Camelia, the Perl 6 bug

IRC log for #parrot, 2008-05-03

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:24 kid51 joined #parrot
00:29 mire joined #parrot
00:33 dalek r27299 | jkeenan++ | trunk:
00:33 dalek : Test of result() needs to be adjusted for change in config/auto/pcre.pm in
00:33 dalek : r27292.
00:33 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27299
00:49 Eevee joined #parrot
00:59 grim_fandango joined #parrot
01:03 particle[ventus] joined #parrot
01:03 tetragon My 10.5 build with '-undefined dynamic_lookup' removed but '-fvisibility=hidden' left intact dies off when building the GLUT callbacks (and I did add the '-lparrot' I needed on the no '-fvisibility=hidden' run)
01:04 tetragon _do_panic is undefined
01:05 tetragon libparrot.dylib is not exporting it (defined internally, though)
01:12 japhb tetragon: I'm not sure what the official way of doing a parrot-friendly panic is.  The intent is that panics from libglutcb.so occur when it detects something wrong with Parrot, like the interp that registered a callback has disappeared or something.
01:12 japhb There actually used to be more failure detection code that I cargo culted from the NCI internals, but that stuff was always broken because it needed access to private data structures,
01:12 japhb so I just ripped most of it out before submitting the patch.
01:13 japhb Something similar should be created (preferably as an exported API) in the future.
01:23 dalek r27300 | jkeenan++ | trunk:
01:23 dalek : Correct spelling error:  Ryby -> Ruby.
01:23 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27300
01:24 tetragon For testing purposes I added PARROT_API to do_panic in exceptions.c.  With that in place, the build and link completes, finding all symbols.
01:24 tetragon And yes, the triangle still spins
01:30 japhb and really, isn't that the most important thing?
01:30 kid51 When will I be able to run the spinning triangle as my wallpaper?
01:31 tetragon I consider building more important than spinning.  It can't spin until it builds
01:31 japhb kid51: When I get enough time to wrap the GLUT GameMode extensions.
01:32 kid51 Hey, it's spinning for me on both OSX 10.4 and Ubuntu!
01:32 japhb kid51: excellent
01:32 Theory joined #parrot
01:32 japhb AFK for a bit, commuting
01:48 sjansen joined #parrot
01:52 tetragon And as expected, removing '-undefined dynamic_lookup' from the compiler flags stopped the (unpatched) readline test from spitting out "strange" errors when faced with editline
01:58 dalek r27301 | jkeenan++ | trunk:
01:58 dalek : Conform to no trailing whitespace coding standard.
01:58 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27301
01:59 japhb joined #parrot
02:12 braceta joined #parrot
02:22 c9s joined #parrot
02:35 janus joined #parrot
03:15 JordanG joined #parrot
03:16 JordanG Hey all, I was wondering if the PIR Language Reference Manual actually exists or if its completely broken
03:23 cotto_home it doesn't appear to exist apart from what's on the wiki
03:23 cotto_home if you want to learn it, http://www.parrotcode.org/​docs/art/pp001-intro.html is a good starting place
03:23 cotto_home pir?
03:23 purl well, pir is a bofh at tufts or http://www.pir.net or Passive Infra Red or Parrot Intermediate Language (and almost an acronym) or the Price is Right.
03:24 cotto_home purl, pir is also http://www.parrotcode.org/​docs/art/pp001-intro.html
03:24 purl okay, cotto_home.
03:26 JordanG cotto_home: great, thanks!
03:27 cotto_home JordanG, why do you want to learn it?
03:28 cotto_home (just out of curiosity)
03:28 JordanG cotto_home: thinking about having a toy langauge i'm working on target parrot
03:29 cotto_home it'll be an interesting exercise
03:30 japhb JordanG: You might want to start with PCT ... much gentler than implementing in raw PIR.
03:31 JordanG cotto_home: thats why I'm doing it ;) I started in python but trying to generate python bytecode is a bit of a pain.
03:31 JordanG japhb: PCT?
03:31 purl it has been said that PCT is a good solution, but heavyweight or the Parrot Compiler Toolkit
03:31 JordanG ah
03:32 JordanG I heard of the compiler toolkit, I wanted to write everything from scratch, just as a learning exercise
03:33 JordanG I'll take a closer look at it though
03:33 JordanG writing parsers and such is just a BIT of a pain.
