jnthn, masak, colomon: anything specific you guys want to see in the annouce from the ChangeLog ?
mathw s/// :) 14:37
smash_ (minor/major improvements, etc.)
i'll write a first draft, and then you guys can update it if you want 14:38
14:38 masak joined
masak haz a last look at the ChangeLog 14:39
smash_ masak: gimme a sec, and you'll be also looking at the announce
jnthn smash_: Yes, give us chance to review the announce. :-) 14:41
smash_: Should include something about the Copenhagen hackathon.
But maybe best if one of us who was there writes that.
smash_ jnthn: of course, i'll commit it in a sec 14:42
masak pushed a few picked nits
I like the ChangeLog now.
what's this Japanese twitterer saying? twitter.com/mattn_jp/status/10668434196 -- he is hoping something, but that's all I get. 14:44
14:44 REPLeffect joined
dalek kudo: b4ed7c1 | masak++ | docs/ChangeLog:
[docs/ChangeLog] added/changed a few quotation marks
 14:44
smash_ annouce file push'ed, please update it as you see fit 14:47
and someone add something about the hackathon, i wasn't where :-(
14:55 ilogger2 joined
smash_ the later annouces don't give info about the number of spectest currently passing 14:56
dalek kudo: 14aeee8 | jonathan++ | docs/announce/2010.03:
Fix next release date.
kudo: a7ff5e8 | jonathan++ | docs/announce/2010.03:
Oops, last patch was wrong.
colomon my two favorite music festivals -- one is on the first weekend in august, the other on the weekend after the first monday in august. so they're the same weekend about half the time... ;)
14:57 Guest4091 left, broquaint joined 14:59 Gothmog_ joined
masak jnthn: I was distracted again by someone who wanted to use my computer to print things. yes, I can write a hackathon description. 15:00
15:00 sunnavy_ joined
jnthn masak: OK. I'm tweaking the specific notable things at the moment. 15:01
15:04 sunnavy_ left, sunnavy joined
smash_ let me know if i did something wrong, and need fixing 15:05
jnthn smash_: Nothing wrong, just some bits I wanted to tweak. :-) 15:06
Pushed.
15:06 sunnavy left
smash_ jnthn: cool, thks 15:07
dalek kudo: f098309 | jonathan++ | docs/announce/2010.03:
Tweaks to the specifics in the release announcement.
 15:08
15:08 sunnavy joined
jnthn All: feel free to review and tweak my tweaks. 15:08
masak ditto. 15:09
jnthn masak: Do we need the [1] ? 15:10
at the start of the paragraph? 15:11
(agree it's good for the link at the bottom)
masak oh!
I might have fibbed there.
jnthn :-) 15:12
masak yes, I did. fixing.
jnthn That nit aside, I like the text. :-)
masak++
masak cool. moritz_++ for the original. :)
smash_ the [1] in the begining of the paragraph should be removed (me thinks) 15:13
masak already pushed :)
jnthn \o/
excellent.
dalek kudo: f343af8 | masak++ | docs/announce/2010.03:
[announce] description of the hackathon
kudo: 6c84d73 | masak++ | docs/announce/2010.03:
[announce] removed redundant [1] footnote
smash_ masak++ jnthn++ # announce and changelog updates 15:14
jnthn masak: Did the specific improvements list capture the main things?
ooh...we should mention series operator improvements too perhaps?
masak aye.
jnthn There was a lot of that
masak I just glanced at them, but I think they capture the main things.
a caveat though is that I'm still an alpha user, so I only know through the commits I see :) 15:15
smash_ i'll continue with the release as soon as everyone is confortable with the announce
15:15 nihiliad joined
jnthn masak: Heh. I'm the ohter way around: people see through the things I commit. :-) 15:16
zoi perl6.................... 15:17
masak zoi: you got that right! :)
zoi: but you don't need all those dots. three is fine.
jnthn std: perl6...
p6eval std 30123: OUTPUT«===SORRY!===␤Missing term at /tmp/NxTEv87ZPa line 1 (EOF):␤------> perl6...⏏<EOL>␤ expecting any of:␤ postfix_prefix_meta_operator␤ prefix or term␤ term␤Undeclared routine:␤ 'perl6' used at line 1␤FAILED 00:01 108m␤»
jnthn Apparently not.
:-)
masak jnthn: there are many variants of 'fine' :P 15:18
jnthn OK, I think I'm happy with the release announcement. :-)
masak me too. smash_++
colomon "Series operator now has basic support for the current Spec." ?
jnthn colomon: If you prefer that, make it so :-) 15:19
Or I can?
(I have the file open here0
colomon go for it (chasing small child here)
zoi masak: i want a perl7 where you have obligatory dots at the end of the package instaed of 1;..........................
dalek kudo: 807c57a | jonathan++ | docs/announce/2010.03:
Also mention the nice work on the series operator in the major items.
colomon or was there already a series announcement I missed?
jnthn lol
colomon ah, that works for me.
smash_ colomon: we're still reviwing the annouce.. so fell free tochange things now 15:20
jnthn colomon: Taking yours.
done
hejki is [+] already in current rakudo?
colomon hejki: no 15:21
smash_ everyone happy with the annouce ?
hejki =(
jnthn I'm happy.
zoi: You don't need an obligatory anything at the end of the package in Perl 6. 15:22
:-)
colomon smash_++
zoi jnthn: that's sooo not perl way man
smash_ ok, moving on with the release, please don't push anything now
zoi i demand obligary dots......
masak zoi: yes, but this is an open-source community. if you demand anything, you have to implement it yourself :) 15:23
zoi wow
smash_ demand equals write a patch !!
zoi i'll pm larry wall
man
colomon hejki: for the moment you can do @a.reduce(&infix:<+>) (for instance) instead of [+] @a
hejki: believe me, as soon as I can figure out how, I'll have [+] implemented again. (or given my current schedule, maybe someone else will beat me to it.) 15:24
masak zoi: sorry -- we're in release mode today, so we're all a bit giddy. at least I am. :)
jnthn colomon: I may, maybe, sort of, in a twisted way, understand operator handling a little better after last night, so may be able to help more with that now. :-)
masak \o/ 15:25
colomon jnthn++
it's mostly just sorting out how to add it to the grammar
jnthn really wants those - and hypers - back too.
dalek kudo: e4e058f | jonathan++ | docs/announce/2010.03:
Tweak series statement in announce thanks to colomon++.
jnthn Well, yeah, once it's in the grammar and we have a sane action method, we're done.
zoi masak: wow
still pming larry wall
jnthn Since really it's "take the list and call .reduce" I guess :-)
hejki colomon: okies :>
smash_ no pushing to git gentleman, before the release it out.. thank you 15:26
jnthn git commit -m "Huge epic change"
;-)
colomon jnthn: like I said yesterday, I want to expand .reduce to make it handle all the possible options for [op]. ;)
jnthn behaves
hejki rakudo: say (1,2,3,4).reduce(&infix:<+>);
p6eval rakudo a7ff5e: OUTPUT«10␤»
hejki nice
jnthn hejki: In a sense, [+] is just sugar for that. :-) 15:27
hejki ye :P
jnthn Very nice sugar though. :-)
hejki likes sugar
zoi would I be able to use swear words in perl6? 15:28
masak zoi: yes, but you have to define them first as operators.
zoi I see
sounds like a nice language
masak Perl 6 is a lot about sugar.
so we prefer if you use sweet words :) 15:29
zoi I made some modules in perl5 for cpan, but they got deleted because i was using "bad" license. I hope that wont happen in perl6
colomon actually, there is a little better sugar for the [+] in place already:
masak zoi: no worried. we don't have a CPAN :P
colomon rakudo: say (1,2,3,4).reduce(*+*);
p6eval rakudo a7ff5e: OUTPUT«10␤» 15:30
masak s/worried/worries/
zoi pretty neat
masak zoi: what is your quest? what's your favourite colour?
zoi my fav. color is sunday. oh wait, that's the day of a week nvm 15:31
my fav color is 6red
my fav color is 7red
my fav color is 4red
ya
15:31 mathw joined
zoi 4red 15:31
masak thanks for sharing.
