Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2006-03-17

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:01 mlh left perl6
00:03 theorbtwo Whoops.
00:03 theorbtwo Somehow, I missed the bless there...
00:03 TimToady np
00:04 TimToady communication lag sucks even for people who think and type perfectly...
00:05 TimToady so those of us who think and type imperfectly are doubly in trouble...
00:10 * ruoso will read the logs later... too much work...
00:12 Ara5n so, does anyone know if: perl -MPugs::Compiler::Rule -e '$rule = Pugs::Compiler::Rule->compile("<word> <ws> <word>"); $rule->match("foo bar")'
00:13 Ara5n should work? or am I doing something mindbuggeringly stupid? :/
00:13 Ara5n (also, where did the <word> rule come from? it doesn't seem to be in S05...)
00:14 GabrielVieira whew uv found <word> ?
00:15 Ara5n first line on Pugs/Grammar/Rule/Rule.pm?
00:15 Quell joined perl6
00:16 GabrielVieira ill look
00:16 GabrielVieira :)
00:17 Ara5n either way, it blows up with a 'Error in evaluation: Undefined subroutine &Pugs::Emitter::Rule::Perl5::concat called at /usr/local/share/perl/5.8.7/Pugs/Runtime/Rule2.pm line 89.'
00:17 * Ara5n goes digging
00:17 GabrielVieira humm
00:24 huhlig left perl6
00:26 wolverian joined perl6
00:26 nothingmuch do we want uniqueness type inference to analyze when data will not need to be garbage collected?
00:33 hcarty joined perl6
00:35 Ara5n what's the relationship between perl5/Pugs-Compiler-Rule and misc/pX/Common/Pugs-Compiler-Rule?
00:35 Ara5n (and whatever's sitting on cpan? :/)
00:36 * Ara5n thinks he's got stuck slap in the middle of some crazy refactoring
00:38 hcarty left perl6
00:38 dvtoo joined perl6
00:44 dvtoo joined perl6
00:45 * Ara5n totally fails to pick apart which version of the libraries's in what state of development & by whom, and gives up when running his oneliner causes perl to hang, seemingly blocked on a pugs waiting on stdin :(
00:46 ayrnieu oneliner?
00:46 Ara5n [00:12] <Ara5n> so, does anyone know if: perl -MPugs::Compiler::Rule -e '$rule = Pugs::Compiler::Rule->compile("<word> <ws> <word>"); $rule->match("foo bar")'
00:46 Ara5n as it doesn't with the P::C::R on CPAN - or the one in perl5/P-C-R
00:47 ayrnieu ah.  You might leave a report at rt.cpan.org ;
00:49 Ara5n hmok
00:50 Ara5n seems that the Pugs::Runtime::Rule2 on CPAN expects to be able to find Pugs::Emitter::Rule::Perl5::concat, whereas in pX/Common it's now in Pugs::Runtime::Rule::concat
00:52 * Ara5n fixes the cpan version up manually
00:52 ayrnieu that also works.
00:53 ayrnieu um, seems possible again that the latest version CPAN sees is out of date, if it depends on pugs.
00:53 Ara5n well, i don't understand to what extent it does depend on pugs
00:54 Ara5n i thought half the idea was that the modules uploaded to CPAN ran natively on p5 using the pmc magic
00:54 ayrnieu I suppose it depends on Pugs::
00:54 Ara5n well, i guess i might have a collision between Pugs:: installed by make install'ing pugs
00:54 Ara5n and cpan installing P-C-R...
00:56 Ara5n either way, fixing the bogus classpath in Pugs::Runtime::Rule2 solves things
00:56 Ara5n s/classpath/location of the concat() sub/
00:57 chouse joined perl6
01:02 stevan_ joined perl6
01:03 Ara5n looks like fglock fixed it in r9605, anyway
01:08 * Ara5n wonders why svn is checking half of these files out with dos line endings :/
01:11 Juerd joined perl6
01:19 Ara5n do pugs hackers tend to set up auto-props for svn:eol-style=native, or fix files which have been committed with the wrong eol terminator by setting the property in the repo?
01:20 Ara5n hm, judging by the number of svn:eol-style's set in the repo...
01:28 svnbot6 r9627 | arathorn++ | setting svn:eol-style as native to fix line-endings on P-C-R and chums
01:29 Ara5n hm
01:29 * Ara5n hopes he didn't !"$% that up
01:30 Frances_ joined perl6
01:31 Juerd Good night, everyone
01:33 Ara5n um, http://rt.openfoundry.org/Foundry/Project/Source​/index.html/pugs/Common/Pugs-Compiler-Rule/TODO doesn't look good
01:33 * Ara5n looks vaguely guilty
01:33 Ara5n night
01:33 Ara5n oh, bogus url
01:35 amnesiac joined perl6
01:37 * Ara5n has a horrible feeling that svn shouldn't recommit the whole file when you set svn:eol-style :/
01:39 FurnaceBoy urk, line ending changes always suck
01:39 Ara5n yeah, that went totally awry :(
01:39 Ara5n fglock: sorry, wherever you are :/
01:39 FurnaceBoy I'm always having to fix and commit, and then cross-revision diffs just won't ever work :~|
01:40 FurnaceBoy so I standardise on UNIX breaks in my own repos
01:40 * Ara5n assumed that svn was smart enough to let you change it for a file which has already been commited incorrectly
01:40 FurnaceBoy heh, I never tried that, actually
01:41 Ara5n and perhaps it is, if you re-checkout after setting svn:eol-style before committing
01:41 Ara5n meh, anyway
01:41 Ara5n what a particularly inglorious commit.
