Parrot 0.6.3 "Beautiful Parrot" Released | parrotcode.org/ | 5/649/88 new/open/stalled tix | logged in irclog.perlgeek.de/parrot/today
Set by moderator on 26 June 2008.
pmichaud thanks, jonathan. great work today. even if nobody else recognizes/appreciates it, I sure do. 00:00
00:09 AndyA joined 00:10 bacek_ joined 00:12 ank joined 00:25 bacek_ joined 00:42 bacek_ joined 00:43 apeiron joined 01:01 bacek_ joined 01:08 ank joined 01:09 Theory joined 01:11 Ademan joined 01:14 bacek_ joined 01:17 bacek__ joined 01:19 bacek___ joined
cotto_work home & 01:22
01:58 ank joined 02:04 bacek_ joined
dalek r28732 | coke++ | trunk: 02:06
: RT#42293
: Remove tools/docs/pod_errors.pl; we have t/doc/pod.t which is more
stupidbot RT 42293: [TODO] t/doc/pod.t vs. tools/doc/pod_errors.pl - open
dalek : accurate, and is already run in "make test"
diff: www.parrotvm.org/svn/parrot/revision?rev=28732
02:32 LordVorp joined
dalek r28733 | jkeenan++ | autojit: 02:36
: Creating autojit in svn.perl.org/parrot//branches
diff: www.parrotvm.org/svn/parrot/revision?rev=28733
r28734 | jkeenan++ | autojit-28732:
: Tagging trunk at r28732 so that the autojit can later be synched to it.
diff: www.parrotvm.org/svn/parrot/revision?rev=28734
r28735 | jkeenan++ | autoicu: 02:37
: Creating autoicu in svn.perl.org/parrot//branches
diff: www.parrotvm.org/svn/parrot/revision?rev=28735
r28736 | jkeenan++ | autoicu-28734:
: Tagging trunk at r28734 so that the autoicu can later be synched to it.
diff: www.parrotvm.org/svn/parrot/revision?rev=28736
02:42 Ademan joined 02:51 rblackwe joined
DietCoke If I want to generate a warning, what's the parrot-approved way to do that? (I'm guessing fprintf(stderr) isn't right. =-) 02:53
02:56 magnachef joined
DietCoke PIO_putps(interp, _PIO_STDERR(interp), s); 02:56
meh. someone can update my patch if they like. =-) 03:02
dalek r28737 | pmichaud++ | trunk: 03:03
: [rakudo]:
: * Add sigil listop contextualizers.
diff: www.parrotvm.org/svn/parrot/revision?rev=28737
03:04 cjfields joined, tetragon joined
DietCoke does NQP have the concept of try/catch? 03:11
pmichaud no. 03:26
we could potentially add it, however.
03:30 Theory joined, bacek__ joined
cjfields hola, pmichaud 03:30
DietCoke (email sent to list) was just asking for that test. 03:39
(which we can just write in PIR)
03:45 bacek_ joined
Coke opbots, names 03:47
opbots, names 03:51
DietCoke Infinoid, danke. =-) 03:52
Infinoid no prob. think we need more opbots
03:53 Andy joined 04:20 bacek_ joined 04:36 TonyC joined 04:44 nopaste joined 04:45 bacek_ joined 04:46 nopaste joined 04:56 cognominal left 05:00 bacek_ joined 05:03 Psyche^ joined 05:09 cognominal joined 05:21 bacek_ joined 05:32 bacek_ joined 05:43 bacek_ joined 07:00 bacek_ joined 07:06 particl1 joined 07:17 apeiron joined
dalek r28738 | fperrad++ | libs4php: 07:32
: [php] fix var_dump()
diff: www.parrotvm.org/svn/parrot/revision?rev=28738
r28739 | fperrad++ | libs4php: 07:33
: [php] add level parameter to error()
diff: www.parrotvm.org/svn/parrot/revision?rev=28739
07:41 Ademan joined 08:05 japhb joined 08:19 iblechbot joined 08:55 masak joined
spinclad pmichaud++ jonathan++ # :lexid appreciation, and cleaning up parrot's closure machinery 10:09
Coleoid cotto_home: Re the prepackaged SVN 1.4.6, I don't need 1.5.0, except insofar as upgrading to TortoiseSVN 1.5.0 has upgraded my working copies in a way not backwards compatible. So my path out is nuke working copies, downgrade TSVN, get fresh working copies. 10:25
cotto_home: Not a calamity, just a few bonus hoops to jump. :/ Thanks again for your help on this. 10:27
10:56 ruoso joined 11:28 kid51 joined 11:41 iblechbot joined 11:53 bacek_ joined 11:59 kj joined
dalek r28740 | jkeenan++ | autoicu: 12:17
: [configure] Begin to refactor auto::icu and write configuration step tests.
