The topic for #perl6 is: pugscode.org/ planetsix.perl.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 TimToady on 25 November 2008.
00:00 Ehtyar left 00:01 preflex joined 00:05 mst joined
mst TimToady: while I remember, Want.pm already has the "how many values do I want" magic, dunno if that'd be any use for your lazy "return foo(), bar(), baz()" case 00:05
00:05 mst left
rakudo_svn r33581 | pmichaud++ | [rakudo]: $tagsets is already an AST, no need to $(...) it again. 00:10
00:12 alester joined 00:16 bacek left 00:18 preflex left, spx2 left, araujo left, nnunley left, szbalint left, ewilhelm left, allbery_b left 00:19 preflex joined, spx2 joined, araujo joined, nnunley joined, szbalint joined, ewilhelm joined, allbery_b joined 00:26 jferrero left, mberends joined 00:27 mberends left 00:32 PZt left 00:39 rhr_ joined 00:44 rhr left 00:45 mberends joined
meppl good night 01:04
01:07 meppl left, zamolxes left
pugs_svn r24183 | pmichaud++ | [t/spec]: #?rakudo skip a test dealing with Match objects 01:33
01:41 Ehtyar joined 01:46 alester left 01:57 DemoFreak left 02:10 _Chillance_ joined 02:11 Chillance left 02:26 slavik joined
slavik hi all 02:27
jhorwitz: are you around? 02:41
open DB, "|mysql --user=user --password=pass db | tail -n+2|"; 02:46
would that work?
02:47 apeiron left
slavik unable to open file :( 02:48
rakudo_svn r33583 | pmichaud++ | [rakudo]: Cleanup Capture definition slightly. 02:50
02:52 apeiron joined
slavik jhorwitz: are you there? 02:58
[Sat Dec 06 21:58:17 2008] [error] [client 127.0.0.1] need AuthType to note auth failure: /my/test/uri 02:59
[Sat Dec 06 21:58:18 2008] [crit] [client 127.0.0.1] configuration error: couldn't check access. No groups file?: /parrot-test/RequestRec-get_basic_auth_pw
03:11 Psyche^ joined 03:27 Patterner left, Psyche^ is now known as Patterner 03:43 xinming joined
slavik @tell jhorwitz: "load_bytecode" couldn't find file 'Protoobject.pbc' 03:52
lambdabot Consider it noted.
04:02 elmex_ joined 04:13 Limbic_Region left 04:17 elmex left 04:18 elmex_ is now known as elmex 04:34 araujo left 04:38 araujo joined 04:51 thei0s left 05:14 _Chillance_ left 05:30 alester joined 05:33 alester left
slavik @tell jhorwitz I got it to load, but now I get null PMC access when it tries to load :( 06:03
lambdabot Consider it noted.
06:25 slavik left 06:27 justatheory joined 06:29 alester joined 06:34 jfredett left
pugs_svn r24184 | hinrik++ | [util/perl6.vim] fix false Error highlighting of closure delimiters, improve folding a bit 07:12
07:15 ab5tract left 07:26 alanhaggai joined 07:35 bacek joined 07:51 alester left 08:00 iblechbot joined 08:02 smtms left 08:19 mberends left
pugs_svn r24185 | hinrik++ | [util/perl6.vim] highlight Pod :config<options> same as in Perl 6 code 08:29
08:30 ejs joined, meppl joined 08:34 justatheory left 08:53 kanru left 08:58 kanru joined
rakudo_svn r33597 | duff++ | [rakudo] rework of :nth() on .subst() to use ACCEPTS 09:00
09:02 eternaleye_ joined 09:03 eternaleye left, smtms joined 09:10 justatheory joined 09:28 eternaleye_ left 09:46 sri_kraih joined
bacek perl6: token Number { \d+ }; rule Math { <Number> [ '+' | '-' ] <Number> }; '20 + 22' ~~ / <Math> /; say $/ 09:55
p6eval rakudo 33597: OUTPUT[20 + 22ā¤]
..pugs: OUTPUT[Error eval perl5: "if (!$INC{'Pugs/Runtime/Match/HsBridge.pm'}) {ā¤ unshift @INC, '/home/evalenv/pugs/perl5/Pugs-Compiler-Rule/lib';ā¤ unshift @INC, '/home/evalenv/pugs/third-party/Parse-Yapp/lib';ā¤ eval q[require 'Pugs/Runtime/Match/HsBridge.pm'] or die $@;ā¤}ā¤'Pugs::Runti...
..elf 24185: OUTPUT[Can't locate object method "cb__RegexDef" via package "EmitSimpleP5" at ./elf_f line 1612.ā¤]
bacek rakudo: token Number { \d+ }; rule Math { <Number> [ '+' | '-' ] <Number> }; '20 + 22' ~~ / <Math> /; say $/<Number> 09:56
p6eval rakudo 33597: OUTPUT[Use of uninitialized valueā¤ā¤]
bacek rakudo: token Number { \d+ }; rule Math { <Number> [ '+' | '-' ] <Number> }; '20 + 22' ~~ / <Math> /; say $/<Math>
p6eval rakudo 33597: OUTPUT[20 + 22ā¤]
bacek rakudo: token Number { \d+ }; rule Math { <Number> [ '+' | '-' ] <Number> }; '20 + 22' ~~ / <Math> /; say $/<Math><+>
p6eval rakudo 33597: OUTPUT[Use of uninitialized valueā¤ā¤]
bacek any ideas how to get '+' or '-' from Match? 09:57
rakudo: token Number { \d+ }; rule Math { <Number> [ '+' | '-' ] <Number> }; '20 + 22' ~~ / <Math> /; say $/<Math><Number>
p6eval rakudo 33597: OUTPUT[2022ā¤]
bacek rakudo: token Number { \d+ }; rule Math { <Number> $<op>=[ '+' | '-' ] <Number> }; '20 + 22' ~~ / <Math> /; say $/<Math><op> 09:58
p6eval rakudo 33597: OUTPUT[+ ā¤]
bacek hmm...
