»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.
00:06 leont joined
dalek osystem: 18ce27a | colomon++ | META.list:
Move GD to perl6-community-modules

This takes Coke++'s changes and adds a few additional tweaks.
00:23
00:48 xenoterracide joined 00:52 test_ joined 00:56 xenoterracide left 00:57 xenoterracide_ joined 01:02 test_ left, thou left 01:06 xenoterracide_ left 01:12 xenoterracide_ joined 01:17 xenoterracide_ left, xenoterracide__ joined 01:30 xenoterracide__ left 01:39 xenoterracide__ joined 01:52 woshty left, virtualsue left 01:53 woshty joined 02:15 dayangkun joined 02:21 mauke_ joined 02:23 mauke left 02:25 chenryn joined, mauke joined 02:28 mauke_ left, mauke_ joined 02:31 mauke left 02:34 mauke_ is now known as mauke 02:41 chenryn_ joined, chenryn left 02:44 bjz left, bjz_ joined 03:16 noganex_ joined, Mso150 joined 03:19 noganex left 03:28 woshty left 03:33 Ben_Goldberg joined, BenGoldberg left, Ben_Goldberg is now known as BenGoldberg 03:38 leont left 03:43 Exodist joined 03:46 woshty joined 03:47 Mso150 left 03:51 xenoterracide__ left 03:52 xenoterracide joined 03:55 jack_rabbit joined 04:01 jack_rabbit left 04:28 azawawi joined
azawawi hi 04:28
can anyone please add github.com/azawawi/totem to panda module list? 04:29
tony-o azawawi: i'll add it, what does it do? 04:33
dalek osystem: 6ab6d85 | tony-o++ | META.list:
azawawi's totem -
04:36
04:36 mauke_ joined
tony-o .tell supernovous i'm trying to use xml::query and when i do .new(xml => $xml) i'm getting 'Cannot create an instance of that type', any clues? Tests are now failing when I install from panda as well 04:37
yoleaux tony-o: I'll pass your message to supernovous.
tony-o .tell supernovus i'm trying to use xml::query and when i do .new(xml => $xml) i'm getting 'Cannot create an instance of that type', any clues? Tests are now failing when I install from panda as well
yoleaux tony-o: I'll pass your message to supernovus.
tony-o carp. 04:38
04:39 chenryn_ left 04:40 mauke left 04:43 mauke_ is now known as mauke 04:46 ggoebel111111114 joined 04:48 ggoebel111111113 left 04:50 BenGoldberg left 04:59 bjz_ left 05:02 chenryn_ joined 05:13 kaare_ joined 05:17 araujo left 05:28 [Sno] left 05:32 dayangkun left
azawawi tony-o: thanks... smart web-based search bar experiment :) 05:42
tony-o: thanks
tony-o: it is basically autocompletion inside your mounted project files 05:43
tony-o: once we index them, you can match modules, files, create files, read docs through it 05:44
tony-o: currently it indexes modules in install/../site/lib
tony-o: think of it as a way to read code faster... 05:45
tony-o: later index all of modules.perl6.org/
azawawi &
05:46 azawawi left 05:53 xenoterracide left 05:54 virtualsue joined 06:02 cognominal left 06:23 [Sno] joined 06:54 ptc_p6 joined 06:56 dayangkun joined 07:03 davido__ joined, kaleem joined 07:05 darutoko joined 07:06 davido_home left 07:32 Isp-sec joined, FROGGS joined 07:34 oskie left 07:35 rindolf joined 07:38 abraxxa joined 07:42 pecastro left 07:49 eiro joined 07:51 rindolf left 07:53 azawawi joined
azawawi hi 07:53
does this look right?
m: my $file = '/dir1/dir2'; my @lib = $*SPEC.splitdir($file); say +@lib;
camelia rakudo-moar 315ec6: OUTPUT«3␤»
FROGGS m: my $file = '/dir1/dir2'; my @lib = $*SPEC.splitdir($file); say @lib.perl; 07:54
azawawi m: my $file = '/dir1/dir2'; my @lib = $*SPEC.splitdir($file); say +@lib; say @lib[0] eq ''
camelia rakudo-moar 315ec6: OUTPUT«Array.new("", "dir1", "dir2")␤»
rakudo-moar 315ec6: OUTPUT«3␤True␤»
FROGGS m: my $file = '/dir1/dir2'; my @lib = $*SPEC.splitpath($file); say @lib.perl; 07:55
camelia rakudo-moar 315ec6: OUTPUT«Array.new("", "/dir1/", "dir2")␤»
FROGGS ahh, that gives you volume + dirs + file
m: my $file = '/dir1/dir2'; my @lib = $*SPEC.splitpath($file, :nofile); say @lib.perl;
camelia rakudo-moar 315ec6: OUTPUT«Array.new("", "/dir1/dir2", "")␤»
azawawi volume like C and D on win32... i see 07:56
azawawi tries it on win32
lizmat S32/IO:597: "Unlike just splitting the directories on the separator, empty directory names (C<''>) can be returned, because these are significant on some OSes." 07:57
synopsebot Link: perlcabal.org/syn/S32/IO.html#line_597
azawawi rakudo.org/ is down btw from my side 07:58
lizmat S32-setting-library/IO:597 rather
azawawi: no problem here 07:59
azawawi now it is working...
i suspect a caching proxy in my ISP...
azawawi begins to use Perl 6 @ work for various parsing & monitoring scripts :) 08:00
08:01 PZt joined
lizmat azawawi++ 08:01
FROGGS ohh, nice :o) 08:03
the stuff I did at the weekend also makes me think that I can use it for several tasks at work 08:04
azawawi m: my $file = 'C:/dir1/dir2'; my @lib = $*SPEC.splitdir($file); @lib.perl.say;
camelia rakudo-moar 315ec6: OUTPUT«Array.new("C:", "dir1", "dir2")␤»
08:04 zakharyas joined
FROGGS I build a Bailador webservice with Template::Mojo and DBIish... works nicely :o) 08:04
azawawi Baliador is built upon HTTP::Easy right? 08:05
Baildor... :)
FROGGS yeah
azawawi damn...
FROGGS *g*
though, it takes about a second to deliver a simple page... I guess I should profile it at some point 08:06
08:06 lizmat_ joined
azawawi FROGGS: HTTP::Easy should be refactored to reuse HTTP::Message imho 08:07
sorry HTTP::Client :)
it is not my day today... :)
m: my $day = "today"; # :) 08:08
camelia ( no output )
azawawi FROGGS: HTTP::UserAgent to be exact 08:09
08:09 rindolf joined
azawawi FROGGS: the parsing in HTTP::Easy is not 100% correct 08:09
08:09 lizmat left, kurahaupo left
FROGGS azawawi: yes, I am aware of that, but it works for now :o) 08:17
masak morning, #perl6
masak is giving a course today
FROGGS but I'll work on that issue when the service is live
masak: moin
08:19 brrt joined
moritz \o 08:20
08:21 virtualsue left 08:23 ptc_p6 left 08:24 ptc_p6 joined
FROGGS hi moritz 08:24
08:27 ugator joined 08:28 spider-mario joined 08:33 cognominal joined 08:36 virtualsue joined 08:39 lizmat joined, mauke_ joined 08:42 lizmat_ left, lizmat_ joined, mauke left, mauke joined 08:44 mauke_ left 08:46 lizmat left 08:48 lizmat joined 08:49 lizmat__ joined, lizmat left 08:50 chenryn_ left, lizmat joined 08:51 lizmat___ joined, lizmat_ left 08:53 lizmat_ joined 08:54 lizmat__ left, dayangkun left 08:55 lizmat left 08:56 lizmat___ left
moritz my $ende = 09:00
$ehour == 24
&& $eminute == 0
? unixtime( $year, $month, $day, 0, 0, 0 ) + 24 * 60**2
: unixtime( $year, $month, $day, $ehour, $eminute, 0 );
who can spot the bug? :-)
09:02 lizmat joined 09:03 chenryn_ joined
brrt unixtime arguments are funky? /me can't recall how they go at any rate 09:04
09:04 mauke_ joined 09:05 lizmat_ left 09:07 kurahaupo joined, kurahaupo left 09:08 mauke left
moritz brrt: no, the bug is the 24 * 60**2 09:09
assuming that a day has 24 hours
yesterday didn't.
(at least in CET/CEST land)
09:10 mauke joined
brrt :-D 09:10
nice
uhm..... hmm
how do you propose fixing it?
