Camelia, the Perl 6 bug

IRC log for #parrot, 2009-04-15

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:03 Xrblackw1 joined #parrot
00:06 megazooch joined #parrot
00:09 AndyA joined #parrot
00:09 Xrblackwe joined #parrot
00:11 TiMBuS joined #parrot
00:14 bacek_ joined #parrot
00:30 rblackwe_ joined #parrot
00:42 wayland76 joined #parrot
00:55 Xrblackwe joined #parrot
01:09 rg do we have any x87 experts?
01:16 amoc joined #parrot
01:16 Infinoid x87?  the math coprocessor?
01:16 rg yep. the one.
01:17 Infinoid not me
01:17 rg that's what's causing TT #530 and TT #501
01:18 rg (and TT #38 and TT #201 before)
01:21 confound joined #parrot
01:22 confound joined #parrot
01:25 Andy joined #parrot
01:41 Xrblackwe joined #parrot
01:42 LylePerl_ joined #parrot
01:42 LylePerl_ hi
01:43 davidfetter hi
01:48 LylePerl_ I'm trying to write a bit of parrot
01:48 LylePerl_ I've found the page http://docs.parrot.org/parrot/l​atest/html/src/pmc/os.pmc.html
01:48 shorten LylePerl_'s url is at http://xrl.us/beo2ad
01:49 LylePerl_ I want to display *cwd and then do a chdir()
01:49 LylePerl_ I've tried print *cwd but that gives error
01:50 cotto LylePerl_, which version of Parrot are you using?
01:51 LylePerl_ latest
01:51 cotto svn?
01:51 purl well, svn is the current svn (see git)
01:51 LylePerl_ yes
01:52 cotto Ok.  Can you nopaste the code you're working on?  Using tools/dev/nopaste.pl is the easiest way.
01:53 nopaste "cosmic" at 78.86.207.240 pasted "cwd" (16 lines) at http://nopaste.snit.ch/16273
01:54 LylePerl_ the commented lines are bits I've got working in the past
01:54 LylePerl_ at the moment I just need to figure out how to use cwd and chdir from that page
01:55 cotto cwd is a method of the OS PMC.
01:55 Infinoid os.'cwd'()
01:56 cotto Look at t/pmc/os.t for some complete examples.
01:56 wayland76 joined #parrot
01:56 nopaste "cosmic" at 78.86.207.240 pasted "cwd" (16 lines) at http://nopaste.snit.ch/16274
01:56 LylePerl_ oops, sorry didn't mean to paste that
01:56 cotto np
01:57 LylePerl_ hmm, getting undefined identifier 'os'
01:58 LylePerl_ I'll look at those test scripts
01:59 LylePerl_ yea, perl :)
01:59 dalek parrot: r38108 | jkeenan++ | branches/gc_options_remove:
01:59 dalek parrot: Create branch to work on https://trac.parrot.org/parrot/ticket/490.
01:59 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38108/
01:59 cotto if you run prove t/pmc/os.t after building Parrot, it'll generate the PIR/PASM files as t/pmc/os_n.pir
02:05 LylePerl_ excellent! Thank you very much. cotto++
02:08 cotto LylePerl_, you're welcome.  ooc, what's your interest in Parrot?
02:19 dalek parrot: r38109 | jkeenan++ | branches/gc_options_remove (5 files):
02:19 dalek parrot: Remove command-line '--gc' option, associated tests and documentation.
02:19 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38109/
02:21 LylePerl_ cotto: Rakudo development
02:21 LylePerl_ :)
02:22 dalek parrot: r38110 | jkeenan++ | trunk/lib/Parrot/Ops2pm.pm:
02:22 dalek parrot: Restoring previously deleted code, per Coke's comment in https://trac.parrot.org/parrot/ticket/489.
02:22 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38110/
02:25 LylePerl_ Hi Infinoid, only just noticed it was you that posted os.'cwd'()
02:37 Infinoid hi LylePerl_
02:37 * Infinoid is doing 10 things at once, but tries to do the occasional flyby
02:42 janus joined #parrot
02:55 szabgab joined #parrot
03:15 eternaleye joined #parrot
03:17 integral joined #parrot
03:21 raiph joined #parrot
03:32 Infinoid rg++
03:33 Infinoid tewk: any chance you can comment on https://trac.parrot.org/pa​rrot/ticket/530#comment:5 ?
04:10 Andy joined #parrot
04:15 tetragon joined #parrot
04:15 wayland joined #parrot
04:16 dngor_ joined #parrot
04:16 tewk_ joined #parrot
04:17 frodwith_ joined #parrot
04:17 lucs_ joined #parrot
04:18 TonyCToo joined #parrot
04:18 Woody4286 joined #parrot
04:21 mmpf joined #parrot
04:21 jsut joined #parrot
04:21 basic joined #parrot
04:21 Aisling joined #parrot
04:21 particle1 joined #parrot
04:21 purl joined #parrot
04:21 Xrblackwe joined #parrot
04:21 Tene joined #parrot
04:21 samlh joined #parrot
04:53 Tene joined #parrot
04:57 masak joined #parrot
04:58 flh joined #parrot
05:19 buildbot joined #parrot
06:06 Tene joined #parrot
06:13 cotto LLVM is certainly the flavor of the month.
