Camelia, the Perl 6 bug

IRC log for #parrot, 2008-02-21

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:12 arcady joined #parrot
01:00 Coke pmichaud: how goes the release?
01:01 pmichaud coke: took a short break, I'll restart in an hour or two
01:03 pmichaud we're hoping to catch a glimpse of the eclipse tonight
01:04 pmichaud but it's overcast here, so might not have much luck
01:04 pmichaud (mostly overcast, that is)
01:05 Coke ok, that's fine. just wondering if I should hold off on some stuff.
01:12 Coke bah. tell me again how I can easily find out when a module was added to perl core?
01:13 lathos There's a module, something corelist. Or you can ask the CPAN bot
01:14 lathos Which appears to have gone away.
01:16 Coke lathos: awesome, thank you.
01:19 Coke lathos++
01:20 Andy joined #parrot
01:20 Coke ... we have a lot of things left in lib that are in core. =-)
01:33 Coke we could rip out 5 more modules from lib based on what's in core as of 5.8.0
01:42 rdice joined #parrot
01:57 kid51 joined #parrot
02:06 particl1 joined #parrot
02:18 kid51 Does anyone know whether we're going to have a release this week?
02:18 Coke kid51: yes, pmichaud is doing it tonight.
02:19 kid51 k.  I just took the ticket about the core modules.  I'll wait until after the release to delete them from distro.  Okay?
02:20 pmichaud I'm working on release now
02:20 chromatic Test::Builder will be tougher.
02:20 pmichaud why, exactly, is this release 0.6.0 ?
02:20 chromatic There's a minor API change we have to work around.
02:20 pmichaud as opposed to 0.5.x?
02:21 chromatic pmichaud, I thought the PMC changes would make it 0.6.0, but they're not in, so I thought it was 0.5.3.
02:21 kid51 No, but the ticket to which Coke posted noted that our minimum Perl 5 version is 5.8.  Hence, anything that was core in 5.8 does not need to be in distro.  Correct?
02:21 pmichaud okay.  the ticket says 0.6.0
02:21 pmichaud but I'll release it as 0.5.3
02:21 kid51 chromatic:  What about Test::Builder?
02:21 pmichaud (chromatic's understanding corresponds with mine.)
02:22 chromatic kid51, I removed it a few days ago in my local tree and there were some failures with Parrot::Test.
02:23 kid51 k.  then how about Test::Simple/More, Class::Struct, Text::Balanced?  Assuming I can get all tests to pass without them in distro, are they good to go?
02:23 chromatic I'm not sure how the Test:: ones will fare until we can clean up the Test::Builder issue.
02:23 chromatic The others should be fine.
02:23 kid51 Okay, I'll work at this in pieces over the weekend.
02:25 pmichaud chromatic: you mentioned earlier today that there were some recent patches or tickets for pge/pct?  or did I misunderstand?
02:25 pmichaud ('recent' meaning...?
02:25 chromatic The past few weeks.
02:25 chromatic http://rt.perl.org/rt3/NoAuth/parrot/Overview.html
02:25 pmichaud thanks
02:26 pmichaud that helps
02:27 chromatic I've been using it myself to clear up newer stuff and patches I missed.
02:27 * pmichaud bookmarks that very useful link
02:34 Andy Y'know, I was going to overhaul something the other day and chromatic +1'd me on it, but I'll be darned if I can remmeber what it was.
02:35 chromatic src/string.c
02:35 purl src/string.c is the builtin string, right? should I create there a method as well?
02:35 chromatic purl, forget src/string.c
02:35 purl chromatic: I forgot src/string.c
02:36 Andy yeah, and what was it...
02:36 Andy oh, and somebody handed me a ticket, too.
02:38 chromatic Pick anything in there and untangle it, and +2 to you.
02:51 Tene Okay, I've got some time before I need to sleep again, lessee if I can push some patches through.
02:51 pmichaud careful, I'm in the midst of making a release
02:52 Tene Oh, that's now.  I was thinking that that was tomorrow.
02:52 Tene I'll hold off, then.
03:00 Andy joined #parrot
03:02 Andy oh, maybe it was string_concat or somethign.
03:08 Coke pmichaud: anything we can do to help at this point? (aside from leaving you alone? =-)
03:08 svnbotl r25908 | pmichaud++ | trunk:
03:08 svnbotl : [nqp]:
03:08 svnbotl : * Fix oddity whereby we ended up with both t/17-class.t and t/17-inline.t
03:08 svnbotl :   (one of them should be 18-*.t).
03:08 svnbotl r25909 | pmichaud++ | trunk:
03:09 svnbotl : [nqp]:
03:09 svnbotl : * Add tests for method call, C<=:=> operator (RT#50520)
03:09 svnbotl : * Patch courtesy Zev Benjamin <zev-perl6@strangersgate.com>
03:09 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25909
03:14 * Coke wonders who committed a manifest patch that had removed developing.
03:15 pmichaud Coke: just tell me if you think there are any steps/items I'm likely to forget or overlook
03:15 Coke "follow the list"
03:15 pmichaud that's basically what I'm doing :-)
03:15 Coke that's pretty much the only way I remember what to do when it's my turn.
03:15 pmichaud paula needs a bit of help for about 30m, then I'm going to cut the release
03:15 Coke chromatic, allison or I can update the website.
03:16 pmichaud everyone has helped a bunch by updating NEWS, PLATFORMs, etc.
03:16 Coke no rush.
03:16 Tene Don't forget to send $10 to Coke.
03:16 * Coke spafs his tea.
03:16 pmichaud Coke, can I give you the $10 at YAPC::NA?  ;-)
03:16 pmichaud bbi30
03:16 Coke that will be fine. But as I recall, I owe you a cookie.
03:24 Coke wknight8111: ENOPATCH on #50884
03:31 chromatic joined #parrot
03:38 Andy HOLY CATS DID I FIND A BUG
03:38 chromatic I hope it's related to constant strings.
03:39 Tene IS IT A BIG BUG?
03:39 chromatic and you fix lots of COW with it
03:39 Andy no, it's in pbc_merge
03:39 Andy oh, not  a bug.
03:39 Andy But it looks like it.
03:41 Andy chromatic: what's your bummer?
03:42 chromatic Trying to fix the Lua segfaults.
03:43 chromatic It's the COW and constant mess again.
03:44 Andy this pbc_merge stuff is makin' me sad
03:44 Andy and I just realized it's not actually used.
