Camelia, the Perl 6 bug

IRC log for #parrot, 2008-02-14

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:00 kid51 Okay, now I'm able to svn up again.  Dunno what happened.
00:01 jonathan Strange.
00:01 purl But true.
00:01 jonathan Ho hum, sleep time here, but I'm on my way to rules/grammars in Rakudo.
00:01 kid51 purl:  Age before beauty.
00:01 purl kid51: excuse me?
00:02 * jonathan is young and ugly
00:02 * jonathan sleeps
00:02 davidfetter g'night, jonathan
00:04 Limbic_Region joined #parrot
00:05 wknight8111 joined #parrot
00:11 svnbotl r25701 | jkeenan++ | tcif:
00:11 svnbotl : Restore all parts of test to uncommented-out status.
00:11 svnbotl r25702 | jkeenan++ | tcif:
00:11 svnbotl : Fix semantic error:  Should have been 'system and croak', not 'system or croak'.
00:11 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25702
00:38 silug joined #parrot
01:13 AndyA joined #parrot
01:27 jjore joined #parrot
01:38 japhb joined #parrot
01:39 Coke purl, Piper is the logbot for #parrot
01:39 purl ...but piper is |er...
01:40 Coke (fixed)
01:51 diakopter purl: piper?
01:51 purl piper is the logbot for #parrot
02:03 svnbotl r25703 | jkeenan++ | trunk:
02:03 svnbotl : 1.  Someone forgot to regenerate the MANIFEST after deleting a file.
02:03 svnbotl : 2.  Add docs/embed.pod to list of skipped files in linelength.t.  Its
02:03 svnbotl : POD has 2 '=item' entries which necessarily exceed 100 characters.
02:03 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25703
02:34 svnbotl r25704 | coke++ | trunk:
02:34 svnbotl : [codingstd]
02:34 svnbotl : We have a list of files here that isn't checked against
02:34 svnbotl : MANIFEST, so there were a few files that got deleted
02:34 svnbotl : from the repo but not deleted here.
02:40 svnbotl r25705 | jkeenan++ | trunk:
02:40 svnbotl : Applying corrected version of patches submitted in
02:40 svnbotl : https://rt.perl.org/rt3/Ti​cket/Display.html?id=50302.  Most of internals
02:40 svnbotl : of t/harness refactored out into lib/Parrot/Harness/DefaultTests.pm,
02:40 svnbotl : Options.pm and Smoke.pm.  Four test files added in t/pharness directory
02:40 svnbotl : (so named to make it easy to distinguish from the t/harness file).
02:40 svnbotl r25706 | jkeenan++ | harness:
02:40 svnbotl : Contents of 'harness' branch have been merged into trunk; okay to delete.
02:40 svnbotl r25707 | jkeenan++ | trunk:
02:40 svnbotl : Tag being deleted because the branch it tagged has been deleted.
02:40 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25707
02:47 svnbotl r25708 | coke++ | trunk:
02:47 svnbotl : [codingstd]
02:47 svnbotl : Several files were being skipped during the linelength test that
02:47 svnbotl : are passing. Don't skip those.
02:47 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25708
02:47 diakopter joined #parrot
02:48 Coke opbots, names
02:50 Andy evening
02:55 Coke ayup.
02:58 kid51 Coke: IIRC, you were looking into something about t/codingstd/pir_code_coda.t.  (It's failing.)
03:00 jjore joined #parrot
03:08 Coke Yes, which is why I removed it from t/harness.
03:08 Theory joined #parrot
03:09 jjore joined #parrot
03:09 Coke It is now checking more PIR files that are using a non-standard coda, as well as files that are *claiming* to be PIR but aren't (and therefore don't have any coda)
03:09 Coke I should open a ticket about that.
03:10 kid51 Yes, please.  I hate it when I get to the end of make test and get failures in the 6 final tests.
03:10 Coke ... but I removed it from the harness.
03:10 Coke It shouldn't be run unless you're running it manually.
03:11 Coke ... oh. you just refactored the harness. did you not track the remove in the refactor?
