Camelia, the Perl 6 bug

IRC log for #parrot, 2008-02-09

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:08 HG` joined #parrot
00:19 Pabellon joined #parrot
00:53 slightlyoff joined #parrot
00:56 slightlyoff joined #parrot
01:04 slightlyoff joined #parrot
01:14 AndyA joined #parrot
01:22 * Coke returns from a day away from IRC.
01:40 kid51 joined #parrot
01:47 svnbotlt r25608 | jkeenan++ | harness:
01:47 svnbotlt : POD correction.
01:47 svnbotlt diff: http://perlsix.org/svn/parrot/revision/?rev=25608
01:55 Coke http://lolthulhu.com/ :: O HAI! BIN SLEEPN. I CAN HAS SOULZ NOW?
02:20 rafl joined #parrot
02:39 Andy joined #parrot
02:55 Tene the problem with using 'null' is that functions can't return null.
02:56 Tene I'm to the 'There are minor logic bugs and fencepost issues' debugging stage.
02:56 Tene Would be nice to be able to write in nqp instead.
02:56 Tene airplane now. bbl.
03:05 amoore joined #parrot
03:27 peepsalot i don't get what this call_list.txt does
03:27 peepsalot it defines parameter and return types for something?
03:29 arcady for NCI calls
03:30 arcady not quite sure why, but maybe it lists all the known NCI signatures?
03:33 arcady because src/nci.c is then generated from that file
03:33 Pabellon joined #parrot
03:35 arcady as an aside, I can't actually figure out what's in that file because Configure.pl appears to be hanging
03:35 slightlyoff joined #parrot
03:37 arcady specifically hanging on git-svn log
03:38 arcady perhaps because I'm using the git mirror on repo.or.cz rather than git-svn
03:39 peepsalot brb
03:43 peepsalot joined #parrot
03:47 zev what mode do people use for editing parrot grammar files in emacs?
03:48 arcady so far, fundamental
03:49 arcady but it would be nice to have a proper mode with colors and all
03:50 arcady especially one that made the {*} stand out nice and bright
04:00 zev I am unfortunately not proficient in making emacs modes
04:03 svnbotlt r25609 | petdance++ | trunk:
04:03 svnbotlt : fixed a proto, and switching to mem_sys_free
04:03 svnbotlt diff: http://perlsix.org/svn/parrot/revision/?rev=25609
04:39 Andy I wish I understood the DOD stuff 'cause then I could do useful things.
04:47 svnbotlt r25610 | petdance++ | trunk:
04:47 svnbotlt : remove an unused initialization
04:47 svnbotlt diff: http://perlsix.org/svn/parrot/revision/?rev=25610
04:49 peepsalot what's DOD?
04:49 purl rumour has it DOD is Dept. of Defense (see 'spook') or Day of Defeat, another Half-Life mod or dead object detection or a company that makes some decent effects pedals or DILDO OF DEATH or beyond stef's understanding or Dead Object Detecton
04:49 peepsalot ah, dead object detection
04:50 peepsalot purl is so smrt
05:10 arcady boy is python's syntax fun
05:10 arcady (a,b,c) is a tuple
05:10 arcady so is (), but not (a)
05:10 peepsalot heh
05:10 Piper joined #parrot
05:10 Piper Hi there.  I am Piper.  I am now publically logging this channel.  If you don't want to be logged, please leave now.
05:10 purl i already had it that way, Piper.
05:10 arcady instead to get a one-element tuple you write (a,)
05:12 arcady I've only gotten as far as making it understand () and (a,) correctly
05:13 arcady oh yeah, and a[1,2,3] really means a[(1,2,3)], likewise a[1,] is a[(1,)]
05:13 arcady but a[] is not a[()]
05:16 peeps joined #parrot
05:16 peepsalot oops
05:54 svnbotlt r25611 | petdance++ | trunk:
05:54 svnbotlt : added mem_allocate_n_typed macro
05:54 svnbotlt diff: http://perlsix.org/svn/parrot/revision/?rev=25611
05:59 svnbotlt r25612 | petdance++ | trunk:
05:59 svnbotlt : consting & localizing
05:59 svnbotlt diff: http://perlsix.org/svn/parrot/revision/?rev=25612
06:10 particl1 joined #parrot
06:26 DarkWolf84 joined #parrot
07:02 Tene arcady: yeah, for the git mirror on git.or.cz, you don't use git-svn, just normal git commands.
07:23 paq joined #parrot