06:13 cotto There are at least 3 separate proposals for LLVM JIT backends for various language-related projects.
06:21 uniejo joined #parrot
06:29 UltraDM joined #parrot
06:52 iblechbot joined #parrot
07:00 krunen joined #parrot
07:32 jimmy joined #parrot
08:50 wayland joined #parrot
09:15 particle2 joined #parrot
09:50 cotto seen darbelo
09:50 purl darbelo was last seen on #parrot 1 days, 17 hours, 46 minutes and 55 seconds ago, saying: �  [Apr 13 16:01:17 2009]
09:50 cotto ³?
10:30 wayland It means that, whatever the previous comment was, he thinks it should be cubed :)
11:01 user joined #parrot
11:01 user hi everyone
11:01 purl Howdy, user, you fantastic person you.
11:02 user I was just playing with parrot, got stuck somewhere need somehelp
11:02 user is that a bot?
11:02 jimmy yes
11:02 user failed the turing test on first go then
11:03 user anyways i was reading through the Episode 1 of implementing Sqaak
11:04 user i tried out  perl tools/dev/mk_language_shell.pl Squaak
11:04 user but the makefile is not getting generated
11:04 user its saying Can't open perl script "/usr/local/lib/parrot/1.0.0-de​vel/tools/dev/gen_makefile.pl": No such file or directory
11:04 user i tried to build cardinal too, but same problem
11:06 user anyone there?
11:08 wayland I'm here
11:09 wayland now, anyway
11:09 purl i guess now is the time for the Kwisatz Haderach to arise and conquer America on a sandworm army
11:09 wayland Question: does that file that it's complaining about exist/
11:09 wayland ?
11:09 user no it does not
11:10 user i compiled parrot from svn
11:10 user should it be there?
11:10 user in fact there is no tools folder in /usr/local/lib/parrot/1.0.0-devel/
11:10 purl Hmm.  No matches for that, user.
11:12 user purl can you shut up
11:12 purl user: what?
11:12 szbalint purl++
11:13 wayland I doubt she can -- she gets annoying, but abuse doesn't seem to help
11:13 user wayland can you help me with my problem
11:14 wayland When you did make install, did you do "make install" or "make install-dev"
11:14 wayland I doubt I can help, but I can try
11:14 user i did make install
11:15 wayland The install system is the only part of parrot that I know anything about
11:15 wayland try make install-dev
11:15 user good i shall pester you once i get started then :)
11:15 user ya will try that
11:15 wayland Doing that helped me with Rakudo one time when there were some files missing
11:17 wayland Well, as soon as I get a Parrot RPM that I can build a Rakudo RPM on top of (and tidy up any messes that I've left :) ), I'm going to leave this channel and just hang out on #perl6, so if you're looking for me, I'll be there
11:18 user yup i see a tools folder, thanks
11:18 user I'm more interested in cardinal, i'm a Ruby kid
11:19 wayland Well, my planned #perl6 projects that are modular will allow hacking in any language :)
11:20 wayland Within reason.  I don't mean Chef, Intercal, or OISC :)
11:20 wayland What's Sqaak?  Google failed me
11:22 user ohh i read this article introducing Parrot VM http://www.parrotblog.org/200​8/03/targeting-parrot-vm.html
11:22 user squaak is there test case
11:23 bacek joined #parrot
11:23 wayland ok
11:23 bacek good evening
11:24 wayland Why do I always want to give the Gandalf response to good evening/morning :)
11:24 bacek :)
11:35 ruoso joined #parrot
12:00 mib_rp1p42 joined #parrot
12:01 mib_rp1p42 tosif, u there ?
12:03 mib_pot1wp joined #parrot
12:04 mib_rp1p42 hi tosif ?
12:06 bacek joined #parrot
12:08 bacek seen jonathan
12:08 purl jonathan was last seen on #parrot 3 days, 20 hours, 33 minutes and 8 seconds ago, saying: I'd love that to change, but like everything, time, effort. :-)  [Apr 11 15:33:17 2009]
12:09 dalek parrot: r38111 | bacek++ | branches/packfile_revamp/i​nclude/parrot/packfile.h:
12:09 dalek parrot: Generate PASM macros for PF_ANNOTATION_KEY_TYPE
12:09 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38111/
12:09 dalek parrot: r38112 | bacek++ | branches/packfile_revamp/src/p​mc/packfileannotationkeys.pmc:
12:09 dalek parrot: Initial naive implementation setters for PackfileAnnotationKeys
12:09 bacek *incoming* :)
12:09 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38112/
12:09 dalek parrot: r38113 | bacek++ | branches/packfile_revamp (3 files):
12:09 dalek parrot: Implement packing of PackfileAnnotations
12:09 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38113/
12:11 wayland bacek: It's nothing like the commit message that autarch/Dave Rolsky had for S32/Temporal :)
12:11 bacek wayland: Is it bigger than actual diff? :)
12:12 wayland Well, no, but it was more than a screenful of IRC
12:12 wayland maybe 30 lines or so
12:13 wayland Very useful description of the changes, though :)
12:37 rg1 joined #parrot
12:48 Util allison: ping
12:51 Util purl: msg allison (From yesterday) In chapter 13 (Appx D), I have much planned, but have committed everything actually written. In other chapters, I only have paper notes of which sections need wholesale revision (like Calling Conventions). So, I expect no conflict. For future (after Ch13) edits bigger-than-a-typo, I will announce my intent, or I will proceed however you advise.