03:11 kid51 [li11-226:parrot] 557 $ grep -n pir_code t/harness
03:11 kid51 [li11-226:parrot] 558 $
03:12 kid51 That's what you want, correct?
03:12 kid51 or is it the other way around?
03:12 Coke ... presuming that check is still done in the actual t/harness script, yes.
03:13 kid51 errrr, that's probably no longer true
03:13 Coke lib/Parrot/Harness/DefaultTests.pm
03:13 Coke yah. you appear to have re-added that back in.
03:14 Coke NBD; just remove that line from that file for now.
03:18 kid51 k
03:22 Coke ticket opened.
03:24 kid51 The file in which to make adjustments to the list of tests is now:  lib/Parrot/Harness/DefaultTests.pm
03:24 svnbotl r25709 | jkeenan++ | trunk:
03:24 svnbotl : Temporarily removing t/codingstd/pir_code_coda.t from @developing_tests.  Changes made elsewhere are causing this to fail on too many files to be useful.
03:24 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25709
03:25 Coke kid51: Yes, I added that to the ticket.
03:25 * kid51 must sleep
03:25 purl $kid51->sleep(8 * 3600);
03:25 Coke (so whoever closes it out will know what to re-enable.)
03:25 Coke Thanks!
04:35 Coke wow. configure --cage sure is noisy. =-)
04:45 svnbotl r25710 | coke++ | trunk:
04:45 svnbotl : [cage]
04:45 svnbotl : Eliminate two unused C variables from a code generator.
04:45 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25710
04:52 Coke Andy: ping
04:52 Andy yessir
04:52 Coke compilers/imcc/pbc.c:187: warning: 'old_blocks' declared 'static' but never defined
04:52 Coke ... but it *is* defined, starting on line 327.
04:53 Coke can you 'splain it to me?
04:57 Coke There seem to be several warnings of this type.
05:05 Coke also, wtf is "return reason=4, 0;"
05:32 Coke -> afk
05:33 Daveman joined #parrot
06:10 c_otto joined #parrot
06:26 iblechbot joined #parrot
07:09 uniejo joined #parrot
07:58 cotto_ joined #parrot
08:44 HG` joined #parrot
09:26 Ademan joined #parrot
10:00 jisom joined #parrot
10:41 dalek joined #parrot
10:43 cognominal_ joined #parrot
12:10 iblechbot joined #parrot
13:11 svnbotl r25711 | jonathan++ | trunk:
13:11 svnbotl : [rakudo] Initial work on parsing rule, token, regex and grammar, plus a first-cut implementation to make regex work. rule and token currently have regex semantics, however.
13:11 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25711
13:31 svnbotl r25712 | jonathan++ | trunk:
13:31 svnbotl : [rakudo] Junctions should not have a pick method, so remove it. Also remove now-unsed attempt at auto-threading.
13:31 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25712
13:48 Andy joined #parrot
13:50 kj joined #parrot
13:57 moritz jonathan: nice talk!
14:05 jonathan moritz: Thanks :-)
14:05 jonathan I finished that regex example like 2 minutes before the talk. ;-)
14:14 svnbotl r25713 | jonathan++ | trunk:
14:14 svnbotl : [rakudo] Tweak to allow you to do all(@array) as well as all(42,45,...).
14:14 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25713
14:17 bphillips joined #parrot
14:18 gryphon joined #parrot
14:33 Coke jonathan: ah, the joys (and stress) of JIT talks.
14:35 cognominal_ jonathan, url of your talk, even if I prefer to see you in person :)
14:35 Coke so, can anyone answer my cage warning of last night?
14:45 svnbotl r25714 | kjs++ | trunk:
14:45 svnbotl : [src] change a lying comment into something truthful. No src is touched, just doc.
14:45 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25714
14:52 svnbotl r25715 | kjs++ | trunk:
14:52 svnbotl : [src] add some documentation.
14:52 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25715
14:59 DarkWolf84 joined #parrot
15:07 IllvilJa joined #parrot
15:18 rdice joined #parrot
15:48 svnbotl r25716 | kjs++ | trunk:
15:48 svnbotl : [src]
15:48 svnbotl : + add some documentation
15:48 svnbotl : + fix some spacing w.r.t. * symbol for pointers
15:48 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25716
16:03 Coke is it me, or does it not make sense to have both a script to check for function docs, and then stub function docs which say "not documented" which look like they'd make the tst pass?