diff: www.parrotvm.org/svn/parrot/revision?rev=28740
kid51 purl seen tewk 12:18
purl tewk was last seen on #parrot 14 hours and 30 minutes ago, saying: I know its late, but I've just started writing production nqp/perl6 code.
12:32 bacek__ joined 13:07 jhorwitz joined 13:15 Andy joined 13:19 Whiteknight joined
dalek r28741 | Whiteknight++ | gsoc_pdd09: 13:21
: [gsoc_pdd09] update to trunk r28740. A few miscellaneous edits thrown in on the GC too.
diff: www.parrotvm.org/svn/parrot/revision?rev=28741
r28742 | Whiteknight++ | trunk: 13:22
: [core] update function-level documentation in src/headers.c, add a few XXX notes concerning questions and suggestions I have.
diff: www.parrotvm.org/svn/parrot/revision?rev=28742
13:23 iblechbot joined
dalek r28743 | fperrad++ | libs4php: 13:23
: [php] now use PhpString
diff: www.parrotvm.org/svn/parrot/revision?rev=28743
DietCoke posts another weight-related blog entry. medifast++ 13:26
dalek r28744 | Whiteknight++ | gsoc_pdd09: 13:34
: [gsoc_pdd09] added a few functions to simplify checking whether a given object is located in a particular pool.
diff: www.parrotvm.org/svn/parrot/revision?rev=28744
DietCoke hurm. is the test harness run with warnings enabled by default? 13:40
moritz Whiteknight: I checked out r28744 (latest) of your gsoc_pdd09 branch, and can't build - is that expected?
DietCoke (getting a metric #@$-ton of deprecation warnings with my local patch)
moritz afk (but will backlog) 13:41
Whiteknight moritz, it's sort of expected, yes. I had it building parrot and libparrot, but not miniparrot for a while 13:48
i've already goobered some things up, so it probably doesn't build right now
13:51 rdice joined
dalek r28745 | Whiteknight++ | gsoc_pdd09: 13:56
: [gsoc_pdd09] macroize and improve *_has_parent_pool functions, add functions to test GC header membership directly. Improve GC_IT_MARK_CHILDREN_GREY macro to use the new simple routines for object differentiation.
diff: www.parrotvm.org/svn/parrot/revision?rev=28745
r28746 | particle++ | trunk: 13:58
: [rakudo] refactor contextualizer code into a subroutine (DRY++)
diff: www.parrotvm.org/svn/parrot/revision?rev=28746
particl1 Whiteknight: you might consider tagging your branch at locations where it builds, or significant features have been added 13:59
particle so you can point people to the latest tagged version to try 14:00
Whiteknight it has never really "built", and I'm still modifying core code. I will tag things once I move on to the testing/debugging stage
particle good 'nuff
Whiteknight even if it "built" parrot, it didn't operate (segfaults everywhere), so there's no sense tagging that as something to pay attention to 14:01
DietCoke so, when can we merge this back to trunk? ^_^ 14:03
14:16 gryphon joined 14:19 jjuran joined 14:22 magnachef joined
dalek r28747 | Whiteknight++ | gsoc_pdd09: 14:27
: [gsoc_pdd09] move private macros into the code file and out of the global header file. Adjust a few macros to account for "special" nodes and doodads. Other misc changes.