09:13 mauke_ left
moritz brrt: in this case I fixed it by calculating the next day using Date::Calc::Add_Delta_YMD 09:13
brrt hmm 09:14
yes
moritz my ($y, $m, $d) = Add_Delta_YMD($year, $month, $day, 0, 0, 1); and then unixtime($y, $m, $d, 0, 0, 0)
09:15 lizmat_ joined, kjs_ joined 09:16 ab5tract joined 09:18 rurban joined, rmgk_ joined, rmgk left, rmgk_ is now known as rmgk 09:19 lizmat left 09:21 sftp left 09:22 sftp joined 09:24 Mso150 joined 09:28 brrt left 09:30 mauke_ joined 09:32 Mso150 left, mauke left 09:36 ugator left 09:37 mauke_ is now known as mauke 09:38 brrt joined, Isp-sec left 09:39 TuxCM left, dayangkun joined 09:45 lizmat joined
tadzik wow, FRA now has a wifi 09:48
09:48 lizmat_ left
FROGGS tadzik: how do you know? 09:50
tadzik FROGGS: I found one :D
good guy t-mobile provides one
FROGGS tadzik: what are you doing in FRA? 09:51
tadzik FROGGS: getting back home from moritz++'s place :)
FROGGS ohh, nice
tadzik FRA is of course on the way from NUE to WAW
FROGGS off course, aye 09:52
tadzik I'm sure that if I flew from Warsaw to Australia FRA would also be on the way :D
metacpan.org/pod/Devel::DidYouMean perl 5 catching up :P
09:54 TuxCM joined
FROGGS *g* 09:57
tadzik: can you help me later this week getting the cpandatesters live on feather or so?
tadzik FROGGS: sure 09:58
FROGGS awesome :o)
tadzik while we're at it we can also move the modules API somewhere outside my /home
10:02 lizmat_ joined 10:06 zakharyas left, lizmat left
masak Devel::DidYouMean only does function and method calls. 10:06
Perl 6 does all declared things :) 10:07
10:10 JimmyZ joined 10:11 pecastro joined
moritz nope, not methods (iirc) 10:15
m: say 'abc'.spit(/b/)
camelia rakudo-moar 315ec6: OUTPUT«No such method 'spit' for invocant of type 'Str'␤ in block <unit> at /tmp/mUhxBPBCs4:1␤␤»
10:16 lizmat joined 10:20 lizmat_ left 10:22 chenryn_ left 10:23 chenryn_ joined 10:29 salv0 joined, JimmyZ left
lizmat methods is on my todo list 10:32
10:33 ab5tract left, kurahaupo joined
FROGGS .oO( methods is on my List $todo ) 10:34
lizmat you french person, you!
FROGGS O.o 10:35
10:35 kurahaupo left
lizmat ok, so I just pushed a fix to File::Find (for when dir() returns Str) 10:36
how do I get that back in ext/File__Find in panda ? 10:37
just do the same change ?
moritz iirc that's a submodule 10:39
so git submodule update or something like that
(it's documented in the star release guide, the same applies for panda too)
lizmat so where does the star release guid live ? can't seem to find it in the rakudo star repo :-( 10:47
FROGGS lizmat: github.com/rakudo/star/blob/master...-guide.pod
10:47 virtualsue left
lizmat ah, so not under docs :( 10:47
FROGGS the command is: git submodule foreach git pull origin master
lizmat and then I need to commit? 10:48
moritz yes 10:49
lizmat FROGGS++ moritz++
FROGGS :o)
dalek nda: c42053e | (Elizabeth Mattijsen)++ | ext/ (3 files):
Bring submodules up to date
10:50
lizmat FROGGS: I'm considering implementing "use fatal" using the same approach you use for "use strict" 10:54
FROGGS lizmat: what does 'use fatal' actually do? 10:55
lizmat throw any warnings in that scope automatically
moritz s/fail/die/ basically
lizmat m: say try +"foo" # should say 0
camelia rakudo-moar 315ec6: OUTPUT«Cannot convert string to number: base-10 number must begin with valid digits or '.' in '⏏foo' (indicated by ⏏)␤ in method gist at src/gen/m-CORE.setting:13672␤ in sub say at src/gen/m-CORE.setting:15862␤ in block <unit> at /tmp/29k0TD3Ee0:1␤␤»
lizmat the reason the above fails, is that the Failure is being returned by the try 10:56
compare with:
m: say try die
camelia rakudo-moar 315ec6: OUTPUT«Nil␤»
moritz then it should also say Nil
because it's not a warning
lizmat m: say try +"foo"; say "this should be said, but isn't" 10:57
camelia rakudo-moar 315ec6: OUTPUT«Cannot convert string to number: base-10 number must begin with valid digits or '.' in '⏏foo' (indicated by ⏏)␤ in method gist at src/gen/m-CORE.setting:13672␤ in sub say at src/gen/m-CORE.setting:15862␤ in block <unit> at /tmp/uuzEe7_n7K:1␤␤»
lizmat actually, it is supposed to return 0 according to spec 10:58
m: say try (+"foo").throw # the effect of "use fatal"
camelia rakudo-moar 315ec6: OUTPUT«Nil␤»
lizmat hmmm....
anyways, first make sure the failure gets thrown, then caught by try
FROGGS, moritz: my question is: for 'use strict', you use a dynamic variable in the grammar 11:00
how can I expose such a variable at runtima?
in Perl6 space, when the exception is being handled ?
FROGGS hmmm
lizmat FWIW, the same approach I would like to use for 'use lib' 11:01
FROGGS I'm not sure
lizmat :-(
moritz lizmat: I guess you'll have to install a new lexical symbol for each scope in which the thinig changes 11:02
lizmat a new dynamic var... 11:03
hmm....
$*W.install_lexical_symbol I assume
or, maybe I should do it the other way around: that the runtime looks up the grammar lexical ? 11:04
moritz iirc EXPORT can return a hash (or pair list) of symbols to be installed in the use'rs lexpad
FROGGS lizmat: I don't think that will work 11:05
moritz dunno if that works for variables yet
but it could be made to work for variables
and then you'd have a pretty generic mechanism
and lib.pm6's EXPORT could just return a new @INC
11:05 kurahaupo joined
FROGGS moritz: EXPORT can export vars 11:05
moritz that's a copy + the new elems 11:06
lizmat problem is that EXPORT is done at runtime
I need the @*INC at compile time
FROGGS problem is, that it explodes for the @INC case, because @INC is already declared in the target
moritz FROGGS: huh? should be declared in an outer scope
FROGGS moritz: not always
moritz lizmat: huh? EXPORT should be called at compile time 11:07
FROGGS lizmat: what moritz says
moritz lizmat: I mean, it needs to know the list of symbols to export at compile time
at run time, the lexpad is immutable
lizmat ok, well, I tried doing that with @*INC some times already, and I couldn't get it to work
in any case, I think installing the lexical feels better 11:08
moritz that seems to work-ish 11:09
FROGGS lizmat: yes, for transferring parser state you'd need a special lexical, aye
moritz lunch& 11:10
lizmat FROGGS: so what is $?STRICT ? 11:11
1071 in Grammar: if $have_outer && $*UNIT_OUTER.symbol('$?STRICT') {
FROGGS that's for the case where the outer has installed that compile time var... I'm not sure if we are installing it at all atm 11:12
but we need something like that for EVAL for example
11:15 kurahaupo left
lizmat it's not installed 11:17
FROGGS: should line 1072 not read $?STRICT as well then ? 11:19
11:20 kurahaupo joined, woshty left
lizmat - $*STRICT := $*W.force_value($*UNIT_OUTER.symbol('$*STRICT'), '$ 11:20
+ $*STRICT := $*W.force_value($*UNIT_OUTER.symbol('$?STRICT'), '$
11:26 woshty joined 11:31 zakharyas joined 11:40 kaleem left 11:52 chenryn_ left, JimmyZ joined
lizmat lunch& 11:56
FROGGS lizmat: I can't tell without tests 11:59
12:05 ghostlines joined, ghostlines left 12:06 ghostlines joined
timotimo www.reddit.com/r/perl6/comments/2k...th_perl_6/ 12:09
12:13 araujo joined, araujo left, araujo joined
moritz timotimo: I've replied 12:14
timotimo thanks :) 12:15
should i go over the whole 5-to-6 document to try and spot out-of-dateness?
JimmyZ authorized Reply!
timotimo "i am moritz lenz, and i approve of this message" 12:16
JimmyZ :P
timotimo moritz: does the "() is a list because special cased" fact have a place at the end of the sentence "1, is a list, (1) isn't."? 12:17
moritz timotimo: wouldn't hurt, though iirc I've done that not-too-long ago
timotimo: IMHO yes
brrt moritz++ 12:18
moritz timotimo: github.com/moritz/perlgeek.de/tree...erl-5-to-6 has the sources, if you want to patch anything
timotimo thanks :)
it'd be interesting to note somewhere that %foo<bar baz> is actually %foo{'bar', 'baz'} rather than %foo{'bar baz'}
FROGGS who blogged recentish about reading material about perl6?
timotimo vendethiel did a lightning talk 12:19
but i guess that's implied with "auto quoting"
FROGGS ohh, right
timotimo moritz: how should i handle the date part of the articles? should i bump the date when i change stuff? 12:20
FROGGS vendethiel: can you respond to [1] with links to reading material about Perl 6 from your talk at the apw? [1] www.reddit.com/r/perl6/comments/2k...th_perl_6/
timotimo like, in one place it says the output of $x.WHAT is 'Int()', whereas nowadays it would be '(Int)' - does that warrant a date bump?
12:22 dayangkun left
moritz timotimo: don't bump it 12:24
timotimo OK
moritz timotimo: also, there are some small things that I have fixed in the .pod files, but haven't re-generated the HTML yet
timotimo fair enough
moritz timotimo: so if you proof-read, please do it with the .pod files 12:25
timotimo ah
makes sense, will do.
12:28 denis_boyun_ joined
brrt brb 12:32
12:32 brrt left
[Coke] colomon: I would like a commit bit to perl6-community-modules 12:34
colomon [Coke]: think I just sent an invitation to you. 12:35
colomon has never done that before, and may have done it wrong
12:37 lizmat_ joined 12:38 kjs_ left
colomon \o/ 12:38
12:40 lizmat__ joined, rurban left, lizmat left
dalek rl6-GD: e53deac | coke++ | t/00-load.t:
fix typo
12:41
[Coke] ^^ I added webhooks for all the community owned modules.
colomon [Coke]++ # typo 12:42
webhooks?
12:42 azawawi left
colomon oh! 12:42
12:42 leont joined
colomon you mean we get reporting on commits now? 12:42
[Coke] a post-commit hook for the web. 12:43
colomon++ yes.
if you want to do that for any new modules, just copy the webhook.