01:41 FurnaceBoy I feel yr pain.
01:41 FurnaceBoy well put. An inglorious commit. at least you can express your agony poetically.
01:42 Ara5n hrm ;)
01:45 Ara5n gah, i blatantly should have just set the property remotely rather than committing at all
01:45 Ara5n what a mess
01:54 FurnaceBoy ah.
01:54 FurnaceBoy you've besmirched an otherwise flawless repo. You animal.
01:55 vel joined perl6
01:57 * Ara5n mails fglock to apologise and crashes for the night
01:57 FurnaceBoy nite
01:57 Ara5n night
02:02 omniuni joined perl6
02:02 omniuni left perl6
02:09 Nouk joined perl6
02:26 justatheory stevan: (Delayed reaction) No, your interpretation sounds right.
02:57 stevan_ justatheory: (also delayed reaction) cool :)
02:57 justatheory Heh, not as delayed as mine!
02:57 stevan_ justatheory: :)
02:58 justatheory 'Course, had there been a Class::MOP when I wanted my introspection API, I might not have bothered with writing Class::Meta.
02:58 stevan_ we set up #moose over at irc.perl.org too, for Class::MOP and Moose related questions/stuff
02:58 justatheory But now, of course, I'm quite fond of it.
02:58 justatheory Cool.
02:58 stevan_ yeah I looked over Class::Meta a few times while I was working on the Perl 6 metamodel
02:58 justatheory Cool!
02:58 justatheory Do I get any credit? :->
02:59 stevan_ you got a shout-out in the SEE ALSO section :)
02:59 hcarty joined perl6
02:59 justatheory W00t!
03:00 justatheory Seems like, what with both of our involvement in meta classes in Perl and JSAN that we have similar interests.
03:01 hcarty left perl6
03:02 stevan_ justatheory: very true, although my JSAN work is fairly limited compared to yours
03:02 stevan_ I gave up doing a lot of Javascript a few years back
03:02 justatheory stevan_: Not so. You sicced Rob onto it.
03:02 stevan_ LOL
03:02 stevan_ actually it was the other way around, he got into it first :)
03:03 justatheory heh
03:03 justatheory But then you supported that, yes?
03:03 stevan_ yup
03:03 justatheory Good man. Hopefully he can resurrect it before OSCON.
03:03 rafl Hello guys
03:03 justatheory whaddup?
03:04 rafl Nothing much. Parrot 0.4.2 is in Debian sid by tomorrow.
03:22 KingDiamond joined perl6
03:48 weinig|away is now known as weinig
04:58 Aankhen`` joined perl6
05:43 szabgab left perl6
06:03 LumberCartel joined perl6
06:04 LumberCartel left perl6
06:20 KingDiamond joined perl6
06:23 justatheory joined perl6
06:47 Quell joined perl6
06:47 GeJ joined perl6
06:48 GeJ Good morning folks
07:07 kanru joined perl6
07:11 iblechbot joined perl6
07:15 KingDiamond joined perl6
07:16 marmic joined perl6
07:18 fandango joined perl6
07:34 Nouk joined perl6
07:46 Quell joined perl6
08:14 sl01 joined perl6
08:35 kane_ joined perl6
08:46 Draconit_ joined perl6
08:49 Frances__ joined perl6
09:02 ghenry joined perl6
09:04 Draconit_ left perl6
09:14 Draconit_ joined perl6
09:14 Draconit_ left perl6
09:54 drbean joined perl6
10:13 jisom joined perl6
10:20 iblechbot joined perl6
10:24 KingDiamond joined perl6
10:25 jisom_ joined perl6
10:34 Draconit_ joined perl6
10:36 Draconit_ left perl6
10:40 jisom_ is now known as jisom
10:45 marmic_ joined perl6
10:48 _cnhackT_ joined perl6
10:49 * Ara4n investigates how to revert his bogus line-ending commit in pX from last night :(
10:50 Khisanth joined perl6
10:52 Juerd rafl: Please reenable automatic pugs building on feather
10:55 leo_ joined perl6
10:56 leo_ is now known as LeTo
11:18 rindolf joined perl6
11:18 rindolf Hi all!
11:18 Arathorn morning
11:20 pasteling "rindolf" at 88.153.140.201 pasted "graham.p6 - latest version" (250 lines, 7.4K) at http://sial.org/pbot/16365
11:20 rindolf Arathorn: hi! What's up?
11:20 rindolf audreyt: here?
11:20 Arathorn rindolf: trying to revert a bogus line-ending 'fix' in the pX branch of the repo
11:21 rindolf Arathorn: OK.
11:21 * Arathorn didn't realise that svn ps svn:eol-style native pX/whatever; svn ci; # would go and totally mangle the line endings in the repo and screw up the files' history :(
11:22 Arathorn and it doesn't work if you s/\r//g; on the files before committing them (after setting the property)
11:22 Arathorn any suggestions from anyone would be
11:22 Arathorn muchly appreciated :)
11:22 jisom edit the property, then change the line endings, then reset the property?
11:23 * Arathorn tries that
11:23 Arathorn (not on the live repo ;)
11:27 ghenry joined perl6
11:27 Arathorn jisom: whilst deleting the property doesn't alter the line endings in the repo (go figure)
11:27 elmex joined perl6
11:27 Arathorn the fact of committing it (with the working version using either \n or \r\n) totally mangles the history by rewriting all the lineendings serverside
11:27 Arathorn which seems ridiculous :(
11:28 jisom weird....