08:40 japhb joined #parrot
09:10 Tene yay, 'make test' passes.
09:16 HG` joined #parrot
09:25 cognominal_ ho, I should try git
09:55 nopaste joined #parrot
10:13 alvar joined #parrot
11:13 spinclad purl, forget Hi there. I
11:13 purl spinclad, I didn't have anything matching hi there. i
11:15 spinclad not so, but never mind.
11:41 mire joined #parrot
11:54 iblechbot joined #parrot
12:30 fridim joined #parrot
13:04 alvar joined #parrot
13:16 alvar_ joined #parrot
13:20 cognominal_ joined #parrot
13:23 amoore joined #parrot
13:37 cognominal_ joined #parrot
14:32 itz joined #parrot
14:38 Alias joined #parrot
14:38 Alias Man, that Piper alert gets annoying
14:38 Alias (ignores it)
14:40 Alias left #parrot
14:40 Alias joined #parrot
14:41 Alias left #parrot
14:41 Alias joined #parrot
14:41 Alias Much better
14:49 Andy joined #parrot
16:10 simcop2387 joined #parrot
16:28 turrepurre joined #parrot
16:35 Ron joined #parrot
16:39 alvar joined #parrot
16:41 peepsalot joined #parrot
16:54 jjore joined #parrot
16:55 peepsalot i asked in here a while ago what I could do to help develop ecmascript, and someone suggested i get the tests from pjs to run on the new ecmascript(which uses pct): http://users.fulladsl.be/spb1622/pjs/
16:57 peepsalot it looks like the original author tested in two different ways.  one was tests that I guess he wrote, these were tested by running a shell script that compares his compilers output with that of the spidermonkey engine
16:57 peepsalot this seems like a bad way to go about testing though
16:59 peepsalot the other way was that he ran the full test suite from spidermonkey on his code.  I'm not sure exactly, but I don't think that this route actually depends on having the spidermonkey engine, just the tests.
16:59 jonathan The second of those would seem to be the more standard approach.
16:59 peepsalot i'm wondering if it would be kosher (if I could get it to work) for ecmascript make test to basically grab the tests from mozilla cvs and try to run parrot ecmascript compiler on them.
17:00 particl1 see languages/perl6/Makefile for an example of that (or tcl for that matter)
17:00 particle breakfast &
17:05 contingencyplan joined #parrot
17:23 peepsalot heh, I suppose 0 tests would probably pass at this point, since comments are not supported
17:49 peepsalot hmm, not sure how to deal with creation of/ testing for this stupid .cvspass in a makefile
17:52 DarkWolf84 joined #parrot
18:08 particl1 joined #parrot
18:44 cognominal_ joined #parrot
18:47 Coke Adding support for comments is pretty easy. =-)
18:47 Coke (JS is just // and /* */, as I recall)
18:48 alvar joined #parrot
18:57 peepsalot i saw rules in the pg file for it, but they are commented out for some reason.  i guess it needs some implementation in action file too or something? I'm still learning all that stuff.  haven't gotten that far yet
18:58 peepsalot Coke ^
19:57 alvar_ joined #parrot
20:36 rdice joined #parrot
20:39 Pabellon joined #parrot
21:00 guru joined #parrot
21:28 turrepurre joined #parrot
21:41 guru left #parrot
21:44 Pabellon joined #parrot
21:48 amoore joined #parrot
22:02 Limbic_Region joined #parrot
22:23 rdice joined #parrot
22:28 contingencyplan joined #parrot
22:46 cognominal_ reading Protoobject.pir, why double-colon classname are legacy?
22:52 jonathan cognominal_: We have heirarchical namespaces now, that is namespaces within namespaces.
22:52 jonathan Syntactically in Perl 6, you'll write Foo::Bar::Baz. But at a PIR level, it'll compile down to a key.
22:52 contingencyplan joined #parrot
22:52 cognominal_ should I read some pdd?
22:53 jonathan http://www.parrotcode.org/do​cs/pdd/pdd21_namespaces.html
22:53 cognominal_ thx
22:53 jonathan np :-)
23:08 IllvilJa joined #parrot
23:59 davidfetter joined #parrot

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

Parrot | source cross referenced