12:51 purl Message for allison stored.
12:58 bacek joined #parrot
12:58 bacek hi again
12:58 purl oh, you're back!
12:59 wayland No, he's bacek, can't you tell the difference?
12:59 bacek She is stupid enough for this...
12:59 wayland (bacek: is your name supposed to have non-ascii markers?)
13:00 bacek wayland: it's can be written in Cyrillic "as is" :)
13:01 wayland ok.  And the pronunciation?  I'm inquiring particularly about the 'c'
13:01 bacek It's read as "s" in Russian.
13:02 bacek roughly "bacek" is "vasek". With accent on "e"
13:03 wayland Ok, kind of like "Vasil*" => "Basil"? ( I put the * there because I think there's some ending, but can't remember)
13:04 Coke bacek - you from .ru ?
13:04 bacek It's "й". Translates as "y" or "iy".
13:05 bacek Coke: originally - yes. Now - .au :)
13:05 docmach joined #parrot
13:05 Coke bacek - apparently my family is from belarus. (Koleda)
13:05 bacek wayland: btw, Vasily is Greek name.
13:06 bacek Coke: Ah! Do you pronounce your surname with accent on "a"?
13:07 Coke we pronounce it with an emphasis on the long e... but we were imported several generations ago.
13:07 Coke (no real connection to the "homeland". =-)
13:07 wayland bacek: I'll keep that in mind.  The guy I know whose father has that name was fairly involved in the Russian Orthodox church; I could see (from my ignorant POV) that resulting in a Greek name
13:08 Coke ha. reading http://en.wikipedia.org/wiki/Koleda, I see :
13:08 Coke Another speculation is that it derived from the Bulgarian/Macedonian word "..../....." (kolia/kolam), which means "to rip, to kill (a man), to cut animal for eating"
13:08 Coke Nice.
13:08 bacek Coke: fair enough. It's easy for English speakers.
13:08 wayland If I leave off my first name, and use the original spelling for my surname, I'm Sven Nilssen (not sure about the double-s)
13:09 Coke . o O (Diado Koleda)
13:11 bacek wayland: "Vasily" derived from "Basilevs" which means something like "kingly" :)
13:11 bacek Coke: :)
13:11 wayland Oh, yes, now I remember that much Greek :).  I only ever did 17 lessons of Greek, and that was 8 years ago
13:22 bacek purl: tell jonathan Packfile PMC are mostly done.
13:22 purl OK, bacek.
13:29 rg bacek: i was wondering, will your packfile rework solve the lexicals problem (RT #60652, RT #60650)?
13:30 bacek rg: not yet. At least packfile_revamp branch targeting only Packfile PMCs.
13:30 bacek rg: OTOH I've got more information about bytecode guts now, so I can take a look at those tickets.
13:31 Infinoid bacek++
13:32 bacek Infinoid: not yet. BTW, branch ready to review and landing back.
13:34 Infinoid ok, I'll try to get some review-shaped tuits today.  It looks like you've made a ton of commits :)
13:34 bacek Infinoid: not actually. Something like 40-50 only :)
13:34 Infinoid Hmm, it'd probably just be easier to just make a checkout and read the code, rather than going sequentially through commits
13:35 bacek Most of them in src/pmc/packfile*pmc and t/pmc/packfile*t
13:35 szabgab joined #parrot
13:35 bacek So you can just check current code :)
13:36 Infinoid Great, will do
13:37 Infinoid hmm, I have a git-svn checkout with all branches, but when I clone that, the new clone only gets the master branch.  how do I mark branches as exported?
13:42 Infinoid in the git-svn checkout, I guess git doesn't consider them "local"
13:42 bacek Infinoid: git svn pull
13:42 bacek ouch.
13:42 bacek git svn fetch
13:43 bacek git checkout packfile_revamp_local origin/packfile_revamp
13:44 bacek But I didn't try it on "local" repos.
13:45 Infinoid "git branch packfile_revamp_local packfile_revamp" is enough to propagate the branch changes
13:45 Infinoid but as I'm trying to maintain a git mirror of everything, I'm looking for a cleaner solution
13:49 bacek Infinoid: "cleaner" as in?
13:50 Tene joined #parrot
13:50 bacek You can always create local branch, reset it to svn branch, and then remove it.
