www.parrotcode.org/ | Last release: 0.7.1 "Manu Aloha"
Set by moderator on 17 September 2008.
Tene Can anyone here help me with ncigen? 00:02
00:09 AndyA joined 00:17 kid51 joined 00:31 Andy joined
Tene Looks like I need either kjs or tewk, but neither is around. 00:35
02:01 dmknopp joined 02:14 jose joined
s1n jonathan: ping 03:04
03:26 ab5tract joined 03:49 johbar_ joined 04:04 Theory joined 04:09 contingencyplan joined 04:12 Debolaz joined
Tene Okay, I think I have a correct throw/EH patch. 04:50
Ontolog :method and :multi are incompatible??? 04:59
why didn't pmichaud say something about it before though?
05:03 Theory joined
nopaste "tene" at 166.70.38.237 pasted "Remove second argument from error handlers. A few test failures. Please test." (1746 lines) at nopaste.snit.ch/14137 05:06
Tene I'm done for now. Be back later.
cotto_home Your nopaste is broken. 05:12
ENONOPASTEPASTE
06:22 Zaba_ joined 06:54 johbar joined
dalek r31293 | allison++ | pdd27mmd: 07:26
: [pdd27mmd] Skipping intermittently hanging test. Most of the rest of this test
: file is skipped anyway, and system needs extensive reworking.
diff: www.parrotvm.org/svn/parrot/revision?rev=31293
Ontolog Is the design of Parrot still being tweaked or is it completely finished? 07:43
cotto_home at what level?
Ontolog like pir for example 07:44
is pir completely defined and frozen?
i'm not saying completely implemented
just, is the design finished and frozen?
cotto_home for the most part, although the target is to replace imcc with pirc
that shouldn't break too much, though 07:45
I'd say mostly defined and quite cold
Ontolog i thought imcc is pir
cotto_home imcc is the current compiler used to compile pir.
Ontolog i just wonder because i am concerned that parrot isn't ready for perl 6 07:46
if things keep changing
like i heard io is being redesigned?
cotto_home pirc is a fairly recent effort to reimplement a pir compiler
IO isn't being redesigned, just properly implemented 07:47
Ontolog i see
someone said that :method and :multi are incompatible
that one precludes the other 07:48
cotto_home how it should work is documented in docs/pdd/pdd22_io.pod
Ontolog is it true?
cotto_home goes to check the pdd
I can't say whether that's supposed to be the case or not. 07:50
If it's causing issues for you, you can ask at the next #parrotsketch. 07:51
Ontolog would it be fair to call multiple dispatch "runtime overloading"? 07:54
is there really any difference? 07:55
cotto_home I think that'd be an accurate way to refer to it.
Ontolog ok
cotto_home mmd is fancier
Ontolog what does it be that "runtime overloading" doesn't imply?
s/be/do/ 07:56
the dispatch system is pluggable, allowing users to swap in their own
type-matching algorithms. ??
that's interesting 07:57
i wonder if Perl 6 does that
Tene cotto_home: pleasedieinafire.net/~tene/0001-Rem...andl.patch 07:59
cotto_home what? 08:00
you want me to die in a fire?
;)
Tene That's the patch that I tried to post to nopaste
Lessee what test failures I get before the patch... 08:02
cotto_home removing redundancy is a good thing 08:03
Tene rakudo still runs fine. I think that most of the errors are errors in the updates I tried to make to fix the tests.
So broken tests, not broken functionality.
Still need to be fixed, though.
cotto_home t/pmc/namespace.t failed 3 tests 08:07
t/pmc/resizablestringarray.t -> 1
nopaste?
clunker3 pasta.test-smoke.org/ or paste.husk.org/ or nopaste.snit.ch:8001/ or rafb.net/paste or poundperl.pastebin.com/ or paste.scsys.co.uk/
purl nopaste is at nopaste.snit.ch/ (ask TonyC for new channels) or rafb.net/paste or poundperl.pastebin.com/ or paste.scsys.co.uk/ or App::Nopaste or tools/dev/nopaste.pl or at www.extpaste.com/
cotto_home t/library/pg.t didn't pass any 08:09
poundparrot.pastebin.com/m5f93886f 08:14
Tene Yeah, all the same ones I'm getting. Lessee... 08:15
cotto_home good news, although I'm not running anything exotic 08:21
sleep & 08:24
Tene Okay, this is much cleaner. 08:27
Still not perfect, though.
Zaba_ *yawn* 08:28
Tene Okay, no more test failures in parrot's 'make test' 08:32
Anyone know how to delete a branch in svn? 08:37
moritz svn rm /path/to/branch 08:43
Tene I'm always nervous dealing with svn. I've broken things more times than I'm comfortable with. 08:44
moritz can very much understand that 08:45
Tene Okay, committing. 08:46
dalek r31294 | tene++ | trunk: 08:48
: [parrot]
: Remove the second argument passed to exception handlers.
: You can get the message by stringifying the exception object.