12:44 lizmat_ left 12:45 brrt joined 12:46 lizmat joined
dalek ast: 2cba5c0 | usev6++ | S02-names-vars/perl.t:
Separate test for RT #123048
12:46
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...?id=123048
colomon [Coke]: how do you do that? 12:48
Ah, think I found it. 12:49
[Coke]++
12:49 lizmat_ joined, lizmat__ left 12:52 rurban joined 12:53 lizmat left, lizmat joined 12:55 ptc_p6 left 12:57 lizmat_ left
timotimo m: my Junction $b = 3 | 2; my $a = 2; say "Yes" if $a != $b; say "Yes2" if !($a == $b); 12:58
camelia ( no output )
12:58 fhelmberger joined
timotimo moritz: that's an example from inside 08-junctions.pod 12:58
the docs expect the output to be Yes
13:00 rindolf left
timotimo m: say [1, 2] === [1, 2] 13:00
camelia rakudo-moar 315ec6: OUTPUT«False␤»
moritz timotimo: that's an out-of-date-o 13:01
timotimo yup
would you like to rephrase that example + paragraph?
lizmat feels stupid
I just managed to upgrade our live web server instead of our test webserver :-( 13:02
moritz timotimo: I'm currently at $work, and can't put much thought into it right now
timotimo: if you don't feel like rephrasing it, please open me a GH issue
timotimo will do
so far, most changes i've made have been turning Perl 6 into Perl 6 13:03
:P
and the github editor doesn't give me a diff before i do the commit, so the commit messages are all useless
i'll rebase them locally and do some squashes/rewords
moritz lizmat: happens to everybody :-)
timotimo i think i feel stupid more often than i feel smart 13:06
arnsholt I definitely feel stupid more often than I feel smart 13:07
13:07 rindolf joined
timotimo i might even feel stupid more often than i feel either normal or smart 13:09
13:09 guru joined 13:10 guru is now known as Guest45519
timotimo m: $_ = "test"; say :e 13:10
camelia rakudo-moar 315ec6: OUTPUT«Unexpected named parameter 'e' passed␤ in sub say at src/gen/m-CORE.setting:15859␤ in block <unit> at /tmp/ZWjierrwry:1␤␤»
timotimo m: $_ = "test"; say (:e)
camelia rakudo-moar 315ec6: OUTPUT«"e" => Bool::True␤»
13:10 Guest45519 is now known as ajr_
timotimo m: $_ = "test"; say (so :e) 13:10
camelia rakudo-moar 315ec6: OUTPUT«True␤»
timotimo 11-basic-operators.pod seems to think :e is still the file test thingie
timotimo rewrites 13:11
m: my $ls = lazy { sleep 10 } 13:17
camelia ( no output )
timotimo m: my $ls = lazy { sleep 5 }; say "nothing done"
camelia rakudo-moar 315ec6: OUTPUT«nothing done␤»
timotimo moritz: should i point out that this is NYI?
[Coke] m: so lazy sleep for now 13:19
camelia rakudo-moar 315ec6: OUTPUT«(timeout)»
rurban p: my $ls = lazy { sleep 5 }; say "nothing done" 13:21
camelia rakudo-parrot 315ec6: OUTPUT«nothing done␤» 13:22
13:22 itz left, itz joined
moritz timotimo: or remove any mention of lazy { } 13:22
13:23 rindolf left
timotimo moritz: what about circumfix operators that "are thus implemented as macros"? 13:25
is there any case at all left where an operator is actually a method rather than a sub? 13:26
lizmat () afaik
timotimo ah, of course
lizmat you did mean postcircumfix, no?
timotimo nope
the 5-to-6 article suggests the circumfix < > operator is implemented as an "is parsed" macro 13:27
moritz that's wrong; it's just a special, syntactic construct
timotimo i'll just kill the whole section
13:28 kaare_ left
moritz +1 13:28
13:29 ugator joined
moritz timotimo++ 13:30
13:32 virtualsue joined
timotimo hm 13:33
std: enum bit Bool <False True>
camelia std 14ad63b: OUTPUT«===SORRY!===␤Illegal redeclaration of symbol 'bit' (see line 1) at /tmp/dchFYiNkST line 1:␤------> enum bit⏏ Bool <False True>␤Two terms in a row at /tmp/dchFYiNkST line 1:␤------> enum bit ⏏Bool <Fa…»
timotimo m: enum Hackers :Larry<Perl>, :Guido<Python>, :Paul<Lisp>; say Larry.perl; 13:35
camelia rakudo-moar 315ec6: OUTPUT«===SORRY!=== Error while compiling /tmp/haDLl1Pm2g␤Undeclared names:␤ Hackers used at line 1␤ Larry used at line 1␤Undeclared routine:␤ enum used at line 1␤␤»
timotimo m: enum Hackers (:Larry<Perl>, :Guido<Python>, :Paul<Lisp>); say Larry.perl;
camelia rakudo-moar 315ec6: OUTPUT«Hackers::Larry␤»
timotimo mhm, mhm.
13:38 lizmat_ joined
brrt timotimo - which paul is that? :-) 13:41
13:41 lizmat left
rurban Paul Graham 13:42
brrt oh.. that's not a real hacker imho
rurban Agree, He was never part of the LISP community. Just wrote books, and his own website 13:43
but he is a serious macro hacker
brrt i have one of his books, and it scar(r)ed me from common lisp 13:45
timotimo moritz: if you tell me when those changes are likely to hit your website, i'll mention my little proofreading effort in that reddit post 13:47
rurban Really? I met him personally and liked that his simple ANI Common LISP book wss so approachable to beginners. The fellow lisp hackers didn't like it for this.
the macro book "On Lisp" is pretty hardcore 13:48
timotimo at the moment i also have the perl6 most wanted project open in two tabs and i'm meaning to link to a few modules from there
moritz timotimo: I can update it tonight
timotimo cool.
13:49 kaleem joined, kjs_ joined
timotimo done 13:50
brrt rurban - it's terribly offtopic, my experience wasn't per se that it wasn't accessible
it just makes clisp come off as a terribly impractical language
as in
i'd rather write the same stuff in perl / python or even java 13:51
because if you're writing stuff wherein the accidental complexitiy dominates the essential complexity, you're not writing the right stuff
13:52 zakharyas left
brrt (the assumption being that macro's help you avoid accidental complexity but not the essential complexity) 13:53
oh, is the flip-flop test flappy?
or does it use bit shifting? 13:54
timotimo has anybody thought about advent calendar posts for this year yet? 13:56
me or tadzik (or both of us) should probably write one (or multiple) articles on game development with what we have nowadays on perl6
FROGGS +1 13:57
no, I'd probably write about cpandatesters
moritz timotimo: I have; I'd be happy to write about the MoarVM profiler (though jnthn++ would get priority here)
FROGGS err, s/no//
13:58 zakharyas joined
[Coke] PerlJam: octover? 13:58
FROGGS there is a lot to write about this year me thinks
13:59 notfix is now known as xfix 14:02 thou joined
brrt yes 14:03
14:04 kaleem left
b2gills If I knew how to write NQP I'm fairly sure I've nailed down how to fix `enum Phonetic (:Alpha<A>, <Beta Charlie>)` 14:07
If I replace the line on github.com/rakudo/rakudo/blob/nom/....nqp#L3292
@values.push($*W.compile_time_evaluate($<term>, $_));
with `@values.push($_[0][0]);@values.push($_[0][1]);` that example works
colomon smoke test keeps on hanging when run from cron. does fine if run by hand. :/ 14:10
dalek rl6-roast-data: ab73248 | coke++ | / (5 files):
today (automated commit)
14:11
rl6-roast-data: c7230e3 | coke++ | / (5 files):
today (automated commit)
[Coke] 44 JVM failures; 32045 failures in moarvm-jit, 37 failures in moarvm-nojit, 45 failures in parrot 14:12
brrt wow! what the hell 14:14
lizmat_ que? 14:15
14:15 lizmat_ is now known as lizmat
brrt moarvm jit being bad 14:15
fwiw, my own spectest run wasn't that bad 14:16
lizmat [Coke] which commit?
rurban 45 for parrot is also pretty heavy. 36 from S22-package-format/local which passes for me
lizmat S22 is probably my fault in some way 14:17
brrt hmm
i want to know why the JIT died
and in what context 14:18
timotimo brrt: it's possibly my implementation of the multicache* ops that tried to pass one more arg than i actually allocated in that buffer
14:18 Ven_ joined
timotimo that was really pretty wrong 14:18
Ven_ FROGGS: I refuse to use reddit :-)
hello, #perl6.
brrt \o Ven_ 14:21
timotimo - let me check those ops :-)
Ven_ brrt: if you tried to start CL using "on lisp", then yes, I feel you
timotimo brrt: i fixed them already
Ven_ brrt: you might want to try "practical common lisp" :) 14:22
brrt i used 'ansi common lisp' iirc
timotimo ./perl6-m: No such file or directory
Ven_ it's very old now, isn't it?
timotimo kind of hard to pass the spec tests if your compiler doesn't build any more 14:23
FROGGS Ven_: np, can you put your slides online?