03:47 chromatic It would be nice to use it.
03:50 Coke tcl was using it for some time.
03:50 Coke I think mdiep removed it.
04:08 AndyA joined #parrot
04:08 DarkWolf84 joined #parrot
04:11 dwolf_ joined #parrot
04:14 wknight8111 what does ENOPATCH mean?
04:15 Tene wknight8111: it means "There's no patch, and that's bad."
04:15 Tene It's a reference to unix error codes, which are represented in all capital letters and start with 'E'
04:16 wknight8111 oh shoot, i must have forgotten to attach the file
04:16 wknight8111 the argument can be made that my marbles aren't all in the right places
04:16 svnbotl r25910 | petdance++ | trunk:
04:16 svnbotl : use the mem_realloc_n_typed macro properly
04:16 svnbotl r25911 | petdance++ | trunk:
04:16 svnbotl : take advantage of str_dup
04:16 svnbotl r25912 | petdance++ | trunk:
04:21 Andy y'know, we still need to put together that list of what yet needs to be done.
04:22 Andy ./perl6 -e'"Hello world".say' doesnt' work.  Waaah.
04:22 Andy but right now, I have a cruel editor to write for.
04:25 pmichaud back again
04:27 Tene Andy: it works if you add a 'say' method to languages/perl6/src/classes/Str.pir
04:27 Tene [tene@kupo perl6]$ ./perl6 -e'"Hello world".say'
04:27 Tene Hello world
04:27 purl Hello world is http://www.roesler-ac.de/wolfram/hello.htm
04:27 pmichaud even better would be to add the 'say' method to Object
04:27 pmichaud that way   3.say  will work also
04:28 pmichaud as well as  <1 2 3>.say
04:28 * Tene tries that.
04:28 Andy Yeah, I'm all over that.
04:28 Andy I'm SO not low-level.
04:29 particle joined #parrot
04:32 Tene When I add it to Object, it works for "Hello world".say but 3.say dies.
04:32 Tene Hmm.
04:33 * Coke writes another coding standard test. :|
04:33 Coke I'll hold off committing this until the release, but if anyone wants to comment...
04:34 nopaste "coke" at 72.228.52.192 pasted "test for RT#47109" (79 lines) at http://nopaste.snit.ch/12370
04:34 Coke (which actually does catch two unwrapped macro args in the current suite.)
04:36 Coke updated the error message slightly.
04:41 Tene Oh, "3.say" doesn't work because rakudo generates pir like: $P34 = 3."say"()
04:42 pmichaud interesting
04:42 pmichaud that's definitely a PCT bug... file a ticket, please?
04:42 Tene purl: parrotbug
04:42 purl parrotbug is mailto:parrotbug@parrotcode.org or http://svn.perl.org/parrot/​trunk/docs/submissions.pod
04:45 * pmichaud needs a release name.  Something related to rakudo would be nice, since this is the release that first contains it.
04:45 Tene pmichaud: sent
04:46 pmichaud Tene++  # thanks
04:46 pmichaud an invocant always has to be a PMC
04:46 pmichaud so 3."say"() is clearly wrong
04:48 Tene Actually, looks like "Hello world".say works fine for me even without adding a method to Object.pir
04:49 Tene You sure it doesn't work, Andy?
04:49 Andy It didn't for me just now.
04:50 Andy lemme try again
04:50 Andy I did it in -e
04:50 Tene So did I.
04:50 Andy k hold on
04:50 Coke Andy: http://www.rakudo.org/cgi-bin/mt/mt-​search.cgi?tag=Rakudo&amp;blog_id=6 :: internal server error
04:50 shorten Coke's url is at http://xrl.us/bgimx
04:50 Andy yeah, I know, Coke.
04:51 Andy the tags get hammered by bots and it hoses the machine it's on and my host chmod 000s mt-search.cgi
04:51 Coke pmichaud: oumu-da.
04:51 Andy nopaste?
04:51 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
04:52 Coke (as opposed to rakudo-da)
04:52 Andy http://scsys.co.uk:8001/13384
04:52 Tene Oh, looks like I'm an idiot.
04:53 Tene I was running 'make', not 'make perl6'
04:54 Coke pmichaud: (way of the parrot. based on a google search of parrot, japanese, camel, dictionary =-)
04:55 nopaste "tene" at 71.16.144.82 pasted "say() method for rakudo" (24 lines) at http://nopaste.snit.ch/12371
04:55 pmichaud coke: oumu ?
04:55 Coke parrot.
04:56 Coke A half-hearted suggestion.
04:56 pmichaud it's good enough for me
04:56 Coke the english "way of the parrot" would work.
04:56 Coke (and be slightly more accessible.)
04:56 pmichaud oh, I like "way of the parrot" too
04:57 Coke I'm -> abed; I can do the website tomorrow before noon if necessary
04:57 pmichaud okay, np.  just watch for the announcement
04:57 pmichaud (which will come in approx 1 hr if all builds well)
05:00 Tene that patch also works for 'my $a = 3; 3.say' and '<1 2 3>.say' and everything else I tried except literal numbers.
05:00 Andy is anyone gonna play with tene's patch?
05:00 Andy Can I apply it?
05:00 pmichaud maybe wait until after I tag the release?
05:00 x joined #parrot
05:00 Andy I thought it wasn'tuntil March 18
05:01 Andy OH, the Feb release.
05:01 Andy Duh
05:01 pmichaud should be ~30 min
05:01 pmichaud maybe less
05:01 Andy k
05:03 Andy wow, I'm acutally writing some Perl 6 code.
05:03 Andy I'm so frightened.
05:10 Andy 3.say seems to be a parsing problmem.
05:10 pmichaud in rakudo, or in the PIR that gets generated?
05:11 Andy error:imcc:syntax error, unexpected FLOATC ('6.')
05:11 pmichaud ah, that's an error in the generated PIR
05:11 pmichaud likely it's the problem Tene just reported
05:15 Andy Come ooooon, TAGGING!
05:15 Andy I wnana commit my first Rakudo test
05:15 Andy !
05:16 * pmichaud hears the fan on his machine grow louder as he invokes more tests
05:17 pmichaud oh, where is the Rakudo test being added?
05:17 Andy languages/perl6
05:18 pmichaud as part of the official tests (t/spec) or just in t/00-parrot or ... ?