mathw hmm, network glitch
jnthn rakudo: sub infix:<f**k>($a, $b) { say "CENSORED" }; 1 f**k 2 15:32
masak mathw! \o/
p6eval rakudo a7ff5e: OUTPUT«CENSORED␤»
jnthn zoi: ^
:-)
masak :)
zoi what's f**k?
masak jnthn: WHAT did the 1 do to the 2?!
zoi fook?
masak zoi: 'fejk'. it's Swedish for 'fake' :) 15:33
jnthn masak: You can't imagine!
masak jnthn: I already know that 7 ate 9, but this... :)
zoi rakudo: sub infix:<h*h>($a) { say "HAHA LOOK AT THIS $a" }; h*h
p6eval rakudo a7ff5e: OUTPUT«Could not find non-existent sub &h␤current instr.: '_block21' pc 29 (EVAL_2:0)␤»
zoi rakudo: sub infix:<h*h>($a) { say "HAHA LOOK AT THIS $a" }; 1 h*h 2
p6eval rakudo a7ff5e: OUTPUT«Too many positional parameters passed; got 2 but expected 1␤current instr.: 'infix:<h*h>' pc 181 (EVAL_2:77)␤»
masak zoi: 'infix' means 'standing between two things'.
zoi: so you need two paramters. 15:34
zoi o
lol
thansk
masak np.
jnthn Three French lolcats were on a frozen lake, but then the ice broke and un deux trois quatre cinq.
15:34 Psyche^ joined
zoi rakudo: sub onfix:<h*h>($a) { say "HAHA LOOK AT THIS $a" }; h*h 2 15:34
p6eval rakudo a7ff5e: OUTPUT«error:imcc:syntax error, unexpected '\n'␤ in file 'EVAL_1' line 58␤Could not find non-existent sub &h␤current instr.: '_block22' pc 29 (EVAL_2:0)␤»
zoi wow
jnthn zoi: prefix
masak submits rakudobug
smash_ i don't think i will have privileges to post on rakudo.org
zoi rakudo: sub prefix:<h*h>($a) { say "HAHA LOOK AT THIS $a" }; h*h 2 15:35
mathw rakudo: sub infix:<=======>($a, $b) { say "These two things are not equal enough for seven equal signs" }; 4 ======= 4;
p6eval rakudo a7ff5e: OUTPUT«HAHA LOOK AT THIS 2␤»
rakudo a7ff5e: OUTPUT«These two things are not equal enough for seven equal signs␤»
masak mathw++
zoi nice 1
mathw wanted to make sure it didn't pick out = or == first and get confused
masak mathw: I've reimplemented the optable parser. it won't :)
mathw \o/ 15:36
zoi jnthn: can i do a bad thing?
please
mathw maybe I'll do some Form.pm tonight
masak \o/
mathw just making usre it works with current rakudo would be a start
zoi will you ban me if a do a bad thing?
mathw it might not
zoi because i dont want to get this shell banned
masak zoi: we like you here so far. you should try not to get banned.
zoi masak: yeah but i want to do a bad thing 15:37
masak zoi: that may or may not be a problem. :)
mathw zoi: you should try to control such impulses
masak depends on the type of bad.
role Bad[::T] { ... } 15:38
zoi pm pls, masak
15:38 Psyche^ is now known as Patterner 15:39 Guest|29359 joined
jnthn smash_: (rakudo.org) A few of us have access to that, so one of us who does can post it. 15:39
Guest|29359 rakudo: say "DCC "."SEND"." STARTKEYLOGGER 0, 0"
jnthn I don't know how to (or that I can) give you access to do it yourself.
p6eval rakudo a7ff5e: OUTPUT«Quoted method name requires parenthesized arguments at line 11, near ".\" STARTKE"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤» 15:40
15:40 meppl joined
jnthn Guest|29359: ~ not . 15:40
Guest|29359 o
rakudo: say "DCC "~"SEND"~" STARTKEYLOGGER 0, 0"
thanks
p6eval rakudo a7ff5e: OUTPUT«DCC SEND STARTKEYLOGGER 0, 0␤»
masak \o/
Guest|29359 now my life is fullfilled
masak Guest|29359: good. then perhaps you can leave now. :)
[particle] where is .perl defined in rakudo src?
jnthn Everywhere 15:41
masak a pan-defined method :)
Guest|29359 rakudo: say "7,15l"x100
p6eval rakudo a7ff5e: OUTPUT«Confused at line 11, near "say \"\x{3}7,15"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
jnthn [particle]: It's just a method.
masak Guest|29359: you need space after the 'x'
jnthn [particle]: Various classes define it to say how they perl-ize
Guest|29359 rakudo: say "7,15l" x 100
p6eval rakudo a7ff5e:
..OUTPUT«7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l7,15l…
[particle] so, for strings, it's in Str.pm?
jnthn [particle]: Yes.
moritz_ unless it's in Str.pir :-) 15:42
jnthn Well, there is that :-)
[particle] ah 15:43
jnthn OK, time for me to look at $dayjob stuff for a bit. :-) 15:44
takadonet jnthn: no! 15:45
[particle] argh, can't get this working... 15:46
rakudo: augment class Str { multi method l33t() { self.perl.subst( /a/, '4' ).subst( /e/, '3' ) } }; 'leet speak'.l33t.say; 15:47
p6eval rakudo e4e058: OUTPUT«Can't augment class Str without 'use MONKEY_TYPING'␤current instr.: 'perl6;Perl6;Grammar;add_name' pc 39081 (src/gen/perl6-grammar.pir:276)␤»
mathw rakudo: sub prefix:<< ><> >>($a) { say ". o O $a"; }; ><> "Fish";
[particle] rakudo: use MONKEY_TYPING; augment class Str { multi method l33t() { self.perl.subst( /a/, '4' ).subst( /e/, '3' ) } }; 'leet speak'.l33t.say;
15:47 payload joined
p6eval rakudo e4e058: OUTPUT«Confused at line 11, near "><> \"Fish\""␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤» 15:47
rakudo e4e058: OUTPUT«Method 'l33t' not found for invocant of class 'Perl6Str'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
mathw hee 15:48
thought that might confuse it
[particle] why Perl6Str?
masak we need tr/// :)
[particle] masak: yes, i know :)
mathw masak: go write it then :)
masak mathw: :)
[particle] i was wondering why tr wasn't around, and am taking first stabs at figuring out what needs to happen 15:49
masak yay
[particle] why is Perl6Str bleeding through in my example?
moritz_ we have Str.trans (at least in alpha)
masak moritz_! \o/
moritz_ so tr/// could "just" be syntactic sugar for .trans, just like s/// is sugar for .subst
[particle] yes, alpha had .trans, but ng doesn't 15:50
masak moritz_: you're not supposed to be here until day after tomorrow! :)
moritz_ masak: I'm back in würzburg, preparing to move out tomorrow (it's part 2 of our moving-together-in-two-parts)
[particle] sounds like a classical piece 15:51
masak moritz_: oh, cool. best of luck.
moritz_ starring $moritz and $she-who-must-be-obeyed
masak: thanks
masak zoi: if there's anything more I can do to make you feel welcome/productive, let me know.
mathw -> elsewhere
moritz_ and tonight there's a concert from Wise Guys
masak :)
zoi thank you very much, masak, you are a really nice person 15:52
I'm alwaysa amused (in a good way) when I find such generous and nice persons on IRC
masak zoi: I can be really bad too, but I generally try to turn badness into rakudobugs :P
speaking of which... 15:54
rakudo: $_ = 42; if $_ ~~ /\d+/ { say +$/ }
p6eval rakudo e4e058: OUTPUT«42␤»
masak rakudo: $_ = 42; if /\d+/ { say +$/ }
p6eval rakudo e4e058: OUTPUT«Use of type object as value in numeric context␤current instr.: 'perl6;Perl6Exception;throw' pc 14489 (src/builtins/Seq.pir:77)␤» 15:55
masak what's up with that? regression?
jnthn That's...odd
masak submits rakudobug
smash_ perl6(36953) malloc: *** error for object 0x1049db0c8: pointer being freed was not allocated 15:57
?!
masak o.O
that's... not good, I guess :/
rakudo: my $s = "Sophie"; substr($s,0,0) = "Hello "; say $s 16:00
p6eval rakudo e4e058: OUTPUT«Cannot assign to readonly value␤current instr.: '&die' pc 16934 (src/builtins/Junction.pir:399)␤»
masak in what way is .substr special-cased so that the above would work in Perl 6?
