Camelia, the Perl 6 bug

IRC log for #parrot, 2012-02-09

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:09 dalek nqp: 2246b9d | jnthn++ | src/HLL/Grammar.pm:
00:09 dalek nqp: Fix preclim handling in operator precedence parser to be consistent with STD's semantics.
00:09 dalek nqp: review: https://github.com/perl6/nqp/commit/2246b9dd83
00:10 dalek rakudo/stdinit: 4f23c1a | jnthn++ | src/Perl6/Actions.pm:
00:10 dalek rakudo/stdinit: Fix assign_op calls.
00:10 dalek rakudo/stdinit: review: https://github.com/rakudo/rakudo/commit/4f23c1af53
00:17 dalek nqp/bs: 19f1639 | moritz++ | / (2 files):
00:17 dalek nqp/bs: [bigint] fix detection of negative numbers from binary ops
00:17 dalek nqp/bs:
00:17 dalek nqp/bs: This should fix RT #109740
00:17 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/19f1639689
00:17 dalek nqp/bs: 2246b9d | jnthn++ | src/HLL/Grammar.pm:
00:17 dalek nqp/bs: Fix preclim handling in operator precedence parser to be consistent with STD's semantics.
00:17 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/2246b9dd83
00:17 dalek nqp/bs: 239b54f | jnthn++ | / (6 files):
00:17 dalek nqp/bs: Merge branch 'master' into bs
00:17 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/239b54ffb1
00:33 tty234 joined #parrot
00:47 whiteknight joined #parrot
00:48 whiteknight good evening, #parrot
00:48 benabik o/ whiteknight
00:48 whiteknight hello benabik
01:08 woosley joined #parrot
01:39 woosley left #parrot
01:40 woosley joined #parrot
01:56 dalek Rosella/gh-pages: f008d11 | Whiteknight++ | index.md:
01:56 dalek Rosella/gh-pages: fix weird formatting issue. benabik++
01:56 dalek Rosella/gh-pages: review: https://github.com/Whiteknig​ht/Rosella/commit/f008d1110d
03:59 Util joined #parrot
04:04 silug joined #parrot
04:10 zby_home_ joined #parrot
06:34 silug joined #parrot
06:36 Infinoid joined #parrot
07:40 he joined #parrot
08:06 mj41 joined #parrot
08:23 alvis msg NotFound if you have some time, please see git://gist.github.com/1778408.git
08:23 aloha OK. I'll deliver the message.
08:23 alvis msg NotFound i keep getting an error I'm unable to figure out. thanks.
08:23 aloha OK. I'll deliver the message.
08:24 alvis cotto: ping
08:26 lucian joined #parrot
08:30 he Found a bug introduced in parrot 3.9.0.  See http://pubs.opengroup.org/onli​nepubs/7908799/xsh/usleep.html, it disallows usleep >= 1000000 microseconds.  ("... must be less than 1,000,000").  I'm on a platform which enforces this restriction.  Ref. to diff follows.
08:31 he There's a ">" which should be ">=".
08:31 nopaste "he" at 192.168.1.3 pasted "Fix for usleep(usecs > 1,000,000)" (13 lines) at http://nopaste.snit.ch/115884
08:32 he (and the comment should be "(usecs >= 1,000,000)", of course)
08:32 alvis cotto: ok, well, it's probably way too late.  i'll try again tomorrow (at a more reasonable hour).
09:02 he Bah: TICKET_CREATE privileges missing for me (user heidnes)
09:03 sorear you seem to have missed the memo where we switched to github
09:03 sorear trac is read-only for everyone
09:03 sorear because we want new tickets to go in one place only
09:05 he The link "Report an issue" on www.parrot.org still points to trac.
09:05 sorear who can fix that?
09:05 he "not me!" :)
09:06 sorear (don't let the @ fool you, I'm mostly outside of the project)
09:22 alvis sorear & he: i'll poke around tomorrow evening and see if i can get things to point to right place.
09:23 alvis make that, this evening.
09:28 lucian joined #parrot
09:31 alvis he: please use https://github.com/parrot/parrot/issues
09:38 alvis msg moritz the release_manager_guide says I should ask "language project leaders," especially Rakudo, if there are any release blockers (to the 4.1.0 release).
