Team Chat Logs

2006 7
Mo Tu We Th Fr Sa Su
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31      
<< <   > >>

August 21, 2006

[00:04:20] * jrydberg has joined #trac
[00:07:44] * hpnadig has joined #trac
[00:16:40] * Jippi has joined #trac
[00:17:39] * ranq has quit IRC
[00:33:16] * flox|away has joined #trac
[00:46:47] * evil_twin has joined #trac
[00:47:07] * Semhirage has joined #trac
[00:47:32] * groogs has joined #trac
[00:47:35] * flox|away has quit IRC
[00:47:35] * hpnadig has quit IRC
[00:47:35] * ged has quit IRC
[00:47:35] * matt_good has quit IRC
[00:47:36] * bartc has quit IRC
[00:47:36] * De_Mon has quit IRC
[00:47:36] * ryepup_ has quit IRC
[00:47:36] * imajes has quit IRC
[00:47:36] * pacopablo has quit IRC
[00:48:02] * xjjk has joined #trac
[00:48:04] * Chutt has joined #trac
[00:48:11] * thansen has joined #trac
[00:49:17] * killing-joke has joined #trac
[00:49:17] * Leentje has joined #trac
[00:49:18] * ping|desert has joined #trac
[00:49:19] * otaku42 has joined #trac
[00:49:20] * fredb has joined #trac
[00:49:41] * seemant has joined #trac
[00:49:45] * mooch has joined #trac
[00:50:10] * pacopablo has joined #trac
[00:50:35] * bartc has joined #trac
[00:51:57] * Tetraboy has joined #trac
[00:52:32] * termie has joined #trac
[00:52:37] * idnar has joined #trac
[00:53:13] * De_Mon has joined #trac
[00:53:21] * Synapse has quit IRC
[00:53:21] * IamLordV1ldemort has quit IRC
[00:53:21] * Epcylon has quit IRC
[00:53:22] * IamLordV1ldemort has joined #trac
[00:54:14] * Epcylon has joined #trac
[00:58:39] * Synapse has joined #trac
[01:00:14] * tic has joined #trac
[01:04:43] * ryepup_ has joined #trac
[01:04:44] * atwork has joined #trac
[01:04:48] * ryepup_ is now known as ryepup
[01:04:58] * imajes has joined #trac
[01:05:02] * matt_good has joined #trac
[01:05:11] * flox|away has joined #trac
[01:05:14] * hpnadig has joined #trac
[01:05:20] * ged has joined #trac
[01:05:32] * hax has joined #trac
[01:06:48] <atwork> i cannot use a svn-repo with 30000+ tags/revisions with trac. Timeline and Browse Repository fails after about ten minutes. is this a known problem?
[01:08:56] <alect> have you tried using trac-admin <trac-env> resync?
[01:09:13] * cmlenz has joined #trac
[01:09:14] <alect> initial repository access triggers a cache run of the changesets
[01:09:37] <alect> if that access occurs through a http request, it could time out before completed
[01:09:57] <alect> completing
[01:13:07] <atwork> alect: tried with resync from console, but it doesn't help
[01:13:32] <alect> how does it fail?
[01:13:44] <alect> @paste
[01:13:44] <evil_twin> alect: "paste" is http://trac.pastebin.com <- paste your text here
[01:13:52] <alect> hmm i think that paste is dead
[01:14:00] <atwork> alect: i'll try it again. it'll take up to 10 minutes before i get an error
[01:15:14] <atwork> alect: ok i have to resync again, because i went back to a smaller repo. it'll take many hours to resync. i'll be back in some hours or tomorrow...
[01:15:54] <alect> ok
[01:15:57] <alect> full backtrace
[01:16:04] <alect> make sure you don't run out of tmp space too
[01:16:31] <alect> sqlite can use tmp for its transaction buffer i believe
[01:24:53] * ranq has joined #trac
[01:28:10] <atwork> alect: any configuration to set to get full trace?
[01:37:04] <alect> urm yeah you could enable full debug logging
[01:37:12] <alect> @logging
[01:37:12] <evil_twin> alect: "logging" is http://projects.edgewall.com/trac/wiki/TracLogging <-- Enable debug logging to file, ensure your environments log/ directory is writeable by your web server user, check for errors.