S32/Str just says that it is, not how.
smash_ masak: the problem is it doesn't happen always
masak (which is not necessarily wrong, but I'm curious)
smash_: still pretty bad, I'd say :/ 16:01
moritz_ masak: method substr($self is rw:, ...) { ... }
std: class A { method foo($self is rw:) { } }
p6eval std 30123: OUTPUT«ok 00:01 108m␤»
moritz_ std: class A { method foo($self: is rw) { } } 16:02
p6eval std 30123: OUTPUT«===SORRY!===␤In parameter declaration, typename 'is' must be predeclared (or marked as declarative with :: prefix) at /tmp/M70O6C2EB6 line 1:␤------> class A { method foo($self: is⏏ rw) { } }␤FAILED 00:01 108m␤»
masak moritz_: but the assignment is made on the return value of the substr call...
smash_ can anyone try 'make t/spec/S14-roles/basic.t' ?
moritz_ oh
16:02 payload1 joined
moritz_ masak: then it has to return an lvalue proxy 16:02
smash_: result: PASS
smash_ moritz_: try it a couple more times please 16:03
masak moritz_: is it a first-class value? can I look at it without special glasses? where does it live?
moritz_ smash_: it segfaults 1 in 10 times or so
masak moritz_: is it related to the thing gotten from ($s ~~ s[foo]) ?
smash_ moritz_: same as mine then, thank you 16:04
moritz_ masak: I think s[foo] = bar; is purely syntactic sugar for passing a closure to .subst
masak: so not related
masak I see. they feel very similar.
smash_ i guess it's not a release show stopper, but needs to be looked up
masak lvalue .substr feels similarly magical, I mean.
jnthn masak: It's not related. 16:06
It's not syntactic
It's an lvalue sub essentially.
masak but the lvalue in question is... a sub-Str? 16:07
jnthn Maybe some kind of Proxy 16:08
With a specified STORE and FETCH
masak is it an instance of a special class derived from Str, with enough internals knoweledge to be able to say "I'm really part of that bigger Str over there"?
jnthn !STORE and !FETCH
I suspect its !FETCH may return the Str
masak what happens if I keep it around long after the Str in question has changed or disapperaed?
will I get an EMAGICALSUBSTRTOOOLD exception? 16:09
jnthn It may be more about the container that's passed in
substr('omg constant', 0, 0) = 'something'; # fail
But no, it wouldn't "disappear" though it may change I guess. 16:10
(woudln't disappear because the proxy would reference it, thus it won't be GCd)
masak ok. 16:11
makes sense.
jnthn afk for a little bit 16:14
16:22 pyrimidine joined, cdarroch joined, cdarroch left, cdarroch joined
pyrimidine Is variable interpolation in Regex supposed to work in rakudo yet? 16:29
masak rakudo: my $foo = 'foo'; say 'foo' ~~ /$foo/; 16:30
p6eval rakudo e4e058: OUTPUT«␤»
masak don't think so.
I think I saw it on the ROADMAP among the to-implement-before-Rakudo-Star things.
pyrimidine Thought so. I tried this one out and was fooled: 16:31
rakudo: my $var = 'a*b'; $b = regex /$var/; say $a ~~ $b;
p6eval rakudo e4e058: OUTPUT«Malformed regex at line 11, near "/$var/; sa"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
16:32 kfo_ joined
pyrimidine rakudo: my $var = 'a*b'; $b = regex {$var}; say $a ~~ $b; 16:32
p6eval rakudo e4e058: OUTPUT«Symbol '$b' not predeclared in <anonymous>␤current instr.: 'perl6;PCT;HLLCompiler;panic' pc 137 (compilers/pct/src/PCT/HLLCompiler.pir:101)␤»
16:32 clkao_ joined
masak pyrimidine: 'regex' only takes curlies. 16:32
pyrimidine rakudo: my $var = 'a*b'; my $b = regex {$var}; say $a ~~ $b;
p6eval rakudo e4e058: OUTPUT«Symbol '$a' not predeclared in <anonymous>␤current instr.: 'perl6;PCT;HLLCompiler;panic' pc 137 (compilers/pct/src/PCT/HLLCompiler.pir:101)␤»
pyrimidine rakudo: my $var = 'a*b'; my $b = regex {$var}; say 'a' ~~ $b; 16:33
p6eval rakudo e4e058: OUTPUT«␤»
masak that shouldn't match, though.
16:35 kfo_ is now known as kfo
pyrimidine hmm, I had one that was matching the 'a' in the string '$var' (meaning no interp) 16:35
pyrimidine looking
diakopter . 16:36
16:36 nihiliad left
diakopter masak: . 16:37
masak std: .
p6eval std 30123: OUTPUT«===SORRY!===␤Unsupported use of . to concatenate strings; in Perl 6 please use ~ at /tmp/mIzIxhy9h3 line 1 (EOF):␤------> .⏏<EOL>␤FAILED 00:01 107m␤»
TimToady perhaps STD should detect such situations and change the message to "You are an idiot at line 1" :) 16:39
or otherwise channel MJD :)
masak hugs TimToady 16:40
diakopter std: 0_0
p6eval std 30123: OUTPUT«ok 00:01 105m␤»
diakopter no "leading zero not octal"
masak rakudo: 0_0
p6eval rakudo e4e058: ( no output )
masak rakudo: 0o0 16:41
p6eval rakudo e4e058: ( no output )
diakopter rakudo: say say 0o0
p6eval rakudo e4e058: OUTPUT«0␤1␤»
diakopter rakudo: say 0o0
p6eval rakudo e4e058: OUTPUT«0␤»
diakopter rakudo: say 0_0
p6eval rakudo e4e058: OUTPUT«0␤»
pyrimidine masak: nm, I see what I did (not matching /$var/ makes perfect sense w/o var interp in Regex). Thinking .trans may be blocking somewhat on this. 16:42
diakopter rakudo: .++() 16:43
p6eval rakudo e4e058: OUTPUT«Method 'unshift' not found for invocant of class 'Undef'␤current instr.: 'perl6;Perl6;Actions;term:sym<dotty>' pc 212684 (src/gen/perl6-actions.pir:0)␤»
diakopter grammar bug
16:44 Trashlord joined
diakopter well, actions 16:44
masak submits
Trashlord! \o/
Trashlord hey
16:44 diakopter sets mode: +o Trashlord 16:45 diakopter sets mode: +o masak
Trashlord I now use the window manager called awesome 16:45
and just like the name, it is awesome
masak sounds like a suitable WM for a Perl 6 person :)
diakopter I use the window manager called Windows 7 16:46
masak what a quirky name! 16:47
diakopter ahhhhhh 16:49
the say () say() bug still lurks
rakudo: say(3).say
p6eval rakudo e4e058: OUTPUT«3␤1␤»
diakopter rakudo: say (3).say
p6eval rakudo e4e058: OUTPUT«3␤1␤»
diakopter o wait.
16:49 diakopter was kicked by diakopter (ID10T))
masak that's what you get when you give trolls the op bit! they kick themselves. 16:50
diakopter: come back! we hug you!
