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.
ep0ch No. 00:49
mncharity /me wonders what my Int Num f() {}; means (to choose two random fulltypename's). 02:20
[particle1 i can't see how that will parse 02:23
[particle] my Int of Num f() {}; will parse, but fail in the semantic analyzer, i suspect 02:25
see fulltypename in svn.pugscode.org/pugs/src/perl6/STD.pm 02:26
mncharity scope_declarator:my -> scoped -> <fulltypename>* <routine_declarator> 02:27
it appears and intentional construct
[particle] token scope_declarator:my { <sym> <scoped> {*} } #= my
mncharity yes 02:28
[particle] ah, i see it now.
that should likely be a ?
mncharity ah. hmm. 02:29
[particle] feel free to update. it'll get reverted if you're wrong
mncharity though isn't my Dog Animal $x; valid?
mncharity or, um, my Dog HeatSource $fido; 02:29
[particle] i'll have to check the synopsis to be sure.
imo it leads to ambiguity 02:30
my Dog|Heatsource $fido; might work
but my Array of Int Array of Num $x; is ambiguous 02:31
or more simply my Array of Int Num
i'm just not sure what multiple typenames allow you to do that it would be valid 02:32
mncharity re changing STD.pm, I don't. On the model that I am usual confused, and TT's few mistakes are of a recognizable flavor (non-local inconsistency), of which this case isn't.
my *very* fuzzy recollection is they might be additional constraints. Eg, $fido can't be a cold dog. 02:33
but that opinion plus five cents will get you... five cents.
err, oops. my example should have been my Int Num sub f() {}; 02:35
i'm puzzling over whether the IR for sub declaration needs a variable node hanging off it, to handle the details of that &f . 02:37
ah well. inline typenames for now, refactor later. 02:40
TimToady: pluralized and scoped have a "'(' <signature> ')' <trait>*". could that be made a subrule? 02:47
mncharity TimToady: variable_decl special cases '=' and '.='. Does that mean my Foo $x += 3; is invalid? for all "+=" ops other than = and .= ? 02:55
/me wonders what to call the "scope typenames plurality" tuple, since its showing up in a lot of decls. eg, DeclareSub scope typenames plurality ident multisig traits block 02:59
mncharity anyone have an opinion on whether to pull out some infix:foos as IR nodes? And which? Bind and Assign being two candidates. an argument being it might simplify IR tree analysis. but I'm not sure it actually does... might be better off with a is_bind? attribute on an Apply("infix:=",...) node. 03:06
s/attribute/annotation/ 03:07
meppl good night 03:21
mncharity good night 03:29
TimToady: in param_var, it seems like <?{ $<sigil> eq '&' }> <?ident> :: <ident=sublongname> eats and discards an ident? 03:30
ingy is bored. know any good projects I can work on? 03:35
mncharity elf! :) 03:36
ingy happy fool's day, #perl6! 03:37
mncharity <render in deep att-tv-commercial voice>"Have you ever wanted to write a p6 backend? A compiler? A prelude? _You will._"</> 03:38
doh
:)
almost made it. only 1/2 hour left. drat.
pugs_svnbot r20206 | putter++ | elf_d_src/Notes.txt: Start of an IRx1 "IR experimental 1" draft. 03:43
diff: dev.pugscode.org/changeset/20206
lambdabot Title: Changeset 20206 - Pugs - Trac
pugs_svnbot r20207 | putter++ | [elf_c] Some refactoring work. 04:49
diff: dev.pugscode.org/changeset/20207
lambdabot Title: Changeset 20207 - Pugs - Trac 04:49
JDlugosz Hello world 11:47
JDlugosz Hello world 11:48
allbery_b whoa. it's audrey! (on list) 15:42
pugs_svnbot r20208 | lwall++ | [STD] inside of () and [] should be semilist, not statementlist 17:34
r20208 | lwall++ | [Cursor5] s:g/warn/print STDERR/ because warn makes debugger too verbose
diff: dev.pugscode.org/changeset/20208
lambdabot Title: Changeset 20208 - Pugs - Trac
[particle] "i don't need a jet to get jet lagged" sounds more like a dylan quote than a wall quote. 17:52
pugs_svnbot r20209 | putter++ | [elf_c] Started adding use(). Refactoring towards runtime-swappable compilers. 18:16
r20209 | putter++ | ../elf_c -x -o ../elf_c1 ElfC.pm now works.