09:38 aloha OK. I'll deliver the message.
09:39 alvis msg moritz are there?
09:39 aloha OK. I'll deliver the message.
09:39 he alvis: ok, logged in and will submit the 1-char bug.
09:39 seme is there some kind of issue with hg or git when checking things out with en_GB.UTF-8 as your locale setting? I checked out another project and it wouldn't build because of that weird << character
09:39 alvis he: thanks!
09:42 alvis seme: you might ask dukeleto. he's a git guru.
09:42 seme thanks
09:42 alvis seme: so is Coke, btw.
09:45 * alvis sleep()
09:47 aloha (parrot/parrot) Issues opened : 714 (Use of usleep in conflict with SUSv2 (fencepost error)) by he32 : https://github.com/parrot/parrot/issues/714
09:51 woosley left #parrot
09:51 moritz aloha: I'm not aware of any blockers, but it's been a few days since I've tested rakudo on parrot/master
10:09 moritz erm sorry, meant alvis
10:35 particle joined #parrot
11:03 schmooster joined #parrot
11:03 he Hm, looks like that bug is a 3.9.0-only thing.
11:47 Hunger joined #parrot
12:17 dalek nqp/bs: ffeec9c | jnthn++ | src/6model/serialization.c:
12:17 dalek nqp/bs: Set SC on objects as we deserialize them.
12:17 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/ffeec9c21a
12:17 dalek nqp/bs: 7275a66 | jnthn++ | src/6model/serialization.c:
12:17 dalek nqp/bs: Serialize/deserialize v-table and method cache from STable.
12:17 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/7275a66b2b
12:17 dalek nqp/bs: 2ea450f | jnthn++ | src/6model/serialization.c:
12:17 dalek nqp/bs: Introduce concept of a cloned code reference; this will be used when implementing closure serialization.
12:17 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/2ea450f2ac
12:19 seme joined #parrot
12:21 seme hey guys... I'm hoping one of you can help... I've been trying to figure out this weird encoding issue I'm having... I checked out the code and haven't made any changes but when I look at the parrot squaak example I can see weird characters  and « »... very odd I looked into resetting my locale which is en_US.UTF-8 and looked at various git settings which may help with no success
12:21 seme any pointers would be greatly appreciated
12:24 dalek rakudo/nom: 828e4d6 | moritz++ | src/core/Exception.pm:
12:24 dalek rakudo/nom: allow numification of X::AdHoc
12:24 dalek rakudo/nom:
12:24 dalek rakudo/nom: This is a bit questionable, but a spectest requires it
12:24 dalek rakudo/nom: In the end we might want to make X::Adhoc a wrapper
12:24 dalek rakudo/nom: around a payload, and only use the stringification
12:24 dalek rakudo/nom: for HLL interop.
12:24 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/828e4d6c4d
12:47 jsut joined #parrot
13:32 dalek rakudo/stdinit: 7b6cf5d | moritz++ | src/ (3 files):
13:32 dalek rakudo/stdinit: Merge remote branch 'origin/has-self' into stdinit
13:32 dalek rakudo/stdinit: review: https://github.com/rakudo/rakudo/commit/7b6cf5da5a
13:32 dalek rakudo/stdinit: 48c50ff | moritz++ | tools/build/NQP_REVISION:
13:32 dalek rakudo/stdinit: bump NQP revision to get preclim fix
13:32 dalek rakudo/stdinit: review: https://github.com/rakudo/rakudo/commit/48c50ff5d3
13:58 Coke joined #parrot
14:00 mtk joined #parrot
14:29 seme is there an alternative for token infix:sym<->  for when your symbol includes < or >
14:30 PerlJam seme: texas quotes probably
14:30 seme what are they?
14:31 seme ooh don't search google for texas quotes ;)
14:39 PerlJam seme: texas quotes are << and >>, but perhaps guillemets work?  («»)
14:40 PerlJam I'm not really sure anymore since I haven't played with things in a while.
14:41 PerlJam seme: btw, you're talking about in NQP, right?
