»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend!
Set by sorear on 4 February 2011.
grondilu Compiling lib/Panda/Ecosystem.pm 00:56
===SORRY!===
When pre-compiling a module, its dependencies must be pre-compiled first.
Please pre-compile /usr/local/src/rakudo/install/lib/parrot/4.10.0-devel/languages/perl6/site/lib/JSON/Tiny/Actions.pm
I'll do it but I thought it was worth mentionning
grondilu I suspect this was the module that prevented reboostrat to work earlier 00:59
grondilu failed installing panda: 01:05
=> Testing panda
t/panda/builder.t .... ok
t/panda/common.t ..... ok
t/panda/ecosystem.t .. ok
t/panda/fetcher.t .... ok
t/panda/installer.t .. Failed 1/8 subtests
t/panda/installer.t (Wstat: 0 Tests: 8 Failed: 1) Failed test: 3 01:06
Files=7, Tests=49, 38 wallclock secs ( 0.17 usr 0.03 sys + 33.40 cusr 1.72 csys = 35.32 CPU)
Result: FAIL
flussence test #3 is file_exists_ok "$destdir/lib/foo.pir", 'pir installed'; 01:09
flussence which means precompilation is failing 01:09
grondilu wipes out the panda directory and tries to reinstall from scratch 01:19
grondilu failed again and gives up. 01:42
skids_ has a laugh at munificent/vigil 04:29
FROGGS morning 09:07
tadzik hello #perl6 09:13
FROGGS hi tadzik 09:14
jnthn morning o/ 09:22
nwc10 wonders what wonders jnthn has in store for us today 09:26
nwc10 is it heresy to use a Makefile for compiling Java? :-) 09:26
(I'm reading the code but it isn't something I can add to) 09:27
tadzik nwc10: if it is, them I'm quilty too ;) 09:29
I refused to ever learned eclipse and every Java project I wrote was in vim and built with Makefiles
bah, I can't english 09:30
jnthn nwc10: I suspect we'll end up with a Makefile fwiw :) 09:33
(And if you were asking for permission to create one, go for it. :)) 09:34
arnsholt tadzik: Ant isn't too terrible actually, even though it's XML 09:35
kresike hello all you happy perl6 people 09:35
timotimo maybe a makefile-to-ant compiler could be written in perl6.
tadzik real-world observations seem to indicate that writing build systems makes you hair go curly 09:40
timotimo i'd like my hair to no longer be flat :| i can't wear it short comfortably 09:42
tadzik but you can tie it when it's long 09:44
it's a terrible mess if they're not straight 09:45
timotimo oh, that's true.
moritz has some experience with that :-)
FROGGS haha 10:45
ph1ur3 sup 10:47
FROGGS $work :/ 10:52
tadzik nonono. It's spelled "hooray, $work!" 10:53
FROGGS tadzik: no, it's not ;o)
tadzik fights with Perl 5 10:54
arnsholt I just fixed a bit of Perl code I wrote. It used a 5.14 feature, but the people who were going to use the had perl < 5.14. Oops =) 10:55
Easily fixed though
tadzik #moose helped me fix scsys.co.uk:8002/222434 but no one know why it works that way :P
pmurias jnthn: how should I put __6MODEL_CORE__? there seems to be 2 objects there but I find references to a STable at index 2 11:59
jnthn: maybe that should be described in the serialization format docs? 12:00
jnthn pmurias: Did you miss github.com/perl6/nqp/blob/master/s...per.c#L331 maybe? 12:12
pmurias: Yes, would probably be good to document that. Please create an nqp ticket for it.
tadzik look what I found in my feather $HOME :) gist.github.com/4483793 13:23
I have to get back to that idea 13:24
tadzik then emmentaler could easily say if (and why not) all the Star modules are functional 13:25
[Coke] yawns
timotimo emmentaler runs all the tests from the ecosystem regularly? 13:38
tadzik er, it's supposed to :P 13:40
timotimo what stops you from deploying it? would be good to have that data visible on modules.perl6.org, no? 13:40
tadzik there seems to be a misunderstanding between cron and sergot++ :)
feather.perl6.nl/~sergot/modules/ says (last update 2012-12-13T05:40:02-0100). 13:41
the coloured circles on the right show the smoketest status
timotimo i think it would be nice to get the output from panda there, too. but what do i know? :P 13:43
tadzik well volunteered! :) 13:44
timotimo er, let me see what i can do
tadzik it's all on github.com/sergot/modules.perl6.org, I think
it doesn't get the full output, but it gets an X::Panda object for the failed builds 13:45
I started working on TAP harness some time ago, to make more info available, but I'm not full of tuits recently to deploy it
timotimo %log{$p} = gen-result($! // 'chocolate cake'); - chocolate cake? :) 13:46
tadzik er, I wrote that? :D
. o O ( why am I surprised ) 13:47
yeah, I see it :) 13:48
yeah, it's a "not an X::Panda value" :)
dalek kudo-js: 4cf209c | (Paweł Murias)++ | / (8 files):
split the setting into parts stored in src/core
 16:12
kudo-js: 1d1e0a7 | (Paweł Murias)++ | / (3 files):
Pass test 53.