r20209 | putter++ | elf_c can no longer be compiled by elf_b, only by itself.
diff: dev.pugscode.org/changeset/20209
lambdabot Title: Changeset 20209 - Pugs - Trac
mncharity so quiet 19:10
nwc10 just the tumbleweed^Wcommits rolling through 19:11
mncharity :) 19:13
pmichaud which is better, lots of discussion and no commits, or vice-versa? ;-) 19:18
nwc10 light is better than heat 19:19
mncharity lol #re which is better 19:22
mncharity hi ruoso 19:23
ruoso hi mncharity
mncharity /me ponders 'light is better than heat'
nwc10 long mailing list threads can end up being referred to as "more heat than light' 19:24
nwc10 the sort that end up as a lot of arguments, and little or no progress 19:25
mncharity right
pmurias mncharity: Apply can be replaced by Call 19:29
mncharity in elf_d's IRx1? 19:30
pmurias yes
mncharity currently Apply is function application, Call is method call. 19:31
pmurias function application is also method call 19:32
mncharity nod
pmurias the hard part is the method name 19:32
mncharity err, un-nod. I'm unclear. 19:33
mncharity I don't have a clear concept of how methods and subs are different. 19:33
/me goes to read Snn... 19:34
feather.perl6.nl/syn/S06.html
lambdabot Title: S06
pmurias &foo(123) is &foo."postcircumfixfix:<( )>"(123) 19:35
&foo(123) is &foo.postcircumfixfix:<( )>(123)
mncharity really? 19:36
pmurias yes 19:37
mncharity spec? 19:38
pmurias indirectly
grep for postcircumfix:<( )> 19:39
and .postcircumfix:<[ ]>
shower& 19:40
mncharity feather.perl6.nl/syn/ "Google Search the Synopses" is broken :( 19:42
lambdabot Title: Official Perl 6 Documentation
pmurias (not yet) ;)
mncharity: i use ack
mncharity ack?
alester ack!
petdance.com/ack/
lambdabot Title: ack -- better than grep, a power search tool for programmers
alester A standard tool for Parrot folks.
[particle] eats ack for breakfast 19:43
pmurias alester: is it especialy needed for parrot work?
alester It's very helpful.
most Parrot people use it.
[particle] c code, perl code, pir code. ack helps me search only the files i'm interested in inside the parrot repo
mncharity so grep-like. but... oh, are the Snn in the pugs tree? 19:44
pmurias i just use as a grep with perl5 syntax
[particle] svn co svn.perl.org/perl6/doc/trunk
lambdabot Title: Revision 14538: /doc/trunk
mncharity ahhhh
mncharity thanks! :) 19:45
TimToady mncharity: <?ident> is just shorthand for <?before <ident> > there, I think 19:49
and yes, all the declarative syntax could probably use a going over 19:50
= and .= are pseudoassignment forms slurped by declrators. my $x += 1 parses as (my $x) += 1 instead
in general, assignops are mutators that make little sense on a new variable 19:53
mncharity ok 19:54
[particle] TimToady: did you see yesterday's comments on <fulltypename>* ? 19:55
TimToady 0 or more is correct 20:00
it's not ambiguous insofar as the parser can know which identifiers are predeclared as typenames
rindolf Hi all! 20:01
TimToady: (and everyone) You may wish to review www.shlomifish.org/same-as-perl.html before I publish it.
TimToady phone &
mncharity pmurias: While S02 says "The C<Callable> role implies the ability to support C<< postcircumfix:<( )> >>.", it also talks of function calls without a pcf(). eg, 'say 3'. Perhaps 'say 3' should be interpreted 'pcf()(say,3)' or 'say.pcf()(3)'. But that's not clear to me. 20:02
mncharity 'say(3)' and 'say.(3)' don't parse using postcircumfix. though that's not definitive, as, for instance, neither does 'while', but one could argue the parser should emit 'statement_control:while(...)' and the IR shouldn't have a While or Loop node. 20:11
s/neither does 'while'/'while' is parsed specially/ 20:12
s/specially/specifically/
hmm. Sub/Method/Macro decl could be collapsed into a RoutineDecl. and Call/Apply/MacroApply collapsed into a Call. RoutineDecl would still need a flag distinguishing them, so it's not really a collapse, just a refactoring. 20:17
Call could be a real collapse. 20:18
but I'm still unclear on the connection between Call and postcircumfix:<( )>, and if they are distinct, whether function application and method call are also distinct, in which case Call is just a refactoring as well. 20:19
And a potentially less interesting one. 20:21
(if emitters/analysis generally wishes to treat the cases differently) 20:22
/me puzzled. thoughts?
