|
Parrot 3.6.0 "P�jaros del Caribe" released | parrot.org | Log: irclog.perlgeek.de/parrot/today | #parrotsketch meeting Tuesday 1930 UTC Set by moderator on 19 July 2011. |
|||
| whiteknight | I can't seem to instal jekyll here, which is annoying | 00:06 | |
| dalek | href="https://parrot.github.com:">parrot.github.com: 0704f29 | Whiteknight++ | / (2 files): quick clearfix hack |
00:07 | |
|
00:11
darbelo_ left
|
|||
| dalek | href="https://parrot.github.com:">parrot.github.com: 0bc8a17 | Whiteknight++ | / (2 files): aesthetics on the sidebar |
00:15 | |
| href="https://parrot.github.com:">parrot.github.com: 9113f35 | Whiteknight++ | / (2 files): fix sidebar list |
|||
| cotto_work | aloha: clock? | 00:16 | |
| aloha | cotto_work: LAX: Wed, 17:16 PDT / CHI: Wed, 19:16 CDT / NYC: Wed, 20:16 EDT / UTC: Thu, 00:16 UTC / LON: Thu, 01:16 BST / BER: Thu, 02:16 CEST / TOK: Thu, 09:16 JST / SYD: Thu, 10:16 EST | ||
|
00:16
rurban_ joined
00:19
rurban left,
rurban_ is now known as rurban
|
|||
| dalek | umage/gh-pages: 8d70f49 | Whiteknight++ | index.html: github generated gh-pages branch |
00:20 | |
| whiteknight | okay, the auto-generated plumage website is officially the ugliest thing ever | 00:23 | |
| ever | |||
| dukeleto | parrot.github.com/plumage/ eek! | 00:25 | |
| whiteknight | check out the main page | 00:26 | |
| I done prettied it up | |||
| Coke | ow. | 00:28 | |
| that color is a hate crime on my eyes. | |||
| (the support email is at patrick's domain??) | |||
| cotto_work | Um, no. It's ugly, but you clearly haven't seen Yvette's Bridal' | ||
| whiteknight | Coke: that's the contact email for the parrot org on github | 00:33 | |
| I think patrick set it up originally | |||
| dalek | umage/gh-pages: 892e626 | Whiteknight++ | index.html: Remove email address |
||
|
00:34
frodwith joined
00:42
kid51 left
00:45
kid51 joined
|
|||
| whiteknight | anybody else can feel free to jump in and embetter the site too | 01:01 | |
| soh_cah_toa | what site? | 01:02 | |
| whiteknight | parrot.github.com/plumage | ||
| look deeply into it, until your eyes melt | 01:03 | ||
| soh_cah_toa | agh! | ||
| Felipe | ouch | ||
| the command part needs some <br/> | 01:06 | ||
| dalek | umage/gh-pages: 82fe14b | soh_cah_toa++ | index.html: Wrapped "Install" section in <pre> tags so it's rendered as code. |
01:10 | |
| Felipe | yeah, <pre> is fine too | ||
| dalek | href="https://parrot.github.com:">parrot.github.com: 1b6830f | Whiteknight++ | pages/ (2 files): +add quick stub pages for projects and compilers |
01:12 | |
| href="https://parrot.github.com:">parrot.github.com: ab7ef33 | Whiteknight++ | stylesheets/parrot.css: fix heading sizes |
01:14 | ||
| href="https://parrot.github.com:">parrot.github.com: 8a795da | Whiteknight++ | _layouts/site.html: Add links to projects and compilers pages |
|||
| umage/gh-pages: bac2179 | soh_cah_toa++ | index.html: Added a link to the Plumage wiki. |
01:16 | ||
| soh_cah_toa | agh, messed it up | ||
| whiteknight | :) | 01:17 | |
| don't worry about it, we're all experimenting | |||
| just play | |||
| soh_cah_toa | i got it | ||
| dalek | umage/gh-pages: 6fbb847 | soh_cah_toa++ | index.html: Fixed <pre> tag from previous commit so that paragraph isn't rendered as code. |
01:18 | |
| soh_cah_toa | i'm thinking about giving our wiki a new look too | 01:19 | |
| right now, it's just one long list | |||
| cotto_work | who's skids? | 01:23 | |
| dalek | href="https://parrot.github.com:">parrot.github.com: 9f36db0 | Whiteknight++ | stylesheets/parrot.css: indent headings to set them apart |
||
| href="https://parrot.github.com:">parrot.github.com: eab3027 | Whiteknight++ | pages/compilers.md: +descriptions of various compilers |
|||
| href="https://parrot.github.com:">parrot.github.com: 0108408 | Whiteknight++ | pages/projects.md: +descriptions and attribution for projects) |