11:28 Arathorn and you can't set properties on a remote repo, apparently
11:28 * Arathorn sighs
11:36 renato joined perl6
11:39 Arathorn turns out this is a mis(sing)-feature of svn
11:39 Arathorn hoorah
11:47 Arathorn oh bollocks
11:47 Arathorn i wonder if this is going to totally mess up Module::Install's checksumming magic
11:48 Arathorn as .pmc's will get checked out on different platforms with different line-endings (with svn:eol-style native)
11:48 Arathorn causing their 32-bit checksums to obviously differ
11:48 Arathorn meh
11:51 kanru joined perl6
12:04 chris2 joined perl6
12:04 TimToady joined perl6
12:06 nnunley joined perl6
12:31 ruoso joined perl6
12:35 * ruoso finally read all the backlogs and understand that Perl 6 will have even more voodoo than Perl 5
12:35 lypanov ruoso: but less of the real crap :)
12:37 arcady but at least it will be new voodoo, there for reasons we mostly still understand
12:39 ruoso :)
12:41 theorbtwo ...at least you understand it if you've read all of #perl6's backlogs and all of perl6-language, and have TimToady handy to ask about the reasons that never left his head.
12:42 ruoso theorbtwo+++ heh
12:43 lypanov ;)
12:49 * ruoso wonders if it makes sense that lrep someday emit parrot assembler...
12:49 arcady so long as it doesn't emit actual parrots...
12:50 * ruoso thinks lrep could be "ported" to parrot as soon as it is completely self-hosted and parrot has the needed features...
12:51 arcady what exactly is lrep?
12:52 ruoso arcady, lrep is lrep.p6 compiled to perl 5. that compiles lrep.p6 to lrep
12:52 ruoso arcady, http://svn.openfoundry.org/pugs​/misc/pX/Common/lrep-compiler/
12:53 ruoso audreyt, do you think it makes sense?
12:57 * Arathorn still doesn't understand why lrep-compiler doesn't use P::C::R for its engine
12:57 Arathorn s/engine/runtime/
12:57 Arathorn or is it just waiting to be updated?
12:57 ruoso Arathorn, it will... fglock is preparing the migration
12:57 Arathorn ah, right
12:57 ruoso Arathorn, but if you want to step ahead, feel free :)
12:58 ruoso I'm still working to self-host lrep
12:58 ruoso as there's still much perl 5 code
12:58 arcady the first job of any good compiler...
12:59 ruoso arcady, yes...
12:59 ruoso arcady, any help is welcome ;)
12:59 ruoso arcady, it's just a matter of creating rules, emitter entries and porting code
12:59 arcady I actually have a week or so of mostly free time
13:00 arcady and I've been waiting years to be able to write perl 6 code...
13:00 ruoso arcady, just take a look at lrep...
13:00 ruoso arcady, and see if you understands...
13:00 ruoso arcady, then just help
13:00 ruoso :)
13:04 * ruoso wonders why svn doesn't print the '?' lines when updating, so I can easily know which files are not in the repository
13:05 arcady probably they consider it to be a feature
13:05 arcady if you have a lot of .o files and such lying around
13:05 arcady then the ? lines are just confusing
13:06 ruoso arcady, that's what .cvsignore is for
13:06 arcady this is true
13:08 jisom perhaps it first queries the server for which files need to be updated between revisions, so it doesn't need to scan your local svn repo until it checks for conflicts
13:10 jisom anyone know how to get sin(pi) to work?
13:10 svnbot6 r9628 | ingy++ |  r5733@bytes:  ingy | 2006-03-17 05:08:51 -0800
13:10 svnbot6 r9628 | ingy++ |  Change all '=over 4' to '=over'. Damian, I think we should remove the indent
13:10 svnbot6 r9628 | ingy++ |  number from the Perldoc model.
13:10 wolverian ingy++ # agreed
13:10 ruoso ingy++
13:13 svnbot6 r9629 | ruoso++ | Better explained TODO list
13:13 ruoso arcady, take a look at the new TODO I made... it's a good starting point
13:15 arcady I have a flight to catch in a few hours, I'll do this on the plane
13:23 fglock joined perl6
13:23 ruoso fglock, hi
13:24 ruoso fglock, I was waiting for you :)
13:27 fglock ruoso: hi!
13:28 ruoso fglock, how is everything... should we start the migration?
13:29 fglock ruoso: you mean make lrep use PCR?
13:30 ruoso fglock, yes
13:33 fglock ruoso: PCR still have a few problems - a few syntax changes, untested features, unstable API...  - can you live with that? :)
13:33 ruoso fglock, I'm in the point to need rules as methods and grammars as classes
13:33 ruoso fglock, almost in the point
13:34 ruoso actually
13:34 Arathorn fglock: 'pologies again for screwing around with lineendings in misc/pX...
13:34 bsb joined perl6
13:35 fglock Arathorn: np!
13:35 ruoso fglock, I still need to make rule closures to be real perl 6 code and not filtered perl 5 code
13:36 fglock ruoso: can you read PCR TODO and see what features you would need first?
13:36 fglock ruoso: per5/PCR
13:36 ruoso in px/Common?
13:36 fglock perl5/PCR
13:36 ruoso ok
13:37 fglock px/Common/PCR is for testing new features - it is broken most of the time
13:45 ruoso fglock, well... it seems there's no missing features...
13:45 ruoso fglock, the things missing in PCR are missing in lrep too
13:46 ruoso fglock, no big deal
13:46 * theorbtwo is getting to be somewhat confused about lrep -- is it supposed to be usable?  The only thing I can get out of it is "Undefined subroutine &grammar1::immediate_statement_exec called at iterator_engine.pl line 184, <GEN0> chunk 1.