rakudo: token Number { \d+ }; rule Math { <Number> $<op>=[ '+' | '-' ] <Number> }; '20 + 22' ~~ / <Math> /; say $/<Math>.exists('op') 09:59
p6eval rakudo 33597: OUTPUT[Method 'exists' not found for invocant of class 'PGE;Match'ā¤current instr.: '_block11' pc 82 (EVAL_16:35)ā¤]
10:00 DemoFreak joined
bacek rakudo: token Number { \d+ }; rule Math { <Number> $<op>=[ '+' | '-' ]? <Number> }; '20 + 22' ~~ / <Math> /; say defined $/<Math><op> 10:01
p6eval rakudo 33597: OUTPUT[1ā¤]
bacek rakudo: token Number { \d+ }; rule Math { <Number> $<op>=[ '+' | '-' ]? <Number> }; '20 22' ~~ / <Math> /; say defined $/<Math><op>
p6eval rakudo 33597: OUTPUT["load_bytecode" couldn't find file 'PGE.pbc'ā¤current instr.: 'parrot;PCT;Grammar;onload' pc 0 (src/PCT/Grammar.pir:41)ā¤]
bacek rakudo: token Number { \d+ }; rule Math { <Number> $<op>=[ '+' | '-' ]? <Number> }; '20 22' ~~ / <Math> /; say defined $/<Math><op>
p6eval rakudo 33597: OUTPUT[1ā¤]
10:06 mberends joined
bacek rakudo: token Number { \d+ }; rule Math { <Number> $<op>=[ '+' | '-' ]? <Number> }; '20 22' ~~ / <Math> /; say ?$/<Math><op> 10:07
p6eval rakudo 33597: OUTPUT[1ā¤]
bacek rakudo: token Number { \d+ }; rule Math { <Number> $<op>=[ '+' | '-' ]? <Number> }; '20 22' ~~ / <Math> /; if $/<Math><op> { say "foo" } else { say "bar" } 10:09
p6eval rakudo 33597: OUTPUT[fooā¤]
bacek rakudo: token Number { \d+ }; rule Math { <Number> [$<op>=[ '+' | '-' ]]? <Number> }; '20 22' ~~ / <Math> /; if $/<Math><op> { say "foo" } else { say "bar" } 10:10
p6eval rakudo 33597: OUTPUT[barā¤]
bacek rakudo: token Number { \d+ }; rule Math { <Number> [$<op>=[ '+' | '-' ]]? <Number> }; '20 + 22' ~~ / <Math> /; if $/<Math><op> { say "foo" } else { say "bar" }
p6eval rakudo 33597: OUTPUT[fooā¤]
bacek ho!
10:16 iblechbot left 10:28 adc_penner joined 10:35 pmurias joined 10:38 spx2 left, xinming left 10:39 spx2 joined 10:40 xinming joined
bacek rakudo: class A { has $.a }; class B is A {}; my $b = B.new(a=>"foo"); 10:42
p6eval rakudo 33597: OUTPUT[You passed an initialization parameter that does not have a matching attribute.ā¤current instr.: 'die' pc 12669 (src/builtins/control.pir:168)ā¤]
10:43 apeiron left
bacek looks like a bug... 10:46
10:48 kanru left
moritz_ bacek: in general the subclass can override the parameter $.a, so there needs to be a way to initialize them in the parent class 10:48
pugs: class A { has $.a }; class B is A {}; my $b = B.new(a=>"foo");
p6eval pugs: RESULT[\B.new((\("a", "foo"),)]
moritz_ pugs: class A { has $.a }; class B is A { has $.a }; my $b = B.new(a=>"foo");
p6eval pugs: RESULT[\B.new((\("a", "foo"),)]
bacek moritz_: bug, indeed 10:49
10:50 kanru joined
moritz_ bacek: then open a ticket, and write a test case ;) 10:51
bacek moritz_: my ticket, your testcase :)
moritz_ bacek: what have I got to do with it? ;-) 10:52
bacek moritz_: commit something like 'class A { has $.a }; class B is A {}; lives_ok(B.new(a=>"foo"));' in proper file :) 10:53
Maybe "eval_lives_ok" 10:54
moritz_ lives_ok is fine... 10:56
bacek moritz_: and add comment to #61120 :) 10:59
moritz_ gives in
bacek moritz_: why? 11:00
moritz_ I try to train people to commit test cases on their own, but it's not easy :/ 11:01
pugs_svn r24186 | moritz++ | [t/spec] tests for inheritance of attributes, bacek++ 11:02
r24187 | moritz++ | [t/spec] svn props for new test file
11:03 justatheory left
bacek moritz_: I've implemented 300+ tests for my XQuery implementation today... 11:04
pmurias bacek: you wrote code to which passes them or wrote them? 11:08
11:08 ejs left
bacek pmurias: both. 11:08
pmurias: actually 200 tests from XQTS. 11:09
but I wrote code to pass them :)
pugs_svn r24188 | hinrik++ | [util/perl6.vim] improve syncing, and work around what looks like a vim bug relating to nested transparent regions 11:20
11:21 ejs joined
moritz_ the new vim syntax hilighting is very nice, but one thing that I find weird is that quoted strings have a different color than their quotes 11:24
pugs_svn r24189 | moritz++ | [t/spec] general clean up of private_methods.t, 11:26
r24189 | moritz++ | add regression tests for [perl #61102]
r24190 | moritz++ | [t/spec] rename private_methods.t to private.t
11:35 riffraff joined 11:37 sri_kraih_ joined
pugs_svn r24191 | moritz++ | [t] clean up oo/attributes/instance.t and move it to spec 11:39
rakudo_svn r33598 | moritz++ | [rakudo] add regression tests for RT #61102 to t/spectest.data, 11:40
r33598 | moritz++ | and restore alphabetical order
11:40 alanhaggai left 11:41 sri_kraih left, adc_penner left
pugs_svn r24192 | hinrik++ | [util/perl6.vim] more syncing stuff 11:43
11:45 schmalbe joined
pugs_svn r24193 | moritz++ | [t/spec] more cleanup of instance.t 11:48
11:55 vixey joined, dr_df0 joined
pugs_svn r24194 | moritz++ | [t/spec] fudge instance.t for rakudo 11:59
12:01 slavik joined
slavik hello 12:01
moritz_ hi 12:02
slavik so, looks like I got it load everything, but I get null PMC access on load :(
moritz_: here's another idea ... use perl5 as a wrapper or something ... 12:03
with mod re-write
moritz_ or work around the error
12:03 drbean left
moritz_ (but that might be time consuming) 12:03
slavik like if you ask for index.p6, it actually fetches index.pl?page=index.p6
or some such
moritz_: thing is, I do not understand what the null PMC access means :( 12:04
moritz_ slavik: it means you've found an error in rakudo
slavik moritz_: I am getting my E90 on tuesday, here's an idea for a port ... Perl6 on Symbian on ARM :D
moritz_: could it be an error in mod_parrot?