[01:43:09] * _idostyle has joined #trac
[01:43:27] * flox|away is now known as flox
[01:43:38] * tuxipuxi has joined #trac
[01:47:41] * Riccc has joined #trac
[01:53:54] * boude has quit IRC
[02:00:09] * idostyle has quit IRC
[02:00:13] * _idostyle is now known as idostyle
[02:14:08] * peppelorum has quit IRC
[02:17:01] * peppelorum has joined #trac
[02:25:31] * peppelorum has quit IRC
[02:29:30] * peppelorum has joined #trac
[02:45:15] <alect> trac!
[03:07:33] * peppelorum has quit IRC
[03:42:31] * converter has joined #trac
[03:59:33] <alect> evening trac people
[04:17:51] * Arnar has joined #trac
[04:17:55] <Arnar> hello there..
[04:19:29] <Arnar> I have a .py file in my svn repository - which is in UTF-8 encoding and has a encoding header to match (# -*- encoding: UTF-8 -*-) but when I view it in trac the file is read as latin1 and converted to UTF-8 before being sent to the browser, resulting in double-utf8 encoding
[04:22:23] <Arnar> how do I specify the files charset, perhaps via a svn property?
[04:25:30] * gakman has quit IRC
[04:28:55] <alect> what version of trac?
[04:31:04] <cmlenz> Arnar: set the "svn:mime-type" property (for one file) or in trac.ini the "default_charset" option (for all files)
[04:34:25] * gakman has joined #trac
[04:39:13] <alect> mmmmmmmmmmm
[04:39:14] <alect> bored
[04:48:15] * TMH_ has left #trac
[04:55:09] <Arnar> cmlenz: thanks - I'll try that
[04:56:19] <Arnar> it works - I changed default_charset since all my files are UTF-8
[05:13:41] * ranq_ has joined #trac
[05:24:03] * ranq has quit IRC
[05:28:23] * agile has quit IRC
[05:34:48] * MasterC has joined #trac
[05:40:24] * peppelorum has joined #trac
[05:44:57] * coderanger_ has joined #trac
[06:17:21] * lisppaste5 has joined #trac
[06:20:25] * agile has joined #trac
[06:30:43] * mjg2 has joined #trac
[06:31:38] * mjg2 has joined #trac
[06:33:49] * milk-it has joined #trac
[06:35:13] * msoares has joined #trac
[06:37:43] * mjg2 has quit IRC
[06:38:18] * mjg2 has joined #trac
[06:38:39] * msoares has quit IRC
[06:39:32] * mjg2 has left #trac
[06:40:14] * mjg2 has joined #trac
[06:42:56] * danbeck has joined #trac
[06:46:33] <atwork> alect: still bored? now i din't get any error when accessing the timeline, but it takes about 30min to open it, when i have 'repository checkings' checked...
[06:48:53] <alect> did resync finish?
[07:00:05] <atwork> yes
[07:01:01] <atwork> it's working but really slow. now i chose to view timeline just one day back. it still takes ages if i have checked "repository changes".
[07:02:09] * xjjk has quit IRC
[07:04:30] <coderanger_> atwork: What if you turn off changeset_show_files (set it to 0)
[07:05:16] <atwork> coderanger_: it has always been 0
[07:07:19] <alect> what version of trac are you using?
[07:07:35] <atwork> 0.9.6
[07:08:02] <alect> can you try trunk?
[07:08:55] <atwork> can i just replace site-packages\trac with trunk?
[07:10:04] <alect> back up your trac repo
[07:10:18] <alect> nuke site-packages/trac
[07:10:32] <alect> and install
[07:10:46] <alect> err, back up your trac environment
[07:11:10] * xjjk has joined #trac
[07:11:36] <coderanger_> "trac repo": Does not computer! :)
[07:12:38] <alect> heh
[07:12:40] <alect> shaddap you
[07:17:39] <atwork> i guess i can install it by just replacing site-packages/trac with the new one
[07:18:08] <alect> yep
[07:18:49] * milk-it has quit IRC
[07:20:47] <coderanger_> Would trac-admin need changing?
[07:21:03] <alect> possibly
[07:21:24] <atwork> all right i clicked timeline with ticket changes checked and 1 day back. i'll go for a coffee and report back if something happens
[07:22:35] <atwork> how do i know trac is using Subversion SWIG Python bindings and not PySVN?
[07:23:14] <coderanger_> atwork: Because it will _only) work with the former ;-)
[07:23:24] <coderanger_> wow, I can't type this morning
[07:23:59] <atwork> swedish keyboard :P
[07:24:04] <nextime> coderanger_ : where are you from? here are 16:22
[07:24:43] <coderanger_> US
[07:25:04] <coderanger_> Its 10:23 here
[07:27:39] <coderanger_> Just settling in for a looooooong day at work
[07:29:48] <atwork> still no timeline coming up. without repository checking is takes less than a second
[07:31:10] * milk-it has joined #trac
[07:31:25] <coderanger_> have you tried nuking the repo cache?