16:04 particle yeah, i disagree with what ptc did there.
16:04 particle it has increased our technical debt.
16:07 svnbotl r25717 | kjs++ | trunk:
16:07 svnbotl : [src] add function documentation and fix copyright year.
16:07 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25717
16:09 Coke Seems like the test is better than the stubs.
16:09 Coke (assuming a spherical test)
16:38 jonathan cognominal_: will pop the talk online when I get chance
16:38 * jonathan is pissed off
16:38 jonathan I don't think I'll bother coming to GPW again.
16:39 jonathan The snipey comments about Perl 6 are annoying. Attacking me personally is unacceptable.
16:40 jonathan This is the only place I have to put up wiht this crap.
16:40 particle wtf?
16:40 particle who's organizing that workshop? you two should have a chat.
16:40 jonathan The organizers are sincere enough.
16:40 jonathan It's just the general attitude here.
16:41 particle the organizers should make it clear that personal attacks are not welcome.
16:41 jonathan I'm probably feeling it more now somebody chose to make it personal and reference to what I did with my wife...
16:41 jonathan Not that I have one.
16:41 particle if it were my workshop, that person would be thrown out and barred from ever returning.
16:43 jonathan It's not everyone. Some people here are genuinely interested.
16:43 particle of course.
16:43 purl Indubitably.
16:44 pmichaud_ so, is this something I should be looking forward to for my p6 talks?
16:45 Coke ... I get that attitude very day from the crowd of perl5 developers I associate with online. it's quite disheartening.
16:45 cognominal_ I have hardly seen any better talks than jonathan ones. I can't understand people critisizing them
16:46 jonathan pmichaud_: Almost everywhere I go to, I get a good response.
16:46 pmichaud_ so do I.  I even get a good response from the python conferences I've attended.
16:46 cognominal_ jonathan talks have always well been received in France
16:47 jonathan Yes, I know. I've enjoyed these conferences.
16:47 pmichaud_ but given that I'm branching out to giving talks in europe, I'm wondering if there will be a difference there
16:47 particle it's individuals, not regions, that are the problem.
16:47 jonathan particle: Very much so. And it's a *minority*.
16:48 jonathan It's just frustrating when I've pretty much taken half a month off $DAUJOB for the sake of the Perl community.
16:48 cognominal_ pmichaud, you are welcome to talk in France.
16:49 pmichaud_ cognominal_: oh, believe me, whenever I can get around to places I definitely want to do so.  I'm hoping to make it to YAPC::EU this summer, but would like to hit a few other places as well on the same trip
16:49 cognominal_ i don't remember the date for the fpw in Paris
16:50 cognominal_ but I am sure we can make a technical meeting in Paris if you come
16:50 cognominal_ without talking of welcoming you in a restaurant
16:50 Andy What's GPW?  And who's sniping on you, Jonathan?
16:51 particle german perl workshop
16:51 cognominal_ strange, I got a ood remembering of YAPC in Munich
16:51 cognominal_ s/ood/good/
16:52 jonathan Andy: I'm not interested in naming names.
16:52 Andy Attendees?
16:52 Andy Or organizers?
16:52 purl organizers are sincere enough.
16:52 jonathan Andy: Attendee.
16:52 Andy I think that's one of those things that can't ever be stoppped.
16:52 jonathan True.
16:53 jonathan The general attuitude amongst many people here isn't great.
16:53 jonathan But yeah, of course, individuals aren't the whole community.
16:54 particle honestly, i think that individual should be asked to leave.
16:55 pmichaud_ normally I'm a very level headed person.  I'm not sure how I would handle a negative reference to my wife at this point, though.
16:55 cognominal_ I like the monger communauty because it is a friendly one
16:56 pmichaud_ I'd probably just walk out of the conference.
16:56 cognominal_ so that must have been a stray guy
16:57 cognominal_ the least the organisers can do is to give you excuses
16:57 jonathan cognominal_: I haven't mentioned it to them.
16:57 jonathan I'm not interested in stirring it up.