: All parrot tests pass. Looks like the languages are all okay too.
diff: www.parrotvm.org/svn/parrot/revision?rev=31294
Tene There we go. I'm pleased with it. 09:00
moritz runs rakudo tests 09:02
Tene In rakudo I only see failure in an io test, which I think failed before. 09:03
moritz maybe we should skip those; I've seen reports of them looping before 09:04
cotto_home (testing infinite loop speed)-- 09:09
Tene karma tene 09:12
purl tene has karma of 211
09:13 rurban_away joined
Tene Prime! 09:13
cotto_home you're wierd
Tene karma moritz
purl moritz has karma of 241
cotto_home *weird
Tene Also prime!
rurban karma rurban
purl rurban has karma of 95
Tene me + 30
rurban :(
Tene ^5 moritz
cotto_home karma prime 09:14
purl prime has karma of 1
Tene karma cotto
purl cotto has karma of 125
Tene 5^5!
cotto_home 5^3
purl 6
Tene erm
5^3
purl 6
Tene yeah
cotto_home 5**3
purl 125
rurban I need to get more karma because I successfully persuaded my new Globetrotter modem to work for me after two days struggle. 09:15
success!
purl somebody said success was finding king size papers
Tene rurban++ # success at modem struggles
cotto_home rurban++ #what Tene said 09:16
Tene I kinda like 211. Maybe I'll stay here for a while.
Ontolog ok I'm trying to understand more how parrot works so i can better help rakudo, i would like to know where to start when one wants to write a language compiler with parrot 09:17
i know there are lots of examples
but is there like a document that outlines all the basics?
cotto_home kjs wrote a tutorial on that exact subject
Tene Ontolog: yes, the squaak tutorial on parrotblog
Ontolog ok i'll look
rurban pbc_merge is failing on windows. elsewhere also? just lua needs it. 09:18
09:18 barney joined
cotto_home www.parrotblog.org/ (check the sidebar) 09:18
moritz rurban: it has been failing somtimes for me
dalek r31295 | rurban++ | cygwin070patches:
: [cage] protect against too old Storable-2.09, which cannot overload stringify
diff: www.parrotvm.org/svn/parrot/revision?rev=31295
moritz rurban: sometimes even only when rum from 'make smoke', not when run directly 09:19
barney cotto_home: pipp.org/ seems to be down
dalek r31296 | rurban++ | cygwin070patches: 09:20
: [cage] update branch-private STATUS
diff: www.parrotvm.org/svn/parrot/revision?rev=31296
cotto_home barney, it might be godaddy's fault 09:21
it looks like they're doing some kind of maintenance
barney k
cotto_home (I'm going to switch away from ASAP, but it can't happen until 60 or 90 days after I first got the domain.)
thanks for noticing, though 09:22
I'll check again on the hour
dalek r31297 | rurban++ | cygwin070patches: 09:25
: [BASIC] add Makefile
diff: www.parrotvm.org/svn/parrot/revision?rev=31297
r31298 | rurban++ | cygwin070patches: 09:37
: [cage] RT#39742 installed-conflict
: * remove SHRPENV
: * cygwin went now the win32 way and puts its dll and importlib into
: build_dir and not into blib/lib.
: So we need no SHRPENV PATH=...blib/lib:$PATH overrides for every parrot
: and pbc_to_exe invocations (as in perl5).
: For the other unix-alikes with libparrot_shared we simply
: export LD_RUN_PATH := @blib_dir@:..., as for darwin.
diff: www.parrotvm.org/svn/parrot/revision?rev=31298
r31299 | tene++ | exceptionmagic:
: Remove the exceptionmagic branch, as it's not needed for now.
diff: www.parrotvm.org/svn/parrot/revision?rev=31299 09:38
moritz my girlfriend says hi and wishes a good sunday localtime() to all of you ;-) 09:39
rurban tene: why? 09:41
dalek r31300 | rurban++ | cygwin070patches: 09:42
: [cage] RT#39742
: * remove "Building a shared parrot library may conflict..." warning
: * link explicitly against library file, as /usr/lib precedes -Lblib/lib path
: * fix cygwin linker only if old ld2. otherwise keep g++ 09:43
: * outcomment cygchkdll as it is not needed anymore.
diff: www.parrotvm.org/svn/parrot/revision?rev=31300
Tene Hi, moritz' girl.
rurban: I'm just going to re-create it the next time I want to work on it. Easier than keeping it up to date.
The next thing I need to do on exceptions is add stuff to the exceptionhandler to tall it which types of exceptions it can handle. 09:44
rurban ok.
dalek r31301 | rurban++ | cygwin070patches:
: [cage] RT#39742
: * same as for r31298: SHRPENV not needed anymore
diff: www.parrotvm.org/svn/parrot/revision?rev=31301
Tene after that, I might add automatic resume for non-fatal exceptions. 09:45
rurban shouldn't we protect parrot_hash_put(INTERP, hash, (void *)key, (void *)value); - I added the (void*) casts
int->ptr warnings
Tene sleep now.
rurban greetings back to moritz' friend 09:46
moritz rurban: thanks. How's your cygwin branch going?
rurban fine as ever.
moritz rurban: any plans for a merge?
