Parrot 2.4.0 "Sulfur Crest" Released | parrot.org | Log: irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | GSOC Students: trac.parrot.org/parrot/wiki/GSoCersStartHere
Set by moderator on 18 May 2010.
00:00 theory joined 00:02 ash_ joined
darbelo Agh, forgot one commit. 00:03
dalek rrot: r46805 | darbelo++ | branches/ucs4_encoding/src/string/encoding/ucs4.c:
Fix typos that got lost in the git-svn merge.
darbelo git-svn--; me--
dalek rrot: r46806 | darbelo++ | trunk/src/string/api.c:
[CAGE] remove a space to make codetest happy.
 00:20
00:21 Whiteknight joined
darbelo I guess I'll need some tests now. 00:23
Coke msg tcurtis nope , doesn't look like it. have your mentor ping a trac admin. (me, whiteknight, half dozen others) 00:24
purl Message for tcurtis stored. 00:25
Whiteknight who is tcurtis' mentor, c?
darbelo I think he was chromatic's student. 00:26
Somebody should probbaly add the mentor-stutent pairs to the wiki. 00:27
00:27 chromatic joined
darbelo Where 'Somebody' == 'Not me' 00:28
chromatic If someone wants to flip the commit but for tcurtis, go ahead.
00:28 seatek joined
chromatic Or I can just have done it. 00:32
chromatic is having had much use out of the time traveler's verb guide he has yet to have purchasing.
msg tcurtis Welcome to your Parrot commit bit! Don't break the build. Increase the awesome. Ask me or any other committer if you have questions. 00:33
purl Message for tcurtis stored.
00:33 Mokurai1 joined
Whiteknight ash_: ping 00:45
dalek rrot: r46807 | darbelo++ | trunk/MANIFEST:
Regenerate MANIFEST
 00:53
rrot: r46808 | darbelo++ | branches/ucs4_encoding/MANIFEST:
Regenerate MANIFEST
00:59 abqar joined 01:00 Andy joined 01:30 plobsing joined 01:33 tcurtis joined 01:34 mikehh_ joined 02:08 snarkyboojum joined
plobsing seen ash? 02:28
purl ash was last seen on #moose 7 hours, 46 minutes and 30 seconds ago, saying: test failure = testing that i get a nicer error for a certain case
dalek rrot: r46809 | plobsing++ | trunk/src/dynoplibs/deprecated.ops:
flag deprecated dynops as :deprecated
 02:34
rrot: r46810 | plobsing++ | trunk (2 files):
svn props
purl svn props are the worst
rrot: r46811 | plobsing++ | branches/ops_massacre:
branch to kill (move to dynlibs) several ops per TT #449
02:37 JimmyZ joined
cotto darbelo would approve 02:43
03:07 janus joined
Coke sleepy. 03:20
cotto stupid 03:25
03:36 LoganLK joined
chromatic sarcastic. 03:44
03:45 theory joined
PerlJam ice cream 03:55
purl now that's what i call a "sticky situation"! or the Supreme Dessert
plobsing confused 03:56
purl You won't be after this episode of Soap!
04:19 snarkyboojum joined
dalek kudo: 8292d45 | (Solomon Foster)++ | src/core/Numeric.pm:
Add stub for Numeric.roots.
 05:00
05:10 skv joined 05:24 khairul joined 05:26 theory_ joined 05:43 parthm joined 05:53 Khisanth joined
dalek rrot: r46812 | tcurtis++ | branches/gsoc_past_optimization:
Creating a branch for my GSoC PAST optimization project.
 05:59
06:03 uniejo joined 06:04 aukjan joined 06:08 parthm left 06:14 uniejo joined 06:29 eternaleye joined 06:32 chromatic joined 06:56 iblechbot joined 07:14 fperrad joined 07:23 fperrad_ joined
dalek rrot: r46813 | plobsing++ | branches/ops_massacre (21 files):
move transcendental ops to dynops
 07:25
07:34 skv_ joined
dalek rrot: r46814 | plobsing++ | branches/ops_massacre (8 files):
move cmod to dynop
 07:43
07:45 TiMBuS joined
dalek rrot: r46815 | fperrad++ | trunk/lib/Parrot/Test/Pod.pm:
output a valid TAP comment
 08:02
tracwiki: v133 | bacek++ | Languages 08:23
tracwiki: fix typo in url
tracwiki: trac.parrot.org/parrot/wiki/Languag...ction=diff
bacek aloha 08:24
moritz good morning, codingrobot 08:26
08:28 mj41_ joined
GeJ Good morning everyone. 08:37
bacek moritz, heya 08:38
moritz, is "too eager materialization of lists" known issue in Rakudo?