05:18 Andy the latter
05:18 purl the latter is better
05:19 pmichaud okay.  t/00-parrot is really meant for very basic tests just to make sure things are working enough to start running the 01-sanity tests
05:19 pmichaud but we can go ahead and put a test there for now.  it'll likely get refactored into the spec tests
05:19 Andy :q
05:20 pmichaud looks like I'm on the 3rd or 4th runcore for 'make fulltest', shouldn't be long now.
05:21 Tene Andy: that's exactly what I reported a bit before you mentioned it.
05:21 Andy ok
05:21 pmichaud it should be an easy fix in PCT, I just don't want to risk it mid-release
05:21 pmichaud I'll look at it as soon as the release is done
05:21 Tene it's a PCT problem.  it generates: 3."say"()
05:24 pmichaud found the bug
05:29 Andy hmm, what should say return?
05:29 Andy oh, bah
05:29 Andy I figured it would return something other than true.
05:31 Andy like characters printed.
05:31 Andy Alas.
05:31 Andy You'd think these would be identical, at least if you were me
05:31 Andy <ok 4>.say;
05:31 Andy say <ok 5>;
05:31 svnbotl r25914 | pmichaud++ | trunk:
05:31 svnbotl : [release]:
05:31 svnbotl : * Updates for 0.5.3 release.
05:31 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25914
05:32 pmichaud Andy:  they aren't?
05:32 Andy nope
05:33 pmichaud well, one would be 'ok 4' and the other 'ok 5'  :-)
05:33 pmichaud but I suspect you mean something other than the digit difference :-)
05:36 Andy yes
05:37 Andy you get
05:37 Andy ok 4
05:37 Andy ok5
05:37 Andy or rather, I get.
05:37 chromatic Abigail will be happy.
05:37 chromatic I suspect you want a string, not a list, quoting construct in the latter example though.
05:38 pmichaud actually, Tene's patch seems to assume string
05:38 pmichaud unless .say implies string context on its invocant, I think that $foo.say() ought to redispatch to say($foo)
05:39 chromatic Yes, but does <ok 5> create a list or a string?
05:39 pmichaud <ok 5> creates a list
05:39 chromatic Yep.
05:39 pmichaud Tene's .say patch  treats the list created by <ok 4> as a string
05:39 chromatic And the active output separator for a list in this context is the empty string.
05:40 pmichaud and a List in string context uses a space for the separator
05:40 pmichaud i.e.,  ~<ok 4>    # "ok 4"
05:40 chromatic The List in the latter case isn't in string context, is it?
05:40 pmichaud correct
05:40 chromatic That's why it prints ok5
05:40 pmichaud say <ok 4>   # "ok4"
05:40 pmichaud say ~<ok 4>   # "ok 4"
05:41 chromatic Feel better about it now, Andy?
05:41 Andy I don't get it, but oh well.
05:41 Andy I'm only picking up the new syntax as we go along.
05:41 Andy pmichaud: Did you tag?
05:42 Andy or was that 25914 just getting ready?
05:42 pmichaud <ok 4> in perl6 is like writing  qw/ok 4/ in perl 5
05:42 pmichaud 25914 is the commit just prior to tagging.  it _should_ be safe to start committing again
05:42 Andy I'll sit and wait if you like.
05:42 pmichaud I'm just waiting for make fulltest to finish on the tarball
05:43 Andy ok, so the ~ is the "stringify this?"
05:43 pmichaud actually, it's far enough through make fulltest that I'm quite sure it's safe to start committing again
05:43 * Tene sleep.
05:43 pmichaud ~ is the generic "something is a string" operator
05:43 pmichaud as an infix op, it does string concatenation
05:44 pmichaud as a prefix op, it coerces its argument to a string  (i.e. imposes string context)
05:44 chromatic Wow, the Capture PMC had bugs.
05:44 pmichaud the one in C?  it's been broken for some time.
05:44 chromatic Gonna be fixed in the PMC branch here shortly.
05:44 pmichaud excellent.  :-)
05:45 pmichaud PGE, NQP, PCT, etc. have all been using Capture_PIR as their base type for captures
05:45 Andy and "@foo" no longer interpolates in the string?
05:45 pmichaud it has to be "@foo[]"  to get the interpolation
05:45 pmichaud (but I don't think Rakudo implements "@foo[]" yet)
05:45 chromatic You might try the C version on the branch just for fun.
05:46 Andy it doesn't, no.
05:46 pmichaud I think that "{@foo}" is also valid in Perl 6 (but also is unimpl in Rakudo thus far)
05:46 pmichaud however, "@foo.join(' ')"  should work.
05:47 pmichaud (er, in Perl 6 still, not in Rakudo yet)
05:47 pmichaud we haven't done a lot with interpolation of things other than scalars
05:47 Andy ok, committed
05:47 purl The chicken is involved, but the pig is *committed*.
05:54 pmichaud hey, what happened to the tags/ branch in SVN?
05:54 svnbotl r25915 | petdance++ | trunk:
05:54 svnbotl : Added tene's patch to Object.pir to do say. Added a perl6 test to test it out.
05:54 svnbotl r25916 | chromatic++ | pdd17pmc:
05:54 svnbotl : [PMC] Converted Capture PMC to PDD17 for PMC attributes.
05:54 svnbotl r25917 | chromatic++ | pdd17pmc:
05:54 svnbotl : [lib] Minor cleanups in PMC parsing and emitting.
05:54 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25917
05:54 particle pmichaud: it's not a branch
05:54 particle it's on the same level as branch/
05:54 pmichaud okay, the tags directory, then
05:54 Andy it's parallel to branch/
05:54 pmichaud http://svn.perl.org/parrot/
05:54 pmichaud I see no "tags/" there.
05:55 Andy wow, it's not there.
05:55 particle whuh?
05:55 pmichaud THAT'S WHAT I SAID!!!
05:55 pmichaud :-)
05:55 Andy $ svn ls https://svn.perl.org/parrot/
05:55 Andy branches/
05:55 Andy trunk/
05:55 Andy well, good thing we have the CPAN!
05:55 particle well, you'd better recreate it for the release, and enter a ticket so we can figure out what happened
05:55 chromatic I don't remember it being there.
05:56 particle it was there.
05:56 pmichaud it was there as of 2008-02-01
05:56 Andy It was there in -r25000
05:56 pmichaud I'm looking to see when it disappeared
05:56 Andy and 25800
05:56 pmichaud it disappeared sometime on 2008-02-18
05:57 Andy somewhere between -r25856  and -r25863
05:57 Andy -r25861
05:58 Andy r25861 | jkeenan | 2008-02-18 20:45:01 -0600 (Mon, 18 Feb 2008) | 1 line
05:58 Andy Removing superseded tag.