03:39 japhb JordanG: I look at it as stages, working your way down -- language to implement -> PCT -> PIR -> PASM -> JIT code.  You don't want to start by trying to do the JIT yourself.  Similarly, trying to do everything in PIR, though nicer than writing C code, can still be a lot of work.  So why not make use of the work pmichaud et al have already done?  That way you get introduced to the "Parrot way of doing things", get the easy stuff out of the
03:39 japhb way, and then you can later replace any part of that with raw PIR if you so desire.
03:39 japhb WTH?
03:39 purl WTH is, like, WhatTheHack and happens 28th july 2005 in .nl
03:39 japhb JordanG: I look at it as stages, working your way down -- language to implement -> PCT -> PIR -> PASM -> JIT code.  You don't want to start by trying to do the JIT yourself.  Similarly, trying to do everything in PIR, though nicer than writing C code, can still be a lot of work.  So why not make use of the work pmichaud et al have already done?  That way you get introduced to the "Parrot way of doing things", get the easy stuff out of the
03:39 japhb way, and then you can later replace any part of that with raw PIR if you so desire.
03:39 japhb That seemed to work a little better.
03:40 JordanG japhb: It is a very good argument.
03:40 japhb Anyway, my point was that in my experience, there is an amazing amount of stuff to learn about Parrot.  Trying to implement a new language using the low level tools is going to make the learning curve into a cliff.  ;-)
03:42 JordanG japhb: sometimes climbing cliffs is fun ;) but your right, if there are decent tools out there its silly to reinvent the wheel
03:42 JordanG japhb: do you know where any good documentation exists for it? Google results are a little sparse
03:43 japhb At least, it's silly if you don't take the time to look over an existing wheel or two first.
03:43 japhb Yes, there is a very good tutorial for it.
03:43 japhb looking for URL
03:43 japhb Start here: http://www.parrotblog.org/200​8/03/targeting-parrot-vm.html
03:43 JordanG japhb: just found the one on parrotblog
03:44 JordanG heh, yeah. Thanks ;)
03:44 japhb The code is all checked into languages/squaak, so that you don't have to retype it yourself if you don't feel like it.
03:44 japhb Personally, I find retyping tutorial code can be good for cementing the concepts.
03:45 JordanG i tend to agree
03:45 japhb There have been a couple minor improvements to PCT in response to the tutorial author finding some grotty bits; I'm not sure if the tutorial articles were updated to reflect those improvements or not.
03:45 japhb (Basically, the tutorial has a couple small workarounds that aren't needed anymore.)
03:46 JordanG I'm sure i'll muddle through. Or come back in here and annoy you guys ;)
03:46 japhb Always a pleasure to get someone new working on Parrot!
03:47 JordanG japhb: so are you a parrot dev or something?
03:47 Ademan_ joined #parrot
03:47 japhb I'm writing the OpenGL bindings for Parrot
03:47 JordanG nice
03:48 japhb :-)
03:49 tetragon joined #parrot
03:51 JordanG japhb: do you know if the Perl6 and Parrot O'Reilly book is any good?
03:51 JordanG i'm browsing the google books preview now, maybe I'll pick it up
03:51 japhb JordanG: I've heard it was good, but the dead tree version is considerably out of date.
03:52 JordanG japhb: yeah, i was kinda suprised they even had one
03:52 japhb However, O'Reilly released the rights on the Parrot portions to the parrot devs, so those chapters of the book are checked into the tree in docs/book/
03:52 JordanG ooh, nice
03:53 japhb I dunno if they have been updated fully yet.  I've mostly been basing my work on docs/pdds/, where the design docs live.
03:54 JordanG man, somebody needs to update the website and make it more obvious where all this stuff lives. I've been clicking around for like half an hour and didn't find any of this stuff
03:55 JordanG i was mostly looking at the examples sections, though
03:57 JordanG well, thanks for the help with everything, I'm sure i'll have more questions ;) Its getting lateish here and i'm exhausted, so I'm gonna head out
03:57 japhb g'night
04:02 JordanG Oh, one more question, is there like a parrot beginners mailing list i could read to learn from other newbies issues?
04:04 JordanG i attempted to google for such a thing, but mostly got results about birds.
04:22 Theory joined #parrot
04:38 iblechbot joined #parrot
04:38 Psyche^ joined #parrot
05:25 particle[ventus] JordanG: parrot-porters@perl.org is the only parrot mailing list. we happily take newbie questions.
06:33 iblechbot joined #parrot
06:33 davidfetter joined #parrot
06:53 braceta joined #parrot
08:17 cognominal joined #parrot
09:01 barney joined #parrot
09:05 schmalbe joined #parrot
09:08 dalek r27302 | bernhard++ | trunk:
09:08 dalek : [docs]
09:08 dalek : Configure.pl no longer checks for 'bc'.
