»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend! | tinyurl.com/p6contest
Set by moritz_ on 28 December 2010.
colomon Ack, he appears to have his own testing module. 00:01
oh, which he also sent me. 00:02
patrickas Is there a reason why TheDamian doesn't release more all the crazy delicous stuff he has? 00:03
colomon he wants to get it all right first, I think.
00:03 yarrack_logging joined
masak yarrack_logging: hi. are you a lumberjack? 00:05
patrickas Oh! I wanted to retype a module of his from blurry screenshots of a video of a presentation :-)
I couldn't careless how all right it was 00:06
I think it was "the computation in negative time" talk
colomon I haven't seen that one.
sorear is that anything like computation with supertasks? 00:07
colomon I dunno, I can kind of understand why he wants to take his time.
But dang it, I want to use his modules. Lots of us are writing Perl 6 now. 00:08
00:09 felliott left
patrickas sorear: I don't think so, iirc a bit more on the humorous side, but I seem to be unable to find the video again 00:11
00:12 bacek joined
colomon hm. 00:13
If I want to get these into ecosystem with minimal modifications, I need to tweak ufo.
But I'm currently using an obsolete hacked version of ufo locally.
Sigh.
patrickas 'night all 00:14
00:15 patrickas left
masak why would ufo tweaks be needed? 00:16
tadzik colomon: tweak module::tools :) 00:17
colomon Well, for one thing, "prove" doesn't work on my machine the way it does in ufo.
tadzik: how is module::tools related? 00:18
tadzik colomon: the mbuild tool is basically ufo w/o a make
colomon you mean, it replaces make?
tadzik sorta 00:19
colomon masak: that's the local tweak. The tweak we need for Damian's modules is to allow "lib6" as well as "lib". (Unless you've already done that and my version is just behind.) 00:21
masak colomon: no, but it sounds like a good addition. 00:23
00:24 bacek left
colomon oh, ack, not as easy as I was hoping. 00:28
00:29 masonkramer left, masonkramer joined 00:30 aindilis left
colomon rakudo: my $a = "lib6/Hello/Kitty.pm"; $a.subst(/^'lib(6?)/'/, ''); say $a 00:33
p6eval rakudo a98c69: OUTPUT«lib6/Hello/Kitty.pm␤»
colomon rakudo: my $a = "lib6/Hello/Kitty.pm"; $a.subst(/^'lib[6]?/'/, ''); say $a
p6eval rakudo a98c69: OUTPUT«lib6/Hello/Kitty.pm␤»
colomon rakudo: my $a = "lib6/Hello/Kitty.pm"; $a.subst(/^'lib"6"?/'/, ''); say $a
p6eval rakudo a98c69: OUTPUT«lib6/Hello/Kitty.pm␤»
colomon rakudo: my $a = "lib6/Hello/Kitty.pm"; $a.subst(/^"lib""6"?"/'/, ''); say $a 00:34
p6eval rakudo a98c69: OUTPUT«===SORRY!===␤Unable to parse postcircumfix:sym<( )>, couldn't find final ')' at line 22␤»
lue I'm starting to remember why I gave up before I started on GCC the first time... 00:36
colomon my %module-to-path = @modules Z=> @module-files; 00:37
00:38 f00li5h joined
colomon got it, I think. 00:45
00:48 kanishka left 00:57 qw3rty joined 00:58 rgrau joined
colomon tadzik: so, what's the module::tools 01:01
?
masak I just put my blogging software on github. it's 250 lines of Perl 6 code: github.com/masak/psyde/blob/master/psyde 01:02
if anyone actually wants to deploy psyde, I have a .zip file where I actually pulled away the specific parts. 01:03
colomon masak++ 01:04
tadzik: found it.
masak I'm currently porting Markdown to Niecza Perl 6. when that's done, my blog will be running pure Perl 6. 01:05
lue is interested
wolverian heh. the required named param style looks kinda smalltalkish.
lue needs a website that can run Perl 6
masak wolverian: that was the intent.
wolverian: assuming you're talking about &nonexistent-or-older 01:06
lue: I think I have your email address. I can send the .zip file over if you want.
lue Sure. Only then I'd have to google "copy files to remote server through ssh" :) 01:08
[alternatively, try compiling Rakudo again]
masak lue: there's scp, which sounds like it does what you want.
colomon tadzik: do I need to use ufo to install Module::Tools? I'm confused here.... 01:09
masak oh, it's ggoebel++ who is already running a psyde instance. :)
lue which bot lets me see the last time someone was on the channel? 01:10
wolverian masak: yes.
masak wolverian: that was directly inspired by the way Smalltalk does named param passing. I like it. 01:11
buubot: seen lue?
buubot masak: I don't think I've seen lue?.
masak buubot: seen lue
buubot masak: I last saw lue saying "perlesque" at Sun May 9 02:27:52 2010 Z.
masak buubot: er. 01:12
buubot masak: Couldn't match input.
lue I'm pretty sure I've never had a reason to say perlesque, much less on 9 May.
wolverian masak: can can you say :from [...] instead of :from[...]? 01:13
or than, I forgot already what the name was. :)
lue aloha: seen bkeeler
aloha lue: Sorry, I haven't seen bkeeler.
lue buubot: seen bkeeler 01:14
buubot lue: I last saw bkeeler saying "bkeeler" at Tue Mar 30 23:58:23 2010 Z.
masak wolverian: no, that'd constitute a two-terms-in-a-row syntax error. 01:15
wolverian ok.
01:15 yarrack_logging left, yarrack_logging joined
masak std: :than\ ["OH HAI"] 01:15
p6eval std 625303c: OUTPUT«ok 00:01 118m␤»
masak that works, though. 01:16
01:18 Fuld left
colomon 's momentum for messing around with other people's code tonight is pretty much gone... 01:21
masak colomon++ 01:22
colomon Not clear if I've accomplished anything useful at all today, other than sending that not carefully tested patch to you, and getting Damian to send me his code. 01:23
Will be out of vicodin tomorrow, and need to get $work done on Monday, I fear. 01:24
Think I might resolve to try to have a Perl 6 day every week this year. 01:25
masak the patch you sent is now applied and pushed.
01:26 jarfhy left
colomon ooo, thank you. 01:26
masak thank *you* :) 01:27
oh wait. the push failed.
trying again.
01:28 jarfhy joined
masak would it be OK if I removed &module-name-to-path, rather than just commenting it out? 01:29
I'm slightly allergic to commented-out code.
colomon Of course. 01:30
I'm just slightly allergic to just deleting other people's code.
masak next time, feel free to delete it. :) 01:31
colomon Also, remember that I programmed for about 20 years before I had routine access to a version control system.
Paranoid habits die hard.
masak I'm very unsentimental about dead code.
they do indeed. 01:32
ruthlessly removing code instead of commenting it out is definitely a habit I picked up after getting into version control systems.
01:33 lukas___ left
colomon Usually I comment stuff out and then a week or two later I delete the comments. 01:34
masak my rule-of-thumb nowadays is that commented-out code is either (1) wrong and should be deleted at sight, or (2) equipped with a one-sentence comment explaining why it shouldn't be removed.
colomon Every now and then I actually manage to lose something despite version control.
masak last time I lost something was when I was robbed and my computer was lost. that must've been 2 years ago. I lost a day's worth of work on Druid. 01:35
colomon I actually lost something today. 01:36
Switched too quickly from my version of something to moritz_++'s, lost about ten minutes of work.