05:58 Andy I think he must have been a little over-rambunctious on that. :)
05:58 particle kid51--
05:58 pmichaud I think I can restore it
05:58 svnbotl r25918 | chromatic++ | pdd17pmc:
05:58 svnbotl : [PMC] Used a defined typename instead of referring to the struct directly.
05:58 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25918
05:58 pmichaud svn copy -r 25860 https://svn.perl.org/parrot/tags https://svn.perl.org/parrot/tags  # maybe?
05:59 particle svn up -r 25861 tags
05:59 particle yah, that may work better
05:59 spinclad [nit: rakudo is rakuda-do in full; way of parrot would be oumu-do  (not that i own much skill at japanese)]
05:59 particle ok, bedtime. pmichaud++ for the release (in advance)
05:59 Andy you guys have a ball, I'm goin' to bed.
05:59 Andy oh and mail me the release note please
05:59 pmichaud okay, will do.
06:00 spinclad 'way of the parrot'++
06:00 jdv79 joined #parrot
06:00 pmichaud yay, 'svn copy' seems to have worked.
06:01 chromatic +    print item
06:01 chromatic +    print "\n"
06:01 chromatic Long way for writing "say item"
06:01 pmichaud right, those are incorrect.
06:01 pmichaud even "say item" is incorrect.
06:01 chromatic I know, but I like the irony.
06:01 pmichaud it needs to be 'say'(self)
06:04 pmichaud and, out of curiosity, what happens with  C<  $foo.say 3, 4, 5;  >  ?
06:04 svnbotl r25919 | pmichaud++ | trunk:
06:04 svnbotl : See if we can restore the tags/ subdirectory of the repository,
06:04 svnbotl : accidentally removed in r25861.
06:04 svnbotl r25920 | pmichaud++ | trunk:
06:04 svnbotl : tagged release 0.5.3
06:05 chromatic There's an arity question there....
06:06 pmichaud or perhaps I really mean   $foo.say(3,4,5) .... is .say assumed to always be zero-arity (excluding the invocant)?
06:06 pmichaud that would be my guess here
06:07 chromatic Ditto.
06:07 chromatic Unless there's a multi.
06:10 jdv79 joined #parrot
06:17 svnbotl r25922 | pmichaud++ | trunk:
06:17 svnbotl : [perl6]:
06:17 svnbotl : * .say method should redispatch to say() function. instead of
06:17 svnbotl :   printing directly.
06:17 svnbotl r25923 | petergibbs++ | trunk:
06:17 svnbotl : Set svn properties to keep file_metadata.t happy
06:17 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25923
06:18 pmichaud uh oh
06:18 pmichaud rgrjr didn't change ownership of some of the Parrot:: modules
06:24 svnbotl r25924 | chromatic++ | pdd17pmc:
06:24 svnbotl : [PMC] Converted NameSpace PMC to PDD 17 attributes.
06:24 svnbotl : This PMC could use more cleanup.
06:24 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25924
06:37 svnbotl r25925 | petdance++ | trunk:
06:37 svnbotl : added some new sneaky ways to invoke say, but ok 4 started failing on me all of a sudden.
06:37 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25925
06:44 svnbotl r25926 | chromatic++ | pdd17pmc:
06:44 svnbotl : [PMC] Migrated NCI PMC to PDD 17 for attributes.
06:44 svnbotl : Using the type PMC * instead of void * for the function pointer is a nasty
06:44 svnbotl : hack, but PDD 17 doesn't (yet) support void * as an attribute.
06:44 svnbotl : For now this only requires some dodgy type casting, but if and when the PMC
06:44 svnbotl : parser handles mark() and destroy() automatically, treating a function pointer
06:44 svnbotl : as a PMC is only asking for trouble.
06:44 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25926
06:51 svnbotl r25927 | chromatic++ | pdd17pmc:
06:51 svnbotl : [PMC] Fixed two line length problems.  Oops.
06:51 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25927
07:08 uniejo joined #parrot
07:57 iblechbot joined #parrot
08:05 svnbotl r25928 | fperrad++ | trunk:
08:05 svnbotl : [Lua]
08:05 svnbotl : aligned with Lua 5.1.3 (part 3)
08:05 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25928
08:38 ewilhelm joined #parrot
08:40 HG` joined #parrot
08:44 IllvilJa joined #parrot
09:56 mire__ joined #parrot
10:04 davidfetter joined #parrot
10:20 davidfetter hello
10:20 purl privet, davidfetter.
10:20 davidfetter anybody going to FOSDEM?
11:12 particl1 joined #parrot
11:30 wknight-away joined #parrot
11:53 particle joined #parrot
11:55 rdice joined #parrot
12:28 pmichaud davidfetter: I'm going to FOSDEM  (I'm presenting there)
12:35 pmichaud afk # making breakfast for kids
12:37 UltraDM joined #parrot
12:50 kid51 joined #parrot
13:28 cognominal_ joined #parrot
13:39 svnbotl r25929 | petergibbs++ | trunk:
13:39 svnbotl : Fix typo in r25897 i.e. changed INTERP to interp.
13:39 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25929
13:46 diakopte1 23:59:47 <@pmichaud> okay, will do.
13:46 diakopte1 ugh; accidental paste
13:46 Coke joined #parrot
13:46 * Coke ~~
13:49 pmichaud good morning
13:49 purl morning jackass
13:49 pmichaud purl, I think you have me confused with my brother.
13:49 purl pmichaud: i'm not following you...
13:51 pmichaud arguing with a bot is like wrestling with a pig -- you still get dirty.  And the pig likes it.
13:51 particle heh
13:51 Coke pmichaud++ fixing tags.
13:52 Coke (svn++ cheap copies)
13:52 pmichaud Coke: need a copy of the announcement?
13:52 particle please let kid51 know what happened, so he can be careful not to do it again
13:52 pmichaud I'm still not sure what happened.  :-)
13:52 Coke I'm guess he did an svn del of tags/ instead of tags/some_tag
13:53 pmichaud normally we don't delete things in tags/, though
13:53 Coke ... he does all the time, sir.
13:53 particle kid51 uses tags to mark his branches all the time
13:53 pmichaud *branches*, yes.  tags/
13:53 pmichaud ahhhh
13:53 pmichaud okay
13:53 pmichaud he should mark his branches in branches, then
13:53 pmichaud (imo)
13:53 Coke it's not a bad system. beats manually keeping track of where you branched from.