e.g. ".say for 1_000_000" 08:39
moritz bacek: 'for' is known to be eager, yes
bacek moritz, any workarounds?
moritz bacek: map
bacek moritz, ok, thanks!
moritz bacek: void/sink context isn't detected yet in rakudo, so you have to explicitly use the return value of .map if you want anything to happen, or call .eager on it 08:41
bacek moritz, ah. got it 08:42
09:10 integral joined 09:21 parthm joined 09:41 mikehh joined 09:59 clinton joined
bacek seen cotto 10:36
purl cotto was last seen on #parrot 7 hours, 11 minutes and 4 seconds ago, saying: stupid
bacek msg cotto what is blocking on merge ops_pct? 10:37
purl Message for cotto stored.
mikehh bacek: testing ops_pct branch now 11:07
bacek mikehh, thanks!
dalek rrot: r46816 | mikehh++ | branches/ops_pct/MANIFEST:
re-generate MANIFEST
 11:10
mikehh bacek: make: *** No rule to make target `ext/nqp-rx/src/stage0/nqp-setting.nqp', needed by `runtime/parrot/library/nqp-setting.pbc'. Stop. 11:17
bacek: makecorevm/make coretest PASS 11:18
bacek mikehh, did you make it from clean checkout (or at least after make realclean; perl Configure.pl)? 11:31
mikehh from make realclean - let me do a clean checkout 11:34
bacek: same story - make corevm/make coretest PASS - make world fails to build - make: *** No rule to make target `ext/nqp-rx/src/stage0/nqp-setting.nqp', needed by `runtime/parrot/library/nqp-setting.pbc'. Stop. 11:46
11:48 parthm left
bacek mikehh, gotcha... Reproduced on latest ops_pct. 11:51
11:59 whiteknight joined
bacek mikehh, interesting. nqp-setting still has .pm extension in branch 12:00
dalek kudo: d855135 | (Solomon Foster)++ | docs/ChangeLog:
Add a few more bits to the changelog.
whiteknight good morning, #parrot 12:11
mikehh hiya whiteknight
All tests PASS (pre/post-config, make corevm/make coretest, smoke (#33946), fulltest) at r46816 - Ubuntu 10.04 amd64 (gcc) 12:13
12:13 iblechbot joined
mikehh but g++ fails to build - ./include/parrot/exceptions.h:120: error: expected unqualified-id before string constant 12:14
12:21 bluescreen joined
whiteknight seen Austin? 12:24
purl Austin was last seen on #parrot 16 days, 6 hours, 24 minutes and 51 seconds ago, saying: gitorious.net/kakapo/kakapo/blobs/m...Syntax.nqp [May 4 05:59:49 2010]
12:28 theory joined 12:36 theory_ joined 12:41 bluescreen joined, ruoso joined
dalek rrot: r46817 | mikehh++ | trunk (2 files):
fix headerizer - misplaced PARROT EXPORT
 12:49
12:49 whiteknight_ joined
dalek kudo: 2e71bc8 | moritz++ | src/metamodel/Attribute.nqp:
beautified code a bit, and add comment about pir::set_vps($meth, $meth_name)

No functional changes.		 12:56
13:00 tetragon joined 13:12 davidfetter joined
dalek kudo: 65893e1 | (Solomon Foster)++ | docs/ (2 files):
Work on release announcements.
 13:14
dukeleto 'ello 13:17
dukeleto is hanging out at PGCon2010, still tweaking slides about PL/Parrot
13:18 theory joined
dukeleto bacek: nice to see pirc/pirate gettting some love 13:22
mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#33948), fulltest) at r46817 - Ubuntu 10.04 amd64 (g++) 13:24
hiya dukeleto 13:25
dukeleto mikehh: howdy
purl hey, dukeleto.
dalek kudo: 9307272 | moritz++ | docs/announce/2010.05:
[announce] list contributors since last release
kudo: 1c6eb2d | (Solomon Foster)++ | docs/announce/2010.05:
Incorporate jnthn++'s suggestions.
kudo: c984346 | Coke++ | docs/announce/2010.05:
fix typo, minor formatting.
 13:31
13:31 Mokurai joined
dalek kudo: 24e28a1 | (Solomon Foster)++ | docs/ (2 files):
Some spell checking, mention Blizkost.