Ontolog tools/dev/mk_language_shell.pl does not create a Makefil
rurban see STATUS-branch
moritz ok
rurban Ontolog: it creates the makefiles/root.in 09:47
Ontolog however it does make config/makesfiles/root.in
yes
moritz Ontolog: I think the Makefile is created by Configure
rurban this is enough
Ontolog moritz: which Configure? 09:48
moritz Ontolog: parrot's
purl parrot's is modelled after perl 5's
rurban I added a lot of stuff to this tools/dev/mk_language_shell.pl in the cygwin070patches branch: pmc's, ops, ...
and useful Makefile additions for make install later.
make uninstall is in still work 09:49
moritz: cygwin070patches should have been merged last week. But I'm not in the position to enforce that. 09:51
Ontolog I ran Configure.pl but still no makefile in languages/mylang
dalek r31302 | rurban++ | cygwin070patches:
: [cage] update private todo: make uninstall
diff: www.parrotvm.org/svn/parrot/revision?rev=31302
Ontolog looking at Configure.pl's help I don't see anywhere how to generate the Makefile for a specific language 09:52
rurban Ontolog: you need to add your language to the list of languages in config/gen/languages.in
only then it is picked up by Configure
Ontolog ohhh
would be nice to add to that Squaak tutorial ;-) 09:53
thanks
rurban and several other places also (needs to be documented somewhere - can you write that down?)
Ontolog What is the format to config/gen/languages.in? The file doesn't exist so I can't see any examples 09:54
just adding the name of the language doesn't work
rurban it is .pm, not .in sorry
do you see it? 09:55
Ontolog yes i'm trying it now 09:56
ok that worked
i wrote it down for future reference
rurban for a full integration of a new language several more templates need to be touched also 09:58
Ontolog i see, for now i'm just trying to follow the squaak tutorial 09:59
dalek r31303 | rurban++ | cygwin070patches:
: [bcg] add some documentation to printPASM, fix endOp doc
diff: www.parrotvm.org/svn/parrot/revision?rev=31303
Ontolog but i gave my language a different name because squaak is already part of the source tree
rurban languages/LANGUAGES_STATUS.pod, config/gen/makefiles/languages.in, config/gen/languages.pm, out of my head
dalek r31304 | rurban++ | cygwin070patches: 10:03
: [cage] partial update to trunk
diff: www.parrotvm.org/svn/parrot/revision?rev=31304
rurban BTW: I registered myself as speaker on the TwinCity perl meeting in vienna/bratislava. 10:04
some parrot talk and some perl-compiler talk
lightning talk: "Short overview of the current parrot status and plans (not perl6 aka "rakudo"), and the yet undecided problems on making parrot and its languages installable" 10:06
and 40min: "Need perl-compiler help: Hack the optree and emit C or JIT" 10:07
cotto_home rurban++ #using "its" correctly 10:16
dalek r31305 | allison++ | pdd27mmd: 10:18
: [pdd27mmd] Add a function 'Parrot_pcc_invoke_sub_from_c_args', which performs a
: PCC invoke using a CallSignature PMC, after building the CallSignature from C
: varargs.
diff: www.parrotvm.org/svn/parrot/revision?rev=31305
r31306 | allison++ | pdd27mmd: 10:20
: [pdd27mmd] Cleaning out deprecated 'n_*' opcodes from NQP and Rakudo.
diff: www.parrotvm.org/svn/parrot/revision?rev=31306
10:21 Zaba joined
Ontolog is docs/pdds/pdd03_calling_conventions.pod mostly deprecated? 10:32
the set_args stuff?
jonathan No, it's current. 10:38
Ontolog but none of the code I see uses set_args set_returns or the like 10:40
is that stuff only pertinent to PASM?
jonathan It does really. :-)
PIR generates those for you from the syntactic sugar for a call.
so foo = bar(baz) is actually four instructions. 10:41
Or maybe 3.
purl somebody said maybe 3 was good enough *shrug*
Ontolog i see, makes sense
how does Parrot know if some PMC is an aggregate type?
jonathan It'll most likely be declared "provides array" or "provides hash". 10:42
Ontolog i see 10:43
is it true that :method and :multi are semantically incompatible?