13:54 Coke er, s/branched from/merged last/
13:54 Coke If you can come up with a better system, by all means. Honestly, I don't think one accidental rm is the end of the world; we are using version control. =-)
13:54 pmichaud anyway, I'm likely to be afk most of today (on a plane), so maybe someone else can mention it.
13:54 Coke but yes, someone should mention it. ... ok. I'll volunteer.
13:54 pmichaud I agree, it's not the end of the world -- I wasn't even going to say anything about it.
13:55 pmichaud anyone have the ability to change module ownerships in PAUSE?
13:55 particle the module owner.
13:55 pmichaud I haven't been able to index the release tarball in CPAN because rgrjr owns three of the modules.
13:56 Coke pmichaud: one of us will track him down.
13:56 pmichaud (I sent him email last night but no response yet)
13:56 Coke ah.
13:56 Coke well, then, he'll get to when he does. it's not a showstopper.
13:56 pmichaud well, I'm *hoping* to have some form of network access at various points throughout the day.
14:03 gryphon joined #parrot
14:09 Andy joined #parrot
14:22 svnbotl r25930 | petergibbs++ | trunk:
14:22 svnbotl : In tonumber, wait until we have finished with the C string before releasing it.
14:22 svnbotl : This significantly improves the 'make test' results for lua.
14:22 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25930
14:24 Coke petergibbs?
14:28 particle coke++ for the c macro args test
14:29 particle i believe i have a ticket for that effort. oh and the copyright is wrong in the nopaste
14:42 kj joined #parrot
14:44 rdice joined #parrot
14:46 Coke particle: yes, I claimed the ticket.
14:46 Coke the copyright'll be updated before commit.
14:47 particle my one concern is that the test should not force extra unneeded parens
14:47 particle however, i don't know if that'll happen
14:48 particle we have so many c scanning code snippets. we should really have a c scanning library instead
14:49 Coke particle: so far the test is catching two things that are actually broken, but the RE I used is naive.
14:50 kj Coke: there are some tickets that seem to be forgotten, and imho, could be closed. Shall I send a mail to list describing them, so that others could have a look and comment on them?
14:50 particle yeah. right.
14:50 particle it'd be nice to use languages/c99/ to parse our c
14:50 Coke kj: sure. You could also reply to the individual tickets.
14:51 kj Coke: yes that's what happens most of the time with these tickets... that's the whole point. There's no feedback.
14:51 kj i only found 5 or so; but at least it's a start to clean up a bit
14:52 Coke the only difference doing it outside of RT is tha there's then no record of this meta-conversation about the ticket. which is, IMO, not a good difference.
14:52 kj yes, you're right. I'll reply on the tickets.
14:53 Coke (cc the list, of course. =-)
14:53 kj aye
14:54 kj particle: shouldn't we use c89 to parse parrot's c?
14:54 particle well... we should enforce c89 standards
14:54 particle we need a c parser
14:55 kj right. but wouldn't it be better to implement a c89 parser?
14:55 kj instead of a c99?
14:57 particle a c parser with 89/99/etc modes sounds more flexible
14:57 particle but anything that works is what we need now
15:03 svnbotl r25931 | petergibbs++ | trunk:
15:03 svnbotl : luastring.pmc: The %S shortcut for %Ss fails at the end of the format string
15:03 svnbotl : os.t: Correct expected output for year 1000 to match lua's output
15:03 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25931
15:04 pmichaud off to the airport -- bbl
15:28 svnbotl r25932 | coke++ | trunk:
15:28 svnbotl : [codingstd]
15:28 svnbotl : Add a (naive?) test to check that args in macro definitions are wrapped.
15:28 svnbotl : RT #47109
15:28 svnbotl : There are currently 2 failures when running this test
15:30 Coke hurm. that test is wrong. I didn't check the error messages; those failures are near unwrapped args, though. damnit.
15:53 svnbotl r25933 | particle++ | trunk:
15:53 svnbotl : [t] ws fixup
15:53 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25933
15:57 Coke particle: ping.
15:57 Coke my Re was more than naive. =-)
15:57 particle pong
15:57 particle yep, seems pretty darned broken :)
15:57 Coke I have a better one now, but it's still not quite right.
15:57 particle commit early!
15:57 Coke I'm getting 56 failures atm.
15:58 nopaste "coke" at 72.228.52.192 pasted "small patch, better results" (13 lines) at http://nopaste.snit.ch/12377
15:58 Coke But I'm still missing a few failures.
15:58 Coke (at least)
16:00 particle yep. i say commit it.
16:00 particle we need line numbers, too
16:01 particle actually, we don't
16:01 particle macro name should be enough
16:01 AndyAway So what was the bummer with 6.say?
16:02 AndyAway I can totally see why the parser would be confused.
16:04 * particle doesn't know what's up with 6.say yet
16:04 Andy I can't see how we could reasonably expect the parser to be happy w/that.
16:05 particle why not? everything in perl 6 is an object
16:05 particle say is a method.
16:05 Andy and because "s" isn't a digit, parser should know that the . isn't a decimal, but a method caller
16:06 kismet Hi folks.  Friendly neighborhood oper here.  I was wondering if it was possible to get the owner of svnbot1 to add just a bit of throttle on how fast that bot spits lines to irc.  the bot keeps accidently setting off network flood alerts.
16:06 kismet Obviously it's not a real flooder.
16:06 particle kismet: i thought we did add throttling. i'll look into it
16:06 particle what's the limit?
16:06 purl the limit is, like, very implementaiton specific
16:07 kismet I think it's over 4 lines in 3 seconds or so
16:07 particle wow. that's low.
16:07 particle ok
16:07 kismet well, just limit it a bit, and we'll see if that helps.  More than happy to work with you.
16:07 particle roger, wilco.
16:08 kismet thanks.  I'll go back to idling now. :)
16:22 Theory joined #parrot
16:35 Coke particle: committed, then.
16:35 bphillips joined #parrot
16:37 svnbotl r25934 | coke++ | trunk:
16:37 svnbotl : [codingstd]
16:37 svnbotl : A definite improvement to the regular expression checking for unwrapped
16:37 svnbotl : macro args.
16:37 svnbotl : Still not perfect.
