pugscode.org/ | nopaste: sial.org/pbot/perl6 | pugs: [~] <m oo se> (or rakudo:, kp6:, elf: etc.) (or perl6: for all) | irclog: irc.pugscode.org/
Set by Tene on 29 July 2008.
cognominal perl6: class H::A { } class H::B is H::A { } 00:00
p6eval rakudo 31096: OUTPUT[Statement not terminated properly at line 1, near "class H::B"␤␤current instr.: 'parrot;PGE::Util;die' pc 119 (runtime/parrot/library/PGE/Util.pir:82)␤]
..elf 22236: OUTPUT[Parse error in: /tmp/tUtCVRjq7I␤panic at line 1 column 0 (pos 0): Can't understand next input--giving up␤WHERE: class H::A { } class H::B is H␤WHERE:/\<-- HERE␤ STD_red/prelude.rb:99:in `panic'␤ STD_red/std.rb:76:in `scan_unitstopper'␤ STD_red/std.rb:224:in `comp_unit'␤
..STD_r...
..pugs: RESULT[undef]
cognominal ouch
s1n perl6: class H::A { }; 00:01
p6eval elf 22236: RESULT[undef␤] 00:02
..rakudo 31096: RESULT[{ ... }]
..pugs: RESULT[undef]
s1n perl6: class H { }; class H::A { }; class H::B is H::A { }; say (new H::B).WHAT 00:03
p6eval rakudo 31096: OUTPUT[Attempt to inherit from non-existent parent class␤current instr.: 'die' pc 12417 (src/gen_builtins.pir:7815)␤]
..elf 22236: OUTPUT[Parse error in: /tmp/TGQ7pa5tWe␤panic at line 1 column 29 (pos 29): Can't understand next input--giving up␤WHERE: class H { }; class H::A { }; class H::B is H::A { }; say (n␤WHERE: /\<-- HERE␤ STD_red/prelude.rb:99:in `panic'␤ STD_red/std.rb:76:in
..`s...
..pugs: OUTPUT[H::B␤]
s1n hmm bummer
cognominal H is a package just like in Perl5. 00:04
00:06 [particle]1 joined
cognominal perl6: class A { my class B {} method doit() { say B.WHAT } } A.new().doit() 00:10
TimToady please get used to putting ; after } if you inline blocks like that
p6eval rakudo 31096: OUTPUT[Statement not terminated properly at line 1, near "method doi"␤␤current instr.: 'parrot;PGE::Util;die' pc 119 (runtime/parrot/library/PGE/Util.pir:82)␤]
..elf 22236: OUTPUT[Parse error in: /tmp/p2gztHjwnK␤panic at line 1 column 11 (pos 11): Missing right brace␤WHERE: class A { my class B {} method doit() ␤WHERE: /\<-- HERE␤ STD_red/prelude.rb:99:in `panic'␤ STD_red/std.rb:255:in `_block_rest'␤ STD_red/std.rb:242:in `block in block'␤
.....
..pugs: OUTPUT[B␤]
cognominal TimToady, thx
are my expectations correct? 00:11
TimToady H::B.WHAT should print H::B
namespace are for naming things
*namespaces
cognominal perl6: class A { my class B {} ; method doit() { say B.WHAT } }; A.new().doit()
p6eval pugs: OUTPUT[B␤]
..rakudo 31096: OUTPUT[Could not find non-existent sub my␤current instr.: 'parrot;A;_block25' pc 160 (EVAL_14:62)␤]
..elf 22236: OUTPUT[syntax error at (eval 121) line 6, near "{ package A::B"␤BEGIN not safe after errors--compilation aborted at (eval 121) line 7.␤ at ./elf_f line 3861␤]
TimToady that should just print B, because it's rooted in a lexical scope, not in a package 00:12
cognominal pugs++ # knows about inner classes
TimToady s1n: naming has nothing to do with inheritance 00:13
cognominal perl6: class H::A { }; class H::B is H::A { }; say H::B.WHAT 00:15
p6eval rakudo 31096: OUTPUT[Attempt to inherit from non-existent parent class␤current instr.: 'die' pc 12417 (src/gen_builtins.pir:7815)␤] 00:16
..pugs: OUTPUT[H::B␤]
..elf 22236: OUTPUT[Parse error in: /tmp/zgGjJkQWU6␤panic at line 1 column 17 (pos 17): Can't understand next input--giving up␤WHERE: class H::A { }; class H::B is H::A { }; say H␤WHERE: /\<-- HERE␤ STD_red/prelude.rb:99:in `panic'␤ STD_red/std.rb:76:in `scan_unitstopper'␤
..STD_r...