moritz_ slavik: dunno, I've never used it 12:05
slavik k
12:06 zamolxes joined
slavik moritz_: I was thinking about SQL injection the other day ... 12:06
and a thought occured, what if there was a specific function for to the DBI that is not a generic execute_SQL() type thing 12:07
like a specific insert(), select() and such
moritz_ slavik: you mean like SQL::Abstract? 12:08
slavik guess so
I mean, it won't do like "select * from blah, (select ...) where such and such
but for simple select and insert/update queries, I think it could work
moritz_ aye 12:09
slavik so that when it tells the RDBMS, it won't be a single SQL statement that has to be escaped and such, but an actual API call
12:09 rff joined
slavik like: insert($table, @keys, @values) or better yet, combine them into a hash 12:10
moritz_ well, with placeholders and prepared statements there's no need for escaping anyway
rakudo_svn r33599 | moritz++ | [rakudo] add instance attribute tests to t/spectest.data
r33599 | moritz++ | (most of them are skipped, but a few actually pass)
pmurias ruoso: hi 12:13
12:13 masak joined
pugs_svn r24195 | moritz++ | [t/spec] regression tests for RT #61100 12:17
12:20 drbean joined
ruoso hi pmurias 12:20
pmurias ruoso: what are the current smopish small TODO tasks? 12:22
ruoso pmurias, well... we're still moving forward to compile ClassHOW.pm 12:23
we need the "for" operator now
that's why I insisted too much on S07 12:24
since "for" and "map" are synonims
12:24 riffraff left
ruoso so it seems we need to compile Map.pm in order to be able to compile ClassHOW.pm 12:25
slavik so: @list = for 1..100 { 1 }; is same as @list = map { 1 } 1..100; ???
ruoso slavik, yes
including the fact that it is lazily evaluated 12:26
slavik so, we don't need map at all then, or is there still some difference between map and for?
ruoso the syntax
pmurias it would be nice to have more small things on the ROADMAP, as i find some of the milestones daunting
slavik also, is it possible to make a function as an iterator? (list generator? list comprehension? not sure what it is called)
ruoso pmurias, alright... maybe we could do that in terms of test cases that we could then move to mildew/t as we implement... 12:27
slavik, I'm not sure I get what you mean
slavik ruoso: like Python's xrange() 12:28
pmurias ruoso: that seems nice
slavik similar to that
ruoso going to find out what xrange is
12:28 thei0s joined
ruoso slavik, how is it different from Perl 6 range? 12:29
slavik ruoso: I mean, can we define our own list generators to be iterated through?
with = operator
ruoso slavik, that's what S07 is about
slavik oh, ok
ruoso slavik, but = is "mostly eager" 12:30
slavik wait ... I know it talks about lists ... like $iter <== @list;
but instead of @list, I want to have a function and then I don't want to have to assign it, so that I can do: for fib(1,10) { say $_; } where fib would lazily generate nth term of the fibonacci sequence 12:31
ruoso slavik, the way you wrote simply DWYM
slavik DWYM?
ruoso Do What You Mean 12:32
and is "mostly lazy"
slavik, you probably want to look at the gather/take operator
slavik ruoso: Python has a 'yield' operator ... that is sort of like a return but isn't 12:33
will do
ruoso: which S is it in?
ruoso S03
slavik ty 12:34
ruoso pmurias, I'll create a ROADMAP directory inside mildew
ruoso bbiab&
slavik anyone have an idea how difficult it would be to port parrot to symbian? 12:35
symbian on ARM
pmurias slavik: is symbian POSIX?
slavik pmurias: there is a POSIX library or some such 12:36
PIPS
pugs_svn r24196 | hinrik++ | [util/perl6.vim] highlight sub/method/regex/etc names
pmurias & 12:37
12:37 pmurias left
ruoso slavik, that shouldn't be so hard 12:39
the bigger problem will be to have the binding to the phone interface
since p5 runs on symbian 12:40
but has no binding to the phone interface as well
slavik ruoso: I am more interested in getting a shell onto symbian and just running regular perl6 code
although making perl6 make phone calls with a txt-to-speech synth would be fun :)
ruoso slavik, well, you should probably see how p5 runs on it 12:42
it should be a good start
12:42 smg left
slavik ruoso: I haven't found a shell/terminal for it yet :( 12:43
ruoso I'm not sure symbian has that
but I really never saw symbian close enough
slavik hmm ... 12:58
pugs_svn r24197 | ruoso++ | [mildew] The mildew ROADMAP is a directory with test cases that do not pass yet; 13:01
r24198 | hinrik++ | [examples/euler] some Pod fixes 13:02
rakudo_svn r33601 | bernhard++ | [codingstd] remove a trailing space 13:10
thei0s write your own terminal in perl5 :P 13:12
ruoso you don't even need to write it 13:13
there's one in CPAN already
(probably more than one, actually)
rakudo: sub foo($code) { $code() }; sub bar { foo({ return 1 }); return 2 }; say bar;
p6eval rakudo 33599: OUTPUT[2ā¤]
ruoso I think that's wrong, isn't it 13:14
?
pugs: sub foo($code) { $code() }; sub bar { foo({ return 1 }); return 2 }; say bar;
p6eval pugs: OUTPUT[2ā¤]
moritz_ ruoso: aye, return() should break out of the sub 13:15
ruoso ok...
moritz_ ruoso: care to submit a bug report? 13:16
ruoso I think I'll implement return usign $?ROUTINE then
thei0s doesn't return break out of the anonymous sub?
ruoso thei0s, it's a bare block, it's not a sub
thei0s ok, then out of foo
moritz_ leave() breaks out of blocks, return() out of subs|methods 13:17
masak moritz_: no, it's not wrong.
foo does return 1
but bar returns 2
nothing wrong there.
ruoso er...
thei0s and then you have in bar: 1; return 2;
ruoso I think the routine is determined lexically
13:17 schmalbe left
ruoso not dinamically 13:17
masak ruoso: I did not grok that. 13:18
but I do think there is nothing wrong with the above code.
ruoso map { return 1 }, 1,2,3; should return from the routine that called map
not return "1" from the map
masak aye
that's right
but that 'raturn' is not within another sub
13:19 justatheory joined
ruoso I think $?ROUTINE is used to determine where to return from 13:19
masak rakudo: sub x { my $code = { return 1 }; $code(); return 2 }; x
p6eval rakudo 33601: RESULT[1]
masak ruoso: that's how you return from the same sub.
ruoso: the first code example had two subs.
ruoso masak, how that applies to map
?
masak ruoso: the map also has a routine block, not a sub 13:20
moritz_ I don't think that map() catches 'return' control exceptions
ruoso masak, er... a sub is a routine
masak moritz_: exactly. but subs do.
ruoso a bare block isn't
masak ruoso: yes, and that's the difference here.
sorry, I don't know the exact terminology.
ruoso masak, the map example is the key to understand 13:21
masak understand what?
ruoso rakudo: sub foo { map { return 1 }, 1; return 2; } say foo
p6eval rakudo 33601: OUTPUT[Statement not terminated properly at line 1, near "say foo"ā¤ā¤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:83)ā¤]
ruoso rakudo: sub foo { map { return 1 }, 1; return 2; }; say foo
p6eval rakudo 33601: OUTPUT[1ā¤]
masak ruoso: yes. that's correct too.