|||
| whiteknight | skids? | ||
| cotto_work | He left a nice comment on something I posted on my blog about implementing M0 in Perl. | ||
| dalek | href="https://parrot.github.com:">parrot.github.com: ee8d68b | Whiteknight++ | _layouts/site.html: fix main header |
01:25 | |
| cotto_work | I'm discovering that a substantial part of what I use gists for is better done by blogging. | ||
| whiteknight | I use gists a lot to jot down ideas for blog posts, then I forget to publish them | 01:26 | |
| soh_cah_toa | hmm...something is wrong. slavorg and slavorgn are not ops anymore which explains why i'm not op status anymore. whiteknight and some others too | 01:28 | |
| maybe it happened b/c of that netsplit recently | |||
| cotto_work | opbots, names | ||
| whiteknight | oh, failz | ||
| cotto_work | ophumans, names | ||
| whiteknight | I'm not ops either | ||
| cotto is, but cotto is at work | 01:29 | ||
| oh, cotto_work is too | |||
| cotto_work | all better | ||
| soh_cah_toa | all better | ||
| ha! | |||
| cotto_work | My client says I won. | ||
| Felipe | :p | ||
|
01:30
kid51 left
|
|||
| atrodo | whiteknight-- # For making such ugly that demands that I do something about it | 01:32 | |
| whiteknight: first, we need to change that color | 01:38 | ||
| Then we need to add unicorns | |||
| whiteknight | atrodo: patches welcome! | ||
| atrodo | Unless parrot wants to adopt another mythical creature, but that seems silly | 01:39 | |
|
01:40
woosley joined
|
|||
| whiteknight | yeah, parrots are mythical enough | 01:42 | |
|
01:46
theory joined
01:51
rdesfo joined
|
|||
| dalek | rrot: 748ed5e | soh_cah_toa++ | t/dynpmc/os.t: Fixed output regular expression for chroot test in t/dynpmc/os.t so that it passes on OpenBSD. |
01:53 | |
|
01:59
whiteknight left
02:00
rdesfo left
|
|||
| soh_cah_toa | dukeleto: ping | 02:42 | |
|
03:17
jnthn__ left
03:29
jnthn_ joined
03:34
theory left
|
|||
| Util | Working on TT#1891 (Problems with acosh, asech, atanh and acoth in Complex PMC) - FYI | 03:42 | |
| Coke wonders if parrot really needs a complex pmc. | 03:49 | ||
| cotto | ~~ | 04:04 | |
| Util | FWIW, working on Parrot's Complex PMC is turning up bugs in Perl 5's Math::Complex and Math::Cephes::Complex | 04:15 | |
| cotto | ... | ||
| <3 | 04:16 | ||
| util++ | |||
|
04:16
daniel-s joined
|
|||
| Coke | well fine, then. ;) | 04:17 | |
| dalek | tracwiki: v4 | soh_cah_toa++ | GettingStartedWithPlumage | 04:30 | |
| tracwiki: Fixed a few grammar mistakes in the opening paragraph. | |||
| tracwiki: trac.parrot.org/parrot/wiki/Getting...ction=diff | |||
| umage/gh-pages: d6b53a8 | soh_cah_toa++ | index.html: Rewrote first paragraph so that it matches the description in the wiki. |
04:34 | ||
| umage/gh-pages: 52669c0 | soh_cah_toa++ | index.html: Added instructions on how to clone the repo. Also changed the commands for installing to 'parrot' instead of 'parrot-nqp'. |
04:42 | ||
| umage/gh-pages: f303a9b | soh_cah_toa++ | index.html: Changed 'setup.nqp' to 'setup.pir' in "Install" section. |
04:44 | ||
| rrot/soh-cah-toa/hbdb: d98b9ec | soh_cah_toa++ | src/runcore/cores.c: Vertically aligned assignments in Parrot_runcore_hbdb_init(). |
04:45 | ||
| rrot/soh-cah-toa/hbdb: a2b156b | soh_cah_toa++ | src/runcore/cores.c: Made the perldoc for runops_hbdb_core() a little more verbose. |
|||
| rrot/soh-cah-toa/hbdb: 729682a | soh_cah_toa++ | lib/Parrot/Test/HBDB.pm: Changed Parrot::Test::HBDB to be a subclass of Test::Builder::Module so that it really acts as a testing library. |
|||
| rrot/soh-cah-toa/hbdb: ed0a12f | soh_cah_toa++ | lib/Parrot/Test/HBDB.pm: Heavily commented lib/Parrot/Test/HBDB.pm for the sake of future readers. |
|||