00:16 [particle] left 00:19 pbuetow joined 00:41 braceta left, sail0r joined 00:49 sail0r left 01:03 justatheory left 01:08 barbeq1 joined 01:09 jferrero left 01:12 barbeq1 left 01:39 wknight8111 joined 01:59 pbuetow_ joined 02:04 wknight8111 left 02:10 pbuetow left 02:20 meppl left 02:36 Chillance left 02:40 alester_ joined
s1n how does the lambabot work? 02:51
lambdabot
nm 02:56
Myoma s1n: It mostly doesn't
s1n Myoma: does the tell command work?
Myoma @tell s1n I don't know
lambdabot Consider it noted.
s1n how do i get those messages? 02:57
lambdabot s1n: You have 2 new messages. '/msg lambdabot @messages' to read them.
s1n ahh
@tell pmichaud i have most of it working, a few things need to be worked out, PM me, i have a few questions 02:58
lambdabot Consider it noted.
03:02 Limbic_Region left 03:05 mbb joined 03:09 mbb left 03:15 eternaleye joined 03:23 xinming joined 03:37 snooper joined 03:43 drbean_ joined 03:44 drbean_ left 03:45 demq joined 03:49 pbuetow_ left 04:09 jinjing joined 04:42 ashizawa joined 04:50 Maghnus joined 05:18 alester_ left 05:22 Psyche^ joined 05:34 Psyche^ is now known as Patterner 05:56 kanru joined 06:08 araujo joined 06:28 xinming left 06:29 xinming joined 06:56 Jedai joined 07:01 ab5tract joined 07:25 xiaoyafeng left 07:26 eternaleye left 07:28 eternaleye joined 07:33 abra joined 07:40 BinGOs joined 07:57 ilogger2 joined 08:08 charsbar_ joined 08:19 hanekomu joined 08:38 kanru joined 08:41 schmalbe joined 08:42 Jedai joined 08:51 Ontolog joined
Ontolog shit 08:51
anyone else get "svn: Decompression of svndiff data failed" when doing a checkout? 08:52
08:53 jinjing_ joined
Ontolog Also for rakudo I can't `make perl6`, I get the error: Null PMC access in get_bool() 08:54
oops forget to mention the "svn: Decompression of svndiff data failed" error is for checking out pugs 08:59
so i'm screwed with both pugs and rakudo
no perl6 for me
anyone know of a working revision of either?
ewilhelm is that error for checkout or update? 09:00
Ontolog checkout, however I just ran it again and the error goes away... go figure
I still have the rakudo error though 09:01
ewilhelm 31053 seemed to work about 24hrs ago
09:02 xiaoyafeng joined
Ontolog i see i'll check it out 09:02
09:10 kanru left 09:11 kanru joined 09:16 pmurias joined 09:25 xuser joined 09:33 iblechbot joined 09:41 eternaleye joined 09:47 pmurias left, pmurias joined, eternaleye left 09:51 larsen_ joined 10:07 masak joined 10:11 eternaleye joined 10:13 eternaleye left 10:15 eternaleye joined, elmex joined 10:20 zamolxes joined 10:21 sri_work joined 10:35 larsen_ left 10:49 pbuetow_ joined
Ontolog Does Rakudo need people to write any pure Perl stuff? 10:49
pure Perl6 stuff
I want to start writing Perl6 and I figured making some kind of contribution to Rakudo would be best
or Pugs for that matter (since it would be pure Perl6) 10:50
10:52 Myoma joined
pmurias Ontolog: would you be confortable with helping with the test suit 10:58
?