I must say I'm very happy to have a time capsule full time on my LAN, so that I have routine hourly backups going on in the background. 01:37
masak wolverian: you could also write :than(@these) or something. 01:38
how will we make Perl 6 great in 2011? what should be our focus, as a community? 01:39
colomon checks on his off-site backup, to make sure his main svn repository is still being backed up each day.
masak: I think we need to ramp Perl 6 up to strength. 01:40
In particular, getting the module ecosystem up to speed.
masak yes. 01:41
I've been a bit negligent of pls, I fear.
colomon And the two implementations converging in terms of speed and features.
masak that'll be nice.
colomon We've all been kind of negligent, I think. I dunno, feels like I've been pretty distracted since August or so.
masak yeah, apart from the Advent Calendar, we've been in a slump lately... 01:42
pmichaud almost entirely distracted, jnthn largely distracted...
colomon sorear aside, I think. He's working miracles.
masak but we had a similar (though shorter) slump around New Year's last year.
yes, sorear++ makes us all look like lazy bums. :P 01:43
I want to reward him by paying attention to his compiler.
lue I would like to help with creating a module ecosystem. I think it's because I have a deep-rooted displeasure towards how it works. 01:44
colomon I keep on meaning to try to install it again.
masak I think the Next Big Thing for the module ecosystem is Emmentaler. but it has pls as a prerequisite, so I need to finish pls first. 01:45
lue looks up Emmentaler
colomon module::tools seems interesting. 01:46
looked like bits of ufo got into it....
masak lue: strangelyconsistent.org/blog/cheese-speleology 01:47
colomon masak: by the way, I've got a port of some C++ suffix tree code I found on the net done but not totally debugged for p5. ;)
masak that reminds me, I should send out an email to all the contestants. 01:48
01:51 qiuhw joined 01:54 qw3rty left, icwiener left
lue How do I mark a test todo or skip? [ I currently just comment out the offending line, but that's not a good solution. ] 01:57
01:58 lao joined
lao P5> @x=( "tom 87 98 90", "bob 78 98 90", "tim 88 98 84", "jon 78 87 90"); 01:59
P5> @x=sort { 02:00
@af = (split /\s+/,$a)[1..3];
@bf = (split /\s+/,$b)[1..3];
$af[2] <=> $bf[2] || $af[1] <=> $bf[1] || $af[0] <=> $bf[0]
} @x;
colomon masak: Getting organizational, it seems like we ought to spend the next couple of months worrying about meta issues with modules: Installing them, testing them, making it easy to make modules which interface with other languages, etc. 02:01
lao I try to convert P5 code to P6 for sort. Need help.
masak lao: please don't do that.
lao: use a pastebin of some sort :)
rakudo: my @x = "tom 87 98 90", "bob 78 98 90", "tim 88 98 84", "jon 78 87 90"; @x .= sort( { my @af = $^a.words; my @bf = $^b.words; @af[2] <=> @bf[2] || @af[1] <=> @bf[1] || @af[0] <=> @bf[0] }); say @x.perl 02:03
p6eval rakudo a98c69: OUTPUT«["jon 78 87 90", "bob 78 98 90", "tom 87 98 90", "tim 88 98 84"]␤»
masak colomon: that's been the hidden challenge all along. 02:04
colomon: thing is, no-one takes on solving those problems voluntarily. they're no fun, not really. 02:05
colomon: so we need to grow into needing to solve them.
colomon I dunno about hidden. But it's been something we needed to do, and we ought to get our acts together. (And I do mean "we", not just you.)
masak right.
my point is that it might actually have been too early thus far.
colomon I reckon that's all irrelevant. The point is, we really should get them done now. 02:07
afk # gotta go back home now. 02:08
happy new year!
02:08 colomon left
masak waves after the departing colomon++ 02:09
02:15 awwaiid left 02:16 redicaps joined
Tene rakudo: my @a = 1,1,2; my @b = 1,2,1; say (@a Z<=> @b).perl 02:18
p6eval rakudo a98c69: OUTPUT«(0, -1, 1)␤»
Tene rakudo: my $a = '1 1 2'; my @b = '1 2 1'; say [||] ($a.words Z<=> $b.words)[2,1,0] 02:20
p6eval rakudo a98c69: OUTPUT«===SORRY!===␤Symbol '$b' not predeclared in <anonymous> (/tmp/MPJ2UcKxaT:22)␤»
masak rakudo: my @a = 1, 1, 2; my @b = 1, 2, 1; say (@a «<=>» @b).perl
Tene rakudo: my $a = '1 1 2'; my $b = '1 2 1'; say [||] ($a.words Z<=> $b.words)[2,1,0]
p6eval rakudo a98c69: OUTPUT«[0, -1, 1]␤»
rakudo a98c69: OUTPUT«1␤»
Tene That looks a bit more expressive than your long version, fwiw. :) 02:21
Huh. I'd never noticed the similarity between Z and «» before. 02:24
What *is* the difference, actually?
masak lazy and hyper, respectively. 02:25
wolverian masak: or than => @these? 02:26
masak wolverian: oh, right. yes.
wolverian: I never use that syntax myself :) I had forgotten about it. 02:27
Tene rakudo: my @x = "tom 87 98 90", "bob 78 98 90", "tim 88 98 84", "jon 78 87 90"; @x .= sort( { [||] ($^a.words Z<=> $^b.words)[2,1,0] } ); say @x.perl 02:29
p6eval rakudo a98c69: OUTPUT«["jon 78 87 90", "bob 78 98 90", "tom 87 98 90", "tim 88 98 84"]␤»
masak Tene++ 02:31
Tene That's still splitting into words repeatedly, so not quite optimal. 02:32
lue afk everyone! o/ 02:34
masak afk, lue \o 02:35
02:38 colomon joined 02:42 rgrau left
masak good night, zebras. 02:45
colomon night! 02:46
02:46 masak left 02:47 florz left, dukeleto left, felliott joined 02:48 dukeleto joined 02:55 lukas___ joined 02:59 florz joined 03:03 jhuni left 03:11 lukas___ left
lao masak, Tene: Thanks. I wrould expect the result of sorting 3rd,3nd,1st field order: tim 88 98 84 jon 78 87 90 bob 78 98 90tom 87 98 90 03:20
03:25 alester joined
lao ok I get it. I miss the correct index [3] [2] [1] for sorting. 03:37
lopnor happy new year! 03:58
how can i declear "local $foo" in perl6? 04:00
04:06 alester left 04:07 alester joined, am0c left, alester left 04:26 dukeleto left 04:27 dukeleto joined 04:28 alester joined
sorear lopnor: temp $foo 04:33
lopnor oo!
sorear it's also possible to do some of the same things with my $*foo
which is better supported
lopnor I wanted to modify some environmental variables inside of method, 04:34
i thought i should use method do_something ($foo is copy), 04:35
which is better?