diff: www.parrotvm.org/svn/parrot/revision?rev=28747
14:39 cognominal joined 14:43 peepsalot joined 14:49 jan joined 14:52 AndyA joined
dalek r28748 | Whiteknight++ | gsoc_pdd09: 15:01
: [gsoc_pdd09] mostly rearrange some comments, changing child-marking for PMC_EXT structures.
diff: www.parrotvm.org/svn/parrot/revision?rev=28748
r28749 | Whiteknight++ | gsoc_pdd09: 15:13
: [gsoc_pdd09] migrate PMC_EXT structures to be stored in the sized_pools, and to be managed normally by the GC.
diff: www.parrotvm.org/svn/parrot/revision?rev=28749
r28750 | fperrad++ | libs4php: 15:49
: [php] rewrite builtin operators
diff: www.parrotvm.org/svn/parrot/revision?rev=28750
15:57 apeiron_ joined
dalek r28751 | fperrad++ | libs4php: 15:59
: [php] move error_reporting()
diff: www.parrotvm.org/svn/parrot/revision?rev=28751
DietCoke hurls search.cpan.org/~coke/ 16:03
Whiteknight particle still here? 16:07
particle here, Whiteknight 16:09
dalek r28752 | fperrad++ | libs4php: 16:10
: [php] refactor builtin
diff: www.parrotvm.org/svn/parrot/revision?rev=28752
Whiteknight particle, did you get the message I sent you on tuesday? 16:11
16:11 slightlyoff joined
Whiteknight chromatic said you had a bug in the 64-bit stack trace code 16:11
16:12 slightlyoff left
particle ah, right 16:13
i'll try to fire up a 64-bit parrot build with head 16:14
oh, crap, i don't have a 64-bit compiler on this box
when i get my desktop up and running again, i'll get in touch 16:15
i've either got a bad stick of ram, or a bad mobo
memtest86 running now, should have some results in an hour or eight
Whiteknight okay, no rush. Since I'm working on that code anyway in my branch, I can try to track down any errors you find there 16:16
DietCoke particle is also one bad mobo. 16:18
purl okay, DietCoke.
dalek r28753 | pmichaud++ | trunk: 16:36
: [rakudo]:
: * spectest-progress update, 75 files, 1080 passing tests, 1 failing test
diff: www.parrotvm.org/svn/parrot/revision?rev=28753 16:37
pmichaud rakudo smashes through the 1000 passing tests: www.pmichaud.com/perl6/rakudo-tests...-06-27.png
16:40 Theory joined
particle yeehah 16:43
pmichaud: do you know offhand of any actions.pm code that could use refactoring? i might have some tuits for that 16:45
pmichaud not offhand, but every time I look in there I find something to refactor.
As soon as we can get Parrot lexicals working I'll do a major refactor of $_, $!, $/, and parameters. 16:46
(I might try doing it even before lexicals work, but I ran into some obstacles last time.)
A really good refactor would be to see about moving some of the List methods into Any.
(e.g., values, pairs, kv, keys, etc.)
I think I'd prefer a different name for "process_contextualizer" -- i.e., something like sigil_method 16:50
"process" is one of those words that gets overused imo
DietCoke finds that his :deprecated warnings are only happening for a very small subset of tests in 'make test' 16:51
particle yes, you're right, process_ isn't a good prefix
DietCoke particle, pmichaud : so the diet peach snapple at yapc has apparently paid off =-) 16:52
particle DietCoke: so it seems :)
if you want to try something different, the jones cola sugar-free cream soda is good 16:53
i don't drink sugar-free sodas, they give me a stomachache 16:54
rjbs Jones++ # watermelon soda 16:55
16:55 jan joined
DietCoke particle: I LOVE sugar free soda. 16:55
Jones, you say. there's a beverage mart on the way home, I'll see if they ahve it. 16:56
(stomacache) *cry*
Whiteknight anything with caffeine drives me completely nutty.