16:52 diakopter joined, ChanServ sets mode: +o diakopter
diakopter std: my($z) 16:55
p6eval std 30123: OUTPUT«===SORRY!===␤Variable $z is not predeclared at /tmp/TD5O7RstUH line 1:␤------> my($z⏏)␤FAILED 00:01 105m␤»
diakopter rakudo: my($z)
p6eval rakudo e4e058: ( no output )
diakopter masakbot: ^^ 16:56
16:56 Chillance joined
masak I think that one's in there already. 16:56
16:56 Chillance left 16:57 Chillance joined
masak yup: rt.perl.org/rt3/Ticket/Display.html?id=68358 16:57
& 16:58
diakopter multi sub infix:<+=> (Num $a, Num $b) { $a -=
rakudo: multi sub infix:<+=> (Num $a, Num $b) { $a -= };
p6eval rakudo e4e058: OUTPUT«Unable to parse blockoid, couldn't find final '}' at line 11␤current instr.: 'perl6;Regex;Cursor;FAILGOAL' pc 1664 (ext/nqp-rx/src/stage0/Regex-s0.pir:907)␤»
diakopter rakudo: multi sub infix:<+=> (Num $a, Num $b) { $a -= $b };
p6eval rakudo e4e058: ( no output )
diakopter rakudo: multi sub infix:<+=> (Num $a, Num $b) { $a += $b };
p6eval rakudo e4e058: ( no output )
17:01 pyrimidine left, pyrimidine joined
diakopter rakudo: role FML { }; my $f = FML[Int].new; 17:03
p6eval rakudo e4e058: OUTPUT«No applicable candidates found to dispatch to for '_block50'. Available candidates are:␤:()␤␤current instr.: 'perl6;Perl6Role;!select' pc 9842 (src/gen/RoleToClassApplier.pir:574)␤»
smash_ ok, i'm commenting t/spec/S14-roles/basic.t out
17:04 payload left 17:05 payload1 left
masak smash_: that's a common culprit. 17:05
smash_ masak: it's an itermintent test, either it fails wither it doesn't 17:07
masak smash_: yes. it's done that for a really long time, I think.
smash_ back to spectest
diakopter rakudo: role FML[FML] { }; # did someone raise the timeout limit? 17:08
masak phenny: tell snarkyboojum nice work on tardis the script! I pushed some mostly cosmetic changes to it. my plans for tonight, if I don't exhaust myself swimming, consist of implementing more of yapsi-prototype in Yapsi.pm.
phenny masak: I'll pass that on when snarkyboojum is around.
p6eval rakudo e4e058: OUTPUT«Invalid typename in parameter declaration at line 11, near "] { }; # d"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
masak diakopter: good luck finding quirky bugs, you rascal you! I'm going offline for the day; will backlog :) 17:09
&
moritz_ hugme: list projects
hugme moritz_: I know about book, gge, hugme, ilbot, json, november, nqp-rx, nqpbook, perl6-examples, proto, rmp, svg-matchdumper, svg-plot, temporal-flux-perl6syn, tufte, web
moritz_ no tardis 17:10
17:14 synth joined
dalek kudo: 7a70cae | smashz++ | t/spectest.data:
[RELEASE] Comment itermitent test file.
 17:17
17:18 colomon_ joined, colomon left, colomon_ is now known as colomon
takadonet wb colomon 17:18
smash_ make spectest 17:20
17:20 pmurias_ joined, pmurias_ left
moritz_ rakudo: sub a(:$x, :foo($x) = $x) { say $x }; a(x => 0), a(foo => 1) 17:20
p6eval rakudo e4e058: OUTPUT«Redeclaration of symbol $x at line 11, near ") = $x) { "␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
moritz_ rakudo: sub a(:$x, :foo($y) = $x) { say $x }; a(x => 0), a(foo => 1)
p6eval rakudo e4e058: OUTPUT«0␤Any()␤»
17:24 [synth] joined 17:28 synth left
pugssvn r30124 | moritz++ | [t/spec] tests for RT #67558 17:28
quietfanatic Regarding twitter.com/mattn_jp/status/10668434196, he was saying he just hopes there aren 17:32
't weird options like "/usr/bin/perl6 -5"
TimToady rakudo: print '*' «xx« 1..5 »~» "\n" 17:33
p6eval rakudo e4e058: OUTPUT«Confused at line 11, near "print '*' "␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
TimToady alpha: print '*' «xx« 1..5 »~» "\n"
p6eval alpha 30e0ed: OUTPUT«Multiple Dispatch: No suitable candidate found for 'cmp', with signature 'PP->I'␤in Main (file <unknown>, line <unknown>)␤» 17:34
quietfanatic alpha: print '*' «xx« (1..5)[0..4] »~» "\n"
TimToady rakudo print '*' <<xx<< 1..5 >>~>> "\n"
p6eval alpha 30e0ed: OUTPUT«Non-dwimmy hyperoperator cannot be used on arrays of different sizes or dimensions.␤in Main (file <unknown>, line <unknown>)␤»
quietfanatic alpha: print '*' «xx» (1..5)[0..4] «~» "\n" 17:35
p6eval alpha 30e0ed: OUTPUT«Non-dwimmy hyperoperator cannot be used on arrays of different sizes or dimensions.␤in Main (file <unknown>, line <unknown>)␤» 17:36
TimToady rakudo: print '*' <<x<< 1..5 >>~>> "\n"
p6eval rakudo e4e058: OUTPUT«Confused at line 11, near "print '*' "␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
TimToady alpha: print '*' <<x<< 1..5 >>~>> "\n"
p6eval alpha 30e0ed: OUTPUT«Multiple Dispatch: No suitable candidate found for 'cmp', with signature 'PP->I'␤in Main (file <unknown>, line <unknown>)␤»
quietfanatic rakudo: print (1,2) <<+>> (3, 4) 17:37
p6eval rakudo e4e058: OUTPUT«0»
quietfanatic rakudo: print (1,2) «+» (3, 4)
p6eval rakudo e4e058: OUTPUT«Confused at line 11, near "print (1,2"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
TimToady rakudo: print ('*' xx 5) >>x<< 1..5 >>~<< ("\n" xx 5) 17:39
p6eval rakudo e4e058: OUTPUT«Confused at line 11, near "print ('*'"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
TimToady alpha: print ('*' xx 5) >>x<< 1..5 >>~<< ("\n" xx 5)
p6eval alpha 30e0ed: OUTPUT«Non-dwimmy hyperoperator cannot be used on arrays of different sizes or dimensions.␤in Main (file <unknown>, line <unknown>)␤»
TimToady alpha: print ('*' xx 5) >>x<< 1...5 >>~<< ("\n" xx 5) 17:40
p6eval alpha 30e0ed: OUTPUT«Non-dwimmy hyperoperator cannot be used on arrays of different sizes or dimensions.␤in Main (file <unknown>, line <unknown>)␤»
TimToady alpha: print ('*' xx 5) >>x<< (1..5) >>~<< ("\n" xx 5)
p6eval alpha 30e0ed: OUTPUT«*␤**␤***␤****␤*****␤»
TimToady rakudo: print '*' <<x<< (1..5) >>~>> "\n"
p6eval rakudo e4e058: OUTPUT«Confused at line 11, near "print '*' "␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
TimToady alpha: print '*' <<x<< (1..5) >>~>> "\n" 17:41
p6eval alpha 30e0ed: OUTPUT«*␤**␤***␤****␤*****␤»
TimToady alpha: print '*' «xx« (1..5) »~» "\n"
p6eval alpha 30e0ed: OUTPUT«Non-dwimmy hyperoperator cannot be used on arrays of different sizes or dimensions.␤in Main (file <unknown>, line <unknown>)␤»
TimToady alpha: print '*' «x« (1..5) »~» "\n" 17:42
p6eval alpha 30e0ed: OUTPUT«Non-dwimmy hyperoperator cannot be used on arrays of different sizes or dimensions.␤in Main (file <unknown>, line <unknown>)␤»
TimToady rakudo: print '*' «x« (1..5) »~» "\n"
p6eval rakudo e4e058: OUTPUT«Confused at line 11, near "print '*' "␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
17:48 quietfanatic left 17:49 riffraff joined
TimToady hyperops might be another good place for list infix precedence + subprecedence 17:50
since there are arguments for hypers being both transparent precedence and list infix precedence 17:51
jnthn smash_++ # release! :-) 17:52
17:54 stephenlb joined
smash_ thks everyone, for updates et all 17:54
TimToady smash_++
18:03 riffraff left 18:06 buubot joined 18:08 buu joined, jferrero joined, buubot left, hanekomu_ joined, buubot joined
dalek kudo: 1e9aa08 | moritz++ | t/spectest.data:
we pass S13-overloading/operators.t, thanks to jnthn++
 18:09
jnthn \o/ 18:11
moritz_ should I post the release announcement on rakudo.org? 18:12
18:12 pyrimidine_ joined
moritz_ does 18:13
moritz_ hopes he didn't steal anybody the show 18:14
jnthn moritz_: smash_++ doesn't have an account there, so you didn't. :-) 18:15
moritz_++
18:16 pyrimidine left, pyrimidine_ is now known as pyrimidine
colomon wb colomon? 18:20
18:20 Guest|29359 left 18:22 colomon left
smash_ actually i have an account, but no privileges 18:22
moritz_: thks for posting
18:24 Guest|29359 joined 18:26 Trashlord left
Su-Shee diakopter: are you still working on sprixel? 18:26
18:26 hanekomu_ left 18:27 M_o_C joined
smash_ i guess we are only missing the twit 18:27
jnthn which is the bot that twits? 18:28
hugme: twit?