[07:31:52] <coderanger_> delete from node_change; delete from revision;
[07:32:26] <atwork> coderanger_: i can nuke the cache straight from the db like that?
[07:32:29] <coderanger_> yep
[07:32:36] <coderanger_> then do a resync from trac-admin
[07:32:49] <coderanger_> thatl make very very sure that everything is in sync
[07:33:25] * nuonguy has joined #trac
[07:36:53] <atwork> wtf Command failed: table node_change has no column named node_type
[07:37:29] <coderanger_> trac-admin /path/to/env upgrade
[07:37:49] <atwork> then i cannopt rollback if something is wrong with trunk, right?
[07:37:57] <coderanger_> itl backup your db for you
[07:38:09] <coderanger_> because Trac is cool like that ;-)
[07:38:39] <coderanger_> (assuming you use sqlite of course)
[07:38:59] <atwork> sure trac is cool, but my boss is not :D
[07:40:08] <coderanger_> in the db folder youll see things like trac.db.<version>.bak
[07:43:27] * dna has quit IRC
[07:43:32] <atwork> this will takes some hours. i'll be back tomorrow
[07:44:36] * atwork has quit IRC
[07:44:37] * dna has joined #trac
[07:55:38] * ged has quit IRC
[07:56:30] * ged has joined #trac
[08:02:14] * klasstek has joined #trac
[08:08:27] * tolsen has quit IRC
[08:19:56] * neoxed has quit IRC
[08:32:43] * neoxed has joined #Trac
[08:33:23] * ged has quit IRC
[08:35:43] * neoxed has quit IRC
[08:44:52] * neoxed has joined #Trac
[08:46:29] <pacopablo> morning
[08:47:27] <coderanger_> afternoon
[08:47:41] <pacopablo> bah, it's not noon there yet :)
[08:48:08] <coderanger_> soon enough
[08:48:21] <coderanger_> (not soon enough actually)
[08:57:54] <coderanger_> sqlite supports the .rowcount thingy, right?
[08:58:15] <pacopablo> in what? SQLAlchemy? or dbapi2 ?
[08:58:26] <coderanger_> dbapi2
[08:58:36] <pacopablo> not sure, but yes, I think so
[08:58:49] <coderanger_> cursor.execute('SELECT * FROM members WHERE user=%s',(user,))
[08:58:50] <coderanger_> assert isinstance(cursor.rowcount, int) and cursor.rowcount >= 0, "Got rowcount as %r"%cursor.rowcount
[08:59:02] * hpnadig has quit IRC
[08:59:39] <coderanger_> dies with .rowcount is -1
[09:00:38] <cmlenz> "This read-only attribute specifies the number of rows that the last executeXXX() produced"
[09:00:43] <cmlenz> see http://www.python.org/dev/peps/pep-0249/
[09:00:54] <cmlenz> "The attribute is -1 in case no executeXXX() has been performed"
[09:01:18] <coderanger_> Yeah, but why doesn't cursor.execute() qualify?
[09:01:22] <cmlenz> what you want is "SELECT COUNT(*)" ;-)
[09:01:31] <cmlenz> oh
[09:01:48] <cmlenz> you're quite right, I guess
[09:05:19] * tolsen has joined #trac
[09:09:50] <coderanger_> I guess I can just use count
[09:10:07] <coderanger_> Just seems odd that that didnt work
[09:11:03] <cmlenz> I think it may be because sqlite never buffers the result set
[09:11:08] <cmlenz> pysqlite2, that is
[09:11:28] <cmlenz> http://lists.initd.org/pipermail/pysqlite/2006-January/000357.html
[09:15:46] <coderanger_> hmm, I guess that makes sense
[09:18:08] * technel has joined #trac
[09:18:16] * otaku42 is now known as otaku42_away
[09:18:17] <technel> How do you create a new Trac wiki page? I can't find the button :\
[09:18:28] <pacopablo> simply create a reference to one
[09:18:36] <pacopablo> or type in a new page URL
[09:18:48] <pacopablo> and then click the "Create this page" button
[09:19:01] <pacopablo> if you dn't see that button, then you don't have WIKI_CREATE permissions
[09:19:18] * MasterC has quit IRC
[09:19:21] <technel> pacopablo, Oh, I see, thanks
[09:19:49] * Arnar has quit IRC
[09:20:48] * ping|desert has quit IRC
[09:31:41] <technel> What is the difference between priorities and severities?