14:46 PerlJam aye, guillemets work in nqp
14:47 PerlJam random example: Perl6/Grammar.pm:2510:    token infix:sym«<==»
14:48 particle https://github.com/perl6/nqp/blo​b/master/src/NQP/Grammar.pm#L691
14:57 PacoAir joined #parrot
15:09 tty234 joined #parrot
15:32 moritz now if you ever want to write an op that includes both > and »...
15:33 masak just surround it with <<< >>> :)
15:44 Psyche^ joined #parrot
15:45 dalek nqp/bs: b06857a | jnthn++ | src/6model/reprs/P6opaque.c:
15:45 dalek nqp/bs: P6opaque should ensure its REPR data is calculated before any serialization of it takes place.
15:45 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/b06857a06c
15:45 dalek nqp/bs: f6f2e8a | jnthn++ | t/serialization/03-closures.t:
15:45 dalek nqp/bs: Add initial test for handling of cloned code refs (not testing outer handling yet, though - just very simple case of cloning).
15:45 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/f6f2e8a09a
15:45 dalek nqp/bs: f31afe1 | jnthn++ | src/6model/serialization. (2 files):
15:45 dalek nqp/bs: Initial changes to add a closures table.
15:45 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/f31afe1b6a
15:45 dalek nqp/bs: 317ff89 | jnthn++ | src/6model/serialization. (2 files):
15:45 dalek nqp/bs: Start sketching out cloned code ref serializtion, in prep for really doing closure support.
15:45 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/317ff89cfe
15:45 dalek nqp/bs: b81e767 | jnthn++ | src/6model/serialization.c:
15:45 dalek nqp/bs: Add initial code for deserializing a row from the closures table; this will eventually grow to be the full closure deserialization, but doing something much simpler for now.
15:45 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/b81e7671e0
15:45 dalek nqp/bs: be303bd | jnthn++ | t/serialization/03-closures.t:
15:45 dalek nqp/bs: Update test case to actually really use the static code object.
15:45 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/be303bd653
15:45 dalek nqp/bs: bf9b58e | jnthn++ | src/pmc/nqplexinfo.pmc:
15:45 dalek nqp/bs: Get NQPLexInfo to remember the Sub it is attached to.
15:45 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/bf9b58e23e
15:45 dalek nqp/bs: 664e87c | jnthn++ | src/6model/serialization.c:
15:45 dalek nqp/bs: Locate static code ref that a closure is based on, plus a few other fixes. First tests for cloned code objects now pass; this is the first pre-requisite for closure serialization done.
15:45 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/664e87c369
15:57 dalek parrot/714/scheduler_bug: 418a184 | dukeleto++ | src/platform/generic/time.c:
15:57 dalek parrot/714/scheduler_bug: Attempt to fix a fencepost error, original patch from he32++
15:57 dalek parrot/714/scheduler_bug: review: https://github.com/parrot/parrot/commit/418a184b83
15:58 PerlJam Where's Parrot's list of gsoc ideas?
16:01 PerlJam (I'd like to include a link in an email to the local university CS club)
16:01 seme thanks guys... I'll have to take a look at that a little more
16:02 seme I actually thought that was an error in the encoding but it was intentional :)
16:03 benabik joined #parrot
16:03 trey joined #parrot
16:16 dmalcolm joined #parrot
16:26 dukeleto PerlJam: don't think we have one yet
16:26 PerlJam Bummer.
16:26 dukeleto PerlJam: we have some of github issues marked with "gci"
16:26 dukeleto PerlJam: and I am sure our idea list from last year is on the wiki and still relevant
16:27 dukeleto PerlJam: https://github.com/parrot/parrot​/wiki/Google-Code-In-Task-Ideas
16:27 dukeleto PerlJam: that is GCI, not GSOC, but many tasks could be in either
16:27 dukeleto PerlJam: care to start a GSOC wiki page?
16:28 PerlJam My email to the cs club today was just to get them a little fired up.   The next one I send will be when they can actually do something besides start a dialog with an open source community.
16:29 PerlJam dukeleto: ENOTIME  (or ... when I do have time I will have likely forgotten about gsoc because something else will be weighing on my mind)
16:47 dukeleto PerlJam: well, just tell the cs club people to come talk to us. I will leave it at that.
16:47 PerlJam will do
16:49 seme hey guys... is there a way to identify on what parrot is failing... I am getting the error "Invalid character in ASCII string" but I can't tell where to look
16:49 seme I can tell that it is occuring while it is processing the Grammar.pg file
16:59 dukeleto seme: inteeresting
16:59 dukeleto seme: can you gist the full output of the command you are running and the full error?
16:59 seme sure one sec
16:59 dukeleto seme: along with OS info and parrot version
17:02 seme http://pastebin.com/bqAhXpCu
17:02 seme let me know if that is enough
17:03 seme I looked at the osutils file but my PIR understanding isn't that great yet... I couldn't devine from that what was happening
17:09 benabik joined #parrot
17:12 benabik o/ #parrot
17:16 dukeleto benabik: hola
17:16 dukeleto seme: that error really sucks
17:17 seme he hee yeah
17:17 dukeleto seme: and pynie hasn't seen any love lately
17:17 seme I'm trying to learn and give it some ;)
17:17 dukeleto seme: thanks!
17:17 dukeleto benabik: can you impart some wisdom on http://pastebin.com/bqAhXpCu ?
17:17 seme I've basically started rewriting it step by step with the intent of leanrning and recreating what is there
17:18 dukeleto seme: that sounds like a good learning exercise
17:18 seme hopefully that will enable me to work with it
17:18 dukeleto seme: part of the problem is that pynie is built on deprecated foundations
17:18 dukeleto seme: PGE (parrot grammar engine) is somewhat elderly and not worked on anymore
17:18 seme well what would be a good example of how to write a language using a modern version of parrot?
17:19 dukeleto seme: the squaak tutorial
17:19 seme perhaps I can just start by following those practices and use the existing pynie and puffin as a guide
17:19 dukeleto seme: let me find a good link
17:19 seme ooh ok great.  I went through that and I think I had a good idea of what was going on
17:19 dukeleto seme: there are old version of it around
17:19 seme there is a bit black magic going on around dealing with indentation... I'm trying to understand that right now
17:20 dukeleto seme: hmm. If squaak is too easy, maybe cardinal would be better suited to you: https://github.com/parrot/cardinal
17:20 dukeleto seme: cardinal is ruby on parrot
17:20 dukeleto seme: yes, i can't say that i understand all the indentation black magic
17:20 dukeleto seme: a good policy is to "git blame" code and see who touched it last
17:20 dukeleto seme: and then send an email to parrot-dev with questions and cc them if you can find their email
17:21 dukeleto seme: you can find peoples contact info in the CREDITS file
17:21 seme thanks... I'm not sure I can write an intelligent question about it just yet... still have a lot of reading to do :) I hate asking until I've exhausted the docs and google :)
17:23 seme I'm thinking of going back and getting my masters... this is really the kind of thing that I love but I'm not sure if getting a degree would allow me to find work related to this kind of thing
17:23 dukeleto seme: sure, i understand that. It is a good way to learn. But don't be afraid to ask questions :)
17:23 dukeleto seme: benabik is working on a grad degree on this stuff, he might be some help
17:23 dukeleto iirc
17:23 dukeleto seme: my degree was in math :)
17:23 seme benabik: any thoughts?
17:24 * dukeleto can't find the new squaak tutorial
17:25 dukeleto this doesn't bode well for outsiders trying to find our docs
17:25 seme I may have followed the older tutorial
17:25 seme http://coolnamehere.com/geekery/parrot/learn/ looks interesting
17:26 dukeleto seme: http://docs.parrot.org/parrot/​latest/html/pct_tutorial.html
17:26 davidfetter joined #parrot
17:26 dukeleto seme: yes, that tutorial is good too. I help write some of it :)
17:26 dukeleto davidfetter: wazzup, stranger?
17:26 dukeleto davidfetter: still in PA?
17:27 davidfetter dukeleto, not atm. WFH so Sarah can use the car for appointments today
17:27 seme is it even possible for people to get paid working on this stuff?
17:27 dukeleto seme: define "this stuff"
17:27 seme language design/implementation, etc
17:27 davidfetter PTO tomorrow--need to bring $son to Milwaukee (the two-e one, not the two-i suburb of Portland) for obligatory showing-off this weekend
17:28 davidfetter seme, just about anything you can imagine, someone gets paid to do :)
17:28 dukeleto seme: i got paid by The Perl Foundation to hack on a parrot grant. Wasn't a full time job, but I did get paid :)
17:28 dukeleto seme: i would say that people who understand languages and compilers at a deep level shouldn't have a problem finding a job
17:28 seme he hee... yeah but as a career?
17:29 seme dukeleto: finding a job isn't a difficult issue.. but doign what you like well now that is a completely different thing
17:29 dukeleto seme: everybody is different. I am pretty sure my extensive involvement in parrot helped me get my current job
17:29 seme :-)
17:30 seme and if you get nothing else out of it, well I think you're swell ;)
17:30 davidfetter dukeleto, so i'm wondering about differential parse tree attacks
17:30 dukeleto seme: thanks!
17:30 dukeleto davidfetter: GO ON....
17:30 dukeleto seme: i try my best. Are you currently enrolled in any kind of university?
17:30 dukeleto seme: if so, we want you in Google Summer of Code :)
17:31 seme dukeleto: no i graduated a long time ago... I've been working in finance but find it less than satisfying...
17:31 davidfetter well, according to meredith patterson, an old pg hand and security expert, finding where different implementations of a spec do different things is a "weaponizable" way to find 0-day exploits
17:31 davidfetter so naturally, i thought of perl6
17:31 dukeleto davidfetter: interesting
17:32 dukeleto davidfetter: so what is up with "parse tree attack"? Describe a simple parse tree attack
17:32 dukeleto davidfetter: is this like a third cousin to hash table DOS?
17:32 davidfetter http://www.cs.dartmouth.edu/~serg​ey/langsec/papers/langsec-tr.pdf
17:34 dukeleto We have developed a powerful technique to enhance code auditing and protocol analysis, known as the parse tree differential attack [8], wherein we give two different implementa- tions of the same specification identical state and input parameters, consider their decodings as concrete parse trees, and enumerate the differences between the trees. Deviations be- tween the trees indicate potential problems,
17:34 dukeleto e.g. an area of implementor discretion due to specification ambiguity or an implementation mistake.
17:35 dukeleto ok, i understand now
17:35 dukeleto davidfetter: very interesting
17:35 dukeleto davidfetter: yeah, perl 6 is like a playground for that kind of attack research
17:35 davidfetter that's what i get for a couple of decades' worth of hanging around with security geeks, even though i've never been one
17:36 dukeleto davidfetter: that attack relies on the ability to actually get a parse tree, though. But usually, that is possible
17:36 davidfetter yes, it is
17:36 dukeleto davidfetter: good thing we change the way our ASTs work frequently enough to keep those blackhats busy ;)
17:37 davidfetter heh
17:37 davidfetter actually, i was thinking in terms of using perl6, etc., for this "weaponization."
17:38 davidfetter the best defense is, in this case literally, a good offense. finding those attacks before the black hats have them makes you more secure, not less
17:39 dukeleto davidfetter: but how do you compare the parse trees of different perl 6 implementations?
17:39 dukeleto davidfetter: they are not going to be in the same formate
17:39 dukeleto format, rather
17:39 davidfetter not quite what i had in mind
17:39 dukeleto davidfetter: for instance, Niecza (perl 6 on mono) has a totally different AST than Rakudo, iirc
17:39 dukeleto davidfetter: what did you have in mind?
17:39 davidfetter more like using p6 to find such attacks
17:39 dukeleto davidfetter: ah! Ok. I like it.
17:40 dukeleto davidfetter: they *do* need a killer app already...
17:40 * dukeleto ducks
17:40 davidfetter although of course if it can find likely attacks on itself, there's value to that, too
17:41 Coke in backscroll - my involvement in open source helped me get at least one of my jobs.
17:41 * davidfetter has this awful feeling he's treading close to things that are UNDECIDEABLE
17:41 Coke (but I find I can do better financially if I work first, and do OS as a subset, or on the side.)
17:41 davidfetter seme, i haven't earned a nickel this century outside what i do around open source
17:42 benabik My ears are burning...
17:42 davidfetter seme, sometimes i work on proprietary software, but neither much nor often, and i'm frequently helping people open the source to stuff that was proprietary
17:44 davidfetter dukeleto, does doing that kind of analysis fit with your understanding of what p6 is good at?
17:44 dukeleto davidfetter: sure, it sounds like something p6 would be good at
17:44 * davidfetter aware that any turing-complete language can do any task that any other can, but there's a suitability question beyond that
17:45 dukeleto seme: any university student can take part in GSoC, so if you go back to school, you could get paid by google to hack on open source
17:45 dukeleto seme: times surely have changed
17:45 dukeleto seme: nobody was paying anybody to learn when I was in school...
17:46 * davidfetter suspects that somewhat larger sums are available from google just by checking the job postings that include open source
17:46 davidfetter seme, what have you been working on in open source lately?
17:46 dukeleto some parts of google are OSS-friendly, other parts, not so much
17:46 dukeleto GSoC offsets all the non-OSS-friendly parts of google, or tries to, anyway
17:47 dukeleto googlers have explained to me that google would like to opensource lots of stuff, but most of their code is so tightly coupled to their crazy infrastructure that it is not worth the effort
17:47 davidfetter dukeleto, an interesting slide deck on $subject: http://www.cosic.esat.kuleuven​.be/publications/talk-186.pdf
17:47 benabik dukeleto: It _looks_ like Parrot thinks there's a non-ASCII character in Grammar/Grammar.pg
17:48 dukeleto benabik: would you agree that that error message should die in a fire?
17:48 dukeleto benabik: a line number would be nice...
17:48 benabik dukeleto: The line it's claiming the error is from is ifh.'encoding'(readall)
17:48 benabik dukeleto: I think it's a from Parrot error, not a from PCT error.
17:49 dukeleto seme: take a look at tools/dev/dedeprecator.nqp
17:49 dukeleto seme: it searches for deprecation code constructs. Running it on pynie might lead to some revelations
17:49 dukeleto s/deprecation/deprecated/
17:50 dukeleto seme: it uses https://github.com/parrot/p​arrot/blob/master/api.yaml
17:57 benabik dukeleto: It _looks_ like parrot thinks there's a byte >= 0x80 in Grammar/Grammar.pg, but I don't see it.
18:01 dukeleto Useless fact of the day: TLDR is also the Texas Department of Licensing and Regulation
18:02 benabik Oh.
18:04 benabik dukeleto: PGE is expecting an ASCII file, but it's getting a UTF-8.  --encoding=utf8 needs to be passed to PGE.
18:05 * benabik has to do $work for a bit..
18:07 dmalcolm joined #parrot
18:13 dukeleto seme: did you catch that?
18:22 dukeleto seme: i think our default encoding changed since pynie was last updated
18:22 dukeleto seme: PGE expects ascii by default, but i think the rest of parrot now defaults to utf8
18:22 benabik dukeleto: setup.pir: add after line 56: $P0['pir_pge_flags'] = '--encoding=utf8'
18:23 dukeleto benabik++
18:23 benabik dukeleto: Untested, of course.  :-D
18:26 benabik As a random note:  To find a range of characters by hex value in vim:  "/[" <Ctrl-V> "x80-" <C-V> "xff]" <Enter>
18:32 lucian joined #parrot
18:45 dalek nqp/bs: e4f5e80 | jnthn++ | t/serialization/03-closures.t:
18:45 dalek nqp/bs: Add a failing test for closure serialization.
18:45 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/e4f5e80c0e
18:45 dalek nqp/bs: 40269bf | jnthn++ | src/6model/serialization.c:
18:45 dalek nqp/bs: Add a column to the closures table for the outer context.
18:45 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/40269bf764
18:45 dalek nqp/bs: 9bdbaa1 | jnthn++ | src/6model/serialization. (2 files):
18:45 dalek nqp/bs: Update serialize/deserialize code to set up for contexts table and data segment.
18:45 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/9bdbaa175b
18:45 dalek nqp/bs: 5ee3a41 | jnthn++ | t/serialization/03-closures.t:
18:45 dalek nqp/bs: Correct test count.
18:45 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/5ee3a41791
19:10 dalek parrot: 0f1422c | jkeenan++ | src/hash.c:
19:10 dalek parrot: [codingstd] Rebreak line to conform to linelength standard.
19:10 dalek parrot: review: https://github.com/parrot/parrot/commit/0f1422c233
19:10 dalek parrot: ad65ada | jkeenan++ | src/hash.c:
19:10 dalek parrot: [codingstd] Rebreak line to avoid cuddled else.
19:10 dalek parrot: review: https://github.com/parrot/parrot/commit/ad65ada64b
19:10 dalek parrot: 2cabfb6 | jkeenan++ | src/hash.c:
19:10 dalek parrot: [codingstd] No space permitted before closing paren.
19:11 dalek parrot: review: https://github.com/parrot/parrot/commit/2cabfb6815
19:11 dalek parrot: 118b706 | jkeenan++ | src/hash.c:
19:11 dalek parrot: Revert "[codingstd] No space permitted before closing paren."
19:11 dalek parrot:
19:11 dalek parrot: This reverts commit 2cabfb6815c86ed86056edabdff0b14cb804b302.
19:11 dalek parrot: review: https://github.com/parrot/parrot/commit/118b706e91
19:13 cogno joined #parrot
19:15 zby_home joined #parrot
19:17 ttbot Parrot 2cabfb68 i386-linux-thread-multi make error http://tt.taptinder.org/cmdinfo/67941
19:22 dukeleto ttbot lives!
19:23 dukeleto msg jkeenan looks like your recent codingstd fixes broke the build
19:23 aloha OK. I'll deliver the message.
19:25 Coke seen ttbot?
19:26 aloha ttbot was last seen in #parrot 8 mins 25 seconds ago saying "Parrot 2cabfb68 i386-linux-thread-multi make error http://tt.taptinder.org/cmdinfo/67941".
19:26 Coke oh, there it is!
19:26 Coke vanished in the dalekspew.
19:58 dmalcolm joined #parrot
20:06 lucian joined #parrot
20:55 seme why is the default for setup.pir to build the artifacts in the same directory as the primary code
20:58 seme nice plumage dies when you type help...
20:58 seme is plumage an active part of parrot any more?
21:00 Coke I believe fperrad was the main developer on that, non?
21:04 seme does parrot have a bug/issue site?
21:05 tadzik it uses github issues
21:05 seme ooh... I looked and only saw 5 closed issues so I didn't think it was being used
21:06 dalek parrot: f4f5025 | jkeenan++ | / (2 files):
21:06 dalek parrot: 'ARGMOD' not appropriate with 'const'; must use 'ARGIN'.
21:06 dalek parrot: review: https://github.com/parrot/parrot/commit/f4f5025c23
21:07 seme never mind... I see
21:07 seme was looking at the wrong repo
21:08 seme do you guys ever get together and talk about the future of parrot... I'd be interested in hearing more about what is going on and where the project is going... I have been reading the blogs but was hoping for more ;)
21:10 tadzik whiteknight++ has written some nice posts about it
21:12 NotFound seme: we are here together ;)
21:13 dalek parrot: 3bc818b | jkeenan++ | / (2 files):
21:13 dalek parrot: [codingstd] No hard tabs.
21:13 dalek parrot: review: https://github.com/parrot/parrot/commit/3bc818bc59
21:25 NotFound msg alvis I can't reproduce your problem
21:25 aloha OK. I'll deliver the message.
21:25 Coke seme: i'd try to drum up interest in BOF again at YAPC.
21:27 * benabik wishes he could go to YAPC::NA this year.
21:32 dukeleto benabik: it was a good time last year
21:33 * dukeleto hasn't figured out if he can make it this year, yet
21:33 seme I've never gone... when/where is the next one... I only see conferences for 2011 on the yapc site
21:33 benabik dukeleto: Indeed it was.  But this summer I'm probably moving, selling a house, and having a kid.  Just no time..
21:33 dukeleto seme: japhb was the main dev of plumage, but he hasn't been around much lately. I hacked on plumage a bunch as well
21:33 benabik seme: yapc2012.us IIRC
21:33 dukeleto seme: we definitely want plumage to work, but it isn't hooked up to any continuous integration, YET
21:34 dukeleto seme: so it might have a bit of bitrot
21:34 seme hrm... would be nice to get something like that going... anyone working on it?
21:34 NotFound I think the problem with --help was already informed somewhere.