* comfortable
Ontolog sure 11:00
pmurias moritz_: ping
Ontolog: i'm not really a test suit hacker, so i can't really help you there 11:01
11:02 sri_work_ joined
Ontolog haha ok 11:02
pmurias Ontolog: there seem to be a t/TASKS
Ontolog: but if you feel like hacking in C/haskell i'm sure i could find you a smop task 11:05
Ontolog haha i'm not too well versed in either C or Haskell 11:10
actually my goal is just to start writing some Perl 6 code 11:11
so anything pure-perl needed for Perl 6 projects I can help
masak Ontolog: have you checked out November? :) 11:12
github.com/viklund/november/ 11:13
lambdabot Title: viklund's november at master — GitHub
masak www.november-wiki.org/
lambdabot Title: Main_Page
Ontolog looks interesting 11:19
is Rakudo ready for a large project like a Wiki? 11:20
masak Ontolog: yes and no
Ontolog last time I checked it didn't actually implement much Perl 6 yet
masak by trying it, we're making it more ready
Ontolog: it does more than you think
Ontolog: use.perl.org/~masak/journal/37303 11:21
lambdabot Title: Journal of masak (6289)
11:23 sri_work left
Ontolog well, my goal really is to learn Perl 6, so I (selfishly) don't want to spend time encountering so many bugs or missing features 11:26
so i am using pugs for now
masak sounds like a wise move 11:27
but don't expect to avoid bugs just by using Pugs :)
they're just distributed differently than in Rakudo
as are the "not-yet-a-feature" and "it-shouldn't-work-exactly-like-this" areas 11:29
11:38 abra joined, xiaoyafeng left
Ontolog many important modules in perl5 are implemented as XS modules; will this also be the case with Perl 6? and if so, how will 'XS' modules be implemented? As parrot vm code? 11:40
masak hm, seems a bit slow here today,,, ordinarily, people would have jumped on such a question :) 11:46
I'm not very qualified to answer, except that I don't think it will work like XS works in Perl 5
it will probably work through Parrot in some way, yes 11:47
&
Ontolog i was hoping the need for writing XS modules would be lesser as Perl 6 is suppose to be a lot faster than Perl 5 yes? (if you strongly type everything, etC) 11:48
11:51 pmurias left
Ontolog I'v always wanted to do something like this: say $_ if $_ ~= /e/ for @names; 11:59
you see I have both a trailing if AND for
I want to be able to 'chain' this way
is Perl 6 suppose to let me?
12:24 larsen_ joined
masak Ontolog: yes. 12:28
Ontolog but it doesn't work on pugs :( can you see if that line works on rakudo? 12:29
12:30 ashizawa joined
masak perl6: .say if $_ % 2 for 1..10 12:31
p6eval pugs: OUTPUT[*** ␤ Unexpected "for"␤ expecting operator␤ at /tmp/11AVX361Yz line 1, column 16␤]
..elf 22236: OUTPUT[Use of uninitialized value in concatenation (.) or string at ./elf_f line 3408.␤syntax error at (eval 119) line 5, near "{␤->say"␤ at ./elf_f line 3861␤]
..rakudo 31102: RESULT[0]
Ontolog haha
masak hm, rakudo parses it
but does the Wrong Thing
actually, I find nothing in the specs or in the test suite right now about stacked if and for statement modifiers 12:37
though I'm certain I've seen discussion about it over the mailing lists
Ontolog man it would be great if we could chain those modifiers 12:44
it would really go a long way into making perl more like a natural language
masak Ontolog: do you know about the modifiable grammar of Perl 6? 12:46
Ontolog nope
masak even if it's not allowed in Standard Perl 6 (and I think it is), you could fairly easily allow it in your programs
just by modifying the compiler grammar at the start of the program 12:47
Ontolog i see, that's interesting... but I really really hope it's in the standard because things like that (adding some non-standard grammar) would never fly in a corporate setting 12:48
masak I can see why not in some cases, yes. 12:51
though the Perl 6 design has been laid out so as to make this type of grammar changes very non-invasive (to other modules, or users of the module) 12:53
anyway, we're not there yet. :)
Ontolog where is the doc on this grammar stuff?