13:46 fglock ruoso: px/Common/PCR fixes some architecture problems (OO, namespaces), but introduces new bugs which are hard to fix - so they will need to be re-implemented in perl5/PCR
13:46 Limbic_Region joined perl6
13:47 ruoso theorbtwo, look at lrep-compiler subtree
13:48 ruoso fglock, were are this bugs listed?
13:50 fglock ruoso: just run the tests - right now, there is an error in the emitter (some commas are missing in the emitted code)
13:50 ruoso fglock, ok
13:51 ruoso inc/Module/Install.pm... is there a reason to use some specific version:
13:51 ruoso ?
13:51 fglock ruoso: can you help with PCR? it has most of the OO infrastructure in place, a single engine (perl5/PCR needs two)
13:51 ruoso fglock, I'll try...
13:52 fglock ruoso: M::I is audreyt's new version, which creates .pmc files
13:52 ruoso fglock, ah ok
13:52 * ruoso has only Common/ checked out
13:53 fglock ruoso: you will need a version of lrep specific for the OO engine - I'll commit it in a while
13:54 ruoso fglock, ok...
13:57 ruoso Failed 4/4 test scripts... ouch
13:57 ruoso Checksum failed for outdated .pmc file
13:57 ruoso oops
13:57 * ruoso inside pX PCR
13:59 fglock ruoso: it needs the updated lrep, in order to compile the grammar - I'm trying to commit it (the version I have was copied from the old directory, and contains invalid .svn info - I'm trying to fix it)
13:59 ruoso Failed 3/4 test scripts... looks better now :)
14:00 Qiang joined perl6
14:01 * ruoso doesn't know what do do...
14:02 fglock ruoso: help with tar - how do I exclude .svn files? I tried --exclude=.svn
14:03 ruoso --excluide=PATTERN
14:04 ruoso it should work
14:04 * fglock wonders what a PATTERN is like
14:05 ruoso fglock, is it not excluding anything? or excluding something more?
14:05 fglock ruoso: it is including the .svn dirs
14:05 ruoso fglock, only them?
14:06 fglock svn will not add the files until I clear up the old .svn info
14:06 ruoso find . -type d -name '.svn' -exec rm -rf {} \;
14:09 fglock ruoso: one sec - I'm moving the files to linux - it will be much easier :)
14:09 ruoso tar czvf teste.tar.gz `find . | grep -v \.svn`
14:11 robkinyon joined perl6
14:18 fglock ruoso: I'm testing my local version and then I'll commit
14:18 * ruoso ansioso
14:21 robkinyon left perl6
14:24 fglock ruoso: sorry - I'm doing a few merges
14:24 svnbot6 r9630 | fglock++ | pX/Common/Pugs-Compiler-Rule - updated grammar file
14:26 ruoso looks worse now :)
14:26 svnbot6 r9631 | fglock++ | pX/../PCR - updated Grammar/Rule.pm
14:27 ruoso looks better again :)
14:28 justatheory joined perl6
14:29 ruoso fglock, is it everything? is there more coming?
14:30 fglock ruoso: I'm trying to commit the new lrep, which compiles grammars to the new engine
14:30 ruoso ok
14:33 fglock ruoso: still trying...
14:34 * ruoso zzzz
14:34 ruoso :)
14:34 fglock "svn: At least one property change failed; repository is unchanged" - it looks like another line-endings problem
14:34 ruoso fglock, gah
14:35 ruoso find . -type f -exec perl -pi -e 's/\r\l/\n/gs' {} \;
14:36 fglock ruoso: thanks - now I'm waiting for svn to timeout :(
14:36 jisom svn's atomic
14:36 jisom can just ^C
14:36 jisom it won't really commit until the full commit's made
14:36 jisom at least that's part of the point of svn
14:37 * ruoso still prefers CVS
14:37 * jisom still prefers svn
14:37 fglock jisom: I was waiting for it to accept ^C
14:38 jisom oh, *that* timeout
14:38 ruoso ^\ should solve
14:38 jisom ^Z kill %1
14:38 jisom sometimes that works better
14:38 jisom for some reason
14:39 ruoso jisom, ^C is kill INT, kill by default is kill TERM
14:39 jisom eh, if it works, it works :)
14:39 fglock done!
14:39 svnbot6 r9632 | fglock++ | lrep again
14:40 ruoso hmmm... looks worse :)
14:40 fglock ruoso: now you can see where I stopped :)
14:41 fglock ruoso: it is missing commas after each "ruleop::concat(...)" - I'm trying to find out why
14:41 * ruoso wonders why fglock took the old lrep instead of the new...
14:42 fglock ruoso: I'm working on this for a week already - the new lrep wasn't around
14:42 ruoso fglock, ah ok
14:43 fglock ruoso: it may be best to just restart implementing OO with a fresh copy of PCR + lrep-compiler
14:43 ruoso Can't locate Pugs/Runtime/Rule.pm
14:43 ruoso weird
14:44 ruoso fglock, could you summarize the changes?
14:46 fglock ruoso: the main change is that rules are methods. but I did some other things at the same time, and this was a mistake
14:47 * ruoso browsing the log
14:48 fglock ruoso: like: there is a single engine, and a lot of redundant code was removed
14:48 ruoso engine == ruleop::*, right?