jonathan No. 10:44
Ontolog i didn't think so 10:45
but if you see my last mail to the perl6-compiler list someone says it is
i was hoping you had some time to respond to "method signature issues" 10:46
as i'm still a bit confused about what is going on there and it's sort of blocking me
jonathan I'm a bit confused by what chromatic said. If that were true, how'd we do multimethods... 10:48
Ontolog yeah, i was hoping someone could clarify it, i'm not sure who is the authority on this particular topic though 10:50
jonathan I've been working on the Perl 6 side of multiple dispatch. 10:51
And as I think was mentioned before, there's some work to do in order to get built-ins written in PIR integrated into that and using the Perl 6 dispatch algorithm. 10:52
Replied to chromatic's post. 10:53
11:08 Zaba joined 11:11 Ontolog joined 11:15 TonyC joined
dalek r31307 | rurban++ | cygwin070patches: 11:15
: [docs] update from trunk: added ch08_pct, ch09_hlls
diff: www.parrotvm.org/svn/parrot/revision?rev=31307
r31308 | rurban++ | cygwin070patches: 11:24
: [install] put all runtime files into lib
diff: www.parrotvm.org/svn/parrot/revision?rev=31308
r31309 | rurban++ | cygwin070patches: 11:29
: [cage] add new files from trunk, fix svn props
diff: www.parrotvm.org/svn/parrot/revision?rev=31309
11:32 clunker3 joined
dalek r31310 | rurban++ | cygwin070patches: 11:32
: [cage] add new files from trunk
diff: www.parrotvm.org/svn/parrot/revision?rev=31310
cotto_home clunker3, owner 11:38
11:48 ruoso joined 11:54 bacek joined
dalek r31311 | allison++ | pdd27mmd: 11:54
: [pdd27mmd] Temporarily override the default generated vtable function for
: 'get_integer'. Later modify Pmc2c generator to call 'pcc_invoke' instead of old
: style Sub invocation from C.
diff: www.parrotvm.org/svn/parrot/revision?rev=31311
12:02 contingencyplan joined 12:05 rurban joined 12:24 Ontolog joined
dalek r31312 | allison++ | pdd27mmd: 12:25
: [pdd27mmd] Line length fix.
diff: www.parrotvm.org/svn/parrot/revision?rev=31312
r31313 | fperrad++ | trunk: 12:39
: [Markdown]
: - implement BulletList & OrderedList
: - and add tests
diff: www.parrotvm.org/svn/parrot/revision?rev=31313
12:42 kid51 joined
dalek r31314 | allison++ | pdd27mmd: 12:58
: [pdd27mmd] Remove deprecated 'n_*' opcodes from LOLCODE.
diff: www.parrotvm.org/svn/parrot/revision?rev=31314
jonathan I CAN HAZ UPDATEZ. 13:00
cotto_home IZ DEPURKATID NAO 13:07
NO CAN UZE
13:11 Ademan_ joined 13:15 masak joined 13:24 Whiteknight joined
masak rakudo: $foo 13:31
polyglotbot OUTPUT[Scope not found for PAST::Var '$foo'ā¤current instr.: 'parrot;PCT::HLLCompiler;panic' pc 156 (src/PCT/HLLCompiler.pir:103)ā¤called from Sub 'parrot;PAST::Compiler;post_children' pc 1743 (src/PAST/Compiler.pir:367)ā¤called from Sub 'parrot;PAST::Compiler;as_post' pc 2008
..(src/PAST/Compiler.pir:495)ā¤called from Sub 'parrot;PAST::Compiler;post_...
masak rakudo: eval '$foo'
polyglotbot RESULT[Null PMC access in find_method()ā¤current instr.: '_block11' pc 25 (EVAL_12:12)ā¤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 806 (src/PCT/HLLCompiler.pir:481)ā¤called from Sub 'parrot;PCT::HLLCompiler;evalfiles' pc 1080 (src/PCT/HLLCompiler.pir:611)ā¤called from Sub
..'parrot;PCT::HLLCompiler;command_line' pc 1259 (src/PCT/HLLCompiler.pir...
masak that can't be right :) 13:32
also,
rakudo: '$'.perl.say
polyglotbot OUTPUT["$"ā¤]
13:32 Zaba joined
masak rakudo "$" 13:32
rakudo: "$"
polyglotbot OUTPUT[Can't use $ as non-variable in interpolated string at line 1, near "$\\""ā¤ā¤current instr.: 'parrot;PGE::Util;die' pc 119 (runtime/parrot/library/PGE/Util.pir:82)ā¤called from Sub 'parrot;Perl6::Grammar;quote_term' pc 97966 (src/parser/quote_expression.pir:319)ā¤called from Sub
..'parrot;Perl6::Grammar;quote_concat' pc 97643 (src/parser/quote_e...
masak IOW, .perl generates un-runnable code in this case 13:33
jonathan :( 13:35
tickets++
Zaba rakudo: "$$" 13:36
polyglotbot OUTPUT[Can't use $ as non-variable in interpolated string at line 1, near "$$\\""ā¤ā¤current instr.: 'parrot;PGE::Util;die' pc 119 (runtime/parrot/library/PGE/Util.pir:82)ā¤called from Sub 'parrot;Perl6::Grammar;quote_term' pc 97966 (src/parser/quote_expression.pir:319)ā¤called from Sub
..'parrot;Perl6::Grammar;quote_concat' pc 97643 (src/parser/quote_...
masak jonathan: I'd submit one, but I'm trying to fool the discoverer into doing it first :)
jonathan Yeah yeah, just so you look less responsible for the queue size :-P 13:37
masak too late for that...
jonathan :-)
I'm not sure I want to think, how much it will grow while I'm away. :-)
masak you're going away?
jonathan Yes.
masak misses jonathan already 13:38
how long?
purl How long must we sing this song?
jonathan About a month.
masak purl, how long must you pester us with your incomplete wit?
purl masak: no idea
jonathan Going to travel across Russia.
masak jonathan: almost November when you get back, then :)
jonathan Yes.
