Camelia, the Perl 6 bug

IRC log for #parrot, 2008-02-28

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:40 cotto_ Is it a bug that a Hash Iterator can only be initialized after its Hash has been assigned an element?
00:41 cotto_ nopaste?
00:41 purl rumour has it nopaste is at http://nopaste.snit.ch/ (ask TonyC for new channels) or http://rafb.net/paste or http://paste.husk.org/ or http://poundperl.pastebin.com/ or http://paste.scsys.co.uk/ or don't bother me while I'm eating
00:43 nopaste "cotto" at 131.107.0.105 pasted "hash iterator oddity demo" (31 lines) at http://nopaste.snit.ch/12412
01:02 svnbotl r26119 | petdance++ | trunk:
01:02 svnbotl : Fixed the test setup
01:02 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26119
01:02 AndyA joined #parrot
01:26 cotto_ nopaste?
01:26 purl nopaste is at http://nopaste.snit.ch/ (ask TonyC for new channels) or http://rafb.net/paste or http://paste.husk.org/ or http://poundperl.pastebin.com/ or http://paste.scsys.co.uk/ or don't bother me while I'm eating
01:29 Khisanth joined #parrot
01:34 cotto_ purl?
01:34 purl cotto_?
02:08 Coke seen chromatic?
02:08 purl chromatic was last seen on #parrot 18 hours and 51 minutes ago, saying: yes
02:10 Coke I'm getting a bus error on something on the freelist.
02:18 kid51 joined #parrot
02:28 svnbotl r26120 | coke++ | trunk:
02:28 svnbotl : [distro]
02:28 svnbotl : Ignore more generated files
02:28 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26120
02:37 c9s joined #parrot
02:55 c9s joined #parrot
03:25 svnbotl r26121 | jkeenan++ | trunk:
03:25 svnbotl : Add two test files to test internals of new config step auto::gettext.  Do
03:25 svnbotl : some refactoring inside auto::gettext::runstep() to increase testability.
03:25 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26121
03:50 Andy ping davidfetter
03:50 purl I can't find davidfetter in the DNS.
03:50 * kid51 must sleep
03:50 purl $kid51->sleep(8 * 3600);
04:00 gabriel joined #parrot
05:00 x joined #parrot
05:08 Pabellon joined #parrot
05:33 Pabellon joined #parrot
05:36 contingencyplan joined #parrot
06:36 cotto purl, karma for purl
06:36 purl A hell of a lot more than you cotto, that's for sure!
06:38 DarkWolf84 joined #parrot
06:39 Andy joined #parrot
07:13 svnbotl r26122 | petdance++ | trunk:
07:13 svnbotl : Changed an ARGMOD to ARGIN
07:13 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26122
07:16 mire__ joined #parrot
07:39 svnbotl r26123 | petdance++ | trunk:
07:39 svnbotl : overhauling how I sniff for splint results.  Most of y'all won't be interested.
07:39 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26123
08:04 svnbotl r26124 | fperrad++ | trunk:
08:04 svnbotl : [Lua]
08:04 svnbotl :  - add tests : getfenv/setfenv for thread
08:04 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26124
08:24 iblechbot joined #parrot
08:34 integral joined #parrot
09:10 ruoso joined #parrot
09:39 cosimo joined #parrot
09:56 slavorg joined #parrot
09:58 clunker joined #parrot
09:59 AndyA joined #parrot
10:19 ruoso joined #parrot
10:21 cotto joined #parrot
10:39 wknight8111 joined #parrot
10:44 mire__ joined #parrot
11:24 kid51 joined #parrot
11:57 kj joined #parrot
12:10 krunen joined #parrot
13:05 ruoso joined #parrot
13:33 krunen joined #parrot
13:58 c9s joined #parrot
14:03 comet joined #parrot
14:05 iblechbot joined #parrot
14:16 zaphod joined #parrot
14:44 Juerd joined #parrot
14:45 MagNET joined #parrot
14:46 wolverian joined #parrot
15:07 gryphon joined #parrot
15:33 particle 65/187
15:33 purl 0.347593582887701
15:47 kjs_ joined #parrot
16:19 jhorwitz joined #parrot
16:21 kj joined #parrot
16:22 skv__ joined #parrot
16:33 zaphod joined #parrot
16:51 DarkWolf84 joined #parrot
17:00 skv__ joined #parrot
17:11 skv__ joined #parrot
17:12 zaphod_ joined #parrot
17:15 skv___ joined #parrot
17:30 cosimo left #parrot
17:50 Psyche^ joined #parrot
17:52 purl joined #parrot
18:02 parrot-poke joined #parrot
18:17 skv___ joined #parrot
18:18 Coke particle: ping
18:19 particle hiya
18:19 skv____ joined #parrot
18:31 particle coke?