14:48 fglock ruoso: yes - it is now called Pugs::Runtime::Rule
14:49 chris2 joined perl6
14:51 ruoso fglock, ok... comparing P::R::R with lrep-compiler R::P5::RuleOps there are small differences
14:52 ruoso the diff has 65 lines
14:53 rindolf joined perl6
14:53 pasteling "ruoso" at 201.29.209.125 pasted "ruleops diffs" (99 lines, 3.6K) at http://sial.org/pbot/16368
14:54 fglock ruoso: the Makefile.PL error has "ruleop::concat(" - but I can't find where 'ruleop' is in the source code
14:55 ruoso fglock, a lot of places
14:55 ruoso lib/Pugs/Runtime/Rule.p6
14:55 ruoso lib/Pugs/Grammar/Rule/Rule.pmc
14:56 ruoso t/01-iterator_engine.t
14:56 ruoso t6/iterator_engine.p6
14:56 rindolf audreyt: here?
14:57 ingy rindolf: she's there.
14:57 fglock ruoso: I think the problem lies in PCR/lrep - there shouldn't be a 'ruleop' in PGR::Rule.pmc
14:57 ruoso fglock, I think so
14:58 ruoso fglock, what about looking for the differences between the new code and the code in lrep-compiler
14:58 ruoso fglock, I didn't change the engine
14:58 fglock ruoso: because it should generate code for the new engine, which is in the 'PRR' namespace
14:59 ruoso we could incorporate the changes to lrep-compiler
14:59 ruoso and then get it back to P::C::R
14:59 ruoso and make lrep-compiler use P::C::R
14:59 fglock ruoso: ok - give me a few minutes...
15:01 ruoso fglock, where is the code that used to be in p6rule_lib?
15:02 fglock ruoso: lib/Pugs/Grammar/Rule.pm - it is much smaller, as things were moved into the grammar
15:02 diotalevi joined perl6
15:04 ruoso fglock, no... that was p6rule.pm
15:04 ruoso fglock, i mean p6rule_lib.pm
15:05 * ruoso is lost
15:06 fglock ruoso: p6rule_lib.pm is now merged into lib/Pugs/Grammar/Rule/Rule.pm, which is the Grammar file
15:06 ruoso fglock, ok
15:09 * ruoso needs to think better... maybe the grammar refactoring is needed now
15:09 svnbot6 r9633 | ingy++ |  r5735@bytes:  ingy | 2006-03-17 07:06:39 -0800
15:09 svnbot6 r9633 | ingy++ |  lowercase the `=doc <dialect>` dialects
15:10 * ruoso almost understanding...
15:10 ruoso ok....
15:11 ruoso first: Some grammars that are in lrep/Grammar::Perl6Init.pm were moved to the standard Grammar::Perl6
15:11 ruoso s/grammars/rules/
15:14 fglock ruoso: re emitter changes: all rules need a '$grammar=shift;' and calls to subrules are now 'sub{$grammar->subrule(@_)}'
15:14 fglock ruoso: the return-block processing is done by 'rule_wrapper()'
15:18 fglock ruoso: I'll commit my local Rule.pmc, as the Grammar compiler is not working...
15:19 * ruoso importing the implemented rules to Grammar::Perl6.p6
15:19 svnbot6 r9634 | fglock++ | pX/.../PCR/.../Rule.pmc - precompiled version
15:19 fglock ruoso: this is what the new lrep would compile the Grammar into
15:20 GabrielVieira joined perl6
15:20 svnbot6 r9635 | ingy++ |  r5742@bytes:  ingy | 2006-03-17 07:19:38 -0800
15:20 svnbot6 r9635 | ingy++ |  Question for TheDamian regarding `=pod` and `=kwid`. Yagni.
15:21 ruoso fglock, I got a syntax error after importing the rules you implemented...
15:21 ruoso fglock, I will investigate
15:23 ruoso fglock, some ',' missing
15:23 ruoso fglock, in alternate ruleop
15:23 fglock ruoso: I'll have lunch
15:23 * ruoso in lrep-compiler
15:23 * ruoso too
15:24 fglock ruoso: back in ~40min
15:24 ruoso fglock, BTW... I'll try to incorporate the changes to lrep-compiler... and then we get it back to P::C::R and makes lrep-compiler uses it
15:24 fglock ruoso: ok!
15:24 fglock &
15:26 ruoso fglock, I think I got an infinite loop
15:29 nothingmuch joined perl6
15:29 _cnhackT_ joined perl6
15:32 _cnhackT_ is now known as cnhackTNT
15:32 Alias_ joined perl6
15:32 Alias_ Any .il people awake?
15:33 ruoso fglock, the new rules dispatch an infinite loop
15:33 ruoso fglock, I'll leave them commented in Grammar::Perl6.p6
15:33 ruoso if someone want something to do, please take a look at the commented rules in Grammar::Perl6.p6 in lrep-compiler
15:34 merlyn joined perl6
15:34 ruoso hi merlyn
15:34 pmichaud joined perl6
15:34 merlyn hey ruoso
15:34 ruoso merlyn, want some fun?
15:35 merlyn in what sense? :)
15:35 nothingmuch Alias_: i'm awake
15:35 ruoso merlyn, debugging :)
15:35 merlyn Oh.  I generally don't consider that "fun".
15:36 Alias_ nothingmuch: I need an English translation for some hebrew words I got for which there is no German translation
15:36 merlyn I'm here because I need to get smarter about eprl6.
15:36 merlyn and even perl6. :)
15:36 Alias_ nothingmuch: (which I got from a german)
15:36 Alias_ :
15:36 nothingmuch ruoso is appearantly a masochist =)
15:36 nothingmuch Alias_: okies =)
15:36 Alias_ chokmah, binah
15:36 ruoso nothingmuch, :)
15:36 Alias_ something kaballa related perhaps?