even the sugar-free stuff makes me crazy 16:57
DietCoke oh, I loves me the caffeine. 16:59
;and the splenda. and the aspartame. and the chromium. mmmmmm, sweet, sweet chromium.
particle betheny found some "sugar-free" twizzlers in new orleans this week 17:00
20 calories per twizzler 17:01
they use a sugar alcohol that metabolizes slowly
so it's good for diabetics
rjbs weird
particle they taste exactly the same, which i didn't expect
DietCoke ooooh. 17:02
PerlJam give me pure cane sugar any day :)
DietCoke I do miss the kosher for passover coke. :|
particle that's the only coke i drink 17:03
rjbs rgeoff hordes that stuff
PerlJam I sometimes think that since I like to eat meat and sugar so much that the cosmic joker will arrange things such that later in life I must watch my sugar intake and I'll be forced to become semi-vegetarian.
DietCoke diabetes, buddy. watchout! 17:05
rjbs: I used to buy cases of it when it came out.
PerlJam yeah, there is some of that on my mom's side of the family.
cotto_work seen barney 17:06
purl barney was last seen on #parrot 20 hours and 24 minutes ago, saying: k. Or put sanity tests for all PMCs in a single file.
DietCoke my mom cleverly waited until after I was diagnosed to turn up diabetic.
PerlJam What exactly is different about kosher coke from regular coke?
DietCoke PerlJam: sugar versus syrup.
rjbs less pork
PerlJam When we had a vegan working here, he said that Coke had some sort of meat-based juices in it.
DietCoke it's the closest thing you'll find to the original coke product.
rjbs there's still rat droppings in it, but the rats aren't pork-fed
DietCoke hey, these, are clean, hand fed rats ! 17:07
particle during the sugar shortage in the 70's coke changed to using corn syrup
DietCoke new coke is www.snopes.com/cokelore/newcoke.asp 17:08
PerlJam sugar-based Dr Pepper is "milder" than syrup-based Dr Pepper. I wonder if kosher Coke is the same.
DietCoke look for the yellow caps on the 2-liters around passover. 17:09
rjbs has everyone here listened to Dispepsia?
negativland.com/?opt=mailorder&...&type=
pmichaud ...really? I always had the impression that the sugar-based Dr Pepper was a bit richer. 17:13
(having regularly and recently consumed both)
rjbs when did coke classic stop saying 'classic'? 17:15
PerlJam pmichaud: richer tasting but milder in flavor (not so hard-core sugary) 17:37
maybe. 17:38
pmichaud PerlJam: yeah, I'd go with that. I can get the sugar-based Dr Pepper at Wal-Mart and Kroger here :-)
17:38 cjfields joined
pmichaud but it's really expensive 17:38
PerlJam rjbs: Why should it say "classic" when there's only one Coke now? :)
rjbs it shouldn't, but that doesn't answer my question, jerk! 17:39
DietCoke Kroger!
PerlJam rjbs: I think it stopped saying "classic" when new coke faded completely into obscurity?
DietCoke <massive Texas flashback> 17:40
rjbs I can't find date on my calendar.
DietCoke hey! I'm right here, buddy!
PerlJam rjbs: or perhaps it stopped saying "classic" when they realized there was no real market value
rjbs DietCoke: you're a real coke ZERO
In April 2007, in Canada, the name "Coca-Cola Classic" was changed back to "Coca-Cola". The word "Classic" was truncated because "New Coke" was no longer in ...
now, when in the real world...
coca-cola orange 17:41
is that like agent orange?
PerlJam rjbs: hmm. I guess you'll also need to specify a market segment. For all I know they still market "coke classic" in Mexico or China or ...
rjbs the real world, I said. the 48 states. 17:42
DietCoke recommends for anyone who goes to epcot to stop in the coke shop and try all the horrendous flavors you don't have to drink.