[09:31:51] <pacopablo> no much of anything
[09:32:06] <pacopablo> it's more of an issue of history
[09:32:20] <pacopablo> I believe that severities is empty by default
[09:33:50] <technel> I think I personally prefer priorities over severities
[09:34:15] <pacopablo> if you remove all entries for either, then that option won't be shown
[09:34:45] <technel> pacopablo, Mmkay
[09:43:47] * chandlerc has quit IRC
[10:03:43] <tuxipuxi> cmlenz, hm wow.. the Trac demo is down because "no such table: system
[10:03:44] <tuxipuxi> " but i've no idea why
[10:03:53] <tuxipuxi> it's the latest trac version, so no known sql injection
[10:05:42] <tuxipuxi> system, node_change, permission and auth_cookie are missing
[10:08:53] <cmlenz> there's a ticket about that
[10:19:32] <cmlenz> tuxipuxi: using sqlite?
[10:19:55] <cmlenz> tuxipuxi: and already fixed? :-P
[10:24:57] * _dmax has joined #trac
[10:25:25] * s0undt3ch_ has joined #trac
[10:30:58] * s0undt3ch has quit IRC
[10:30:58] * s0undt3ch_ is now known as s0undt3ch
[10:31:07] <pacopablo> anyone have a snippet of code that will return the permutations of agiven set of numbers?
[10:36:23] * flox has quit IRC
[10:37:44] * dmax has quit IRC
[10:37:48] * _dmax is now known as dmax
[10:39:10] <coderanger_> you mean (1,2) => [(,), (1,), (2,), (1,2)]
[10:39:10] <coderanger_> ?
[10:40:47] <pacopablo> yes, though I don't actually care about the empty set
[10:41:08] <pacopablo> and order doesn't matter, so (1,2) would be the same as (2,1)
[10:46:34] <coderanger_> hmm
[10:46:35] <coderanger_> one sec
[10:47:36] <tuxipuxi> cmlenz, i've given the project a fresh start :) but yes, with sqlite
[10:57:06] <coderanger_> pacopablo: Got it
[10:57:07] <coderanger_> I think
[10:57:12] <pacopablo> hehe :)
[10:57:59] <lisppaste5> coderanger pasted "Permutation code" at http://paste.lisp.org/display/24502
[10:58:07] <coderanger_> Not perfect, but a good base
[10:58:24] * osimons has joined #trac
[10:59:44] <coderanger_> the lists end up nested oddly
[11:01:37] * cboos has joined #trac
[11:03:23] <coderanger_> aha
[11:03:41] <lisppaste5> coderanger annotated #24502 with "Better permutations" at http://paste.lisp.org/display/24502#1
[11:03:55] <coderanger_> enjoy :)
[11:04:20] <pacopablo> thanks ;)
[11:04:42] * osimons has quit IRC
[11:05:07] <coderanger_> Probably a better way to do it, but that seems to work
[11:05:22] <pacopablo> efficiency is not my aim here :)
[11:07:50] <coderanger_> its good for me to get back into the functional programming mindset since I am doing a class in scheme this semester :)
[11:08:23] * osimons has joined #trac
[11:09:43] * ryepup_ has joined #trac
[11:10:50] * ryepup has quit IRC
[11:10:55] * ryepup_ is now known as ryepup
[11:13:13] <coderanger_> hmm, I wonder if I could do this without the copy though
[11:13:37] <pacopablo> hmm, not good when running with a list of 12
[11:13:46] <coderanger_> "not good"?
[11:13:56] <coderanger_> as in very slow?
[11:13:57] <pacopablo> almost out of physical mem
[11:14:00] <coderanger_> yeah
[11:14:01] <pacopablo> and yes, bery slow
[11:14:02] <coderanger_> one second
[11:14:04] <pacopablo> er, very
[11:14:43] <pacopablo> hehe, killed it and immediately freed 400 MB of memory :)
[11:15:59] <lisppaste5> coderanger annotated #24502 with "Now with less memory suckage" at http://paste.lisp.org/display/24502#2
[11:16:55] <s0undt3ch> sorry to bug ppl here with generic python questions, but I always seem to get better and quiker answers here...
[11:16:58] <s0undt3ch> is it possible to start the python shell with some attributes assiged? ie, some modules imported, etc?