13:51 Infinoid first problem is that "git svn --std-layout" created a bunch of remote branches without the "origin/" prefix
13:52 Infinoid second problem is I don't know how to automatically make (and update) local branches from the remote ones, such that those changes are propagated to other clones when pulled
13:52 Infinoid --stdlayout, I mean
13:52 PerlJam Infinoid: that's why you should always use the --prefix optino to git-svn
13:52 bacek Infinoid: ah. Ok. git checkout -b tempbranch; git svn reset --hard remotes/svn_branch
13:52 Infinoid PerlJam: nice, thanks.  time to make a new checkout then
13:52 bacek or it's just "git reset"... Let me check
13:53 PerlJam I don't know of a way to automatically make local tracking branches from remote ones.
13:53 Infinoid I can write a script to do that
13:53 PerlJam you probably have to.
13:53 Infinoid looping around the output of git branch -r, I guess
13:54 Infinoid thanks, guys.  I think I can make a working parrot-all repo from these ideas, to complement the parrot-trunk one I'm already running
13:56 bacek It's almost tomorrow here. See you soon :)
13:56 Coke Infinoid: a wiki page on setting up a local svn-git mirror on trac.parrot.org would probably be helpful.
13:56 wayland I'm glad someone understands git :)
13:56 Coke (or a pointer on how to use yours if that's appropriate)
13:56 bacek afk # Zzzz...
13:57 Infinoid Coke: local git-svn is easy, but I can certainly make such a thing.  I'm just trying to make this one useful to the general public
13:57 Infinoid Still not a very good solution, this is all read-only unless I want to allow everyone to make commits under my name (which I don't).
13:58 PerlJam Infinoid: you could setup gitosis.
13:58 wayland That sounds like a disease :)
13:59 wayland bacek: 'night
14:01 PerlJam Infinoid: do you have access to the parrot svn repo?  I.e., can you make hooks?
14:01 Infinoid no
14:02 PerlJam bummer.
14:02 PerlJam you could probably setup a two-way gateway if you did.
14:02 PerlJam you could do it without access to the svn repo, but I'd worry a bit about consistency.
14:02 PacoLinux joined #parrot
14:02 davidfetter joined #parrot
14:03 PerlJam (I was thinking of a post-commit hook that triggers "git svn rebase --all" in the git repo)
14:04 Infinoid maintaining a personal git-svn mirror is a significant barrier to entry, anything I can do to reduce that overhead is good for parrot I think
14:04 Infinoid but I'd prefer not to mess with svn directly, I don't want to open any security problems
14:06 PerlJam Infinoid: you could use git-svn to create the git repo, turn it into a bare repo, setup a cron job to run "git svn rebase --all" once a minute, and a post-receive hook in the git repo to run "git svn dcommit".
14:06 PerlJam then people should be able to clone from that repo and have up-to-date parrot changes via git.
14:06 PerlJam and when they commit, their changes will be reflected in the svn repo
14:07 PerlJam (as well as the git repo of course)
14:07 PerlJam But there's probably all sorts of problems with that scheme.
14:07 PerlJam Though it makes a good first approximation I think
14:08 Infinoid Yeah, that's pretty close to what I'm doing, without the dcommit part
14:10 PerlJam If you add the post-receive hook, then it won't be so read-only.  Any changes pushed to the git repo will be pushed upstream to svn as well.
14:11 Infinoid yeah.  I'm doing that by hand at the moment
14:11 PerlJam (of course, you still have to worry about authentication)
14:11 Infinoid and of course, it's readonly to everyone but me
14:14 Infinoid msg bacek Is the intention that Packfile.get_string() should ignore the packfile format attributes?  I think it might not be reasonable to expect the old packfile code to generate pbcs with foreign byte orders and such... but perhaps the attrs should be reset to defaults in this case?
14:14 purl Message for bacek stored.
14:17 Infinoid msg bacek Also, it looks like the defaults for these things are all 0, if you're creating a new pbc (rather than importing one from a STRING) they will never be set.  I'm not sure whether that matters or not, what's your opinion?
14:17 purl Message for bacek stored.
14:21 Infinoid hmm, pdd13 seems a bit unclear about what the valid values of uuid_type are, and whether the uuid is stored in packed binary format or ascii (hex) encoded.  packfile.t seems to suggest it's hex
14:29 dalek parrot: r38114 | fperrad++ | trunk/t/codingstd/svn_id.t:
14:29 dalek parrot: [codingstd] add check for PIR files.
14:29 dalek parrot: See TT #554
14:29 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38114/
14:33 LylePerl joined #parrot
14:33 LylePerl hi
14:34 Coke LylePerl: hello
14:34 Coke (hooks to parrot svn repo) such things could conceivably be added.
14:39 Infinoid Thanks, Coke.  I'll keep that in mind, but I'd prefer solutions that stay as far away from the svn repo as possible, to reduce the risk of breakage
14:40 Coke as do i. =-)
14:42 Coke t/codingstd/svn_id.t now has a bunch of low hanging fruit.
14:43 Coke ... did not wait long enough.
14:43 Coke nevermind.