IMHO. 13:22
ruoso that is correct... but
rakudo: sub bar($code) { $code() }; sub foo { bar { return 1 }, 1; return 2; }; say foo
p6eval rakudo 33601: OUTPUT[too many arguments passed (2) - 1 params expectedā¤current instr.: 'bar' pc 78 (EVAL_14:39)ā¤]
ruoso rakudo: sub bar($code) { $code() }; sub foo { bar { return 1 }; return 2; }; say foo
p6eval rakudo 33601: OUTPUT[2ā¤]
ruoso but that's wrong
map is not a special case
masak man, take a step back and look at your code!
there's a second sub in your last example!
that sub catches the return exception. 13:23
that's it.
ruoso map is a sub
masak huh?
ruoso just as bar
masak that's news to me.
ruoso there are no core operators in Perl 6
everything is dispatched the same way
masak even if it's a sub, I don't expect it to catch returns.
thei0s what is map then? just a block?
ruoso well it just mean that the both examples should behave the same
masak ruoso: I see your point now. 13:24
but I think that map and grep deserve special treatment somehow.
that's what I expect as a Perl programmer.
ruoso not in Perl
Perl 6*
masak yes, I expect it there too, to be honest. 13:25
ruoso in Perl 6 there are no "special treatment"
everything is consistent
masak if you say so.
ruoso TimToady says so ;)
masak but that goes against my expectations in this case.
ruoso at every talk he presents
thei0s pugs: sub foo { map { return 1 }, 1; return 2; }; say foo
p6eval pugs: OUTPUT[1ā¤]
thei0s if you ask me, this with map seems weird, I would expect such behavour that current implementations do with bar 13:26
masak ruoso: what if map were to explicitly re-throw return exceptions?
ruoso in S06: "The return function notionally throws a control exception that is caught by the current lexically enclosing Routine" 13:27
masak can it do that?
slavik is the following legal in Perl6? open $FILE, "| command1 | command2 |"; say $FILE "input"; my $output = =$FILE; say $output;
ruoso *lexically*
thei0s what is map, grep are macros?
ruoso map and grep are subs
masak ruoso: see my quesion above about re-throwing.
slavik @seen jhorwitz 13:28
lambdabot jhorwitz is in #perl6. I last heard jhorwitz speak 15h 13m 1s ago.
ruoso masak, map doesn't even notice it
masak slavik: I doubt that works today in Rakudo.
ruoso: ok.
ruoso because the return is *lexically* determined
masak ruoso: I don't grok the term 'lexically' in this context.
I know what it means in general, but... 13:29
slavik masak: :( ... I was thinking of making a ghetto DBI::MySQL by using mysql command line
masak slavik: use run() or something.
pugs_svn r24199 | ruoso++ | [mildew] two more tests in the ROADMAP
ruoso masak, it means it resolves the same way a lexical variable would
slavik masak: any idea if classes are "finished", last time I tried, I had problems with constructor and such
masak ruoso: I'm having difficulty applying that to return statements. 13:30
ruoso masak, think in return as something called in $?ROUTINE
masak slavik: there are some things left to do. but much is in place.
ruoso where $?ROUTINE is a lexical variable
slavik k
ruoso that is implicit to every sub or method keyword
masak ruoso: aye.
slavik what's the site for the current spec? 13:31
moritz_ svn.pugscode.org
erm, syn.pugscode.org
ruoso moritz_, how do I report that bug?
rakudo.org doesn't mention it 13:32
masak perlcabal.org/syn/
moritz_ ruoso: [email@hidden.address]
ruoso ok
masak I think not being able to return from within a map block is a bit of a loss.
also, how can for and map be equivalent if one cannot return from within a map block? 13:33
moritz_ is fairly sure that map shouldn't cat 'return' control exceptions 13:34
thei0s rakudo: sub foo { map { say "map $?ROUTINE"; return 1 }, 1; say "foo $?ROUTINE"; return 2; }; say foo
rakudo: sub bar($code) { say "bar $?ROUTINE"; $code() }; sub foo { bar { say "block $?ROUTINE"; say "foo $?ROUTINE"; return 1 }, 1; return 2; }; say foo
p6eval rakudo 33601: OUTPUT[Lexical '$?ROUTINE' not foundā¤current instr.: '_block23' pc 270 (EVAL_13:106)ā¤]
rakudo 33601: OUTPUT[too many arguments passed (2) - 1 params expectedā¤current instr.: 'bar' pc 78 (EVAL_14:39)ā¤]
moritz_ but I said that already, no?
ruoso masak, you probably want to "leave" instead of "return" in that case
masak er.
no, I want to return. from the surrounding sub or method.
ruoso just posted the bug to rakudo 13:35
masak ruoso++ 13:36
ruoso later &
13:37 ejs left 13:39 ejs joined 13:48 rindolf joined
dr_df0 hi, I'm looking the place in rakudo source tree, where to implement simple slices (eg. @a[0..2], @a[0,2,5] etc. 14:03
any hints?
moritz_ dr_df0: pmichaud is working on that one, I think 14:04
dr_df0: maybe you'll have some luck searching for `postcircumfix'
dr_df0 i'll look at that, and talk to pmichaud 14:05
tt
ty
14:12 smg joined 14:13 mberends left 14:18 rff left
thei0s rakudo: say Task.WHAT; 14:33
p6eval rakudo 33603: OUTPUT[Moduleā¤]
thei0s what is this?
jnthn rakudo: say ResizablePMCArray.WHAT 14:34
p6eval rakudo 33603: OUTPUT[Moduleā¤]
jnthn Probably just namespaces.
Leaking in from Parrot. 14:35
slavik hi, jnthn
thei0s aha, it seems I'll then have to rename my class
jnthn hi slavik 14:36
thei0s: Yeah. Once we do HLL map stuff in Parrot, that problem will go away. 14:37
masak rakudo: class Task {} 14:39
p6eval rakudo 33603: OUTPUT[Class Task already registered!ā¤ā¤current instr.: '!keyword_class' pc 13963 (src/builtins/guts.pir:352)ā¤]
masak thei0s: thanks for that one :)
masak resports
s/s//
jnthn masak: Attach HLL map as a dependency to it...
dr_df0 @tell pmichaud I've heard (from moritz_) that you are working on slices in rakudo. can i help somehow?
lambdabot Consider it noted.
masak jnthn: what does that mean? what's a HLL map? 14:40
slavik high level language? 14:41
masak slavik: yes, thank you. I know that.
I don't know what 'Attach HLL map' means. it's probably another ticket, but which one? 14:42
slavik does anyone have a working class example?