jnthn pmurias++ 16:34
timotimo i recall having asked this before, but how do i get the stdout of a shell command? (and can i also get the return code, as well?) 17:32
FROGGS[mobile] I believe there is a module for that 17:35
look modules.perl6.org and search for shell
*at* 17:36
colomon that's builtin to p6
rn: say qx/ls/
p6eval rakudo ffde43: OUTPUT«qx, qqx is disallowed in restricted setting␤ in sub restricted at src/RESTRICTED.setting:2␤ in sub QX at src/RESTRICTED.setting:9␤ in block at /tmp/9svWPqefEV:1␤␤»
..niecza v24-16-g89e47b6: OUTPUT«Unhandled exception: rungather may not be used in safe mode␤ at /home/p6eval/niecza/lib/CORE.setting line 0 (rungather @ 0) ␤ at /tmp/6WBdgcQQxx line 1 (mainline @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 4218 (ANON @ 3) ␤ at /home/p6eval/niec…
colomon niecza> say qx/ls/
4th.pl
Bag.pl 17:37
etc
same works in rakudo
FROGGS[mobile] colomon: what about stderr,l and returncode?
timotimo huh. i did q:x{echo "foo bar"} and the result was just 0
oh, haha, i re-used $foo and put my in front, that can't work of course 17:38
colomon FROGGS[mobile]: it's not supposed to capture stderr, but I believe it does in rakudo at least.
FROGGS[mobile] colomon wrote qx, not q:x... is that a difference?
timotimo doesn't seem so? 17:39
timotimo i'd like to teach sergots modules.perl6.org website to display the output of the test or build process when a module didn't make it through installation and testing 17:40
but i'd need to either parse "prove"s output to see if the result was success or failure or find a way to both capture stdout and get the return value. 17:41
FROGGS thats better than phone 17:42
timotimo but then who was phone?
FROGGS ?
I mean typing on the pc is better than on my android phone
timotimo ;) 17:43
it was an obscure meme reference 17:44
FROGGS well, post the image link next time ;o)
jnthn qx and q:x are the same, fwiw 17:46
q followed by any quote modifier is the same as q :the_mod
Thus qw and q:w are the same too, etc.
FROGGS ahh, Shell::Command is not what we need here: github.com/tadzik/perl6-File-Tools...Command.pm 17:47
jnthn: cool, thanks
IMO it makes sense that qx{} returns a parcel that contains STDOUT, STDERR, and the retcode, and stringifies to STDOUT only 17:48
but I guess thats not the way it is specced :/
timotimo is that how it happens or is that how you'd like it the most? 17:49
FROGGS thats how I would like it
since I need perl 5's Capture::Tiny and friends 17:50
timotimo sounds sensible, albeit a little bit non-obvious
jnthn Or we could return an object ;) 17:51
FROGGS that follows the rule "keep simple things simple"... I like that rule...
or that
lunch &
timotimo so, there's not really a way to do it thus far, except for adding > /tmp/foobar.txt into the shell command and reading that? 17:52
FROGGS 1>STDOUT.txt 2>STDERR.txt 18:26
timotimo: but I have no idea how to get the return code
colomon use shell instead of qx 18:29
rn: my $result = shell "ls 1>STDOUT.txt 2>STDERR.txt" 18:30
p6eval rakudo ffde43: OUTPUT«shell is disallowed in restricted setting␤ in sub restricted at src/RESTRICTED.setting:2␤ in sub shell at src/RESTRICTED.setting:7␤ in block at /tmp/bae9brDMqh:1␤␤»
..niecza v24-16-g89e47b6: OUTPUT«Potential difficulties:␤ $result is declared but not used at /tmp/hACx1Eqska line 1:␤------> my ⏏$result = shell "ls 1>STDOUT.txt 2>STDER␤␤Unhandled exception: shell may not be used in safe mode␤ at /home/p6eval/niecza/lib/CORE.setting…
colomon does not know if shell fully works under both niecza and rakudo, but it does appear to at least shell out the command on both systems. 18:35
masak evening, #perl6 18:36
diakopter o/
flussence huh, panda's working normally today... 18:37
dalek p: 6a9828f | rurban++ | tools/build/Makefile.in:
[GH #71] Add -L. -X. to parrot for newer parrots.

