pugscode.org/ | nopaste: sial.org/pbot/perl6 | pugs: [~] <m oo se> (or rakudo:, kp6:, smop: etc.) || We do Haskell, too | > reverse . show $ foldl1 (*) [1..4] | irclog: irc.pugscode.org/
Set by TimToady on 25 January 2008.
00:04 Fokez left 00:12 eternaleye_ joined 00:22 eternaleye left, Jedai left 00:38 thestarslookdown joined
pugs_svnbot r20262 | Darren_Duncan++ | ext/Muldis-DB/ : part 1 of 20080411 resync with Perl 5 version 01:15
diff: dev.pugscode.org/changeset/20262
lambdabot Title: Changeset 20262 - Pugs - Trac
01:15 eternaleye_ left 01:20 meppel-san joined
pugs_svnbot r20263 | Darren_Duncan++ | ext/Muldis-DB/ : part 2 of 20080411 resync with Perl 5 version 01:20
diff: dev.pugscode.org/changeset/20263
lambdabot Title: Changeset 20263 - Pugs - Trac
01:26 penk joined 01:28 meppl left 01:30 Greyscale joined 01:36 penk left 01:37 penk joined 01:39 syle left 01:54 pen1 joined, penk left 01:57 ofer joined 02:02 pen1 left
meppel-san good night 02:43
02:47 meppel-san left 02:52 ofer left 02:54 thestarslookdown left 03:05 elmex_ joined, elmex_ left 03:10 elmex left 03:12 eternaleye_ joined 03:31 eternaleye_ left 03:32 eternaleye_ joined 03:57 eternaleye joined 03:58 eternaleye_ left, justatheory left 04:14 mj41___ joined 04:29 mj41 left 05:02 Psyche^ joined, alester joined 05:07 alanhaggai joined 05:10 Patterner left, Psyche^ is now known as Patterner
pugs_svnbot r20264 | Darren_Duncan++ | ext/Muldis-DB/ : part 3 (done?) of 20080411 resync with Perl 5 version 05:22
diff: dev.pugscode.org/changeset/20264
lambdabot Title: Changeset 20264 - Pugs - Trac
05:25 alanhaggai_ joined 05:30 eternaleye left, eternaleye joined 05:37 alanhaggai left 05:38 eternaleye_ joined 05:41 wtgee left 05:54 s3ir left 06:03 sri_work joined 06:06 eternaleye left 06:11 alanhaggai_ left 06:19 alester left 06:29 alanhaggai_ joined 07:14 yewenbin joined 07:26 ilbot2 left 07:33 alanhaggai_ left 07:34 alanhaggai joined 07:52 kanru2 left 07:55 kanru joined 07:57 Jedai joined, iblechbot joined 08:20 xinming left, kcwu left, araujo left 08:21 xinming joined, araujo joined, kcwu joined 08:27 Auzon left 08:42 barney joined 08:46 xinming left, kcwu left, araujo left, xinming joined, araujo joined, kcwu joined 08:56 alanhaggai__ joined 09:00 yewenbin left 09:07 alanhaggai left 09:22 eternaleye_ left 09:32 kanru2 joined, kanru left 09:36 cognominal_ left 09:40 xinming left, kcwu left, araujo left 09:41 xinming joined, araujo joined, kcwu joined 09:42 wknight8111 joined 09:45 cognominal_ joined 10:10 wknight8111 left 10:40 chris2 joined 10:43 nirz joined 10:45 jan_ left 10:47 nirz left 10:57 Zygo joined 11:03 literal joined
pasteling "literal" at 194.144.99.91 pasted "pugs svn build fails" (744 lines, 65.7K) at sial.org/pbot/30800 11:05
literal ^-- help!
11:05 chris2 left
pugs_svnbot r20265 | putter++ | [elf_e] Move global $*variables to GLOBAL package. 11:09
diff: dev.pugscode.org/changeset/20265
lambdabot Title: Changeset 20265 - Pugs - Trac
11:32 Fokez joined 11:38 jan joined 11:48 alanhaggai__ is now known as alanhaggai 11:55 Dave2 left 12:00 mncharity joined 12:01 chris2 joined
mncharity moritz_: irclog is down 12:04
lambdabot mncharity: You have 1 new message. '/msg lambdabot @messages' to read it.
mncharity literal: wierd, let's see...