09:08 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27302
09:33 purl joined #parrot
09:52 ambs joined #parrot
10:01 Ivatar joined #parrot
11:07 ambs joined #parrot
11:13 dalek r27303 | bernhard++ | trunk:
11:13 dalek : [Eclectus]
11:13 dalek : Add src/builtins/all.pbc as dependency for driver_nqp.pbc
11:13 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27303
11:15 dalek r27304 | allison++ | trunk:
11:15 dalek : [pdd] Strings PDD: fix a typo, move a few sections around.
11:15 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27304
11:43 dalek r27305 | allison++ | trunk:
11:43 dalek : [pdd] Delete docs/strings.pod after merging content that's still relevant into
11:43 dalek : Strings PDD.
11:43 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27305
11:47 dalek r27306 | bernhard++ | trunk:
11:47 dalek : [Eclectus]
11:47 dalek : Move the NQP->PIR step from tests-driver.scm to driver_nqp.pir
11:47 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27306
12:01 kid51 joined #parrot
12:11 ruoso joined #parrot
12:25 barney joined #parrot
12:50 dalek r27307 | bernhard++ | trunk:
12:50 dalek : [Eclectus]
12:50 dalek : Add test for nested 'begin'.
12:50 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27307
12:51 dalek r27308 | bernhard++ | trunk:
12:51 dalek : [Eclectus]
12:51 dalek : Using 'infix:' in name of builtin subs looks funny.
12:51 dalek : Renaming to 'eclectus:'
12:51 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27308
13:02 dalek r27309 | bernhard++ | trunk:
13:02 dalek : [Eclectus]
13:02 dalek : Remove redundant comments
13:02 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27309
13:20 * Jonathan finally gets his talk submissions in for NPW and FPW.
13:21 kid51 Jonathan:  Will we be seeing you in the Western Hemisphere any time this year?
13:22 cognominal jonathan++
13:23 Jonathan I fear not...not this summer, at least. Flights to the US are just too expensive in the summer.
13:23 Jonathan If there's any autumn conference, I maybe consider it.
13:23 Jonathan cognominal: I hope they don't overlap too much with what you plan on doing - I submitted more than I expect to give so there is some choice for the organizers, anyway.
13:24 mire joined #parrot
13:25 cognominal when I start parrot in the debugger and do  >  r  languages/perl6/perl6.pbc   it dumps (apparently) the perl6 pbc
13:26 cognominal this is weird because it works well other wise
13:26 Jonathan Which debugger? gdb or Parrot's own one?
13:27 dalek r27310 | bernhard++ | trunk:
13:27 dalek : [Eclectus]
13:27 dalek : Add support for 'newline'
13:27 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27310
13:27 cognominal warning: Could not find object file "/Users/stef/svn/parrot/src/pdump.o" - no debug information available for "src/pdump.c".
13:28 cognominal my bad
13:28 cognominal thx Jonathan
13:30 Jonathan OK, I need to take some exercise and go to the shopping mall...back later.
13:34 ambs joined #parrot
13:37 dalek r27311 | bernhard++ | trunk:
13:37 dalek : [Eclectus]
13:37 dalek : whitespace changes
13:37 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27311
14:13 wknight8111 joined #parrot
14:28 ambs joined #parrot
15:36 Zaba joined #parrot
15:37 Coke .
15:38 spinclad ,
15:39 Infinoid ;
15:39 * Coke was alerting purl to his presence in case of messages.\
15:40 ambs .
15:40 ambs ,
15:40 ambs ;
15:40 ambs !
15:40 ambs |
15:40 Infinoid ¡
15:40 spinclad ?
15:40 spinclad `
15:40 Infinoid happy Saturday :)
15:41 ambs indeed :)
15:41 * Zaba will go to school TOMORROW :<
15:42 spinclad we live in exciting times
15:42 Coke sunday?
15:42 purl sunday is looking tight or Mother's Day
15:42 Zaba Coke, yeah :(
15:42 Coke there's two phrases I didn't need in close proximity
15:42 ambs Mother's day, in fact.
15:42 Coke that's in a week!
15:42 ambs not here
15:42 ambs :D
15:43 Coke ... freaky europeans, not following our greeting-card manufacturer standards.
15:43 ambs :D
15:43 Coke my five year old is snorting at your craziness.
15:43 Coke (snorting. literally.)
15:47 Theory joined #parrot
16:00 ambs and now, I will make a choir sing
16:01 ambs see you later, folks
16:12 rdice joined #parrot
16:31 Andy joined #parrot
16:48 cotto_home good idea
16:48 purl cotto_home: Good Idea: Giving a small child a balloon. Bad Idea: Giving a small child a bunch of balloons.