18:33 Coke particle: regarding your config question: find an existing config system that does 90% of what we want and steal that.
18:33 particle yes
18:33 particle i'm not sure perl is that system, but it's free :)
18:33 particle *perl's system
18:36 Tene Wow.  Quiet today.
18:38 particle yeah. too quiet....
18:39 skv____ joined #parrot
18:40 cotto_ Is it a bug that a Hash Iterator can only be initialized after its Hash has been assigned an element?
18:40 cotto_ http://nopaste.snit.ch/12412
18:40 Coke Tene: need some tickets? =-)
18:40 skv_____ joined #parrot
18:40 Tene Coke: you promised me some tickets ages ago.
18:41 parrot-poke GNU autoconf config seems to finally be going out of style
18:41 Coke and being replaced with/
18:41 Coke ?
18:41 parrot-poke layers of custom scripts in various languages :p
18:41 Coke ISTR we can't use autoconf because of the lack of sh on windows. (which always made me want to encourage someone to port sh.)
18:42 Theory joined #parrot
18:42 Coke we really shouldn't be re-inventing the config wheel. There's plenty of other wheels here.
18:42 parrot-poke the perl config is actually relatively clean and portable in comparison to all others I know
18:42 Coke rephrase: we shouldn't *need* to.
18:42 Tene Coke: we could just write sh on top of parrot, right?  ;)
18:43 Coke *thwap*
18:44 particle after parrot 1.0, we can use whatever config we want
18:44 particle for now, i'd like to make it easier to extend, without a massive rewrite
18:45 Coke true. rephrase: we shouldn't have needed to. =-)
18:45 parrot-poke This is kind of an odd question, but is there any idea of what will serve as CPAN for perl6?  Are modules going to be similar at all?  Has this been vetted yet?
18:45 particle there is an idea, some folks have in the past worked on it
18:45 Coke There was a lot of discussion about C6PAN before I dropped out of the perl6 pre-spec discussion. what was that, six years ago?
18:46 pmichaud it comes up from time to time, but afaik no hard decisions have been made
18:47 parrot-poke I'll look up C6PAN stuff, thanks for the pointer ... just soaking up ideas for now
18:47 Tene I doubt any hard decisions *will* be made.  The only decision I anticipate seeing is someone choosing an idea to try implementing.
18:48 pmichaud I think we're still waiting for at least one more robust implementation
18:48 Tene Nobody who is seen as having any sort of authority seems inclined to mandate anything of this nature.
18:48 parrot-poke Yes, theory is no good for this, it needs to work.  Rakudo is quickly reaching that capacity, but I wouldn't want to implement and try to keep something like that as things are right at this moment
18:48 Tene Was cpan itself created by authority?
18:48 pmichaud I think it grew out of the community
18:49 pmichaud my impression (and I could be wrong here) is that the design team is explicitly not trying to be authoritative about a perl 6 cpan
18:49 parrot-poke excellent
18:49 Tene Exactly what I'm saying.
18:51 TimToady except insofar as S11 mandates particular meta-behavior
18:53 pmichaud correct.
18:53 purl no, it's not!
19:03 IllvilJa joined #parrot
19:06 ewilhelm particle: last I looked, the perl5 config still had troubles cross-compiling
19:06 * AndyAway is listening to Allison talking to a MS guy
19:07 particle ewilhelm: parrot config has trouble with that, too :(
19:07 * ewilhelm sees zero mention of 'cross' in the docs or configure bits
19:08 ewilhelm there was some talk on p5p a while back about it, but I'm not sure where that got us
19:11 particle you mean the perl 5 docs/config bits?