Ven_ FROGGS: they are online already. link is on the apw website
FROGGS ohh
Ven_ *g*
brrt they look good to me timotimo 14:24
FROGGS Ven_: where? I just see that: act.useperl.at/apw2014/event/1616
Ven_ FROGGS: in "presentations". act.useperl.at/apw2014/talk/5833 14:25
FROGGS Ven_: thansk 14:26
thanks*
timotimo brrt: keep in mind i just fixed them a few hours ago 14:27
whereas cokes spec test run may have run before that
brrt ah ok
i did not know that
you can excuse me for feeling responsible for all that is JIT
[Coke] lizmat: a372592 was good jit, a07bdab was bad jit
timotimo :)
brrt could well be that [Coke]++ uncovered some edge case I haven't seen yet
lizmat a07bdab is a docs commit ??? 14:28
[Coke] there was a commit in the past few hours that might fix it? re-checking with latest...
lizmat: those are just boundaries.
whenever I happened to kick off the runs.
lizmat ok 14:29
brrt let's hope so
gtodd nqp is pretty cool ... :-) 14:30
gtodd just figured that out
soo .... why do we need perl6 if we have nqp? :-D 14:31
moritz gtodd: because NQP doesn't have listops
gtodd :-D
lizmat or assignment ?
gtodd I know there was something the "NQ" part hehe
moritz also, I get tired of typing nqp:: eventually :-) 14:32
gtodd :)
but since it is so useful .... it would seem NQP would essential for implementing any Perl6 compiler 14:33
or is it just the way to implement a rakudo backend ? :-)
JimmyZ there is a rubyish and phpish compiler too 14:34
14:34 kaare_ joined, rindolf joined
JimmyZ *are 14:34
gtodd right
moritz gtodd: well, niecza tried without an intermediate step
gtodd moritz: yeah that's sort of what I was thinking about ... if another .NET variant of perl6 was going to come to life would nqp be in the mix 14:36
I'm guessing there are advantages to each approach ... 14:37
[Coke] timotimo: yes, HEAD seems fine.
JimmyZ
.oO(6model was firstly implemented in .NET)
brrt pfew
timotimo thank you, [Coke] 14:41
[Coke] spectests still using "winner"
gtodd JimmyZ: It's all a blur ... I was just going through the Edument slides on 6model and had some slides about Moose open on another desktop at the same time .... :-) 14:43
14:43 molaf joined 14:49 zombcode left 14:50 bcode joined
gtodd anyway I look forward to future installments in the nqp-[<backend>] story 14:52
colomon woah, JSON::RPC depends on a LOT of modules 14:53
and works. :\ 14:56
14:57 chenryn joined
colomon just installed a cron to be run every hour, so he can test this closer to the source 14:58
FROGGS colomon: how are your feelings about emmentaler vs. cpandatesters btw? 15:00
colomon FROGGS: I've got no clue about cpandatesters
FROGGS (I know, cpandatesters is not online yet, so it is probably hard to judge)
colomon it's got a cute name
FROGGS colomon: it like like this: froggs.de/perl6/cpantesters/ 15:01
and when all goes well it will be online this week
but it is only valuable when we get reports from different machines/users/compiler versiosn
versions*
colomon Seems valuable but (at least at first glance) completely different in scope? 15:03
PerlJam cpan-date-sters? :)
FROGGS colomon: what's the scope of emmentaler then? 15:04
PerlJam: :P
15:05 immortal joined, immortal left, immortal joined
moritz lexical! 15:06
FROGGS lexical cheese? 15:07
15:07 erkan left
moritz with the holes analyzable at compile time 15:08
FROGGS I bet the holes also get autovivified over time 15:10
15:10 ajr_ left
FROGGS .oO( resists to make a pun about Brie and runtime, because that probably only work in German ) 15:10
15:11 guru joined, guru is now known as Guest45432 15:12 kjs_ left, Guest45432 is now known as ajr_ 15:15 treehug88 joined
colomon FROGGS: sorry for disappearing there 15:17
pmichaud good morning, #perl6
15:17 [Sno] left
FROGGS colomon: np, I do not expect immediate reponses on irc :o) 15:17
moritz good am, pm
FROGGS hi pmichaud
colomon FROGGS: Am I correctly understanding the idea is to capture information on module installs? 15:18
just as a for instance difference, the smoke tester is running all modules against the HEAD of rakudo moar. 15:19
FROGGS colomon: it captures module install attempts, aye 15:20
colomon: but I don't see why panda would forbid attempting a (re)installation of all modules
colomon so ( at least as shown there ) cpandatesters is telling you how things work against the latest relase 15:21
FROGGS I mean, we can allow it easily
colomon: not only against the latest release, against the compiler version one is currently using
15:21 bcode is now known as zombcode
FROGGS colomon: the shown page is just a mockup; the generated page on my box already show more information 15:22
colomon I dunno, cpandatesters is clearly an awesome idea. 15:23
it seems more like it's something that goes alongside emmentaler rather than replacing it. but maybe I'm wrong and I'll not have to check each morning to find out what's gone wrong with the emmentaler. :) 15:24
FROGGS colomon: that's why I want to talk to you... I don't want to kill emmentaler, but I also want to know what I can steal from it :o) 15:25
pmichaud I'm getting spectest errors on my system (using moarvm)... are these known/expected?
flip-flop.t and local.t 15:26
moritz not known to me
FROGGS flip-flop was mentioned today already (either here or on #moarvm)
lizmat S22 is locat.t, right?
pmichaud yes, S22
lizmat I probably messed that up :-(
chenryn I try to run perl6-bench, and the `bench time $CHECKOUTS` hang at `15/65: Testing while_pushme ...` of nqp-moar/2014.10 fro several hours. Anyone knows the reason? 15:27
moritz flip-flop.t was modified yesterday
lizmat it also appears that my inadvertent upgrade of our web server to Yosemite, removed all of the database files that were on the system
thank you &^$%&^%*&Y()*_ much, Apple!
FROGGS timotimo: do you know about chenryn's problem? 15:28
lizmat goes offline a bit to consider her sins
15:28 lizmat left
FROGGS eww 15:28
15:28 MilkmanDan left 15:30 MilkmanDan joined
colomon FROGGS: … yeah, trying to think how I can explain what I know. 15:33
FROGGS we can also do that as soon cpandatesters is online...
colomon FROGGS: I guess the first thing is, expect noise. :) 15:36
FROGGS yeah, probably
dalek ast: f4efeee | pmichaud++ | S03-operators/flip-flop.t:
Change flip-flop.t test from 'todo' to 'skip' to avoid spectest failure.
colomon switching subjects (slighjtly)
here's the command that's hanging for me in my emmentaler runs (but only from cron):
(whoops) /bin/sh -c /home/smoker/.rakudobrew/moar-HEAD/install/bin/perl6 --target=mbc --output=blib/lib/JSON/RPC/Server.pm.moarvm lib/JSON/RPC/Server.pm 15:37
guess you can't start an IRC line with a slash
arnsholt /Yes you can =)
But you have to send "/ /Stuff"
colomon anyway, my (big linux box with lots of RAM) machine has spent the last 30 minutes trying to execute that 15:39
FROGGS colomon: that could be an issue with your ENV, because that's quite different when running using cron
brrt / /foo
hmm 15:40
colomon FROGGS: yes, I'm guessing that, but what could cause a perl6 compile to hang like that?
japhb chenryn: What OS are you running on, what CPU, and how much RAM?
FROGGS colomon: BEGIN blocks?
moritz colomon: precompilation bugs, maybe 15:41
brrt lizmat, wrt to: "thank you &^$%&^%*&Y()*_ much, Apple!thank you &^$%&^%*&Y()*_ much, Apple!"
brrt so totally feels
japhb .ask timotimo Can you make a perl6-bench mini-benchmark out of the AST benchmark you've been playing with?
yoleaux japhb: I'll pass your message to timotimo.
FROGGS m: say <0.0.3 1.0.0>.sort(:by({ Version.new($^a) cmp Version.new($^b)}))
camelia rakudo-moar 315ec6: OUTPUT«0.0.3 1.0.0␤»
japhb brrt: What was her rant in reference to?
FROGGS m: say <0.0.3 1.0.0>.sort(:by({ Version.new($^b) cmp Version.new($^a)}))
camelia rakudo-moar 315ec6: OUTPUT«0.0.3 1.0.0␤»
FROGGS why is that the same?
15:41 psch joined
chenryn 2.6.32-431.11.2.el6 Intel(R) Xeon(R) CPU E5-2620 0 @ 2.00GHz 32GB ram 15:42
Ven_ "NYM"?
brrt something about an update removing all database files
that's not my personal problem with os x, but i have others
moritz m: say <0.0.3 1.0.0>.sort({Version.new($^b) cmp Version.new($^a)}))
camelia rakudo-moar 315ec6: OUTPUT«===SORRY!=== Error while compiling /tmp/1QUu29WBFL␤Unexpected closing bracket␤at /tmp/1QUu29WBFL:1␤------> {Version.new($^b) cmp Version.new($^a)})⏏)␤»
moritz m: say <0.0.3 1.0.0>.sort({Version.new($^b) cmp Version.new($^a)})
camelia rakudo-moar 315ec6: OUTPUT«1.0.0 0.0.3␤»
moritz m: say <0.0.3 1.0.0>.sort({Version.new($^a) cmp Version.new($^b)})
camelia rakudo-moar 315ec6: OUTPUT«0.0.3 1.0.0␤»
FROGGS hmmm
moritz FROGGS: seems that :by is ignored
looks like a fossil from the days where positional arguments could be filled by name 15:43
FROGGS rakudo/src/core/List.pm:399: method sort($by = &infix:<cmp>) {
ahh, it's not a named
moritz FROGGS: yes, it's not :$by
FROGGS moritz++
japhb chenryn: Woah, they're still patching RHEL/CentOS 6? Sheesh.
But OK, that should be no problem CPU or RAM wise, hmmm. 15:44
And you're not running anything else that would cause swapping or chew all the CPU?