Well, hope it will give me lots of fresh energy for hacking. :-) 13:39
masak indeed
jonathan I went away for October last year, and it wasn't that long after this, that I got involved in Rakudo.
masak we'll try to pull ourselves along without you
jonathan You'll do fine without me. :-)
masak who knows, maybe that will force me to learn more about Parrot
frustration is the mother of knowledge 13:40
jonathan Yeah! Then you can fix some of your own bugs. ;-)
13:40 rurban joined
masak cetere censeo #58392 delendam esse 13:40
jonathan Well, going to be in Vladivostok at start of my trip, so will meet Ilya there, who hacks on November. 13:41
masak Oh! Nice!
so you're doing Russia backwards? :)
jonathan I'd not heard it put that way before... 13:42
But yes, start in Vladivostok, and head to Moscow.
masak definitely backwards. 13:43
but, as they say, "In Soviet Russia..."
jonathan "In Soviet Russia, Perl 6 finishes you." 13:45
masak :) 13:46
rurban Soviet Russia is my big hope for the real optree gore (fixing the perl5 compiler) 13:53
jonathan Ah yes, you're coming to the twin city workshop looking for talent, I hear. :-) 13:55
14:18 Zaba joined
dalek r31315 | rurban++ | unknown: 14:26
: rename cygwin070patches branch to pdd30_install
diff: www.parrotvm.org/svn/parrot/revision?rev=31315
rurban jonathan: yes. I always loved to work with russian hackers on former projects (lisp and perl) 14:28
jonathan I'm looking forward to my time in Russia, and meeting more of them. :-) 14:32
rurban parallelinfo.mur.at/ru/novo.html and more (I did the webpages and mailinglist) 14:38
jonathan Novosibirsk is one city I'm not going to make it to on this trip. 14:41
14:47 Zaba joined 14:56 Zaba joined 14:57 Limbic_Region joined 14:59 tetragon joined
rurban What was the pdd09gc branch for? Switchable gc's? 15:03
Ah I see: "...to try to continue some of my GC work from the summer with a blank slate." 15:04
rurban_away going to the movies now 15:08
15:16 iblechbot joined 15:21 grim_fandango joined
dalek r31316 | fperrad++ | trunk: 15:34
: [Markdown]
: - implement EmphStar & StrongStar
: - and add tests
diff: www.parrotvm.org/svn/parrot/revision?rev=31316
15:35 jhorwitz joined
moritz masak: re #59112, looks like it might be a configure problem - did you try 'make realclean#? 15:40
masak moritz: is the final '#' significant? otherwise, yes 15:41
I always do `make realclean && perl Configure.pl && TEST_JOBS=2 make -j 2 test` nowadays
moritz it's not 15:42
same here, except that I also use --cc='ccache gcc'
masak ISTR that didn't work over here 15:43
15:46 apeiron joined
dalek r31317 | fperrad++ | trunk: 15:49
: [Markdown]
: - implement EmphUI & StrongUI
: - and add tests
diff: www.parrotvm.org/svn/parrot/revision?rev=31317
r31318 | allison++ | pdd27mmd:
: [pdd27mmd] Remove deprecated 'n_*' opcodes from Cardinal.
diff: www.parrotvm.org/svn/parrot/revision?rev=31318
16:06 Whiteknight joined
Zaba EmphUI? StrongUI? 16:07
moritz Zaba: markdown is a markup language ;) 16:08
Zaba oh, I think I did hear of it somewhre
where*
moritz used in some wikis, I think
masak aye 16:09
many "Use of uninitialized value" warnings when running make spectest_regression nowadays
moritz right 16:10
some of them are bogus
masak in what sense?
moritz that they shouldnn't warn
I'll try to find an example... 16:12
dalek r31319 | allison++ | pdd27mmd:
: [pdd27mmd] Don't multiply dispatch add_float, nothing defines a multi for it.
diff: www.parrotvm.org/svn/parrot/revision?rev=31319
masak my goal for today is to see how much of t/spec/S29-conversions/ord_and_chr.t starts passing in Rakudo if I just implement ord and chr as simple wrappers to their Parrot opcode equivalents 16:13
moritz masak: irclog.perlgeek.de/perl6/2008-09-16#i_570702 16:14
masak moritz: ah, that parsing error
moritz is that a parsing error?
masak moritz: well, I guess it's a question of how you define it... 16:15
moritz $something ~~ undef should be same as !defined($something)
masak a counts as a listop in that expression
so ~~ parses as two prefix ~
moritz ah, right
masak pmichaud says it a bit down in the discussion 16:16
irclog.perlgeek.de/perl6/2008-09-16#i_570741
moritz right, I should read on
masak what unsettles me is that I get the impression that there is no error involved, that 'a ~~ undef' simply means 'a ~(~undef)' 16:20
moritz yes, subname expression is the same as subname(expression) 16:21
masak well, that's ok 16:22
but ~~ was clearly what was intended here
moritz clearly = human perception 16:24
masak there's the rub, I guess 16:25
sushi &
moritz I don't like how it behaves, but I have no idea how to fix it without introducing ambiguty or inconsitency
bonne apetite ;) 16:26
16:26 Theory joined
pmichaud to fix a ~~ undef change to a() ~~ undef 16:29
then there's no ambiguity.