sorear lopnor: not sure how is copy would work. 04:51
niecza: my $*foo = "Hello"; say "abc $*foo.uc() def"
p6eval niecza v1-35-gc9d17a7: OUTPUT«abc HELLO def␤»
04:54 justatheory left
lopnor k, i will try someway 04:58
05:01 orafu left, orafu joined 05:09 redicaps left 05:14 redicaps joined
lue phenny: tell masak I still get that malformed UTF-8 string error, and AFAICT my .markdown file is in utf-8. Maybe I'll stumble upon the cause in the meantime :) . 05:15
phenny lue: I'll pass that on when masak is around.
sorear niecza_eval: sub foo(*%_) { say %_.dump }; my $x = 5; foo(:$x) 05:21
niecza: sub foo(*%_) { say %_.dump }; my $x = 5; foo(:$x)
p6eval niecza v1-35-gc9d17a7: OUTPUT«{"x" => 5}␤»
sorear What's the best way to run some code and test the exception thrown? 05:25
05:31 HarryS left 05:41 gbacon left 05:58 HarryS joined 06:05 risou joined
sorear niecza: class Base { class Derived is Base { } }; my $b = Base::Derived.new; say $b ~~ Base; 06:08
p6eval niecza v1-35-gc9d17a7: OUTPUT«Bool::True␤»
06:20 x19290 left
dalek ecza: c23321d | sorear++ | v6n/ (2 files):
[v6] Translate first 1/3 of Metamodel
06:20
06:20 alester left
diakopter what is GameDay 07:07
oh, that was a irclong time ago
07:11 yarrack_logging left
sorear diakopter: tournament of roses 07:12
diakopter what's that 07:15
sorear it's one of the highest level football games in the country 07:16
diakopter I mean, I googled those terms, but I don't see how they're.. oh.
tournament of roses brought up a bunch of stuff about a parade 07:17
but yeah if there's a football game too, I can see how a tv show on sports would mention it 07:18
sorear niecza: my @foo = 1, 2, 3; for @foo { $_++ }; say ~@foo 07:22
p6eval niecza v1-35-gc9d17a7: OUTPUT«2 3 4␤»
07:23 ROlds joined
simon how do I join a predefined character group with a custom one? <[123] + <alpha>>? 07:34
07:37 kaare joined, kaare is now known as Guest29967
sorear simon: yes 07:40
wait
no, it's just <[128]+alpha>
diakopter niecza: my $foo = (1, 2, 3); for $foo { $_++ }; say ~$foo
p6eval niecza v1-35-gc9d17a7: OUTPUT«Unhandled exception: Unable to resolve method Numeric in class Parcel␤ at line 0 (ExitRunloop @ 0)␤ at (eval) line 1 (MAIN mainline @ 4)␤ at /home/p6eval/niecza/lib/SAFE.setting line 1064 (SAFE C381_ANON @ 1)␤ at /home/p6eval/niecza/lib/SAFE.setting line 1064
..(SAFE module-SAFE…
simon sorear, thanks. 07:44
sorear need is better than use because need does not affect the local language 07:55
a slightly smart compiler can respond to "need Foo; need Bar; " by loading Foo and Bar in parallel 07:56
this is not possible with use Foo; because Foo needs to be parsed for macros
dalek ecza: ce5553f | sorear++ | v6n/ (3 files):
Translate second 1/3 of Metamodel
08:11
lao Help: What is the equivalent statement in Perl6 for: substr($s1,3,4,"CAD") (perl5) 08:13
simon in grammars, how do I specify precedence? using ||s and ordering? 08:22
tadzik rakudo: "CAD".substr(1, 2, 3) 08:25
p6eval rakudo a98c69: OUTPUT«No applicable candidates found to dispatch to for 'substr'. Available candidates are:␤:(Mu : Any $start, Any $length?;; *%_)␤␤ in main program body at line 22:/tmp/JB6XgVRy0h␤»
mberends lao: see the penultimate paragraph about substr (left hand side) in perlcabal.org/syn/S32/Str.html 08:26
rakudo: my $a="the cat sat"; substr($a,4,3)="dog"; $a.say; 08:28
p6eval rakudo a98c69: OUTPUT«Cannot modify readonly value␤ in '&infix:<=>' at line 1␤ in main program body at line 22:/tmp/i79vJtJd4x␤» 08:29
mberends so it's NYI :-(
lao mberends: Thank. I looked the web. substr just returns substring. I do not see replacement. I just want to change "theDOGHouse" to "theCADhouse" using substrng index rather than regex. 08:34
sorear I'd rather see 4-argument substr than lvalue substr 08:38
the absense of error recovery in STD is irritating 08:40
std: $GLOBAL::x 08:43
p6eval std 625303c: OUTPUT«ok 00:01 118m␤»
sorear std: GLOBAL::Foo.new 08:44
p6eval std 625303c: OUTPUT«===SORRY!===␤Undeclared name:␤ 'GLOBAL::Foo' used at line 1␤Check failed␤FAILED 00:01 118m␤»
sorear std: Foo::Bar.new
p6eval std 625303c: OUTPUT«===SORRY!===␤Undeclared name:␤ 'Foo::Bar' used at line 1␤Check failed␤FAILED 00:01 118m␤»
tadzik isn't #81562 a feature? 08:45
or is it a duplicate; I think I recall talking about it with someone here, and there was a reason I didn't file a bug report 08:46
dalek ecza: 3523595 | sorear++ | v6n/Metamodel.pm6:
[v6] Fix compile errors
08:48
sorear tadzik: search for "Interface consistency" in S12 or S06
08:49 sftp left 09:01 Chillance joined, roen joined 09:03 lao left 09:05 neroxx joined 09:15 isBEKaml joined 09:19 Alias__ left 09:34 redicaps left 09:40 neroxx left 09:48 isBEKaml left 09:52 neroxx joined 10:08 amkrankruleuen left, amkrankruleuen joined 10:09 mikehh joined
sorear rakudo: sub flue($x is rw) {}; flue(5); say "alive" 10:35
p6eval rakudo a98c69: OUTPUT«alive␤»
10:35 roen left 10:40 molaf joined
moritz_ rakudo's "is rw" doesn't work as it should 10:41
it should fail to bind
rakudo's 'is rw' is what should be 'is ref' 10:42
dalek ecza: 51aa3cd | sorear++ | v6n/Sig.pm6:
[v6] Translate Sig
10:47
pumphaus hm, github still gives a 403 when I try to download 2010.12 -.-
sorear is ref isn't actually documented
moritz_ wait 10:49
the 'is ref' I meant was renamed to 'is parcel' in 4bd6b5e
+You may modify the argument, but only if argument is already a suitable 10:50
+lvalue since, unlike C<rw>, no attempt at autovivification is made,
+so unsuitable lvalues will throw an exception if you try to modify
+them within the body of the routine.
sorear OH! 10:51
a lot of stuff is a lot clearer now that I have that connection.
moritz_ \o/ 10:52
sorear niecza: sub foo(\$x) { my @arr = $x; say +@arr }; foo((1,2,3)) # Discuss.