.tell timotimo Actually, both before and after microoptimizations would be good to have, because that shows how much work the user is having to do that the compiler should be doing ... and how much better it performs with user assistance. 15:46
yoleaux japhb: I'll pass your message to timotimo.
psch Ven_: "not yet merged" - but further discussion delayed the PR 15:51
Ven_: so i guess it should read 'NYI' again
Ven_ psch: mmh
oh.
psch also hi #perl6, and sorry for the oversight in flip-flop.t 15:52
chenryn of course no. And when I delete nqp-moar from $CHECKOUTS, I saw `rakudo-moar` run at `56/65: Testing parse-json-no-obj-creation ...` right now.
my $CHECKOUTS='perl5/v5.20.1 nqp-moar/2014.10 rakudo-moar/2014.10' before, and only 'perl5/v5.20.1 rakudo-moar/2014.10' now. 15:54
15:55 grettis joined 15:56 kurahaupo left
TimToady m: 42 16:01
camelia ( no output )
TimToady m: 42;43
camelia rakudo-moar 315ec6: OUTPUT«WARNINGS:␤Useless use of constant integer 42 in sink context (line 1)␤»
TimToady m: 42.0; 43.0
camelia ( no output )
TimToady missing diagnostic for rats
LHF for someone
16:03 Hor|zon left
pmichaud m: ($_ * 2 if .odd for 0..10).perl.say 16:04
camelia rakudo-moar 315ec6: OUTPUT«No such method 'odd' for invocant of type 'Int'␤ in block <unit> at /tmp/7C86gN0n4u:1␤␤»
pmichaud m: ($_ * 2 if $_ %% 2 for 0..10).perl.say
camelia rakudo-moar 315ec6: OUTPUT«(0, 4, 8, 12, 16, 20).list␤»
16:06 JimmyZ left
TimToady pmichaud: I note that change the current Nil to () there will tend to spit out extra () under our new list semantics 16:08
pmichaud TimToady: except I'd expect the 'for' to flatten those out. 16:09
TimToady for flattens its input, not its output, I think 16:10
pmichaud oh
yes.
anyway, yes, that's what I'm working on ATM
TimToady an argument could still be made that we need a return value that always disappears in lists 16:11
pmichaud feel free to make that argument. :)
FROGGS which has not another meaning or is used in other cases, like it happened for Nil 16:12
pmichaud and then be careful of "always disappears" versus "disappears when flattened"
TimToady also, only disappears itself, without implication of lower flattening
so it's not equiv to .flat
16:13 isBEKaml joined, H2O1 joined 16:14 FROGGS[mobile] joined, FROGGS left
TimToady These-aren't-the-droids-you're-looking-for is a bit unweildy 16:14
FROGGS[mobile] we could call it HarryPotter 16:15
16:16 H2O1 left
japhb force-wave 16:16
TimToady NIL!
Nothing, Nada, Zilch 16:17
pmichaud m: (1, 3..5).list.perl.say
camelia rakudo-moar 315ec6: OUTPUT«(1, 3, 4, 5).list␤»
pmichaud post-GLR, would we expect that to remain (1, 3..5) ? 16:18
TimToady I'd think .list on a list would be a no-op 16:19
japhb chenryn: OK, nothing I know of that would cause it to go to sleep like that. Can you run the benchmark for nqp-moar/2014.10 while running e.g. atop in another window and see if something lookings amiss?
Gah, cell reception SUCKING
japhb is now typing blind, wheeee
TimToady probably just space weather 16:20
pmichaud TimToady: so it remains (1, 3..5), and the way to get the Range to interpolate is to do .flat ?
isBEKaml japhb: if you're typing blind - atleast you know how to type. :)
TimToady or whatever sugar we decide means that
pmichaud okay.
(I'm agreeing with all of these, fwiw. :) 16:21
Ven_ mmh, how much much would it be to have a .each? 16:23
16:25 denis_boyun_ left 16:26 rurban left
chenryn @japhb: there is a process like `perl /data1/git/perl6-bench/timeall --outfile=/data1/git/perl6-bench/timings/nqp-moar/2014.10.json nqp-moar` 16:27
cann't use `strace -p $pid` to follow it... 16:28
TimToady Ven_: do you mean something that flattens only the top level? 16:29
16:30 chenryn left
Ven_ TimToady: no, I mean our currently-specced .each (>>. but with order) 16:30
TimToady ah, well, up till now everyone has thought it was very much much :)
japhb chenryn: Yeah, bench launches timeall, which then launches the compiler in question for each run of each test. So you'd pretty much have to trace the process group or somesuch. 16:31
Ven_ well :)
decommute &
16:31 Ven_ left
japhb Hence my suggestion of atop or another "sees all the perf info at once" program 16:31
16:33 brrt left
japhb timotimo: p6weekly this week? 16:33
16:34 kjs_ joined 16:35 salv0 left 16:46 FROGGS joined 16:47 zakharyas left, Hor|zon joined 16:52 ghostlines left, ugator left
timotimo japhb: give me a bit :) 16:53
yoleaux 15:41Z <japhb> timotimo: Can you make a perl6-bench mini-benchmark out of the AST benchmark you've been playing with?
15:46Z <japhb> timotimo: Actually, both before and after microoptimizations would be good to have, because that shows how much work the user is having to do that the compiler should be doing ... and how much better it performs with user assistance.
16:53 djanatyn left, djanatyn joined
japhb ++timotim 16:55
++timotimo
timotimo %)
i ran a car-accident-related errand today
that kind of took most of my time :\
being a supposedly grown-up is kind of complicated sometimes
16:55 ajr_ left 17:10 isBEKaml left 17:12 isBEKaml joined 17:15 Mso150 joined
raydiak g'morning 17:16
timotimo hello raydiak 17:20
TimToady timotimo: just got back from just such an errand himself; someone smacked our car while it parked in the driveway while we were out of town 17:21
17:21 ugator joined
TimToady *was 17:21
timotimo ah. well, at least it wasn't your fault in that case
i'm 100% to blame for what happened to my car. the other party was a concrete pillar, you see? 17:22
moritz a fast-moving concrete pillar :-)
japhb timotimo: Clearly it jumped out at you ....
TimToady or they installed it incorrectly
japhb TimToady: Why in the world was someone travelling into your driveway at sufficient velocity to continue on and hit your car? 17:23
TimToady TAWUI I suppose
(turning around while...) 17:24
17:24 guru joined
timotimo i should have asked the person at the shoppe what it would have cost if they just used gaffer tape 17:24
17:24 guru is now known as Guest37560, Guest37560 is now known as ajr_
timotimo and if that price could have been reduced if i supplied my own gaffer tape 17:24
tony-o lol
masak g'd'evening, #perl6 17:25
abraxxa hi masak! 17:26
17:27 isBEKaml left, FROGGS[mobile] left, cognominal left 17:28 cognominal joined, Akagi201 left
moritz timotimo: 5-to-6 updates are live now. Thank you! 17:29
timotimo no, thank *you*! 17:30
for writing that stuff in the first place 17:31
17:33 Mso150 left, kaleem joined, Mso150 joined, lizmat joined
lizmat is feeling a bit better 17:36
turns out I changed MySQL database storage engine from MyISAM to InnoDB a while ago 17:37
and had forgotten about it
when I accidentally did a upgrade to Yosemite on our live server, instead of our test server
17:38 [Sno] joined
lizmat and afterwards got a system that a. didn't run mysql anymore, and b. had a limited apache that could not handle the config it was given 17:38
I freaked out when I realized that the last successful image backup was in January
how time flies :-(
raydiak takes note, goes to back up his server 17:39
lizmat yup
please do
:-)
raydiak wow, my last was february... lizmat++ :) 17:40
lizmat my work machine has a daily image and an hourly time machine backup :-)
raydiak that'd be a bit overkill for most of my purposes, but at least sticking an e-mail to the effect of "run a backup, fool!" in a weekly cronjob might be wise 17:43
17:46 erkan joined, erkan left, erkan joined
japhb Also remember that unless you've done a successful restore, you haven't really done a backup. 17:48
woolfy is happy lizmat is a bit less stressed: most of the past hours were no fun
17:49 immortal left, FROGGS_ joined
lizmat is restoring the last image backup now on another MacMini (fine so far) and will then copy the necessary data files from the upgraded server 17:50
and then swap MacMini's
moritz hopes it works out fine for lizmat++ 17:51
japhb Ditto that
17:53 grettis left, FROGGS left, Sqirrel left
lizmat moritz japhb FROGGS_ : thank you for your kind words 17:54
17:54 Sqirrel joined, kaleem_ joined
lizmat ah well, moving our SDSL connection to a VDSL connection earlier today, worked out almost flawlessly 17:54
timotimo i really ought to freshen up my backups on my laptop and desktop system ... 17:56
17:56 pecastro left, kaleem left 18:04 immortal joined 18:06 erkan left 18:08 Mso150 left, Mso150 joined, itz left 18:10 kjs_ left
dalek kudo/nom: cb4a1b1 | (Elizabeth Mattijsen)++ | src/core/CompUnitRepo/Locally.pm:
Fix test breakage in S22-package-format
18:11
raydiak are there things I need to know about circular references in rakudo, other than .perl not working? 18:14
18:14 kurahaupo joined
raydiak e.g. will such things be garbage-collected appropriately, for one 18:15
[Coke] lizmat: still better support than I'm used to. :) 18:17
japhb raydiak: circular references that are not referenced externally should be garbage-collected, yes.