19:12 ewilhelm no, I meant "discussion of improving cross-compile support for perl5"
19:12 x joined #parrot
19:12 ewilhelm in particular, I would like to be able to build win32 perl from linux
19:13 particle can gcc or some other compiler do that?
19:13 ewilhelm or really, I would like to build anything that needs to be built for win32 from *not win32*
19:13 ewilhelm yeah, mingw32 runs on linux
19:14 particle ah
19:14 ewilhelm and ships in debian since at least etch
19:16 jq- joined #parrot
19:16 Tene Coke: If you give me tickets, I can look at them while I'm giving an exam tomorrow.
19:16 ewilhelm so most apps with autoconf can be built for win32 by passing ./configure '--host' and '--build' arguments
19:16 Coke Tene: muahahah
19:17 Coke tene; do you fear C?
19:17 barney joined #parrot
19:17 Coke (if I gave you a ticket regarding some PMC stuff in one of the languages, for example....)
19:17 ewilhelm Tene: re perl6pan packages: http://www.perlfoundation.org/pe​rl6/index.cgi?distributionformat
19:17 shorten ewilhelm's url is at http://xrl.us/bgwgt
19:17 Coke extjs++ # mmmm, shiny.
19:18 ewilhelm or was that parrot-poke asking about that?
19:19 parrot-poke yes, thank you I was asking about packages and was reading the old Perl6 FAQ which was kind of a trippy document I must say
19:19 Tene My C is a bit rusty, but not too much.  I'd be glad to at least take a look.
19:20 ewilhelm parrot-poke: most of the people responsible for the workings of cpan and the various perl5 installers are on #toolchain
19:21 ewilhelm we all have lots to say about what's wrong with the current stuff and where we've gotten backed into a corner by various bits of interacting reverse compatibility :-D
19:21 Tene Coke: I'd like C tickets.  I need more stuff that I need to work to do.
19:22 particle tene: you could help us complete some conversion in the pdd17pmc branch
19:22 Coke tene;ok. gave you one for APL's pmc that *should* be straightforward, hoisting up constant strings into statics.
19:22 Tene particle: Is there documentation on what that means?  If not, can you produce some for me?
19:24 particle svn co http://svn.perl.org/parrot/branches/pdd17pmc pdd17pmc && gvim docs/pdds/pdd17_pmc.pod # also see  http://perlsix.org/svn/parrot/revision/?rev=r25862 and http://perlsix.org/svn/parrot/revision/?rev=r25832
19:24 particle the remainder of src/pmc/*.pmc needs conversion to the style in those commits
19:25 Tene particle: send email to tene@allalone.org saying exactly that?
19:25 particle ewilhelm: i'm definitely interested in improving configure.pl. i wish it was paid work, i'd have at it. you know how that goes, i'm sure.
19:25 particle tene: roger, wilco
19:25 Tene Fantastic.
19:34 peeps[work] joined #parrot
20:07 schmalbe joined #parrot
20:30 svnbotl r26125 | bernhard++ | trunk:
20:30 svnbotl : Set svn properties.
20:30 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26125
20:42 schmalbe coke: Is rt42679 still depecrecated? The note was removed from DEPRECATED.pod in r23067
20:42 marmic joined #parrot
20:45 Coke 42679 doesn't appear to be in the rt queue.
20:45 Coke (for parrot)
20:47 schmalbe rt42769 it is, deprecated synonyms for '.local pmc'
20:51 cotto_ why is get_iter called during PMC initialization?
20:52 Coke schmalbe: looks like it was never fixed.
20:52 Coke I just tried ".local Array foo" and it worked.
20:52 * Coke checks 23067
20:54 Coke yah. I think I removed that note; but I seem to have done so in error.
20:55 Coke someone should go through and verify that all the tickets that say [DEPRECATED] are listed in the pod.
20:57 schmalbe I'll add a TODO-test for rt42769 in t/compilers/syn/errors.t
20:58 Coke danke.
20:58 Coke schmalbe?
20:58 Coke (barney?)