masak modifying the grammar on-the-fly is far down on the lists of all the current implementations
Ontolog ohhh 12:54
i see
masak Ontolog: perlcabal.org/syn/S06.html#Macros
lambdabot Title: S06
Ontolog which is the official mailing list for Perl 6 devs? I would like to email about this chained modifiers thing
masak Ontolog: perl6compiler has a lot of Perl 6 devs 12:57
if you have time though, I'd advice you to hit the perl6language archives first
just to see what the consensus was to this question 12:58
12:59 meppl joined 13:05 xiaoyafeng joined 13:07 sri_work joined 13:28 sri_work_ left
Ontolog perl6: .say "Parent: $_" for ('Chris','Wendy','Deek') eq 'Chris'|'Wendy'; 13:34
p6eval rakudo 31112: OUTPUT[Statement not terminated properly at line 1, near "\"Parent: $"␤␤current instr.: 'parrot;PGE::Util;die' pc 119 (runtime/parrot/library/PGE/Util.pir:82)␤]
..pugs: OUTPUT[*** ␤ Unexpected "\""␤ expecting operator␤ at /tmp/ESkdXvrbwE line 1, column 6␤]
..elf 22236: OUTPUT[Parse error in: /tmp/jTmKTS0GW2␤panic at line 1 column 0 (pos 0): Can't understand next input--giving up␤WHERE: .say "Parent: $_" for ('Chris'␤WHERE:/\<-- HERE␤ STD_red/prelude.rb:99:in `panic'␤ STD_red/std.rb:76:in `scan_unitstopper'␤ STD_red/std.rb:224:in `comp_unit'␤
..STD_r...
Ontolog perl6: for ('Chris','Wendy','Deek') { if $_ eq 'Chris'|'Wendy' { say $_ } } 13:35
p6eval elf 22236: RESULT[␤]
..pugs, rakudo 31112: OUTPUT[Chris␤Wendy␤]
masak Ontolog: and you don't need parens around the list if you don't want to 13:38
13:39 Jedai left, schmalbe left 13:40 Jedai joined 13:45 missingthepoint joined 13:47 larsen_ left
Ontolog i see 13:48
13:51 pmurias joined 13:53 jinjing_ left
pmurias Ontolog: re XS i think it will be implementation dependend 13:53
Ontolog there will hopefully be a dominant implementation yes? 13:54
and i would imagine that is the parrot one
but who knows ... it seems there is still a lot to go
pmurias parrot is a vm
13:56 jinjing joined
pmurias it is possible there will be only one implementation of the Perl 6 compiler in the future as the AST will have to be specced and the grammar shared 13:57
14:02 alester_ joined, Chillance joined
pmurias Ontolog: why do you hope that there will be one dominant implementation? 14:03
14:03 apeiron joined
Ontolog for a couple reasons 14:04
pmurias ;) 14:05
Ontolog so, eventually, everyone who wants to contribute works on the one implementation
also, there is consistency among platforms
i can see a reason for having competing implementations during this phase but i don't see any reason for it in the future 14:06
pmurias i think most of the components of the final Perl6es will be interchangable 14:07
Ontolog you mean most of the components will be pure perl 6? 14:08
pmurias all of the components of the compiler should be pure perl 6
and there a lots of reasons for running on other runtimes like the JVM,CLR or javascript 14:09
s/reasons for/benefits of
14:13 alester_ left 14:22 jinjing left 14:23 hanekomu left 14:24 jinjing joined
masak wants Perl 6 on the JVM 14:24
pretty please, with sugat on top?
s/sugat/sugar/
14:28 pbuetow_ left
pmurias masak: why do you use the JVM? 14:30
masak pmurias: in my work, we develop on top of it 14:31
or rather on top of the Eclipse RCP platform 14:32
I'm about to sneak some Groovy into the Java environment
but sneaking in some Perl 6 would of course be even better :) 14:33
pmurias have you tried Scala?
masak not really
just heard interesting things about it
pmurias it seems to be a better java 14:34
masak well, it's the usual situation with tools and market share, I guess 14:36
Scala may be a better Java, but it doesn't have the Eclipse RCP platform
cognominal RCP?
masak "Rich Client Platform" 14:37
that's what they call their extensible basis to Eclipse
14:37 rindolf joined
pmurias masak: can't Scala use everything java does? 14:37
masak pmurias: I don't know. maybe in theory, it can. 14:38
14:40 jan_ joined
ruoso masak, I'm pretty confident that the JVM can't support Perl 6 features... 14:41
masak ruoso: I believe you.