16:37 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25934
16:43 svnbotl r25935 | petdance++ | trunk:
16:43 svnbotl : fix the "say <ok 4>"
16:43 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25935
16:46 Coke purl, forget the limit
16:46 purl Coke: I forgot limit
16:59 DarkWolf84 joined #parrot
17:02 BinGOs joined #parrot
17:02 BinGOs left #parrot
17:05 jhorwitz joined #parrot
17:17 sjansen joined #parrot
17:24 pmichaud the problem with 6.say is that PCT keeps the '6' as an integer constant, instead of making it into a PMC
17:24 pmichaud thus it generates PIR like  6.'say'()
17:24 pmichaud this is rt#51052
17:26 pmichaud the signature used to generate the children nodes in PAST::Compiler needs to take into account the difference between 'call' and 'callmethod' and make sure that the invocant is a PMC
17:27 pmichaud I'll try to fix it from the airport (while eating lunch)
17:30 Coke pmichaud: this is something I struggled with in tcl; I pretty much never get to use the non PMC registers. :|
17:30 pmichaud well, in this case it's not a problem of non-PMC registers -- it's actually generating a constant
17:30 pmichaud i.e., the PIR literally reads   6.'say'()
17:31 pmichaud however, PCT is smart enough to be able to know when it has to convert things to a PMC, and when it can leave them as a constant or register
17:31 Coke constant is no different than $I0 there.
17:31 Coke (PCT smart enough) ah, that's cool.
17:31 pmichaud I just have the signature wrong for callmethod, which needs to force the invocant into a PMC
17:32 Coke Enjoy your trip.
17:32 pmichaud plane doesn't leave for another 90 mins.  right now I'm having fish and chips
17:34 Coke -> errans
17:34 Coke -> errands, even.
17:40 svnbotl r25936 | pmichaud++ | trunk:
17:40 svnbotl : [pct]:
17:40 svnbotl : * Fix 'callmethod' nodes so that they coerce the invocant into a
17:40 svnbotl :   PMC instead of sometimes using a constant.
17:40 svnbotl : * Resolves RT#51052, reported by Tene++ and Andy++ .
17:40 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25936
17:42 nnunley joined #parrot
17:43 Tene pmichaud++
17:43 particle there's a test that can be changed now, andy
17:44 particle tene: btw we use [rakudo] instead of [perl6] now in bug reports, and they can be submitted to rakudobug@perl.org
17:44 particle that's all pretty new. i still screw it up.
17:50 Tene Thanks.
17:51 particle actually, if submitted to rakudobug@, you don't even need [rakudo]
17:52 particle pmichaud: any specific hacking plans on the planes?
17:53 Andy test committed
17:56 kj woohoo! c99 parses a simple struct :-)
17:56 svnbotl r25937 | particle++ | trunk:
17:56 svnbotl : [include] ws cleanup
17:56 svnbotl r25938 | petdance++ | trunk:
17:56 svnbotl : "6.say"
17:56 svnbotl :  now works
17:56 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25938
17:57 pmichaud particle:  was mostly going to clean up stuff.  perhaps play with .HLL as I've been meaning to for a while.
17:57 pmichaud I might also catch up on some sleep
17:57 pmichaud and save the hacking for the ord->bru flight
17:57 particle (hacking sleep)++
17:57 Andy I'm inordinately tickled that I wrote a test that turned up a bug.
17:57 pmichaud (tests that expose bugs)++
17:58 particle andy: tickle yourself more often. write more tests!
17:58 pmichaud (self tickling)++
17:58 pmichaud er...hrm.
17:58 rdice joined #parrot
17:58 Andy I just don't know what's been done yet.
17:58 Andy which is why I wanna get the list together of "This is done, this isn't/"
17:59 particle that list is in the pugs repo. we call it t/
17:59 pmichaud actually, t/spec/
17:59 particle well, eventually it'll be t/spec/
17:59 pmichaud oh, I might also play with listop parsing
17:59 Andy I don't have a pugs repo
17:59 particle but there's some conversion yet to do
17:59 particle our testing Great Need is converting tests to t/spec/
18:00 particle s/converting/migrating/
18:00 kj are there people around who are very much attached to the current c99 implementation in languages/?
18:01 particle svn co http://svn.pugscode.org/pugs/t/ pugs-tests
18:01 particle kj: nobody.
18:01 kj particle: I started a clean implementation using the mk_language_shell script
18:01 kj and I start to get things running now
18:01 particle excellent!
18:01 purl Smithers, release the hounds!
18:02 kj the current implementation is not working at all. so I guess I can overwrite it?
18:06 particle yes, please
18:07 particle or if you want, create languages/c89/
18:07 particle c99 was initially created with the ability to parse c++ headers in mind
18:07 Andy so t/spec is in the Pugs repo?
18:07 particle andy: yes
18:07 particle and rakudo's 'make spectest' does svn co
18:07 pmichaud although doing 'make spectest' in languages/perl6/ (parrot repo) will also check out a copy of t/spec
18:08 pmichaud maybe I'll review tests on the plane
18:08 kj particle: but it's based on a reference for c99...
18:08 pmichaud although sleep is starting to seem far more likely
18:08 particle kj: okay, then.
18:08 pmichaud anyway, time to head to the gate
18:08 particle safe trip!
18:09 * particle will likely hack perl6doc more, and migrate some tests to t/spec/
18:09 Andy Surprised there's no branches/trunk in pugs
18:09 particle andy: yes, it's a problem with that repo, imo
18:21 * spinclad 2
18:21 Theory joined #parrot
18:41 jhorwitz particle: were you ever able to match the contents of a POD block?
18:41 particle jhorwitz: yes.
18:41 jhorwitz particle++
18:41 * jhorwitz checks his working copy...
18:42 nopaste "particle" at 24.19.3.148 pasted "a brief example of perl6doc" (89 lines) at http://nopaste.snit.ch/12378
18:43 particle that's with a small local patch
18:44 particle there, patch committed
18:45 particle now the bodies can be reformatted to squash whitespace
18:45 svnbotl r25939 | particle++ | trunk:
18:45 svnbotl : [rakudo perl6doc]capture pod bodies, and process pod options
18:45 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25939
18:46 particle and of course the actions are wrong now, as you can't emit html etc
18:46 jhorwitz so $<body> contains the meaty goodness?
18:46 particle indeed
18:46 * jhorwitz is off to play
18:46 particle --target=parse is your friend
18:47 jhorwitz don't i know it...