21:02 schmalbe yes, barney
21:05 Coke ~~ ^_^
21:07 kj joined #parrot
21:11 mire__ joined #parrot
21:17 svnbotl r26126 | kjs++ | trunk:
21:17 svnbotl : [docs] add a todo-review note to pct doc; I think there's something not correct there.
21:17 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26126
21:20 x joined #parrot
21:21 wknight8111 joined #parrot
21:24 svnbotl r26127 | bernhard++ | trunk:
21:24 svnbotl : #42769: Remove unneeded type names from PIR
21:24 svnbotl : Add two tests, one succeeding and the the other TODO.
21:24 svnbotl : Put deprecation note back into DEPRECATED.pod
21:24 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26127
21:32 particle now... what error was it that pmichaud was looking for to provide a better message?
21:35 particle ah, too few arguments.
21:37 Coke at the general parrot level?
21:37 Coke or somewhere in PCT?
21:37 particle parrot level
21:37 particle src/inter_call.c:1025
21:37 Coke there's a ticket for that or something like it.
21:38 Coke ah, yes, you gave me some pointers on it. =-) Feel free to steal it, I haven't had time to get far.
21:38 particle i'll look in your queue
21:42 particle hrmm... the function is in the backtrace
21:43 particle too few arguments passed (0) - 1 params expected
21:43 particle current instr.: 'foo' pc 12 (..\t-too-few.pir:5)
21:43 particle called from Sub 'main' pc 7 (..\t-too-few.pir:2)
21:43 pmichaud ah, yes
21:43 pmichaud apparently that wasn't pasted
21:43 pmichaud in yesterday's discussion
21:43 pmichaud although, it may be that the call to the PIR compiler wouldn't produce a backtrace
21:43 particle it may, or may not, be worth putting it in the error, too
21:43 particle right
21:43 pmichaud (since that was ultimately the issue)
21:45 particle i'd like to see "too few arguments: passed 0, expected 1 in sub 'foo'"
21:47 particle hrmm, that error code is wrong anyway, wrt optional params
21:48 silug joined #parrot
22:03 peeps[work] joined #parrot
22:14 svnbotl r26128 | bernhard++ | trunk:
22:14 svnbotl : #42769: Remove unneeded type names from PIR
22:14 svnbotl : Replace a couple of builtin type names with 'pmc'.
22:14 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26128
22:15 wknight8111 joined #parrot
22:15 particle int
22:15 particle Parrot_store_arg(SHIM_INTERP, ARGMOD(call_state *st))
22:15 particle {
22:15 particle return store_current_arg(st);
22:15 particle }
22:15 particle i'm sorry, what?!?!?
22:16 particle it just so happens store_current_arg is a static function in the same file
22:16 Coke what's the wtf? that it says ARGMOD?
22:17 particle that it's simply a tailcall dropping the unused interp, adding a Parrot_, and calling a static function
22:17 particle why not just rename the static func, make it non-static, and use it everywhere?
22:18 pmichaud perhaps the static func is called from elsewhere?
22:18 particle static functions can only be called in the file in which they're declared
22:18 pmichaud perhaps the static func is called from elsewhere (in the file)
22:18 particle yep, it's called in one place
22:19 pmichaud in one other place, besides Parrot_store_arg?
22:19 particle yes. one other place
22:19 particle src/inter_call.c
22:19 pmichaud well, there ya go then :-)
22:19 particle but the place where it's called has a valid interp
22:19 particle there's no use having two names for the same thing.
22:22 * particle fixes & rebuilds
22:49 davidfetter joined #parrot
23:00 svnbotl r26129 | particle++ | trunk:
23:00 svnbotl : [src] remove duplicated function
23:00 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26129
23:07 purl joined #parrot
23:24 dalek joined #parrot
23:24 wolverian joined #parrot
23:36 svnbotl r26130 | jonathan++ | trunk:
23:36 svnbotl : [rakudo] Start to implement pairs. Add grammar rules and partly implement actions. Tweak Pair PMC so it's usable in Perl 6. Still lots to do, and this doesn't work with calls yet, so don't expect to do anything especially useful with it right now. :-)
23:36 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=26130

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

Parrot | source cross referenced