still, even a cut-down version would probably rock 14:42
pmurias ruoso: hi
ruoso hi pmurias
14:43 zamolxes left
pmurias ruoso: i'm writing a smop to haskell binding so i can write a mold debugger in haskell 14:43
it's a bit funny that we are using both C and haskell which are opposites 14:46
cognominal I found funny that the haskell and Perl community could mix together. Haskell was originally all about researh and Per about raw practicallity. 14:48
pmurias the goals of the reasearch very propably to achieve praticallity 14:49
14:49 jferrero joined
pmurias s/very/were 14:50
* practicallity
pmurias hates typos 14:51
cognominal that rarely suceeds. In fact, until recently SPJ was not wanting his language to be too successful to avoid to deal with a user base.
ruoso pmurias, I'm pretty curious to see how smop and haskell would look like :) 14:52
cognominal makes many typos and is bad at writing English
rindolf cognominal: SPJ == Simon Peyton Jones?
cognominal yup 14:53
I meant his implementation
14:55 larsen_ joined 14:59 iblechbot left 15:02 Ontolog left
pmurias ruoso: where should i put the the smop in haskell embeding stuff when i check it in? 15:02
ruoso it's embedding haskell? or being embedded in haskell?
pmurias it's being embedded in haskell 15:03
15:03 keepguessing joined
ruoso reading groups.google.com/group/perl.perl6....31e4830f42 15:03
lambdabot Title: arrayref/hashref in spectest suite - perl.perl6.language | Google Groups, tinyurl.com/68okmh
ruoso if @a = @b imply a copy of the array @b to the array @a... does that mean eager evaluation? 15:04
keepguessing Please help me interpret the following RegEx in perl ([^\[]+)\s+
ruoso I think '@a= @b' can't always mean a copy... otherwise 'my @a = map { ... }, @b' would always be eager... 15:06
pmurias (anything except ]) one or more times and one or more whitespace
pmichaud rakudo doesn't understand multi-level namespaces yet, I don't think.
lambdabot pmichaud: You have 1 new message. '/msg lambdabot @messages' to read it.
pmichaud oops, sorry, was reading scrollback and didn't realize it was scrollback 15:07
keepguessing thank you
pmurias: I think u meant (anything except [) 15:08
pmurias keepguessing: yes ;)
keepguessing alright thanks ...
pmurias keepguessing: se perlre manual man page or Programming Perl for more detail
* see 15:09
15:14 Jedai left 15:15 Jedai joined
keepguessing thanks :) 15:18
15:19 [particle] joined
pmurias ruoso: we could require my @a := map {...},@b 15:21
or @a = map {...},@b would do a COW 15:25
15:26 iblechbot joined
pmurias ruoso: got hello world working from haskell where should i put it in 15:30
rakudo_svn r31116 | pmichaud++ | [rakudo]: Make 'defined' into a named unary. (See related RT #58830, masak++) 15:32
15:36 [particle]1 joined 15:39 jinjing left 15:41 masak left, Myoma left 15:44 Myoma joined 15:46 [particle] left 15:51 cognominal joined 15:54 kanru left 16:05 keepguessing left 16:18 missingthepoint left 16:51 ab5tract joined 16:53 smtms joined
moritz_ re 17:13
17:16 wknight8111 joined 17:24 mncharity joined
pmurias mncharity: hi 17:28
17:30 larsen_ left
pugs_svn r22237 | putter++ | [elfish/on_sbcl] Fleshed out development environment. 17:32
r22237 | putter++ | Synced with current elf_g (from elf_f). Added CL-PPCRE download aid.
r22237 | putter++ | fib() runs.
mncharity hi pmurias :) 17:33
btw, fib(37) is 10x p5 speed.