14:43 dalek parrot: r38115 | fperrad++ | trunk (165 files):
14:43 dalek parrot: [codingstd] add missing $Id
14:43 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38115/
14:43 Coke now it only has one failure. =-)
14:44 Infinoid My "less" command tries to render .cue files by running them through cdrecord, but it doesn't know how to render .pod files with perldoc.  How useful.
14:45 amoc joined #parrot
14:48 szabgab_ joined #parrot
14:49 Coke Infinoid: see the "INPUT PREPROCESSOR" section of 'man less'
14:49 Infinoid Thanks, the default configuration just seems silly :)
14:50 Infinoid Do parrot PMCs support multiple inheritance?  Is there a performance penalty?
14:51 Coke I think you can say 'extends' more than one PMC, yes.
14:51 Coke I have no idea how it works, though.
14:51 Infinoid I seem to remember some issues with inherited attribute ordering
14:53 particle- yes, but i'm not sure since we switched to ATTRs if that's still an issue
14:53 iblechbot joined #parrot
14:55 Infinoid There are quite a few vtable functions in bacek++'s PackfileDirectory that just proxy to Hash, which I thought might be cleaned up a bit by "extends Hash".  But I don't feel like opening that particular can of worms right now
14:56 szabgab joined #parrot
15:06 gryphon joined #parrot
15:07 particle- does PD expose the entire hash interface?
15:10 Coke There's no penalty for using lying and using UNUSED(foo), is there?
15:10 Coke aside from potentially confusing code?
15:10 particle- s:1st/using// ??
15:10 Coke I am pondering just slapping in UNUSED(interp) /* YMMV */ for the generated code.
15:11 Coke particle-: yes.
15:11 rg unfortunately, from what i've seen yesterday, no.
15:11 particle- why not #define MAYBE_UNUSED(x) UNUSED(x)
15:13 Coke heh.
15:13 Coke I suppose that's slightly less confusing for someone looking at the generated code.
15:14 particle- then document the macro, and the curious can make tags-vi and ctrl-t
15:14 particle- i mean vim -t MAYBE_UNUSED
15:15 Coke ... I wrote the comment first, damnit. =-)
15:16 particle- coke++
15:22 Tene joined #parrot
15:22 Andy joined #parrot
15:23 particle- .1486 * 365.25
15:23 purl 54.27615
15:24 particle- feb 23
15:38 szabgab joined #parrot
15:39 flh joined #parrot
15:40 nopaste "infinoid" at 75.5.243.250 pasted "minor_packfile_doc.patch" (172 lines) at http://nopaste.snit.ch/16276
15:41 Infinoid msg bacek The Packfile PMCs in the packfile_revamp branch look good.  http://nopaste.snit.ch/16276 is a minor patch, with some pod and comment fixes.
15:41 purl Message for bacek stored.
15:48 rg infinoid: why don't you just commit it to the branch (and point to the commit)?
15:53 Infinoid the code review was the important part and he'll probably wake up before I'm done with $realjob
15:56 dalek parrot: r38116 | fperrad++ | trunk/lib/Parrot/Distribution.pm:
15:56 dalek parrot: [codingstd] handles PASM file as PIR file
15:56 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38116/
16:00 dalek parrot: r38117 | fperrad++ | trunk (8 files):
16:00 dalek parrot: [codingstd] remove trailing space
16:00 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38117/
16:13 dalek parrot: r38118 | fperrad++ | trunk (9 files):
16:13 dalek parrot: [codingstd] add missing $Id
16:13 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38118/
16:16 iblechbot joined #parrot
16:18 Tene_ joined #parrot
16:23 particle- time
16:23 particle- time/
16:23 particle- time?
16:23 purl time is probably 16:21:55 2009 and (did you mean "clock"?) or flowing like a river
16:23 particle- clock?
16:23 purl particle-: LAX: Wed 9:23am PDT / CHI: Wed 11:23am CDT / NYC: Wed 12:23pm EDT / LON: Wed 5:23pm BST / BER: Wed 6:23pm CEST / IND: Wed 9:53pm IST / TOK: Thu 1:23am JST / SYD: Thu 2:23am EST /
16:35 dalek parrot: r38119 | fperrad++ | trunk (43 files):
16:35 dalek parrot: [codingstd] add missing code coda
16:35 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38119/
17:03 iblechbot joined #parrot
17:23 barney joined #parrot
17:41 dalek parrot: r38120 | allison++ | trunk (5 files):
17:41 dalek parrot: [book] Reordering chapters as they will appear in the book.
17:41 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38120/
17:48 dalek parrot: r38121 | allison++ | trunk/docs/book (8 files):
17:48 dalek parrot: [book] Edits to existing chapters.
17:48 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38121/
17:51 dalek parrot: r38122 | allison++ | trunk/docs/book (2 files):
17:51 dalek parrot: [book] Add new chapters for PGE and NQP.
17:51 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38122/
17:54 dalek parrot: r38123 | allison++ | trunk/docs/book (4 files):
17:54 dalek parrot: [book] Adding appendices.