DietCoke is still pissed he can't get vault zero out here.
rjbs heh
SUURRRRRRGE
DietCoke is going to have to scour PA when he goes for the workshop.
(we'll have the minivan and the credit card, WATCH OUT!) 17:43
rjbs dude, when you're in PA, you drink Pepsi Scrapple
pmichaud I have to be in pretty dire straits to drink a Pepsi 17:47
PerlJam you'd have to be in worse straits to drink something called "Pepsi Scrapple" I think 17:48
pmichaud yup. 17:49
rjbs made with every part of the kola nut!
flickr.com/photos/rjbs/2394669497/
PerlJam koala nuts?!? 17:50
;>
apeiron Is that the kind of cola you step in? 17:51
jonathan thinks this was a bad channel to glance at while hungry 18:00
TimToady ooh, don't say that!
now I'm hungry...
davidfetter passes the food 18:01
pmichaud I'm hungry also, but my wife hasn't returned from the grocery store yet :-|
DietCoke ponders just getting some hot sauce. 18:02
cotto_work so I guess this would be a bad time to start singing "His Cheeseburger"
jonathan pmichaud: I've been pondering the bug you discovered before I went to sleep last night - the fix may even be as simple as re-ordering a block of conditionals in Closure's invoke... 18:03
davidfetter I CAN HAZ CHEEZBURGER?
TimToady and now, it's time for Silly Songs with Larry, the part of our show in which Larry sings a Silly Song.
pmichaud jonathan: yes, I'm hoping that's the case. I haven't been able to summon up the motivation to go look. (Part of me is afraid I'll find something else :-)
"Silly Songs with Larry" sounds like a good lightning talk.
cotto_work Friday++ 18:04
pmichaud although it could easily degrade into "The Gong Show".
I really can't believe it's Friday already. This week (month!) has just disappeared.
jonathan pmichaud: I've gotta do the eating/drinking thing now, then I have some $dayjob left to worry about, but the final part of that is just supervising some automated tools Doing Stuff, so I'll have a look then. 18:05
It may well not be that simple to, but if it is, well, it's simple. :-)
s/to/too/
cotto_work DietCoke, I'm about to buy pipp.org for Plumhead (which will soon after be renamed to Pipp) 18:07
Is this something the Parrot Foundation should be part of? 18:08
DietCoke Silly Songs++ 18:10
cotto_work: that's a good question. I would tend to say "only if you need them involved", but you could email the board.
(me, particle, chromium, allison and horwitz) 18:11
cotto_work ok. I don't foresee any issues that would require your involvement. 18:12
Thnaks.
DietCoke if you get stuck, holler!
cotto_work s/na/an/ 18:13
will do
DietCoke and see if partcl.org is available. =-0
cotto_work it is
rjbs but cpp.org is not 18:14
particle notices pmichaud said "wife" :)
DietCoke 's wife is at the beach today. :| 18:15
DietCoke wonders if there's a way to me with an immediate apostrophe. 18:16
particle nope
cotto_work rjbs, good luck getting that one from Cleveland Public Power ;)
rjbs I'll make them an offer they can refuse. 18:17
cotto_work "That's a nice power line ya got there. It'd be a shame if it were to fall over." 18:18
rjbs "Sir, that's a subsurface line."
DietCoke "damnit. why aren't I notified about these things!" 18:22
anyone have a good tool for applying patches on windows?
(as long as I'm using the GUI coolaid with tortoisecvs) 18:23
(Guess I can use the command line one with strawberry perl for now.) 18:27
particle dietcoke: you can right-click a .patch file and apply with svn 18:33
18:33 Theory joined
particle er, tortoisesvn 18:34
tortoisecvs likely has the same functionality
DietCoke nope. 18:44
might be able to do the apply from tsvn which I also have installed, though.