14:42 pmurias joined
jnthn masak: It's the thing that'll put Rakudo (and other languages) into a language namespace, separate from all of Parrot's guts. 14:42
slavik masak: then I am as confused as you :P
masak slavik: November source code?
lots of working examples. 14:43
slavik masak: huh?
oh
where do I get it?
jnthn masak: Don't worry about it if you can't find a ticket on HLL Map, i just thought we had one.
masak hold on.
slavik: github.com/viklund/november/
jnthn: we might. I'll look.
slavik masak: ty 14:44
masak jnthn: rt.perl.org/rt3/Search/Simple.html?q=hll
any of those?
jnthn 49812 14:45
masak dz
masak attached the dep 14:48
14:53 apeiron joined
jnthn -> concert (Opeth! :-)) 14:54
wolverian ooh, opeth. 14:55
have fun. :)
14:57 adc_penner joined
jnthn Thanks, it should be good! :-) 14:58
jnthn really afk now
14:58 ejs1 joined
masak rakudo: my $a = "foo" x 5; $a .= subst(/foo/, "bar", :nth(2), :x(2)); say $a 15:01
p6eval rakudo 33603: OUTPUT[foobarfoofoofooā¤]
masak shouldn't that replace two 'foo's?
15:01 ejs left
dr_df0 pugs: my $a = "foo" x 5; $a .= subst(/foo/, "bar", :nth(2), :x(2)); say $a 15:04
p6eval pugs: OUTPUT[Error eval perl5: "if (!$INC{'Pugs/Runtime/Match/HsBridge.pm'}) {ā¤ unshift @INC, '/home/evalenv/pugs/perl5/Pugs-Compiler-Rule/lib';ā¤ unshift @INC, '/home/evalenv/pugs/third-party/Parse-Yapp/lib';ā¤ eval q[require 'Pugs/Runtime/Match/HsBridge.pm'] or die $@;ā¤}ā¤'Pugs::Runti...
ruoso jnthn, have you seen the rakudobug I've just posted? 15:05
pmurias ruoso: do you have a plan how to solve the smop leaks? 15:06
masak ruoso: jnthn just left.
ruoso masak, ah ok...
pmurias, well... I'll take some time and do the old fashion hunting
pmurias, have you seen the tests I added on mildew/ROADMAP? 15:07
pmurias yes
15:19 apeiron left, apeiron joined
spx2 I'd like to ask about perl6's OO system 15:19
how are copy-constructors implemented ?
pmurias spx2: do we have copy-constructors in perl6? 15:22
spx2 I don't know
what I know is some stuff from perl5
like
sub new
{
my ($class, %args) = @_;
or there is a Clone module on CPAN
that's all I know from perl5
how is this stuff translated to perl6 ?
I'm talking about this module search.cpan.org/~rdf/Clone-0.29/Clone.pm 15:23
pmurias perl6 objects have a .clone method
spx2 oh nice :)
15:24 spx2_ joined, spx2_ left 15:25 spx2 left, spx2 joined
pmurias i'm pretty sure we don't have the C++ style copy constructors 15:25
slavik pmurias: they aren't need though, are they? 15:26
pmurias we conceptually always pass a pointer to an object 15:27
spx2 that's great
I could just write a dirty method 15:29
that would check ref(parameter)
to see if it's the class it expects
15:29 rindolf left
slavik pmurias: what about references to other objects? 15:29
spx2 and then takes the members one by one and copies them
slavik is deep copy done? or no?
pmurias forgot about the container stuff
spx2 but that would be the most lame thing possible
*lamest 15:30
rakudo_svn r33605 | pmichaud++ | [rakudo]: spectest-progress.csv update: 231 files, 4717 passing, 0 failing
pmurias slavik: you mean references to variables
?
slavik pmurias: I guess so ...
C++ example: classa A { classb *B }; 15:31
pmurias classa being?
slavik just classa
and you have an object of classa
pmurias the example you mentioned isn't valid C++ 15:32
slavik when the pointer gets copied, does it make a copy of whatever it points to? or no or what happens?
pmurias slavik: the pointer just gets copied
it would be class A { has $!B } in perl6 15:33
slavik pmurias: so, whatever $!B points to doesn't get copied? 15:34
pmurias when you do a .clone or just $a = $b?
slavik clone 15:36
actually, both
pmurias i don't think .clone is deep
slavik please don't tell me that $a = $b is same as java (ref assignment)
pmurias $a = $b just copies the content of $b into $a
slavik err, what would be the diff between $a=$b and $a=$b.clone ? 15:37
pmurias $b.clone creates a new object and stores it in $a, $a = $b stores the object which is stored in $b in $a 15:38
slavik pmurias: neither refers to the same exact object, right? 15:39
pmurias $a = $b does it works the same as in java 15:40
(as far as i understand how java works)
slavik dude, that's crap! 15:41
pmurias what's wrong with that approach
slavik rakudo: my $a; my $b = 5; $a = $b; say $a; $b = 6; say $a;
p6eval rakudo 33604: OUTPUT[5ā¤5ā¤]
slavik oh, nvm ... good
rakudo: my @a; my @b = 1,5; @a = @b; say @a; @b = 2,6; say @a; 15:42
p6eval rakudo 33604: OUTPUT[15ā¤15ā¤]
15:42 alester joined
avar rakudo: my @a; my @b = 1,5; @a = @b; say @a; @b = 2,6; say @b; 15:42
p6eval rakudo 33604: OUTPUT[15ā¤26ā¤]
slavik pmurias: the latter in java would output 15\n26 15:43
because array=array does reference assignment, not value
pmurias arrays are a special case
slavik java arrays 15:44
:P
in Perl, we have lists :P
btw, are lists doubly linked lists as in Perl5?
doubly linked circular* list
spx2 will the Clone module copy references or copy the objects inside the refferences as well ?
what is the difference between shallow , deep copy ?
is this the difference ?
pmurias yes
slavik: lists in Perl5 are doubly linked? 15:45
slavik pmurias: I read somewhere that they were double linked circular lists
pmurias doubts it
slavik which is neat since you get negative index support without any extra effort
pmichaud rakudo: say 'hello'; 15:46
lambdabot pmichaud: You have 1 new message. '/msg lambdabot @messages' to read it.
p6eval rakudo 33610: OUTPUT[helloā¤]
pmichaud rakudo: say [+] 1..7; 15:47
p6eval rakudo 33610: OUTPUT[Statement not terminated properly at line 1, near "[+] 1..7;"ā¤ā¤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:83)ā¤]
pmurias slavik: circular linked lists have much worse performance for most stuff
slavik versus regular lists?