p6eval niecza v1-35-gc9d17a7: OUTPUT«3␤»
sorear my 'is parcel' preserves the list bit
moritz_ right; it doesn't force $x to be a scalar, even though it has the $ sigil 10:53
which is mightily confusing, but probably right
sorear the niecza setting couldn't exist without that
moritz_ right 10:54
I just wish there was a distinction between "uncontexted" and "scalar" variables
sorear now, STD's placeholder system turns { ... } into -> $_ is ref = OUTER::<$_> { ... }
moritz_ s/ref/parcel/
sorear does this mean that $_ is potentially not a scalar in a bare block?
moritz_: it still uses ref there actually. Which made no sense until todya. 10:55
moritz_ I guess that's a fossile
and yes, $_ could be a list
dalek kudo: db11af9 | moritz++ | .gitignore:
gitignore tarball created for smoke reports, mikehh++
10:56
sorear which means that $_ and $^x are potentially very different 10:57
moritz_ yes :(
moritz_ wants the § sigil for variables with unknown context
sorear niecza: my $h = { map { ($_ => 1) }, 1, 2, 3 }; say $h.dump 11:11
p6eval niecza v1-35-gc9d17a7: OUTPUT«UnknownSub␤»
sorear bleh
niecza: my $h = %( map { ($_ => 1) }, 1, 2, 3 ); say $h.dump 11:12
p6eval niecza v1-35-gc9d17a7: OUTPUT«Unhandled exception: Unable to resolve method hash in class List␤ at (eval) line 1 (MAIN mainline @ 2)␤ at /home/p6eval/niecza/lib/SAFE.setting line 1064 (SAFE C381_ANON @ 1)␤ at /home/p6eval/niecza/lib/SAFE.setting line 1064 (SAFE module-SAFE @ 29)␤ at
../home/p6eval/niecza/lib/…
sorear niecza: my $h = _hash_constructor( map { ($_ => 1) }, 1, 2, 3 ); say $h.dump
p6eval niecza v1-35-gc9d17a7: OUTPUT«{"1" => 1, "2" => 1, "3" => 1}␤»
sorear ponders
how should I translate this line of Perl 5? 11:13
11:13 Guest29967 left 11:14 Guest29967 joined 11:26 coldhead left
moritz_ doesn't see a line of perl 5 11:27
11:31 masak joined
masak \o/ 11:31
phenny masak: 05:15Z <lue> tell masak I still get that malformed UTF-8 string error, and AFAICT my .markdown file is in utf-8. Maybe I'll stumble upon the cause in the meantime :) .
masak lue: check again :)
either your UTF-8 is correct, or it's malformed. there's no third option. 11:32
moritz_ has a crazy idea: parallelized automatic bisecting 11:38
masak as long as you have the computing power, sure. 11:39
you need to be able to tell a bunch of nodes: "guys! stop! it's over here!"
moritz_ well, I have more CPUs than are typically used during a rakudo rebuild, and enough mem 11:40
and two machines in my local network
arnsholt Is it just me, or is checking out Parrot faster now that it's on git? 11:42
masak lots faster. 11:45
both the whole repo and comparable updates.
11:52 kanishka joined
arnsholt Yeah, that was my impression 11:57
masak S32/Str/substr is a joke. one can't just posit that .substr returns an lvalue substring, without also saying what type machinery needs to be in place for that to work. (we already know that .substr is not allowed to "cheat" by using non-p6 primitives.) 11:58
I'd say that that part of S32/Str is void by virtue of its naivety.
*naïvité
12:03 wamba joined 12:04 Alias joined
masak Alias! \o/ 12:04
Alias: it might please you to know that as of yesterday, 'done_testing' is called 'done' in the Perl 6 specification. :) 12:05
Alias Oh thank the fucking lord
masak :D
Alias It pisses me off more than I can possibly describe when people take a perfectly good API that describes things in concise since terms and start adding new things with extra superfluous words 12:06
Usually because they don't speak english as a first language, and so don't trust themselves to get it right with just one word
ok_test 12:07
is_test
masak :(
Alias testing_plan
12:07 kanishka left
Alias use_module Foo::Bar; 12:07
masak please stop... :/
Alias Perl Language 6
:)
masak urgh.
Alias Personally though, I also hate done() 12:12
I think it makes people lazy
But I see how it might be slightly better in some cases
masak I think it's not black-and-white.
Alias My main concern is the existance of it will make people try less hard to actually make proper test counts
Which means less tests with test counts
masak why is that a concern? 12:13
Alias Which means we've traded improved-knowledge-of-early-stopping-for-unknown-plans for increased-likelyhood-someone-won't-use-a-plan-and-the-test-count-will-be-off-and-nobody-will-notice
12:14 apeiron left
Alias I've had test counts go wrong more times than I've had non-plan tests end prematurely with no obvious failure 12:15
But I'm mostly a -0 on the issue
I'm against it, but not enough for it to count
12:15 pmurias joined
masak I'm for it, but I don't mind when people use test plans, or argue for test plans. 12:17
I like not having to re-count the test plan each time I add or remove a few tests in my file.
I've not had much trouble with premature failures.
Alias That you know of
oh wait
You mean for no_plan ones 12:18
masak right.
had I had much trouble with premature failures in no_plan test files, I might be more sceptical as well.
but I haven't, so I'm not. 12:19
pmurias premature failures are taken care of by done()
masak right, that's why we need &done 12:20
12:20 noganex_ joined
masak this all assumes a TAP harness to keep us informed, of course. 12:20
12:24 noganex left
pmurias masak: test's are usually run by a harness, or manually for inspection 12:27
Alias: doesn't updating the plan annoy you? 12:28
masak mst once pointed out that updating the plan works particularly badly with version system merges.
12:29 masonkramer left
masak that blog post of his was what initially made me lean towards 'done()' in my test files. 12:29
moritz_ masak: that sounds logical, but I haven't had that problem in practise really
12:29 masonkramer joined
pmurias had once 12:29
masak moritz_: no, I think it's worse when many people crowd around a particular test file. 12:30
moritz_: some test files in the p5 repo have upwards of 17k tests.
colomon "The top referring sites [for justrakudoit] in 2010 were perlfoundation.org, reddit.com, planetsix.perl.org, Google Reader, and irclog.perlgeek.de." -- moritz_++
moritz_ kinda hopes in vain that many people crowd around roast 12:31
12:31 kanishka joined
pmurias masak: imagine having to count them up to see if the plan count is wrong or the tests are 12:31
masak pmurias: I don't need to imagine very hard.
moritz_ I guess most people either only count the diffs, or just trust the count from TAP::Harness 12:32
masak pmurias: though usually what I do is make an idempotent s/is /is / replacement in vim and count the number of replacements. :)
s/count/read off/
pmurias moritz_: if you trust the count from TAP::Harness the whole point of the plan is broken 12:33
masak as soon as we have AST analyzers, we can start playing around with automated test counting.
moritz_ pmurias: right 12:36
Alias pmurias, sure. Doesn't declaring variables annoy you? 12:37
masak I don't think the analogy is very accurate.
Alias Test plans take more work, but give more reliable result
With some extra typeing they remove a catagory of potential errors 12:38
Via some additional strictness
I think it's a reasonable analogy
masak declaring variables so that they stay in their scopes will help me as my application grows. keeping a test plan up-to-date will become increasingly difficult as my test file grows.