I agree that fixing this so that ~~ undef isn't parsed as ~ ~ undef is a good idea, but that needs discussion on p6l. 16:30
(unless it's already in the spec somewhere and I missed it.)
moritz STD.pm parses it as two prefixes as well 16:34
16:34 Zaba joined
dalek r31320 | allison++ | pdd27mmd: 16:57
: [pdd27mmd] Go ahead and do the proxy dispatch on 'pow' from the Object PMC.
diff: www.parrotvm.org/svn/parrot/revision?rev=31320
16:58 ruoso joined 17:09 szabgab joined
szabgab good evening 17:11
jonathan hi szabgab 17:12
szabgab I am trying to add parrot related syntax highlighting to Padre
I thought I'll try to add highlighting for pasm 17:13
jonathan PIR would be much more relevant/useful (and not vastly different).
Would be a nice thing, though. :-)
See the editors directory in Parrot - may provide some hints. 17:14
szabgab I'll try to look at PIR then
how can I parse a PIR file from perl5 and get know which part is what kind of a token? 17:15
btw the first version of my PASM support was just committed to svn of Padre 17:16
need to go, back in a few minutes...
jonathan Hmm. I'm not sure we've got a Perl 5 parser - the parsers that exist are in C (IMCC and PIRC) and Perl 6 Rules (languages/pge/, IIRC)
pmichaud languages/PIR, maybe. 17:19
jonathan oh, yes! 17:20
Duh!
szabgab back 17:25
so all I need to do is sg like what I have here : svn.perlide.org/padre/trunk/lib/Padre/Document/ 17:26
jonathan $REGISTER => qr/I0|N\\d+/, 17:30
Making that
$REGISTER => qr/I0|N\\d+/,
oops
$REGISTER => qr/\\$?[ISPN]\\d+/, 17:31
Is probably more correct.
Also, whitespace is allowed before labels, at least in PIR.
So $LABEL => qr/^\\w*:/m,
dalek r31321 | rurban++ | pdd30_install:
: update STATUS-branch after rename
diff: www.parrotvm.org/svn/parrot/revision?rev=31321
jonathan Likely wants to be $LABEL => qr/^\\s*\\w*:/m,
17:32 rurban_away joined
szabgab thanks, I fix those thought this is really a perliminary version and I hope some people who are better equipted to ad a parser of pasm and pir will do this ;-) 17:34
I just show how the coloring interface in Padre works
jonathan Aha, cool. 17:41
I'm about to head off on a trip to Russia, so cna't help too much myself at the moment.
szabgab :-(
jonathan Maybe when I get back. :-) 17:42
szabgab in pir what is the difference between end and .end ?
and print and .include?
are they all keywords?
jonathan One interesting thing though, is whether we can use the PGE parse trees or even what they boil down to in PAST to do highlighting for many languages, more easily... 17:43
end is an opcode
Whiteknight things with a . are directives, and may translate to multiple PASM instructions
jonathan .end isn't an instruction itself.
szabgab so I can call .end as a directive?
jonathan Yes, or maybe no instructions. Like .sub 'name' ... .end marks the instructions that make up a sub of a given name.
szabgab for now?
purl For now, it works but once I figure out how to get the lube to stick to the mouse I'll be able to market it.
Whiteknight which is actually inconsistent with the "say" command, which is also a directives
jonathan Yes.
Oh? 17:44
szabgab so ^\\.\\w is a directive?
just for the colouring
jonathan I'm struggling to think of a better name.
szabgab so it is good for now
jonathan But for me, works OK. It distinguishes it from being an instruction.
szabgab It is more so I can brag that Padre supports pir :-)
jonathan :-) 17:45
So, when do we get Perl 6 support? ;-)
szabgab but I also hope that one of you will pick it up and add real parsing to it and real calltips
it is in there already :-)
moritz really?
szabgab just very preliminary ....
moritz has to try
jonathan OH RLY?
jonathan too 17:46
...wow, you just might have got two new users. ;-)
Whiteknight we can has syntax highlighting?
szabgab it is just a few keywords now
jonathan szabgab: This is the bit where you learn that half of #parrot speak like lolcats...
Whiteknight whatever, better then nothing
szabgab that's another thing I wanted to ask, how can I parse perl6 from perl5?
moritz szabgab: STD.pm already can ;)
szabgab: it can be compiled to perl5, and then run 17:47
szabgab is that perl5?
moritz but dead slow
jonathan Yes, there's that. Another thing I'd really, really like to see tried, is embedding Parrot and using PGE itself to do the parsing.
szabgab jonathan: right, that would be great
Whiteknight I asked my girlfriend "I can has chocolate" and she was like "You can has chocolate later!". I'mrubbing off on her
szabgab so if there is an Inline::Parrot ..
jonathan szabgab: I'm not promising it'll be easy. ;-) 17:48
szabgab so where is STD.pm ?