particle yep
avar a/w 3 18:45
19:10 tewk_ joined
nopaste "tewk" at 155.97.237.62 pasted "ncigen generated ast" (1204 lines) at nopaste.snit.ch/13419 19:17
particle looking good, tewk! 19:20
some problems, still, though
AtEOXact_SPI
<source> => "Hash[0xb674f388]"
looks like a problem in c99AST::Param action
tewk_ Structs and unions work too, but not in that dump. 19:21
DietCoke tewk++ 19:22
tewk_ I just have to write a quick unparse routine to create the signatures .pir file and I can start looking at auto generating stubs. 19:24
tyring to create a good dynamic language interface from a c .h file is probably not reasonable, (case in point swig), but ncigen will at least provide pir signatures for c functions that one can use to build good interfaces for dynamic languages. 19:26
DietCoke closes a year old ticket that particle left lying about. 19:28
cotto_work DietCoke, do you see any reason I should close rt.perl.org/rt3/Ticket/Display.html?id=46387 and rt.perl.org/rt3/Ticket/Display.html?id=46389 ? 19:36
s/should/shouldn't/ 19:38
pmichaud # Failed test 'no trailing whitespace' 19:39
# in t/codingstd/trailing_space.t at line 50.
# Trailing space or tab char found in the following files:
# /home/pmichaud/parrot/trunk/src/headers.c 1028
dalek r28754 | fperrad++ | libs4php: 19:41
: [php] getenv & sleep
diff: www.parrotvm.org/svn/parrot/revision?rev=28754
DietCoke Whiteknight: ping. 19:44
Whiteknight DietCoke, pong 19:45
oh shoot, did I screw up a whitespace? me fixey...
DietCoke cotto_work: I see no object. just make sure you remove the comment from the code.
Whiteknight: if you are able to salvage anything from the old gc branch, it's not going to to trunk and then to your branch, you'd just steal the good bits into your branch, neh? 19:46
Whiteknight Yeah, if I were going to salvage it, I would move it right into my branch, not trunk
DietCoke (by hand, presumably, since I'd phear any merges from a branch that old.)
Ok. I'm going to close that ticket, then.
Whiteknight Give the ticket to me 19:47
I'll double-check everything, and close it when I'm done with everything
cotto_work DietCoke, thanks
dalek r28755 | Whiteknight++ | trunk: 19:53
: [core] Fix trailing whitespace error I introduced earlier.
diff: www.parrotvm.org/svn/parrot/revision?rev=28755
19:58 dolmen joined, purl joined
dolmen hi purl 19:59
davidfetter HAI PURL
hrm
19:59 purl joined
dolmen hi purl 20:00
purl? 20:01
purl yes, dolmen?
dalek r28756 | coke++ | trunk:
: minor comment update
diff: www.parrotvm.org/svn/parrot/revision?rev=28756
particle tewk: you haven't checked *any* code into the gsoc_nci branch 20:02
DietCoke oh, here comes the smackdown! ^_^ 20:05
(seriously, whiteknight's branch only barely compiles, check in whatever you like. =-)
particle i need to write a mid-term evaluation this coming week
Whiteknight my branch probably doesnt even compile today, I've checked in a lot of quick garbage off-the-top-of-my-head code this morning 20:06
I realized a conceptual mistake that I need to fix before I try compiling again 20:07
I hope we aren't being graded based on the does-it-compile condition of our branches!! 20:11
dalek r28757 | Whiteknight++ | trunk:
: [core] change an unhelpful PARROT_ASSERT(0) into a more helpful real_exception.
diff: www.parrotvm.org/svn/parrot/revision?rev=28757
particle no, but "has made sufficient progress" is a prerequsite for the midterm payment 20:23
dalek r28758 | Whiteknight++ | trunk: 20:27
: [core] refactor function src/headers.c:fix_pmc_syncs
diff: www.parrotvm.org/svn/parrot/revision?rev=28758
20:31 Ademan joined 21:13 Theory joined
dolmen Order::? 21:21
Order? 21:22
purl rumour has it Order is only semi-defined
Whiteknight "has made sufficient progress"? 21:38
dalek r28759 | Whiteknight++ | trunk:
: [gsoc_pdd09] add function "object_lives" as an alternative to pobject_lives for objects which are not isomorphic with pobjects
diff: www.parrotvm.org/svn/parrot/revision?rev=28759
cotto_work no, order is <reply>I'm the one who's giving the orders around here!
purl okay, cotto_work.