21:34 dukeleto seme: i have plans to make another parrot developer summit happen, but that is currently a pipedream
21:35 dukeleto seme: http://travis-ci.org/#/parrot/parrot
21:35 dukeleto seme: that is a free CI system for open source projects
21:35 dukeleto seme: i have parrot.git on there, but not the rest of the ecosystem, yet
21:35 dukeleto seme: it is all controlled by this config file: https://github.com/parrot/pa​rrot/blob/master/.travis.yml
21:37 benabik I keep pondering writing a "my ideal parrot" blog post.  I think plumage should go on the list.  Good modules make for a good VM.
21:38 seme he hee... there is a lot of knowledge out there about parrot... it would be nice if more of this were organized/maintained so new guys like me can get involved easily....
21:38 benabik I should probably add a link to my cish "tutorial" to the wiki or something.
21:39 dukeleto seme: and adding a travis post-receive hook
21:39 dukeleto seme: that is what parrot.github.com is supposed to be
21:39 dukeleto seme: we want that to be a portal of useful docs
21:39 benabik And a tutorial based around using winxed would be nice too.  :-
21:39 benabik :-/
21:39 dukeleto benabik++ # winxed tutorial
21:39 dukeleto i would prefer if all people brand new to Parrot saw Winxed first
21:39 dukeleto that is our shiny shiny
21:40 dalek nqp/bs: 44d674d | jnthn++ | src/6model/serialization. (2 files):
21:40 dalek nqp/bs: First crack at one-level-deep closure serialization; survives, though not done deserialization yet.
21:40 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/44d674d1f3
21:40 benabik dukeleto: It shouldn't be too hard to do.  I'm fairly sure that you could use NQP for a grammar and then winxed for everything else..
21:40 seme he hee... I tried walking through winxed but the docs didn't really help much so I just kept on with teh squaak tutorial :)
21:41 NotFound seme: winxed is not particularly intended to writing languages.
21:42 dukeleto NotFound: but it *is* great for writing libraries for Parrot :)
21:42 dukeleto NotFound: and it *can* be used to write languages :)
21:43 NotFound dukeleto: of course... to write winxed, for example.
21:43 * dukeleto mumbles about turtles all the way down...
21:43 seme well gents.. I will try to get back on in a bit to work on pynie :)
21:44 dukeleto seme: awesome!
21:45 dukeleto seme: feel free to email parrot-dev questions. we don't bite :)
21:45 dukeleto seme: and many smart people are on the list but don't do IRC
21:49 NotFound Pesky macros
22:00 dalek parrot: 06e7078 | NotFound++ | include/parrot/hash.h:
22:00 dalek parrot: make parrot_hash_iterate_... usable like functions to decreeas dangling else and semicolon placing problems
22:00 dalek parrot: review: https://github.com/parrot/parrot/commit/06e70789b3
23:16 dalek nqp/bs: 0505b08 | jnthn++ | src/6model/serialization.c:
23:16 dalek nqp/bs: Don't attempt to serialize an outer that'll simply never work out.
23:16 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/0505b08177
23:16 dalek nqp/bs: cf55512 | jnthn++ | src/6model/serialization.c:
23:16 dalek nqp/bs: Start to sketch out overall closure deserialization process.
23:16 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/cf55512f22
23:16 dalek nqp/bs: 3b9c867 | jnthn++ | src/6model/serialization.c:
23:16 dalek nqp/bs: First crack at closure deserialization. Passes one of the tests, fails another...something still a bit off.
23:16 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/3b9c86796c
23:16 dalek nqp/bs: 84c2256 | jnthn++ | src/6model/serialization.c:
23:16 dalek nqp/bs: Fix a couple of silly thinkos; basic closure serialization tests now pass.
23:17 dalek nqp/bs: review: https://github.com/perl6/nqp/commit/84c2256ba6
23:37 whiteknight joined #parrot
23:39 whiteknight good evening, #parrot
23:40 tadzik good evening
23:45 whiteknight hello tadzik, how are you doing tonight?
23:45 tadzik not bad. I have the One Last Exam tomorrow on 8 AM, and I'm now considering whether to go to sleep at all, or not

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

Parrot | source cross referenced