Whiteknight An Inline::Parrot module would be nice to have for Perl 5
moritz szabgab: pugs repository, src/perl6/
jonathan Inline::Rakudo could be cool too. 17:49
I guess that could be a thin layer on top of a more general Inline::Parrot, perhaps.
szabgab I think I'll release Padre 0.10 with the support as it is now and then see how to add better support 17:50
jonathan Sure. Step by step. :-) 17:51
jonathan goes to cook dinner, learn more Russian and continue packing
szabgab hey std.pm is Perl 6 17:52
I cant run that yet ...
moritz szabgab: but it can be compiled to perl 5
szabgab: just type 'make'
szabgab: and you need some dependencies... like perl-5.10.0 in /usr/local/bin/perl, and re::engine::TRE
szabgab in the src/perl6 dir, it did notlike me 17:53
ah I see
that will take some time to build
so for now I skip this but I'll come back for more help with this
in the meantime if anyone wants to try Padre, it is on CPAN, and help to install is on padre.perlide.org/wiki/Download 17:56
the pir/pasm/perl6 colouring is in svn right now
Whiteknight Padre requiures Wx, right? 17:58
moritz right
Whiteknight is that going to be a big pain in the ass to install on ActivePerl? 17:59
szabgab yes, that's the biggest issue with it
Whiteknight (gtk2 was, so I've become leery of new graphics kits)
szabgab no, on ActivePerl windows you have instructions on the web site
Whiteknight Tk was easy to install, but Tk looks terrible
szabgab there are ppd files in a separated repo
Whiteknight oh, maybe I'll go hunting for them 18:00
szabgab see the padre Download page for instructions
Whiteknight ok
szabgab but I think Strawberry is better :-)
Whiteknight This computer is linux, so that isn't a problem. my other computers are all windows though
szabgab actually it is a bit harder to install on linux right now 18:01
we already have instructions for windows both AP and Strawberry but the linux instructions are a bit harder / less existant
Whiteknight oh really? So I can't just do "install padre"? 18:02
szabgab if you have Wx installed you can 18:03
Whiteknight ok
szabgab the problem is with the Wx installation
Whiteknight oh, that installation fails?
szabgab but for linus I have a binary you can download 18:04
18:04 Zaba joined
moritz is that PAR packed? 18:04
szabgab Whiteknight: there are all kinds of magic and spells you need to use to get Wx running on linux
yes par packed but exe
Whiteknight i'll play with it a little bit. I know a little magic 18:05
szabgab and it is for 0.09 without the pasm/pir thing
and there is the wxperl mailing list and #wxperl 18:06
but I have to go now
&
18:07 rurban_away joined
Whiteknight wx *appears* to be building fine on my system 18:08
purl Try a 4-letter station code (see weather.noaa.gov/weather/curcond.html for locations and codes)
rurban Whiteknight: when do you add gc_it to the new pdd09gc branch? 18:09
Whiteknight I dont know, probably soon
I need to add a few new support functions first to support it 18:10
I'll probably work on that this week a little bit, if there are no more pdd27mmd tasks for me to work on
rurban ah, pdd27. that's why 18:11
masak rakudo: say eval("\\$foo".perl)
polyglotbot OUTPUT[Null PMC access in get_string()ā¤current instr.: 'print' pc 14071 (src/gen_builtins.pir:8773)ā¤called from Sub 'say' pc 14093 (src/gen_builtins.pir:8783)ā¤called from Sub '_block11' pc 44 (EVAL_12:15)ā¤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 806 (src/PCT/HLLCompiler.pir:481)ā¤called
..from Sub 'parrot;PCT::HLLCompiler;evalfiles' pc 108...
moritz masak: rakudo doesn't do lexicals in eval()s yet 18:19
masak I can tell. 18:21
but note -- somewhat subtle, I know -- that I didn't evaluate the program $foo
I evaluated the string "$foo"
moritz that tries to access the lexical $foo as well 18:22
masak yes, due to the non-escaping of $
if Rakudo implemented Perl 6 a bit more, it wouldn't
moritz right 18:23
.oO( we're working on it ;)
masak aye
didn't mean to sound harsh :)
masak hugs Rakudo
moritz: where did this [Book] mail in p6c drop down from? 18:25
moritz masak: I just copied & pasted the change message and some parts of the diff in 18:26
Whiteknight that [Book] mail was about a change I made to pugs/docs/tutorial
masak aha, so it's not from some other list. I see
Whiteknight i added some details which might not be true, depending on how well I've read some information
moritz and I thought I'd cc p6c in case somebody knew the answers
Whiteknight I didn't cc my reply to p6c though 18:27
masak moritz: FWIW, I agree with you. inner classes are not automatically subclasses
Whiteknight i probably should have
moritz Whiteknight: also the matching against a grammar will be done differently
Whiteknight: stuff ~~ Grammar ist just a type check
masak in fact, inner classes in Perl 6 are quite pointless as I understand them.