Auzon order 21:39
order?
purl I'm the one who's giving the orders around here!
davidfetter HAI PURL
cotto_work Whiteknight, I don't think that diff does what you said it does. 21:40
Whiteknight yeah, it doesnt 21:41
an undo diff should be coming any moment now...
dalek r28760 | Whiteknight++ | trunk:
: [core] Ooops, last commit was wrong. Commenting out code that I shouldn't have added to trunk yet (but likely will add later).
diff: www.parrotvm.org/svn/parrot/revision?rev=28760
Whiteknight ...there we go
cotto_work happy!
Whiteknight I committed the wrong repository!
cotto_work (although an explanation of the commenting-out might prevent confusion if the code's going to stay there for long) 21:42
Whiteknight yeah, I just wanted to kill it before people started submitting bug reports 21:43
pmichaud bug! bug! bug!
Whiteknight I'm working on comments now
pmichaud :-)
21:43 mire joined
Whiteknight this is what I get for working on two projects in two different repositories at the same time 21:44
dalek r28761 | Whiteknight++ | gsoc_pdd09:
: [gsoc_pdd09] add function "object_lives" as an alternative to pobject_lives for objects which are not isomorphic with pobjects (this should actually be correct this time)
diff: www.parrotvm.org/svn/parrot/revision?rev=28761
pmichaud you just wanted the extra karmas :-)
cotto_work pmichaud++
Whiteknight I'm a karma whore like that 21:45
Whiteknight--
HA! I tried to fix it, and Purl pm'd me to say I shouldn't do that 21:46
I just can't win
cotto_work you can always go through polyglotbot 21:51
jonathan cotto_work: Hey, I'd not thought of that! 21:54
perl6: say "I think jonathan++ deserves karma!"
polyglotbot OUTPUT[I think jonathan++ deserves karma!ā¤]
21:54 teknomunk joined
jonathan won't write japh's any more...he'll write secret karma-incrasing scripts. 21:55
cotto_work you could even justify it if those scripts showed off new features you'd recently completed 22:10
dalek r28762 | pmichaud++ | trunk: 22:11
: [rakudo]:
: * Clean up handling of Array, eliminate Arrayref cheats
diff: www.parrotvm.org/svn/parrot/revision?rev=28762
jonathan cotto_work: Yeah, that a few moments ago was blatent whoring. 22:17
jonathan svn up's to play with the closure PMC
pmichaud: ping 22:35
pmichaud pong
jonathan I have a patch, that I think fixes the lexical issue you discovered.
pmichaud yay!
jonathan It passes all the Parrot tests.
pmichaud yay!
jonathan Making sure Rakudo passes all its tests too.
pmichaud does it pass the test file I sent in the ticket? 22:36
jonathan Is it's desired output:
outer foo try 1
inner foo try 1
outer foo try 2
inner foo try 2
outer foo try 3
inner foo try 3
pmichaud yes.
and since it passes all of parrot's test, I'm guessing it passes the :outer :lexid test I added to t/pmc/sub.t ?
jonathan Would seem so. 22:37
pmichaud excellent
if it passes rakudo's tests then I think we may have a winner (for now, at least :-)
I say that with a lot of trepidation, though, because every time I think I have a winner I find a new bug about 5 minutes later
jonathan :-( 22:38
t\\spec\\S02-builtin_data_types\\anon_block - fails, don't know if it did before my patch?
pmichaud it's been passing for me.
that may mean that newclosure isn't working properly. 22:39
jonathan All the sanity ones pass...will see what else happens in spectest....
22:39 zostay joined
pmichaud the sanity tests don't do much with closures, I don't think. 22:39
jonathan t\\spec\\S03-operators\\comparison..................FAILED tests 1, 3-18
Into the S29 ones now. Up to there, those two are the failures. 22:40
Hmmm. So I may have broken newclosure. 22:42
pmichaud: Can you verify what is passing for you there, just so I know my patch is to blame?