/msg lambdabot @messages 12:05
:/
stevan_: re "package B its a reserved "package"", oy, right, thanks. bad, bad naming. um, I've forgotten context... aside from "my test case was odd" and "someday package names should be mangled to avoid 'B'", was there an impact? 12:09
literal: are you setting PUGS_EMBED? 12:12
12:41 jan left, yewenbin joined, jan_ joined 12:45 barney left
pugs_svnbot r20266 | pmurias++ | [elf] 12:48
r20266 | pmurias++ | removed the unrefactored ElfD_NoMoose and put the refactored one it it's place
r20266 | pmurias++ | added an elf_d_nomoose executable
diff: dev.pugscode.org/changeset/20266
lambdabot Title: Changeset 20266 - Pugs - Trac
spinclad mncharity: '< mncharity> /msg lambdabot @messages': i see a box (non-break space?) before the /. 13:05
stevan_ mncharity: I suspect your test case would pass if you removed B 13:07
13:09 meppl joined
meppl good morning 13:11
mncharity good morning meppl. 13:15
13:15 meppl left
mncharity stevan_: ah. /me no longer remembers failing test case, goes backlogging... 13:15
13:16 meppl joined
meppl good morning mncharity 13:16
mncharity perl -we '{package A;use Moose;sub f{print 3}}{package C;use Moose; extends qw(::A);}A->f;B->f;' 13:18
Can't call method "can" without a package or object reference at /usr/local/lib/perl5/site_perl/5.10.0/Moose/Meta/Class.pm line 278.
oh, duh. sigh
perl -we '{package A;use Moose;sub f{print 3}}{package C;use Moose; extends qw(::A);}A->f;C->f; 13:19
still fails. no typos. ;)
stevan_ s/:://
perl -we '{package A;use Moose;sub f{print 3}}{package C;use Moose; extends qw(A);}A->f;C->f;'
works fine
mncharity re 's/:://', works fine, yes, that was the point. :) @ISA is happy with ::A. extends() is not. I got briefly burned by the difference, so thought I'd mention it. 13:21
stevan_ @ISA is an internal mechanism though
lambdabot Unknown command, try @list
stevan_ extends is just a "macro" really
and honestly, you are the first person I have seen use ::A for a package name in @ISA 13:22
:)
if you really want it to work for Moose though, you can patch Class::MOP::Class::superclasses to do it
I will even give you a commit bit :)
mncharity re 'really want it', no, no, just noting an oddity. 13:23
re 'first person', /me points finger at code generator.
re '@ISA is an internal mechanism though'... huh? 13:24
stevan_ well sorry,.. method dispatch is an internal mechanism
and it reads @ISA
and I would assume (based on your example) that method dispatch sees ::A and A as equic
equiv
mncharity ah, so the observation is "you can use ISA directly, instead of extent(), and Moose will still work"? nifty, I was wondering about that. 13:25
stevan_ sure 13:26
mncharity hence the extends(grep{!/^Moose::Object$/}@ISA,'AddANewClass') hack
stevan_ extends is just a helper
use base Foo; @ISA = qw(Foo),.. it all works
mncharity ok. thanks :) 13:27
13:27 araujo left
stevan_ one of the primary design goals of Moose was to make sure it interoperated as nicely as possible with existing Perl idioms/techniques and modules 13:28
mncharity spinclad: re 'box (non-break space?) before the /', ahhhh. thanks. /me wonders if lambdabot changed, or gaim.