 13:36
kudo: 296a56c | (Solomon Foster)++ | docs/release_guide.pod:
Add #29 Erlangen to the release list, and mark that [Coke] has the July release.
 13:48
13:52 gbacon_ joined 13:53 Andy joined 13:55 tetragon joined 14:01 theory joined 14:02 plobsing joined 14:05 tetragon joined
dukeleto msg bacek i forked your pir github repo, and I am adding some issues as I play around with it. Let me know how I can help. 14:18
purl Message for bacek stored.
14:19 davidfetter joined
dalek kudo: 1c32466 | Coke++ | docs/release_guide.pod:
remove extra punctuation.
 14:19
parrot: 85d131f | dukeleto++ | t/sql/test.sql:
Add tests for STRICT and IMMUTABLE
 14:34
14:34 iblechbot joined, whiteknight joined
dalek kudo: 9021d4c | (Solomon Foster)++ | docs/ROADMAP:
Paradoxically update ROADMAP to note when the last real update to it was.
 14:36
14:45 patspam joined 14:57 theory joined 15:18 tetragon joined 15:23 bubaflub joined
cotto bacek, implementing ops renumbering is all that's left. It should be done in the next day or two. 15:25
15:30 theory joined 15:33 gbacon joined
mikehh cotto: nqp is out of sync with trunk on ops_pct branch (and fails to build) 15:34
or rather nqp_rx branch 15:36
15:38 wagle joined 15:43 tcurtis joined
dalek TT #1650 created by dukeleto++: Parrot needs Date/DateTime Object 15:47
TT #1650: trac.parrot.org/parrot/ticket/1650
parrot: 4bc41c5 | dukeleto++ | (6 files):
Rename plparrot to plparrotu
 16:14
16:17 mmcleric joined 16:24 ash_ joined
ash_ as a GSoC student, how would I make a blog post on the parrot page? 16:25
I made one to talk abut my work, because everyone else seems to, and I wana be cool
darbelo ash_: Do you have a parrot.org username?
ash_ but I just posted it on my blog, not the parrot one
darbelo (Not the same as trac.)
ash_ i have a trac usename, is that the same thing?
nope then
darbelo You need to create one, and the ask someone to give you the right privileges. 16:26
dalek parrot: 88d8784 | dukeleto++ | .gitignore:
Improve .gitingore
darbelo I think Coke is the one that ussualy does it.
Andy holy crap, I love this nickclifton.livejournal.com/5630.html
-Wsuggest-attribute=const !!! 16:27
ash_ ping Coke 16:28
16:36 cotto_work2 joined
cotto_work2 mikehh, I had to make clean but everything looked fine after that. 16:36
darbelo Andy: You are going to const functions as well now? Nice. 16:39
Andy darbelo: I have been. 16:40
are you talking about const functions or const parameters?
darbelo const functions.
Andy We already have PARROT_CONST and PARROT_PURE function annotaters
ack for them.
ack -A5 PARROT_CONST
darbelo Yeah, but they are a bit under used.
Andy If you find a function you think could be const but isn't, I'd love to know. 16:41
I've been through every .c file there is.
darbelo Offhand, I can't think of any, but I don't ussually pay a lot of attention to annotations :) 16:42
Andy Then you might want to rethink saying that they are underused. 16:43
darbelo I know I don't use them when I should.
PerlJam It would be interesting to see what -Wsuggest-attribute=const has to say. We'll find out how good a const'er Andy is when gcc has no suggestions :) 16:44
Andy :-P
It's not clear to me what compiler has those in it.
if I have to build from a snapshot or something.
darbelo Andy: It could be added to our warings detection step, when you get a compiler that suports them they'll JFW. 16:56
Andy Yes, I know that.
But it's not clear from the posting I linked to where to find them to work on them.
If I have to build from a snapshot, I will, I just haven't for a while. 16:57
heck, it might not even be on trunk.
darbelo Ah, ok. My guess is that it'll warn about our generated code, since that's not very annotation friendly. You seem to have taken care of everything else AFAICT. 16:58
Andy darbelo: If it complains about generated code, then we should make the annotated code use PARROT_CONST and PARROT_PURE, too. 16:59
darbelo Our code generation tools aren't smart enough to know where to use it. 17:00
Andy no, but we can also annotate.
I've done it before, I can do it again. :-)
darbelo Inside the function, sure. The function itself... Our code generation tools aren't smart enough. 17:01
17:01 ash_ joined
dalek rrot: r46818 | petdance++ | trunk/src/gc (2 files):
changed function pointers to NOTNULL instead of ARGIN
 17:01
Andy Yes, I know, but i'm sure we can annotate where necessary.
Let's see what the compiler complains about.
dalek kudo: e3fe329 | moritz++ | docs/release_guide.pod:
[docs] warn about potential damage from "make release"
 17:07
17:17 theory joined
dalek rrot: r46819 | petdance++ | trunk/src/utils.c:
more arg consting
 17:18
17:23 Mokurai joined 17:26 tetragon joined 17:31 mmcleric joined
dalek rrot: r46820 | petdance++ | trunk (2 files):
Don't complain about lexer-generated functions
 17:36
17:37 davidfetter joined 17:41 Casan joined 17:42 allison joined 17:52 Mokurai joined
Coke ZOMG it's allison 17:59
how's classes?
18:01 joeri joined
davidfetter hello 18:02
how do i compile some HLL, ideally a simple one, into PIR?
dukeleto examples/languages/squaak 18:03
Coke davidfetter: are you using PCT?
if so, you can do parrot hll.pbc --target=PIR code.hll > code.pir 18:04
davidfetter oh, interesting :)
Coke otherwise, depends on your compiler.
dukeleto --target=pir should work 18:05
but that will include definition of namespaces, which won't quite work when PL/Parrot wraps the code you pass it in an anonymous function 18:06
it is doing with manual massaging, i think
s/doing/doable/
davidfetter well, i'm ok with running it through a filter. a manual one, even. at least for today :)
dukeleto i am recompiling parrot now, since my tests for squaak failed 18:07
i am seeing the tests for examples/languages/squaak blowing up, anybody else seeing this? 18:10
i made a TT: trac.parrot.org/parrot/ticket/1651 18:13
whiteknight dukeleto: I have a question about your parrot mirror on github: Why don't branches there disappear when they are deleted from the SVN repo? 18:15
the project lists hundreds of old branches which have since been deleted
dalek kudo: e6863e4 | moritz++ | (2 files):
bump PARROT_REVISION, and load obscure_ops
 18:16
18:18 ash_ joined
dalek TT #1651 created by dukeleto++: Squaak tests fail 18:20
TT #1651: trac.parrot.org/parrot/ticket/1651
ash_ ping Coke? 18:21
Coke pong ash 18:22
ash_ Coke: I made an account on the parrot.org site (the drupal one), I was wondering if I could get added to the user group that the other GSoC guys are in
i think its for posting blog posts
dukeleto whiteknight: yeah, i need to fix that 18:24
whiteknight dukeleto: is that a svn mirroring tool of your own design? 18:25
dukeleto whiteknight: indeed, i have since learned better ways, but haven't gone back and made it better 18:26
Coke: could you add squaak as a language to trac?
Coke: in the languages dropdown
whiteknight I can do that
dalek rrot: r46821 | darbelo++ | branches/ucs4_encoding (2 files):
Add charset and encoding options to the stringinfo opcode.
dukeleto whiteknight++ 18:27
18:27 fperrad joined
whiteknight ...actually, I can't see where to add it 18:28
Coke note that the languages trac has its own trac. 18:30
.. not that we're using that.
moment.
whiteknight a second trac?
moritz as if one wouldn't be too many already :-) 18:31
it's software, and therefore hateful
Coke ash_: what is your parrot.org id? 18:32
ash_ ash 18:33
Coke ash is now a poster.
ash_ cool, thanks 18:34
Coke++ 18:35
GSoC++ ?
as a GSoC person, am I supposed to do a weekly blog post?
whiteknight ash_: we would very much prefer it if you did
ash_ kk, I'll do my best to make sure the content is good
should I just repost my blog post i put on my website to the parrot blog? or...? 18:36
Andy It would be highly awesome if there were some kind of GSoC summary each week that Perlbuzz could point to, too.
Coke dukeleto: done.
ash_: up to your mentor, but what whiteknight said.
whiteknight ash_: parrot-related blog posts should be getting mirrored to the parrot aggregator
Andy Visibility = good
Coke is leery of summaries.
Andy Repetition = good
Coke (esp. of summaries)
Andy A summary that points to individual blog psots. 18:37
Coke the languages dropdown is controlled by the trac.ini file on the server. 18:38
whiteknight Coke: ah, thats why I couldn't find it
ash_ greaterthaninfinity.com/2010/05/gsoc-work/ is the blog post I made today, i guess for now I'll just repost it... 18:39
Andy ash_: We can't tell your name. 18:41
18:41 iblechbot joined
Andy I look at greaterthaninfinity.com/2010/05/gsoc-work/ and have no idea who you are. 18:41
ash_ hmmm thats a good point, i should introduce myself on the parrot blog at least 18:42
Andy and put your name on each post 18:45
and contact information
ash_ kk 18:46
Andy Consider the reader stumbling on your work without any idea who you are?
s/.$/./
ash_ does drupal obfuscate emails?
Andy ash_: I dunno.
dukeleto Andy: i can try to make a gsoc summary blog post each week, but I make no promises. how do you want me to notify you of them? 18:55
ash_ Okay, I posted a blog post about my GSoC work, questions, comments, concerns, etc. please let me know 18:56
dalek website: ash++ | GSoC Project: NCI and Stack Frame Improvements
website: www.parrot.org/content/gsoc-project...ovements-0
Andy dukeleto: Make them visible to whereever people will see them, and if it's interesting I
will point a Perlbuzz tweet at it.
Coke ... and /if it's interesting/ 18:57
Andy Coke: Yes? 18:58
Coke so you're asking us to do extra work on spec. =-)
Andy No, not at all.
I don't run anything on Perlbuzz that isn't interesting.
And yes "interesting" is subjective.
darbelo has an interest in uninteresting topics. 18:59
Coke Andy: yes, but my point is, dukeleto is now going to make those summary posts, even if they're not interesting.
Andy Coke: OK.
Presumably he's not doing them only for Perlbuzz
dukeleto Andy: your definition of interesting seems to be an undefined variable 19:00
Coke yes. we weren't doing them. then you asked us to do them.
so now he's doing them, but you might not like them. =-)
Andy Well, I didn't imagine that you'd only be doing them for Perlbuzz.
ash_
.oO( should post more explosions... )
Andy ash_: Your intro thing is interesting.
Coke dukeleto: have fun. =-)
Andy If you next week post "Well, I started working on such and such, and I made some documents about blah blah" then that's probly not very interesting to the general public, is it? 19:01
dukeleto Andy: even if you don't think it is interesting, you did a disservice to the Perl community last year by not mentioning GSoC stuff
Andy dukeleto: Bummer.
I don't think that's a fair assessment in the slightest, however.
Perlbuzz is very specifically NOT a firehose. 19:02
There are more than enough planet.* aggregators for that.
planet.parrotcode.org/
Stuff like this, dukeleto: perlbuzz.com/2009/03/get-paid-for-w...-2009.html ? 19:04
What should I have pointed to that I didn't? 19:05
19:06 silug joined
Coke Andy: there is a difference between "make your individual posts interesting so people care about them" and the summary postings we were talking about. 19:07
at least in my mind.
Andy I'm not talking abotu summaries in specific.
I just mean in general we need to make things interesting.
Coke 14:36 <@Andy> It would be highly awesome if there were some kind of GSoC summary each week that Perlbuzz could point to, too.
Andy Right, itwould be. 19:08
Coke (that's why I keep referring to summary)
Andy But don't do it for PB's sake.
Coke let me try to start over:
Andy If there are a dozen projects, and each of those projects has a weekly blog post, then something that aggregates those posts with pointers would be sweell.
Maybe I jumped past the assumption that each project would have its own weekly posting.
Coke GSOC students - when posting your blog entries for the week, remember that you have a potentially wider audience than your mentor and the folks in #parrot.
Andy sure
Coke Andy: as you say, we already have aggregators. why make another one? 19:09
Andy becuase it would be GSoC-specific
moritz gnah, I get a PIR line number reported outside the file
Andy "Here's what Perl/Parrot have done with GSoC this week"
The aggregators are all unfiltered.
Coke GSOC students - please tag your posts with "GSOC"
Andy That'd be a quick fix perhaps. 19:10
Coke certainly easier than trying to make even more content interesting by rehashing existing content, yes.
Andy No, i'm not suggesting making the summary interesting.
The summary I just see as a link pile of pointers.
ash_ Coke: how do you add tags in drupal?
Coke Andy: ok. then tagging seems to be much easier. 19:11
ash_: checking.
Andy The big thing is "make your posts interesting"
Coke ash_: when I edit a post, I see a title, then categories, and that block contains "tags"
Andy "make your posts interesting other than to someone who already knows everything about what you're doing'
ash_ Umm... I don't have a categories, i only have title, body, input format, url path settings 19:12
19:12 mmcleric left
ash_ so... as a design consideration, currently the NCI system uses strings to identify types, like Int's and what have you, but how should I do structures? 19:13
Andy dukeleto: ping
ash_ I am thinking a new object that represents a structure, but it might make things difficult since currently to make an NCI function call you have to say like "v" for void, how would I interpolate the structure into that? hmmm.... anyone have any comments? I could do some sort of seperators like "v{tf}" for a "void, struct { char*, float}" 19:17
hmm
darbelo ash_: You want to pass structures into a function, not pointers? 19:18
ash_ I want full support libffi offcers, so that includes both 19:19
Coke () and [] for struct vs. pointer to struct?
darbelo From what little I know about our NCI capabilities we don't handle that type of call. 19:20
moritz that's why we need ash_'s project :-)
ash_ darbelo: no, but thats my GSoC project
(or part of it)
darbelo Closest thing we have are ManagedStruct and UnmanagedStruct, but those do a different thing.
Andy ManagedProject and UnmanagedProject 19:21
ash_ darbelo: i'll look at how those two work 19:23
dukeleto Andy: pong 19:26
Andy dukeleto: A couple of things to clear up. First, I think "disservice" is pretty harsh to call "you didn't run articles that I think you should have." Disservice implies that I hurt Perl somehow, which NOT running an article certainly doesn't. Second, I don't recall ever rejecting anything from you. 19:28
dukeleto Andy: it is water under the bridge. I will attempt to write interesting blog posts this summer and let you know about them. 19:29
Coke plobsing: trac.parrot.org/parrot/ticket/1652
(more imcc goodness for you.)
Andy Not sure I'd call it "water under the bridge" when you throw accusations at me today.
dukeleto Andy: I do remember you telling me that some of my posts were not interesting enough for Perlbuzz 19:30
Andy OK, and why was that?
dukeleto Andy: it is water under the bridge because I don't feel like beating a dead horse
Andy Do you recall which? I'm sure if I did I'd have said something along the lines of "What's the angle, how can we make it more interesting to the general reader?"
19:30 bubaflub joined
dalek TT #1652 created by moritz++: Line numbers in a particular IMCC stack trace are all wrong 19:31
TT #1652: trac.parrot.org/parrot/ticket/1652
ash_ for design decisions that might impact functionality/usability of parrot how should I go about resolving the design decisions? start a parrot-dev mailing list message? or ask here (at a specific time? the parrot sketch?) 19:32
Andy I'm right now pulling down a GCC 4.6 snapshot to see if I can constify/pureify more funcs.
dalek parrot: f9f4dba | dukeleto++ | plparrotu.c:
Include bytea.h if PG is new enough
 19:33
dukeleto Andy: perlbuzz.com/2009/03/how-to-write-a...ement.html
Coke ash_: mailing list is a good place to beat the drum.
Andy dukeleto: What about it?
ash_ Coke: thanks, i'll write up my thoughts then send an email for criticism 19:34
should i make a ticket for a RFC too? or no because its a GSoC not trunk parrot?
Coke ash_: pinging people in here is fine for insta-feedback, you'll get more info on the list, and you can usually get resolution there or in #parrotsketch.
ash_: depends on who you ask. I think RFCs are fine (since presumably we hope to eventually merge your branch to trunk) 19:35
else why even ask, right? you could just do whatever you wanted on branch with no consequences. =-)
whiteknight adds another line to his list of technologies he hates: silverlight 19:37
ash_ alright, once i get my thoughts sketched out so i have had time to think through my changes i'll email the mailing list and and make an RFC ticket (that can be resolved when my changes get hopefully merged into the trunk) 19:38
19:38 allison joined
ash_ Coke: also, i don't think i have privileges to tag my blog post as any sort of category, just so you know 19:38
Coke ok. anyone with editor or admin privs can re-tag an existing blog post. I'll let mentors do that until they revolt and make me figure it out more. 19:41
ash_ lol, drupal confuses me too, i have to work with it occasionally at work
Coke ah.
no, blogs != news
whiteknight up != down 19:43
black != white
particle NaN != NaN
dukeleto particle++
Coke leave my nana out of this. 19:44
NotFound Let's break things! 19:45
whiteknight ++ 19:46
Coke anyone know what "administer taxonomy" permission in drupal does. this looks like the setting I want, but it has no docs.
HA! 19:47
ww.
19:48 [1]Casan joined
dalek rrot: r46822 | NotFound++ | trunk (3 files):
store string encoding in PBC and change semantic of wide char encodings in PIR string literals
 19:53
rrot: r46823 | NotFound++ | trunk/src/packfile/pf_items.c:
c++ fix
moritz I *HATE* IMMC 19:59
erm, IMCC
I get a syntax error in a routine
I comment it out
now I get a syntax error *before* that routine
Andy Am I the only one that cares about new comilers? Does anyone need an account on uniqua?
19:59 sorear joined
whiteknight uniqua? 19:59
darbelo moritz: Join the club.
Andy whiteknight: My dev Linux box. 20:00
whiteknight Andy: which compilers you peddling?
Andy Or am I the only one pushing bleeding edge GCC
icc, gcc, trying to do clang
and splint
whiteknight I do icc and clang regularly 20:01
Andy awesome.
whiteknight I don't even know what the "bleeding edge" version of GCC is right now
Andy ftp://ftp.nluug.nl/mirror/languages/gcc/snapshots/4.6-20100515/gcc-4.6-20100515.tar.bz2
whiteknight oh, no. I think I use 4.1 on one of my boxes, but certainly not 4.6 20:02
darbelo Andy: My main platform is stuck on 3.3.5, how's that for "bleeding edge"?
Andy nickclifton.livejournal.com/5630.html is what started all this
particle darbelo: that's closer to the bleeding tail.
atrodo My server at home has gcc 3.3.3 20:03
whiteknight damn, and I thought it was bad that my recent install of Solaris had 3.4.5 20:05
Andy atrodo: So many awesome warnigns you're not seeing!
Warnings are fun!
darbelo Technically, I have 4.2.4 available in the ports colection, but the system compiler is still 3.3.5
atrodo Andy> I know! I long for the day that I could see all the code I compile warn me
GeJ Good morning everyone. 20:06
Andy Need to play with gold, too. 20:07
20:09 mmcleric joined
dalek rrot: r46824 | darbelo++ | branches/ucs4_encoding/src/string/encoding/ucs4.c:
Update string length after manually filling the buffer.
 20:10
rrot: r46825 | NotFound++ | trunk/t/native_pbc (4 files):
update native_pbc
ash_ Andy, i have a development version of clang (svn trunk, relatively recent) that i use for most of my compiling, although for the GSoC I plan on working with llvm 2.7, the most recent stable, since it is stable 20:22
its clang 1.5, currently for me 20:23
darbelo NotFound++ # Wide chars. 20:25
NotFound darbelo: It was time!
dalek rrot: r46826 | darbelo++ | branches/ucs4_encoding (25 files):
Pull changes from trunk.
 20:27
20:39 bluescreen joined 20:45 hercynium joined
darbelo allison: ping 20:56
dalek rrot: r46827 | darbelo++ | branches/ucs4_encoding (2 files):
Undo r46821, those pointers aren't telling us anything new.
 21:00
21:15 mikehh_ joined
bacek Good morning, humans. 21:15
mikehh_ yay bacek 21:16
darbelo Good morning, magical coding robot.
dalek rrot: r46828 | darbelo++ | branches/ucs4_encoding/src/string/encoding/ucs4.c:
Add some ASSERT_ARGS() I had missed.
rrot: r46829 | darbelo++ | branches/ucs4_encoding/src/string/encoding (2 files):
[CAGE] Adjust svn props.
bacek dukeleto, ping
GeJ G'Day bacek. 21:17
bacek G'Day GeJ
aloha mikehh, darbelo
21:18 mikehh joined 21:20 Whiteknight joined
dalek rrot: r46830 | darbelo++ | branches/ucs4_encoding/src/string/encoding/ucs4.c:
[CAGE] Shuffle whitespace about to make codetest happy.
 21:32
bacek msg dukeleto I added you to bacek/pir collaborators. Feel free to hack it :) 21:36
purl Message for dukeleto stored.
darbelo bacek++ # Getting more necromancers.
bacek darbelo, I started from scratch... 21:37
darbelo Oh, my copy was stale. PIRATE? Nice. 21:39
21:41 elmex joined
bacek darbelo, It was first not fully terrible name that came to my mind :) 21:42
darbelo I like it. Arr. 21:43
bacek We can use one eyed parrot as logo :)
dalek rrot: r46831 | darbelo++ | trunk/compilers/imcc/pbc.c:
[CAGE] Tab police.
 21:49
rrot: r46832 | bacek++ | trunk/examples/languages/squaak/t/00-sanity.t:
Update tests to avoid multiple declarations of variables. Closes #1651
NotFound Ups 21:50
21:55 Mokurai joined
Coke another pirate project? 21:59
Coke oh mys at his google search for parrot pirate.
dalek TT #1651 closed by bacek++: Squaak tests fail 22:00
TT #1651: trac.parrot.org/parrot/ticket/1651
bacek How I can add this github/bacek/pir to dalek? 22:03
dukeleto bacek: i think you bug Infinoid 22:04
darbelo Adding it to the Languages page in the wiki should be enough.
If that doesn't work, bug Infinoid.
bacek It's on Languages page.
dukeleto bacek: thanks for the commit bit, i will attempt to write some tests for pirATE
darbelo dukeleto: We had tests, but the PIR ate them. 22:06
:)
bacek dukeleto, don't put too much effort now. Everything is broken.
Only one construction works: empty subroutine without pragmas and content :)
dukeleto bacek: what is the most useful thing that i can do? 22:07
Whiteknight There are only so many plays on the words "pir" and "parrot" we can make 22:09
bacek dukeleto, for now - almost nothing. In about week or two - testing, extending grammar, etc.
Whiteknight Eventually we're going to go the gnome/kde route and just prepend "p" on front of everything
pperl6, pruby, pphp
Tene or follow Coke and prepend Par 22:11
NotFound Or play a different game, like Winxed. 22:13
dukeleto bacek: i can fiddle with docs. A lot of docs talk about PGE, that is not outdated, correct? 22:16
s/not/now/ 22:17
darbelo dukeleto: Yep. It's (re)written in NQP. 22:18
dukeleto darbelo: that is what I thought, but wanted to verify. I will hack on the docs, then 22:20
darbelo dukeleto++ 22:21
bacek dukeleto, yes
dukeleto that is pretty interesting. we are creating a PIR parser in a language written in PIR and which boils down to PIR. Just feels really meta.
bacek dukeleto, it will boil down to PBC directly :) 22:22
Eventually....
Tene Meta? Us? No way.
darbelo starts pulling his bootstraps. 22:23
dukeleto how do I make setup.pir look at tests in subdirectories of t/ ? 22:25
i.e. what is the equivalent of Module::Build's recursive_test_files => 1 ? 22:26
Whiteknight NotFound: pWinxed 22:27
darbelo $P0['prove_files'] = 't/*.t t/whatever/*.t'
dukeleto darbelo++ 22:28
darbelo dukeleto: Add that line under the " $P0['prove_exec'] = $S0" one and it'll work out of the box.
I think.
dukeleto darbelo: yes, that worked. and sure enough, every test fails :) But now they run! 22:40
darbelo dukeleto: SHIP IT!
purl: msg allison I've created the ucs4_encoding branch to add ucs4 support to parrot. It has no tests yet, but it should be ready for review. Other than scattered support code here and there, the work happens in src/string/encoding/ucs4.c so I'd like to hear your thought on it. 22:42
purl Message for allison stored.
22:46 theory joined 22:50 kid51 joined 22:59 Psyche^ joined
Whiteknight darbelo++ 23:31
darbelo Huh? What did I break now?
Whiteknight the ucs4 branch
except you didn't break it
darbelo Yeah, I feel weird *adding* so much code to parrot ;) 23:32
ash_ gist.github.com/408274 does that seem crazy for an idea for NCI functionality? I am going to email the mailing list for more feedback, just wanted anyone that had time to glance at it and tell my where I am doing silly things 23:39
the example of the pir code is at the bottom 23:43
23:43 mmcleric_ joined
bacek_at_work ash_, op dlfunc should accept array for types binding. We don't have variable-size ops. 23:44
ash_ okay, i can change that
dalek rrot: r46833 | darbelo++ | branches/ucs4_encoding/src/string (2 files):
Update the encoding promotion logic in Parrot_str_concat() to correctly handle ucs4.
ash_ hmm, dang, got to go, i'll email the mailing list for feedback, thanks bacek_at_work++ for your quick response 23:46
bacek_at_work ash_, and take a look at (Un)ManagedStruct PMC. May be you can use it.
mmcleric, привет :)
ash_ yeah, i'll look at those, my cursory glance at them seemed like they may do part of what I want but not all of it, so I figured i'd work from scratch until told otherwise 23:47
anyway, gotta go
NotFound A problem with that way is size. A signature string is just a few bytes, an array of opcodes is a lot more space. 23:53
darbelo Also, opcodes get bigger on 64-bit platforms. 23:54
Whiteknight I need to find a small project to do, get my feet wet again 23:57
any ideas?
darbelo Whiteknight: Hack on pirate! 23:58
NotFound Whiteknight: rewrite pbc_checker in a parrot language.
Even beter, write a parrot C++ compiler able to compile it X-) 23:59

Powered by Cro and the Raku® Programming Language.

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

Thank you!