raydiak thanks japhb++, I guess that was my main concern, off the top of my head 18:18
18:18 kaleem_ left 18:29 Akagi201 joined
dalek rlito: 2df3521 | (Flavio S. Glock)++ | / (2 files):
Perlito5 - Dumper tweak
18:29
18:33 Akagi201 left 18:36 leont left
timotimo i can't find terribly many changes to list for last week :S 18:36
.o( i should have done more, damnit! )
PerlJam timotimo: it was release week. people were traveling. et cetera. :) 18:37
timotimo right, the release will be favourably mentioned, PerlJam++
do we already have a victim^Wvolunteer for this month's Star release? 18:43
PerlJam Isn't it almost always moritz anyway? ;) 18:44
18:45 abraxxa left
timotimo poor moritz ;( 18:45
18:46 abraxxa joined 18:47 abraxxa left
avuserow are things mostly in a good state? I'd be willing to give it another try if so :) 18:47
18:47 abraxxa joined
PerlJam avuserow++ (Even if just for the thought :) 18:47
FROGGS_ avuserow: take a look at the star-daily smokes done by [Coke]++ 18:49
moritz PerlJam: these days, it's mostly been FROGGS_++ who did the star releases 18:55
I helped preparing some, but my last "real" one is a few months back
PerlJam FROGGS++ moritz++ 18:56
vendethiel Any reason the awesome power we have in multis isn't available in other places? 18:58
18:58 eternaleye joined
vendethiel That's directly related to having a "match" 18:58
PerlJam vendethiel: for instance?
vendethiel PerlJam: given @a { when :(Bool where True, @b) { say 1 } } 18:59
lizmat afaik, 2014.11 is on my plate 19:00
19:00 baest left 19:01 baest joined 19:04 baest left 19:05 baest joined
japhb timotimo: It would be useful to have not just the committed changes, but the big discussions/decisions that happened as well, since that seems like an equal part of the "weekly news". OTOH, I know that's a lot of additional work, unless you happen to be taking notes, or we have some collaborative way for people to identify these big discussions for you. 19:06
masak "Sure, Python isn’t Perl 6, so you can’t add arbitrary syntax, but you can often find a clever way to get the job done." -- just found this gem in www.aaronsw.com/weblog/rewritingreddit 19:07
this is the ideal that we should be striving for. :)
(with macros and slangs)
19:08 baest left
colomon masak: wait, which one? 19:08
dalek nda: 85c6e01 | (Elizabeth Mattijsen)++ | ext/Shell__Command:
Bring Shell::Command up to date
19:09
19:10 baest joined 19:11 denis_boyun_ joined
PerlJam colomon: turning python into Perl6 obviously. 19:12
colomon :p
wait, read that backwards. 19:14
dalek kudo/nom: b130ad0 | (Elizabeth Mattijsen)++ | src/core/Perl.pm:
Make $*PERL initialization non-lazy

If it *is* lazy, the next commit will break precomp. I've traced this back to something being serialized when it shouldn't, or vice-versa (being in the serialized blob without there being a de-serializer). This actually generates an error similar to when we have lazy @*INC initialization.
19:15
kudo/nom: 737ec3c | (Elizabeth Mattijsen)++ | src/core/Str.pm:
Str.path marked DEPRECATED

If $*PERL is lazy, then this change will break precomp, and thus panda. 0decd7b | (Elizabeth Mattijsen)++ | src/core/IO/Path.pm: dir() now returns Str, as specced
19:16 baest left, baest joined, ghostlines joined 19:17 kjs_ joined, psch left 19:18 abraxxa left, kurahaupo_ joined 19:21 ptc_p6 joined
ast: 6396528 | (Elizabeth Mattijsen)++ | S32-io/dir.t:
Adapt dir() tests, because now Str is returned
19:24
19:24 darutoko left
vendethiel mutters something about implementing match using macros anyway 19:26
timotimo and the cat sayeth: vxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxi
vendethiel did he. 19:28
timotimo yes, he did 19:29
pmichaud m: ($_ * 2 if $_ %% 2 for 1..10).perl.say 19:30
camelia rakudo-moar 315ec6: OUTPUT«(4, 8, 12, 16, 20).list␤»
masak colomon: which ideal? that of being the language that people say "sure, [language X] isn't Perl 6, but [some sour grapes]" :) 19:31
pmichaud TimToady (and others): Having that output ((), 4, (), 8, (), 12, (), 16, (), 20) feels wrong to me.
it feels like the false cases of the 'if' ought to return something undefined
19:32 Ven joined
colomon masak++ 19:32
19:32 prammer joined, erkan joined, erkan left, erkan joined
timotimo pmichaud: FWIW, i find it pretty neat that i can have a map that can fall through the false branch of an if and throw values out 19:34
it kind of feels natural as in: you could have made that give multiple values instead just as well
as in:
19:34 immortal left
TimToady pmichaud: this is why I was wondering about something special to mark the absence of a value in list comprehensions 19:35
timotimo m: (($_ * 2, $_ * $_) if $_ %% 2 for 1..10).perl.say 19:36
camelia rakudo-moar 315ec6: OUTPUT«((4, 4), (8, 16), (12, 36), (16, 64), (20, 100)).list␤»
pmichaud timotimo: I agree with the neatness of throwing values out. I'm just thinking that the values to be thrown out ought to be undefined somehow.
moritz Nil but defined :-) 19:37
pmichaud as opposed to defined.
moritz erm
() but undef
timotimo oh, you mean () is a defined value?
pmichaud m: say ().defined
camelia rakudo-moar 315ec6: OUTPUT«True␤»
pmichaud of course it's a defined value. :)
timotimo mhm
TimToady that's what I meant the other day by an Empty primitive
pmichaud yeah, I'm warming to the idea of Empty
I wish I had a better intuition of what "Nil" actually means. 19:38
TimToady but maybe we should be thinking harder about not producing the empty value in the first place
PerlJam Empty == Nil but undefined?
TimToady instead of weeding it out after the fact
pmichaud PerlJam: Nil is already undefined.
m: say Nil.defined
camelia rakudo-moar 315ec6: OUTPUT«False␤»
PerlJam oh, I'm still thinking of () and Nil as synonyms.
moritz sort of like () with the OKNess bit set to 0? 19:39
pmichaud Haven't been synonyms since 2010.
19:39 fhelmberger left
PerlJam Shows you how much I use them :) 19:39
pmichaud well, to me it relates to "what does a function return when it doesn't return anything?"
m: sub xyz() { return; }; say xyz().WHAT 19:40
camelia rakudo-moar 315ec6: OUTPUT«Nil␤»
masak m: sub foo {}; say foo.WHAT
camelia rakudo-moar 315ec6: OUTPUT«Nil␤»
pmichaud currently we have them returning 'Nil'. It seems to me that the false-side of an 'if' modifier ought to do the same
masak m: my %h; say %h<foo>.WHAT
camelia rakudo-moar 315ec6: OUTPUT«(Any)␤»
TimToady maybe statement modifier loops just establish a context inside of which any else-less conditional implies an else { next } 19:42
19:42 Ven left
TimToady then we never put the value into the list in the first palce 19:42
pmichaud is it only in statement modifier loops that we have this issue?
moritz nope
pmichaud (for 1..10 { $_ * 2 if $_ %% 2 }).perl.say
TimToady map too 19:43
pmichaud m: (for 1..10 { $_ * 2 if $_ %% 2 }).perl.say
camelia rakudo-moar 315ec6: OUTPUT«(4, 8, 12, 16, 20).list␤»
pmichaud is it just loops in general?
moritz my @list =^Hmdo for 1..10 { $_ * 2 if $_ %% 2 }}
pmichaud I wouldn't have a problem with saying that LoopIters throw out any Nils they encounter (unless itemized)
as opposed to saying that Lists always do it.
TimToady earlier seems better 19:44
moritz m: say do for 1..10 { $_ * 2 if $_ %% 2 }}
camelia rakudo-moar 315ec6: OUTPUT«===SORRY!=== Error while compiling /tmp/NdGFtCs0EU␤Unexpected closing bracket␤at /tmp/NdGFtCs0EU:1␤------> say do for 1..10 { $_ * 2 if $_ %% 2 }⏏}␤»
moritz m: say do for 1..10 { $_ * 2 if $_ %% 2 }
camelia rakudo-moar 315ec6: OUTPUT«4 8 12 16 20␤»
19:44 ggherdov left, isacloud left, Ven joined
pmichaud perhaps the distinction we want is Nil.item versus Nil 19:45
i.e., Nil.item doesn't flatten/disappear
19:45 isacloud______ joined 19:46 rindolf left
pmichaud although that's kinda weird since it's not really possible to stick Nil into a scalar ATM 19:46
timotimo posted a pretty short post on the weekly
pmichaud (because it has another overloaded meaning)
19:46 rindolf joined, rindolf left, kurahaupo_ left
pmichaud using Nil.item to mean "don't discard this" feels good somehow, though. 19:46
19:47 ggherdov_ joined
pmichaud fwiw, using $foo = Nil to restore Foo to its uninitialized state has always felt a little off-kilter to me. 19:47
timotimo it seems like rakudo isn't building properly on moarvm master at the moment :o 19:48
pmichaud retyping that....
fwiw, using $foo = Nil to restore $foo to its uninitialized state has always felt a little off-kilter to me. 19:49
Ven m: role Pet[::ID;; $tag] { has ID $.collar .= new($tag); }
camelia rakudo-moar 315ec6: OUTPUT«===SORRY!=== Error while compiling /tmp/AeAfCM6ZzV␤Cannot use .= to initialize an attribute␤at /tmp/AeAfCM6ZzV:1␤------> ;; $tag] { has ID $.collar .= new($tag)⏏; }␤»
Ven m: role Pet[::ID;; $tag] { has ID $.collar = new($tag); }
camelia rakudo-moar 315ec6: OUTPUT«===SORRY!=== Error while compiling /tmp/NjX3PfToru␤Undeclared routine:␤ new used at line 1 (in Perl 6 please use method call syntax instead)␤␤»
Ven m: role Pet[::ID;; $tag] { has ID $.collar = Pet.new($tag); }; 1
camelia ( no output )
Ven m: role Pet[::ID;; $tag] { has ID $.collar = Pet.new($tag); }; say Pet[Int, 5]; say "alive"; say Pet[Str, 5]
camelia rakudo-moar 315ec6: OUTPUT«(Pet[Int,Int])␤alive␤(Pet[Str,Int])␤»
Ven oh yea. attr constraints nyi 19:50
masak pmichaud: completely agree about the off-kilter.
pmichaud: the fewer such special cases, the better.
m: class Dog {}; my Dog $dog; say $dog.defined; $dog .= new; say $dog.defined; $dog = Dog; say $dog.defined 19:51
camelia rakudo-moar 315ec6: OUTPUT«False␤True␤False␤»
moritz rakudo-p spectest with parrot 6.9.0 gives me *lots* of errors 19:54
bartolin moritz: I suspect most of those errors are segfaults (maybe related to gc): perl Configure.pl --backends=ALL 19:55
erm: irclog.perlgeek.de/perl6/2014-10-26#i_9568218
s/perl.*$//
pmichaud what's the canonical example of where we don't want Nil to automatically disappear from lists? 19:57
bartolin moritz: I tried to run several of the failing test files manually and always got a segfault 19:58
20:01 ghostlines left, Rounin left 20:04 rindolf joined
moritz so looks like parrot b0rk 20:04
bartolin rurban++ is hunting GC bugs currently: irclog.perlgeek.de/parrot/2014-10-27#i_9569886 20:06
20:06 ptc_p61 joined 20:08 ptc_p6 left, ecocode joined 20:11 itz joined, rindolf left 20:12 rindolf joined, kurahaupo is now known as kurahaupo_away
timotimo can other people build a rakudo-moar with all latest masters from scratch? 20:13
i get segfaults during the core setting :(
20:13 ecocode left
itz perl Configure.pl --backends=moar --gen-moar=master --moar-option="--enable-jit" --gen-nqp 20:14
=master ?
without the line break of course?
timotimo no need for --enable-jit any more
itz oh yeah its default
timotimo hello itz :)
that should do it for you
itz works for me 20:21
[Coke] takes longer for me to git via http proxy than it does to build rakudo.
itz what about the parrot one? 20:22
moritz parrot is b0rked.
20:22 ecocode joined
TimToady pmichaud: when you're going to bind it positionally, and don't want the wrong positional to show up 20:23
20:23 rindolf left 20:24 rindolf joined
pmichaud TimToady: I think I'd need a code example. 20:25
20:25 ecocode left
TimToady m: sub foo ($a,$b,$c?) { say $c.defined }; foo 'A', Nil, 'C' # currently says True 20:28
camelia rakudo-moar 315ec6: OUTPUT«True␤»
TimToady need to preserve the arg vs item distinction
if the Nil disappears too early, we're hosed on that 20:29
pmichaud you're saying that should've been false?
lizmat but isn't that because Nil currently isn't implemented correctly?
TimToady no, it should say True
lizmat ah, ok *phew* 20:30
pmichaud lizmat: I'm working on nailing down the post-GLR semantics for Nil
20:30 Akagi201 joined
TimToady if 'A', Nil, 'C' turns into 'A', 'C' too soon, then foo never binds $c 20:30
lizmat will remove fingers between hammer and wood
pmichaud TimToady: well, I never think of that as being a list in the first place. 20:31
*List
moritz with moar master, I get a test failure in t/spec/S32-io/dir.t
TimToady right, but we're getting rid of parcels
pmichaud that's not even a Parcel in the current implementation
TimToady which are the feedstock for Captures
lizmat moritz: even after update of roast ?
and most recent rakudo ? 20:32
moritz lizmat: ah no, my rakudo was out of date
lizmat *phew*
moritz waitwaitwait, why does dir return strs now?
lizmat because it is in the spec ? 20:33
moritz then I'm going to change the spec
lizmat so we don't need to make an .IO for each dir entry when you don't need it ?
pmichaud (dir) yeah, we already went through the pain of converting from dir returning Str to IO... why are we changing back?
masak .oO( ...are doomed to repeat it... )
lizmat was not aware that this had gone the other way before 20:34
FROGGS_ .oO( ...are doomed to repeat it... )
moritz lizmat: because it's very convenient to be able to do all sorts of useful file ops on them
TimToady
.oO( ...are doomed to have it repeated to them... )
moritz lizmat: like checking .d, and .dir()ing them again
masak .oO( DOOMED! )
pmichaud ChangeLog from 2012.07: + Changed &dir to return IO::Path objects, not strings
TimToady and how much slower are they? 20:35
lizmat well, not as slow as they used to be
moritz github.com/perl6/doc/blob/master/htmlify.p6#L88
20:35 Akagi201 left
TimToady
.oO( use Dir::Tiny; )
20:35
pmichaud
.oO( The Synopses are not "spec". )
20:36
lizmat well, I'm glad someone is paying attention to what I'm writing :-(
lizmat is having a bad day and decides to get some rest 20:37
20:37 lizmat left
moritz lizmat: it's not the first time I mentioned it: irclog.perlgeek.de/perl6/2014-09-09#i_9322300 20:37
20:39 rindolf left, rindolf joined
pmichaud TimToady: (Nil) So, you're saying we'd need to make sure that something like foo |('A', Nil, 'C') doesn't lose the Nil 20:41
TimToady which are the feedstock for Captures
oops
that one, is, of course, more arguable both ways
pmichaud well, the case of foo 'A', Nil, 'C' already doesn't involve a Parcel or List, so I'm having trouble with it as a useful example. 20:42
TimToady stupid touchpad autoscrolls my history too easily
pmichaud If we have it so that loops remove any bare Nils they encounter, does that get what we're after? 20:44
TimToady could be
and a loop optimizer might be good enough to turn that into 'else { next }' 20:45
pmichaud maybe
20:45 kurahaupo joined
pmichaud I'm not sure that optimizes. 20:45
20:45 kurahaupo left
TimToady whether it's worth it depends on the bookkeeping costs, either way 20:46
pmichaud well, I'd feel really good if I knew that { next } resulted in a Nil, just like { return } does now.
TimToady eh? if you next, you don't need a Nil at all 20:47
pmichaud I mean conceptually. 20:48
TimToady you bypass whatever code stores the value
pmichaud ummmm, not exactly.
the canonical example is @list.map( &block )
well, maybe not, since next throws an exception 20:49
anyway
(I'm still rusty.)
20:50 ajr_ left
TimToady but yes, conceptually I don't see a big problem with value-producing lists throwing away nulls, except maybe for internal ones processing argument lists 20:50
pmichaud what I'm trying to get at is this: at some level, "next" in a loop serves the same purpose as "return" does in a Routine -- it exits the current block and optionally sends a value to the caller. It'd be conceptually nice if 'next' and 'return' with no argument sent back the same sort of thing :) 20:52
20:52 kurahaupo_away left, kaare_ left
pmichaud moving away from 'next', though: 20:53
m: sub xyz($_) { if $_ == 3 { return }; $_ }; (for 1..5 { xyz($_) }).perl.say 20:54
camelia rakudo-moar 315ec6: OUTPUT«(1, 2, 4, 5).list␤»
pmichaud would we expect the Nil to be preserved there or discarded?
TimToady they could always return () if they mean the other thing, I guess 20:55
pmichaud yes.
or even perhaps:
TimToady so I think I'm okay with it disappearing there 20:56
pmichaud m: sub xyz($_) { if $_ == 3 { return Nil.item }; $_ }; (for 1..5 { xyz($_) }).perl.say
20:56 rindolf left
camelia rakudo-moar 315ec6: OUTPUT«(1, 2, Nil, 4, 5).list␤» 20:56
pmichaud or even:
m: sub xyz($_) { if $_ == 3 { return Nil.item }; $_ }; (for 1..5 { xyz($_).item }).perl.say
camelia rakudo-moar 315ec6: OUTPUT«(1, 2, Nil, 4, 5).list␤»
pmichaud m: sub xyz($_) { if $_ == 3 { return }; $_ }; (for 1..5 { xyz($_).item }).perl.say
20:56 rindolf joined
camelia rakudo-moar 315ec6: OUTPUT«(1, 2, Nil, 4, 5).list␤» 20:56
TimToady seems pretty okay to me 20:57
pmichaud I think we're in agreement that we want something that means "nothing". In the case of a "return" with no argument, we've decided that's Nil, and it's undefined. 20:59
the undefinedness is a really useful property.
to me, that rules out using () to mean "nothing", because () is defined.
(and I really don't think I want to head down the path of an empty List being undefined. :) 21:00
TimToady well, it's 0 values as a defined concept
pmichaud so, extrapolating a bit further, would I expect
for @values { foo($_) }; 21:01
to always produce a list with the same number of elements as @values ?