Whiteknight it will be? Calling $string ~~ Grammar doesnt call Grammar.TOP?
moritz Whiteknight: no. I'll look for the discussion in the #perl6 logs, just a sec 18:28
Whiteknight i read that somewhere, but my source may have been very old
that's also the way it's working in PGE now, so you can understand my confusion
moritz it used to work that way, but it's a conflict with how other stuff works in Perl 6 18:29
Whiteknight: irclog.perlgeek.de/parrot/2008-05-31#i_322490
18:53 ruoso joined 18:58 ruoso joined 19:06 ruoso joined 19:09 Zaba joined 19:23 Maddingu1 joined, masak joined, baest joined 19:24 Whiteknight joined 19:25 ruoso joined 19:29 ruoso joined 19:44 leo_ joined 19:47 jonathan_ joined 19:50 ruoso joined 19:51 Zaba_ joined 19:52 jan joined 19:53 slavorg joined 19:54 PacoLinux joined, baest joined 19:57 Zaba joined 19:58 Ontolog joined 20:01 ruoso joined 20:06 grim_fandango joined 20:08 nopaste joined 20:14 ruoso joined 20:32 Theory joined 20:35 Zaba joined
dalek r31322 | fperrad++ | trunk: 20:37
: [Markdown]
: - refactor: now use only token (no rule)
diff: www.parrotvm.org/svn/parrot/revision?rev=31322
20:39 Zaba_ joined 20:42 Andy joined
dalek r31323 | fperrad++ | trunk: 20:45
: [Lua]
: - fix exception
diff: www.parrotvm.org/svn/parrot/revision?rev=31323
20:54 Ontolog joined 20:56 Zaba joined 20:58 nopaste joined
dalek r31324 | julianalbo++ | trunk: 21:01
: add delete_keyed_int to resizableintegerarray
diff: www.parrotvm.org/svn/parrot/revision?rev=31324
21:02 Zaba_ joined 21:09 ruoso joined 21:28 nopaste joined 21:29 Zaba joined 21:31 grim_fandango joined
dalek r31325 | julianalbo++ | trunk: 21:50
: fix C90 violations in core.ops
diff: www.parrotvm.org/svn/parrot/revision?rev=31325
21:57 Zaba_ joined 22:09 apeiron joined
dalek r31326 | julianalbo++ | trunk: 22:13
: many fixes and improvements to pirric
diff: www.parrotvm.org/svn/parrot/revision?rev=31326
22:13 nopaste joined 22:32 dmknopp joined 22:43 apeiron joined
dalek r31327 | coke++ | pdd30_install: 22:57
: [docs] add pointer to old RT for this item.
diff: www.parrotvm.org/svn/parrot/revision?rev=31327
cognominal pmichaud, can you clarify what you said here two days ago about .item and .list in the context of slurp() 23:02
23:04 baest joined
cognominal cocoacast++ 23:16
23:16 Whiteknight joined
pmichaud cognominal: I'm not sure I can clarify it fully, no. 23:22
Whiteknight what's the easiest way to delete a branch?
pmichaud I just know that 'want' is generally an indication that we're not thinking about things correctly.
Tene Whiteknight: svn rm whatever/foo/bar 23:23
pmichaud just be sure to get the url correct :-)
Whiteknight okay, I'll try really hard
pmichaud we've had it occur where someone accidentally deleted svn.perl.org/parrot/branches
(OOPS!)
Whiteknight yeah, that sounds bad
NotFound pmichaud: except when someone say "I want you", I suppose.
Tene svn rm svn.perl.org/parrot/trunk
oops, wrong window 23:24
purl sudo rm -rf /
Tene I need to figure out this PMC_pmc_val, PMC_data, etc. stuff. 23:25
Whiteknight I completely garbaged up the pdd09gc branch somehow, probably in a merge somewhere 23:26
Tene I know what that's like. :)
Whiteknight so I'm going to just delete it because there haven't been many changes yet and I'm not ready to work on it full-time yet
Tene I've done that more times than I'm comfortable admitting.
Whiteknight yeah, svn seems to make it very easy to screw things up
jonathan has a slight fear of branches, so bullies other people into maintaining his ;-) 23:28
23:28 AndyA joined
Tene I can do branches just fine, as long as I get to deal with them through git. I always harass other people into doing svn stuff for me. 23:29
cognominal pmichaud, do you mean that we would want to multi a sub on context? because my initial motivation was to fix the slurp sub 23:30
I don't think the type is part of the signature, but somehow the context should be. btw I don't think that TimToady speaks of a hash contect. 23:32
I meant "the return type"
Whiteknight ...okay, I think I deleted the branch without causing any major problems 23:33
Whiteknight crosses his fingers
dalek r31328 | Whiteknight++ | pdd09gc:
: Branch is old, I haven't been focusing on it like I should have, and I screwed it up in a merge somewhere so I'm deleting it
diff: www.parrotvm.org/svn/parrot/revision?rev=31328
23:36 baest joined
cognominal reread the synopses part that deal with context 23:38
23:47 TiMBuS joined
Whiteknight where is the pdd27 tasklist at? 23:54
i think it's been lost somewhere in the move to the new site 23:58

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!