15:36 Alias_ maybe chockmah
15:36 nothingmuch chokmah == חוכמה == smartness (distinct from wisdom)
15:36 merlyn But at the moment, I need to drive from the hotel to $worksite... be back later
15:37 nothingmuch more like chochmah
15:37 fglock joined perl6
15:37 nothingmuch or hohma
15:37 Alias_ ok
15:37 kanru is now known as chchihun
15:37 Alias_ nothingmuch: So intelligence then?
15:37 chchihun is now known as kanru
15:37 nothingmuch binah = knowlege, but more elemtary... maybe intelligence
15:37 nothingmuch בינה
15:37 theorbtwo Intersting, I could have sworn chokmah was one of the names of the sepheroph (nodes on the tree of life), generally translated as "wisdom".
15:37 nothingmuch שן בינה is
15:37 nothingmuch the name for wisdom tooth
15:38 nothingmuch so i'm not 100% sure on binah
15:38 nothingmuch but it's closer to intelligence than chochmah is
15:38 nothingmuch chacham is "smart"
15:38 Alias_ So basically they are very subtle specific forms of "smartness" in general
15:38 nothingmuch chochma is the adjective
15:39 cdpruden joined perl6
15:41 svnbot6 r9636 | ruoso++ | There is an infinite loop dispatched by the rules commented in Grammar::Perl6.p6 in lrep-compiler... The current p5 implementation of the rules does not dispatch this loop
15:42 * ruoso needs help with that problem
15:43 * ruoso needs to lunch... and have a meeting this afternoon... so... no more lrep work for today... :(...
15:44 Alias_ nothingmuch: So would one of those two be close to what IQ measures?
15:44 Alias_ nothingmuch: I gather it's basically IQ versus experience/wisdom, but more subtle than both those terms
15:44 nothingmuch Alias_: no
15:44 nothingmuch they are more intagiable, i think
15:44 Alias_ darn
15:45 nothingmuch do you need such a word?
15:45 nothingmuch well, actually
15:45 nothingmuch hmm
15:45 Alias_ Well, the author of the editor I'm basing my refactoring Perl thing on needs a new name
15:45 nothingmuch Gaon is "genius" but it's either IQish or a theological genius
15:45 Alias_ Because the original name is taken
15:45 Alias_ And he's very religious, although the specific one has changed a few times
15:45 Alias_ And he's stuck(ish) for a name
15:46 Alias_ ooo, I like Gaon though
15:46 nothingmuch it's Ga'on, as in too distinct syllables
15:46 Alias_ ok
15:46 nothingmuch i can dish out a few more
15:46 nothingmuch even without a thesuarus
15:46 nothingmuch let me think
15:46 Alias_ :)
15:47 nothingmuch hmm
15:47 nothingmuch i'm stuck
15:47 Alias_ ok :)
15:47 nothingmuch =)
15:47 Alias_ never mind
15:47 nothingmuch wait wait
15:47 eric256_ joined perl6
15:47 Alias_ heh
15:48 * Alias_ waits as nothingmuch runs for the hebrew thesaurus
15:48 nothingmuch i don't have one
15:48 nothingmuch not even an english one
15:48 nothingmuch except for a crappy pocket version of MW
15:48 Juerd There are too many vowels in thesaurus to be hebrew :)
15:48 nothingmuch that's about as thick as my pinky
15:49 Juerd Exactly how thick is your pinky?
15:49 * eric256_ wonders how much of a new POD parser exists and where he might find it. hmmmm
15:49 nothingmuch pretty thin
15:49 * nothingmuch has small hands
15:50 Juerd Do you have a small keyboard and a small mouse too?
15:55 * particle_ hurls http://www2.multithread.co.uk/im​ages/tiny_usb_keyboard_large.jpg at nothingmuch
15:57 nothingmuch joined perl6
15:57 Alias_ nothingmuch: can't prv message. Yes, I do a little
15:58 nothingmuch Alias_: i'll spam you then
15:58 Alias_ ok
15:58 nothingmuch Alias_: cpan.org addy?
15:59 Alias_ adamk
16:00 Juerd particle_: It's incomplete. Sholud be tiny_usb_keyboard_large_cable ;)
16:00 particle_ :)
16:00 Juerd But whoa, that's really tiny.
16:01 Juerd Is it water proof? If so, I want it for my shower radio project ;)
16:02 particle_ Love your CrackBerry(tm) keyboard? Rejoice! Now you can get one for your laptop, too!
16:09 pasteling "rindolf" at 88.153.140.201 pasted "graham.p6 - latest version - can anyone advise how to make it more perl6ixy?" (246 lines, 7.3K) at http://sial.org/pbot/16372
16:10 rindolf nothingmuch: can you read http://sial.org/pbot/16372 and give some advice?
16:10 Juerd rindolf: Can you send private messages on freenode/
16:10 Juerd ?
16:11 rindolf Juerd: I can, yes.
16:11 nnunley Juerd: Only if you've registered your nick.
16:11 Juerd nnunley: That was the implied question :)
16:12 * nnunley grins.
16:12 nnunley Brains.  I'm going home. ;)
16:18 svnbot6 r9637 | ingy++ |  r5747@bytes:  ingy | 2006-03-17 08:16:36 -0800
16:18 svnbot6 r9637 | ingy++ |  Note to Damian about =for
16:20 Juerd Hmmm
16:20 Juerd svn as a messaging system :)
16:20 ingy Juerd++
16:21 ingy to bad Damian doesn't do irc
16:21 Juerd Yea
16:21 ingy s/to/two/
16:21 Juerd Though perhaps PBP is a bit too recent for it to be safe ;)
16:21 particle_ s/two/too/
16:21 ingy s/too/2/
16:21 obra I hear ingy and damian are planning on a high-bandwidth hacking session soon
16:21 ingy s/2/II/
16:22 ingy obra: correct
16:22 obra woot
16:22 Juerd Don't hack my high bandwidth
16:22 obra <- unsure if he gets to go to japan
16:22 ingy obra: really? what up?