| soh_cah_toa | msg dukeleto i was able to fix the t/dynpmc/os.t test (commit 748ed5e) but how do you think i should fix the two in t/src/extend_vtable.t? | 04:47 | |
| aloha | OK. I'll deliver the message. | ||
| soh_cah_toa | msg dukeleto if Parrot_PMC_get_pointer() returns null only on openbsd, should i just do something like: unless $openbsd { normal test } else { same thing except expect null } ? | ||
| aloha | OK. I'll deliver the message. | ||
|
04:53
daniel-s left
|
|||
| soh_cah_toa | msg whiteknight is there any reason the source files for plumage are in the gh-pages branch? this branch doesn't really seem like the place for them unless they serve some hidden purpose i can't think of | 04:59 | |
| aloha | OK. I'll deliver the message. | 05:00 | |
| Util sleeps | 05:03 | ||
| dalek | umage/gh-pages: cfb2bd4 | soh_cah_toa++ | .gitignore: Added .gitignore file. |
05:05 | |
| cotto | oh noes! commits! | ||
| soh_cah_toa | yup | 05:06 | |
| dukeleto | soh_cah_toa: no, there is a bug in get_pointer on openbsd. Not running the test on openbsd doesn't solve the problem, just sweeps it under the rug | 05:07 | |
| soh_cah_toa | dukeleto: i wasn't proposing to not run the test, just run the same thing as non-bsd except check for null instead of > 0 | 05:08 | |
| but if it's a bug in get_pointer then i guess that's not the right route | 05:09 | ||
|
05:11
daniel-s joined
|
|||
| dukeleto | soh_cah_toa: perhaps the test is wrong. I am tweaking it | 05:13 | |
| cotto | That carpet's going to get pretty lumpy. | ||
|
05:18
daniel-s left,
woosley left
|
|||
| dalek | umage/gh-pages: a206081 | soh_cah_toa++ | / (4 files): Moved CSS code to an external file and linked it from index.html. Also started using Jekyll templates. |
05:29 | |
| soh_cah_toa | whoops | 05:30 | |
| dalek | umage/gh-pages: 0d8e247 | soh_cah_toa++ | / (2 files): Let's see if this fixes the stylesheet error in the last commit. |
05:35 | |
| umage/gh-pages: 9f35e9c | soh_cah_toa++ | / (3 files): Moved CSS code back into index.html because GitHub is stupid. |
05:41 | ||
| cotto | soh_cah_toa, github.com/parrot/parrot/commit/ed...ent-488495 | 05:44 | |
| soh_cah_toa | cotto: github.com/parrot/parrot/commit/ed...ent-488498 | 05:45 | |
| cotto | touche | ||
| If someone removes all the code, you're in trouble. ;) | 05:46 | ||
| soh_cah_toa | yeah but i mean if they don't understand what a function does or something (pretty much all of parrot's c code). those i suppose it's easy to figure out what _close_fh() does | 05:47 | |
| *though | |||
| cotto | I assume that anyone who reads the comments can read the code that goes along with them, barring unusual code. | 05:48 | |
| soh_cah_toa | not necessarily, much of my frustration is b/c of thousands of uncommented function/macros in parrot code. i admit, i took my own guidelines way overboard | 05:49 | |
| cotto | Don't react to one extreme by moving to the other. | 05:50 | |
| I agree it's lta. | |||
| soh_cah_toa | actually, looking through this, there's more comments than actual code. not cool | 05:51 | |
| cotto smells a learning experience | 05:52 | ||
| soh_cah_toa | no, that's my litter boxes that need to be cleaned | 05:53 | |
| cotto | better take care of that then | ||
| dalek | rrot/soh-cah-toa/hbdb: c82e793 | soh_cah_toa++ | lib/Parrot/Test/HBDB.pm: Removed comments that were practically hiding the code. |
05:55 | |
| rrot/soh-cah-toa/hbdb: 0b1cea1 | soh_cah_toa++ | lib/Parrot/Test/HBDB.pm: Fixed typo in lib/Parrot/Test/HBDB.pm. |
05:56 | ||
| rrot: 3078cc6 | dukeleto++ | t/src/extend_vtable.t: [t][src] Refactor get_pointer test to have a more useful failure message |
05:58 | ||
|
06:02
simcop2387 left
06:17
spinclad left,
spinclad joined
|
|||
| dalek | umage/gh-pages: 32c1d09 | soh_cah_toa++ | index.html: Trying a different color scheme. |
06:20 | |
|
06:24
preflex_ joined
|
|||
| dalek | umage/gh-pages: d2952a0 | soh_cah_toa++ | index.html: Added link to Parrot repo in "Dependencies" section. |
06:25 | |
|
06:26
woosley joined
|
|||
| soh_cah_toa | enough of that... | 06:26 | |
|
06:26
soh_cah_toa left
06:28
preflex left,
contingencyplan joined
06:32
simcop2387 joined
|
|||
| dalek | kudo/nom: ddf95b3 | moritz++ | src/Perl6/Actions.pm: move detection of indirect name lookups into a method, and check for it in sigiled variables |
06:55 | |
|
07:12
woosley1 joined
07:15
woosley1 left
07:16
woosley left,
woosley joined
07:23
woosley left
07:30
mj41 joined
07:44
fperrad joined
|
|||
| cotto | dukeleto, you around? | 08:00 | |
|
08:17
rurban_ joined
08:18
rohit_nsit08 joined
08:19
rohit_nsit08 left,
rurban left,
rurban_ is now known as rurban
08:34
Kulag left
08:37
Kulag joined
08:38
woosley joined
|
|||
| dalek | rrot/m0-prototype: 77f1ba6 | cotto++ | src/m0/perl5/m0_assembler.pl: better debugging messages, bugfix in constant serialization |
08:42 | |
| rrot/m0-prototype: da4e0cb | cotto++ | src/m0/perl5/m0_ (2 files): start to implement full-on C int semantics in M0 I tried to limit the hackiness and make it readable, but the concept of implementing low-level semantics in Perl implies a certain amount of hackery. A basic test to add two ints and print the result works, but anything more complex fails. |
|||
| rrot/m0-prototype: 588ca65 | cotto++ | src/m0/perl5/m0_ (2 files): add a note about how to fix the assembler and some temporary debugging code |
|||
|
08:51
contingencyplan left
09:07
dod1 left
09:10
dod joined
09:12
dod left
|
|||
| NotFound | zampor& | 09:16 | |
| Ups | |||
|
09:17
dod joined
10:01
woosley left
10:56
jsut_ joined
10:57
jsut left
11:04
he joined
11:29
preflex joined
11:30
preflex_ left
11:51
JimmyZ joined
11:55
kid51 joined
|
|||
| kid51 | msg whiteknight parrot-dev mail is slow; have filed ticket; please see lists.parrot.org/pipermail/parrot-d...06071.html | 12:03 | |
| aloha | OK. I'll deliver the message. | ||
| Felipe | morning devs. | 12:08 | |
|
12:10
jsut joined
12:11
preflex_ joined
12:13
preflex left
12:15
jsut_ left
12:24
kid51 left
12:45
whiteknight joined
|
|||
| whiteknight | good morning, #parrot | 12:48 | |
| msg soh_cah_toa the source files shouldn't be there. I won't remove anything that isn't tracked. | 12:49 | ||
| aloha | OK. I'll deliver the message. | ||
|
13:04
smash joined
|
|||
| smash | hello everyone | 13:04 | |
|
13:06
preflex joined
|
|||
| whiteknight | good morning, smash. How are you doing today? | 13:07 | |
| smash | whiteknight: fine thank you, and you? | ||
| whiteknight | pretty well, thanks. Been hacking on anything cool? | 13:08 | |
| smash | whiteknight: nothing fancy or interesting, currently hacking on my pre-thesis | ||
|
13:09
preflex_ left
|
|||
| whiteknight | pre-thesis? What's your topic? | 13:09 | |
| smash | whiteknight: the title is An Ontology Toolkit for Problem Domain Concept Location in Program | 13:10 | |
| Comprehension | |||
| Felipe | morning | ||
| moritz | no offense meant, but that does sound like bs bingo to me, without further explanation :-) | 13:11 | |
| smash | moritz: non taken | 13:12 | |
| it's just a title | |||
| moritz | I guess most thesis titles are rather meaningless to somebody outside the specific field | 13:13 | |
| smash | i agree | 13:14 | |
| moritz has/had "fun" explaining two thesis titles to friends and family | 13:15 | ||
| dalek | kudo: 495578b | moritz++ | build/PARROT_REVISION: bump parrot version to 3.6 |
13:20 | |
|
13:21
AzureSto_ joined
13:22
AzureStone left
|
|||
| dalek | kudo: 1ebab75 | moritz++ | docs/ChangeLog: changelog |
13:24 | |
| whiteknight | smash: Can you explain that at all? | 13:29 | |
| it sounds decidedly abstract | |||
| smash | whiteknight: its' abou tways to enhance current program comprehension techniques by linking the problem domain with the source code and produced world effects using ontologies | 13:34 | |
| Coke | smash:O_o; | 13:37 | |
| ... that sounds like fancy talk for "use 4gls" | 13:38 | ||
| or DSLs, as I've also heard them called. | |||
| smash nods. | 13:42 | ||
| Coke | \\o/ | 13:45 | |
| ISTR chromatic has written some screed on that on his modern perl blog; might be worth adding to the pile of reading material. | 13:46 | ||
| dalek | kudo: 53a9da1 | moritz++ | docs/announce/2011.07: initial release announcement |
13:52 | |
| smash | Coke: yeap, i recall something | ||
|
13:55
ambs joined
13:56
plobsing joined
14:01
benabik joined
14:03
benabik left
14:06
benabik joined
14:12
tewk_ joined
14:24
he left
14:28
daniel-s joined
14:30
nbezzala joined
|
|||
| dalek | kudo: a73d73c | moritz++ | docs/announce/2011.07: name our release Beijing |
14:40 | |
|
14:50
darbelo joined
|
|||
| benabik | So I now have a duel-booting Lion/Win7 laptop… but I forgot that Fink is slow to update for new OSes and now I have to find another way to get dependencies for git/parrot/etc | 14:57 | |
| dalek | kudo: cc373ff | moritz++ | / (2 files): update release_guide.pod, bump VERSION |
15:06 | |
| kudo: 085a4b7 | moritz++ | build/Makefile.in: make spectest_update nonfatal, it fails in the release (because t/spec/ is not a git checkout) |
|||
| nxed: b6437a3 | NotFound++ | winxedst1.winxed: refactor builtin creation avoiding intermediate storages |
15:07 | ||
|
15:15
preflex_ joined
15:16
lucian joined
15:17
preflex left
15:36
Kulag left
15:39
Kulag joined
15:47
daniel-s left
|
|||
| Coke | if I'm using git svn, and I have local branches tracking remote branches, | 15:47 | |
| and I want to do some local branch merging, do I need to do anything special, | |||
| or can I just use "git merge" locally happily and do a 'git svn dcommit' once | |||
| I'm satisfied? | |||
| JimmyZ | well, I don't like git svn too much, because I can't commit to svn, and my git commits always on head | 15:55 | |
| benabik | Coke: I don't remember how git-svn handles committing merges. It used to be that it didn't go too well, but that may have gotten better. | ||
| dalek | kudo: 61eef48 | pmichaud++ | docs/announce/2011.07: Revise release announcement a bit. |
16:11 | |
| Coke | I did a "git commit" on master, git checkout of branch, git merge --no-ff master, git svn dcommit, and it worked pretty much like I wanted. (like having individual commits from trunk on the branch, rather than a mess of "merge latest changes from trunk" commit. | 16:15 | |
| benabik | git-svn has been getting smarter. | 16:16 | |
| Coke | much less painful than last I used it. probably 1/3 git-svn, 2/3 me using git for a year. | ||
|
16:16
rurban_ joined
|
|||
| benabik | Although I kinda wish it would be integrated into the new vcs-helper framework… But that would mean updating it to use fast-import, which is non-trivial. | 16:16 | |
| Huh. '...' is becoming '…' automatically now | 16:17 | ||
|
16:17
JimmyZ left
|
|||
| whiteknight | where? | 16:17 | |
| benabik | I think it's OS level in Lion. | ||
| Coke | hurm. I wonder if I wanted --ff or --no-ff there. | ||
| benabik | --no-ff mean "I always want a merge commit" | 16:18 | |
| Coke | ah, so I really wanted --ff. whoops. | 16:19 | |
|
16:19
rurban left,
rurban_ is now known as rurban
|
|||
| dalek | kudo: 9225003 | pmichaud++ | docs/release_guide.pod: Add some release name suggestions for future releases. |
16:22 | |
| cotto | ~~ | 16:30 | |
| whiteknight | hello cotto | 16:32 | |
| cotto | It is a good day for pie. | 16:39 | |
| Coke | Perhaps today IS a good day to PIE! <worf/> | 16:49 | |
|
16:59
theory joined
|
|||
| benabik | errands & | 17:01 | |
|
17:01
benabik left
17:02
mj41 left
17:08
nbezzala left
|
|||
| cotto_work | ~~ | 17:16 | |
|
17:32
mj41 joined
17:35
particle joined
17:37
dmalcolm joined
17:43
RobertLJ joined
|
|||
| RobertLJ | questin: where do I submit parrot build problems ? have been attempting to build parrot 3.6.0 on Windows 7 using straberry perl. | 17:46 | |
| whiteknight | RobertLJ: Here is good. You can send an email to parrot-users or parrot-dev mailinglists, or you can open a ticket on trac.parrot.org | 17:47 | |
| RobertLJ: are you on 64-bit or 32-bit windows? | |||
| RobertLJ | 64 bit .. and i am using the 32 bit strawberry perl installation | 17:51 | |
| 64 bit windows 7 that is | |||
| whiteknight | okay | 17:52 | |
|
18:01
contingencyplan joined
18:07
ambs left
18:19
darbelo_ joined,
darbelo left
|
|||
| RobertLJ | thanks whiteknight .. i think i may have some more info about the problem .. I will do some more testing before I send an email to the parrot-user list | 18:25 | |
| whiteknight | RobertLJ: okay, awesome. | 18:28 | |
| I've got windows here, but I've got MSVC, not strawberry | |||
| cotto_work | strawberry and msvc aren't mutually exclusive | 18:30 | |
| whiteknight | cotto_work: I already have one, and am not installing another | ||
| cotto_work | your call | 18:31 | |
| whiteknight | I already have activeperl on here. Installing strawberry on the same machine has never been a pleasant experience | ||
| maybe it's gotten better, I don't know | 18:32 | ||
| cotto_work | gotcha | ||
|
18:39
benabik joined,
silug left
18:42
benabik left
|
|||
| RobertLJ | what dev environment is used to build parrot windows ? | 18:44 | |
| cotto_work | RobertLJ: I use strawberry/msvc and strawberry/mingw | ||
| RobertLJ | ok .. just wanted to make sure I was compiling parrot with the right dev envirnment | 19:00 | |
| dukeleto | we have lots of broken links and wrong statements on our wikipedia page | 19:02 | |
| dalek | website: lucian++ | Delays ... | 19:08 | |
| website: www.parrot.org/content/delays-... | |||
| dukeleto | lucian: can you make that image a tiny bit smaller ? it is 3,375px × 2,239px, which is unreasonable | 19:10 | |
| lucian: i do like the picture, though :) | 19:12 | ||
|
19:14
plobsing left
|
|||
| lucian | dukeleto: oh. i just tried to make it 800k | 19:17 | |
| whiteknight | lucian++ It's good to see you blogging again! | ||
| lucian | doesn't it get sized to 700-ish pixels on parrot.org? it does on wordpress | 19:18 | |
| dukeleto | lucian: our drupal is dumb | ||
| whiteknight | lucian: also, I'd love to hear your feedback for Rosella.Test if you have time. I don't want it to be giving anybody problems | ||
| dukeleto | lucian: there are some tough critiques in your blog post, but I think we need to hear them | 19:19 | |
| lucian | whiteknight: it's ok, distutils is the problem | ||
| whiteknight | I've wanted to rewrite distutils for a long time now | ||
|
19:19
alester joined
|
|||
| lucian | dukeleto: it's easy to remember what's annoying | 19:20 | |
| dukeleto | lucian: for a non-trivial setup.winxed, bubaflub++'s parrot-gmp might help you: github.com/bubaflub/parrot-gmp/blo...tup.winxed | 19:21 | |
| lucian | dukeleto: yeah, i've used his for inspiration already :) | ||
| whiteknight | There is supposed to be some kind of way to specify what test harness program to use, and what arguments to pass it, but I never quite figured that out | 19:26 | |
| the Rosella and PLA setup scripts detect when the first argument is "test" and executes the harness manually | |||
| the downside is that when we test, it doesn't build first. I need to work on that | 19:27 | ||
| the problem with rewriting distutils is that I will probably rewrite it as part of Rosella, but Rosella is the project I most want the improvements for | 19:28 | ||
| bootstrapping is a bitch | 19:29 | ||
| dukeleto | what are the actual biggest problems with our distutils, currently? | ||
| whiteknight | I don't want to say "problems". It's well-written and very useful | ||
| it's all written in PIR, all in one big file, and it's not nearly flexible enough | 19:30 | ||
| plus it's missing things like building NCI thunk libs, or easily compiling things from other languages | |||
| Rosella provides a wrapper library for distutils that adds winxed support, but having to do that for every language we want to build could become expensive | 19:31 | ||
| it also doesn't do anything to detect prerequisites. Integration with plumage would be awesome in that regard | |||
| In fact, a merger of the two, in a hypothetical rewrite that I was already told not to do, would be awesome | 19:32 | ||
| lucian: I am glad to see you are on the 6model bandwagon now. I'm really looking forward to getting it moved into core and helping update puffin to use it | 19:33 | ||
| won't be done before the end of the summer probably, unless it all goes *much* easier than I expect | 19:34 | ||
| lucian | dukeleto: i think it just doesn't do enough | ||
| whiteknight: well, after john put some docs up and explained what was going on, it made sense | |||
| whiteknight | I think what I want to do with rosella is put together a library for building and traversing a DAG. Then we can use that to build and traverse dependency graphs | 19:35 | |
| If we have a library that does that, all we need is an interface to add data to the DAG, a loop, and a collection of builder objects to take each node and do something with it | 19:36 | ||
| plumage is the same. Put together dependencies in a DAG, then traverse it. Any Make clone would be the same | 19:38 | ||
| Right now, distutils does everything in waves. It does all the pbc_pir compilations first, then the pbc_pbc merges, etc | 19:40 | ||
| so if you have to add new steps to it, you need to know the ordering of the stage as a whole | 19:41 | ||
| could probably use that as the basis for an expert system library too, but I don't feel like getting into a Rete algorithm nonsens | 19:45 | ||
|
19:48
RobertLJ left
|
|||
| cotto_work | A good first step for distutils would be to break it into multiple files. That'd hardly even be a rewrite. ;) | 19:54 | |
| NotFound | whiteknight: a minor problem in distutils: no way no specify a path for an include file different than the installed destination. | 19:55 | |
|
19:56
plobsing joined
|
|||
| whiteknight | I don't want to change distutils too much as it is, because a lot of projects rely on it | 19:58 | |
| and it does work. No questions there. It works very well for the projects that are using it | |||
| Getting something that works well for other projects as well is the big trick | 19:59 | ||
| NotFound | The problem is its lack of tests, is har to add something without risk of breaking something. | 20:00 | |
|
20:00
benabik joined
|
|||
| whiteknight | ah, that's a good point. Either increasing test coverage, or increasing base testability would be a good step | 20:02 | |
| NotFound | But for testing we may need a way to specify an alternate install path. | 20:04 | |
| Now that I think about it... we can use it to install some of the core libraries | 20:05 | ||
| cotto_work | lack of tests is one of its biggest problems | 20:15 | |
| that and the fact that it's one gigantic PIR file | |||
| NotFound | The simpler way to reduce its size can be rewriting it in winxed. | 20:18 | |
| whiteknight | yeah, I wouldn't write anything in PIR now | ||
| nobody should | |||
| and if we're going to do it in winxed, using namespaces and proper OO guts is a no-brainer | |||
| NotFound | Ms | 20:19 | |
| whiteknight | ? | ||
| NotFound | Most of its length are long string concatnations and array initializations that in winxed will become one line. | 20:20 | |
| whiteknight | yeah | ||
| NotFound | PIR is an assembler, after all, is not appropiate for that things. | 20:21 | |
| whiteknight | s/that things/any thing/ | ||
| s/an assembler/a piece of crap/ | 20:22 | ||
| s/after all/kill it now/ | |||
| cotto_work | s/PIR/PIR/ | 20:24 | |
| whiteknight | aloha: PIR is a piece of crap, kill it now, is not appropriate for anything | 20:25 | |
| aloha | whiteknight: ... but PIR is just pasm with a lot of sugar ... | ||
| whiteknight | damnit | 20:26 | |
| aloha PIR is also a piece of crap, kill it now, is not appropriate for anything | |||
| aloha | whiteknight: Okay. | ||
| whiteknight | botsnack | ||
| alohaP: botsnack | |||
| >:( | |||
| aloha: botsnack | |||
| damnit | |||
| I hate bots | |||
|
20:29
whiteknight left
20:36
silug joined
20:44
plobsing left
20:49
soh_cah_toa joined
|
|||
| soh_cah_toa | ~~ | 20:52 | |
| tadzik | ~~ | ||
| Felipe | ~~ | 21:02 | |
|
21:08
mj41 left
21:10
theory left
21:19
theory joined
21:28
fperrad left
21:32
benabik left
|
|||
| dalek | umage/gh-pages: a16fd14 | soh_cah_toa++ | index.html: Moved "Download" section to be right before the "Install" section. This makes more sense since you download Plumage before you install it. |
21:40 | |
| umage/gh-pages: a6178eb | soh_cah_toa++ | / (2 files): Modified layout to make use of Jekyll templates. Also changed footer to use new HTML 5 <footer> tag instead of a <div>. |
|||
| umage/gh-pages: 00eaf0f | soh_cah_toa++ | / (2 files): Made source code a little more clean and organized. |
|||
| umage/gh-pages: 2fac907 | soh_cah_toa++ | index.html: Removed "Contact" section because contacting "Parrot Virtual Machine" makes no sense and it's also redundant to list it along with the "Authors" section. |
|||
|
21:43
Psyche^ joined,
Patterner left,
Psyche^ is now known as Patterner
21:48
whiteknight joined
|
|||
| dalek | umage/gh-pages: 7437336 | soh_cah_toa++ | / (2 files): Refactored some more code into the "default" layout to reduce duplication in future pages. |
21:53 | |
| umage/gh-pages: 44504ff | soh_cah_toa++ | index.html: Made "Authors" section into an undordered list so things are a little nicer looking. |
|||
|
21:54
darbelo_ left
|
|||
| whiteknight | heh, that page is starting to look half-nice | 21:56 | |
| soh_cah_toa++ | |||
| soh_cah_toa | yeah, it's not too bad :) | ||
| Felipe | good | ||
| dalek | umage/gh-pages: 81af96c | soh_cah_toa++ | index.html: Added a "Usage" section explaining how to get started using Plumage. |
22:17 | |
| cotto_work | kid51++ got himself quoted on LWN's weekly edition: lwn.net/Articles/452091/ | 22:38 | |
| Felipe | haha nice | 22:39 | |
| cotto_work | right below jwz and Rasmus, though I wouldn't read too much into that | ||
| msg kid51 You have some takers: lwn.net/Articles/452177 | 22:40 | ||
| aloha | OK. I'll deliver the message. | ||
| rblackwe | I say that too. Go kid51 | 22:41 | |
| I should take him something nice to FOSSCON. | |||
|
22:42
dmalcolm left
22:52
alester left,
Limbic_Region joined
|
|||
| dalek | umage/gh-pages: 0e9f513 | soh_cah_toa++ | .gitignore: Removed entries for source files in .gitignore since they're no longer in the repo. |
22:59 | |
| website: tcurtis++ | GSoC: LALR Parsing: Midterms over! | 23:05 | ||
| website: www.parrot.org/content/gsoc-lalr-pa...terms-over | |||
| rrot/Util/tt_1891_complex_trig: 6e2c0cf | util++ | src/pmc/complex.pmc: [TT#1891] Fix calculation of acosh in complex.pmc |
23:43 | ||
| rrot/Util/tt_1891_complex_trig: 4900212 | util++ | t/pmc/complex.t: [TT#1891] Restore TODOed tests that now pass |
|||
| rrot/Util/tt_1891_complex_trig: 4e505d1 | util++ | t/pmc/complex.t: [TT#1891] Correct a bad test value |
|||
| whiteknight | util++ | 23:48 | |
|
23:53
benabik joined
23:56
whiteknight left
|
|||