:) # re goal
allbery_b LB didn't change (and would have no control over your client sending garbage before a `/') 13:30
13:33 araujo joined
mncharity allbery_b: ok. gnome/gaim then. tnx. :) 13:38
allbery_b gaim/pidgin is fairly well known for odd /msg (and /-command in general) behavior
mncharity sigh. :) 13:40
moritz_ re 13:46
13:47 ilbot2 joined
moritz_ mncharity: thanks, now started agian 13:47
mncharity :) 13:50
pugs_svnbot r20267 | putter++ | [elf_d] Tweaks in support of elf_e: HASH::dup, and a second copy of irbuild_ir.
r20267 | putter++ | README: Working elf_d_nomoose build instructions.
diff: dev.pugscode.org/changeset/20267
lambdabot Title: Changeset 20267 - Pugs - Trac
13:55 chris2_ joined 13:57 chris2 left, chris2_ is now known as chris2
pugs_svnbot r20268 | putter++ | [elf_e] Moved global subs to GLOBAL. The current emitter hack for f() is "if I've seen a sub decl for f, emit 'f()', else emit 'GLOBAL::f()'". 14:02
diff: dev.pugscode.org/changeset/20268
lambdabot Title: Changeset 20268 - Pugs - Trac 14:03
14:03 meppl left
mncharity maybe prelude next... 14:04
@tell rouso ping? it should now be possible to build a smop compiler on elf. how goes work? 14:10
lambdabot Consider it noted.
14:25 IllvilJa left 14:40 chris2 left 14:45 chris2 joined 14:48 pmurias joined
pmurias mncharity: hi 14:48
why did you have to create elf_e instead of adding analysis to elf_d? 14:49
14:49 meppl joined
mncharity here's a todo list for someone: tweak autobox to permit mapping subtypes of SCALAR. eg STRING, NUMBER, perhaps INTEGER, UNSIGNED, FLOAT. can support the old SCALAR with no performance hit, and the new subtypes at the cost of a hash lookup. 14:50
hi pmurias. because elf_c doesn't support the class inheritance used in Analysis, and I didn't want elf_d to be self hosting only, given that it is already carrying the non-trivial step of switching IRs. 14:52
I think that was it... let's see... 14:53
yes. "@ISA can be incrementally extended." 14:54
s/the cost of a/the cost of a _extra_/ 14:55
pmurias using Elf instead of ElfLetter would help moving stuff over ;)
mncharity: re autobox how would you do it? 14:56
14:57 alanhaggai left
mncharity in search.cpan.org/src/CHOCOLATE/autob...autobox.xs if the reftype is SCALAR, and the hash does _not_ have a SCALAR entry, use SvIOK and friends to create a more specific reftype, and check that. 14:58
pmurias i see 15:02
mncharity that would give a usage of use autobox SCALAR => undef, STRING => 'STring', NUMBER => 'Number';
hmm, let's see, what are the p6 number types... 15:03
TimToady: S02 mentions Rat, but then doesn't include it in the Im/mutable type list. Does it exist? 15:07
Also, S02 lists Int, Num, Complex, Rat, Bit, Bool, but S29 says Num is a role. What type is a float? 15:09
I also have a very fuzzy recollection that the type of a literal 3 was some weird Int/Num junction thing? 15:11
[particle] Num is float 15:12
mncharity so S29 is wrong, Num is a class not a role, but... classes can also serve as roles? 15:13
[particle] roles are composition-time only
if you request my SomeRole $x, 15:14
it autovivifies a class out of the role in order to type $x
it's in S12 somewhere iirc
mncharity so role RA {...} is identical to class CA {...} except RA has a "can be used as a role at composition-time" bit set? 15:16
[particle] well, not exactly 15:17
if you want to use a class at composition time, the object system tries to make a role out of it for you
it's DWIMmy too
sometimes this may fail
RA and CA differ in their levels of definedness 15:18
15:18 meppl left
mncharity s12: "You can use a role name as a type, but only for constraints, not for declaring actual objects. (However, if you use a role as if it were a class, an anonymous class is generated that composes the role, which provides a way to force a role to test its crony composition for infelicities.)" 15:19
so every instance of a Num gets a unique anonymous class? 15:20
[particle] looks in S29 for the "Num is a role" bit 15:21
mncharity feather.perl6.nl/syn/S29.html#Num
lambdabot Title: S29
mncharity "The following are all defined in the Num role:"
if a class Num can be used as a role, then maybe that's just a "not quite right" phrase. 15:22
TimToady no, there's only one anonymous class generated for Num 15:23
it memoizes
which is fine, since role is immutable
going the other way, you can only make a role out of a class by taking a snapshot of the class's current state
it's still debatable whether that is generally allowed 15:24
[particle] if we can pull it off, it's allowed. 15:25
mncharity so to modify Num's memoized class at runtime, one gets the class with something like $num_class = 3.14.HOW; ? 15:26
15:26 yewenbin left
[particle] why is the class anonymous, if the role is Num and it's immutable? 15:26
TimToady well, that's the metaclass, which presumably has some way of dealing with its bits
it's anonymous so we can pun the name 15:27
15:27 yewenbin joined
mncharity class NumDeriv is 3.14.HOW { ... } 15:27
TimToady and so that JMD doesn't have to have Tie::Array class to go with Array role
it's just Array class and Array role
no, a class is not a metaclass 15:28
mncharity oh... .WHAT ?