18:47 particle ../../../../parrot perl6doc.pbc --target=parse foo # generates and displays the parse tree from file foo
18:48 particle --target=[parse|past|post|pir]
18:48 jhorwitz PCT is the greatest thing since sliced arrays.  ;-)
18:57 Ron joined #parrot
19:02 kj mm. c's grammar is not as straightforward as I thought :-/
19:02 destiney joined #parrot
19:03 svnbotl r25940 | petergibbs++ | trunk:
19:03 svnbotl : Add tag to newly-added struct in line with coding standards.
19:03 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25940
19:22 svnbotl r25941 | kjs++ | trunk:
19:22 svnbotl : [c99]
19:22 svnbotl : add new action/grammar for c99.
19:22 svnbotl : update MANIFEST
19:22 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25941
19:29 svnbotl r25942 | kjs++ | trunk:
19:29 svnbotl : [c99]
19:29 svnbotl : -update main file
19:29 svnbotl : -add builtins file and directory
19:29 svnbotl : -update MANIFEST
19:36 svnbotl r25944 | kjs++ | trunk:
19:36 svnbotl : [c99]
19:36 svnbotl : fix root.in; I forgot to change a few "c"s into c99.
19:36 svnbotl r25945 | kjs++ | trunk:
19:36 svnbotl : [c99]
19:41 mire__ joined #parrot
19:48 HG` joined #parrot
19:53 kj particle: what was your idea with the c99 language? what kind of back-end, one that checks for Parrot coding standards?
19:53 particle kj: there's a few things we want to do. one is coding standards. another is generating nci stubs from headers.
19:54 svnbotl r25946 | kjs++ | trunk:
19:54 svnbotl : [c99]
19:54 svnbotl : + finish operator table
19:54 svnbotl : + finish some other rules.
19:54 svnbotl : + grammar mostly complete now. Needs testing and refactoring a bit.
19:54 particle C::Critic ala Perl::Critic would be great
19:55 kj mm .ok. I don't know what the latter does, but I'm sure it can be done :-)
19:55 * particle is currently working on a yaml dumper for past
19:55 kj at least languages/c99 is building now, which is better than when I started :-)
19:55 * jonathan has made it to Ukraine.
19:55 particle yes, definitely better :)
19:56 kj I'm hoping for  <%hash> to be working sometime in PGE
19:57 kj that would fix one ambiguity in C's grammar.
19:57 jonathan Oh, interesting. I have a half-finished patch that attempts to implement indirect method calls...
19:57 * jonathan wonders when he coded this
20:04 Coke jonathan: it's from the future.
20:10 jonathan Can the future send me more half-complete patches, so I only have to do half the work?
20:10 particle you'll have to wait and see.
20:11 jonathan On PAST::Var nodes, you can only set :name to a string constant?
20:12 jonathan How will we support indirect (dynamic) lookups?
20:12 svnbotl r25947 | kjs++ | trunk:
20:12 svnbotl : [c99]
20:12 svnbotl : It's important to set the precedence of "term:" special rule tighter than the tightest operator; otherwise this won't work. Note to self: add this to operator table guide.
20:12 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25947
20:12 jonathan Should PAST::Var with no name take a first child which gives the name>
20:12 jonathan s/'>'/'?'/
20:13 kj jonathan: that would make the :name attribute rather special. Of course, one could argue that the name is the most important thing of a Var
20:13 kj but there's also a :scope attribute, that is obligatory.
20:14 jonathan Yeah.
20:14 jonathan Hmm.
20:14 jonathan Not sure what a Good Solution is. :-)
20:14 kj IMHO, it's not really the same as a PAST::Op :pasttype('call'), where the first child is the sub to be invoked
20:14 jonathan Sure.
20:14 kj but what do I know :-)
20:15 jonathan Would be good to have some way to do this.
20:15 kj OTOH, if it solves your problem
20:16 kj it's no longer an exception, is just The Way To Do It (whatever you're doing)
20:16 kj I mean, if it's a common language construct
20:16 kj where common is used by at least one language :-)
20:16 kj s/is/means/
20:16 chromatic joined #parrot
20:17 kj afk for a bit
20:19 PacoLinux joined #parrot
20:20 cout joined #parrot
20:23 sjansen joined #parrot
20:44 Ron_ joined #parrot
20:51 davidfetter joined #parrot
20:56 Coke there. i think the c_macro_args tests is now usable.
20:59 * Coke finally finds U+236A, and asks larry for a patch to APL on Parrot.
21:00 Coke crap, ww, ws, ...
21:00 svnbotl r25948 | coke++ | trunk:
21:00 svnbotl : [codingstd]
21:00 svnbotl : Instead of writing a single RE that catches the bad case while ignoring the
21:00 svnbotl : good cases, remove all the good cases first, and whatever's left is broken.
21:00 svnbotl : This brings us to a more reasonable and accurate 50 incorrect #defines.
21:01 TimToady Coke: metathesis is one of the few prohibited operations on Unicode characters :)
21:07 svnbotl r25950 | kjs++ | trunk:
21:07 svnbotl : [docs]
21:07 svnbotl : add a note about is tighter in 'term:' special rule in optable.
21:07 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25950
21:12 Coke question: in define foo(a) (*a)
21:12 Coke is the *a allowable there? or should that be: *(a) ?
21:13 particle *(a) seems righter
21:13 kj sorry, don't know...
21:16 PerlJam *a should be fine as it's just straight text substitution, but *(a) is probably "safer"
21:16 Coke ok. just wondering how smart the dumb macro catcher has to be. sounds like it's dumb enough.
21:17 particle perljam: the parens are necessary for safety
21:17 chromatic Remember how dumb the preprocessor is in C.
21:17 svnbotl r25951 | chromatic++ | trunk:
21:17 svnbotl : [Lua] Cleaned up the LuaAny PMC.  It shouldn't leak any memory now.
21:17 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25951
21:18 Coke chromatic: your mother's prepocessor is sooooo dumb...
21:18 Ron__ joined #parrot
21:21 Coke if you have define foo (a,b) (a)[b] is the b safe?
21:21 Coke or should that be (a)[(b)] ?
21:23 chromatic the latter
21:23 purl well, the latter is better
21:25 chromatic Coke, are you planning to update the website, or should I stab it?
21:27 Coke ah, crappity. please do.
21:28 Coke I had meant to do it and was distracted.
21:29 cognominal_ joined #parrot
21:30 svnbotl r25952 | chromatic++ | trunk:
21:30 svnbotl : [PMC] Minor cleanups of the LuaNil PMC.