[11:16:59] <coderanger_> that ought to be somewhat better
[11:17:17] * technel has left #trac
[11:17:45] <coderanger_> s0undt3ch: http://python.org/doc/2.4.2/tut/node15.html#foot1939
[11:17:47] <coderanger_> ;-)
[11:18:25] <coderanger_> also http://python.org/doc/2.4.2/lib/module-user.html
[11:19:22] <s0undt3ch> 1st link only adds history, which is not bad to start with
[11:19:39] <coderanger_> s0undt3ch: Read the footnote
[11:19:57] <coderanger_> If you are just working on debugging something you can put it in a file and use python -i too
[11:20:01] <s0undt3ch> ha ha
[11:20:20] <pacopablo> coderanger_: mucho less memory suckage
[11:20:27] <pacopablo> and a hell of a lot faster too
[11:20:35] <coderanger_> pacopablo: :)
[11:20:42] <pacopablo> btw, 8191 permutations for a list of 13
[11:20:46] <s0undt3ch> thanks coderanger_, your the man, and it verifies, I did got a better and faster answer here :)
[11:21:09] <pacopablo> s0undt3ch: codranger is a python beast!
[11:21:34] <coderanger_> pacopablo: I could optimize it a bit more, but that should be okay
[11:21:49] <pacopablo> returns in about a second, that's good enough for me :)
[11:22:09] <coderanger_> pacopablo: the b = list(...) can be inlined right into the for loop
[11:22:19] <coderanger_> so it doesnlt have to hold the whole list in memory
[11:22:24] <pacopablo> true
[11:22:43] <pacopablo> but I didn't notice any nasty memory consumption in the last iteration
[11:22:57] * jborg has joined #trac
[11:24:24] <coderanger_> s0undt3ch: Its funny to think that I only learned python about this time last year
[11:27:23] <coderanger_> but yeah, #python is usually about as much help as a hole in the head
[11:27:48] <pacopablo> unless you've got a very techinal interesting question
[11:27:56] <pacopablo> then I can usually get some answers
[11:28:19] <pacopablo> of course then they'll debabte about the best wayt o do it, etc. but hey.
[11:29:49] <coderanger_> pacopablo: The kinds of "very technical" questions I usually ask just get the IRC equivalent of blank stares ;-)
[11:29:58] <pacopablo> hehe
[11:31:04] * cmlenz has quit IRC
[11:33:29] <pacopablo> hmm, coderanger_ how can I convert the perm function to limit the result set to permutations of x lenght or less?
[11:33:51] <coderanger_> hmmm
[11:33:53] <pacopablo> I got it, I'll jsut check the length of the resulting list
[11:34:02] <pacopablo> and throw out those that are longer
[11:34:19] <coderanger_> heh, that works
[11:34:39] <coderanger_> [x for x in perm(...) if len(x)>y]
[11:34:45] <pacopablo> yup
[11:35:06] <coderanger_> There is probably sime tricky way to do it so it doesn't even bother computing those
[11:35:08] <coderanger_> hmm
[11:35:21] <pacopablo> well, that's your homework then :)
[11:35:30] <pacopablo> got to get you preped for school :)
[11:35:56] <coderanger_> heh, Ive already warned my AI prof that I am going to be having fun in the class
[11:36:07] <coderanger_> (he knows me and knows how insane I can be when I feel like it)
[11:36:17] <pacopablo> lol
[11:37:48] <s0undt3ch> coderanger_: but you have a general programing background right?
[11:38:32] <pacopablo> he's a CS student
[11:38:34] <coderanger_> s0undt3ch: Yeah, been doing thing since 2
[11:42:08] <s0undt3ch> coderanger_: well, that a huge start, progaming logics are all about the same thing, implementation on the several languages is what's diferent, but if you have a good background, you just adapt easely
[11:42:27] <pacopablo> unless the language is perl
[11:42:33] <s0undt3ch> YEAH!!!
[11:42:38] <s0undt3ch> dam, perl
[11:42:38] <pacopablo> then you just run away screaming
[11:43:06] <s0undt3ch> unfortunatly for the moment I can, that's why I'm using pyperl, the python perl module ;)
[11:43:11] <s0undt3ch> *can't
[11:43:57] <s0undt3ch> which simplefies perl a bit, and we get to continue to use python as the coding language
[11:44:17] <coderanger_> pacopablo: Hmm, the way this is generating the permutations it may not work too well
[11:44:30] <pacopablo> FIX IT! :)
[11:44:45] <pacopablo> checking len() after the fact is sufficient
[11:45:51] * pfn has joined #trac
[11:46:11] <coderanger_> pacopablo: The problem is it needs the smaller lists to build the bigger lists
[12:00:15] * jborg has quit IRC
[12:09:15] * cmlenz has joined #trac
[12:23:43] * klasstek has quit IRC
[12:34:33] * flox has joined #trac
[12:35:43] * flox is now known as flox|away
[12:42:16] * gopokes64 has joined #trac
[12:43:47] <gopokes64> I'm trying to calculate a weighted average of custom fields in trac. Is there a function that can be employed in trac?