16:22 obra family stuff
16:22 Juerd ingy: In ASCII, do you prefer "doet Net" or "dot Net"?
16:23 ingy heh
16:23 ingy dot Net
16:23 Juerd ok :)
16:23 ingy I don't just prefer it I use it
16:23 Juerd The latter is less important :)
16:24 Juerd I constantly use a name I don't like.
16:28 ingy obra: freenode is not letting me identify (thus msg you)
16:29 obra  urgh
16:29 ingy obra: msged for magnet
16:29 ingy I think
16:32 fglock ruoso: the new PCR is very small - there should be only a few bugs left, if any - I guess I'll just finish the grammar by hand, and fix the grammar compiler later (the grammar compiler can then be rewritten using the working PCR)
16:33 fglock &
16:44 merlyn Chuck Norris Perl6 joke:  Chuck Norris can *have* the colon.
16:52 arcady joined perl6
16:54 svnbot6 r9638 | ingy++ |  r5749@bytes:  ingy | 2006-03-17 08:54:03 -0800
16:54 svnbot6 r9638 | ingy++ |  Beg Damian to make utf8 the one true encoding in Perldoc.
16:54 rindolf merlyn: I don't get it.
16:55 merlyn first, are you familiar with www.chucknorrisfacts.com ?
16:55 merlyn it's an internet meme
16:55 rindolf merlyn: no, I'm afraid not.
16:55 merlyn then you wouldn't get it.
16:55 eric256_ hey ingy....will items be heirerchial (god i can't spell that word) in Perldoc?
16:55 eric256_ is now known as eric256
16:57 estel joined perl6
17:02 nnunley joined perl6
17:05 ingy eric256: you mean can you nest?
17:05 ingy the answer is yes
17:05 ingy or maybe you meant something else
17:05 eric256 yea. i guess.  i want to be able to store data inside it. like for the builtin functino documentation we would want to be able to store information about the caller type, and first argument type for SMD and MMD index pages.
17:06 ingy ok, well let's not overload the fuck out of perldoc
17:07 eric256 the point is i don't think that is
17:07 eric256 =first_arg: String
17:07 eric256 thats already syntax you support, i just wanted to see if when its parsed if it well nest properly
17:08 eric256 then it would be easy to make a script that parses the builtin function page and produces two index pages sorted by what they want on each page.
17:12 ruoso joined perl6
17:14 vel joined perl6
17:14 DesreveR joined perl6
17:17 bernhard joined perl6
17:17 * eric256 looks around.... lol
17:17 eric256 anyway ingy  making sure the features it already has can be used for things you havn't forseen isn't overloading the crap out of it...its just being creative
17:18 eric256 besides the ability to add attributes to an item would certain be usefull for documentation. because it would be nice to store information like ownership, date added, api supported, whatever to Perldoc in a way that can be accessed and usefull to scripts/editors/etc. ;)
17:18 eric256 time to go play outside. later
17:19 eric256 left perl6
17:19 sahadev ?eval_bot my %h = (a => 1, b => 2, c => 3); say "%h.keys.reverse()"
17:20 sahadev ?eval my %h = (a => 1, b => 2, c => 3); say "%h.keys.reverse()"
17:20 svnbot6 r9639 | ruoso++ | bootstrap target now really reacompiles everything it can
17:24 nothingmuch møøsage
17:24 rindolf nothingmuch: yø!
17:25 nothingmuch yo mc parsec yo
17:25 nothingmuch wrappin to you mama like a
17:25 nothingmuch one sc
17:25 nothingmuch sec
17:25 * nothingmuch opens pugs/misc/rap.txt
17:30 SamB joined perl6
17:53 svnbot6 r9640 | ruoso++ | bootstrap target now checks is lrep can compile itself and result in the same thing
17:54 rindolf nothingmuch: here?
17:54 nothingmuch rindolf: slightly
17:55 rindolf nothingmuch: can you look at http://sial.org/pbot/16372 and find ways to improve it?
17:55 nothingmuch rindolf: use type annotations for clarification
17:56 nothingmuch don't use references when you don't need to
17:56 nothingmuch @4
17:56 nothingmuch @$n etc:
17:56 nothingmuch sub multiply_squaring_factors( @n, @m )
17:57 nothingmuch instead of using my %gsf_cache use 'state %gsf_cache' inside get_squaring_factors
17:57 nothingmuch it's like 'static' in C
17:57 nothingmuch use type annotations for documentations
17:57 nothingmuch e.g. sub Graham( Int $n ) if it's an int, or Num $n if it's any number
17:57 nothingmuch that usually helps to understand the algorithm
17:58 nothingmuch you can use 'where { }' subtyping for more constraints
17:58 pdcawley_ joined perl6
17:59 nothingmuch rindolf: again, wrt array references: $n_sq_factors = ...;
17:59 nothingmuch it's a collection of items
17:59 nothingmuch so my @n_sq_factors makes more sense
17:59 stennie joined perl6
18:00 nothingmuch while(1) is spelled loop { } now
18:00 nothingmuch     if ($low_square_val != $high_square_val)
18:00 nothingmuch        {
18:00 nothingmuch            return ($lower_bound, [$n, ($low_square_val+1)*($low_square_val+1)*​$rest_of_factors_product,$lower_bound]);
18:00 nothingmuch        }
18:00 nothingmuch that's just scary =)
18:00 nothingmuch use more temporary variables
18:00 nothingmuch to help readability
18:01 nothingmuch if ....