Alias: could you please explain in some more detail what the potential errors are?
it seems neither pmurias or I understand that.
Alias my @output = $some->function; 12:39
foreach ( @output ) { isa_ok($_, 'Something') }
How do you tell when it's returned the wrong number of things
That's one notional example
But there's tons of others
masak oh, I think you've blogged about this once. 12:40
colomon is +@output, right number of things.
moritz_ with is scalar(@output), $expected_number, '...';
Alias If I don't have a plan, then all of a sudden I don't need to skip any more either
I just shortcut
12:40 kanishka1 joined
masak ...and that's bad because...? 12:40
colomon but yeah, loops which exit prematurely are harder to detect without a fixed plan.
Alias Loops which exit prematurely, or conditionals which can silently evaluate wrong 12:41
masak one could argue that testing for the length of the array is a good test to have in there anyway, plan or no plan.
Alias In the old situation, the untested case was a test script which did a hard-non-error exit(0) prematurely
masak: Sure it is, but you don't have to
Just like using ? placeholders in DBI
masak even with a plan, there might be subtle errors which offset each other, ending up on the right test count.
Alias That is true 12:42
But then you have an two-factor error case
Which is likely to be exceedingly rare
masak right. it's a weak argument that way.
12:42 kanishka left
Alias Personally, I can't remember more than 1 or 2 cases in my entire career that a test script has done a premature silent exit() which was also in a test script that did not have an inherently plan 12:43
I can remember lots where I got a loop count wrong
masak Alias: if you get loop counts wrong so often, why don't you test the length of the array?
Alias So in my opinion, we're fixing a very rare case and as a side effect encouraging greater use of unsafe behaviour 12:44
masak: Because sometimes I'm lazy?
12:44 kanishka1 left
Alias Same thing with SQL and placeholders 12:44
And I have a lot of faith in programmers tendency to be lazy 12:45
As a collective userbase of Test::More
12:45 arthur-_ joined
masak arthur-_! \o/ 12:45
arthur-_: happy new year!
arthur-_ masak: you too !
masak Alias: putting in a test plan as a safeguard against laziness. I buy that. 12:46
Alias right
And part of this thing with done() is that it's been promoted
I was reaonsably ok to have it, but I think you should HAVE to write explicitly no_plan in order to get it 12:47
It should take MORE work to access the lazy option
I'd even go so far as to leave it out of the SYNOPSIS
Encouraging it's use annoys me more than having it there for the one case it does defend against 12:49
dalek osystem: df83462 | (Solomon Foster)++ | projects.list:
Add Damian Conway's Testing module to ecosystem.
12:50
arnsholt Making &done undocumented would be a bad idea 12:52
12:52 zby_home_ joined
Alias Not undocumented 12:52
Just not in the SYNOPSIS
arnsholt Especially since that would mean it would be non-standard
The synopsis is the spec
Alias No it isn't
masak arnsholt: you're attacking a strawman. 12:53
Alias The SYNOPSIS is the bit at the top of the module
arnsholt masak: Good point
masak arnsholt: not undocumented, just removed from the SYNOPSIS section.
Alias That shows the most normal and common usage
arnsholt Oh, that synopsis
Alias right
masak yeah.
Alias (Capital SYNOPSIS, not Synopsis)
masak Alias wasn't shouting, he was using a technical term. :)
12:57 neroxx left 13:04 qiuhw left
moritz_ things I wish perl 5 had: a switch to log all IO 13:10
like the system() or chdir() or exec() arguments
I guess in Perl 6 it's easier to do, because run() isn't parsed as specially as system() is in p5
masak Arc has this nice kind of function that takes a closure, and every 'say' or other output to $*OUT is gathered up into a string and returned from that function. 13:13
moritz_ not hard to do in Perl 6 either 13:14
in fact that's a nice module to write :-) 13:15
masak aye.
colomon go moritz_, go! 13:16
moritz_ it just needs a class that supports .print and .say (and maybe .write), and then set $*OUT and $*ERR to instances of that
masak the benefits of designing stuff right :)
13:17 shi joined
moritz_ rakudo: class Dummy { method print(*@_) { }; method say(*@_) { } }; my $*ERR = Dummy.new; die "should not show up" 13:18
p6eval rakudo db11af: OUTPUT«should not show up␤ in main program body at line 22:/tmp/iMWFJotqZL␤»
moritz_ am I right?
masak submits rakudobug
think so.
well, depends on in which scope the error is printed, I guess. 13:19
but it's worth getting a verdict on.
moritz_ there might be some in-the-setting exception handler that catches and prints to $*ERR 13:20
so it's not the localized $*ERR
masak right.
in fact, for the error to boil up all the way out of the program, it's pretty much certain that we've left the scope where $*ERR was overridden. 13:21
so this is probably a non-bug.
moritz_ unsubmits the rakudobug :-) 13:22
masak *poof*
moritz_ tries his new script for bisecting parrotbugs through rakudo 13:24
arnsholt rakudo: class Dummy { method print(*@_) { }; method say(*@_) { } }; $*ERR := Dummy.new; die "should not show up" 13:27
p6eval rakudo db11af: OUTPUT«should not show up␤ in main program body at line 22:/tmp/fWXB7EOyH1␤»
masak hm... not declaring $*ERR *would* change matter, wouldn't it?
dalek osystem: 7f3cf76 | (Solomon Foster)++ | projects.list:
Add Damian Conway's IO::Prompter to the ecosystem.
masak whirrs up the bug-submitter again
arnsholt++ 13:28
moritz_ not at all
masak whirrs down
arnsholt I suppose it depends on where and how $*ERR is set up 13:29
colomon moritz_: I hate to say it, but Damian's Testing code is beautiful. 13:31
moritz_ I imagine the setting to look like this: try { YOU_ARE_HERE }; $*ERR.print($!) if $!;
colomon OK have => prompt(:integer, in => IN('a','1.2','-42a','-12')),
want => -12,
desc => "Require an integer";
arnsholt masak: Consolation prize 13:32
rakudo: class Dummy { method print(*@_) { }; method say(*@_) { } }; my $*IN = Dummy.new; say "should not show up"
p6eval rakudo db11af: OUTPUT«should not show up␤»
masak arnsholt: $*IN?
arnsholt >.< 13:33
I r an idiot =D
masak hugs arnsholt
arnsholt I blame your competition
I've been thinking too much about parsing all the input from stdin ;)
Given that, I guess it's just as well I don't have to do actual work 'till tomorrow 13:34
masak arnsholt: make sure you submit as solutions some of that thinking you've been making. :)
arnsholt *giggle* 13:35
I just need to implement some of that thinking as well
13:36 ZeroCoder joined 13:41 Layla joined
moritz_ git question: why does 'git describe' work out of the box in the rakudo repo, but needs the --tags option in the parrot repo? 13:45
masak moritz_: from the manpage: "By default (without --all or --tags) git describe only shows annotated tags." 13:46
maybe that's a clue.
moritz_ what's an annotated tag?
masak I don't know what an annotated tag is, but it does sound relevant.