class NumDeriv is 3.14.WHAT { ... }
TimToady that is likelier to work
pmurias isn't .HOW the metaclass instance? 15:30
TimToady it's something that does the metaclass role, I suppose... 15:31
mncharity re Array class and Array role... so a class "is for object management", and a role "is for code reuse"... who handles "type which signifies an Array-like interface"? Array::Like ? where Array does Array::Like?
Array::Interface 15:32
TimToady roles also work for interfaces
if you just ignore their suggested implementations
mncharity how does one say 'does Array :but_ignore_its_suggested_implementations' ? 15:33
TimToady $x ~~ Array
mncharity class A2 does Array(:but_ignore_its_suggested_implementations) { ... } ? 15:34
TimToady you mean composition...
just override all the methods
pmurias if you miss one you'll end in trouble 15:36
s/one/any/ 15:37
mncharity a cpan module Array::Interface with all the method bodies overwritten with yadas?
[particle] you can loop over the list of methods and redefine them 15:38
Tene I dunno... sounds like it might be nice to have a convenient way to say "I already provide the interface of this role..."
pmurias mncharity: but_ignore_method_implementations(::Foo) function in a cpan module would be better 15:39
Tene Perl could even check to confirm.
TimToady seems like a pretty easy thing to do
mncharity re loop, feels like ruby. :) (Array.methods - Object.methods).each{...}
moritz_ you can just write a class that 'does Positional', can't you?
you don't have to use the Array class at all 15:40
Tene class MagicArray provides Array { ... }
mncharity re bimi function, nice. though... applied to roles it seems straightforward. applied to classes, less so.
re roles, or no? compile-time role composition has multiple definition as an error. so does()ing an Array::Interface at compile-time... you can't say class A2 does AnImperfectArrayImplementation does Array::Interface {} ? that would go boom, yes? 15:43
TimToady a class can override any colliding methods 15:44
mncharity that requires A2 to mention them all, rather defeating the point of code reuse 15:45
use Xlib2 does PartialXlibImpl does XlibSpec { ...400 pointless functions... }
TimToady so don't use imperfect role models :) 15:46
pmurias mncharity: a partial implementation dosn't match the spec for obvious reasons ;) 15:47
mncharity well, I'm not sure were at the bottom of that one, but mission creep
15:47 yewenbin left
mncharity pmurias: so it looks like a STRING, INTEGER, NUMBER distinction might be most useful 15:48
unless one wants to potentially expose p5 types as the int and uint p6 types...