www.weitz.de/cl-ppcre/ 17:34
lambdabot Title: CL-PPCRE - Portable Perl-compatible regular expressions for Common Lisp
mncharity note www.weitz.de/cl-ppcre/#filters 17:35
lambdabot Title: CL-PPCRE - Portable Perl-compatible regular expressions for Common Lisp
mncharity well, that's my "nibble at p6" for the day. cheers. 17:38
rakudo_svn r31123 | pmichaud++ | [rakudo]: spectest-progress.csv update: 161 files, 3332 passing tests, 4 failing 17:41
pugs_svn r22238 | moritz++ | [t/spec] fudged gather.t for rakudo 17:42
pmurias mncharity: got to move a computer, i'll look at the commit later 17:43
mncharity pmurias: so the current elf v STD.pm plan is for STD_blue to try using gimme5 directly, rather than creating a STD.pm {*} derivative. Making tracking STD changes much easier. Hopefully will be able to do that though 17:44
nifty :)
...through STD_blue becoming a STD_red equivalent (ie, elf can backend bootstrap on it), and ideally until elf can run STD.pm (or a tweak of it) itself (ie, elf frontend bootstrap). 17:45
As STD_red became increasingly out of sync with STD.pm, it lost the ability to provide feedback to STD development. Hopefully we have that back now. 17:47
pmurias: apropos smop, I'd appreciate your thoughts on the sbcl backend doing true p6 oo. Last time I checked, p6l was still arguing over what .WHAT() would return. So I was leaning towards a p5-backend-like "anything simple which is good enough for now". But unlike with p5, 17:52
there isn't the pressing performance motivation to not do the right thing. So... is the Right Thing oo wise, p6opaque, BUILD, etc, something which is currently well spec'ed? Is anyone (smop) actually using it (ie, would it be a 1st-implementation spec debugging exercise)? 17:54
tnx. will backlog (eventually). 17:56
17:56 mncharity left 18:00 ashizawa left
pugs_svn r22239 | moritz++ | [t/spec] unfudged type-based.t for rakudo 18:05
18:05 pmurias left
pugs_svn r22240 | moritz++ | [t/spec] split quoting.t into quoting.t and quoting-unicode.t 18:06
r22240 | moritz++ | (This is so that non-Unicode aware implementations can run at least parts of
r22240 | moritz++ | the tests). Also fudged a bit for rakudo
rakudo_svn r31127 | moritz++ | [rakudo] add quoting tests to spectest_regression 18:13
pugs_svn r22241 | cjfields++ | add simple test for RT# 58678 (along with TODO information re: possible smart matching changes with grammars, moritz++ ) 18:19
ab5tract where can i learn more about STD_red? 18:30
elf source, i assume, but wiki / website, etc? 18:31
moritz_ here, hopefullly ;)
all I know is that 1) it's a ruby translation of STD.pm and 2) mncharity doesn't keep that much external documentation about his elf/ stuff 18:32
ab5tract right on. thats what i was intuiting but you know what they say about about assuming
STD_red as the ruby translation of STD.pm, i mean 18:33
moritz_ perl6: my Str $x; $x ~= 'a'; say $x 18:35
p6eval rakudo 31128: No output (you need to produce output to STDOUT)
..elf 22241: OUTPUT[Unknown rule: infix_postfix_meta_operator:=␤It needs to be added to ast_handlers.␤ at ./elf_f line 1918␤]
..pugs: OUTPUT[Stra␤]
18:40 pmurias joined 18:41 wknight8111 is now known as wknight-away 18:48 luqui joined 18:56 apple-gunkies joined, apple-gunkies is now known as hudnix 18:59 wknight-away left, Auzon joined
ewilhelm thinks he's spotted a logic error in E06 19:05
moritz_ ewilhelm: only the Synopsis are maintained, iirc 19:06
ewilhelm "if %person{title} ne $known_title" should be "unless %person{title} eq $known_title"
moritz_, this appears to be the best explanation of junctions at the moment -- and it is a *logic* error, not a syntax change 19:07
moritz_ ewilhelm: I then guess "patches welcome" ;)
ewilhelm perl6: my $known_title = 'Mr' | 'Mrs' | 'Ms' | 'Dr' | 'Rev'; say "unknown" if "Mr" ne $known_title; 19:08
p6eval elf 22241: OUTPUT[No viable candidate for call to multimethod infix__124($,$,$,$,$) at (eval 119) line 3␤ at ./elf_f line 3861␤]
..pugs, rakudo 31128: OUTPUT[unknown␤]
ewilhelm moritz_, to where should the patch go?