[12:43:55] <gopokes64> This is for a custom report
[12:44:07] <De_Mon> gopokes64 you mean other than SQL?
[12:45:10] <De_Mon> select (val*weight + val2*weight2)/2 AS avg from moo
[12:45:12] <gopokes64> De_Mon: No, whenever I create the sql for the report I wind up with 0.0 as a result. seems to be a floating point issue or something.
[12:45:51] <De_Mon> fix your sql then
[12:46:30] <gopokes64> the sql is (c1.value * (25/100)) + (c2.value * (5/100)) + (c3.value * (20/100)) + (c4.value * (50/100)) AS total
[12:47:18] <coderanger_> try putting .0s in there
[12:48:06] <gopokes64> rather than the division by 100?
[12:48:36] <coderanger_> 25.0/100.0 for example
[12:48:49] <gopokes64> k...
[12:48:50] * peppelorum has quit IRC
[12:49:14] <De_Mon> gopokes64 you're using SQLite?
[12:49:31] <gopokes64> De_Mon: yes
[12:49:58] <gopokes64> De_Mon: that fixed it...thanks
[12:52:01] <De_Mon> ... + c1.value * .25 + c2.value * .05 + ... would work just as well
[12:52:21] <gopokes64> I tried that, but it complained about the period
[12:53:00] <gopokes64> ah, must include a leading zero, as in 0.25
[12:53:23] <De_Mon> oh, sorry use to using postgresql ;)
[12:56:42] * Riccc has quit IRC
[13:05:12] * gopokes64 has quit IRC
[13:29:15] * chandlerc has joined #trac
[13:36:00] * Semhirage has quit IRC
[13:36:37] * Semhirage has joined #trac
[13:40:21] * matt_good_ has joined #trac
[13:41:11] * matt_good has quit IRC
[13:45:19] <coderanger_> Have any of you guys worked with Oracle via Python?
[13:45:57] <De_Mon> working with oracle via oracle was bad enough
[13:51:57] <coderanger_> Yeah, but I don't have much of an option in this
[13:57:48] * chandlerc has quit IRC
[14:16:39] * milk-it has quit IRC
[14:17:31] * coderanger has joined #trac
[14:17:51] * Gruf has quit IRC
[14:23:06] * coderanger_ has quit IRC
[14:29:51] <matt_good_> coderanger: yeah, I've used cx_Oracle
[14:31:05] <matt_good_> one thing that sucks about it is that when you're dealing with LONG RAW columns you need to do some extra calls to set the input or output sizes
[14:33:43] <coderanger> Thats the one I'm thought looked best
[14:36:34] <matt_good_> I ended up hacking the setup.py script a bit too so I could get it to build with the way I installed the Oracle libs and headers
[14:37:42] <matt_good_> I'll be biab if you need any tips on it
[14:43:07] * felis_ has quit IRC
[14:52:31] * Ricccc has joined #trac
[14:53:57] * Elephant has joined #trac
[14:55:32] <Elephant> I'm trying to get permissions set up using the methods listed on http://trac.edgewall.org/wiki/FineGrainedPermissions, but it doesn't actually work when I try it on a test repository
[14:56:01] <Elephant> Does anyone here tinker with permissions in trac?
[14:57:48] * matt_good_ has quit IRC
[15:02:36] * chandlerc has joined #trac
[15:06:55] <coderanger> Elephant: What isn't working?
[15:09:01] * matt_good_ has joined #trac
[15:10:15] * matt_good_ is now known as matt_good
[15:13:56] * danbeck has quit IRC
[15:20:57] * agile has quit IRC
[15:21:22] <Elephant> Trac won't actually take permission settings from the auth file I set up
[15:21:39] <coderanger> What are you trying to set up?
[15:22:54] <Elephant> I want to define a set of users and groups in the access file - so I only have to do it once - and then have those permissions propagate to all new Trac projects created in the future
[15:23:49] * cboos has quit IRC
[