»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg camelia perl6: ... | irclog: irc.perl6.org | UTF-8 is our friend!
Set by sorear on 25 June 2013.
00:00 lizmat joined 00:02 adu joined 00:07 gfldex joined 00:10 exixt_ is now known as exixt, colomon left 00:11 cognominal joined
timotimo hrmpf. lots of fiddling with no good results :\ 00:23
00:34 thou left 00:36 dayangkun joined, huf joined 00:53 colomon joined 01:01 huf left 01:02 sftp_ joined 01:04 sftp left 01:05 sftp_ is now known as sftp 01:07 araujo joined 01:09 raiph joined 01:12 grondilu left 01:16 gfldex left 01:20 xenoterracide_ joined, araujo left 01:26 xenoterracide_ left 01:27 xenoterracide joined 01:29 vendethiel left 01:30 lizmat left, lizmat joined 01:36 vendethiel joined 01:37 mirjam joined 01:41 Guest48053 left 01:46 raiph left 01:47 huf joined 01:56 thou joined 01:58 vendethiel left 02:00 kaleem joined, thou left 02:02 mirjam left
raydiak am I the only one who finds it odd that we numify w/.Numeric but stringify w/.Str, not .Stringy? 02:07
colomon we just don't have much that's Stringy but not Str at the moment. 02:08
while we have lots that's Numeric but not Num
raydiak I figured it was something like that, it just seems asymettrical in a seemingly needless way 02:09
I noticed the docs for Stringy mentioned a Cat class which doesn't exist...wonder what that used to be 02:11
colomon it's not been yet
Cat is a lazy string 02:12
raydiak ah, interesting
colomon so (for instance) it might make more sense for slurp to return a Cat than a Str 02:13
but I think right now everyone is more interested in getting what we already have working well enough, if you know what I mean. 02:14
raydiak yep...glad we don't *just* run around in circles of reinvention :)
colomon ;) 02:15
I'm lazily interested in Cat because I've had to sometimes deal with 1 GB text files for work, and it seems like Cat combined with a parse that threw away the "string so far" after ratcheting would be ideal for that. 02:17
but right now Rakudo is still so slow that 1 GB files are kind of out of the question even with extremely clever handling, so I'm not worrying about it. 02:18
raydiak that would be really slick and seamless though, compared to handling small parsable chunks at a time 02:21
in most cases, laziness is transparent/automatic...why would we need a Cat class?
02:26 vendethiel joined
colomon to implement the transparent/automatic stuff 02:27
the idea, of course, is that if you're using a Stringy interface, what you write works with Str or Cat. 02:28
raydiak I mean e.g. Lists can be lazy, but there's not like a List and a LazyList, it's just some private attrs and so forth that control whether/how much an instance is lazy 02:29
(if I understand how it all works correctly)
02:29 xinming left
colomon huh 02:30
now I'm halfway to convincing myself that's a flaw in List. ;)
raydiak hah :)
02:30 xinming joined
colomon well, really that we could use a list-ish role that allows us to cleanly consume list-y things 02:31
might be that Iterable is that already.
I dunno.
anyway, no one has really explored Cat yet 02:32
and Stringy just sort of hangs around based on the notion that it might be a good abstraction someday.
raydiak yeah I didn't really have a point, I just often learn new stuff with these meandering trains of thought 02:33
02:36 raiph joined
colomon nothing wrong with that in the least. :) 02:41
raydiak got any ideas or suggestions for math::symbolic other than the obviously missing peices? 02:46
*pieces
dalek ast: 9888da3 | (Rob Hoelz)++ | S10-packages/precompilation.t:
Unfudge test for 122773 on MoarVM

It's passing now
02:49 xenoterracide left 02:50 xenoterracide joined
colomon raydiak: does it work with panda yet? :) 02:56
raydiak colomon: um...yes? unless you know something I don't...
colomon nope. :(
resolve stage failed for Math::Symbolic: Project Math::Symbolic not found in the ecosystem
raydiak do a 'panda update' recently? 02:57
colomon 30 seconds ago
raydiak weird it just worked for me earlier today
colomon can you install it from panda?
raydiak yep, just confirmed it still wfm
colomon :\ 02:58
raydiak other modules install fine still? 02:59
colomon yes
wonder if panda update is somehow broken on my machine? 03:00
raydiak huh...other than running 'update', I'm kinda clueless right now
oh look the github urls are broken 03:01
raw2.github.com/raydiak/Math-Symbo.../META.info <-- 404
03:02 kaleem left
raydiak must not be affecting me b/c supposedly I ran update before they broke 03:02
we've had stuff break b/c of github changing its urls or even their html before
colomon can you run panda info Math::Symbolic and post the results?
errr… maybe you need to do a panda update and see if you still have Math::Symbolic ? 03:03
raydiak Math::Symbolic
Symbolic math for Perl 6
colomon raw.githubusercontent.com/raydiak/.../META.info works
raydiak State: installed
colomon for me
raydiak Source-url: git://github.com/raydiak/Math-Symbolic.git
Author: raydiak
yes, but all the urls in the ecosystem use raw2.github.com 03:04
03:05 atta left, atta joined
colomon just did panda install git://github.com/raydiak/Math-Symbolic.git and that seems to be working. :) 03:05
raydiak every single line in github.com/perl6/ecosystem/blob/ma.../META.list is currently a 404 :P 03:06
03:06 grettis left 03:12 Mso150 joined 03:14 chenryn joined
raydiak remember that day we went through this b/c panda wasn't handling redirects or something? github.com/perl6/ecosystem/commit/...2650d17810 03:22
raydiak goes to s:g/// it again 03:23
dalek osystem: 7ff21ca | raydiak++ | META.list:
Update github URLs (again)
 03:25
raydiak there, now if I didn't break the whole world, I fixed it :)
colomon :) 03:34
03:44 thou joined 03:46 raiph left 03:49 thou left 03:54 Mso150 left 04:01 rmgk is now known as Guest15015, rmgk_ joined, Guest15015 left, rmgk_ is now known as rmgk
dalek href="https://modules.perl6.org:">modules.perl6.org: 8d8a68c | raydiak++ | web/build-project-list.pl:
Update github URL (again)
 04:05
04:20 BenGoldberg left 04:31 dayangkun left 04:41 slavik joined 04:44 tardisx` is now known as TardisX 05:10 adu left 05:13 chenryn left 05:16 xenoterracide left 05:32 thou joined 05:35 chenryn joined 05:37 thou left 05:40 kaare_ joined 05:42 tinyblak left, tinyblak joined 05:43 tinyblak_ joined 05:47 tinyblak left
dalek osystem: 0b7644a | ugexe++ | META.list:
Update META.list

Text::Table::Simple		 06:10
06:24 kaleem joined 06:30 virtualsue joined 06:40 hagiri joined
raydiak ugexe++ 06:41
06:47 hagiri left 06:54 aborazmeh joined 07:10 bowtie joined, bowtie is now known as Guest25839 07:14 Guest25839 left 07:15 jfredett left 07:18 FROGGS_ left 07:20 thou joined 07:22 aborazmeh left 07:24 chenryn left 07:25 thou left 07:26 gfldex joined 07:29 chenryn joined, bowtie_ joined 07:30 DelmoreTheDonkey joined
DelmoreTheDonkey r: any(11,7,5)-all(1,3) 07:30
camelia ( no output )
DelmoreTheDonkey r: say any(11,7,5)-all(1,3)
camelia rakudo-{parrot,moar} 315ec6: OUTPUT«all(any(10, 6, 4), any(8, 4, 2))␤»
DelmoreTheDonkey why is the output "all(any(10, 6, 4), any(8, 4, 2))" 07:31
I was expecting "any(all(10, 6, 4), all(8, 4, 2))"
07:34 bowtie_ left
moritz DelmoreTheDonkey: not saying that it's wrong, but why did you expect that? 07:39
07:39 ptc_p6 joined 07:48 FROGGS joined 07:49 brrt joined
DelmoreTheDonkey If I take everything in B away from anything in A should I expect to have everything(all) or anything(any)? 07:54
07:54 rindolf joined 07:56 pecastro left
moritz DelmoreTheDonkey: you're arguing as if junctions were sets 07:57
DelmoreTheDonkey: they aren't
junctions only make sense as matchers
DelmoreTheDonkey ok.
moritz all(1, 3) means "a matcher that's true if the number I compare it to is both 1 and 3"
DelmoreTheDonkey what current uses do we have for junctions of junctions? 07:58
moritz for example if $filename ~~ any(none('README', 'LICENSE'), none(/^\./)) 07:59
imagine you have several files that you want to filter out by junctions
and one matcher none('README', 'LICENSE'), and one matcher none(/^\./) for dotfiles
and then you want to test if a file is exluded by either of those 08:00
DelmoreTheDonkey thanks 08:01
r: $filename = "README" ; if $filename ~~ any(none('README', 'LICENSE'), none(/^\./)) { say "YES" } 08:04
camelia rakudo-{parrot,moar} 315ec6: OUTPUT«===SORRY!=== Error while compiling /tmp/tmpfile␤Variable '$filename' is not declared␤at /tmp/tmpfile:1␤------> $filename⏏ = "README" ; if $filename ~~ any(none␤ expecting any of:␤ postfi…»
08:04 bowtie joined
DelmoreTheDonkey r: my $filename = "README" ; if $filename ~~ any(none('README', 'LICENSE'), none(/^\./)) { say "YES" } 08:04
camelia rakudo-{parrot,moar} 315ec6: OUTPUT«YES␤» 08:05
08:05 bowtie is now known as Guest41269
DelmoreTheDonkey r: my $filename = "COPYRIGHT" ; if $filename ~~ any(none('README', 'LICENSE'), none(/^\./)) { say "YES" } 08:05
camelia rakudo-{parrot,moar} 315ec6: OUTPUT«YES␤»
FROGGS m: say "README" ~~ any(none('README', 'LICENSE'), none(/^\./)) 08:06
camelia rakudo-moar 315ec6: OUTPUT«True␤»
FROGGS m: say ".git" ~~ any(none('README', 'LICENSE'), none(/^\./))
camelia rakudo-moar 315ec6: OUTPUT«True␤»
FROGGS well, that's always true...
moritz m: say 'something else' ~~ any(none('README', 'LICENSE'), none(/^\./))
camelia rakudo-moar 315ec6: OUTPUT«True␤»
FROGGS m: say (".git" ~~ any(none('README', 'LICENSE'), none(/^\./))).perl
camelia rakudo-moar 315ec6: OUTPUT«Bool::True␤»
moritz huh 08:07
oh
raydiak it's right
moritz it must be all()
FROGGS any of them is always true :o)
m: say (".git" ~~ all(none('README', 'LICENSE'), none(/^\./))).perl
camelia rakudo-moar 315ec6: OUTPUT«Bool::False␤»
moritz m: say 'something else' ~~ all(none('README', 'LICENSE'), none(/^\./))
camelia rakudo-moar 315ec6: OUTPUT«True␤»
moritz m: say '.git' ~~ all(none('README', 'LICENSE'), none(/^\./))
camelia rakudo-moar 315ec6: OUTPUT«False␤»
moritz so the all() check tells if you something isn't ignored
and this, ladies and gentlemen, is why it's probably best not to use nested junctions :-) 08:08
FROGGS moritz: well, it makes sense in case the none() parts are in variables 08:09
08:09 Guest41269 left
moritz possibly 08:09
FROGGS so you keep an array of junctions that specify what should be ignored, and push to that array depending on the platform you are running on or so
moritz actually
the junctions specify what to *not* ignore in my example 08:10
which is why we need all(), not any()
FROGGS or that
moritz if specify what to ignore, you use any() junctions, and then again any() junctions to combine them 08:11
which is a lot more intuitive, IMHO
and less spectacular from a "nested junctions" POV, because any(a, any(b, c)) is equivalent to any(a, b, c)
08:15 bowtie_ joined 08:16 rindolf left 08:18 tinyblak_ left, tinyblak joined 08:20 darutoko joined 08:28 daxim left 08:30 rindolf joined 08:32 gfldex left 08:34 bowtie_ left 08:40 telex left, bowtie_ joined, telex joined 08:44 molaf__ joined 08:48 molaf_ left, sqirrel__ joined, bowtie_ left 08:57 Mso150 joined 09:00 tinyblak_ joined 09:01 skarn left 09:03 bowtie joined, bowtie is now known as Guest41562 09:04 tinyblak left 09:09 ghostlines joined 09:13 kjs_ joined 09:23 pdcawley left 09:24 pdcawley joined 09:25 jack_rabbit left 09:26 jack_rabbit joined, tinyblak_ left 09:27 tinyblak joined 09:29 skarn joined 09:33 chenryn left 09:40 grondilu joined
grondilu hello, if someone wants to help writing a C parser: github.com/grondilu/C 09:40
09:42 jack_rabbit left, kjs_ left 09:43 jack_rabbit joined
FROGGS grondilu: I've seen a (huge) C grammar written in Perl 6 already 09:43
grondilu ok 09:44
where is it?
FROGGS I am already searching for it
09:45 sqirrel__ left 09:46 dakkar joined
FROGGS grondilu: irclog.perlgeek.de/perl6/2013-03-29#i_6643290 09:48
grondilu ok thanks 09:50
09:58 pecastro joined
brrt grondilu: this because of the c4 thingy? 09:59
09:59 fhelmberger joined 10:00 bjz_ joined, bjz left
FROGGS c4? 10:00
10:01 vike left
brrt github.com/rswier/c4/blob/master/c4.c 10:03
10:04 vike joined
FROGGS interesting 10:04
10:07 MalliParitala joined
MalliParitala Hello 10:08
I want to help Perl6 10:09
FROGGS hi MalliParitala
MalliParitala Hello FROGGS 10:10
peteretep MalliParitala: Great! Just send over your CC details, and we'll get started
10:10 thou joined
brrt hi MalliParitala, what do you want to do? there is a lot to do :-) 10:10
one of the best ways to help perl6 is to start using it, and report to us any issues that you find 10:11
MalliParitala I know Perl5
i can code in Perl5
Ok 10:12
10:12 jack_rabbit left
FROGGS MalliParitala: you could for example port a module from Perl 5 to Perl 6, if you want to... 10:12
MalliParitala Okay.
Do we have list of modules need to be ported to Perl6? 10:13
10:15 thou left
brrt well, anything you like, really 10:15
we have modules.perl6.org
modules.perl6.org
if you find that any perl5 module you frequently use is not represented there, you can try and port it
MalliParitala Cool. 10:16
brrt of course, we already have a list like that: github.com/perl6/perl6-most-wanted...modules.md
MalliParitala Thanks brrt
10:16 chenryn joined
brrt you're welcome :-) i hope you can find something to your liking 10:17
MalliParitala Do i need to provide my mail ID so that i can get updates about perl6 work?
10:17 kjs_ joined
brrt no, you can subscribe to one of the mailing lists 10:17
MalliParitala Okay.
brrt perl6.org/community/ this page gives you all the information you might want 10:18
or perhaps, are likely to want
MalliParitala I will start using Perl6 first. Thanks much.
10:18 kjs_ left
brrt no problem. please keep us updated with anything you might encounter 10:18
10:22 exixt is now known as exixt_, vike left 10:23 daxim joined 10:29 kjs_ joined 10:30 vike joined
grondilu brrt: not really. More that I'd like a C parser to automate X11 bindings 10:31
brrt i see 10:32
:-)
10:33 jack_rabbit joined
dalek ecs: edb565d | (Elizabeth Mattijsen)++ | S32-setting-library/IO.pod:
Revert "Spec dir() default exclusion to all /^\./"

It appears the consensus is to just ignore . and ..
This reverts commit 91c184fbb816b642b9ac724db65352e6cfd2eb3a.		 10:41
kudo/nom: 1ad58fc | (Elizabeth Mattijsen)++ | src/core/IO.pm:
Revert "dir primitives exclude anything starting with ".""

It appears the consensus is to just ignore '.' and '..' by default
This reverts commit f7c7b06a98fdd89ab8686ed52d4d9746fb0b0189.		 10:42
grondilu FROGGS: added a comment on your fork: gist.github.com/FROGGS/5268178#comment-1331609
10:43 chenryn left
grondilu meh nevermind. I removed it. I suspect it was dumb. 10:43
10:44 araujo joined, araujo left, araujo joined 10:48 raiph joined
grondilu still, a line like 'token integer-constant:radix(8) { <octal-constant> <integer-suffix>* }' does not compile for me. Gets 'Colon pair value '8' too complex to use in name' 10:55
not sure about this :radix() thing means. 10:56
std: grammar { token integer-constant:radix(8) { <octal-constant> <integer-suffix>* }
camelia std 14ad63b: OUTPUT«===SORRY!===␤Unable to parse block at /tmp/bX0PtG7jJe line 1:␤------> grammar ⏏{ token integer-constant:radix(8) { <oct␤Couldn't find final '}'; gave up at /tmp/bX0PtG7jJe line 1 (EOF):␤------> { <octal-constan…»
grondilu std: grammar { token integer-constant:radix(8) { .* } }
camelia std 14ad63b: OUTPUT«ok 00:00 141m␤»
10:56 jack_rabbit left
FROGGS m: grammar { token integer-constant:radix<8> { .* } } 11:00
camelia ( no output )
FROGGS you can use that, it is the same 11:01
m: grammar { token integer-constant:radix('8') { .* } }
camelia rakudo-moar 315ec6: OUTPUT«===SORRY!=== Error while compiling /tmp/io14Bn_gyv␤Colon pair value ''8'' too complex to use in name␤at /tmp/io14Bn_gyv:1␤------> mmar { token integer-constant:radix('8')⏏ { .* } }␤»
FROGGS m: grammar { token integer-constant:radix['8'] { .* } }
camelia rakudo-moar 315ec6: OUTPUT«===SORRY!=== Error while compiling /tmp/qzLW5hSsJb␤Colon pair value ''8'' too complex to use in name␤at /tmp/qzLW5hSsJb:1␤------> mmar { token integer-constant:radix['8']⏏ { .* } }␤»
FROGGS m: grammar { token integer-constant:radix[8] { .* } }
camelia rakudo-moar 315ec6: OUTPUT«===SORRY!=== Error while compiling /tmp/XbYIEfq_XH␤Colon pair value '8' too complex to use in name␤at /tmp/XbYIEfq_XH:1␤------> rammar { token integer-constant:radix[8]⏏ { .* } }␤»
FROGGS hmmm
dalek ecs: 1d0834c | (Elizabeth Mattijsen)++ | S16-io.pod:
Remove :test parameter from chdir/tmpdir/homedir
11:04 virtualsue left, Mso150 left 11:05 Mso150 joined, esaym153 left 11:06 chenryn joined, MalliParitala left 11:09 Grrrr left 11:12 esaym153 joined 11:13 pdcawley left, Grrrr joined
dalek p: 92f6783 | (Tobias Leich)++ | tools/build/MOAR_REVISION:
bump moar rev (#122896)
 11:18
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...?id=122896
dalek kudo/nom: d2e301e | (Tobias Leich)++ | tools/build/NQP_REVISION:
bump nqp/moar rev (#122896)
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...?id=122896
11:22 psch joined
psch hi #perl6 o/ 11:22
lizmat pulls, builds and starts a spectest
psch \o 11:23
psch FROGGS: there's a ticket for the colon pair thingy you fiddled with in the clog, but RT is really slow for me atm
lizmat o/
FROGGS psch: I remember solving a bug in that area...
psch FROGGS: 50c765 has you putting in the exception 11:25
which was about 2 months ago 11:26
FROGGS yeah
psch i *think* the ticket was noticeably older, let me see if i can convince my phone to let me search on RT... :)
11:28 sqirrel__ joined 11:30 grondilu left
dalek ast: 11575cc | (Tobias Leich)++ | packages/Test/Util.pm:
prefer $*EXECUTABLE over $*EXECUTABLE_NAME

  ...since $*EXECUTABLE gives us an absolute path to the perl6 binary, rather than just "perl6".		 11:30
11:31 denis_boyun_ joined 11:33 Mso150 left 11:35 brrt left
psch m: (:a['b']).WHAT.say 11:39
camelia rakudo-moar 315ec6: OUTPUT«(Pair)␤»
psch m: (:a['b']).say
camelia rakudo-moar 315ec6: OUTPUT«"a" => ["b"]␤»
psch m: (:a('b')).say
camelia rakudo-moar 315ec6: OUTPUT«"a" => "b"␤»
psch those semantics aren't quite right, are they?
specifically, the first one shouldn't have an array as value 11:40
oh, it should 11:41
S02:3652
synopsebot Link: perlcabal.org/syn/S02.html#line_3652
psch (assuming lines in vim translate correctly...)
nope, they don't 11:42
11:43 kaleem left 11:47 bcode left 11:48 psch left 11:49 azawawi joined
azawawi hi #perl6 11:49
im getting "resolve stage failed for $module-name" from panda installs. Is there a problem with latest panda?
11:50 psch joined, bcode joined, virtualsue joined
azawawi panda install Farabi6 11:50
resolve stage failed for Farabi6: Project Farabi6 not found in the ecosystem
latest github panda, and rebootstrapped... 11:52
colomon according to my smoke test run, there's nothing in the ecosystem at all. though that was as of 5 hours ago.
lizmat FROGGS: looks like the spectest is significantly slower after this update
before: Files=930, Tests=33836, 172 wallclock secs ( 8.98 usr 3.36 sys + 1041.29 cusr 163.08 csys = 1216.71 CPU)
azawawi i guess it was the latest rakudo when i pulled it
lizmat after: Files=930, Tests=34048, 208 wallclock secs ( 8.74 usr 3.21 sys + 1276.04 cusr 184.99 csys = 1472.98 CPU) 11:53
azawawi rebuilds rakudo
the first thing i dont when i log on to play with perl6 is a 'cd ~/rakudo && git pull && (git log | less)' :) 11:54
lizmat looks like bare startup time almost *doubled*
before ~ .21 sec, now ~ .34 sec
and I'm still seeing flappy tests :-(
azawawi hmmm projects.json has only [] 11:56
lizmat (ok, more like 1.6x slower)
azawawi after a perl6 rebootstrap, projects.json has only '[]' 11:57
panda regression?
lizmat feels to me this version of moar is more like using more and giving less :-(
github URL change wtf ?
11:58 thou joined, tinyblak_ joined 11:59 denis_boyun_ left 12:01 denis_boyun_ joined 12:02 woolfy left, woolfy1 joined 12:03 thou left, tinyblak left 12:11 sqirrel__ left 12:15 darutoko- joined
azawawi has anyone tried latest panda with latest rakudo? 12:15
12:18 darutoko left
FROGGS lizmat: the consensus was to deprecate $*EXECUTABLE, so, they way forward might be to speedup $*EXECUTABLE if that is possible 12:23
lizmat: in what areas are the flapping tests?
12:23 chenryn left
lizmat azawawi: what are you seeing ? 12:27
fwiw, a rebootstrap worked for me just now 12:28
azawawi really?
sorry i was away a bit
lizmat as well
azawawi same problem as irclog.perlgeek.de/perl6/2014-11-05#i_9612344 12:29
lizmat needs to get some other things done before travelling to London for LPW this weekend
FROGGS problem is that feather.perl6.nl:3000/projects.json is empty (and empty json array) 12:30
tadzik: ping 12:32
tadzik: the service here needs to be adjusted: feather.perl6.nl:3000/projects.json you need to apply a patch similar to github.com/perl6/modules.perl6.org...8d8a68cd18 12:33
azawawi alright i caught one... FROGGS++
azawawi fishes for other bugs :)
lizmat the pond is still pretty full :-) 12:35
azawawi which is good, right? :)
12:36 kaleem joined
azawawi panda install rakudo # that would be fun :) 12:38
i was working on integrating perl6-debug-m with farabi6 to provide a gui 12:39
so far im near its completion... stepping in/out, ...etc 12:40
ironing some bugs here & there
lizmat FROGGS: flapping tests are all over the map 12:45
FROGGS: not sure what you mean with "the consensus was to deprecate $*EXECUTABLE, so, they way forward might be to speedup $*EXECUTABLE if that is possible" 12:46
FROGGS lizmat: I have not a single failing test
lizmat guesses you're not on OSX
FROGGS lizmat: we want to replace $*EXECUTABLE_NAME by $*EXECUTABLE 12:47
lizmat aha :-)
FROGGS sorry :o)
correct, I'm using ubuntu
lizmat on JVM and Moar, EXECUTABLE depends on $*VM, and that's the costly bit, I guess 12:48
EXECUTABLE creation is not particularly expensive 12:49
FROGGS tadzik: looks like your modules/updatelist.pl needs to run under PERL_LWP_SSL_VERIFY_HOSTNAME=0
12:52 woolfy1 left
azawawi was rakudo star 2014.10 released or not? 12:54
lizmat not that I'm aware of 12:55
azawawi it is strange that we deprecate something in 2014.10 and then we dont release it to the public :)
lizmat azawawi: afaik we only deprecate rakudo compiler versions... not rakudo star 13:01
13:08 woolfy joined
moritz azawawi: though if you want to make a star release for 2014.10, feel free! 13:11
13:12 tinyblak_ left, daxim left, tinyblak joined 13:15 guru joined 13:16 guru is now known as Guest40587, Guest40587 is now known as ajr_ 13:17 Guest41562 left 13:26 bowtie joined, bowtie is now known as Guest90992
azawawi moritz: :) 13:28
azawawi is waiting for his home AMD PC to be returned from the repair shop today :) 13:29
13:31 Guest90992 left
[Coke] m: say (10,6,4 X- 8,4,2).sort.uniq 13:35
camelia rakudo-moar 315ec6: OUTPUT«-4 -2 0 2 4 6 8␤»
13:42 lucas__ joined 13:44 raiph left
lucas__ hello #perl6 13:44
lizmat lucas__ o/ 13:45
PerlJam greetings lucas__
lucas__ thanks!
13:46 thou joined
dalek ecs: 4e09ab8 | (Elizabeth Mattijsen)++ | S16-io.pod:
Some more :test eradication
 13:47
lucas__ So, following the conventions of traditional Unix and Perl 5 manpages, would it be interesting to move AUTHORS sections to the very end of the file, instead of leaving them at the very beginning?
Having said that, I have to say I fully admire the hard work of the original authors. I hope this is not disrespectful. 13:48
But I think nowadays, the specs are community-property, rather than the work of sole-authors like it was "In The Beginning". 13:49
lizmat I guess the AUTHORS section also predates anything like "git blame" :-) 13:50
13:51 thou left, FROGGS left 13:53 kaleem left
lizmat or automatic author fetching from the git blobs 13:53
13:56 tinyblak_ joined, jfredett joined 13:59 azawawi left 14:00 tinyblak left, bowtie_ joined 14:01 raiph joined 14:04 bowtie_ left
[Coke] moar nojit - 10 failing tests. :( 14:09
colomon m: say FatRat.new(1, 10*1000).floor.perl 14:14
camelia rakudo-moar 315ec6: OUTPUT«0␤»
colomon m: say FatRat.new(10*1000,3).floor.perl
camelia rakudo-moar 315ec6: OUTPUT«3333␤»
14:15 daxim joined 14:16 bowtie_ joined 14:20 aborazmeh joined, bowtie_ left 14:21 notfix is now known as xfix
[Coke] moar jit failing a different 10 tests. no confidence it's fixed, though. 14:24
arnsholt r-m: my $a = 1; { say $a; my $a = 2; say $a; }; 14:25
camelia rakudo-moar 315ec6: OUTPUT«===SORRY!=== Error while compiling /tmp/X8vZFxRGjM␤Lexical symbol '$a' is already bound to an outer symbol;␤the implicit outer binding must be rewritten as OUTER::<$a>␤before you can unambiguously declare a new '$a' in this scope␤…»
arnsholt Aha. That's sensible 14:26
14:30 Guest95483 joined
psch m: 6; $a = 1; { say $a; my $a = 2; say $a } 14:36
camelia rakudo-moar 315ec6: OUTPUT«===SORRY!=== Error while compiling /tmp/CnCEdQ9PNl␤Lexical symbol '$a' is already bound to an outer symbol;␤the implicit outer binding must be rewritten as OUTER::<$a>␤before you can unambiguously declare a new '$a' in this scope␤…»
psch m: 6; $a = 1; say $a
camelia rakudo-moar 315ec6: OUTPUT«1␤»
14:36 muraiki left, Guest95483 left 14:41 JimmyZ joined
lucas__ Do I have an "OK" from someone for moving the AUTHORS section to the end? I think that would be an easy change. Then, I would submit a pull request. I think ownership discourages contributions. I'm just trying to clean up the specs a little bit. This change would unclutter the top of the documents for more useful information like introduction/overview/summary of the text. I won't change anything in the section, just move it to the end. Would such change be ac 14:41
14:43 bowtie_ joined
lucas__ And... what about my current spec pull request, any opinions yet? 14:43
moritz lucas__: truncated after "Would such change be" 14:44
lucas__: and yes, you have my +1
14:44 muraiki joined
lucas__ ... Would such change be accepted or be wasted effort? Should I just do it to see how things look? Anyway, it should be easy to hit the close button on my pull request, right? 14:44
I didn't knew this thing truncates... thanks. 14:45
moritz lucas__: re your pull request, IMHO the feedback infomration belongs to the index page, perlcabal.org/syn/
lucas__: that page is currentl in the perl6/mu repo, and IMHO should go into the perl6/specs repo, but that requires updating cron-jobs too 14:47
14:47 bowtie_ left
dalek ecs: 487e7d4 | (Lucas Buchala)++ | S (12 files):
Move source/repository information to S01
 14:48
ecs: 8f5f83a | (Lucas Buchala)++ | S (12 files):
Move contributions/additions information to S01
ecs: d7bd63b | moritz++ | S (15 files):
Merge pull request #72 from lucasbuchala/move-notes

Move notes around
moritz lucas__: but they are an improvement nonetheless, so I'm merging them
lucas__ moritz: Thanks, again.
14:49 ClarusCogitatio left 14:50 ClarusCogitatio joined
moritz lucas__: also you should have an invitation for a commi bit already 14:51
14:51 BooK left 14:52 abraxxa left
lucas__ moritz: Oh, that github organization thing? :D I didn't know it was a commit bit. 14:53
14:53 ghostlines left, ghostlines joined 14:54 bowtie_ joined
moritz lucas__: it's actually a whole bunch of commit bits (there are many repos in there) 14:54
colomon did we ever get an "is symmetric" or "is commutative" trait?
psch ISTR not being able to commit to perl6/nqp even with belonging to the perl6 org. i think that makes sense, because nqp probably also wants the CLA? 14:56
moritz psch: it simply has separate teams
psch: if you need an nqp comit bit, let me know
psch: (and tell me your github ID)
14:58 bowtie_ left
psch moritz: i don't usually need it, but i'll speak up again in case it looks like i'm doing more in nqp. thanks 15:01
15:02 adu joined 15:04 thou joined
psch fwiw, what i did in nqp up to now was just a failed refactoring attempt and a few overloadable methods for typed exceptions, which is quick enough to review 15:04
in contrast to e.g. the rakudo cliparser or tr/// branches 15:05
15:06 chenryn joined, chenryn left
psch .oO( although removing the neccessity of reviewing frees up time for others, and we do have at least reasonable test coverage... ) 15:07
15:07 bowtie_ joined 15:09 aborazmeh left 15:11 Ven joined
Ven o/, #perl6. 15:12
psch Ven \o
PerlJam just saw xkcd.com/1443/
15:14 bowtie_ left 15:16 guru joined, ajr_ left 15:17 guru is now known as Guest70470, Guest70470 is now known as ajr_ 15:22 dj_goku_ joined, Gruber joined 15:23 ClarusCogitatio left 15:24 ClarusCogitatio joined, dj_goku left, Grrrr left, TimToady left 15:25 TimToady joined, FROGGS joined
FROGGS tadzik: ping 15:28
15:29 ClarusCogitatio left 15:32 ClarusCogitatio joined 15:33 tinyblak_ left, tinyblak joined
psch huh. so for 'sub infix:["a"] { }' World::colonpair_nibble_to_str gets an NQPMatch from the grammar, but NQPMu from the actions...? 15:37
tadzik FROGGS: pong 15:43
catching up with backlog :)
15:43 adu left
FROGGS tadzik: please read my last lines :o) 15:45
and rerun the updatelist.pl under that env var 15:46
15:46 ClarusCogitatio left
timotimo o/ 15:47
i'm glad that it wasn't my fault the ecosystem suddenly broke
tadzik oh
the URL is actually correct
but the env var is necessary, yes 15:48
15:48 JimmyZ left
FROGGS tadzik: sorry that I had to a foot in you home directory :o) 15:49
timotimo disabling https certificate validation sounds like a huge smell ...
tadzik FROGGS: no worries :)
timotimo: yeah, my thoughts exactly
FROGGS tadzik: I did not look at your p0rn, promise :P
timotimo what's wrong? is their certificate self-signed? or expired?
15:49 ClarusCogitatio joined
FROGGS we can install their cert, that should also do I guess 15:50
timotimo i would prefer that, but i don't know how to handle certificate storage and stuff ...
dalek kudo/nom: 0f77ebe | (Elizabeth Mattijsen)++ | src/core/IO.pm:
Introduce CHANGE-DIRECTORY primitive
kudo/nom: 2b57f3f | (Elizabeth Mattijsen)++ | src/core/ (2 files):
Eradicate :test from chdir/indir/tmpdir/homedir

As per current consensus / spec
PerlJam lizmat++ yay! 15:51
tadzik github cert in my browser verifies perfectly well
FROGGS tadzik: for me too
15:51 tinyblak_ joined
tadzik is our LWP too old on feather? 15:52
15:52 zakharyas joined
tadzik and using some version of SSL that's not cool anymore? 15:52
(so, any of them? :P)
FROGGS ahh btw, depending on the perl 5 ssl module you use, it does not do cert verification at all
15:53 sqirrel__ joined 15:54 tinyblak left
tadzik I'm using Perl from /usr/local/bin/perl, no local::lib as far as I can see 15:59
so it's just using feather's system libraries
I blame LWP, perhaps it's missing proper TLS support, lacking underlying libraries? 16:00
FROGGS tadzik: do you explicitly use an ssl library before you use lwp? 16:01
tadzik no
should I? :)
FROGGS I am... not sure right now 16:04
16:05 atta left
FROGGS maybe that helps? stackoverflow.com/questions/74358/h...rtificates 16:05
i.e. setting HTTPS_CA_DIR=/usr/share/ca-certificates/ (if that path exists)
16:06 rindolf left, telex left
FROGGS no, that does not help 16:07
16:08 telex joined
lucas__ Hi there. I've just finished moving authors downstairs. It's a spec pull request now. I hope it's correct. 16:08
16:10 atta joined
[Tux] Failed running perl Configure.pl --backends=moar --gen-moar at /pro/3gl/CPAN/rakudobrew/bin/rakudobrew line 25. 16:10
16:11 ClarusCogitatio left
FROGGS tadzik: I found it! 16:11
tadzik: that works: HTTPS_CA_FILE=/etc/ssl/certs/ca-certificates.crt perl updatelist.pl 16:12
tadzik \o/
FROGGS so, set it in your script and you are done
tadzik yeah, should work
FROGGS tadzik: I tested it on that server :o)
[Tux] rakudobrew 514 > rakudobrew build moar 16:13
Already up-to-date.
Your branch is up-to-date with 'origin/nom'.
make: *** No rule to make target 'm-clean', needed by 'clean'. Stop.
Failed running make realclean at /pro/3gl/CPAN/rakudobrew/bin/rakudobrew line 25.
16:13 kaleem joined
tadzik yay, works :) 16:13
FROGGS tadzik: please close github.com/perl6/ecosystem/issues/34 when the projects.json contains data :o)
[Tux]: is that error fatal?
16:14 ClarusCogitatio joined
[Tux] yes: and it happens on the system that was recently upgraded from i686 to x68_64 16:14
make distclean and/or git clean -dfx don't work
16:15 Ven left
tadzik FROGGS: huh, the frontend complains about Cwd.pm 16:15
the lakc of it, that is
FROGGS O.o
tadzik restarted it, now it works :)
FROGGS yeah, it does not complain in my shell
[Tux]: I nearly know nothing about rakudobrew, but the other guy here does :o) 16:16
tadzik weird
16:18 denis_boyun_ left, bowtie joined 16:19 bowtie is now known as Guest48172 16:20 lucas__ left 16:24 tinyblak_ left, cognominal left
tony-o how would i represent this structure for nativecall? msdn.microsoft.com/en-us/library/wi...85%29.aspx 16:29
[Tux] starts from scratch …
16:31 Guest48172 left
psch m: sub infix:{a} { } # /o\ 16:32
camelia rakudo-moar 315ec6: OUTPUT«===SORRY!===␤Internal error: find_var_decl could not find $_␤»
dalek href="https://cpandatesters.perl6.org:">cpandatesters.perl6.org: 53ca4a4 | (Tobias Leich)++ | / (2 files):
implement client side caching via Etags
 16:34
ecs: 3d64f1a | (Lucas Buchala)++ | S (38 files):
Move AUTHORS sections to end of file
ecs: 25d0973 | lizmat++ | S (38 files):
Merge pull request #73 from lucasbuchala/move-authors

Move AUTHORS sections to end of file
[Tux] lizmat, feedback on speed for Text::CSV6 16:35
i686 i686 x68_64
20141007 20141014 20141105
------------------ -------- --------- ---------
Text::CSV::Easy_XS 0.022 0.021 0.037
Text::CSV::Easy_PP 0.018 0.022 0.028
Text::CSV_XS 0.038 0.039 0.046
Text::CSV_PP 0.651 0.535 0.533
Pegex::CSV 1.947 1.420 1.387
csv.pl 12.543 8.437 9.152
test.pl 256.985 201.621 71.847
tony-o FROGGS: have you run across representing a union of structs for nativecall yet?
lizmat [Tux]: will look at it tonight
[Tux] (the last two are perl6)
16:35 fhelmberger left
[Tux] my approach gained a factor 3 without changing a single character: pure gain in perl6 16:36
lizmat [Tux]: in case other people want to look at it: what was the repo again?
[Tux] one sec
github.com/Tux/CSV 16:37
FROGGS tony-o: timotimo was working on that
tony-o FROGGS: i will hold off on these 2 functions then :-) 16:38
FROGGS tony-o: what two functions?
[Tux] carefully pokes tadzik about his promised slang
tony-o FROGGS: win32 calls to QueryPerformance[Frequency|Counter]
tadzik [Tux]: I promised a slang? :o 16:41
[Tux] yes: i can haz a space between the function name and the opening paren 16:43
psch m: say\ ("spaaace")
camelia rakudo-moar 315ec6: OUTPUT«spaaace␤»
tadzik sure that wasn't timtoady? :D
[Tux] yes, you promised me in Liz' room in Sofia 16:44
FROGGS [Tux]: are you talking about a grammar by any chance?
[Tux] did the name change? :P
FROGGS no :o) 16:45
[Tux] fwiw that promise might have been the main reason I got back into trying perl6
FROGGS a slang let you write in another language literally subsequent blocks
in subsequent* 16:46
[Tux] doesn't care what its called, as long as I can write «function (arg, arg);» instead of «function( arg, arg );»
FROGGS and a grammar is just a class-like regex
ahh
that's a slang then
16:47 ghostlines left 16:48 treehug88 joined
FROGGS [Tux]: creating a slang that inherits from the Perl 6 grammar is not easily doable as of today 16:49
[Tux] I want it yesterday, but next week will be good enough :} 16:50
FROGGS :S
[Tux] seriously: that is the most important issue that is holding me from having FUN in perl6
and I know more people that have similar issues
FROGGS I am thinking about it... gimme some time 16:51
[Tux] I will
I won't be here very often, but lizmat will know how to reach me
woolfy FOSDEM 2015 in Brussels will be quite special for us. Larry Wall will be there (I'm so proud and happy, wow). 16:52
Look at this schedule: fosdem.org/2015/schedule/track/languages/
You might see the first talk, just accepted by the FOSDEM organisation. Get ready to party. Yay!
lizmat fosdem.org/2015/schedule/event/get..._to_party/ 16:53
woolfy Only (minor) problem: the Perl-devroom is also on a Saturday. Whoever has his/her presentation from 15:00-16:00, might nog have many attendees... 16:54
I can live with such a problem. :-) 16:55
PerlJam "2015 will be the year that Perl 6 officially launches for production use" is kind of a bold claim. 16:56
(I'm not saying it won't happen, just that any hiccups will be magnified)
colomon woah, that's only three months away! 16:57
woolfy PerlJam: bold claims can be good!
psch $ perl6 -e'sub infix:["a"]($,$) { "foo" }; say 1 a 1' 16:58
foo
now for the spectest to see if anything broke :s
[Tux] having been present with perl6 launch (mug event in Monteray), it would be a good Idea to also be present at fosdem
PerlJam woolfy: agreed. But the delivery must be close to flawless otherwise we get all the naysayers coming out of the woodwork *worse* than I expect they already will
psch oh, that talk looks exciting 16:59
lizmat I have no doubt that a lot of shit is going to hit the fan
PerlJam [Tux]: Do you think there will be flying, breaking mugs at fosdem too? : )
er, :-)
lizmat but that's only to be expected
[Tux] PerlJam, nope 17:00
lizmat (thinking about all of the shit that's already flying around on perlmonks, reddit, etc.)
.oO( note to self: bring disposable mugs)
PerlJam lizmat: you know what would make me really happy? (and this is totally irrational) If chromatic gives Rakudo a try again, likes it, and recants on the negative vibes he's been sending our way. 17:01
lizmat is reminded of an Eagles album: hell freezes over
I'm just afraid chromatic suffers too much by the "not invented by me" syndrome 17:02
*from
cycling& 17:03
PerlJam chromatic: and if you're watching us from afar (because, say, a google alert was triggered when we mention your name :) ... *hugs* :) 17:04
tadzik [Tux]: ah, I remember now :) 17:05
17:05 sqirrel__ left
[Tux] :) ☺ 17:05
woolfy PerlJam: we might even want to ask him for a "Modern Perl 6" book. Along with many other interesting books... Programming Perl 6 (Larry), Beginning Perl (Curtis 'Ovid' Poe), Perl 6 Best Practices (Damian Conway), etc 17:06
Flying and breaking mugs... we buried the meme in Perl, august 2012. :-) 17:07
arnsholt masak: As our resident category theorist (I guess), what would you recommend as an introductory text to the field?
PerlJam woolfy: I think we'll have to wait until Perl 6 has been released to the wild for several years before there's enough material for a "Modern Perl 6" book. 17:08
woolfy: Though, a "Modern Perl" book that has a chapter on Perl 6 doesn't seem too far out there. 17:09
woolfy PerlJam: either way, chromatic coming back would be lovely 17:11
cycling&
17:15 dbane joined 17:16 ajr_ left 17:17 guru joined, guru is now known as Guest25022, Guest25022 is now known as ajr_
tony-o is there a way to get a handle for the current process and for a 'start'ed async block (is there a separate handle for that)? 17:18
17:21 raiph left
moritz tony-o: the return value of start { } is a Promise 17:21
tony-o: which you can use to wait for the thread, extract the result, and maybe even cancle the thing
tony-o moritz: i'm looking to pass a p6 version of C's HANDLE to NativeCall, I'm just not sure where i can get that 17:22
moritz tony-o: oh, so file handles?
tony-o Process and Thread handles
moritz rmpf, no idea
tony-o rmpf?
moritz noise of dissatisfaction
tony-o oh, hard to tell what is and isn't an acronym now 17:23
moritz :-)
lol I blug: perlgeek.de/blog-en/perl-6/2014-com...nding.html
PerlJam (moritz's employer)++ 17:24
looks like I have to make sure I can still use my paypal account now 17:25
17:25 oetiker_ joined 17:26 oetiker left, oetiker_ left, oetiker joined
arnsholt tony-o: I don't think there's a way to get the underlying C stuff relating to a Perl 6-level thread ATM 17:28
Not sure if that's even a good idea (or feasible, certainly near impossible on JVM I suspect)
ajr_ If I started working through Perl 5 examples, converting them to Perl 6, and asking why when they break, would that be a valuable guide to writing an introduction to Perl 6, or an annoying display of stupidity?
moritz ajr_: I'd be interested
tony-o ajr_: sounds pretty useful
arnsholt: i'm trying to use NativeCall to get process times on windows, it wants to get Process HANDLE or Thread HANDLE and I'm having trouble figuring out how to use those. (I'm caling GetProcessTimes and GetThreadTimes specifically) 17:30
ajr_ (I tried working through "Mastering Algorithms" a while back, and crashed and burned so fast it didn't feel worth continuing.)
arnsholt tony-o: I think you'll have to hack deeply into the NQP guts to get at that, if it's even possible 17:31
17:31 pdcawley joined 17:32 bowtie joined, bowtie is now known as Guest1933 17:33 psch left, ajr_ left 17:34 psch joined, treehug8_ joined 17:35 ajr joined, ajr is now known as Guest22120 17:36 Guest22120 is now known as ajr_ 17:37 treehug88 left 17:39 Guest1933 left 17:43 leont joined, psch left 17:44 psch joined 17:46 bowtie_ joined, treehug8_ left 17:47 treehug88 joined 17:49 pdcawley left, treehug8_ joined 17:50 FROGGS_ joined 17:52 ghostlines joined, treehug88 left, ghostlines left 17:53 ghostlines joined 17:54 pecastro left, FROGGS left 17:56 rurban joined 17:57 spider-mario joined, dakkar left 17:58 zakharyas left 17:59 dbane left 18:02 kjs_ left 18:13 daxim left 18:15 pat_js joined 18:18 daxim joined 18:19 bowtie_ left 18:24 bowtie_ joined
vendethiel I don't really like python, but I have to admit "python gangsta rap" is the best thing ever 18:24
( www.youtube.com/watch?v=FJ7QsEytQq4 )
18:25 ajr_ left 18:26 virtualsue left 18:28 kaleem left 18:31 bowtie_ left, Sqirrel joined 18:32 bowtie joined, FROGGS_ is now known as FROGGS, bowtie is now known as Guest75553 18:33 kjs_ joined
timotimo hm, what was i working on? 18:33
FROGGS spesh? 18:34
profiling json::tiny?
union support?
sjn memory recall techniques? :) 18:37
18:37 grettis joined
FROGGS sjn++ 18:37
vendethiel: :P 18:38
timotimo ah, unions in nativecall 18:46
i've thrown some ideas around, none of them seeming thoroughly satisfying
18:50 rindolf joined 18:52 gfldex joined
arnsholt timotimo: Might be possible to make them work more or less identically to structs 18:57
Except how the read memory, obviously =)
*they
18:57 guru joined 18:58 guru is now known as Guest36392, Guest36392 is now known as ajr_
timotimo yes 18:58
well, only how they lay out the memory
i thought i could have a trait "is union" that would have a flag set in the REPRData before compose gets called
and that would just make the CStruct begin each member at offset 0
18:58 rurban left 19:00 ghostlines left
arnsholt Oooh, clever idea! 19:02
19:06 kjs_ left, treehug88 joined 19:08 treehug8_ left
dalek rl6-roast-data: 61d6f11 | coke++ | / (4 files):
today (automated commit)
 19:10
vendethiel okay, so, the GLR semantics aren't going to change now, right? 19:15
19:15 raiph joined
vendethiel I've been slacking off on the tutorial lately, and by now "waiting to explain the list stuff" is no more an excuse :) 19:15
19:17 ptc_p6 left, ugator joined, ptc_p6 joined 19:18 Mso150 joined 19:23 kjs_ joined
timotimo arnsholt: if you like it that much, why not implement it? :)) 19:33
19:41 kjs_ left 19:45 antigremlin joined
dalek rlito: edff284 | (Flavio S. Glock)++ | / (2 files):
Perlito5 - fix BEGIN block namespace
 19:45
19:46 antigremlin left
FROGGS [Tux]: how would you call such a slang... say, in case I just made one up 19:50
19:55 raiph left 19:58 Guest75553 left 19:59 bowtie joined 20:00 bowtie is now known as Guest83856 20:01 pecastro joined
[Tux] FROGGS, I'd go for "banner" - en.wikipedia.org/wiki/Indent_style#Banner_style 20:03
FROGGS hmmm, then I have to rename it :o) 20:04
[Tux] thought that is more about indentation than space-use
FROGGS github.com/FROGGS/p6-Slang-Tuxic
[Tux] Tuxic is fine too ! :)
FROGS++
TimToady
.oO( Tuxecology :)
FROGGS *g* 20:05
and that is how it works: github.com/FROGGS/p6-Slang-Tuxic/b...g/Tuxic.pm
[Tux] and yes, I was aware of the need of parens with if/while
FROGGS kids: don't try that at home :o)
hmmm, it makes the right choice anyway: 20:06
perl6-m -Ilib -e 'use Slang::Tuxic; sub if (Int) {}; if (21) { say "ohh" }'
ohh
[Tux] how do I install that? 20:07
FROGGS clone it, and then: panda install p6-Slang-Tuxic/
vendethiel wow :P
FROGGS I'll add it to the ecosystem, and I hope it does no harm :P 20:08
psch FROGGS++
m: say %*LANG.keys
camelia rakudo-moar 315ec6: OUTPUT«0␤»
psch right, those only exist in Perl6::Grammar i think as of now? 20:09
iirc the user-facing interface is $~LANG..
[Tux] AWESOME!
psch or was it $~MAIN
FROGGS psch: you get your hands at %*LANG only in an EXPORT, because the Perl6::Grammar calls your EXPORT sub directly 20:10
there is no $~LANG yet
[Tux] "Wilt u meer of minder spaties toestan?" MEER! MEER! "Dan gaan we dat regelen!"
FROGGS O.o
I understand everything but "minder spaties toestan" 20:11
[Tux] toesta*a*n (I was typing too fast)
psch FROGGS: i knew about $~LANG etc not existing yet, but i hadn't realized that we can get there via EXPORT 20:12
FROGGS [Tux]: that... does not help at all :o)
psch: that's what v5 does for ages :o)
[Tux] it is based on some Dutch politician, doing political incorrect statements about Maroccans
FROGGS ahh 20:13
spaties eq species?
[Tux] spaces
FROGGS ahh, now I got it 20:14
dalek osystem: 22a14d9 | (Tobias Leich)++ | META.list:
added Slang::Tuxic
 20:15
FROGGS but nice to know that we can hack the Perl 6 grammar that way...
masak: that might be interesting for you too
20:20 kjs_ joined 20:21 virtualsue joined
[Tux] ===SORRY!=== Error while compiling test-t.pl 20:22
Two terms in a row
at test-t.pl:76
------> my $msg = $reason.sprintf ⏏(@args);
expecting any of:
www.xs4all.nl/~hmbrand/test-t.pl
FROGGS ohh
subs only atm... need to look into method calls
psch m: pi.sprintf("1.5%d").say 20:23
camelia rakudo-moar 315ec6: OUTPUT«Too few directives: found 0, fewer than the 1 arguments after the format string␤␤»
FROGGS that's probably handles by the 'name' token instead of the 'identifier' token
psch ^^ seems unintuitive 20:24
FROGGS m: pi.sprintf('1.5%d').say
camelia rakudo-moar 315ec6: OUTPUT«Too few directives: found 0, fewer than the 1 arguments after the format string␤␤»
psch oh, also the directive is clearly out of order.. :)
m: sprintf("%1.5d", pi).say 20:25
camelia rakudo-moar 315ec6: OUTPUT«3␤»
psch ...right
m: sprintf("%1.5f", pi).say
camelia rakudo-moar 315ec6: OUTPUT«3.14159␤»
FROGGS m: say pi.sprintf('1.5%d')
camelia rakudo-moar 315ec6: OUTPUT«Too few directives: found 0, fewer than the 1 arguments after the format string␤␤»
FROGGS m: say pi.fmt('1.5%d')
camelia rakudo-moar 315ec6: OUTPUT«1.53␤»
FROGGS there is no sprintf method, is there? 20:26
psch m: say (~pi).sprintf("%1.5f")
camelia rakudo-moar 315ec6: OUTPUT«Too few directives: found 0, fewer than the 1 arguments after the format string␤␤»
psch m: Str.^methods.grep("sprintf")
camelia rakudo-moar 315ec6: OUTPUT«Unhandled exception: No exception handler located for warn␤ at src/gen/m-CORE.setting:770 (/home/camelia/rakudo-inst-1/languages/perl6/runtime/CORE.setting.moarvm:warn:29)␤ from src/gen/m-CORE.setting:766 (/home/camelia/rakudo-inst-1/languages/perl…»
psch i'm not thinking good right now it seems...
FROGGS: that's probably it, in any case 20:27
i blame map and grep and so on...
they've been misguiding me! 20:29
vendethiel soooo, mysql's MATCH sucks. Wow. 20:30
dalek ast: cd74a7c | TimToady++ | S04-statement-modifiers/for.t:
ensure that "{$_}" for ... sets correct $_
 20:32
20:35 darutoko- left
psch m: "%1.5f".sprintf(pi).say 20:39
camelia rakudo-moar 315ec6: OUTPUT«3.14159␤»
psch i don't know what to think of that
FROGGS ahh -.- 20:45
PerlJam m: pi.fmt("%1.5f").say 20:49
camelia rakudo-moar 315ec6: OUTPUT«3.14159␤»
PerlJam TMTOWTDI! :)
psch m: say floor(pi * 50000) / 50000
camelia rakudo-moar 315ec6: OUTPUT«3.14158␤»
psch aww
PerlJam m: say floor(pi * 500000)/ 500000
camelia rakudo-moar 315ec6: OUTPUT«3.141592␤»
psch m: say floor(pi * 50000 div 10) / 5000
camelia rakudo-moar 315ec6: OUTPUT«Cannot call 'infix:<div>'; none of these signatures match:␤:(Int:D \a, Int:D \b)␤:(int $a, int $b --> int)␤ in block <unit> at /tmp/o73WG3BfaG:1␤␤»
psch i don't grok IEEE 754 20:50
20:52 raiph joined 20:53 denis_boyun_ joined 20:54 kjs_ left 20:58 exixt_ is now known as exixt
psch just to reassure myself: we'll never do 'my $foo = "bar"; sub infix:<<$foo>> { $foo }; say 1 bar 1 # prints "foo\n"', because we can't right? 20:59
PerlJam "can't" is such a strong word ... ;>
TimToady compile time can't rely on run time like that 21:00
that's time travel
21:01 spider-mario left
TimToady now if you said, 'constant $foo = "bar";' that's a different story 21:02
21:03 Mso150 left 21:05 Mso150 joined
psch there's definitely more thinking to be done how to get all the different sub and token colonpair notations to work 21:06
FROGGS [Tux]: I am trying to implement that tomorrow 21:07
21:07 Mso150_o joined
FROGGS gnight #perl6 21:07
21:07 FROGGS left
psch what i have (kinda messily) solves infix:['a'] and infix:['a', 'b'], but doesn't work for token q:sym['q'], to say nothing of the constant bit above... 21:07
same from me, g'nite o/ 21:08
21:08 raiph left, psch left 21:09 denis_boyun_ left 21:10 exixt is now known as exixt_, Mso150 left 21:12 kjs_ joined 21:15 rurban joined 21:16 colomon left 21:17 kaare_ left
masak m: my $foo = 42; constant bar = $foo; say bar 21:32
camelia rakudo-moar 315ec6: OUTPUT«(Any)␤»
21:33 kjs_ left, cibs left
masak fantasizes about a sufficiently good IDE that would detect and warn on that kind of situation 21:33
21:34 cibs joined
PerlJam masak: Maybe Padre could if given some help 21:34
masak mebbe. 21:35
vendethiel masak: Perl6 will be insanely hard for IDEs, btw :P
21:36 ptc_p6 left
masak I can see how that's true for the parsing part, because Perl 6 has a lot of fun with the parse/run barrier. 21:36
more fun than Perl 5, in some senses. 21:37
21:37 colomon joined
masak after having an AST and something PSI-like, I don't see how it's harder than most other languages, though. 21:37
vendethiel Right. But that's not the hard part -- haskell and scala IDEs are pretty good (seen haskforce?)
21:37 raiph joined
vendethiel the hard part is macro fun and metamodel fun 21:37
masak granted, those bits are challenging 21:39
moritz back from a very fine dinner
arnsholt I imagine Perl 6 IDE will work more or less like Lisp IDE stuff: The editor talks to a running program image in a background process
vendethiel well, I certainly hope we can get a rakudo-slave for the IDE :)
21:40 xfix is now known as notfix 21:42 user3 joined
arnsholt Bah. My clever Python variable handling scheme wasn't clever enough >.< 21:43
21:43 user3 left
arnsholt Action methods are of course executed inside-out, which breaks it 21:43
masak arnsholt: I've made peace with action methods working that way -- maybe if you provided more information, I could help? 21:44
arnsholt I want to mark free variables (that is, those that are referenced before being assigned to) as such 21:45
Doing it in action methods mostly works, but something like the for loop also introduces a binding
masak oh, right.
arnsholt And marking the variable as declared in the for loop action is too late
masak yes, I see the problem.
arnsholt Easy way out is to add the decl out-of-order in the grammar
masak I think so, yes. 21:46
you probably want to keep a dynamical.
21:46 rindolf left
arnsholt Yeah, already added $*BLOCK 21:46
Only sane way to keep track of that
masak arnsholt++ 21:47
vendethiel agrees
but I made peace with the order as well :-)
arnsholt Yeah, I'm at peace with the order. I just brainoed and expected things to work the way I needed them to =) 21:48
(Something that happens more frequently than I'd like to admit, I guess =)
21:50 kjs_ joined
masak .oO( brain in a way that sticks ) 21:51
21:52 kjs_ left
arnsholt masak: Do you have a recommended category theory introduction, BTW? 21:52
vendethiel arnsholt: "homotopy type theory" 21:53
masak arnsholt: it really started to click for me with math.mit.edu/~dspivak/CT4S.pdf
vendethiel (sorry, I'm kidding, that's a terrible book to start...)
masak: I think I'd start off with set theory 21:54
masak arnsholt: that helped me get further into www.amazon.com/Categories-Working-M...387984038/ -- which I still find dense.
vendethiel "Algebra chapter 0" by Aluffi is pretty good for a set theory introduction
vendethiel doesn't read a lot of programming books, because most of those are costly... 21:55
masak arnsholt: currently, I'm filling in some glaring gaps in group theory, universal algebra, and algebraic topology, all of which help me appreciate category theory better.
vendethiel masak: you might be interested in this :). www.cs.uoregon.edu/research/summer...culum.html
www.cs.uoregon.edu/research/summer...culum.html
www.cs.cmu.edu/~fp/courses/15816-f0...douts.html
www.cs.cmu.edu/~cangiuli/talks/hpt-icfp.pdf
Those contain a lot, category theory, logic, and whatever else. 21:56
arnsholt masak: Cheers!
vendethiel isn't going through to arnsholt, apparently :/
arnsholt It's one of those things I suspect I'd be better off knowing
vendethiel: Getting there =)
vendethiel arnsholt: my links are for you
arnsholt My set theory is a bit rusty, but mostly adequate I think 21:57
I should probably brush up on that too, though
vendethiel alright. I don't have *any* science-y background whatsoever, so... I just learned stuff on top of stuff, and then later I started reading proper stuff
I have a paper version of Homotopy type theory, but I'm still not reading it yet
arnsholt Yeah, in that case I can see how set theory doesn't come first
vendethiel it comes before category theory :) 21:58
arnsholt Oh, wow. A linear logic course. Neat!
vendethiel yes, a friend of mine linked those to me. You might be interested by answers from #idris and #agda, even maybe #nothaskell
masak haha, #nothaskell :) 21:59
vendethiel ;-)
arnsholt Linear logic is yet another thing I should learn more about
One of my master's advisors uses it to compute semantics for natural language
The resource sensitive part of the logic is very useful for that
(Also, having an operator called "lollipop" is pretty neat =D) 22:00
vendethiel yeah, logical frameworks are very interesting. Dependently-typed languages and proof assistants use those to great effect
also, yay for logical frameworks where ¬(¬a) is not a :D
arnsholt Yeah, I remember those
I took a course on (essentially) strategies for automated theorem proving (in FOL) once. That was pretty neat 22:01
Also what made me actually grok how models in FOL work
timotimo o/ 22:04
arnsholt \o
vendethiel arnsholt: when I hear people say "I took a course on *subject*" I always feel sad :P 22:05
timotimo arnsholt: are you interested in building CUnion support for CStruct?
PerlJam vendethiel: why? 22:06
vendethiel PerlJam: I never took any courses :)
PerlJam vendethiel: and that makes you sad?
vendethiel PerlJam: yes.
PerlJam vendethiel: so take some courses! :)
arnsholt timotimo: It sounds like a reasonable plan, certainly
vendethiel PerlJam: I can't get in university
s/y$/ies/
masak 'night, #perl6 22:07
arnsholt g'night masak o/
colomon \o
PerlJam vendethiel: there are online courses like from coursera an ocw.mit.edu 22:08
vendethiel PerlJam: yes, I've followed some of those
22:13 colomon left
arnsholt heads sleepwards too 22:14
Night, night!
lizmat gnight arnsholt
22:14 lucas_ joined
timotimo gnite arnsholt 22:15
lizmat moritz: why is your blogpost not on pl6anet.org ?
vendethiel g'night 22:16
timotimo hmm. what should i do next? 22:17
lucas_ hi
About S15 and S26:
a) convert everything to Perl 6 Pod in the future 22:18
b) convert S15 and S26 to Perl 5 POD
c) leave it as is at the moment
Have you are decided?
*already
lizmat also calls it a night 22:24
22:25 Mso150_o left
raiph "I'm also curious how this would look in perl6." Anyone want to respond? perlmonks.org/?node_id=1105972 (I plan to do so tomorrow if no one else does.) 22:26
22:31 telex left 22:32 telex joined 22:36 virtualsue left
lucas_ I guess I'll just write here my opinion, ok? Yes, I would like to have all the files in the same Perl 5 POD format. 22:37
22:37 colomon joined
lucas_ I think Perl 6 Pod format experiments could be done in a different repository. 22:38
Also, I don't think it's good to right a specification for some format in its own format.
Think about a YAML specification written in... YAML. :) 22:39
If something changes, it would have to be applied recursively in the document.
Once in its content, and once in its own format. 22:40
So... I would like to convert those files to Perl 5 POD. Will such thing be accepted and merged? 22:42
22:43 ajr_ left
raiph "2015 will be the year that Perl 6 officially launches for production use." <-- I hate that. 22:43
timotimo we should definitely nominate some random outsider to be our spokesperson 22:44
since they seem to know everything better than us 22:45
Ulti doing it now
:P
raiph timotimo: ? 22:48
lucas_ *good to write
22:49 gfldex left 22:51 BenGoldberg joined
lucas_ Well, iff anyone wants to answer me that later, write here and I'll read the logs later. Thanks. Bye. 22:51
22:51 lucas_ left 22:57 rurban left
tony-o .tell lucas_ pod is for writing documentation, yaml isn't. that isn't really a good comparison 22:59
yoleaux tony-o: I'll pass your message to lucas_.
Ulti hmmm I have reputation 0 on PerlMonks not sure anyone can see my posts
raiph Ulti: Wow, nice answer. 23:02
Ulti it's a good answer because there is enough material floating about for it to be a good answer ;) 23:04
also because Perl6 has really nice stuff that people still spend pointless amounts of time faffing with in Perl5 whilst trying to avoid deps
23:06 baest left
raiph Heh. That's about my all time favorite PerlMonks answer to someone asking something about P6. 23:08
Ulti: You can edit posts. And you can make those text only links clickable. 23:09
Ulti k k
also PerlMonks utterly sucks 23:10
23:10 baest joined
Ulti as in the interface to posting 23:11
not the content
maybe we need PerlEntomologist 23:12
23:16 bonsaikitten left 23:18 rmgk left, kjs_ joined, xiaomiao joined 23:19 treehug88 left, kjs_ left 23:20 rmgk joined 23:39 xiaomiao left 23:41 xiaomiao joined 23:48 dayangkun joined 23:55 raiph left

Powered by Cro and the Raku® Programming Language.

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

Thank you!