17:54 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38123/
17:54 dalek parrot: r38124 | allison++ | trunk/docs/book (2 files):
17:54 dalek parrot: [book] Splitting out the reference chapter into several.
17:55 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38124/
17:58 dalek parrot: r38125 | allison++ | trunk/docs/book/ch04_pir_subroutines.pod:
17:58 dalek parrot: [book] Merging PIR into one chapter.
17:58 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38125/
18:01 dalek parrot: r38126 | allison++ | trunk/MANIFEST:
18:01 dalek parrot: [book] Adding new chapters, merging and splitting other chapters.
18:01 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38126/
18:04 dalek parrot: r38127 | allison++ | trunk/docs/book/README:
18:04 dalek parrot: [book] Update the chapter list.
18:04 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38127/
18:05 uniejo joined #parrot
18:13 * Coke hopes dukeleto is cool with jerry taking over =-)
18:18 dalek parrot: r38128 | allison++ | trunk/docs/book/ch10_opcode_reference.pod:
18:18 dalek parrot: [book] Split command-line options, directives, and PIR operators into
18:18 dalek parrot: separate chapters.
18:18 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38128/
18:19 davidfetter erm, how do i build the book?
18:21 dalek parrot: r38129 | allison++ | trunk/docs/book/ch10_opcode_reference.pod:
18:21 dalek parrot: [book] Revert an inadvertent downcasing of the the entire chapter.
18:21 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38129/
18:28 dalek parrot: r38130 | allison++ | trunk/docs/book/ch10_opcode_reference.pod:
18:28 dalek parrot: [book] Integrate changes from book.
18:28 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38130/
18:29 slavorgn joined #parrot
18:29 particle- coke: if we didn't show up at the meeting, we lose the student and slot automatically, so all i did was the same thing, but with a smile :)
18:31 allison purl: msg Util Thanks! I've merged in my edits to the book so far, and will continue working on trunk, so you can commit your changes freely. Thanks for the help!
18:31 purl Message for util stored.
18:32 Coke particle++
18:33 davidfetter allison, um, how to build the book?
18:34 allison davidfetter: You need Pod::PseudoPod and Pod::PseudoPod::LaTeX
18:34 davidfetter k
18:34 davidfetter then what?
18:34 allison davidfetter: just Pod::PsuedoPod will get you a nice htmldump, which is plenty for general use
18:34 allison davidfetter: then run ppod2latex
18:34 allison or ppod2html
18:35 allison (both are included in the CPAN modules)
18:38 japhb joined #parrot
18:42 dalek parrot: r38131 | coke++ | trunk (2 files):
18:42 dalek parrot: [cage] fix unused parameter warning and build dependencies for src/nci.c
18:42 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38131/
18:43 dduncan joined #parrot
18:45 dduncan left #parrot
19:03 acajou joined #parrot
19:08 Coke (build book) perhaps a make target?
19:18 allison Coke: a good idea, but it has too many external dependencies
19:19 dalek parrot: r38132 | chromatic++ | branches/headercleanup (355 files):
19:19 dalek parrot: Brought the headercleanup branch up to date with trunk at r37950.
19:19 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38132/
19:19 Coke I'm not saying we have to include the dependencies.
19:24 Coke Infinoid: ping.
19:26 Coke hurm. we have functions that don't exist if we don't have ICU - they should probably exist and throw and exception, yes?
19:26 Coke "throw an"
19:32 * davidfetter eats, shoots, and leaves
19:39 cotto allison, ping
19:41 riffraff joined #parrot
19:42 dalek parrot: r38133 | chromatic++ | branches/headercleanup (6 files):
19:42 dalek parrot: Moved src/interp/interpreter.c to src/runcore/main.c, as most of its functions
19:42 dalek parrot: pertain to runops.  Also moved src/interp/interp_guts.h to
19:42 dalek parrot: src/runcore/runcore_guts.h, though it'll move to a real header file shortly.
19:42 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38133/
19:46 allison cotto: pong
19:47 cotto allison, privmsg
19:50 acajou left #parrot
19:51 allison Coke: yes, dummy functions that throw an exception are sensible
19:56 Infinoid Coke: pong
20:07 Coke Infinoid: if we can enable -Wunused-macros, do we need t/codingstd/c_*assert* ?
20:08 Infinoid ASSERT_ARGS() (and c_arg_assert.t by extension) is a useful seatbelt, even with -Wunused-macros.
20:09 Infinoid Taking a pointer argument (which was null) and passing it to some other function would satisfy -Wunused-macros but not ASSERT_ARGS()
20:10 Infinoid hmm.  -Wunused-macros would help enforce the SHIM stuff, ASSERT_ARGS is meant to enforce the NONNULL stuff
20:10 Infinoid oh, wait.
20:10 * Infinoid had read that as -Wunused-args
20:11 Infinoid how does -Wunused-macros even work?  if a #define exists in a header that was included by a source file, but the source file didn't use the #define, it will issue a warning?