pmurias/et al :) 15:49
15:49 alanhaggai joined
mncharity but for now, I'll fudge it in a SCALAR prelude, which checks the magic "is it a string or number" phrase, and ... redispatches. +sub_call_cost on each string op. :/ 15:51
~$obj&$obj <- the magic. 15:53
from Class/Multimethods.pm
if anyone knows a better approach... I'd be most interested. "better" == "faster on 5.10" :) 15:54
16:01 justatheory joined
mncharity @perl [0,1].shape 16:08
lambdabot Maybe you meant: keal part pl spell tell url vera
mncharity perl: [0,1].shape
TimToady nobody does S09 yet
mncharity [].end is ... undef? s29 says index of final element. 16:11
final subscript of the first dimension 16:12
TimToady [] is one dimensional as a term 16:14
so just -1
.[] is 0-dimensional
I suppose Array is 0 dimensional :) 16:15
16:17 FurnaceBoy joined
mncharity lol 16:17
s29 Array exists() and delete() appear to be handling their @indices in a confusingly inconsistent manner. 16:19
16:20 ruoso joined
mncharity hmm, or maybe not. nm 16:20
TimToady are they the newer :exists and :delete forms? If not, they're probably fossils 16:23
sorry, don't have time to check it all myself--still trying to catch up with p6l backlog... :)
mncharity newer :exists and :delete? sigh. sounds like I'm doing fossils 16:27
the s29 Array pop/shift/push/unshift sigs should probably be all @array _is rw_. 16:28
anyone have a pointer to "newer :exists and :delete forms"? 16:29
TimToady invocant is rw already 16:30
S12:755
and in fact, splice doesn't need the rw 16:32
presumably the exported versions of methods also mark the first arg as rw
mncharity ah, ok 16:34
pmurias does compile (or copy over if timestamp/checksum matches) and follow all the @DEPENDENCIES outputing a directory of files sound like a sane way to do seperate compilation? 16:39
16:42 wtgee joined
mncharity I don't know of a sane way to do separate compilation. because, 16:46
pmurias mncharity: continue please ;) 16:47
mncharity: would replacing Compiler be the right way to experiment with seperate compilation? 16:49
mncharity timestamp is just a proxy for content. so the "traditional" foo.c -> foo.o is simply a mapping from <foo> to an output file. with the timestamp of the Makefile, and dependencies from *.o to the Makefile, encoding "what compiler and options were used". 16:51
so could add a .p5 suffix recognition to... 16:52
yeah, it would be Compiler. but 16:59
hmm 17:00
17:02 ujwalic joined
mncharity since you are going to need a makefile anyway, I'm wondering whether it might be easier to -s the individual files, and just cat them on to the end of one generated by compiling an empty file (to get the prelude). 17:03
pmurias mncharity: i'm not going to use a makefile
mncharity separate build dirs?
17:03 sri_work left 17:04 japhb joined
pmurias mncharity: sort of 17:04
mncharity and adding a builddir option to elf? or using builddirs via -I, and teaching use() about .p5?
pmurias yes
mncharity yes, which? 17:05
pmurias all of them
the first one really 17:06
mncharity re sane, how long does it take your moose elf to start up? 17:09
(for me, separate compilation would actually take longer... oh, right. compile multiple files in invocation. nm)
is there any kind of perl "mips" module? Ie, so we can tell that my machine is running N perl ops per second, and yours is running M? 17:11
pmurias don't know of any
17:12 ujwalic left
pmurias my current view of seperate compilation is elf --seperate-compilation Elf.pm -o elf creating a directory named "elf" with compiled Elf.pm and all the p6 stuff it uses in it 17:13
mncharity re mips, ah well. re sane, I guess it sound sane, given the performance you are still seeing? I just wish we were sure it's your machine being inherently slow, before adding complexity... . 17:17
17:18 Jedai left
mncharity re api, it would be nice if -o didn't change meaning. so elf --foo=builddir ...everything else is the same as usual... 17:18
17:20 justatheory left
mncharity elf --cachedir=bar/ -o elfx ElfX.pm ? 17:20
pmurias --builddir maybe 17:22
17:23 Auzon joined 17:25 ujwalic joined
mncharity if you use checksums, then you don't have to duplicate a directory tree in bar, and "cache" would be a good name. if you mirror the tree, then "build dir" is a good name. 17:26
pmurias you mean renaming the filenames to checksums? 17:28
i think duplicating the tree is better
mncharity re checksums, yes. eg, what STD_red is doing in STD_RED_CACHEDIR. 17:29
pugs_svnbot r20269 | putter++ | [elf_d] Compiler now has hooks for caching to avoid compilation.