21:30 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25952
21:34 Coke oh. hee. fixed another macro arg checker bug... bringing us to 502 errors.
21:34 particle ah, that's better.
21:34 particle are you skipping generated files?
21:34 Coke I'm just getting a list of C files.
21:35 Coke if the C file were generated by perl, we'd have no way of checking the perl code.
21:35 particle yes. some of which are generated.
21:35 Coke yes. and we should be checking them. =-)
21:35 particle right, but not fixing them inplace.
21:35 svnbotl r25953 | coke++ | trunk:
21:35 svnbotl : [codingstd]
21:35 svnbotl : ... Might be nice to warn about all the errors in a file, not just the first.
21:35 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25953
21:35 particle that's all i'm saying
21:36 particle anyway, it's looking like a much better test now
21:36 Coke actually, I think were only reporting errors if the first define if the file happened to have an error; and there were *still* 50.
21:50 Coke there. I gave up #47109 for now in case anyone wants to fix the macro errors now that there's a test to narrow down the search.
21:53 particle awesome. coke++
21:53 wknight-away joined #parrot
21:54 Coke meh. I dislike spending time on things that aren't features, but better I do it and leave the features to someone who can implement them. :|
21:56 Coke http://groups.google.com/group/bsdmailingli​st/browse_thread/thread/5182e6eeab833de4/3a​47f3b7d4e985da?hl=en&amp;q=parrot+-site%3Ap​erl.org+-site%3Aparrotcode.org+-site%3Acpan​.org+-perl.perl6+-perl.cvs#3a47f3b7d4e985da
21:56 shorten Coke's url is at http://xrl.us/bgjv7
22:17 chromatic Did FreeBSD send us a bug report?
22:18 Theory joined #parrot
22:28 chromatic joined #parrot
22:29 wknight-away What should be done about http://rt.perl.org/rt3/Publi​c/Bug/Display.html?id=46635 ?
22:29 wknight-away raising an overflow error would probably make the most sense
22:34 chromatic Silently promote to BigInt?
22:34 * davidfetter hands chromatic a copy of the source for libdwim.so ;)
22:35 particle should likely be controlled by a pragma
22:35 particle or we should have separate Integer PMCs, one that promotes, one that doesn't
22:36 particle or perhaps s/pragma/attribute/ depending on whether you want to make it scope controlled or object controlled
22:37 wknight-away see? there are lots of possibilities
22:37 wknight-away that's why i didn't want to submit a patch yet
22:38 chromatic Hey, silent truncation is the source of a great deal of suck in C and its derivatives.
22:38 particle ok, well how about this... the least surprise for today's current behavior is what?
22:38 particle overflow exception, probably.
22:39 particle so move forward with that, based on 'least surprise'
22:39 particle (re)design is definitely warranted. if you want to take a stab at designing a better solution, try that next
22:41 wknight8111 okay, i'll work on a patch that calls an overflow error
22:41 particle actually, an overflow type exception can be caught and can cause a promotion
22:42 particle so, try generating an exception object and throwing it
22:42 wknight8111 so it shall be, now and forevermore
22:44 wknight8111 what would be the best way to do that here, real_exception()?
22:46 particle see what the 'throw' opcode does
22:46 particle ...or use it.
22:48 wknight8111 okay
22:49 wknight8111 where are the opcode actions defined?
22:50 particle src\ops\*.ops
22:50 wknight8111 yeah, that's what i figured. i'll find it all from there
22:50 chromatic AHA a bug in Lua!
22:51 chromatic A GC bug no less.
22:52 particle anchoring?
22:52 svnbotl r25954 | chromatic++ | trunk:
22:52 svnbotl : [Lua] Minor cleanups to the LuaFunction PMC.
22:52 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25954
22:52 chromatic LuaThread stores a Parrot coroutine PMC in PMC_pmc_val, then sets the mark flag.
22:52 chromatic mark() marks PMC_struct_val.
22:52 chromatic Goodbye, coroutine.
22:53 chromatic That doesn't fix the strange string problem, but it fixes one problem.
23:04 chromatic (refactoring code to coding standards)++
23:04 svnbotl r25955 | chromatic++ | trunk:
23:04 svnbotl : [Lua] Fixed a GC bug in LuaThread -- mark() marked the wrong thing as live.
23:04 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25955
23:08 cotto_ Are there any autovivification tests in the Array code, because I'm not sure it does what it thinks it does.
23:14 Limbic_Region joined #parrot
23:16 particle cotto_: ResizablePMCArray tests should include autovivification
23:16 particle hrmm, seems Hash always stores values as PMCs....
23:18 cotto_ What about vanilla Array?
23:18 cotto_ it says it does in the source, but the implementation doesn't look like it would work
23:18 chromatic joined #parrot
23:18 particle i don't know anything that uses Array.
23:19 particle we recommend using Fixed* or Resizable*Array
23:19 cotto_ I was just looking at the code as an example
23:19 cotto_ Perhaps the other arrays would serve me better
23:19 particle Array PMC is *very* old.
23:20 particle and mostly deprecated, however unoffical
23:20 cotto_ now I find out
23:20 cotto_ are the others better?
23:20 particle iirc Array may have some sparse handling in it, others don't yet.
23:21 particle otherwise, the others are much better
23:21 particle ResizablePMCArray is widely used by HLLs
23:21 svnbotl r25956 | chromatic++ | trunk:
23:21 svnbotl : [Lua] Cleaned up LuaThread PMC.
23:21 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25956
23:21 cotto_ better to find out later than never
23:21 particle ResizableBooleanArray has good space-saving optimizations
23:21 cotto_ Thanks
23:21 cotto_ particle++
23:24 cotto_ premature optimization is the root of all evil
23:47 Limbic_Region salutations all
23:49 particle hey there
23:50 Limbic_Region just had to do realclean, presumably it is because of the release
23:50 particle yep
23:51 particle now... to find the magic spot in pct that does the dumping for past et al
23:52 svnbotl r25957 | particle++ | trunk:
23:52 svnbotl : [runtime] YAML/Dumper.pir library stolen and adapted from Data/Dumper.pir
23:52 svnbotl r25958 | particle++ | trunk:
23:52 svnbotl : manifesto
23:52 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25958
23:58 svnbotl r25959 | chromatic++ | trunk:
23:58 svnbotl : [Lua] Minor cleanups in LuaClosure PMC.
23:58 svnbotl diff: http://parrotvm.org/svn/parrot/revision/?rev=25959

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

Parrot | source cross referenced