I just get failures in the two I mentioned. 22:43
Oh, argh.
I have a bunch of local changes too.
(To Rakudo)
22:47 Andy joined
jonathan pmichaud: The failures in t\\spec\\S02-builtin_data_types\\anon_block are just due to my local diffs here on Rakudo, not the closure patch. 22:48
pmichaud: And t\\spec\\S03-operators\\comparison.t fails the same way here with my closure patch taken back out again. So they're not related. 22:50
pmichaud jonathan: sorry, I could've saved you the time on comparison.t 22:56
Auzon++ switched the -1, 0, 1 to be Order::Increase, Order::Same, Order::Decrease
and we don't have those in rakudo yet :-)
I'll add them shortly.
If you think it's clean enough to commit, I say go for it :-) 22:57
jonathan pmichaud: The other thing is, that it fixes the problems my :immediate subs were having too. 22:58
pmichaud yay!
jonathan I pondered last night it might be the same issue...I'm happy it boiled down to the same thing.
So anyway, that means my patch to get the PMCs out and replace them with setting properties and laying the foundations for storing signatures, is not far off ready. 22:59
22:59 AndyA joined
jonathan There's one missing piece, which I will leave for tonight. Maybe tomorrow. 22:59
pmichaud excellent work, as usual 23:01
jonathan Next week, I have some rollout preparations for a big $dayjob I've been working on for quite a while.
I'll see how things go with it - if it looks like I can get a day without people all over me to do things, I'll do a Rakudo day next week. If not, I'll do two days the week after. 23:02
Also, the multiple dispatch grant starts in July - I have 40 hours to spend during July / August, so expect I'll generally do about 2 days a week or so on Rakudo for most weeks in July. 23:03
I plan to get all of the time under that grant in before YAPC::EU.
Sounds like a reasonable plan? 23:04
dalek r28763 | jonathan++ | trunk: 23:07
: [core] Fix RT#56398; lexical inner scope always kept the first lexpad they got (unless newclosure was explicitly used) rather than taking that of the latest invocation of their outer. Introduces no new failures in Parrot's and Rakudo's tests.
diff: www.parrotvm.org/svn/parrot/revision?rev=28763
stupidbot RT 56398: [BUG] lexical inner scope always keeps first lexpad (or something) - new
23:07 tetragon joined 23:15 ank joined
pmichaud jonathan: yes, very reasonable plan 23:20
jonathan pmichaud: OK, great. And above is the patch. Hope this lets you finally rid yourselve of the newclosure's. :-) 23:22
OK, that's enough hours in front of the computer for me today...time for some reading/beer and then sleep.
pmichaud well, at least the ones that aren't supposed to be closured :-)
jonathan Yes, the immediate block ones is what I meant. :-) 23:23
jonathan afk 23:24
23:26 kid51 joined
dalek r28764 | jkeenan++ | autoicu: 23:26
: [configure] Revert to state of file at start of branch (except that result is
: empty string rather than undef).
diff: www.parrotvm.org/svn/parrot/revision?rev=28764
23:30 rdice joined
dolmen in PIR, in rakudo, how can check that a PMC has '&iterator' ? 23:31
pmichaud $I0 = can $P0, 'iterator' 23:32
also, for most objects, $P1 = $P0.'HOW'(); $I0 = $P0.'can'('iterator')
23:32 japhb joined
pmichaud (i.e., for those that are using P6object) 23:33
er, that should be
also, for most objects, $P1 = $P0.'HOW'(); $I0 = $P1.'can'('iterator')
dolmen ok 23:34
pmichaud (the latter is the PIR equivalent of $x.^can('iterator') in p6 23:36
23:50 TiMBuS joined
japhb notices perlgeek's automatic acronym marks ... very cool 23:51

Powered by Cro and the RakuĀ® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!