diff: dev.pugscode.org/changeset/20269
lambdabot Title: Changeset 20269 - Pugs - Trac
mncharity btw, you are setting STD_RED_CACHEDIR, yes?
so you should be able to just inherit from Compiler now, and use whichever scheme you want. 17:30
17:30 Greyscale left
pmurias mncharity: yes ;) 17:30
mncharity :) 17:31
feel free to change the hooks. I was just sketching them in to see what it would look like.
anyone know if there is a standard method name for shallow copy? eg, dup or clone in ruby? 17:34
pmurias i think it's clone 17:35
mncharity ah, ok. 17:38
tnx
17:44 ujwalic left 18:08 barney joined 18:19 aindilis left
pmurias mncharity: does perl5.10 run elf faster then perl5.8.8 18:21
18:24 aindilis joined 18:26 stevan_ left
mncharity pmurias: don't know, checking... 18:27
18:28 Fokez left
mncharity 9.5 vs 10.9 sec 18:29
so not by much
pmurias for elf_d 18:39
?
18:41 chris2 left
pmurias i get ~10sec for nomoose 18:43
18:44 meppl joined
mncharity for elf_e moose 18:49
lunch. bbl & 18:52
re ~10sec for nomoose, :/
could it be something silly like perl was built unoptimized? 18:53
19:00 eternaleye_ joined
mncharity pmurias: note that when working on the emitter, you can avoid one compile cycle by doing elfw EmitWhatever.pm -o elfx ElfX.pm The EmitWhatever is used to generate elfx, rather than the emitter compiled into elfw. 19:03
19:04 cmarcelo joined 19:46 barney left 19:51 stevan_ joined 19:56 ujwalic joined 19:57 meppl left 20:19 eternaleye joined, eternaleye_ left
literal <mncharity> literal: are you setting PUGS_EMBED? 20:25
yes, to "parrot perl5"
moritz_ don't embed parrot 20:27
it's kinda b0rked atm
the PIL/PIR emitter, that is
literal moritz_: my buils still fails in the same way if I don't embed parrot 20:37
build
sial.org/pbot/30800
lambdabot Title: Paste #30800 from "literal" at 194.144.99.91
moritz_ literal: did you try a 'make realclean' first? 20:46
*** Wasn't able to find 'Parrot_stub.o' (this may be a problem)...
this looks like it shouldn't happen without parrot
literal it's a clean make
moritz_ pugs: say $?VERSION 20:47
exp_evalbot OUTPUT[6.2.13␤]
moritz_ pugs: say $?REVISION
exp_evalbot OUTPUT[*** Undeclared variable: ("$?REVISION",MkPad (padToList [("$_",PELexical {pe_type = (mkType "Scalar"), pe_proto = <Scalar:0xb689e30c>, pe_flags = MkEntryFlags {ef_isContext = True}, pe_store = <ref:0xb68a66d8>}),("@_",PELexical {pe_type = (mkType "Array"), pe_proto = <Array:0xb689f7f8>,
..pe_flags = MkEntryFlags {ef_isContext = False}, pe_s...
TimToady pugs: say $?PUGS_VERSION 20:49
exp_evalbot OUTPUT[Perl6 User's Golfing System, version 6.2.13, October 17, 2006 (r18093)␤]
moritz_ literal: current versions are broken. r18093 is known to compile with GHC 6.6.1 at least 20:52
literal the whole thing actually compiles, as you see (right?), seems to be something else a matter 20:53
moritz_ aye 20:55
and r18093 actually *works*
like, it executes code. Smoking works etc.