moritz_ ewilhelm: to p6l 19:09
source is at svn.perl.org/perl6/doc/trunk/design/exe/E05.pod
ewilhelm thanks 19:10
moritz_, should I update the print to say and such or just switch the logic? 19:12
pugs_svn r22242 | pmurias++ | [smop] smop embedding in haskell, ony hello world atm
ewilhelm perl6: my %hash = (a => 2); say "%hash{a}" 19:13
p6eval rakudo 31130: OUTPUT[Could not find non-existent sub a␤current instr.: '_block19' pc 150 (EVAL_13:53)␤]
..elf 22241: OUTPUT[%hash{a}␤]
..pugs: OUTPUT[*** No such subroutine: "&a"␤ at /tmp/tsjPnvrN9B line 1, column 33-34␤]
pmichaud ewilhelm: you probably mean %hash<a>
or %hash{'a'}
ewilhelm for interpolation?
pmichaud (rakudo doesn't recognize either of these at the moment, there it's {%hash<a>} 19:14
yes, for interpolation
ewilhelm perl6: my %hash = (a => 2); say "%hash{'a'}"
p6eval rakudo 31130: OUTPUT[%hasha␤]
..elf 22241: OUTPUT[%hash{'a'}␤]
..pugs: OUTPUT[2␤]
ewilhelm perl6: my %hash = (a => 2); say "%hash<a>"
p6eval elf 22241, rakudo 31130: OUTPUT[%hash<a>␤]
..pugs: OUTPUT[2␤]
moritz_ pugs++
ewilhelm perl6: my %hash = (a => 2); say "{%hash<a>}"
p6eval elf 22241: OUTPUT[{%hash<a>}␤]
..pugs, rakudo 31130: OUTPUT[2␤]
ewilhelm ah, so {a} is a sub call 19:15
pmichaud anything in braces generally a closure.
*is generally
ewilhelm so, what do you recommend for fixing this bit in E06?
<strikethrough> ?
pmichaud see the top of the file re: "update notes" 19:16
but moritz's note about being out of date definitely applies
ewilhelm yeah. argh. So I'll just hope to fix the logic error in the explanation 19:17
19:22 abra left 19:24 pmurias left
pugs_svn r22243 | moritz++ | [t/spec] unfudge tests for rakudo 19:29
moritz_ rakudo: sub x { take 4 }; say gather { x }; 19:36
p6eval rakudo 31131: OUTPUT[4␤]
moritz_ rakudo++ # lexically scooped gather/take 19:37
wolverian scoop { ... }
moritz_ erm, dynamically
19:39 luqui left 19:42 larsen_ joined 19:45 pmurias joined
ewilhelm oh, {} not just in the interpolation, but for all hash addressing? 19:49
perl6: my %a = (foo => 6); say %a{foo}
p6eval rakudo 31131: OUTPUT[Could not find non-existent sub foo␤current instr.: '_block11' pc 41 (EVAL_13:17)␤]
..pugs: OUTPUT[*** No such subroutine: "&foo"␤ at /tmp/tRDvfIWmwb line 1, column 28-31␤]
..elf 22243: OUTPUT[Undefined subroutine &GLOBAL::foo called at (eval 119) line 4.␤ at ./elf_f line 3861␤]
ewilhelm perl6: my %a = (foo => 6); say %a<foo>
p6eval elf 22243, pugs, rakudo 31131: OUTPUT[6␤]
19:50 Jedai left 19:51 Jedai joined
pugs_svn r22244 | pmurias++ | [m0ld] added a --print-bytecode to print bytecode 19:54
19:59 hudnix left 20:25 iblechbot left
pugs_svn r22245 | pmurias++ | [m0ld] --print-bytecode prints out constants too 20:33
20:34 eternaleye left
pugs_svn r22246 | pmurias++ | [m0ld] fixed indentation bug 20:37
20:51 eternaleye joined 21:00 wknight8111 joined
rakudo_svn r31136 | tene++ | [rakudo] 21:09
r31136 | tene++ | * Use a newclosure workaround for gather for moritz++.
r31136 | tene++ | * rt #58854
21:18 wknight8111 left
pugs_svn r22247 | tene++ | [t/spec] 21:26
r22247 | tene++ | S04-statements/gather.t was also testing for proper behavior of nested for loops accessing outer variables.
r22247 | tene++ | It now instead relies on Array.join().