20:12 Infinoid headerizer puts those defines in headers (for public functions) and in the source file (for private functions).  If you were to hack headerizer to put them all in the local function, which sounds very appropriate to me, then we wouldn't need t/codingstd/c_*assert* indeed
20:12 Infinoid s/local function/local file/
20:13 Infinoid (sorry, I'm pretty out of it today)
20:13 dalek rakudo: a4535c1 | pmichaud++ | docs/spectest-progress.csv:
20:13 dalek rakudo: spectest-progress.csv update: 362 files, 10357 passing, 0 failing
20:13 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/a​4535c13b7d89c61e756938c8ee11f545ccc107c
20:13 shorten dalek's url is at http://xrl.us/beo5q8
20:13 rg infinoid: manual says: "Warn about macros defined in the main file that are unused." so it doesn't apply to anything defined in headers.
20:14 Infinoid yeah, that's the concern.
20:14 Infinoid It would be nice if headerizer put all these things in the local source file anyway, rather than cluttering up include/
20:17 Infinoid hmm.  except for functions inside of #if JIT or #ifdef WIN32 or whatever, those wouldn't work very well with -Wunused-macros
20:19 rg and i think for jit it would complain *a lot*
20:20 Infinoid there are a lot of both of those.  I was thinking of src/io/socket_win32.c too
20:20 rg ah no, i'm misremembering that. those emit definitions are just a header file.
20:20 Infinoid there are several cases where the whole file are in #if blocks
20:20 Infinoid though... hmm.  Suppose you could just put the headerizer block inside the #if too
20:22 Infinoid a fix for the old_blocks warning in compilers/imcc/pbc.c could fix this too.
20:24 * Coke came here for an argument.
20:26 Infinoid I'm pretty good at arguing with myself, as you can see
20:29 Coke hurm. PARROT_DOES_NOT_RETURN does not seem to satisfy -Wmissing-noreturn
20:32 Coke (though I could be misreading. some pmc methods that don't return aren't decorated)
20:33 GeJ Good morning everyone
20:40 Coke sixcall?
20:43 uniejo joined #parrot
20:44 Coke there. I think I have updated the sixperl event in the parrot google calendar properly (had to delete and re-add)
20:49 particle- missed the darned call again
20:52 Coke particle-: now you can set a reminder on the CLP calendar.
20:55 pmichaud I missed the call also.
20:56 Coke ah, I think we thought you were travelling to ... nordia.
20:56 pmichaud well, I'm here, yes.
20:56 pmichaud so I have a valid excuse.  :-)
20:57 particle- they have skype there
20:57 pmichaud I figured it'd be okay for me to miss this one call :-)
20:58 pmichaud uUnless I missed something important.  In which case, too bad for me (I'll read the minutes).
20:58 particle- three in a row for me, and i have news of gsoc :(
20:58 pmichaud What's the CLP calendar?
20:58 Coke comp.lang.parrot calendar visible at www.parrot.org
20:58 Coke you can click on the big "+" there to sub to it.
20:59 particle- search for comp.lang.parrot
20:59 pmichaud just added it to my google calendar.
21:00 pmichaud The times for the #parrotsketch meeting seem off in my calendar.  The sixperl conference is correct, though.
21:00 particle- aha! i need to re-add goosync to my phone after a firmware update
21:01 Coke pmichaud: I just fixed sixperl.
21:02 Coke I may need to fix parrotsketch, also.
21:02 Coke (I think the original elements were added in a DST field)
21:02 Coke parrotsketch?
21:02 purl parrotsketch is a status meeting for parrot core committers held every Tuesday at 18:30 UTC in #parrotsketch
21:02 particle- i thought i fixed that, but i may not have
21:02 Coke could be google sucks. yah, it's off by an hour, fixing...
21:04 Coke updated.
21:05 Coke there is also https://www.google.com/calendar/hosted/coleda.​com/embed?src=ngctmrd1cac35061mrjt3hpgng@group​.calendar.google.com&ctz=America/New_York
21:05 shorten Coke's url is at http://xrl.us/beo5xn
21:05 Coke er..
21:05 Coke that's not the original url. =-)
21:05 Coke perl review has a calendar that includes workshops, etc. just found that today.
21:06 Coke that cal name is "The Perl Review: Perl Community Events"
21:32 donaldh joined #parrot
21:45 Limbic_Region joined #parrot
22:04 Whiteknight joined #parrot
22:06 donaldh left #parrot
22:08 szabgab joined #parrot
22:12 rakudohudson joined #parrot
22:20 ruoso joined #parrot
22:27 dalek parrot: r38134 | allison++ | trunk/docs/book/ch10_opcode_reference.pod:
22:27 dalek parrot: [book] Clean up references to the chapter number.
22:27 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38134/
22:29 kid51 joined #parrot
22:34 dalek parrot: r38135 | allison++ | trunk/docs/book/ch09_pasm.pod:
22:34 dalek parrot: [book] Update chapter references in PASM chapter.