21:10 jferrero joined 21:18 eternaleye_ joined 21:20 syle joined 21:23 eternaleye left 21:26 ez joined
pugs_svnbot r20270 | Darren_Duncan++ | ext/Muldis-DB/ : temp removed Value.pm , it will return later 21:26
diff: dev.pugscode.org/changeset/20270
lambdabot Title: Changeset 20270 - Pugs - Trac
21:39 wtgee left
pugs_svnbot r20271 | Darren_Duncan++ | renamed ext/Muldis-DB/ to ext/Muldis-Rosetta/ 21:47
diff: dev.pugscode.org/changeset/20271
lambdabot Title: Changeset 20271 - Pugs - Trac
21:47 pmurias left 21:49 syle left 21:53 alanhaggai__ joined 21:59 eternaleye joined
pugs_svnbot r20272 | Darren_Duncan++ | ext/Muldis-Rosetta/ : part 1 of renaming Muldis DB to Muldis Rosetta - file/path renames 22:05
diff: dev.pugscode.org/changeset/20272
lambdabot Title: Changeset 20272 - Pugs - Trac
22:06 IllvilJa joined 22:07 alanhaggai left 22:08 FurnaceBoy left 22:11 ez left 22:26 eternaleye_ left 22:30 meppl joined
mncharity wow. amazing weather in BOS. the expression is "weather fronts collide", but how often are things lit so you can watch "there's a cloud zooming along... slam, smoosh, rip, shred, blam... no more cloud" in a few 10's of seconds. 22:35
22:42 iblechbot left
mncharity anyone know if ref() is still part of p6? the few instances of it in t/ look fossil-like. 23:01
Auzon I'm pretty sure it's replaced by .WHAT and such, but I'm looking 23:02
23:03 thestarslookdown joined 23:06 wtgee joined
mncharity thanks! :) 23:07
Auzon I can't actually find the specs in the Pugs SVN repo though :-/
You can also use "$var ~~ HASH"
pugs: $var = {}; $var ~~ HASH
exp_evalbot OUTPUT[*** ␤ Unexpected " ="␤ expecting "::"␤ Variable "$var" requires predeclaration or explicit package name␤ at /tmp/ahQ3Jtuqic line 1, column 5␤]
Auzon pugs: $var = {}; $var ~~ HASH;
exp_evalbot OUTPUT[*** ␤ Unexpected " ="␤ expecting "::"␤ Variable "$var" requires predeclaration or explicit package name␤ at /tmp/NPywboueAz line 1, column 5␤]
Auzon hm... I'm missing something 23:08
[particle] rakudo: my $var = {}; say $var ~~ Hash;
exp_evalbot OUTPUT[0␤]
[particle] rakudo: my %; say %var ~~ Hash; 23:09
Auzon perl6: $var = {}; $var ~~ Hash;
exp_evalbot OUTPUT[Statement not terminated properly at line 1, near "%; say %va"␤current instr.: 'parrot;PGE::Util;die' pc 120 (runtime/parrot/library/PGE/Util.pir:82)␤called from Sub 'parrot;Perl6::Grammar;statementlist' pc 16390 (src/gen_grammar.pir:2404)␤called from Sub
..'parrot;Perl6::Grammar;statement_block' pc 13561 (src/gen_grammar.pir:1390)␤called f...
kp6: OUTPUT[no method 'APPLY' in Class 'Undef'␤ at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 345␤ KindaPerl6::Runtime::Perl5::MOP::__ANON__('HASH(0x824c084)', 'APPLY') called at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 169␤
..main::DISPATCH('HASH(0x824c084)', 'APPLY') called at compiled/perl5-kp6-mp6/lib/Kin...
..pugs: OUTPUT[*** ␤ Unexpected " ="␤ expecting "::"␤ Variable "$var" requires predeclaration or explicit package name␤ at /tmp/K7PqhQK2zI line 1, column 5␤]
Auzon Oops.
exp_evalbot ..rakudo: OUTPUT[Scope not found for PAST::Var '$var'␤current instr.: 'parrot;PCT::HLLCompiler;panic' pc 158 (src/PCT/HLLCompiler.pir:103)␤called from Sub 'parrot;PAST::Compiler;post_children' pc 1207 (src/PAST/Compiler.pir:144)␤called from Sub 'parrot;PAST::Compiler;inline' pc 4310
..(src/PAST/Compiler.pir:1062)␤called from Sub 'parrot;PAST::Compiler;post...
Auzon sorry [particle]
[particle] rakudo: my %var; say %var ~~ Hash;
exp_evalbot OUTPUT[1␤]
[particle] rakudo: my $var = {}; say $var.WHAT; 23:10
exp_evalbot OUTPUT[Code␤]
Auzon Then again, hashrefs are dead and captures should be used instead, which I don't know yet 23:11
23:13 Limbic_Region joined
mncharity ok, thanks. 23:15
23:18 ujwalic_ joined
mncharity okay... WHAT's up. 23:24
23:25 justatheory joined
mncharity so who's going to translate the classic "Who's on first?" joke to p6? 23:26
Auzon pugs: $.WHO 23:28
exp_evalbot OUTPUT[*** Undeclared variable: ("$__SELF__",MkPad (padToList [("$_",PELexical {pe_type = (mkType "Scalar"), pe_proto = <Scalar:0xb699ee40>, pe_flags = MkEntryFlags {ef_isContext = True}, pe_store = <ref:0xb67de42c>}),("@_",PELexical {pe_type = (mkType "Array"), pe_proto = <Array:0xb699e2dc>,
..pe_flags = MkEntryFlags {ef_isContext = False}, pe_st...
Auzon I think we need .WHO first ;)
mncharity lol 23:29
23:35 justatheory left
mncharity pugs: Str ~~ Object 23:39
exp_evalbot RESULT[Bool::True]
mncharity pugs: Str ~~ Any
pugs: Str.does(Str) 23:40
Auzon hm, it didn't like that one :P
mncharity :)
pugs: Str.isa(Object)
[particle] rakudo: Str ~~ Any
exp_evalbot RESULT[Method 'perl' not found for invocant of class 'Bool'␤current instr.: '_block10' pc 35 (EVAL_11:16)␤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 785 (src/PCT/HLLCompiler.pir:458)␤called from Sub 'parrot;PCT::HLLCompiler;evalfiles' pc 1067 (src/PCT/HLLCompiler.pir:587)␤called from Sub
..'parrot;PCT::HLLCompiler;command_line' pc 1246 (src...
[particle] rakudo: say Str ~~ Any
exp_evalbot OUTPUT[0␤]
[particle] rakudo: say Perl6Str ~~ Any 23:41
exp_evalbot OUTPUT[0␤]
[particle] hrm
Auzon rakudo: say Any ~~ Any
[particle] rakudo: say String ~~ Any
exp_evalbot OUTPUT[1␤]
OUTPUT[0␤]
mncharity rakudo: say Str ~~ Object
wolverian mncharity, leading space :)
rakudo: say Str ~~ Object 23:42
exp_evalbot OUTPUT[1␤]
mncharity sigh. thanks
23:42 ujwalic left
wolverian np 23:42
[particle] looks like rakudo still needs a little jiggering in its type hierarchy
mncharity pugs: Str ~~ Any
pugs: 3
[particle] rakudo: say Junction ~~ Any
exp_evalbot OUTPUT[0␤]
Auzon rakudo: say Object ~~ Any
exp_evalbot OUTPUT[0␤]
[particle] rakudo: say Junction ~~ Object
exp_evalbot OUTPUT[1␤]
mncharity pugs: 3 23:43
exp_evalbot RESULT[3]
[particle] rakudo: say Any ~~ Object
exp_evalbot OUTPUT[1␤]
mncharity pugs: Str ~~ Any
exp_evalbot RESULT[Bool::True]
mncharity pugs: Str.isa(Any) 23:44
exp_evalbot RESULT[Bool::True]
23:45 tango1 joined
tango1 hey everyone 23:46
Auzon Hi.
tango1 wow a long time since I been in an IRC chat room
Tene Hi! 23:47
wolverian welcome back, then :) 23:48
tango1 been at least over 4 years
23:53 xdg left, xdg joined 23:56 meppl left

Powered by Cro and the Raku® Programming Language.

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

Thank you!