16:57 particle i am.
16:57 pmichaud_ oh, I agree with jonathan here -- take the high road.
16:57 cognominal_ I suppose that kind behavior has been reported to them.
16:58 particle there should be a no tolerance policy for personal attacks.
16:58 pmichaud_ the organizers should be made aware that it happened, and then let it go at that.
16:58 cognominal_ If such thing happened in a french meeting, I would immediately raise up to say that the guy does not belong here and is free to leave.
16:59 Andy I dunno, I figure something like that is just best ignored.
17:00 Coke was this during the actual presentation!?
17:01 pmichaud afk, time for paula and I to get some lunch
17:02 cognominal_ BooK says in #perlfr  Jonathan is the " nicest guy on earth ou presque" (or close to it)   :)
17:03 jonathan Coke: No, afterwards over coffee
17:03 jonathan Well, after the coffee really...
17:03 jonathan Anyway, thanks for putting up with my ranting.
17:04 Coke pmichaud: enjoy!
17:05 Coke jonathan: np whatsoever.
17:05 cognominal_ well, you should know that all the communauty appreciates you as a person and for your contributions. So a lone remark should not affect you much.
17:06 cognominal_ I dont get the right word in English for communauté
17:06 particle pmichaud: i'm looking at writing .pm source files now for inclusion in perl6.exe
17:06 Coke cognominal_: sounds like "community"
17:06 particle "community"
17:06 purl "community" is only download, without commercial apps
17:06 cognominal_ yup
17:06 Coke no, purl community is all these jokers.
17:06 purl okay, Coke.
17:07 jonathan For whoever - I put in parsing of regex, rule and token, but am having some...fun...figuring out how to pass along :sigspace and :ratchet as needed.
17:08 cognominal_ is there heredoc in rakudo?
17:08 * jonathan is far from familiar with PGE
17:08 Coke anyone see the pugs branch announcement?
17:09 cognominal_ jonathan, I get two problems with     undef.WHAT  in rakudo.
17:09 cognominal_ one it does not compile
17:10 cognominal_ two, if I write   say undef.WHAT is returns  "Undef" instead of "Failure"
17:11 jonathan cognominal_: I'm nt sure how you are producing problem 2 given problem 1...
17:11 jonathan But problem 1 is certainly the case, verified that here.
17:12 cognominal_ oos
17:12 cognominal_ i meant  say undef().WHAT
17:12 cognominal_ or say (undef).WHAt
17:13 cognominal_ I am not yet familiar enough with rakudo to understand what happens
17:14 jonathan cognominal_: In STD.pm undef is there in the grammar, as a term.
17:14 jonathan In Rakudo it is currently implemented as a unary sub.
17:14 jisom joined #parrot
17:14 jonathan I'm not sure undef().WHAT should work.
17:15 cognominal_ me neither
17:15 cognominal_ but if it is defined as a sub...
17:16 TimToady undef.WHAT should return Failure
17:16 cognominal_ TimToady discussed at lenght yesterday in #perl6 about undef and failure
17:16 cognominal_ speaking of the wolf...
17:25 pmichaud right now rakudo is probably treating   undef.WHAT like   &undef.WHAT
17:26 pmichaud jonathan: the PGE::P6Regex parser understands :ratchet and :sigspace as adverbs (options)
17:27 pmichaud sorry, Perl6Regex
17:27 pmichaud so does the regex rule
17:27 pmichaud afk, lunch for real this time
17:42 particle so, i've got a file written in perl 6, and i want to compile it to bytecode.
17:43 particle however, i don't want the bytecode to include the perl 6 runtime
17:43 particle there's no --target=pbc
18:11 jjore joined #parrot
18:22 Theory joined #parrot
18:36 pmichaud (--target=pbc)  afaict, there's not a way in PIR to produce a .pbc file
18:37 pmichaud the only way I've been able to manage that is to do --target=pir, and then re-invoke parrot with --output-pbc
18:39 Coke ./parrot perl6.pbc --target=pir foo.p6 | parrot -o script.pbc -
18:39 Coke (untested)
18:41 particle i'm not sure the output from --target=pir is enough to work with parrot -o foo.pbc
18:41 particle but i'll try that and see
18:43 Coke sounds like a bug with the pir target, then.
18:44 Coke but I think you might be SOL on not including the perl6 runtime; if the runtime is used by any of the generate code... well, it's got to come from somewhere.
18:45 particle well, i'm generating a .pbc file from source written in perl 6
18:45 particle to be included as the perl 6 runtime
18:46 particle 1) build basic compiler. 2) use compiler to compile runtime. 3) include runtime in new compiler
18:46 pmichaud I haven't been wanting to set up the runtimes as separate .pbc
18:46 pmichaud I've been thinking
18:46 pmichaud 1) build basic perl6.pbc, 2) use it to compile .p6 code into pir, 3) rebuild full perl6.pbc that includes .pir versions of runtime
18:47 particle .include rather than load_bytecode.
18:47 pmichaud yes
18:47 pmichaud it avoids the path problems for now
18:48 particle ok, i'll try that.
18:48 particle however, the pir that's generated doesn't look right to me
18:48 pmichaud describe "doesn't look right"
18:48 particle src/runtime/foo.pm contains "sub foo($bar) { say $bar }"
18:49 particle src/gen_runtime.pir: $(LIBS_PERL) perl6.pbc
18:49 particle $(PARROT) $(PARROT_ARGS) perl6.pbc --target=pir --output=$@ $(LIBS_PERL)
18:50 particle LIBS_PERL = \
18:50 particle src/runtime/foo.pm \
18:50 pmichaud fwiw, the current perl6.pbc doesn't understand multiple .pm files on the command line
18:51 pmichaud unless the corresponding option is given to HLLCompiler
18:51 particle yes, that's not a problem yet
18:52 pmichaud (--combine might work)
18:52 nopaste "particle" at 24.19.3.148 pasted "src/gen_runtime.pir" (26 lines) at http://nopaste.snit.ch/12324
18:53 particle "_block10" isn't going to be unique, is it?
18:53 particle also, nothing calls "_block10"
18:53 pmichaud _block10 is "main"
18:53 particle but it's default :main
18:53 particle right.
18:54 particle if we do multiple files, what happens?
18:54 pmichaud I don't know about multiple files
18:54 particle i guess that's all part of multi-file support
18:54 particle if we're doing a .include, then :main is bad news
18:55 pmichaud ...but there's no :main in the output, so it isn't a problem
18:55 particle so, it's never used?
18:55 pmichaud it's used if the .pir/.pbc is invoked directly from parrot, yes
18:55 particle yes. but in this case.
18:55 pmichaud if it's .included, then it's never used
18:56 particle right. however... if there are multiple '_block10' subs, due to multiple files being .included...
18:56 pmichaud that's probably an issue, yes.
18:56 particle then the subs that have :outer("_block10") will have problems
18:56 particle ok.
18:58 pmichaud I don't have a good answer off the top of my head, short of uuids of some sort
18:59 AndyA joined #parrot
19:00 pmichaud also, I would tend to expect most .pm files in this sense to be modules or packages, in which case the '_block10's would all appear in separate namespaces.
19:01 pmichaud (doesn't resolve the overall problem, but ameliorates it somewhat for this case)
19:01 particle yes, true.
19:02 pmichaud and perhaps the real answer is to use load_bytecode instead of .include for this stuff
19:02 pmichaud we have to figure out where the .pbcs belong, though
19:15 cotto__ joined #parrot
19:19 spinclad joined #parrot
19:20 mj41 joined #parrot
19:26 jjore joined #parrot
19:27 Coke I tend to prefer load_bytecode because I'm under the untested delusion that it's faster once you've done the initial compile.
19:27 pmichaud .include still results in the initial compile being part of the .pbc
19:28 pmichaud with .include, the initial compiled code is part of the perl6.pbc file itself
19:28 pmichaud with load_bytecode, the (initial compiled) code is a separate .pbc that gets loaded at runtime, as part of loading perl6.pbc
19:28 pmichaud either way we're working with bytecode files, it's just a question of whether the bytecode is part of perl6.pbc or part of a pbc that perl6.pbc loads
19:29 pmichaud (perl6.pbc uses .include because pbc_merge is traditionally unreliable)
19:32 Coke tcl had a habit of jumping right and using the unreliable stuff.
19:32 Coke "in"
19:33 pmichaud anyway, most of the parrot compiler tools and compilers I've built have used .include simply as a better way of doing what pbc_merge would have otherwise handled for us -- i.e., packaging multiple items into a single .pbc
19:33 pmichaud I suspect that .include is also faster, since there's not the overhead of the extra load_bytecode
19:35 Coke fair enough.
19:35 Coke on the other hand, going with .include forces a recompile where load_bytecode would not.
19:36 pmichaud yes, but it's just a recompile once when building.
19:36 pmichaud whereas load_bytecode is overhead on each execution
19:37 Coke so it depends on the use case. =-)
19:37 pmichaud well, in the case of perl6.pbc, I think it's fairly safe to say that we execute far more often than we build :-)
19:38 * Coke cries that tcl is still borked.
19:38 Coke Now would be an excellent time for people to start working on those DEPRECATION tickets...
19:39 HG` joined #parrot
19:39 pmichaud agreed
19:39 pmichaud I need to get started on those
19:39 pmichaud others are feel free to pitch in
19:40 Coke I've started a half dozen of them and each time I find a small snag. :|
19:40 Coke (which i tried to document in the ticket.)
19:40 Tene marked with [DEPRECATION]?
19:40 Coke Tene: see DEPRECATED.pod for a list.
19:41 Tene Ahh.
19:41 Coke Each entry there points to a ticket.
19:41 * Tene will look
19:41 Coke danke1
19:41 Coke !
20:12 HG` joined #parrot
20:35 IllvilJa joined #parrot
20:39 svnbotl r25718 | coke++ | trunk:
20:39 svnbotl : [DEPRECATED]
20:39 svnbotl : Remove warning about long removed NCI call signature; Resolves RT #48733.
20:39 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25718
20:42 gryphon joined #parrot
21:07 jmcadams joined #parrot
21:07 jmcadams hi parrot hackers.  i just wanted to make sure that you all knew that the yapc::na cfp is open
21:07 jmcadams we'd love some parrot talks
21:08 jmcadams and i was talking to jim keenan about a possible workshop
21:11 * davidfetter waves to jmcadams
21:12 * davidfetter has a talk submitted, but not on parrot
21:12 jmcadams :)
21:12 jmcadams excellent
21:12 jmcadams i haven't looked in the queue for a few days
21:13 jmcadams it's been silently accepting submissions because there were a few folks who wanted to get there's in early
21:13 davidfetter o/` qu-e-u-e // woa, baby // me gotta go now // yeah, yeah, yeah, yeah, yeah, yeah o/`
21:14 jmcadams dbi link
21:14 jmcadams cool
21:14 jmcadams it's amazing how much faster submissions are coming in this year
21:16 Coke jmcadams: any chance of a pre-workshop hackathon?
21:16 * particle wishes he could make yapc::na this year :(
21:17 jmcadams Coke: it can be done in the dorm that we are getting
21:17 jmcadams they have some large rooms that we'll have 24x7 access too
21:17 jmcadams probably Saturday and Sunday
21:17 jmcadams i could help promote it, but not organize it per say
21:17 jmcadams you volunteering?
21:17 jmcadams :)
21:18 jmcadams particle... other plans?
21:30 suzy35 joined #parrot
21:36 particle jmcadams: a wedding in state college, pa
21:37 jmcadams chances are they'll both get married again... just tell them you'll catch the next round :)
21:38 jmcadams ... i actually think that the conference is conflicting with a few anniversaries too
21:38 jmcadams :(
21:38 jmcadams so many things to try to plan around
21:38 particle seriously.
21:38 purl Totally!
21:38 particle couldn't you have it on canada day or something?
21:39 davidfetter o/` oh, canada o/`
21:39 jmcadams we could go canda day through independence day
21:39 jmcadams there'd be around 5 people at the conference
21:39 particle i'll give a talk.
21:40 particle then nobody will show.
21:41 * particle can't believe there are no direct flights from seattle to pittsburgh anymore
21:41 * Coke thinks his wife is booking a flight for him for yapc now.
21:42 grim_fandango joined #parrot
21:50 rblackwe Particle Pittsburgh's flight situation is terrible.
21:50 particle yeah. us air sure messed things up.
21:50 particle great airport.
21:51 rblackwe Yes much work is being done to make it better.
21:51 rblackwe AirTran adds Pittsburgh to Florida flights.
21:51 rblackwe http://www.bizjournals.com/pittsburgh/sto​ries/2008/02/11/daily37.html?ana=from_rss
21:51 shorten rblackwe's url is at http://xrl.us/bf8eh
21:51 rblackwe whish is good for me since I go to FL.
21:52 rblackwe also new international flights are coming says our county executive.
21:52 particle maybe i'll talk to sir richard about getting virgin america to fly from pittsburgh
21:52 rblackwe We have lots of space available.
21:53 rblackwe I think one terminal is closed right now.
21:54 rblackwe jmcadams are you reaching out to people to get them to talk?
21:54 rblackwe For the Pittsburgh Perl Workshop we reached out to a few shy folk and it worked out well.
21:54 rblackwe Some people just need a poke.
21:54 particle i wonder if i'll finally be able to make ppw this year
21:55 rblackwe I wonder if we well get to anounce the date soon :)
21:55 particle it'd be nice to come home with some pepperoni from jimmy&nino's
21:55 jmcadams rblackwe: bdfoy was doing it, as was jrockway i believe
21:55 rdice joined #parrot
21:55 Coke I am pondering doing a parrot-related talk; suggestions welcome. =-)
21:56 jmcadams it is a good idea to ask people.... helps convince them if they are on the fence about it
21:56 rblackwe very cool jmcadams
21:56 rblackwe also many people do perl stuff that just are not connected to the perl machine.
21:57 jmcadams yeah, it turns out that there is a decent bioinformatics department at iit
21:57 pmichaud I'm hoping to do talks at yapc::na
21:57 * confound too
21:57 jmcadams steve lembark was going to get some of them to come
21:57 pmichaud coke, we should do "APL, part 2" or something :-)
21:57 jmcadams pmichaud: i know i'm hoping that you do :)
21:58 pmichaud I'm eagerly looking forward to yapc
21:58 jmcadams confound: you said your submitting at talk too... parrot related?
21:58 jmcadams btw, in regards to the workshop
21:58 rblackwe Benjamin C. Pierce the guy that wrote Types and Programming Languages which I understand was Audrey's motivation for Pugs is at U of Penn.
21:58 confound no
21:58 rblackwe which is not far away.
21:59 rblackwe I am conserding asking him for a talk but fear no one would go.
21:59 jmcadams jim had thought that we could take some time during yapc to teach people to compile parrot and how to poke around in the source
21:59 jmcadams so that they might come to the post-yapc hack-a-thon
22:00 rblackwe pmichaud I want a parrot workshop out of you for PPW :)
22:00 PerlJam jmcadams: Got anything interesting in the perlcast queue?  :)
22:00 jmcadams got a few edited from npw
22:01 jmcadams need to get jrockway for a catalyst book talk
22:01 jmcadams i've missed a time with him twice now
22:01 jmcadams shame on me :(
22:05 Coke pmichaud: heh. "it's been two years. What have we done? Nothin'."
22:09 rblackwe well nothing like CDD, Confernece Driven Development to get things moving along.
22:12 particle jmcadams: patrick and i are planning a "contributing to rakudo perl" talk for oscon
22:13 jmcadams maybe patrick can try a one-man practice run at yapc ;)
22:15 * jonathan is back
22:15 jonathan Though sleeping in just a moment.
22:17 * Coke ponders a talk where we develop a new language during the talk.
22:18 particle hmm, i may be able to make the conf
22:19 * jonathan will be at YAPC::Europe, all being well.
22:19 particle coke: sounds like a workshop, and a fun one
22:19 jonathan If pmichaud comes, I'll leave him to talk on Rakudo; if not, I'll do it.
22:20 Coke jonathan: I realllly hope to be there this year. Need to ask my boss about it next week.
22:20 * Coke also needs to get his passport in order.
22:20 davidfetter Coke, if you're carrying a US passport, move soon
22:21 davidfetter Coke, they're slow as molasses. took me 4 weeks just to get a few pages addes :(
22:21 jonathan Coke: Hopefully see you there.
22:22 davidfetter added*
22:26 particle hrmm, we certainly have some work to do with rakudo attributes
22:27 particle all attributes are stored as '$!foo'
22:27 particle if you declare with $!, you must use $!. if you declare with $, you may call '$foo'
22:28 particle i don't see any smartlinked tests for that in S12
22:29 particle pmichaud: think nqp should get *@foo or just rakudo?
22:29 particle i'm thinking just rakudo
22:33 particle so, i may be able to make yapc::na after all, but i can only do a pre-conf hackathon
22:42 Limbic_Region joined #parrot
22:45 wknight8111 joined #parrot
22:48 parrot-poke joined #parrot
23:25 cotto__ what would I use to determine equality between two PMCs of unknown types?  VTABLE_is_equal() doesn
23:25 cotto__ 't seem to be working
23:26 particle define equality
23:26 TimToady in p6 terms, do you mean eq, ==, ===, eqv, or =:=    :)
23:27 TimToady or ~~, for that matter :)
23:27 particle i think that last one is dwequality
23:27 kid51 joined #parrot
23:28 simcop2387 ok i'm sure the answer is very simple, but how would i do something like, s{\r\n|\n\r|\r}{\n}g in parrot if i've got a string in $S0
23:29 simcop2387 PIR specifically i mean sorry
23:31 particle cotto__: do you mean to test that a string or num value is equal, or that the pmcs are in the same memory location, or what?
23:31 Theory joined #parrot
23:32 cotto__ I mean their values
23:33 particle ok. string or number value?
23:34 cotto__ same type and same value
23:34 particle ah. same type, and value.
23:34 cotto__ yes
23:35 particle then is_equal is the way to go
23:35 particle you call is_equal on one of them, pass the other.
23:35 cotto__ using VTABLE_is_equal compiles fine, but gives the following error when I try to run it
23:35 cotto__ ../../parrot: symbol lookup error: /usr/src/parrot/parrot-svn/runt​ime/parrot/dynext/php_group.so: undefined symbol: VTABLE_cmp
23:36 cotto__ (except for VTABLE_is_equal)
23:36 cotto__ that's for a trivial case where I'm using it on two String PMCs
23:38 cotto__ http://pastebin.com/m602afa6f
23:40 particle simcop2387: see runtime/parrot/library/String/Utils.pir
23:42 particle cotto__: you're using VTABLE_is_equal incorrectly there
23:42 particle hang on...
23:42 purl Hang on... are you an incredibly stacked, egoless, chivalrous man?
23:42 cotto__ I suspected something like that
23:46 particle you probably want p1.is_equal(INTERP, p2);
23:47 cotto That won't compile.
23:48 particle are you writing a pmc?
23:48 cotto yes
23:49 particle are you inside a pmclass {...} block?
23:50 cotto no.  that code is part of helper function
23:50 particle ah.
23:50 cotto *a helper function
23:50 particle maybe it's p1.VTABLE_is_equal(INTERP, p2)
23:51 particle crud. sorry, i'm frazzled today.
23:51 svnbotl r25719 | jkeenan++ | trunk:
23:51 svnbotl : Tagging trunk at r25718 so that the tcif can be synched to it.
23:51 svnbotl r25720 | jkeenan++ | trunk:
23:51 svnbotl : Deleting superfluous tag tcif-25669
23:51 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25720
23:51 cotto I appreciate the help.
23:52 cotto that doesn't work either, though
23:52 particle could you paste the whole file? tools/dev/nopaste.pl -c #parrot -t "whatever" file.pmc
23:53 particle yeah, i have to start thinking in pmc again. too much switching contexts lately.
23:58 cotto CPAN puked.  The file is here: http://mksig.org/parrot/phparray.pmc
23:59 particle ah, whoops. i forgot you might need cpan for that
23:59 cotto the code in question is on line 141

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

Parrot | source cross referenced