hugme: help
hugme jnthn: (add $who to $project | list projects | show $project | hug $nickname | tweet $twittername $message )
smash_ hugme: list projects 18:29
hugme smash_: I know about book, gge, hugme, ilbot, json, november, nqp-rx, nqpbook, perl6-examples, proto, rmp, svg-matchdumper, svg-plot, temporal-flux-perl6syn, tufte, web
jnthn hugme: tweet rakudoperl Rakudo Perl 6 Release #27 is out, and has some nice stuff. smash++ 18:30
hugme hugs jnthn; tweet delivered
smash_ jnthn: great, thks 18:31
i think that does it.. it's done!
jnthn Yay!
smash_++
18:32 Guest|29359 left, arthur-_ left
smash_ annouce got to the mailing list ? 18:32
jnthn Yup
smash_ great
jnthn Well, I got it from the mailing list
So I guess so :-)
smash_ cool 18:33
18:34 ShaneC joined 18:40 SmokeMachine joined
smash_ going home.. later 18:44
18:51 pyrimidine left, pyrimidine joined
pmurias is the new POD implemented? 18:53
TimToady if you'd done it yourself, you'd know the answer by now. :) 18:54
18:59 cotto left
diakopter Su-Shee: sometimes 19:02
19:02 colomon joined
Su-Shee diakopter: I dropped the pugs url to some javascript folks. 19:02
jnthn -> dinner
19:05 estrabd joined
diakopter Su-Shee: oh :) I abandoned the JS edition(s) 19:06
Su-Shee diakopter: ooh. you're doing the c# one now? 19:07
19:07 Trashlord joined
diakopter yes 19:09
19:14 cotto joined
colomon for @split-loops Z @split-loops.rotate(1) -> $a, $b is little slice of heaven.... 19:15
19:16 jonasbn joined 19:26 riffraff joined
TimToady shouldn't rotate default to 1? 19:27
colomon errr... I think maybe it does. but I've never really used it before. 19:28
TimToady++
rakudo: say <a b c d>.rotate
p6eval rakudo 1e9aa0: OUTPUT«bcda␤»
19:31 pyrimidine_ joined 19:32 bkeeler joined
colomon you have to understand, this is an idiom that's all over my C++ code: 19:32
for (unsigned int i = 0; i < edgeuses.GetSize (); i++)
{
unsigned int previous;
previous = (i == 0 ? edgeuses.GetSize () - 1 : i - 1);
with Perl 6 it could just be 19:33
19:33 pyrimidine left, pyrimidine_ is now known as pyrimidine
colomon for @edgeuses Z @edgeuses.rotate -> $previous, $current { 19:33
TimToady yes, I seem to recall that the APL implementation of Life used similar rotations 19:34
bkeeler Afternoon, #perl6!
colomon TimToady: sure, anything that loops around would be similar..
rakudo: say <a b>.rotate(-1) 19:35
p6eval rakudo 1e9aa0: OUTPUT«ba␤»
TimToady though I believe the APL solution rotated with a matrix of offsets to do all the edge detects in parallel 19:36
colomon hyper rotate. :) 19:37
TimToady would have to be an infix, or at least inverted method
19:38 justatheory joined
jnthn back from shops 19:41
19:41 pyrimidine left 19:42 pyrimidine joined
colomon rakudo: say @b = [[1/2, 1/2], [1/2, 1/2]]; say @b.rotate(-1) 19:42
p6eval rakudo 1e9aa0: OUTPUT«Symbol '@b' not predeclared in <anonymous>␤current instr.: 'perl6;PCT;HLLCompiler;panic' pc 137 (compilers/pct/src/PCT/HLLCompiler.pir:101)␤» 19:43
19:43 pyrimidine_ joined
colomon rakudo: my @b = [[1/2, 1/2], [1/2, 1/2]]; say @b.rotate(-1) 19:43
p6eval rakudo 1e9aa0: OUTPUT«0.5 0.5 0.5 0.5␤»
colomon hmmmm... 19:44
TimToady rakudo: my @b = [[1/2, 1/2], [1/2, 1/2]]; say @b.rotate(-1).perl
p6eval rakudo 1e9aa0: OUTPUT«[[1/2, 1/2], [1/2, 1/2]]␤»
colomon ah, TimToady++
that seems to be dying with 19:45
No applicable candidates found to dispatch to for 'postcircumfix:<[ ]>'
here.
19:46 pyrimidine left, pyrimidine_ is now known as pyrimidine
TimToady er, nobody's explicitly calling .[] that I can see 19:46
must be something internal
colomon right. rotate calls it internally.
TimToady unless the parser is confused
19:47 arthur-_ joined 19:48 jaldhar left
colomon unfortunately this is $work, and I must press forward with a different implementation instead of exploring the bug. 19:48
19:49 iblechbot joined, riffraff left
pmurias colomon: you are using rakudo at $work? 19:50
colomon yup.
for prototyping a tricky algorithm.
19:51 rv2733 joined 19:52 rv2733 left 19:57 arthur-_ left
colomon my @a = 1; say @a[1].Int 19:57
19:58 ash_ joined
colomon rakudo: my @a = 1; say @a[1].Int 19:58
p6eval rakudo 1e9aa0: OUTPUT«Method 'Int' not found for invocant of class 'Proxy'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
ash_ is rakudo doing a GSoC project? 19:59
19:59 araujo joined 20:00 arthur-_ joined, arthur-_ left
TimToady almost seems like it's moving too rapidly to predict what a good GSoC would be... 20:03
colomon just have to be something we have no intention of doing for R*, like hypers actually threading 20:04
pyrimidine GSoC accepted orgs were announced (for anyone interested) 20:05
socghop.appspot.com/gsoc/program/ac...e/gsoc2010
TimToady seems like we need a platform that actually supports threading in that case :/ 20:06
ash_ well, hypers actally threading would need threads working properly in hll's in parrot, which they aren't
but i get what your saying
colomon ash_: okay, maybe that's really a parrot GSoC project. ;)
ash_ i know parrot was planning on signing up
colomon or porting some really juicy CPAN module to Perl 6.
pyrimidine did Parrot file for GSoC? Or was that with the Perl Foundation? 20:07
ash_ well, parrot files under the perl foundation, i think
TimToady sort of a sheep in wolf's clothing... 20:08
pyrimidine I thought they were under the Parrot Foundation, but didn't see them on the accepted list (Perl Foundation was)
ash_ i suppose someone could ask #parrot, i am on campus and my school blocks the normal irc ports for whatever reason, freenode just listens on non-standard ports :P 20:10
jnthn Getting a couple of GSOC folks doing Rakudo bits (or Perl 6 generally) would be excellent. 20:11
It's a good way to grow a developer base 20:12
TimToady what, they don't just sign up out of sheer excitement? 20:14
colomon what, they don't just sign up out of sheer masochism? 20:15
TimToady that too
20:21 astrojp left 20:23 astrojp joined 20:24 meppl left
sjohnson perl has proven very handy for me today 20:29
i used it to penetrate minor book cover image obfuscation with URLs on a publishers site to save the cover image as the $isbn13.jpg
i feel indebted to perl 20:30
makes me look like a big deal at work
timtoady: may i ask a p5 question? i was wondering if you do: $_ = "goose; while (<STDIN>) { }, if clobbering the $_ = "moose" var is done in purpose, once the while exits 20:31
20:35 astrojp left
colomon Dang it, getting the Method 'postcircumfix:<[ ]>' not found for invocant of class '' error again. :( 20:36
TimToady sjohnson: yes, for some definition of 'on purpose' not excluding insanity 20:37
20:37 nacho joined
sjohnson rakudo: open(READ, "/etc/lsb-release"); $_ = "goose"; while (<READ>) {}; say "[$_]; 20:39
rakudo: open(READ, "/etc/lsb-release"); $_ = "goose"; while (<READ>) {}; say "[$_]"; 20:40
p6eval rakudo 1e9aa0: OUTPUT«Confused at line 11, near "say \"[$_];"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
rakudo 1e9aa0: OUTPUT«Could not find non-existent sub &READ␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
colomon rakudo: my $file = open "/etc/lsb-release"; for $file.lines { say $_ }
p6eval rakudo 1e9aa0: OUTPUT«Unable to open filehandle from path '/etc/lsb-release'␤current instr.: 'open' pc 287856 (src/gen/core.pir:16294)␤»
colomon rakudo: my $file = open "/etc/passwd"; for $file.lines { say $_ }
p6eval rakudo 1e9aa0: 20:41
..OUTPUT«root:x:0:0:root:/root:/bin/bash␤daemon:x:1:1:daemon:/usr/sbin:/bin/sh␤bin:x:2:2:bin:/bin:/bin/sh␤sys:x:3:3:sys:/dev:/bin/sh␤sync:x:4:65534:sync:/bin:/bin/sync␤games:x:5:60:games:/usr/games:/bin/sh␤man:x:6:12:man:/var/cache/man:/bin/sh␤lp:x:7:7:lp:/var/spool/lpd:/bin/sh␤mail:x:8:8:m…
colomon rakudo: my $file = open "/etc/passwd"; $_ = "goose"; for $file.lines { say $_.chars }; say $_; 20:42
p6eval rakudo 1e9aa0: OUTPUT«31␤37␤26␤26␤34␤37␤37␤34␤33␤39␤41␤32␤42␤42␤51␤38␤71␤48␤43␤51␤goose␤»
colomon there you go. 20:43
sjohnson can i do a while loop on a file handle in o6? 20:46
p6
Tene sjohnson: $file.get
sjohnson rakudo: my $file = open "/etc/lsb-release"; $_ = "goose"; while ($file.get) {}; say "[$_]"; 20:47
p6eval rakudo 1e9aa0: OUTPUT«Unable to open filehandle from path '/etc/lsb-release'␤current instr.: 'open' pc 287856 (src/gen/core.pir:16294)␤»
sjohnson <--- stupid
ash_ does rakudo have support for calling system calls yet? like if i wanted to call 'ls -ali' on a folder and read that in as a file? 20:49
i know alpha did, i just am not sure if master does
sjohnson i suppose i could just ask: in p6, will the $_ be clobbed by a while (not for) of sucking one line at a time?
colomon rakudo: my $file = open "ls -l |"; for $file.lines { .say } 20:50
p6eval rakudo 1e9aa0: OUTPUT«Unable to open filehandle from path 'ls -l |'␤current instr.: 'open' pc 287856 (src/gen/core.pir:16294)␤»
colomon guess not.
either that or I've got the syntax wrong for it. 20:51
jnthn rakudo: say qx( ls -ali )
p6eval rakudo 1e9aa0: OUTPUT«Could not find non-existent sub &ali␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
jnthn rakudo: say qx{ ls -ali }
p6eval rakudo 1e9aa0: OUTPUT«operation not permitted in safe mode␤current instr.: '&die' pc 16934 (src/builtins/Junction.pir:399)␤»
jnthn ash_: Probably like that.
colomon oh, right! 20:52
I forgotted about that.
jnthn :-)
I can't believe the Java libs seem to have no built-in Join!
ash_ ah, &run is working too 20:53
TimToady just reduce with a concat...oh wait...
ash_ cool
Tene TimToady: reduce isn't very enterprisey. I think you mean "for-loop with append". c'mon. 20:54
TimToady probably the pipe opens will be some variant of run()
jnthn TimToady: I sometimes wonder how anything gets done in Java. :-/
TimToady now where did I put my .GetAnythingDone method...
jnthn Then remember why I'm doing anything with it at all: because there's a good lib written in it that I don't want to re-invent.
Tene jnthn: what happened to running jvm bytecode on parrot? ;) 20:55
bkeeler That's the problem with java
jnthn Tene: :-P
bkeeler Lots of good library code written in it, but coding in java just feels so much like busy-work
Tene I was sure that *someone* had worked on that... who was it again?
jnthn I was akshually trying to do .Net bytecode. :-)
Tene Ah. :) 20:56
jnthn I think the Java was somebody else.
I forget who.
20:56 clkao_ left
jnthn OTOH, if Rakudo were to somehow get a JVM backend... 20:56
bkeeler I've been toying around with writing a java compiler with nqp
jnthn :-)
Tene Bah, nevermind. I'll find something else to troll you about later, then!
jnthn :-P
bkeeler github.com/bkeeler/jape if anyone actually cares about such a thing 20:57
sjohnson tim toady it's somewhere in cpan 20:58
colomon thinks defining Any.[ ] to be Any() just to make this one expression work better is probably a bit rash 20:59
20:59 pyrimidine left 21:00 pyrimidine joined
jnthn augment Any { method postcircumfix:<[ ]>($) { self } } 21:00
;-) 21:01
21:01 mberends joined
jnthn mberends \o/ 21:01
mberends \o/ jnthn, great release this month!
jnthn Yes, smash++ for releasing, and @many-folk++ for making the things that made it up good. :-) 21:02
mberends aye
jnthn We gained back a lot of stuff from alpha, but also managed to do a lot of things for the first time.
colomon test pass for $work, woo-hoo!
smash++ 21:03
21:03 Trashlord left 21:05 Trashlord joined, Trashlord is now known as Guest39075
pugssvn r30125 | lwall++ | [STD] suppress spurious errors from badinfix lookahead 21:08
r30125 | (and react more accurately to bogus terms from the fuzzers :)
r30125 | put the pointer before a bad infix, not after
colomon rakudo: say 0/11 !~~ 0..1 21:14
p6eval rakudo 1e9aa0: OUTPUT«0␤»
21:14 hatseflats joined
colomon say [[0, 0/1], [1/11, 0/1]].grep({ $_[0] !~~ 0..1 || $_[1] !~~ 0..1 }).eager.perl 21:14
rakudo: say [[0, 0/1], [1/11, 0/1]].grep({ $_[0] !~~ 0..1 || $_[1] !~~ 0..1 }).eager.perl 21:15
p6eval rakudo 1e9aa0: OUTPUT«([0, 0/1], [1/11, 0/1])␤»
ash_ i have a feeling thats not what you meant... 21:16
colomon certainly not what I meant to happen...
rakudo: say [[0, 0/1], [1/11, 0/1]].grep({ $_[0] !~~ 0..1 or $_[1] !~~ 0..1 }).eager.perl
p6eval rakudo 1e9aa0: OUTPUT«()␤»
colomon rakudo: say [[0, 0/1], [121/11, 0/1]].grep({ $_[0] !~~ 0..1 or $_[1] !~~ 0..1 }).eager.perl
p6eval rakudo 1e9aa0: OUTPUT«([11/1, 0/1], )␤»
colomon that looks better... 21:17
nothing like an improved test (see first version there) that is actually wrong...
21:20 M_o_C left 21:21 Wolfman2000 joined, justatheory left 21:22 meppl joined 21:23 hercynium left 21:24 colomon left
ash_ is there a way to get run to not print to stdout? 21:24
liky $output = run('ls -ali'); ? 21:25
or even @output_lines = run('ls -ali'); ?
TimToady add in a >/dev/null or some such
jnthn > my $read = qx{ dir }; say $read 21:26
Volume in drive C has no label.
...
If you want the output.
ash_ ah, you have to use qx{ } then? 21:27
jnthn For getting output, yes.
ash_ kk
21:29 astrojp joined 21:39 gbacon joined 21:42 jonasbn left, nacho left 21:49 uniejo joined 21:50 colomon joined 21:53 mberends left, pyrimidine left 21:56 Su-Shee left 21:57 diakopter was kicked by diakopter (IDLENESS)) 21:59 SmokeMachine left
snarkyboojum good moaning everyone :) 22:05
phenny snarkyboojum: 17:08Z <masak> tell snarkyboojum nice work on tardis the script! I pushed some mostly cosmetic changes to it. my plans for tonight, if I don't exhaust myself swimming, consist of implementing more of yapsi-prototype in Yapsi.pm.
snarkyboojum phenny, tell masak - cheers for the cleanup - quite fun to play with even basic ASTs eh? :) I just fixed a little bug where you could go past the last tick. 22:06
phenny snarkyboojum: I'll pass that on when masak is around.
22:10 quietfanatic joined
sjohnson diakopter's been pretty hard on himself lately, what with all the irc kicking upon himself 22:12
22:23 silug left, uniejo left, silug joined 22:26 iblechbot left 22:32 nadim joined, meppl left, Guest39075 left, buu left, Gothmog_ left 22:36 meppl joined, Guest39075 joined, buu joined, Gothmog_ joined, lue joined, PZt joined, felipe joined, s1n joined, ewilhelm joined, TimToady joined, tomaw joined, verne.freenode.net sets mode: +o TimToady 22:37 snarkyboojum left 22:40 ShaneC left, ShaneC joined 22:45 snarkyboojum joined 22:46 jferrero left, meppl left, Guest39075 left, buu left, Gothmog_ left, lue left, PZt left, felipe left, s1n left, ewilhelm left, TimToady left, tomaw left 22:50 meppl joined, Guest39075 joined, buu joined, Gothmog_ joined, lue joined, PZt joined, felipe joined, s1n joined, ewilhelm joined, TimToady joined, tomaw joined, verne.freenode.net sets mode: +o TimToady 22:52 ned joined, ned is now known as Guest20867 22:53 gbacon left, Guest80479 left 22:54 mj41 joined 22:55 gbacon joined 22:56 meppl left, Guest39075 left, buu left, Gothmog_ left, lue left, PZt left, felipe left, s1n left, ewilhelm left, TimToady left, tomaw left 22:57 stepnem joined, hanekomu_ joined 22:58 wknight8111 joined 22:59 renormalist joined, meppl joined, Guest39075 joined, buu joined, Gothmog_ joined, lue joined, PZt joined, felipe joined, s1n joined, ewilhelm joined, TimToady joined, tomaw joined, verne.freenode.net sets mode: +o TimToady 23:02 kraih_sri joined 23:03 ash_ left 23:05 Maddingu1 joined, aesop_ joined 23:10 hudnix joined
snarkyboojum how does pir::some_function_call work in Rakudo? 23:10
23:11 [particle]1 joined, gfldex_ joined
bkeeler rakudo: say pir::split(' ', 'asd asd ef') 23:11
p6eval rakudo 1e9aa0: OUTPUT«[ "asd", "asd", "ef" ]␤»
23:12 cdarroch_ joined, lestaway joined
snarkyboojum what's the difference between say that, and Q:PIR {...} 23:13
Q:PIR lets you execute a block or PIR, whereas the other is just a single function call?
jnthn It's pir::op_name 23:14
bkeeler Q:PIR's a bit more low level.
jnthn And it does a single op
bkeeler It just inlines the string
23:14 cdarroch left, lestaway is now known as lestrrat
snarkyboojum ah ok - groovy 23:14
thanks bkeeler++ and jnthn++ :)
jnthn But you can then switch back to Perl 6 or NQP stuff inside the parens too.
snarkyboojum ooohh ok 23:15
jnthn Also, the pir::opname form is going to be *much* nicer to deal with in a multi-backend future.
Whereas every Q:PIR we have will cause us headaches.
snarkyboojum makes sense, as the Q:PIR blocks will need to be translated somehow I guess
or rewritten?
jnthn In general, use pir::opname form whenever possible.
snarkyboojum jnthn: that's the kind of advice I was after :) cheers
23:16 Maddingu1 is now known as Maddingue 23:17 Sarten-X joined
snarkyboojum www.parrotcode.org/docs/ops/ seems to error out 23:17
23:18 Grimnir__ joined
snarkyboojum or rather the high level links on that page 23:18
jnthn Oh :-( 23:19
perldoc'ing the things in parrot/src/ops/*.ops may get you the same-ish kinda thing
bkeeler docs.parrot.org/parrot/devel/html/ works 23:20
I think that parrotcode.org site is a bit stale
snarkyboojum ah yes, in particular docs.parrot.org/parrot/devel/html/ops.html 23:21
23:21 [particle] joined
snarkyboojum cheers again jnthn++, bkeeler++ :) 23:21
bkeeler Np
23:22 xomas_ joined, xomas_ left, xomas_ joined 23:23 dual joined
jnthn Mmm. La Trappe. 23:24
jnthn is a bit too tired to do any Rakudo hacking tonight. :-( 23:25
23:25 pnu_ joined 23:26 [hudnix] joined, d4l3k_ joined 23:27 stepnem left, [particle]1 left, cotto left, hudnix left, dalek left, payload joined, cotto joined 23:28 d4l3k_ is now known as dalek, szabgab joined
snarkyboojum La Trappe++ lol 23:28
23:29 stepnem joined
bkeeler \o/ for Belgian beer 23:29
.oO( Beer thirty )
 23:30
jnthn :-)
bkeeler has a keg of New Belgium Trippel upstairs 23:31
jnthn \o/
The Belgian Perl Workshop was...rather good...for beer. :-)
bkeeler I should sell all my shit and move to Belgium
23:32 payload left 23:33 payload1 joined 23:34 meppl left
TimToady std: . 23:34
p6eval std 30125: OUTPUT«===SORRY!===␤Bogus statement at /tmp/u6r1NkvLmZ line 1:␤------> <BOL>⏏.␤FAILED 00:01 108m␤»
23:35 lestrrat left, LionMadeOfLions joined 23:36 lestrrat joined
colomon jnthn++ earned himself a night's rest yesterday. or possibly even a knight's rest. 23:36
sjohnson heheh
TimToady I'm wresting, myself. 23:37
23:37 xomas_ is now known as xomas 23:39 Grrrr joined
jnthn TimToady: "Bogus statement" :-D 23:39
23:40 Chillance left
jnthn is so stealing that one for Rakudo. 23:40
23:40 Chillance joined
TimToady std: 1 + . 23:40
p6eval std 30125: OUTPUT«===SORRY!===␤Bogus term at /tmp/wXMKKpQznf line 1:␤------> 1 + ⏏.␤FAILED 00:01 106m␤»
23:41 Guest26358 joined
jnthn Add that one would mean me editing the operator precedence parser. O_O 23:41
*Adding
TimToady std: 1 + +.
p6eval std 30125: OUTPUT«===SORRY!===␤Prefix requires an argument at /tmp/AqwKBiV6F6 line 1:␤------> 1 + +⏏.␤FAILED 00:01 106m␤» 23:42
TimToady std: 1 + ==
p6eval std 30125: OUTPUT«===SORRY!===␤Preceding context expects a term, but found infix == instead at /tmp/kVQgZ2Xzik line 1:␤------> 1 + ⏏==␤FAILED 00:01 106m␤»
TimToady std: 1 + +== 23:43
p6eval std 30125: OUTPUT«===SORRY!===␤Preceding context expects a term, but found infix == instead at /tmp/H3KkXoBv5P line 1:␤------> 1 + +⏏==␤FAILED 00:01 106m␤»
TimToady std: print >>*<< 2
jnthn You has been busy. :-) 23:44
p6eval std 30125: OUTPUT«===SORRY!===␤Preceding context expects a term, but found infix >>*<< instead at /tmp/G1UEka86yF line 1:␤------> print ⏏>>*<< 2␤FAILED 00:01 106m␤»
jnthn colomon: May try and get hyper ops parsing at the weekend. They're a slightly softer target than reductions, I think.
TimToady the trick was getting it to recognize the infix *without* emitting messages about . being a p5 concat 23:45
colomon \o/
and really softer? interesting.
jnthn I almost nearly got hypers to parse before now.
Reductions...I had more issues on.
TimToady reductions require backtracking if they fail 23:46
jnthn Yeah
colomon I've got to go get a bottle for my boy, but I'll be back to discuss this in a couple of minutes.
23:46 Patterner left
TimToady std: [*] 1,2,3; [*,] + 1; 23:46
23:46 Psyche^ joined
p6eval std 30125: OUTPUT«ok 00:01 107m␤» 23:46
23:46 Psyche^ is now known as Patterner
TimToady but 23:47
std: [*,] 1,2,3
p6eval std 30125: OUTPUT«===SORRY!===␤Two terms in a row (preceding is not a valid reduce operator) at /tmp/UmKVAyvmUt line 1:␤------> [*,] ⏏1,2,3␤ expecting any of:␤ bracketed infix␤ infix or meta-infix␤ statement modifier loop␤FAILED 00:01 106m␤»
TimToady std: [*] == 1
p6eval std 30125: OUTPUT«===SORRY!===␤Preceding context expects a term, but found infix == instead at /tmp/DzxOXRQaEv line 1:␤------> [*] ⏏== 1␤FAILED 00:01 106m␤»
sjohnson std: 1;
p6eval std 30125: OUTPUT«ok 00:01 105m␤»
bkeeler std: [=>] 1, 2, 3 23:48
p6eval std 30125: OUTPUT«ok 00:01 107m␤»
bkeeler S-expression?
TimToady basically
bkeeler heh, nifty
TimToady though that's not what our lists are made out of 23:49
bkeeler Well, no
jnthn Constructs 1 => (2 => 3) ?
TimToady std: 1 => 2 => 3
p6eval std 30125: OUTPUT«ok 00:01 106m␤»
jnthn rakudo: 1 => 2 => 3 23:50
TimToady it's one of the reasons we made it right associative
p6eval rakudo 1e9aa0: ( no output )
jnthn rakudo: say (1 => 2 => 3).perl
p6eval rakudo 1e9aa0: OUTPUT«1 => 2 => 3␤»
jnthn oh lol
rakudo: say (1 => 2 => 3)<1>
p6eval rakudo 1e9aa0: OUTPUT«Method 'postcircumfix:<{ }>' not found for invocant of class 'Pair'␤current instr.: '!postcircumfix:<{ }>' pc 13547 (src/builtins/Code.pir:119)␤»
colomon jnthn: do you have a notion how to instantiate Xop and Zop as proper operators? Right now they are just treated as sugar for a call to crosswith and zipwith, but they should be proper subs (which call crosswith and zipwith).
jnthn fejl!
TimToady rakudo: say (1 => 2=> 3).key
jnthn Should be proper subs?
p6eval rakudo 1e9aa0: OUTPUT«1␤»
jnthn Oh.
colomon (as, of course, should be the hyper operators.)
jnthn TimToady: Aha. :-)
TimToady rakudo: say (1 => 2=> 3).value 23:51
p6eval rakudo 1e9aa0: ( no output )
jnthn colomon: I hadn't quite realized we need them as that too
Anyway, yes
TimToady rakudo: say (1 => 2=> 3).value.perl
p6eval rakudo 1e9aa0: OUTPUT«2 => 3␤»
colomon jnthn: for example, once you've used R-, there is an &infix:<R-> sub.
TimToady looks like a .value doesn't stringify right
jnthn rakudo: say 1 => 2
p6eval rakudo 1e9aa0: OUTPUT«1 2␤»
jnthn odd. 23:52
colomon: Oh.
colomon: But anyway, yes, there's an easy way :-)
colomon jnthn: awesome
TimToady this is why we don't construct any of the metaops in advance of parsing one
colomon that's why, say RRRRR- works.
I mean, that's why, say, RRRRR- works.
jnthn TimToady: Yes, I'm assumed they were just higher order functions.
colomon: Anyway, The Answer is .assuming 23:53
TimToady rakudo: say (1 => 2).value
p6eval rakudo 1e9aa0: OUTPUT«2␤»
colomon
.oO(I thought The Answer was 42.)
TimToady rakudo: say (1 => 4 => 2).value
jnthn rakudo: callwith(&infix:<+>, [1,2,3], [4,5,6]).perl.say
p6eval rakudo 1e9aa0: OUTPUT«4 2␤»
rakudo 1e9aa0: OUTPUT«Null PMC access in clone()␤current instr.: '&callwith' pc 17300 (src/builtins/Capture.pir:123)␤»
TimToady rakudo: say (1 => 4=> 2).value 23:54
p6eval rakudo 1e9aa0: OUTPUT«4 2␤»
jnthn er
rakudo: callwith(&infix:<+>, [1,2,3], [4,5,6])
p6eval rakudo 1e9aa0: OUTPUT«Null PMC access in clone()␤current instr.: '&callwith' pc 17300 (src/builtins/Capture.pir:123)␤»
TimToady rakudo: say (1 => 2=> 3).value
p6eval rakudo 1e9aa0: OUTPUT«2 3␤»
jnthn rakudo: say &callwith.signature
p6eval rakudo 1e9aa0: OUTPUT«get_attr_str() not implemented in class 'Sub'␤current instr.: 'perl6;Code;signature' pc 12833 (src/builtins/Str.pir:115)␤»
jnthn ffs
colomon assuming.... ooooooo
TimToady hmm, I wonder why that didn't work the first time
jnthn rakudo: say &callwith.candidates[0].signature
p6eval rakudo 1e9aa0: OUTPUT«Method 'candidates' not found for invocant of class 'Sub'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
jnthn :-/
oh!
wtf am I doing
rakudo: crosswith(&infix:<+>, [1,2,3], [4,5,6]).perl.say 23:55
p6eval rakudo 1e9aa0: OUTPUT«GatherIterator.new()␤»
jnthn rakudo: crosswith(&infix:<+>, [1,2,3], [4,5,6]).eager.perl.say
p6eval rakudo 1e9aa0: OUTPUT«(5, 6, 7, 6, 7, 8, 7, 8, 9)␤»
colomon crosswith.assuming(&infix:<+>), something like that?
jnthn rakudo: my $opsub = &crosswith.assuming(&infix:<+>); $opsub([1,2,3], [4,5,6]).eager.perl.say
p6eval rakudo 1e9aa0: OUTPUT«(5, 6, 7, 6, 7, 8, 7, 8, 9)␤»
jnthn Like that. 23:56
colomon \o/
jnthn
.oO( and now everyone knows why I'm just drinking beer and not writing code tonight :-))
TimToady well, but hypers are zipwith, not crosswith
colomon the next question is, how do we stick that into the symbol table?
colomon gets confused because he knows how to do half of this in PIR and half in Perl 6.... 23:57
jnthn Well...where do they live?
Are they lexical, or package-y, or?
colomon TimToady: hypers had better not be zipwith or crosswith, right?
TimToady I only know how to do it in STD (but haven't yet)
eager zipwith is a strict hyper
kinda sorta
colomon emphasis on the kinda sorta. 23:58
:)
jnthn :-)
TimToady zipwith guarantees order of execution, while a hyper doesn't
colomon because the rules for what to do with different length lists are very different.
jnthn TimToady: Any thoughts on where we install generated meta-op subs?
TimToady how so?
colomon TimToady: on the lengths? 23:59
jnthn Or on the installig?
.oO( this is what happens when you have hyper conversations )
TimToady well, for a strict hyper, you did if the lengths aren't the same, I guess
*die
jnthn ...not that I'm saying one of us should zip it. :-)

Powered by Cro and the Raku® Programming Language.

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

Thank you!