16:53 * Jonathan wonders if that's because they're Helium balloons...
17:06 particle[ventus] joined #parrot
17:30 * spinclad wonders how big they have to be
17:42 Jonathan mwahahaha...
18:17 guru joined #parrot
18:23 AndyA joined #parrot
18:26 ambs joined #parrot
18:30 ambs back
18:32 cotto_home gah
18:32 cotto_home The information contains incomplete or invalid entries.
18:32 cotto_home Please select a Username that contains at least one letter and one number.
18:32 cotto_home someone needs to die
18:35 Zaba I hope it's not me
18:36 cotto_home not unless you committed that particular bit of UI braindeadness on United Healthcare's site
18:39 Zaba uh, no I didn't..
18:39 * Zaba survives!
18:39 Infinoid You die.  Die? (Y/N)
18:40 Zaba both Y and N default?
18:40 ambs lol
18:40 Infinoid yes, if you're a cat in a box
18:40 Zaba lol
18:41 * Zaba does not know how to spell that cat's owner's name
18:41 davidfetter schrödinger
18:41 Zaba right.
18:42 Infinoid problem is, I don't know how to type a quantum superposition into nethack's dialog box
18:42 Zaba lol!
18:43 Zaba the only games I play are roguelikes, and I horribly fail on 'em
18:44 ambs lolol
18:45 Zaba nethack, crawl (one at crawl-ref.sf.net)
18:45 davidfetter you are in a maze of twisty superposed passages
18:46 Zaba eh
18:46 davidfetter you may or may not be eaten by a grue
18:46 Zaba heh
18:47 ambs why are we talking about cats, mazes, mices and the like? :-P
18:47 Eevee joined #parrot
18:59 davidfetter i suppose we could work the parrot sketch into this somehow
19:00 particle joined #parrot
19:01 ambs :)
19:07 paco joined #parrot
19:14 particle[ventus] joined #parrot
19:43 * Tene tries to shuffle today's schedule around to make room for cardinal hacking.
19:53 mire joined #parrot
19:55 tetragon joined #parrot
20:12 Andy joined #parrot
20:16 ambs joined #parrot
20:16 ambs !
20:33 guru joined #parrot
20:33 dalek r27312 | infinoid++ | pdd13pbc:
20:33 dalek : [pdd13] Merge changes from trunk (-r27252:27311) into pdd13pbc branch.
20:33 dalek diff: http://www.parrotvm.org/svn​/parrot/revision?rev=27312
20:45 particle joined #parrot
20:46 Andy joined #parrot
20:57 particle[ventus] joined #parrot
21:22 desertmax joined #parrot
21:30 Theory joined #parrot
22:08 AndyA joined #parrot
22:16 particle[ventus] joined #parrot
22:20 Andy joined #parrot
22:40 tetragon joined #parrot
23:02 Infinoid what kind of exception should I toss if PackFile_unpack() fails to parse the packfile data?
23:02 Infinoid I want a sort of "malformed data" exception, but the closest I can find is E_ValueError, which is mainly used by utf8 encoding stuff.
23:04 Jonathan Infinoid: If there's nothing appropriate, I suggest adding one.
23:05 Jonathan That's what I did when there was no good error code to use for role coposition collision exceptions...
23:07 Infinoid two questions:  1.  is adding an entry to the exceptions enum all I have to do?  2.  do I have to bump the bytecode version for that change?
23:07 Infinoid (for now I'm just using E_ValueError, plenty of time to change that later)
23:07 Jonathan 1. yes, I think so. 2. no, because it doesn't affect the bytecode
23:08 Jonathan Well, easier to have a distinct value for it now maybe?
23:08 Infinoid perhaps.
23:08 Jonathan Then at least if people dislike the name you give it or decide it should be one of the existing exceptions, it's an easy renaming job.
23:08 Jonathan Find/Replace
23:09 Infinoid I've implemented PackFile->set_string_native and PackFile->get_integer_keyed_str, now its time to start writing tests, to see if the header parsed correctly
23:09 Jonathan And I suspect we want a code for it.
23:09 Jonathan Nice! :-)
23:09 Infinoid E_InvalidBytecode?  E_MalformedPackFile?
23:10 Jonathan I think the second one is better - the first could be confused with trying to execute an unknown opcode or something.
23:10 Infinoid good point, thanks.
23:12 Jonathan OK, I need sleep...laterz.
23:13 Infinoid sleep well

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

Parrot | source cross referenced