22:34 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38135/
23:02 mikehh joined #parrot
23:10 dalek parrot: r38136 | whiteknight++ | trunk/docs/book (8 files):
23:10 dalek parrot: lots of misc fixes for the book, a lot of whitespace cleanups (I know whitespace policies in docs/* aren't as rigorous as they are in src/* but I'm a stickler), etc
23:10 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38136/
23:11 Whiteknight oh, I didn't realize allison++ was working on the book too
23:11 szabgab joined #parrot
23:11 * Whiteknight will find a different hobby for the rest of the evening
23:12 kid51 duelling edits!  May the best red pencil win!
23:14 kid51 allison:  Can you set your editor to wrap at 78 chars per line in .pod files?
23:15 kid51 We're getting failures on the linelength.t test.
23:16 kid51 Example:  docs/book/ch05_pge.pod
23:16 allison kid51: I think it's because some of the files I split out don't have the standard "coda" with editor information in them. will fix
23:16 allison oh, that's because I imported content from the old book
23:20 dalek parrot: r38137 | allison++ | trunk/docs/book/ch01_introduction.pod:
23:20 dalek parrot: [book] Adding index entries for the various roles.
23:20 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38137/
23:21 allison kid51: I also (unfortunately) lost Coke's "TODO" markings on some of the code examples because they couldn't be parsed by an ordinary Pod parser.
23:22 Whiteknight I fixed a few long lines too, I'll go back through later
23:22 kid51 thx
23:22 Whiteknight allison, when are you looking to get this book more "finished"?
23:23 allison Whiteknight: in the few weeks
23:23 Whiteknight I can spend more energy on it now if it's higher-priority
23:23 Whiteknight okay
23:23 allison Printing by June 1st
23:23 Whiteknight oh wow. Sooner then I expected. What publisher?
23:24 allison Onyx Neon Press
23:24 Whiteknight Okay, I sort of assumed it would have been O'Reilly
23:25 Whiteknight not that it matters, of course
23:25 allison O'Reilly loses money if a book sells fewer than 10,000 copies
23:26 Whiteknight oh, wow. I didn't suspect their margins were so small. That makes more sense then
23:27 Whiteknight (probably why my book proposal got rejected)
23:27 allison what was your proposal?
23:27 Whiteknight It was a book about open-education technologies, like wikis
23:28 Util allison: Were the Perl 6 chapters of the book released to us by O'Reilly, or just the Parrot chapters?
23:28 allison Util: the entire book was released by O'Reilly
23:28 Whiteknight I was working with researchers from ODU and Manchester U on the topic, and proposed a book about it
23:28 allison the perl chapters live at svn.perl.org/perl6
23:28 Tene_ allison: progress on the HLL PMC issues?
23:28 allison Util: it could also be updated into a separate book
23:29 allison Tene: I got your message about the hll_info branch, is that what you mean?
23:30 Tene_ No.  I can't move rakudo to .HLL 'perl6' because its pmcs break.
23:30 allison Whiteknight: sounds like an interesting book, but yes, not enough sales for O'Reilly to publish
23:30 Tene_ they go in the 'parrot' HLL namespace or something.
23:30 Tene_ there's a ticket, I think...
23:30 allison Tene: ah, because there's no way to tell C PMCs to associate with a particular HLL
23:31 allison Tene: it's not something I've looked at lately
23:31 kid51 Those docs files will also need to be fixed for svn:keywords and svn:eol-style native tests in t/distro/file_metadata.t.
23:31 allison kid51: I'll do that now
23:32 Tene_ allison: it's a blocker in getting rakudo into a .HLL, which is necessary if any other language wants to use Perl 6 code.
23:33 Whiteknight Tene_: What are the requirements? How do we fix it?
23:33 Tene_ I haven't had the focus to figure out how to do it myself for a few months now... haven't been sleeping.
23:35 Tene_ Whiteknight: C PMCs need to be accessible from code in an HLL.  I think this is namespace related.  They always go in the 'parrot' hll ns, and then nothing can find them.
23:35 Tene_ i.e. perl6scalar, etc.
23:36 Whiteknight doesn't the "group" specification do that?
23:36 Whiteknight or, shouldn't it?
23:36 Tene_ there's also an HLL specification
23:36 Tene_ but, dunno
23:36 allison Whiteknight: the 'group' is for compiling the C shared lib
23:37 dalek parrot: r38138 | allison++ | trunk/docs/book (8 files):
23:37 dalek parrot: [book] Setting svn properties on new book files.
23:37 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38138/
23:37 allison The problem is that C PMCs don't actually pay any attention to namespaces
23:38 Whiteknight Okay, I guess I just don't understand why they would need to follow namespaces at all
23:38 Whiteknight I wasn't aware that namespaces were involved at all
23:39 * Whiteknight apparently needs to browse a little bit more code
23:39 Tene_ lemmeook
23:56 Coke_afk (lost pod markers on code) that sucks, as now your editor will have to manually insure that all the code snippets work.
23:57 allison Coke: It kept the POD markers, it just lost the TODO additions
23:58 allison so, =begin PIR and =end PIR are still there
23:58 allison Coke: but it couldn't figure out what to do with "=begin PIR TODO some text"
23:59 allison Coke: maybe we can mark TODOs another way?

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

Parrot | source cross referenced