TimToady well, not in sink context :)
if you want that, you should return () instead of Nil from foo 21:02
pmichaud okay then: my @a = (for @values { foo($_) };)
TimToady but certainly we've never guaranteed that a flat map returns the same number of values 21:04
pmichaud so, if the block of a 'for' returns Nil, that's an indication to not record a value for that iteration, and if you wanted to preserve the Nil you should do so using .item or something like that.
or ()
TimToady that seems good 21:05
21:05 Mso150_f joined, Mso150 left
PerlJam pmichaud: so ... @a above does not preserve the Nils that foo may return? 21:05
pmichaud pmichaud: not as I wrote it, no.
if I wanted to keep the Nils, I'd do:
my @a = (for @values { foo($_).item }) 21:06
EXCEPT
that assigning a Nil into @a would cause that element to revert to its base type object
PerlJam my @a := (for @values { foo($_) }); # would also preserve the Nil too?
pmichaud in that case, no, since the Nil's get stripped by the 'for'
(on output) 21:07
PerlJam ok.
pmichaud but
my @a := (for @values { foo($_).item } );
would cause @a to have the same number of elements as @values (barring any funny control exceptions in &foo) 21:08
PerlJam And map's behavior is identical to for?
pmichaud on the output side, yes.
TimToady except method form doesn't flatten input 21:09
pmichaud this would be for all looping constructs
my @a := (while $i < 10 { foo($i++).item })
raydiak we return () when we *do* want the Nil preserved in the output list?
pmichaud raydiak: you return () when you want an empty list in the output
("Nil" ne "empty list") 21:10
raydiak okay, I was just lost momentarily then
21:10 rindolf left
pmichaud I kind of like this outcome; it prevents us from having to introduce an Empty, and puts the onus of removing items on the looping constructs instead of the List interpolators 21:12
21:12 woshty left
pmichaud more precisely, it puts the onus of removing the Nils on the things that are generating them in the first place :) 21:12
21:12 ptc_p61 left
pmichaud so that "removing" is really more "skipping" 21:13
21:13 woshty joined
pmichaud or "omitting" 21:13
or "never generating" :)
never generating is way more efficient than "generate and remove" :) 21:14
21:14 molaf left 21:17 gfldex joined 21:19 kjs_ left
dalek href="https://cpandatesters.perl6.org:">cpandatesters.perl6.org: 6d22454 | (Tobias Leich)++ | app.pl:
add shebang and `use v6`
21:21
21:21 Ven left
sergot FROGGS++ 21:26
FROGGS_ :o)
21:26 FROGGS_ is now known as FROGGS
FROGGS though, I need to implement nqp::closefhi for parrot and jvm before I can merge my panda branch, which implements the reporter client 21:26
sergot hi o/ btw :) 21:27
FROGGS hi sergot
pmichaud afk for a while, more thinking to do
dalek nda/reporter: 9497994 | (Elizabeth Mattijsen)++ | / (8 files):
Fix 2014.10 deprecations
21:29
nda/reporter: c42053e | (Elizabeth Mattijsen)++ | ext/ (3 files):
Bring submodules up to date
nda/reporter: 85c6e01 | (Elizabeth Mattijsen)++ | ext/Shell__Command:
Bring Shell::Command up to date
nda/reporter: 45be099 | (Tobias Leich)++ | / (11 files):
Merge branch 'master' of github.com:tadzik/panda into reporter
21:35 Tekk_` is now known as Tekk_ 21:39 fhelmberger joined
dalek ast: ff58ad1 | usev6++ | S04-declarations/state.t:
Add test (fudged 'skip') for RT #115614
21:39
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...?id=115614
21:40 MilkmanDan left, MilkmanDan joined 21:42 kjs_ joined 21:43 fhelmberger left
tony-o FROGGS++ 21:44
FROGGS thanks :o) 21:45
21:45 erkan left 21:46 kjs_ left, Akagi201 joined 21:51 Akagi201 left 21:57 virtualsue left 22:15 Rounin joined 22:18 ugator left 22:23 Mso150_f left
dalek ast: 2c34a36 | (Elizabeth Mattijsen)++ | S19-command-line/dash-e.t:
-e as a stopper appears to work
22:29
22:33 colomon left 22:34 colomon joined 22:35 bronco_creek joined
bronco_creek I have a question about reading from a file. When I use, for example: for $in.lines -> $line {say $line} is see a "?" prepended to the first line. What's up with that? 22:38
FROGGS that sounds like a replacement character.. 22:40
geekosaur windows by any chance? a utf8 file with a byte order mark at the start (like windows text editors like to prepend), sent to a non-utf8 display, would do that
FROGGS bronco_creek: does that file start with a BOM? 22:41
bronco_creek Yes, Windows.
FROGGS bronco_creek: you've got notepad++ by any chance? 22:42
bronco_creek I'm using Padre.
FROGGS ahh, hmmm
dunno if that allows you to save that file without a BOM 22:43
bronco_creek The file I'm trying to process is xml output from another program, but the leading "?" does not show up in Padre.
timotimo do you have some kind of hex editor? 22:44
now the question becomes
do we want to support byte order marks in rakudo?
geekosaur I think you'll need to; notepad in at least some versions of windows will insist on prepending a BOM
some other text editors as well (and there's a handful of programs, including some from Microsoft, that depend on it!) 22:45
even though you're not supposed to use a BOM with UTF-8
bronco_creek I tried to strip the "?" with ~~ s/^\?// but that did not do the trick.
geekosaur yes, it's getting mapped to a replacement character on output; it's not actually a question mark 22:46
timotimo that's right
how about opening the file as binary and outputting the first few bytes as hexadecimals and looking at what exactly it is
geekosaur actually on display, not on output
bronco_creek Hmm. Drove me nut trying to figure out why all of my xml files were invalid when I tried to parse with XML or XML::Parser::Tiny 22:47
timotimo damn, that's annoying 22:48
very sorry to hear that, bronco_creek :(
bronco_creek timotimo: I don't think I have an editor that will let me look at the hex. Maybe I need to install one. 22:49
timotimo try code like this: 22:50
m: open("foobar.txt", :bin).perl.say
camelia rakudo-moar 315ec6: OUTPUT«open is disallowed in restricted setting␤ in sub restricted at src/RESTRICTED.setting:1␤ in sub open at src/RESTRICTED.setting:5␤ in block <unit> at /tmp/GKZm7GJt9Q:1␤␤»
22:55 spider-mario left
bronco_creek timotimo: I tired: $in = open("$file", :bin); $in.perl.say; and got IO::Handle.new etc. 22:56
Looks like user error on my part.
22:56 treehug88 left
geekosaur slightly wrong instructions actually 22:56
it showed you what you were told to get, which is the handle :) you actually need to read a line and dump it. 22:57
timotimo oh, sorry, you need to .read(5) or so from it
bronco_creek geekosaur, timotimo: thanks, I'll try that 22:58
FROGGS we should open an rt ticket for that 22:59
timotimo yeah
FROGGS will do that now 23:00
bronco_creek: you're using rakudo on moarvm? 23:01
bronco_creek geekosaur, timotimo: Nope, even with the :bin in the open statement, I still get the leading BOM 23:02
timotimo yeah, of course you would
but at least we know it's a BOM now, right? 23:03
bartolin for the records: I just had clean spectests for latest rakudo on all backends (running on debian 7)
bronco_creek timotimo: I'll take you word for it. I'm still new to perl6.
timotimo what exactly is the beginning of the output, then?
23:03 denis_boyun_ left
FROGGS RT #123070 - rakudo should strip a BOM, but it does not 23:04
synopsebot Link: rt.perl.org/rt3//Public/Bug/Displa...?id=123070
bronco_creek timotimo: I get: ?<?xml version="1.0" encoding="utf-8"?>
FROGGS bronco_creek: hopefully we will have a fix for that within a few days
geekosaur hm 23:05
you did not use .perl i take it?
FROGGS bronco_creek: can you also pass :bin to the read() all?
call*
geekosaur that should have given you a visualization of a Buf
timotimo hold on, :bin may be wrong or something?
i don't have a local perl6 at this very moment >_<
FROGGS hold on
geekosaur pyanfar:60977 Z$ ./install/bin/perl6 -e 'open("/dev/random", :bin).read(5).perl.say'
Buf[uint8].new(102, 234, 161, 85, 205)
bronco_creek Thanks all. This isn't a mission-critical issue for me. Just playing around trying to get familiar with the language.
TimToady ain't we all :) 23:06
FROGGS *g*
bronco_creek Still a bit of a moving target, but awesome progress lately. perl6++ 23:07
timotimo bronco_creek: i'm kind of wondering where those xml files have come from such that they started with a BOM
23:08 leont joined
FROGGS timotimo: I bet that parrot and jvm get it right though 23:08
timotimo so far i thought BOM was deprecated, but unicode still allows it (but does not recommend)
bronco_creek They are DITA/XML docs that I extracted from my CMS at work.
FROGGS and I also think it is a three lines patch to handle it
geekosaur timotimo, it's not supposed to be used with anything but utf16 and utf32. but many windows programs use it to detect utf8 vs. a windows code page encoding 23:09
this is noncompliant (BOM is *never* valid in UTF8) but windows "standard" :/
23:10 FROGGS left
geekosaur so de facto for windows interop you need to deal with a BOM at the start of a text file 23:11
23:12 timotimo left
bronco_creek Bye for now. I'm on the road for a week - flying to northern CA then driving back to CO. Thanks again. 23:12
geekosaur (microsoft has officially deprecated its use in windows utf8... but some of the programs that rely on it to detect utf8 are part of visual studio)
23:13 timotimo joined
bronco_creek One more thing. On backlogging I see I missed FROGGS' question. I am using the moarvm version from last month. Hope that helps. 23:19
23:22 virtualsue joined 23:24 virtualsue left 23:25 bronco_creek left 23:33 chenryn joined 23:35 Akagi201 joined 23:37 gfldex left 23:40 Akagi201 left 23:42 zjmarlow joined 23:43 zjmarlow left