18:01 nothingmuch {
18:01 DesreveR is now known as r0nny
18:01 nothingmuch    last;
18:01 nothingmuch }
18:01 nothingmuch use 'last if  .. ;'
18:01 nothingmuch it looks like your code might want to use hyper operators
18:01 nothingmuch e.g. reduce
18:01 nothingmuch for @$n_times_lb -> $i
18:01 nothingmuch        {
18:01 nothingmuch            $rest_of_factors_product *= $i;
18:01 nothingmuch        }
18:01 nothingmuch $rest_of_factors_product = [*] @n_times_lb;
18:02 nothingmuch loop($i=$n+1;;$i++)
18:02 nothingmuch screaming for whitespace
18:02 nothingmuch use 'say
18:02 nothingmuch '
18:02 nothingmuch instead of print
18:03 nothingmuch hard coded "\n"s are not fun =)
18:03 nothingmuch and say is shorter
18:03 nothingmuch print "G($n) = " ~ $g_val ~ " [" ~ join(",", @$composition) ~ "]\n"
18:04 nothingmuch say "G($n) =  $g_val  [ {join(",", @$composition)} "]"
18:04 nothingmuch while ( $n_vec.elems() ) {
18:04 * ruoso wonders if there is a rule modifier that turns "trace" on
18:04 nothingmuch if it were an array not an array ref then '+@n_vec'
18:04 nothingmuch i'm not sure if that works well for arrays
18:05 nothingmuch or just 'while ( @n_vec ) {'
18:05 nothingmuch rindolf: done
18:06 rindolf nothingmuch: OK.
18:15 nothingmuch joined perl6
18:18 svnbot6 r9641 | ruoso++ | nailing down the new rules bug... the bug is in "quantifier" or "alt" rule
18:18 nothingmuch rindolf: ping
18:19 rindolf nothingmuch: pong
18:19 nothingmuch rindolf: did you see my comments?
18:19 rindolf nothingmuch: yes, I did.
18:19 nothingmuch okies
18:19 nothingmuch have fun
18:19 utku joined perl6
18:19 rindolf nothingmuch: OK.
18:20 * ruoso have to work... bye
18:21 nothingmuch also
18:21 nothingmuch style comment
18:21 nothingmuch for $start..$end -> $n
18:21 nothingmuch {
18:21 nothingmuch that reads very clumsily to me
18:22 nothingmuch if you put the brace on the same line as the for it looks better IMHO
18:22 hexmode joined perl6
18:22 nothingmuch for start to end as in in here
18:22 nothingmuch that's how i read the brace on the same line
18:22 nothingmuch for start to end as n
18:22 nothingmuch now this
18:22 nothingmuch that's how I read the brace on the same line
18:22 nothingmuch i think larry's style preference kind of optimized Perl 6 into something that looks better using his style =)
19:13 Arathorn is now known as Aragone
19:14 FurnaceBoy joined perl6
19:30 Quell joined perl6
19:48 DHGE joined perl6
19:55 gaal Alias_: I wouldn't insist on a direct gloss, since these concepts don't necessarily have one, unless this is indeed in the context of kabbala specifically, in which case there are probably several standard glosses to choose from. :)
20:31 nothingmuch joined perl6
20:32 fglock joined perl6
20:33 elmex joined perl6
20:34 nothingmuch joined perl6
20:34 ruz joined perl6
20:52 fglock ruoso: I found the 'missing comma' bug - it was in file 'p5hacks.pl' (fixed)
20:55 penk joined perl6
21:19 meppl joined perl6
21:23 GabrielVieira hello ther
21:23 GabrielVieira hello there
21:25 GabrielVieira fglock which is your OS?
21:27 merlyn hey fglock
21:29 GabrielVieira merlyn hey.. can i talk private?
21:29 merlyn to yourself?  Sure. :)
21:29 merlyn you mean with me? :)
21:30 GabrielVieira u know what i mean
21:30 GabrielVieira :P
21:30 GabrielVieira that about yapc brasil
21:30 GabrielVieira :)
21:30 GabrielVieira *thats
21:33 penk left perl6
21:34 Quell joined perl6
21:36 Quell joined perl6
21:38 fglock GabrielVieira: sorry, I was in a meeting - need to go back (I use Ubuntu, debian, windows XP)
21:39 GabrielVieira hum
21:39 GabrielVieira ill ask u somethings later
21:39 GabrielVieira :)
21:39 GabrielVieira i need to go now
21:39 whiteg joined perl6
21:39 GabrielVieira meet the wife.. u know
21:39 GabrielVieira hehe
21:39 GabrielVieira night all
21:39 GabrielVieira ;)
21:41 larsen joined perl6
21:47 merlyn hey fglock
21:49 ilogger2 joined perl6
21:55 sysfault joined perl6
21:55 sysfault left perl6
22:06 Lorn__ joined perl6
22:20 samv_home joined perl6
22:21 samv_home left perl6
22:37 fglock merlyn: ping
22:40 svnbot6 r9642 | fglock++ | pX/.../PCR - grammar compiler is fixed
22:44 jisom joined perl6
22:53 FurnaceBoy_ joined perl6
22:57 |mjk| joined perl6
22:59 Kattana joined perl6
22:59 hlen joined perl6
23:04 qu1j0t3 joined perl6
23:04 svnbot6 r9643 | fglock++ | pX/.../PCR - fixed OO; AST/emitter still broken
23:04 fglock audreyt: re pX/.../PCR - currying all rule method calls with the grammar works, but it "smells" wrong
23:05 fglock &
23:08 hlen_ joined perl6
23:35 pdcawley joined perl6

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo