Camelia, the Perl 6 bug

IRC log for #parrot, 2009-11-01

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 bacek src\ops\core.ops(128) : error C2037: left of 'pred_offset' specifies undefined struct/union 'Parrot_Context_attributes'
00:00 bacek What does it mean???
00:01 dukeleto bacek: trolls are in your machine. don't feed them.
00:01 dalek parrot: r42198 | NotFound++ | branches/context_auto_attrs:
00:01 dalek parrot: remove already merged branch
00:02 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42198/
00:06 dukeleto bacek: i didn't know you were on the twitterwebs!
00:08 jonathan bacek: Means the compiler doesn't think the thingy you tried to look up pred_offset in isn't a pointer to a struct.
00:09 jonathan oh, one too many negatives
00:11 bacek dukeleto, I'm everywhere. It's part of my plan to Conquer the World!
00:11 bacek jonathan, but it does work on gcc...
00:12 payload joined #parrot
00:14 bacek dukeleto, I'm usually twit in Russian :)
00:14 dukeleto bacek: all the more fun!
00:15 dalek parrot: r42199 | bacek++ | trunk (3 files):
00:15 dalek parrot: Expose Parrot_pcc_allocate_register and free_registers. Use it in
00:15 dalek parrot: Context.destroy
00:15 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42199/
00:16 dukeleto bacek: can you add stuff to Changes whenever you add to the PARROT_API ? that would greatly ease the job of the release manager who compiles the Changelist
00:17 dukeleto i would like to see this as a policy in docs/project/ somewhere
00:17 dukeleto bacek: does that seem reasonable?
00:17 bacek dukeleto, yeah... But my laziness winning atm :)
00:18 dukeleto bacek: i can understand. adding sometime before a few hours before the release would be fine :)
00:18 dukeleto but the sooner you do it, the less likely you are to forget
00:18 bacek dukeleto, that's why I don't opt-in as release manager :)
00:19 dukeleto bacek: you don't have to make the language sound good in the Changes file, just start a list under the heading: "Parrot API Changes"
00:19 dukeleto bacek: all we need to know is if something is added, deprecated or removed
00:19 dukeleto the release manager can make it sound awesome ;)
00:20 bacek dukeleto, will do. After merging Context and CallSig. API is likely to change during this merge.
00:20 ttbot Parrot trunk/ r42199 MSWin32-x86-multi-thread make error http://tt.ro.vutbr.cz/file/cmdout/128968.txt ( http://tt.ro.vutbr.cz//buil​dstatus/pr-Parrot/rp-trunk/ )
00:20 dukeleto bacek: sounds good. is there a good way for me to find out when branches are merged, other than read every svn commit?
00:21 bacek dukeleto, use grep!
00:21 dukeleto i could write a cronjob that looks for merge commits, i guess
00:21 dukeleto bacek: do all merge commits msgs have a pattern that they must match?
00:22 bacek dukeleto, they are usually stating "Merge something"
00:22 bacek docs/project/branching_guide
00:23 jonathan bacek: Meh. :-/
00:23 dukeleto so if I attempt match the pattern m/merge/ i will find all merge commits, but I may find false positives
00:23 bacek jonathan, I found it. Trying to fix
00:24 jonathan bacek: Guess I can try and hunt it down next time I need to get a later Parrot, if nobody finds it first.
00:24 jonathan oh, nice. :-)
00:24 jonathan bacek++
00:24 dukeleto how do I weed them out? does svn have any metadata associated to a commit that identifies it as a merge as opposed to a normal commit?
00:24 dukeleto jonathan: what's up?
00:24 purl The birds, the sky, and the ceiling.
00:24 jonathan What purl said. ;-)
00:25 dukeleto can i have ops, please?
00:25 jonathan dukeleto: Depends. Can you resist the incredible temptation to kick purl? ;-)
00:25 kid51 dukeleto:  There is no requirement that merge commits identify themselves as such, though most do.
00:26 jonathan bacek: BTW, still working on context/call sig merge?
00:26 kid51 And I doubt there is anything in the svn metadata.
00:26 bacek jonathan, I'll scrap old branch and start new one. Maybe tonight
00:26 kid51 After all, you merge into your sandbox, and then commit from the sandbox.
00:26 dukeleto kid51: yeah, I was pretty sure of that already, but I asked in vain hope that y'all know something about svn that I don't :)
00:26 jonathan bacek: OK, I'm not in a hurry for it at all.
00:27 jonathan bacek: I was just curious of hte status and what roadblocks you might have hit.
00:27 jonathan It's not something I'd want to attempt. ;-)
00:27 Topic for #parrotis now Parrot 1.7.0 "African Grey" is out! | Fix issues caused by the pcc_reapply and context_auto_attrs merges | find out what's up with the slice opcode
00:27 dukeleto not to imply that there are issues with the context_auto_attrs merge, just taking a precautionary measure ;)
00:28 bacek jonathan, one roadblock I just cleaned in context_auto_attrs :)
00:28 dukeleto so is anybody seeing threads.t fail intermittently?
00:28 bacek jonathan, next one - is shiny weather outside :)
00:28 dukeleto kid51: what kind of write up are you looking for to the pm mailing list?
00:30 kid51 dukeleto:  For problems with threads.t, see https://trac.parrot.org/parrot/search?q=threads.t
00:30 jonathan bacek: Come here! -3C outside! :-D
00:30 kid51 Now, switch to pmgroups question
00:31 bacek jonathan, no way! I like current +22 more!
00:31 dalek parrot: r42200 | bacek++ | trunk/lib/Parrot/Ops2c/Utils.pm:
00:31 dalek parrot: Include "pmc_context.h" in all ops preamble.
00:31 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42200/
00:32 kid51 dukeleto:  What I'd be interested in is hearing:  Where did it take place?  How was it organized?  Who organized it/attended?  What topics were hacked on?  Were there any logistical barriers (e.g., net access) to be overcome?
00:33 kid51 Am interested because the Twitter feeds and other stuff I read in the last few days suggested that these have been somewhat regular events.
00:33 kid51 All other hackathons I've known about/tried to organize have been *special* events.
00:35 kid51 purl coverage?
00:35 purl i think coverage is http://cv.perl6.cz
00:36 kid51 purl Parrot coverage?
00:36 purl i don't know, kid51
00:37 dalek parrot: r42201 | bacek++ | trunk/include/parrot/context.h:
00:37 dalek parrot: Fix optimized build by include pmc_context.h inside parrot/context.h
00:37 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42201/
00:38 ttbot Parrot trunk/ r42200 MSWin32-x86-multi-thread make error http://tt.ro.vutbr.cz/file/cmdout/129022.txt ( http://tt.ro.vutbr.cz//buil​dstatus/pr-Parrot/rp-trunk/ )
00:38 dukeleto kid51: the pdx hackathon is a weekly event, every thursday
00:39 dukeleto kid51: according to calagator: http://calagator.org/events/search?query=hackathon , we have had 19 announced hackathons
00:39 dukeleto in a row
00:41 bacek msg chromatic Can you benchmark trunk (at r42201 it should build with --optimized)
00:41 purl Message for chromatic stored.
00:41 dukeleto it is a loose knit group of perl mongers, PUGS people (postgres), some ruby folks, parrot hackers and random portland tech people that use calagator and/or twitter
00:41 dukeleto bacek: you should be asking me that question
00:41 dukeleto bacek: benchmarked against what?
00:42 bacek dukeleto, against before merge
00:42 dukeleto bacek: i have access to a server to use for parrot benchmarks, donated to me by moritz++
00:42 bacek dukeleto, is benchmark results published somehow?
00:42 dukeleto bacek: how about vs 1.7.0 ?
00:43 dukeleto hmm, that would include pcc_reapply
00:43 dukeleto yeah, I can do it against an arbitrary rev. which one would you prefer?
00:43 dukeleto bacek: compare it against the trunk commit directly before the merge?
00:43 bacek dukeleto, r42192 - just before merge
00:43 kid51 dukeleto:  Looked at calagator site.
00:44 kid51 Looks like raison d'etre for moving to Portland ;-)
00:44 dukeleto bacek: they are not published in a way better than me telling whoever wants to hear about it, in an adhoc manner. i have plans to change this
00:45 dukeleto kid51: yeah, there is often more than 10 tech events per day in PDX
00:45 dukeleto most of which don't suck
00:45 dukeleto except for the SEO ones :)
00:45 bacek dukeleto, good-good.
00:45 kid51 SEO?
00:45 purl SEO is, like, search engine optimizer or lol, suicide by Google: http://forums.searchenginewat​ch.com/showthread.php?t=5800 or spammy bastards or Spammer, Execrably Overpaid
00:45 bacek Bulli beach is bloody good reason to go offline
00:45 bacek See you!
00:46 dukeleto bacek: this may be of use to you: http://timtowtdi.org/~leto/valgrind_output/
00:46 dukeleto bacek: that is valgrind output for every release version of parrot and trunk (from about 2 days ago) run against fib.pir
00:47 dukeleto bacek: fib.pir has gotten slower on linux recently :(
00:47 dukeleto kid51: PDX also has PDX.pm, which mostly spawned PDX Hackathon
00:48 dukeleto kid51: at first it was the PDX.pm Hackathon, but we decided to open it up to anybody, not just Perl Mongers
00:48 dukeleto PDX.pm is possibly the best/most fun Perl Monger group in existence
00:48 dukeleto don't let particle hear that, though
00:49 * kid51 has long had interest in organizing hackathons:  http://thenceforward.net/perl​/yapc/YAPC-NA-2007/houslight/
00:50 kid51 ... but hasn't succeeded in doing it in NYC since late lamented Phalanx project
00:50 kid51 4.5 years ago
00:50 dukeleto kid51: pdx hackathon has *very* little organization
00:50 dukeleto kid51: we all agree to show up at the same time/location to drink and nerd out. anything else is optional.
00:50 kid51 Do you have your own room in that bar?
00:51 dukeleto kid51: we usually occupy a large portion of it. but it is in a huge warehouse, and the "european board gamers" meet there the same night and consistently dwarf our numbers
00:52 kid51 Ah, that gives me a mental picture.
00:52 dukeleto like, pdx hackathon could have 10-15 people, while 100 people are there for board games. sometimes there is even a non-zero intersection between the groups :)
00:52 kid51 What makes a board game european?
00:52 dukeleto we usually grab the back wall which has lots of power outlets
00:53 dukeleto there have been some pictures of hackathons, one sec
00:54 NotFound kid51: Bids are in euros
00:55 NotFound (not in pounds, because british throw darts instead)
00:57 jonathan NotFound: Heh, you beat me to making a comment about the British... :-P
00:58 jonathan NotFound: I think the darts world champion actually is that British guy who drinks ten pints of beer a day or something.
00:59 NotFound And he championed because they pay beers to the winner.
01:00 jonathan Well, at least British beer is good. ;-)
01:00 NotFound That's right
01:01 jonathan The selection of beer is one of the things I miss about living there. OTOH, I now live close to the Czech border, so lots of good beer flows over the border from there. \o/
01:02 NotFound The beer contribution to good international relations is understimated.
01:02 NotFound The UNO must create an agency or something
01:05 * kid51 finds yet another error in the PIR book
01:07 kid51 Good, that one was already fixed in docs/book/pir/
01:09 Austin_away joined #parrot
01:13 dukeleto Portland beer is better than British beer. And it is pretty darn close to being as good as Czech beer.
01:14 dukeleto kid51: Picture of Eric Wilhelm and Ben Hengst at a recent PDX hackathon: http://www.facebook.com/photo.php?pid=35​71773&l=b4bc8098f3&id=697737500
01:15 dukeleto Portland has the most brew pubs per capita than any other US city.
01:16 dukeleto pretty bad grammar, but hopefully that made sense. I am enjoying some fine PDX brew right now :)
01:17 jonathan dukeleto: It sounds like the brew works well. :-D
01:18 dukeleto jonathan: I take that both as an insult and a compliment ;)
01:18 * jonathan was only in PDX once.
01:18 jonathan For OSCON. I had epic jet-lag for much of it though.
01:18 jonathan But still got to enjoy some beer.
01:19 jonathan Didn't get to see much of the city, mind. Was like, in for OSCON, attend, then out again.
01:21 jonathan It seemed like a pretty cool place, though.
01:21 nopaste "kid51" at 71.246.106.224 pasted "Book says last line of output should be '5.0' rather than '0'" (16 lines) at http://nopaste.snit.ch/18523
01:21 kid51 Cf: lines 1101-1104 of docs/book/pir/ch04_variables.pod
01:22 dukeleto jonathan: next OSCON is in PDX, as well as Open Source Bridge, which is cheaper and better. stay in town a little longer, I will buy you a beer :)
01:22 jonathan kid51: Heh. I'm inclined to agree with the book.
01:22 jonathan dukeleto: tbh, OSCON was kinda too big for me.
01:22 kid51 jonathan:  Why?
01:22 NotFound I have a window with the winxed parser c++ source and other with the parser example source code and sometimes I confuse them. The goal of having a familiar looking syntax has bee reached X-)
01:22 jonathan kid51: I dunno. Just how I felt.
01:22 dukeleto jonathan: OSCON is all enterprisy and commercial
01:23 dukeleto kid51: the book is correct, that is a bug
01:23 kid51 Is that bug tracked in a ticket?
01:23 jonathan kid51: To put it into context, I do a lot of workshops, which are more on the scale of hundred or so people or less, or YAPCs, which are more like several hundred.
01:23 jonathan So OSCON was a bit of a shock to the system.
01:24 jonathan :-)
01:24 kid51 jonathan:  No, I wasn't wondering why you thought OSCON was too big.  I was wondering why you were agreeing with the book rather than the results
01:24 jonathan kid51: oh.
01:24 jonathan :-)
01:24 kid51 I'm working my way thru the PIR book ...
01:24 jonathan kid51: Well, because something int-ifying to something, but the numification being so different, feels...odd.
01:24 dukeleto kid51: hmm, wait a sec. let me verify that
01:24 kid51 ... having successfully avoided learning PIR for my first 3 years in Parrot project :-)
01:25 jonathan kid51: I mena, if they both gave 0 consistently, of gave 5 and 5.0, I would kinda follow. It just feels inconsistent having one do 5 and the other not numify the string at all.
01:25 dukeleto no, it is not a bug, unless large swaths of parrot are not acting as they should
01:25 purl okay, dukeleto.
01:26 kid51 Docs say:  Converting a string to an integer or number only makes sense when the contents
01:26 kid51 of the string are a number. The C<String> PMC will attempt to extract a number
01:26 kid51 from the beginning of the string, but otherwise will return a false value.
01:26 dukeleto $P1 = box "5"
01:26 dukeleto $N1 = $P1
01:26 dukeleto say $N1
01:26 dukeleto prints "5\n"
01:26 jonathan dukeleto: OK, that's what I'd expect.
01:26 jonathan So why doesn't it in the larger example?
01:26 dukeleto jonathan: it does
01:27 jonathan Wait...I thought kid51 said that the book claimed the output for $N1 was 0?
01:27 jonathan erm
01:27 jonathan wait
01:27 jonathan s/the book/parrot/
01:27 NotFound It gives 5.0 as a number, but that doesn't mean that say emits '5.0'
01:27 dukeleto jonathan: it actually could still be a bug
01:28 jonathan NotFound: OK, fine but it seems to emit 0 in the nopaste.
01:28 dukeleto boxing a string which is not a valid number should give NaN, in my humble opinion
01:28 jonathan dukeleto: And for the integer case?
01:28 NotFound NaI X-)
01:29 jonathan lol
01:31 dukeleto jonathan: don't you laugh at my numeric and integral complications! ;)
01:31 dukeleto sadly, there is no NaI, only MAX_INT
01:31 kid51 The *biggest* problem here, IMO, is the comment in line 1104 of docs/book/pir/ch04_variables.pod
01:32 dukeleto which is platform dependent, but at least documented in ieee754-2008. certain integer operations return MAX_INT when something fishy happens
01:32 kid51 It leads me to believe that the output of 'say $N1' would be "5.0\n" ... which it clearly is not
01:32 kid51 platform is darwin/ppc mac os x 10.4.11
01:32 jonathan kid51: I must admit that I'm surprised that Parrot doesn't do that.
01:32 jonathan Or at least 5 (decimal part truncated)
01:33 NotFound $ winxed -e 'var p= "5 birds"; int i= p; say(i) '
01:33 dukeleto kid51: there is a bug there. say $N1 in that case should return "5\n"
01:33 dukeleto not 0
01:33 NotFound Unfortunately, winxed has no float type yet.
01:34 kid51 Speaking naively (i.e., not knowing much about PMCs), I wouldn't expect conversion of strings to floating-point numbers to be defined
01:34 dukeleto but numeric registers do not add a .0 when an integer is stored in a numeric register
01:34 kid51 Am filing TT
01:34 dukeleto there is no way to know that "it is an integer". unless a flag is set at modification time, which is not done
01:35 dukeleto kid51: yes, please add a test case to the TT, as an attached patch to a currently existing test file or a new one
01:35 dukeleto kid51: thanks for your work, this is important
01:35 NotFound dukeleto: a flag for what?
01:36 dukeleto NotFound: a flag saying " i am an integer being stored in a numeric register". none exists currently, to my knowledge
01:37 jonathan dukeleto: No, it's coerced to a float.
01:37 NotFound dukeleto: there is no such flag because such a thing doesn't make sense.
01:37 dukeleto NotFound: exactly.
01:37 dukeleto jonathan: exactly.
01:38 NotFound They are registers, they just contain values, no some metadata about how they were initialized or assigned
01:38 dukeleto i am trying to explain why a numeric register *can not know* that an integer is being stored in it
01:38 dukeleto NotFound: yes.
01:38 dukeleto kid51: does what we said make sense?
01:41 dalek TT #1168 created by jkeenan++: PMC to Float conversion not working as documented
01:42 kid51 dukeleto:  See the ticket, which documents the extent to which I understand the issue.
01:42 kid51 I cc-ed you all :-)
01:42 NotFound String.get_number calls Parrot_str_to_num
01:43 NotFound And Parrot_str_to_num is... uh... long and ugly.
01:46 jonathan Using the evalbot on #perl6 I did...
01:46 jonathan nqp: my $x := pir::box__PS('5 birds'); pir::say(pir::set__NP($x))
01:46 NotFound The before_dot part stops on any unexpected char by setting check_nan, and check_nan sets 0 on non NaN
01:46 jonathan And got output 0.
01:46 pmichaud I'm very surprised that doesn't work.
01:47 jonathan pmichaud: yeah same!
01:47 pmichaud I'm very surprised that hasn't shown up to be a problem before now :)
01:47 jonathan pmichaud: yeah same!
01:47 jonathan ;-)
01:47 jonathan I still can't believe that (a) Parrot does this and (b) there aren't tests that fail as a result...
01:47 jonathan I guess it could be by design. :-|
01:47 NotFound And doesn't make sense to check for "NaN" after having seen some digits.
01:52 dalek parrot: r42202 | mikehh++ | trunk/include/parrot/context.h:
01:52 dalek parrot: fix codetest failure - incorrect indenting in preprocessor directive
01:52 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42202/
01:52 * jonathan -> sleep
01:56 dukeleto Patches Welcome!
01:56 purl patches welcome is ponies welcome or Set Objectives, Achieve Results! or swahili for "Put up or shut up."
01:59 kid51 Well, before we can apply a patch, we need to clarify what the correct behavior should be.
01:59 kid51 Please post your thoughts to https://trac.parrot.org/parrot/ticket/1168
02:02 dalek parrot: r42203 | mikehh++ | trunk/compilers/pirc/src/bcgen.c:
02:02 dalek parrot: fix codetest failure - unused assert macro
02:02 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42203/
02:04 Austin_away joined #parrot
02:07 dukeleto kid51: i put my 2 centavos in
02:17 NotFound One more time: what print or says outs has nothing to do about how the register has been assigned.
02:19 bubaflub joined #parrot
02:19 dukeleto NotFound: yes. but the output of printing a numeric register should be specced and tested
02:19 dukeleto bubaflub: welcome!
02:19 purl Heh, not likely around here you won't be. or
02:19 bubaflub hola.
02:20 dukeleto purl, go play in traffic
02:20 * purl wanders off to dent some cars.
02:21 NotFound dukeleto: but mixing things unrelated in the same ticket doesn't help
02:21 mikehh All tests PASS (pre/post-config, smoke (#29558), fulltest) at r42203 - Ubuntu 9.10 amd64
02:22 dukeleto NotFound: I agree. New tickets welcome!
02:23 Topic for #parrotis now Parrot 1.7.0 "African Grey" is out! | Fix issues caused by the pcc_reapply and context_auto_attrs merges | find out what's up with the slice opcode | Latest modified TT's: http://icanhaz.com/parrotbugs
02:25 TiMBuS joined #parrot
02:41 RobertLJ joined #parrot
02:42 RobertLJ joined #parrot
03:08 Psyche^ joined #parrot
03:12 dukeleto languages?
03:12 purl languages is https://trac.parrot.org/parrot/wiki/Languages
03:20 dalek tracwiki: v103 | dukeleto++ | Languages
03:20 dalek tracwiki: https://trac.parrot.org/parrot/wiki/L​anguages?version=103&amp;action=diff
03:30 dukeleto Kea is the newest Parrot-based HLL. Factor of the Parrot VM: http://github.com/leto/kea . Patches and tuits welcome!
03:30 dukeleto ask me for commit bits to the github repo, and ye shall receive
03:31 dalek TT #1169 created by bubaflub++: [PATCH] lives_ok() for Test::More
03:31 bubaflub dukeleto: what exactly is kea? i.e. what does it do?
03:32 dukeleto bubaflub: it is a re-implementation of the Factor programming language on top of the Parrot Virtual Machine
03:32 dukeleto purl, factor?
03:32 purl no idea, dukeleto
03:33 dukeleto purl, factor is http://factorcode.org - Factor : A practical stack language
03:33 purl OK, dukeleto.
03:37 dukeleto bubaflub: i am running a test suite on top of trunk with your patch applied
03:38 dukeleto bubaflub: if it passes, patch will be committed :)
03:40 dukeleto is make -j still broken on trunk? that makes me really angry
03:40 dukeleto (whoever did that and didn't fix it)--
03:40 bubaflub yeah, i couldn't compile it with -j on trunk
03:40 bubaflub just plane ole make for me
03:40 dukeleto it wastes so much time, for no fucking reason
03:41 dukeleto chromatic told me how to fix it, i mostly forgot already. something doesn't have the correct dependencies listed in Makefile.in. i think
03:41 janus joined #parrot
03:41 dukeleto a pmc was added recently, so the list is not up to date. or somesuch
03:41 dukeleto can someone bisect the issue, to see when it started failing?
03:42 dukeleto that would help a bunch
03:43 dukeleto i meant to say "Factor on the Parrot VM" before. oh well.
03:43 dukeleto prepositions are the devil
03:46 dukeleto THE DEVIL, I SAY
03:53 * Tene just got google wave invite.
03:53 Tene someone's working on Factor on Parrot?
03:54 Tene Oh, you.
03:57 dukeleto Tene: me :)
03:57 dukeleto Tene: and at least one other. would you like a commit bit?
03:59 dukeleto Tene: i've got friends in low places
04:21 dalek parrot: r42204 | dukeleto++ | trunk (3 files):
04:21 dalek parrot: [t][TT# 1169] Add lives_ok() to Test::More, which tests if a bit of PIR runs without throwing an exception. bubaflub++
04:21 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42204/
04:21 dukeleto karma bubaflub
04:21 purl bubaflub has karma of 3
04:27 dalek TT #1169 closed by dukeleto++: [PATCH] lives_ok() for Test::More
04:28 dukeleto karma dukeleto
04:28 purl dukeleto has karma of 281
04:29 dukeleto does parrot ever use SSA form for anything, internally ?
04:29 dukeleto msg chromatic does parrot ever use SSA form for anything, internally ?
04:29 purl Message for chromatic stored.
04:34 Tene dukeleto: I would like.
04:36 dukeleto Tene: you now have a commit bit
04:48 dukeleto There was a screaming in the sky.
04:54 Tene karma tene
04:54 purl tene has karma of 572
04:56 dukeleto we now have a welcome wagon: http://twitter.com/parrotvm/status/5331285388
05:05 dukeleto i need to add instructions for @parrotvm on the release manager docs. some one bug me about that when I am less drunk
06:20 bubaflub joined #parrot
06:38 dalek tracwiki: v104 | dukeleto++ | Languages
06:38 dalek tracwiki: https://trac.parrot.org/parrot/wiki/L​anguages?version=104&amp;action=diff
06:45 dalek tracwiki: v7 | dukeleto++ | AbandonedLanguages
06:45 dalek tracwiki: add stuff about blizkost and the relation it has to PONIE
06:45 dalek tracwiki: https://trac.parrot.org/parrot/wiki/Aban​donedLanguages?version=7&amp;action=diff
07:04 nbrown joined #parrot
07:07 eternaleye joined #parrot
07:31 fperrad joined #parrot
07:38 mikehh joined #parrot
07:41 dalek tracwiki: v105 | fperrad++ | Languages
07:41 dalek tracwiki: typo
07:41 dalek tracwiki: https://trac.parrot.org/parrot/wiki/L​anguages?version=105&amp;action=diff
08:07 dalek tracwiki: v12 | barney++ | ListOfPlatforms
08:07 dalek tracwiki: Upgraded to Karmic
08:07 dalek tracwiki: https://trac.parrot.org/parrot/wiki/Lis​tOfPlatforms?version=12&amp;action=diff
08:14 iblechbot joined #parrot
08:46 mikehh msg bacek: rakudo don't like r42201/2
08:46 purl Message for bacek stored.
08:59 cotto mikehh, those are pretty innocent changes.
09:00 cotto well past time for bed
09:24 dalek parrot: r42205 | fperrad++ | trunk/compilers/pirc/src/bcgen.h:
09:24 dalek parrot: [pirc] fix build broken by merge r42197
09:24 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42205/
09:32 mikehh cotto: not from an installed parrot
09:34 mikehh msg bacek: with installed parrot it can't find '../src/pmc/pmc_context.h'
09:35 purl Message for bacek stored.
09:45 mokurai1 left #parrot
09:48 bacek joined #parrot
10:40 bacek o hai
10:44 mikehh hi there
10:46 bacek mikehh, "fixed" :)
10:46 mikehh bacek :-}
10:47 dalek parrot: r42206 | bacek++ | trunk/include/parrot/context.h:
10:47 dalek parrot: Fix including pmc_context.h in installed parrot
10:47 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42206/
10:47 dalek parrot: r42207 | bacek++ | trunk/MANIFEST.generated:
10:47 dalek parrot: Install pmc_context.h
10:47 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42207/
10:49 ttbot Parrot trunk/ r42206 i386-linux-thread-multi make error http://tt.ro.vutbr.cz/file/cmdout/129409.txt ( http://tt.ro.vutbr.cz//buil​dstatus/pr-Parrot/rp-trunk/ )
10:50 bacek oh shi...
10:51 ttbot Parrot trunk/ r42207 i386-linux-thread-multi make error http://tt.ro.vutbr.cz/file/cmdout/129424.txt ( http://tt.ro.vutbr.cz//buil​dstatus/pr-Parrot/rp-trunk/ )
10:52 bacek Why on Earth installed parrot has so different directory structure...
10:53 dalek parrot: r42208 | bacek++ | branches/context_unify2:
10:53 dalek parrot: New branch for merging Context and CallSignature
10:53 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42208/
11:00 joeri joined #parrot
11:23 dalek parrot: r42209 | bacek++ | trunk (33 files):
11:23 dalek parrot: Don't try to include pmc_context.h in sane way for installed and non-installed parrot. Just include it explicitely in core.
11:23 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42209/
11:33 dalek parrot: r42210 | bacek++ | trunk/compilers/pirc/src (2 files):
11:33 dalek parrot: Fix PIRC build
11:33 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42210/
11:33 dalek parrot: r42211 | bacek++ | trunk/compilers/pirc/src/pircapi.c:
11:33 dalek parrot: [pirc] Fix call to PF_create_default_segs
11:33 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42211/
11:37 Austin_away joined #parrot
12:21 mdevan joined #parrot
12:22 mdevan is this a good place to ask about pynie?
12:23 Austin mdevan: It would be an excellent place to ask, except there's nobody here right now.
12:26 mdevan Austin: maybe 5-6 hours later?
12:27 Austin During the week, probably. It's a weekend, so I'm less sure. Maybe evening, California time. (GMT-8)
12:28 NotFound joined #parrot
12:28 NotFound hi
12:28 purl hello, NotFound.
12:28 mdevan Austin: right, thanks. I'll be back later.
12:30 Coke joined #parrot
12:37 Austin purl, clock?
12:37 purl Austin: LAX: Sun 4:37am PST / CHI: Sun 6:37am CST / NYC: Sun 7:37am EST / LON: Sun 12:37pm GMT / BER: Sun 1:37pm CET / IND: Sun 6:07pm IST / TOK: Sun 9:37pm JST / SYD: Sun 11:37pm EST /
12:48 bacek joined #parrot
13:05 desertm4x joined #parrot
13:06 Whiteknight joined #parrot
13:07 Austin Good morning, Andrew.
13:07 Austin How was your Halloween?
13:19 Whiteknight good morning Austin
13:19 Whiteknight halloween was alright, didn't see too many kids
13:20 Austin Ah. I did.
13:20 Austin 2 years ago -- which was the last time I was in town for Halloween -- there were like 15-20 kids all night.
13:20 Austin So I bought accordingly this year, and got cleaned out in about 45 minutes. :(
13:31 Patterner joined #parrot
14:43 kid51 joined #parrot
14:49 mdevan joined #parrot
15:01 nopaste "kid51" at 71.246.113.243 pasted "Iterator does not work as documented" (25 lines) at http://nopaste.snit.ch/18524
15:03 kid51 Based on PIR book, page 34, I would have expected both versions to have output "10"
15:05 kid51 See:  /docs/book/pir/ch04_variables.pod lines 1284-1292
15:10 dalek TT #1170 created by jkeenan++: /docs/book/pir/ch04_variables.pod:  Iterator does not work as documented
15:12 Austin kid51: The "new iterator" approach was removed a while back, on purpose. I want to say it was in June, +/- a month. I think Bacek did the work.
15:13 Austin msg kid51 The "new iterator" approach was removed a while back, on purpose. I want to say it was in June, +/- a month. I think Bacek did the work.
15:13 purl Message for kid51 stored.
15:31 Psyche^ joined #parrot
15:32 dalek nqp-rx: 7df80c2 | pmichaud++ | README:
15:32 dalek nqp-rx: README update with note that double-quoted strings now interpolate $-vars.
15:32 dalek nqp-rx: review: http://github.com/perl6/nqp-rx/commit/7​df80c2a61ba34166ba48044cc341cc817264f7d
15:32 dalek nqp-rx: 2a9430b | pmichaud++ | src/HLL/Actions.pm:
15:32 dalek nqp-rx: EXPR should prefer $/.ast to $<OPER>.ast when building node.
15:32 dalek nqp-rx: Fix .panic message to occur on cursor object instead of Match.
15:32 dalek nqp-rx: review: http://github.com/perl6/nqp-rx/commit/2​a9430b2e463bc07b4a10983053e3aa86ef0f319
15:40 pmichaud (iterator) -- it would help if the error message suggested "use iter opcode instead"
15:42 theory joined #parrot
15:43 dalek nqp-rx: fd51bde | pmichaud++ | src/cheats/hll-grammar.pir:
15:43 dalek nqp-rx: Eliminate bogus debugging 'say' in cheats/hll-grammar.pir .
15:43 dalek nqp-rx: review: http://github.com/perl6/nqp-rx/commit/f​d51bde0a2f0b7db169697df7482e8e5deebfaa8
15:43 dalek nqp-rx: 1ee030e | pmichaud++ | src/stage0/ (3 files):
15:43 dalek nqp-rx: Update stage-0 files.
15:44 dalek nqp-rx: review: http://github.com/perl6/nqp-rx/commit/1​ee030e7d0a8717e6a316eac4ed3848ec7fb8fd0
16:13 dalek parrot: r42212 | mikehh++ | trunk/include/parrot/context.h:
16:13 dalek parrot: fix codetest failure - incorrect indenting in preprocessor directive
16:13 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42212/
16:26 iblechbot joined #parrot
16:42 dalek TT #1171 created by chromatic++: Lexicals and :outer not Thawed Properly from PBC
16:43 desertm4x joined #parrot
16:45 dalek parrot: r42213 | coke++ | trunk/t (3 files):
16:45 dalek parrot: ticket moved from RT to TT
16:45 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42213/
16:49 dalek parrot: r42214 | coke++ | trunk/t (2 files):
16:49 dalek parrot: ticket moved from RT to TT
16:49 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42214/
16:49 dalek TT #1172 created by chromatic++: Lexical Associations Not Thawed with Tailcalls
16:52 dalek TT #1173 created by jhorwitz++: NCI segfaults with null pcc_params_signature
17:10 Coke joined #parrot
17:21 ash_ joined #parrot
17:29 ash_ joined #parrot
17:56 davidfetter joined #parrot
18:01 mikehh I am getting a test failure in t/src/warnings.t - Failed test: 2 (smoke #29571) but smoke #29569 at same revision is ok
18:02 mikehh the only difference seems to be gcc version mine 4.4 theirs 4.3 (both linux amd64)
18:02 mikehh well perl 5.10.1 vs 5.10.0
18:16 dukeleto mikehh: can you nopaste the failure?
18:18 bostongeek24 joined #parrot
18:18 bostongeek24 hi
18:18 purl que tal, bostongeek24.
18:18 bostongeek24 is this a spanish chat room?
18:20 bostongeek24 hello?
18:20 purl Raise your hand in the back if you can't hear me.
18:20 bostongeek24 anyone here
18:20 purl look around and see, you dolt
18:20 bostongeek24 are you a bot?
18:20 purl Damn straight, and proud of it.
18:20 MinorToken joined #parrot
18:21 bostongeek24 are there any humans here?
18:21 bostongeek24 i don't really want to talk to a bot
18:21 mikehh smoke - http://smolder.plusthree.com/app/pu​blic_projects/report_details/29572 - go to first failure - look at tap output
18:23 mikehh or for that matter #29571 as well
18:23 bostongeek24 how do i install 1.7.0? if i go to the download section on the website it still has 1.4.0
18:24 jonathan bostongeek24: hi :-)
18:24 jonathan I'm not a bot.
18:24 jonathan ;-)
18:24 moritz re
18:25 jonathan ftp://ftp.parrot.org/pub/pa​rrot/releases/devel/1.7.0/
18:25 bostongeek24 but why does the website still have 1.4.0?
18:27 jonathan Were you looking in ftp://ftp.parrot.org/pub/pa​rrot/releases/stable/1.4.0/ ?
18:27 jonathan erm, just the stable relesae link?
18:27 jonathan If so, that just lists the every six month releases.
18:28 jonathan The developer release link should be up to date.
18:28 bostongeek24 is the svn up to date?
18:30 jonathan svn HEAD is the very latest, bleeding edge code, yes.
18:31 bostongeek24 how do i know which version i have installed?
18:31 mikehh bostongeek24: svn co https://svn.parrot.org/parrot/trunk parrot gets you the latest in parrot
18:32 jonathan bostongeek24: -V
18:32 jonathan parrot -V that is
18:35 bostongeek24 it says parrot is not installed
18:35 bostongeek24 after i download the source from svn do i need to compile it?
18:36 mikehh bostongeek24: i.e 'svn co https://svn.parrot.org/parrot/trunk parrot' retrievs the latest source and places it in dir parrot
18:38 mikehh yes - perl Configure.pl, make, and then make test etc
18:38 mikehh you might need some extra perl and library modules
18:39 bostongeek24 i have perl installed
18:39 ash_ joined #parrot
18:40 mikehh bostongeek24: what platform are you on
18:44 bostongeek24 linux
18:44 bostongeek24 ubuntu specifically
18:47 mikehh do you use apt get or synaptic or whatever
18:50 NotFound cls winxed styled: $ winxed -e "var f= new 'FileHandle'; f.open('tput clear', 'rp'); string s=f.readall(); print(s);"
18:54 bostongeek24 i used svn to get the latest version of parrot
18:57 mikehh bostongeek24: you need bison and flex, libgmp3-dev, libicu-dev, libreadline6-dev (or 5), libpcre3-dev, libglut3-dev maybe GDBM and PQ
18:58 jan joined #parrot
18:58 mikehh and some perl modules - depending on what you want to do
19:01 Whiteknight joined #parrot
19:03 mikehh Test::Base, Test::Pod::Coverage, Pod::Spell, File::HomeDir, File::Which, Readonly, RegExp::Parser
19:03 mikehh and Perl::Critic, Test::Pertl::Critic - that's if you want to run tests
19:04 bostongeek24 so should i just install it via synaptic?
19:06 mikehh that's what I did - the perl modules are under Sections, Perl Programming Language
19:09 mikehh the other stuff under Development and Libraries - Development
19:12 mikehh bostongeek24: I'll be around for a while, so if you have any problems, yell and I'll see if I can help
19:15 mikehh oh he dissapeared
19:21 buildbot joined #parrot
19:33 tetragon joined #parrot
19:46 chromatic joined #parrot
19:47 masak joined #parrot
19:56 chromatic Looks like a 4.405% slowdown from r42197, bacek.
20:00 chromatic left #parrot
20:00 chromatic joined #parrot
20:02 mokurai joined #parrot
20:15 eternaleye joined #parrot
20:42 dalek parrot: r42215 | chromatic++ | trunk (2 files):
20:42 dalek parrot: [context] Made Parrot_pcc_calculate_registers_size() inlineable to make
20:42 dalek parrot: destroying Contexts slightly less expensive.
20:42 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42215/
21:02 bacek_at_work good morning.
21:04 Whiteknight hello bacek_at_work
21:05 bacek_at_work alloha Whiteknight
21:08 brooksbp joined #parrot
21:19 payload joined #parrot
21:20 jonathan lol. Even our opening curly braces segv. http://gist.github.com/223728
21:21 jonathan (This is got by somebody who is running Parrot for an extended period.)
21:21 jonathan So probably hard to re-produce. :-(
21:21 ash_ joined #parrot
21:33 dalek parrot-linear-algebra: d2b5ece | Whiteknight++ | t/10-nummatrix.t:
21:33 dalek parrot-linear-algebra: add a bunch of tests for accessing by linear index
21:33 dalek parrot-linear-algebra: review: http://github.com/Whiteknight/parrot-linear-algebr​a/commit/d2b5ece6f68920d746ae1aee2356989e468cae86
21:33 dalek parrot-linear-algebra: a28414f | Whiteknight++ | src/pmc/nummatrix2d.pmc:
21:33 dalek parrot-linear-algebra: update the fill method to take optional sizes. Update the get_string VTABLE to actually print things out the way we expect
21:33 dalek parrot-linear-algebra: review: http://github.com/Whiteknight/parrot-linear-algebr​a/commit/a28414f95ce39af96b37216228ccd23816627a60
21:36 payload joined #parrot
21:52 Whiteknight joined #parrot
22:04 xenoterracide_ joined #parrot
22:07 xenoterracide joined #parrot
22:09 Zak joined #parrot
22:21 brooksbp joined #parrot
22:29 xenoterracide_ joined #parrot
22:38 ash_ joined #parrot
22:51 kid51 joined #parrot
22:54 dalek rakudo: 33111d4 | moritz++ | t/spectest.data:
22:54 dalek rakudo: remove two junction related tests from spectest.data
22:54 dalek rakudo: One was deleted, the other one now assumes we can call $j!eigenstates, which
22:54 dalek rakudo: is not so easy, and needs some magic and scary preamble, which we don't know
22:54 dalek rakudo: yet. So from a Rakudo perspective that test is dead, but since it's based on a
22:54 dalek rakudo: spec change that's OK for now.
22:54 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/3​3111d4345cc91abdc9e4c340e73dd914e3f9274
23:10 dalek parrot: r42216 | jkeenan++ | trunk/docs/book/pir/ch04_variables.pod:
23:10 dalek parrot: Correcting documentation to reflect discontinuation of 'new "Iterator"' approach.  Cf.:  https://trac.parrot.org/parrot/ticket/1170.
23:10 dalek parrot: review: https://trac.parrot.org/parrot/changeset/42216/
23:35 ash_ joined #parrot
23:38 cotto joined #parrot

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

Parrot | source cross referenced