pmichaud argh. Have to wait for r33611 for [+]
pmurias slavik: versus arrays
slavik pmurias: in Perl5, arrays are not exactly arrays, are they? since they are resizable 15:48
pmurias the are STL deque
's
rakudo_svn r33611 | pmichaud++ | [rakudo]: Fix assignment metaoperators, add reduction operators (e.g., [+]). 15:50
pmurias slavik: you can resize arrays if you have spare space and copy over the elements when it runs out
slavik pmurias: with realloc(), isn't that expensive? 15:52
pmurias you don't do it very often
i'm not familiar with realloc internals 15:53
slavik I mean realloc() from C 15:54
pmurias that's what i meant too ;)
slavik oh
realloc() tries to extend the space or gets it from somewhere else ...
not sure if it copies the content 15:55
pmurias only when mallocing the new block 15:56
15:56 dr_df0 left
slavik so yea ... expensive I think 15:56
pmichaud rakudo: say [+] 1..7; 16:09
p6eval rakudo 33610: OUTPUT[Statement not terminated properly at line 1, near "[+] 1..7;"ā¤ā¤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:83)ā¤]
slavik [] can contain any operation/sub? 16:10
hmm
jhorwitz slavik: ping 16:13
lambdabot jhorwitz: You have 2 new messages. '/msg lambdabot @messages' to read them.
slavik hey 16:14
messages :)
jhorwitz ya
did you add PARROT_RUNTIME to your envvars script?
pmichaud slavik: any infix operator, yes.
slavik no
jhorwitz: no
pmichaud: no chance for subs?
jhorwitz: the first issue was fixed (not finding stuff)
jhorwitz: the second issue is the problem now 16:15
pmichaud slavik: there's always .reduce
jhorwitz can you paste the error log?
slavik k
jhorwitz: not atm ... at work :(
jhorwitz ah. well, send it along when you can and i'll take a look. i'm sure it something simple. 16:16
did all the tests pass?
16:16 iblechbot joined
slavik jhorwitz: except for something creating a socket or something 16:17
jhorwitz that's odd 16:18
slavik yeah, anyway, it did not seem like a showstopper, so I continued
jhorwitz *any* test failures worry me 16:19
slavik oh
it was something about binding to a socket
jhorwitz that would probably be Apache::Test doing its thing
slavik mebbe
jhorwitz anyway, ping me when you have a chance to grab the logs. :) 16:20
slavik k
waiting for system to boot, will ssh in and see 16:21
rakudo_svn r33617 | pmichaud++ | [rakudo]: Add [,], [&], [|], and [^] reduction operators. 16:30
slavik pmichaud: so what is the point of [,] 1..100 ??? 16:32
doesn't that just make a list?
16:32 smg left
masak slavik: it does. so you understand the semantics of the operator. what's your complaint? 16:47
that it doesn't do enough?
slavik masak: seems useless
what's the difference between ([,] 1..100) and (1..100)
masak slavik: the first is a list, the second is a range.
slavik hmm 16:48
masak slavik: remember that () is not a list constructor in Perl 6
, is
slavik masak: parenthesis are for grouping ;)
masak slavik: aye. not for makin lists.
s/makin/making/
slavik masak: yes, but in all uses for ranges I've seen is to make a list ... 16:49
masak what? no.
if $a ~~ 1..100 { ... }
slavik err, what does that match? 16:50
if $a is in that range?
masak aye.
(1..100).pick 16:51
slavik hmm, ~~ is really neat
masak neither of these two examples "makes a list"
slavik what's .pick? rand in that range?
masak aye
slavik # Failed test 'joeuser:password granted access'
# at t/legacy/03AuthenHandler.t line 18.
# got: '<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 2.0//EN">
masak: ok, you need to tell us such cool things 16:52
masak slavik: read Moritz' blog posts about it.
and the spec.
slavik masak: spec is big ...
masak and the November source code. :)
slavik: suit yourself. there are cool things in there.
slavik masak: there needs to be "cool and neat stuff"
I am trying to accomplish this with Dive into Perl6 ... 16:53
heh
masak and a pony.
slavik pwnies rule
btw, is it possible given an object to get all the methods it has?
like .perl but for methods
masak $obj.HOW.meth
(I think)
slavik: if you're writing "Dive into Perl 6", you really should read through the specs at least once... 16:54
slavik masak: I will
masak they tell you this type of things.
slavik masak: basically, each chapter will correspond to a synopsis
and grab the most essential stuff 16:55
masak sounds like the Exigeses.
slavik guess so, but with real code examples that compile and work ;)
not just theoretical code that will someday work, but something that work now
masak aye.
slavik and not just "you can say this or that" but "here's what the entire chapter looks like" 16:56
masak: like the grammar article that I wrote, that you have trouble loading
masak slavik: the Exigeses also give full examples.
slavik oh ...
epic fail
16:57 adc_penner left
masak slavik: yes, I got the Google cache, but it was from 28 November. 16:57
slavik it's fairly up to date ;)
masak slavik: don't worry, the Exigeses are old, and as you say, they don't contain runnable code.
slavik so, it's not like exigeses
these are exigeses++ ^^
masak something like that. 16:58
except that one can't reach your page. :P 16:59
slavik not my fault
masak I'm going to tease you about it until you do something real about it.
pasteling "slavik" at 12.130.86.9 pasted "make test = FAIL" (43 lines, 2.1K) at sial.org/pbot/33531
slavik masak: until I get my own mod_perl6 environment ... no :P
jhorwitz: see the paste :)
masak slavik: you could just move the page to a server that is reachable. 17:00
slavik masak: I don't got one
masak slavik: I'd like to read your posts, too.
slavik: there are free hosting solutions on the web.
slavik masak: not with mod_perl6
and not with root access :P
masak slavik: no, but people can reach them. that's worth something, too. 17:01
slavik masak: I set up that server with mail and everything :P
masak slavik: too bad it's not reachable.
slavik @tell jhorwitz sial.org/pbot/33531
lambdabot Consider it noted.
slavik masak: try: 67.100.227.140/
do you get a hello?
masak aye
jhorwitz slavik: let's not worry about that failure for now. 17:02
lambdabot jhorwitz: You have 1 new message. '/msg lambdabot @messages' to read it.
slavik k
jhorwitz: I didn't :P
jhorwitz: want the paste from my parrot config and the error?
jhorwitz yes please
pasteling "slavik" at 12.130.86.9 pasted "mod_parrot FAIL" (22 lines, 1.5K) at sial.org/pbot/33532 17:05
slavik jhorwitz: there :)
33317 revision of rakudo
0.8.1-devel version of parrot 17:06
jhorwitz is it a recent svn checkout of parrot? 17:07
17:08 jan_ left
jhorwitz slavik: can i see your envvars script? (from apache's bin dir) 17:10
17:11 masak left 17:19 apeiron left 17:22 apeiron joined
slavik jhorwitz: sure, 1 sec 17:29
jhorwitz: just default stuff 17:30
jhorwitz: I didn't add anything to it
jhorwitz ok, the eg/perl6/README should tell you what to add 17:31
slavik k
jhorwitz you'll need PARROT_RUNTIME and PERL6LIB
17:31 ThierryM joined
slavik ty 17:34
17:42 dr_df0 joined 17:50 jan_ joined 17:51 cknapp joined
zamolxes www.computerworld.com.au/article/26...?eid=-6787 .. 18:02
18:02 smg joined
slavik jhorwitz: question 18:08
the runtime dir: is ti the one with include/ and library/ subdirs?
pmichaud I only added [,] because it's in the spec. :-) 18:12
(and it was relatively easy to do)
rakudo: say [+] 1..10; 18:13
p6eval rakudo 33619: OUTPUT[55ā¤]
pmichaud yay
rakudo: say [*] 1..10;
p6eval rakudo 33619: OUTPUT[3628800ā¤]
slavik nice
rakudo_svn: say [*] 1..100
rakudo: say [*] 1..100
p6eval rakudo 33619: OUTPUT[9.33262154439441e+157ā¤]
slavik rakudo: I hate you!!!!!! 18:14
p6eval rakudo 33619: OUTPUT[Statement not terminated properly at line 1, near "hate you!!"ā¤ā¤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:83)ā¤]
slavik GAAAAAAAAAAAAAHHHHHHHH!!!!!!!!!!
18:18 vixey left, vixey joined 18:20 vixey left 18:37 awwaiid left, dr_df0 left
pugs_svn r24200 | moritz++ | [t/spec] some unfudges for rakudo (pmichaud++) 19:07
19:21 eternaleye joined 19:25 jferrero joined 19:29 Limbic_Region joined
jhorwitz slavic: PARROT_RUNTIME is the directory containing "runtime". should be your parrot source dir 19:39
kinda confusing 19:40
19:41 ejs2 joined
slavik jhorwitz: even though I installed parrot? 19:42
jhorwitz you probably shouldn't install parrot 19:43
slavik why not?
/home/slavik/src/parrot/runtime/parrot/
does that seem like it will work?
jhorwitz /home/slavik/src/parrot
slavik Can't find ./mod_perl6 in @INC 19:44
:D
jhorwitz good
now the README should say you need PERL6LIB
slavik yes
=/usr/local/runtime/parrot/library 19:45
jhorwitz no -- /path/to/mod_parrot/languages/perl6/lib
slavik got it
ty
jhorwitz np -- it's confusing when there's no docs!
slavik aww it failed 19:46
jhorwitz ?
slavik Null PMC access in find_method()
current instr.: 'infix:=' pc 11214 (src/gen_builtins.pir:6893) called from Sub 'parrot;ModParrot;HLL;perl6;post_config_handler' pc 5729 (EVAL_16:1820)
that's in the log 19:47
19:48 dr_df0 joined
jhorwitz hm 19:48
pugs_svn r24201 | moritz++ | [t/spec] more tests in indirect_notation.t, for example for RT #61106 19:51
19:51 ejs1 left
moritz_ should I retitle #61106 to something like 'using closure as a method doesn't pass invocant as first positional arg'? 19:52
slavik moritz_: ETF?! 19:53
jhorwitz slavik: got it -- mod_parrot bug
bombs out if you have no mod_perl6 apache directives 19:54
moritz_ slavik: "etf" being "Exchange Traded Fund"?
jhorwitz slavic: try throwing in a handler configuration (doesn't have to work yet)
^slavik
slavik lol 19:56
was supposed to be wtf
moritz_ ah
slavik jhorwitz: where should I through it and what should it look like?
jhorwitz: parrot.load?
jhorwitz see the counter in eg/perl6 19:57
slavik k
jhorwitz mod_perl6 is loaded and working
just has no config, and it doesn't know how to deal with that.
20:03 cknapp left
slavik hmm 20:07
how can I have no response handler or w/e it is
jhorwitz not sure what you mean 20:09
slavik I did addhandler and perl6requesthandler
20:09 nacho joined
slavik no dice, or do I have to do sethandler? 20:09
jhorwitz no -- follow the README exactly 20:10
slavik :(
jhorwitz actually, the README stops short -- the examples have configurations
it looks exactly like mod_perl
slavik what about the counter file 20:11
jhorwitz for a response handler in a directory, say SetHandler perl6-script, then Perl6ResponseHandler Your::Module
slavik what about global? 20:12
isn't it AddHandler perl6-source .p6, then the Perl6ResponseHandler?
jhorwitz the "counter" file is an example of a registry script
slavik right
jhorwitz the modules in ModPerl6 are examples of real handlers
the config in counter will set up a directory for registry scripts. 20:13
slavik AddHandler perl6-script .p6
Perl6ResponseHandler ModPerl6::Registry
I want perl6 to be global
jhorwitz ah
you're reading ahead. :) 20:14
slavik oh
that's what I have for root
I have similar for Perl5
jhorwitz i wanted to make sure the examples worked first
slavik jhorwitz: let's assume they do :)
jhorwitz you assume a lot. :)
20:15 cknapp joined
slavik jhorwitz: I trust in your code writing skill :) 20:15
jhorwitz famous last words...
slavik otherwise I will go to PHP ... and we know it sucks
jhorwitz LOL
well, i have that working too. :
:)
slavik no pointy blocks, arrays that thing they are hashes ... etc
so do I, your point?
jhorwitz i meant under mod_parrot
slavik Counter.pm? 20:16
jhorwitz anyhoo...
php (uses parrot's pipp compiler)
slavik php is crap, perl6 is where it's at 20:17
jhorwitz giddyup
slavik sure thing, Cosmo
jhorwitz :)
20:17 cknapp left
jhorwitz ok, to make a global handler for .p6 files... 20:17
slavik don't stop 20:18
:-\
jhorwitz i'm not sure if mod_perl6 will support that yet. it's not hard, just have to tweak it. 20:19
slavik err, why would mod_perl6 need to support it? isn't it an apache config issue?
given a *.p6, it should call mod_perl6 or w/e 20:20
jhorwitz mod_perl6 looks for a handler name it recognizes (perl6-script, etc), but it expects to also have a Perl6ResponseHandler setting
you *can* do it with a <Files *.p6> block
slavik so, I set it to registry :)
the handler is ModPerl6::Registry 20:21
jhorwitz not sure if Perl6ResponseHandler is valid in the main server config
slavik seems to be for Perl5 20:22
jhorwitz well, that's mod_perl2
:)
i can make that happen though
slavik good :D
jhorwitz see if <Files *.p6> works for you right now
slavik where do I put it? 20:23
to add handler?
I am not an apache guru :(
or is that a sectoion? 20:24
pugs_svn r24202 | moritz++ | [t] move operator_overloading.t to spec/
jhorwitz it's a section. put "SetHandler perl6-script" and "Perl6ResponseHandler ModPerl6::Registry" in there
slavik k and this section is under a directory?
jhorwitz no, it's on its own 20:25
slavik k
jhorwitz says "use this config for all files ending with .p6"
brb 20:27
slavik jhorwitz: segfault 20:29
20:31 ft left
pugs_svn r24203 | moritz++ | [t] cleanup of flip-flop.t 20:31
jhorwitz poop
that really shouldn't happen 20:32
pugs_svn r24204 | moritz++ | [t] move flip-flop.t to spec/
jhorwitz slavik: works for me 20:33
can you paste your config?
slavik k 20:34
pasteling "slavik" at 12.130.86.9 pasted "config" (15 lines, 514B) at sial.org/pbot/33539 20:35
slavik der
jhorwitz when did it segfault? startup or when you made a request? 20:39
20:39 riffraff joined
slavik start up 20:40
jhorwitz hm 20:41
can you run this: source /path/to/envvars 20:43
then gdb /path/to/httpd
then run -X
you should see where it's segfaulting. if so, type bt and paste the back trace
pmurias ruoso: is there a memory leak debugging tool we are lacking? 20:47
i'm thinking of writing a thing which would print the stack trace during the allocation of the missing objects 20:48
slavik bt 20:49
?
pmurias gdb command 20:50
slavik oh, ok
pugs_svn r24205 | moritz++ | [t] move submethods.t to spec/, clean up some eval clutter 20:51
r24206 | moritz++ | [t] no need for a Synopsis dir, it was mostly outdated anyway 20:53
slavik no debug symbols :( 20:54
20:55 sudvabalis joined
sudvabalis help me surprise my girlfreind, watch this video on youtube, thanx! www.youtube.com/watch?v=LHsitCVMbgE 20:56
20:56 sudvabalis left
slavik wtf? 20:56
I say we spam him
moritz_ ignores spam
20:57 slavik left
jhorwitz @tell slavik set MP_DEBUG environment variable to -g, rebuild and reinstall mod_parrot, then run the gdb session again. 20:58
lambdabot Consider it noted.
20:58 stephens joined 21:01 ruoso left, lambdabot left 21:03 stephens left 21:04 lambdabot joined
bacek perl6: token a { 'a' }; token b { 'b' }; rule Foo { <a> | <b> }; given $/<Foo> { when .<a> { say "a"; }; when .<b> { say "b"; }; }; 21:04
p6eval rakudo 33630: OUTPUT[get_pmc_keyed() not implemented in class 'Undef'ā¤current instr.: '_block11' pc 46 (EVAL_16:27)ā¤]
..pugs: OUTPUT[*** Can't modify constant item: VUndefā¤ at /tmp/lgGduPl0kw line 1, column 80-85ā¤ /tmp/lgGduPl0kw line 1, column 80-85ā¤]
..elf 24206: OUTPUT[Can't locate object method "cb__RegexDef" via package "EmitSimpleP5" at ./elf_f line 1612.ā¤]
bacek is it legal syntax?
or I always should use if/elsif/else? 21:05
moritz_ bacek: yes, but until you match your regex you won't get any meaningful result
bacek perl6: token a { 'a' }; token b { 'b' }; rule Foo { <a> | <b> }; 'b'~~ / <Foo> /; given $/<Foo> { when .<a> { say "a"; }; when .<b> { say "b"; }; };
p6eval rakudo 33630: OUTPUT[Method 'ACCEPTS' not found for invocant of class 'PGE;Match'ā¤current instr.: 'infix:~~' pc 14429 (src/builtins/match.pir:18)ā¤] 21:06
..pugs: OUTPUT[Error eval perl5: "if (!$INC{'Pugs/Runtime/Match/HsBridge.pm'}) {ā¤ unshift @INC, '/home/evalenv/pugs/perl5/Pugs-Compiler-Rule/lib';ā¤ unshift @INC, '/home/evalenv/pugs/third-party/Parse-Yapp/lib';ā¤ eval q[require 'Pugs/Runtime/Match/HsBridge.pm'] or die $@;ā¤}ā¤'Pugs::Runti...
..elf 24206: OUTPUT[Can't locate object method "cb__RegexDef" via package "EmitSimpleP5" at ./elf_f line 1612.ā¤]
bacek moritz_: thanks
looks like another TODO ticket for rakudo.
moritz_ let me check the S03 smart match table first...
yes, Any .foo - method truth 21:09
21:11 ruoso joined 21:14 Entonian joined 21:27 jferrero left 21:28 jhorwitz left
riffraff does perl6 specify some builtin way to use a string as an IO object? 21:32
moritz_ .open, or what do you mean? 21:34
riffraff well, something that can be read and written to but has data in a string instead of a file 21:36
pretty useful for testing, in my case
21:40 iblechbot left
moritz_ like perl5's open($handle, '<', \$scalar) ? 21:43
don't know if that works in Perl 6
riffraff I was thinking of IO::String on cpan, but just cause I didn't know this trick in perl5 :) 21:46
moritz_ ;-) 21:47
22:04 justatheory left 22:18 pcbuetow joined 22:19 dr_df0 left 22:22 sail0r joined 22:23 sail0r left, ejs2 left, rindolf joined
rindolf Hi all. 22:24
rakudo_svn r33637 | infinoid++ | [cage] Fix some trailing spaces. 22:30
22:41 alester left 22:54 pcbuetow left
meppl good night 22:59
23:00 meppl left
pugs_svn r24207 | moritz++ | [t/spec] tests for Str.subst with :g, :x() and :nth(), both single and combined 23:04
23:08 pmurias left 23:09 apeiron left
riffraff moritz_, interesting, I found an io_string.c in parrot sources which behaves as I was saying before.. I wonder where it surfaces in perl 23:11
moritz_ riffraff: either it's specced in S16, or it waits for your spec work ;-) (S16 is heavily under-specced, and needs a lovin' hand) 23:14
riffraff :) 23:18
riffraff adds "check s16" at the end of his eternal todolist 23:19
moritz_ I could imagine a constructor IO.new(:scalar($source_scalar)) and IO.new(:list(@source_list))
the first case would split $source_scalar on the input record separator... 23:20
and the second one would take the items in @source_list as input recrods
23:22 rindolf left 23:24 apeiron joined 23:25 Entonian left
riffraff yes, but the latter would be confusing when stuff like seek() is used, I think, easier to just join() it with a suitable record separator I think 23:25
anyway, it seems that io_string.c is useless in parrot, cool 23:26
moritz_ useless? because it's not exposed to Perl 6?
there are other languages out there... 23:27
riffraff no because the functions defined there do not appear anywhere else when I do grep funcname . -r 23:28
but probably I'm mistaken
23:42 nothingmuch left 23:59 Entonian joined

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!