Tene Someone slap me if that was inappropriate. I feel awkward changing the tests to avoid bugs in the implementation. I think this is a bit more reasonable, though. 21:27
moritz_ Tene: it's fine as long as the buggy behaviour is tested elsewhere
pugs_svn r22248 | moritz++ | [t/spec] unfudged gather.t a bit, Tene++ 21:28
Tene perl6: for 0..1 -> $i { for 0..1 -> $j { for 0..1 -> $k { print "$i$j$k " } } } 21:29
p6eval pugs, rakudo 31136: OUTPUT[000 001 010 011 100 101 110 111 ] 21:30
..elf 22246: OUTPUT[$i$j$k $i$j$k $i$j$k $i$j$k $i$j$k $i$j$k $i$j$k $i$j$k ]
Tene That works in rakudo? Huh? Weird. I guess I don't understand the lexicals issue as well as i should.
pugs_svn r22249 | pmurias++ | [smop] 21:34
r22249 | pmurias++ | redeclaration in m0ld are ignore instead of causing a bug
r22249 | pmurias++ | added failing test for modified output of pugs -Cm0ld '$*OUT.print("ok")' - trying to manually make it run see what's changes are needed to smop/pugs
21:35 pmurias left, pmurias_ joined
pmichaud Tene: that works in rakudo because there aren't any closures involved 21:38
i.e., we aren't taking reference to a block and passing it to another function
the lexicals issue in Parrot and Rakudo is a really deep problem, which is why it's been around for months (years?) 21:39
the length of the discussion in #56398 gives some indication of why :-) 21:40
Tene Rather.
Hmm. Is rt.perl.org/rt3/Ticket/Display.html?id=58828 actually valid according to the spec? 21:41
lambdabot Title: #58828: Make calleable &-sigil parameters work in rakudo
21:41 pmurias_ left
Tene perl6: my &foo = sub { say "foo" }; foo(); 21:42
p6eval elf 22249: OUTPUT[/home/evalenv/pugs/misc/elf/elf_f_src/STD_red/match.rb:141:in `block in to_dump0': undefined method `to_dump0' for true:TrueClass (NoMethodError)␤ from /home/evalenv/pugs/misc/elf/elf_f_src/STD_red/match.rb:140:in `each'␤ from
../home/evalenv/pugs/misc/elf/elf_f_src/STD_red/match.r...
..rakudo 31136: OUTPUT[foo␤]
..pugs: OUTPUT[*** Can't modify constant item: VStr "MkCode {isMulti = True, subName = \"&\", subType = SubPrim, subOuterPads = [], subInnerPad = MkPad (padToList []), subPackage = , subAssoc = AIrrelevantToParsing, subParams = [], subBindings = [], subSlurpLimit = [], subReturns = (mkType \"An...
moritz_ pugs: my &foo := sub { say "foo" }; foo();
p6eval pugs: OUTPUT[foo␤]
21:42 jferrero left
moritz_ Tene: I think it should work, yes 21:42
Tene Huh. Okay. That's surprising to me.
pmichaud I think that Rakudo simply doesn't know how to process & in parameter definitions yet
one of the generic "parameters in Rakudo need a big-time refactor" issues
but yes, afaik anytime we have my &foo that should cause foo() to work. 21:43
moritz_ and the other way round
pmichaud or, phrased differently -- there are plenty of examples in S29 where parameters are being declared as &foo :-)
moritz_ if you have 'sub foo', you can also reference &foo 21:44
pmichaud I have to take family to library -- bbl
21:58 larsen_ left 22:00 ceasless joined 22:12 jan_ left 22:14 jan_ joined, PZt joined 22:17 justatheory joined 22:19 alester joined 22:25 ceasless left 22:36 hanekomu joined 22:42 rindolf left
rakudo_svn r31138 | pmichaud++ | [rakudo]: Convert :inline newclosure on gather to :pirop 22:42
23:08 justatheory left
ewilhelm is pugs/util/perl6.vim the most up-to-date vim syntax file? 23:09
23:12 Limbic_Region joined 23:25 jamtech joined 23:26 jamtech left 23:35 hanekomu left 23:40 Jedai left 23:41 Jedai joined 23:45 Chillance left 23:54 BinGOs_ joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!