See github.com/parrot/parrot/issues/368		 18:53
p: c52df3f | rurban++ | tools/build/Makefile.in:
Merge branch 'dot-libpath-gh71'
FROGGS ya, shell gives you the return code 19:02
in which synopsis do I have to look for qx{} ? 19:06
I find it hard find the right one 19:07
found feather.perl6.nl/syn/S02.html#Adverbs_on_quotes 19:11
q:x q :exec Execute as command and return results
they dont say how it should be returned 19:12
FROGGS .oO( MUHAAHAAHAA )
flussence ...that's a good point. It could be binary. 19:13
FROGGS r: class A { method Str { "Uhh" } }; my $a = A.new; say $a; say ~$a;
p6eval rakudo ffde43: OUTPUT«A.new()␤Uhh␤»
FROGGS r: class A { method Str { "Uhh" } }; my $a = A.new; say "$a"; say ~$a; 19:14
p6eval rakudo ffde43: OUTPUT«Uhh␤Uhh␤»
FROGGS but it would break existing code if it would return an object 19:15
masak 'night, #perl6 21:09
dalek ast: 51d98f3 | moritz++ | S06-multi/proto.t:
RT #116164: proto signature was not checked on dispatch
 21:11
moritz lives_ok { $*PROGRAM_NAME = "coldfusion" }, '$*PROGRAM_NAME is assignable'; 21:12
I can kinda guess who wrote that test :-) 21:13
diakopter who? 21:16
moritz [Coke]++
dalek ast: e140846 | moritz++ | S02-magicals/progname.t:
RT #116164: $*PROGRAM_NAME reported wrong argument
 21:18
p-jvm-prep: 20ed33c | jnthn++ | / (7 files):
Provide some HLL config infrastructure.

This will allow creating correct box types and slurpy array/hash types in NQP, but using the BOOT types in NQP's MOP until the others are made available.
p-jvm-prep: a760822 | jnthn++ | t/qast_arguments.t:
Test for slurpy positional arguments.
p-jvm-prep: c56fc92 | jnthn++ | lib/QAST/JASTCompiler.nqp:
Code-gen for slurpy params.
p-jvm-prep: 7566191 | jnthn++ | src/org/perl6/nqp/runtime/GlobalContext.java:
Fix a GlobalContext build-up bug.
p-jvm-prep: 97b6a66 | jnthn++ | src/org/perl6/nqp/runtime/Ops.java:
Implement posslurpy op.

Positional slurpy params now work.
moritz r: say <<<\>'n'>>.perl 21:20
p6eval rakudo ffde43: OUTPUT«("<>", "n")␤»
diakopter remembers doing that one 21:21
moritz why is that two tokens?
moritz expected one
jnthn Shell words treats quotes specially 21:21
diakopter I don't remember which way the bug expects it to be :)
er, test
moritz jnthn: ah, that explains it
jnthn So you can do like 21:22
r: << 'foo bar' baz >>.perl.say
p6eval rakudo ffde43: OUTPUT«("foo bar", "baz")␤»
moritz r: << 'foo bar'baz >>.perl.say
p6eval rakudo ffde43: OUTPUT«("foo bar", "baz")␤»
jnthn r: << 'foo bar''baz' >>.perl.say 21:23
p6eval rakudo ffde43: OUTPUT«("foo bar", "baz")␤»
jnthn It's behaving the way I expect, but my expectations may be off... :)
dalek ast: aa20719 | moritz++ | S02-literals/quoting.t:
RT #115272: texas quote edge case
 21:24
moritz I'm testing it as-is
if somebody expects differently, ey can open a new bug report :-)
diakopter I remember rakudo did it differently from niecza
jnthn It's hard to tell either way from STD, I'm not aware of the spec saying anything about this case, and I'm pretty sure Niecza doesn't handle the << ... >> form yet really.
moritz r: BEGIN { try 0 }; say "alive" 21:26
p6eval rakudo ffde43: OUTPUT«(timeout)»
jnthn That's...weird.
r: try 0
p6eval rakudo ffde43: ( no output )
moritz r: BEGIN { CATCH { } }; say "alive"
p6eval rakudo ffde43: OUTPUT«alive␤»
moritz even weirder :-)
jnthn r: eval 'try 0' 21:27
p6eval rakudo ffde43: ( no output )
moritz that's rt.perl.org/rt3/Ticket/Display.html?id=115502 fwiw
jnthn r: BEGIN try 0
moritz no luck with -ll-exception, fwiw
p6eval rakudo ffde43: OUTPUT«(timeout)»
jnthn wonders where on earth it hangs...
r: BEGIN { say 1; try 0; say 2; }; say "alive" 21:28
p6eval rakudo ffde43: OUTPUT«(timeout)1␤»
moritz so it hangs at run time
jnthn r: BEGIN { say 1; try say 1.5; say 2; }; say "alive"
p6eval rakudo ffde43: OUTPUT«(timeout)1␤1.5␤» 21:29
moritz r: BEGIN { if 0 { try 1 }; say 'alive' }
p6eval rakudo ffde43: OUTPUT«alive␤»
diakopter r: eval BEGIN { say 1; eval 'try 0'; say 2; }; say "alive"
p6eval rakudo ffde43: OUTPUT«1␤2␤Cannot call 'eval'; none of these signatures match:␤:(Str $code, :lang(:$lang) = { ... }, PseudoStash :context(:$context))␤␤ in block at /tmp/qPdC5Olm6z:1␤␤»
diakopter r: BEGIN { say 1; eval 'try 0'; say 2; }; say "alive"
p6eval rakudo ffde43: OUTPUT«1␤2␤alive␤»
jnthn Very weird
moritz r: BEGIN { try die 42; say "alive" } 21:30
p6eval rakudo ffde43: OUTPUT«(timeout)»
tadzik r: BEGIN { say 1; CATCH {} }; say 2
p6eval rakudo ffde43: OUTPUT«1␤2␤»
diakopter r: BEGIN { say 1; eval 'die'; say 2; }; say "alive"
p6eval rakudo ffde43: OUTPUT«1␤===SORRY!===␤␤»
diakopter haha 21:31
colomon how do I say "not at the beginning of a line" in a regex?
diakopter .
moritz colomon: <!before ^>
erm 21:32
jnthn ^^
?
moritz <!before ^^>
colomon this is on the output of lines, so ^ and ^^ should be the same, right?
moritz++ jnthn++
moritz if the string is one line, then it's the same 21:33
colomon right
colomon is trying to write a script to translate a STEP file into an HTML file for ease of browsing. 21:34
diakopter r: BEGIN { say 1; eval 'die'; say 2; CATCH { say 3 } }; say "alive" 21:42
p6eval rakudo ffde43: OUTPUT«1␤===SORRY!===␤Cannot assign to a non-container␤»
diakopter <.<. 21:43
r: BEGIN { say 1; eval 'die'; say 2 }; say "alive"
p6eval rakudo ffde43: OUTPUT«1␤===SORRY!===␤␤»
colomon moritz, jnthn: so if I wanted to specify "is not followed by an equals sign" that would be something like <!after "=">? 21:44
flussence I think I've found another precomp bug, but golfing it down is really hard so I'll just throw this out there while I figure that out... 21:45
so here's a Buf: github.com/tadzik/Bailador/blob/ma...tent.t#L11
when there's pir files involved, at some point that Buf forgets it has a .decode method and I end up with a runtime error.
tadzik oh 21:47
that's too bad 21:48
colomon actually it's <!before "="> there too, correct? # off in his own world... 21:50
jnthn colomon: If it should not be the next thing, then !before, yes 21:51
dalek p-jvm-prep: 75246ba | jnthn++ | src/org/perl6/nqp/runtime/ (2 files):
Refactor named param ops in prep for slurpy.
p-jvm-prep: 84b1d69 | jnthn++ | t/qast_arguments.t:
Add a test for slurpy named params.
p-jvm-prep: 331bb37 | jnthn++ | src/org/perl6/nqp/runtime/Ops.java:
Implement slurpy named params.
colomon jnthn: danke
flussence wait, I think I screwed up and this error's somewhere else entirely... 22:06
request.env seems to be undefined here using precomp - github.com/tadzik/Bailador/blob/ma...ntent.t#L7 22:07
tadzik hmmm 22:13
I've no idea
dalek p: cd827db | jnthn++ | src/QAST/Operations.nqp:
Add nqp:: mappings for some 6model operations.
 22:57
p: 591e59b | jnthn++ | src/stage0/ (9 files):
Update bootstrap to get new nqp:: ops.
p: 808a0d9 | jnthn++ | src/ (3 files):
Some more pir:: => nqp:: in the MOP.
jnthn 'night, #perl6 23:13
diakopter o/

Powered by Cro and the Raku® Programming Language.

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

Thank you!