mncharity rindolf is a Llama? 20:25
pmurias mncharity: pcf = ? 20:28
mncharity postcircumfix 20:28
Juerd Does Perl 6 have a way to indicate that of a certain set of parameters, only one may be given? 20:30
e.g. :html or :text or :xhtml or :yaml 20:31
Or does it have to be :type<htlm>, :type<text>, :type<xhtml>, :type<yaml> (lots of typing)
mncharity so there are two optional parameters, and only one may be set? don't know of such.
Juerd Yes
pmurias TimToady: could you clarify the corespondence between function application and postcircumfix:<( )>? 20:32
mncharity assert [$html,$text,$xhtml,$yaml].count_defined == 1 ;) 20:34
assert_only_one_defined $html,$text,$xhtml,$yaml;
s/only/exactly
TimToady Juerd: just use two different multi sigs 20:59
Juerd TimToady: Oh, of course!
TimToady: Makes me happy ):
Eh
:)
Smileys should have parity 21:00
and error correction
TimToady pmurias: I can't clarify it unless I know what the question is :) 21:01
pmurias TimToady: is &foo.postcircumfix(...) equivalent to &foo(...) 21:05
? 21:06
spx2 hello 21:08
Juerd IIRC, it's &foo.postcircumfix:«( )»(...)
Hi
spx2 how is perl6 doing today ?
Juerd spx2: Rakudo got IO.
spx2 Juerd: cool !
Juerd So I guess Perl 6 is happy.
spx2 Juerd: you mean all packages in IO namespace ?
Juerd: or just "generic" IO ? 21:09
TimToady it is likely that the arguments to &foo.postcircumfix:«( )»(...) would be the same as the arguments to &foo.(...) 21:11
Juerd spx2: Simple IO
TimToady though that doesn't necessarily generalize to other postcircumfixes
spx2 Juerd: ok thanks
Juerd TimToady: I see recursion ;) 21:12
TimToady for instance, there might be some remapping from .[] to .postcircumfix:<[ ]>()
yes, well, the buck stops somewhere in there... 21:13
in general, the concept of .postcircumfix:<( )> as a standalone operator is rather useless 21:15
mncharity class A { sub f($obj;$y){} }; A.new.f(3) # valid? 21:17
TimToady I presume you meant : there, but no in any case 21:18
subs are not visible through the single dispatcher
mncharity re meant, yes. re rest, ah, ok.
TimToady and the colon is probably illegal in an ordinary sub declaration 21:19
spx2 may I ask a question ?
TimToady no 21:20
spx2 is there a page describing how perl6 funds are distributed ?
the funds from donations
I am interested in seeing the priorities of the short-term goals etc
is that written somewhere online ?
pmichaud spx2: depends on what you mean by "perl6 funds" 21:21
spx2 pmichaud: the funds raised by donations 21:22
pmichaud I don't think those are tied specifically to "Perl 6"
spx2 pmichaud: what do you understand by perl6 funds and what possible meanings would you find for that expression ?
pmichaud i.e., those are just "perl funds"
or, better still, simply "TPF donations".
(note: I'm not an officer or even particularly knowledgeable about any of this, so take everything I say with a good size chunk of salt.) 21:23
spx2 pmichaud: ok , I'm interested in where do TPF donations go as in concrete work of someone
spx2 pmichaud: ok no problem 21:23
pmichaud: you need not be an officer 21:24
pmichaud spx2: I think you want www.perlfoundation.org/grants
pugs_svnbot r20210 | putter++ | [elf_d] A first cut at elf_d, with a new IR. 23:48
diff: dev.pugscode.org/changeset/20210
lambdabot Title: Changeset 20210 - Pugs - Trac

Powered by Cro and the Raku® Programming Language.

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

Thank you!