"For more information about creating annotated tags see the -a and -s options to git-tag(1)." :)
13:47 Guest29967 is now known as kaare_
masak nom & 13:47
13:48 masak left
moritz_ git-tag only mentions that it can create annotated tags 13:48
wwow
13:50 glow left 13:51 Layla left
lopnor hi, I finally made fully perl6-written blog online! 14:00
journal6.soffritto.org/
superslow so you have to wait about 3min.
14:05 sftp joined 14:06 qw3rty joined
moritz_ lopnor: cool idea. What do you think of a caching proxy? :-) 14:11
lopnor moritz_: i should:)
14:12 yarp joined
lopnor but the best is perl6 is enought fast to return all request in time! 14:12
moritz_ or generate static HTML in the first place (though that would probably require many changes)
lopnor: is the source code online somewhere?
lopnor github.com/lopnor/p6-journal 14:13
this is blog code
github.com/lopnor/p6-plackdo
this is firmware code
moritz_ lopnor: care to add it the official module list? 14:14
lopnor and this is same blog retuned with perl5 journal.soffritto.org/ 14:15
moritz_ lopnor: you now have commit access to github.com/perl6/ecosystem/
lopnor great!
moritz_ (and most other repos in the perl6 organization, for that matter. Like test suite, mu, ...) 14:16
lopnor moritz_: i'm warrying plackdo conflict some specs with 'web'
flussence re: tags; annotated tag is just a tag with the tag's "commit message" filled in 14:17
(which is what -a/s prompt you for)
(also, those messages show up in qgit's statusbar but I don't think github shows them...) 14:19
moritz_ lopnor: don't worry; I haven't seen any real world application of 'web', and "works" wins over "specs" :-) 14:23
at least regarding modules
lopnor k, will work harder!
moritz_: thank you! 14:24
moritz_ lopnor: you're very welcome
fwiw I get 502 proxy errors from your blog
14:24 rindolf joined
moritz_ "The proxy server received an invalid response from an upstream server. 14:24
The proxy server could not handle the request GET /."
rindolf Hi all.
lopnor moritz_: yes, this blog runs with single process, i will enhance handler with prefork strategy. 14:25
moritz_ rakudo: pir::fork(); say 3
p6eval rakudo db11af: OUTPUT«error:imcc:syntax error, unexpected '\n', expecting '('␤ in file 'EVAL_10' line 86␤===SORRY!===␤syntax error ... somewhere␤»
lopnor currently thinking fork with NativeCall 14:26
moritz_ right, that's probably the only working solution today
lopnor++ # cool perl 6 code
lopnor almost every code is stolen from plack in perl5. 14:27
so coolness is for miyagawa. 14:28
moritz_ well, you ported it and made it run
14:29 yarp left
lopnor :) 14:29
now you can see journal6.soffritto.org/ , i think 14:35
14:52 rgrau joined 14:56 ZeroCoder left
moritz_ lopnor: yes, works now 15:00
though I seem to lack fonts to display all of it properly
lopnor moritz_: written in japanese so you would need japanese fonts.
moritz_ even then I couldn't read it :( 15:01
lopnor if you see yellow squares it's css brought by perl6.
moritz_ just the favicon is a yellow square 15:02
flussence ooh, nice 15:03
lopnor moritz_: ah, its writen by perl6 :) 15:04
i will work harder to deliver thins more faster (in application layer). 15:05
flussence btw, "* { size: 90% }" is probably meant to be "body { font-size: 90% }" :)
lopnor flussence: yes!
my css is very bad work :P
15:06 am0c joined
flussence (Chrom[e|ium] puts a little warning symbol next to unknown css in the inspector now, nice) 15:07
15:08 c1sung left 15:09 c1sung joined
pmurias Alias: for me the main reason why i hate test plans is not the extra typing but the fact that when is that i tend to forget updating them and i waste time counting tests 15:15
Alias: the error you ran 11 tests instead of 12 isn't a very helpfull one
Alias: if you keep track of the plan easily and do tricky stuff in the tests the plan might be helpfull 15:16
Alias: but for me it's a tiny gain for a huge annoyance 15:18
15:20 carlos__ joined 15:21 lambda-nil left
flussence it's a lot like having to manually set the size of an array 15:21
15:21 jedai joined
flussence what I usually do is just stick all my tests in an array and plan +@stuff; :) 15:22
pmurias likes that 15:23
15:30 carlos__ left 15:38 qw3rty left
dalek ecza: 67a3e6f | pmurias++ | / (2 files):
added Niecza::Frontend::STD which will be responsible for the parsing
15:55
ecza: 77dab6d | pmurias++ | v6n/ (5 files):
Merge branch 'master' of github.com:sorear/niecza
pmurias argh forgot git pull --rebase :(
15:55 pmurias left 15:57 hcchien joined 16:01 Patterner left 16:03 envi left 16:06 am0c left, orafu left, orafu joined 16:07 Psyche^ joined, Psyche^ is now known as Patterner 16:10 _twitch joined
colomon hmmm. 16:17
16:19 kanishka joined, orafu left, orafu joined
colomon Fixed ufo to handle sources in lib6 directories last night. 16:20
But it installs them into .perl6/lib6 -- and Rakudo doesn't recognize that.
Change ufo or change Rakudo? 16:21
tadzik ufo imno 16:24
tadzik tunes module::tools 16:25
colomon tadzik: I think you might be right, but unfortunately that's the hard answer. :) 16:26
tadzik is it?
16:27 kraai_ joined
kraai_ Hi. The top-level README in Rakudo * 2010.12 says that it's from the 2010.08 release. 16:29
tadzik . o O ( amap: appenging map )
kraai_: that sounds wrong indeed, thanks for letting us know 16:30
kraai_ tadzik: You're welcome. 16:31
pumphaus oh wow, we can finally download it :) 16:37
colomon tadzik: I'd certainly rather tweak Rakudo than try to sort out how to make a significant modification to ufo's makefile generation. 16:41
tadzik the "Suprise lib6" case made me think about some sorf of a META file, with info about where to find stuff 16:48
I feel like a blag post promoting the idea, or rather provoking a discussion 16:49
fixing every possible tool in a module toolchain because there appeared a module with different dirs feels LTA 16:50
16:53 jimfuller joined, gbacon joined, shi left 16:55 colomon left, silug left
jimfuller lazy request: whats the state of visual gui (gtk, etc) integration with perl6 these days ? 16:55
tadzik tk is reported to work on blizkost or zavolaj 16:56
I know of no other attempts, feel free to try
(I think that was blizkost)
jimfuller ok, cheers for the info 16:57
tadzik you're welcome 16:58
17:02 masak joined
masak lopnor: wow! 17:02
that is seriously impressive.
lopnor: I think your best chance at speed at this point is to make your code work on niecza. 17:03
17:08 MayDaniel joined
masak moving & 17:09
17:09 masak left 17:10 kanishka1 joined, colomon joined 17:11 kanishka1 left
colomon tadzik: sorry for disappearing there. 17:11
I think we need to set standards.
17:11 kanishka left
colomon and in the meantime, I'm just going to change Damian's modules to use lib instead of lib6. 17:11
tadzik good eye deer 17:14
17:15 _jaldhar is now known as jaldhar
TimToady is still recovering from a spiral write on my disk drive yesterday morning, or maybe a minor headcrash 17:16
17:16 MayDaniel left
TimToady it's a bad thing when your OS decides to remount your filesystem ro 17:16
17:16 MayDaniel joined
moritz_ oh dear yes 17:17
TimToady and of course it clobbered the inode table and most of my top-level directories
everything ended up in lost+found; well, everything it could get back 17:18
colomon oh noez!
TimToady still haven't seen my home directory anywhere...
fortunately, most of what I care about is backed up on the net
and though I'm running in a different partition now (and used the excuse to upgrade to 64-bit ubuntu), I don't really trust my disk anymore 17:19
17:20 _twitch left
colomon tadzik: lib6 to lib done and pushed. 17:21
17:24 icwiener joined, kraai_ left, orafu left, orafu joined 17:29 _twitch joined 17:34 jimfuller left 17:43 dual left 17:50 yarrack_logging joined
dalek kudo: fc2db18 | KodiB++ | / (2 files):
Added a Perl 5 script to update the leap-second tables in tai-utc.pm.

I tried to do it in Perl 6 first, but I couldn't get perl6-lwp-simple to work.
17:53
17:55 WonTu joined, WonTu left
moritz_ my hacking on a bisecting script for parrot found a perl 5 bug :-) 18:03
(well, an autodie bug, to be exact) 18:04
maybe we should put instructions into the release managers guide to run the script after a release, and commit the results 18:05
where by "the script" I meant KodiB++'s last commit
18:07 MayDaniel left
pmichaud I'm not sure that should be part of the release process 18:08
regenerating and/or updating src/core/tai-utc.pm should be in response to bugfix/maintenance, not part of release 18:09
TimToady We should name our Metamodel Metalmode 18:16
pmichaud yamm
(yet another meta model) 18:17
TimToady sorear: STD does do a certain amount of error recovery from semantic errors--that's what sorry() is for 18:21
perigrin MeatAlaMode 18:33
18:34 jaldhar left 18:37 dukeleto left 18:38 dukeleto joined, jaldhar joined 18:40 justatheory joined 18:41 _twitch left 18:57 risou left
sorear TimToady: how impossible would it be for STD to recover from a misplaced ) and report errors further along the file? 18:59
good * #perl6 19:00
19:00 MayDaniel joined
sorear phenny: tell masak What do you mean that substr cannot use non-p6 primitives? How else is it supposed to inspect an opaque Str? 19:03
phenny sorear: I'll pass that on when masak is around.
19:07 wamba left 19:11 kraai_ joined 19:18 amkrankruleuen left
sorear TimToady: I've never heard of a spiral write before. 19:26
19:27 dual joined 19:31 dorlamm joined 19:41 MayDaniel left 19:44 lukas___ joined
diakopter metamodel -> Told A Meme 19:46
rindolf I met a model. 19:50
dalek rixel: 6bbccc5 | diakopter++ | / (13 files):
another wip commit.

previous perlesque); still need to validate/debug them on mono.
19:54
19:55 shi joined 19:56 Vlavv` joined 19:57 Vlavv_ left 20:10 rgrau left 20:12 orafu left, orafu joined
lue ohai fellow zebras o/ 20:15
moritz_ \o
colomon o/ 20:17
moritz_ has now submitted a solution for each of masak's five tasks, each good enough to pass base-test 20:36
colomon what is base-test? 20:39
my p5 is still problematic. 20:40
and I didn't see any tests included for it? the writeup seemed incomplete next to the others, though the concept is obviously straightforward.
20:42 LeifD joined 20:45 lukas___ left
lue Just out of curiosity, do I need the line module A::B; in the file A/B.pm? [doesn't [seem to] hurt anything, just wondering if it's redundant] 20:46
20:48 dorlamm left
moritz_ yes 20:51
though class A::B { ... } will do too
20:51 MayDaniel joined
moritz_ eventually the contents of the module file will be authorotative, not the name 20:51
lue how do you mean? 20:52
flussence filesystems are just an implementation detail :P 20:53
lue Does that mean the name won't tell you everything, eventually?
moritz_ correct 20:55
the spec says that we must be able to map module names that aren't directly mappable to file names
like case sensitivity on case insensitive file systems
or Unicode names on ASCII-only file systems
and then there's this whole thing about versions and authorities 20:56
lue gets the sneaking feeling that S11 says more than he thinks if he were to just read it like a gripping adventure novel. 20:57
20:59 explorer__ joined 21:00 jferrero left 21:02 LeifD left
moritz_ we had a brainstorming session on it, and then a hackathon to implement parts of it 21:02
21:03 rindolf left 21:04 colomon left 21:05 dorlamm joined
lue There's something about the module system I don't like. I'll read S11 first, and then see if I still don't like it. 21:05
lue reads
21:06 colomon joined
tadzik lue: what about the module system? 21:10
21:11 awwaiid joined
lue I honestly don't know [yet]. Maybe it's because it's (Not Yet|Not Well) Implemented. 21:16
dalek rixel: 3d6211f | diakopter++ | / (7 files):
fixed if/else without intervening elsif, and various other things.
21:18 pmurias joined
diakopter pmurias: hi 21:19
tadzik lue: you mean a CPAN-like infrastructure? 21:20
pmurias diakopter: hi 21:21
21:21 am0c joined 21:22 lukas___ joined
moritz_ it certainly isn't well implemented yet 21:23
21:23 plobsing left 21:24 plobsing joined
pmurias sorear: how often does the compiled by niecza STD use coroutines? 21:28
21:30 molaf left
sorear pmurias: every call to &map, &grep creates one; every 'regex' creates one 21:31
lue moritz_: eventually, would I be able to store module A::B in places other than A/B.pm ? 21:32
moritz_ lue: yes 21:34
though how "other" exactly they are going to be, I don't know yet
dalek ecza: 7a14f4f | pmurias++ | src/Niecza/Backend/NAM.pm:
added the backend which outputs the ast as NAM
ecza: 09646fd | pmurias++ | niecza:
started creating the script for the refactored compiler
pmurias sorear: so the mono specific continuation stuff would slow us down? 21:35
lue Is A:name<B> always referred to as A:name<B>? 21:38
diakopter pmurias: yay; man-or-boy.t is working again :D
sorear: does man-or-boy work on niecza? 21:39
moritz_ lue: my $x = A:name<B> # now you can refer to it by $x
diakopter (I'm not challenging, merely curious b/c I don't know)
21:40 lukas___ left
lue Let me see if I got this right: if module A comes with a shortname (e.g. :name<B>), then you can use B; instead of use A; , right? 21:40
pmurias diakopter: is copy is not implemented yet in niecza 21:41
diakopter pmurias: why would 'is copy' matter?
moritz_ you can easily change the test not to use "is copy"
diakopter (right)
my version is quite different, b/c it uses primitive types only 21:42
well, and strongly-typed closures
moritz_ I get
Can't use an undefined value as an ARRAY reference at /home/moritz/p6/niecza/src/NAMBackend.pm line 96.
while trying to compile man-or-boy.t with niecza
moritz_ -> sleep
diakopter nite 21:43
lue My question was, does the longname ever get shown as anything other than A:name<B> ?
'night, moritz_ o/
pmurias moritz_: rm -fr obj/
diakopter lue: sorry; I don't know :( 21:44
lue that's alright, it's just my curiosity, I don't actually need the info :) . 21:48
diakopter pmurias: e.g. here's my edition: perl6.pastebin.com/raw.php?i=AXhUA23d 21:49
pmurias sorear: i'm thinking ->run is not a good name for the method in passes, what would a good one be? ->tranform,->process?
dalek rixel: bfd7f45 | diakopter++ | sprixel/ (3 files):
finish getting man-or-boy.t to work. strongly-typed edition works. :)
sorear pmurias: invoke? 21:54
diakopter phenny: ask TimToady in the following primitive-typed edition of man-or-boy perl6.pastebin.com/raw.php?i=AXhUA23d I postulated a shorthand to write type annotations for "generic" Callables (that are strongly-typed)... could you please [in-]validate that such a construct could work? 21:55
phenny diakopter: I'll pass that on when TimToady is around.
sorear lue: you could also use A:name( "B" )
er, :name[ "B" ]
diakopter phenny: ask TimToady but ignore the '&' "sigil" before 'B' - I'm removing that curiosity from this perlesque dialect 21:57
phenny diakopter: I'll pass that on when TimToady is around.
pmurias sorear: we could have passes as subs but i thing code reuse between the different passes will be important
dalek ecza: 756b65f | pmurias++ | / (5 files):
passes and Niecza::Compiler that will tie stuff together
21:59
ecza: 8f0056c | pmurias++ | src/Niecza/ (4 files):
change the ->run method of passes to ->invoke
pmurias sorear: changed
sorear pmurias: are you planning to kill CompilerDriver and niecza_eval ? 22:01
pmurias once niecza and the refactored stuff yes 22:02
22:03 coldhead joined
pmurias i don't want to break stuff so i'm letting CompilerDriver live for now 22:03
sorear pmurias: why does the 'niecza' script need to know about the internal passes? I call most of them ugly implementation details 22:04
Why doesn't Niecza::Compiler.compile use named arguments?
pmurias it should use those
niecza is a placeholder for testing 22:05
what should specify the passes?
sorear Niecza::Compiler? 22:06
pmurias different backends/flag combinations can have different passes 22:07
in mildew i used Bread::Board but i don't to pay the price for it's loading time
s/to/want to/
but i'm trying to do dependency injection manual 22:08
dalek ecza: eb2e337 | pmurias++ | / (2 files):
make Niecz::Compiler use named arguments
22:12
22:12 spam joined 22:13 spam is now known as Guest45477
pmurias sorear: but we can store all the ugly implementation details in a seperate file 22:13
22:13 Guest45477 left 22:15 colomon left
lue
.oO(S11:535 is closely related to the bit in S01 about P5-by-default, just so you know.)
22:15
22:15 colomon joined
pmurias sorear: what does stagetime argument to CompilerDriver do? 22:17
22:19 patrickas joined
patrickas rakudo: my $r = 2..4;say pick 2,3,4;say $r.pick;say pick($r); 22:20
p6eval rakudo fc2db1: OUTPUT«43␤3␤␤»
sorear pmurias: implements the -v compiler option; try ./niecza_eval -v -e 'say 2+2'
patrickas should the last pick act like the first or the second?
22:20 rgrau joined
pmurias so it should be named verbose? 22:20
sorear pmurias: that would work too, but it prints stage times, so stagetime seemed like a good name 22:21
22:40 jarfhy left 22:44 explorer__ left, dual left, jaldhar left, dukeleto left, icwiener left, gbacon left, Patterner left, masonkramer left, felliott left, knewt2 left, mberends left, JodaZ left, skangas left, robinsmidsrod left, aloha left, sjohnson left, flatwhatson left, elb0w left, sukria left, PacoLinux left, sbp left, barika_ left, oha left, __sri left, DarthGandalf left, pmurias left, VXZ left, Sarten-X left, Tene left, zorgnax left, cognominal left 22:46 explorer__ joined, dual joined, jaldhar joined, dukeleto joined, icwiener joined, gbacon joined, Patterner joined, masonkramer joined, felliott joined, knewt2 joined, mberends joined, JodaZ joined, skangas joined, robinsmidsrod joined, aloha joined, sbp joined, sjohnson joined, flatwhatson joined, elb0w joined, sukria joined, PacoLinux joined, Util joined, hillu joined, sunnavy joined, tadzik joined, toebu joined, jerome joined, phenny joined, nsh_ joined, barika_ joined, oha joined, __sri joined, DarthGandalf joined, kjeldahl left, kjeldahl joined, cognominal joined, pmurias joined, VXZ joined, Sarten-X joined, Tene joined, zorgnax joined 22:51 jarfhy joined 22:52 takadonet joined, takadonet left 22:53 yarrack_logging left 22:54 mikehh left, Eevee left, sorear left, tomaw left, felipe left, Gothmog_ left, larsen left, jdv79 left, pmurias left, VXZ left, Sarten-X left, Tene left, zorgnax left 22:56 mikehh joined 22:57 kaare_ left, takadonet joined 22:58 Eevee joined, jdv79 joined, sorear joined 22:59 c9s joined, jmmills joined, shachaf joined 23:01 [Coke] joined 23:03 larsen_ joined, pmurias joined, VXZ joined, Sarten-X joined, Tene joined, zorgnax joined
sorear nice netsplit. 23:04
23:05 shi left
diakopter wb all 23:08
23:08 dorlamm left 23:10 zby_home_ left 23:12 kraai_ left 23:13 kraai_ joined 23:14 shi joined 23:16 envi joined 23:26 pmurias left 23:27 explorer__ left, dual left, jaldhar left, dukeleto left, icwiener left, gbacon left, Patterner left, masonkramer left, felliott left, knewt2 left, mberends left, JodaZ left, skangas left, robinsmidsrod left, aloha left, sjohnson left, flatwhatson left, elb0w left, sukria left, PacoLinux left, Util left, sunnavy left, hillu left, tadzik left, toebu left, jerome left, phenny left, nsh_ left, sbp left, barika_ left, oha left, __sri left, DarthGandalf left, envi left, plobsing left, orafu left, jenk left, arlinius left, spq1 left, IllvilJa left, literal left 23:29 tomaw joined, envi joined, literal joined, IllvilJa joined, spq1 joined, arlinius joined, jenk joined, orafu joined, plobsing joined, explorer__ joined, dual joined, jaldhar joined, dukeleto joined, icwiener joined, gbacon joined, Patterner joined, masonkramer joined, felliott joined, knewt2 joined, mberends joined, JodaZ joined, skangas joined, robinsmidsrod joined, aloha joined, sbp joined, sjohnson joined, flatwhatson joined, elb0w joined, sukria joined, PacoLinux joined, Util joined, hillu joined, sunnavy joined, tadzik joined, toebu joined, jerome joined, phenny joined, nsh_ joined, barika_ joined, oha joined, __sri joined, DarthGandalf joined 23:30 lukas___ joined 23:31 Gothmog_ joined 23:33 ggoebel left 23:38 ggoebel joined 23:41 yarp joined 23:48 patrickas left