Team Chat Logs

2007 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 08, 2007

<
[00:11:52] * em-dash has joined #trac
[00:14:57] * stevegt_ has quit IRC
[00:28:01] * Federico2 has joined #trac
[00:36:13] * Federico2 has quit IRC
[00:40:41] * Tetraboy has quit IRC
[00:47:45] * mbuf has joined #trac
[00:49:51] <mbuf> can we make projects listed in Trac either publicly or make it visible only to internal LAN?
[01:00:27] * coling has joined #trac
[01:14:20] * cbalan has joined #trac
[01:33:07] * cmlenz_ has joined #trac
[01:33:07] * cmlenz has quit IRC
[01:43:12] * cmlenz has joined #trac
[01:46:26] * cmlenz has quit IRC
[01:54:19] * TinoW has joined #Trac
[01:54:22] <TinoW> hell-o
[01:54:27] <asmodai> oya
[01:57:41] * cmlenz_ has quit IRC
[02:10:07] * Muelli has quit IRC
[02:11:11] * Muelli has joined #trac
[02:15:45] <TinoW> where would I find a trac API spec?
[02:21:05] * Tetraboy has joined #trac
[02:23:40] * Muelli has quit IRC
[02:24:05] * tank has quit IRC
[02:26:30] * cmlenz has joined #trac
[02:27:18] * cmlenz has joined #trac
[02:28:00] * prologic has quit IRC
[02:28:49] * prologic has joined #trac
[02:37:29] * columbo has quit IRC
[02:38:48] * columbo has joined #trac
[02:38:52] * mbuf has quit IRC
[02:39:26] * cmlenz_ has joined #trac
[02:40:27] * cmlenz has quit IRC
[02:43:08] * prologic has quit IRC
[02:43:28] * prologic has joined #trac
[02:43:37] * columbo has quit IRC
[02:44:26] * columbo has joined #trac
[02:46:31] <evil_twin> New news from t.e.o: Ticket #5837 (defect created): Using different file-names <http://trac.edgewall.org/ticket/5837> || Ticket #5836 (defect created): Dropdown editor <http://trac.edgewall.org/ticket/5836> || Ticket #5835 (defect created): Error pages <http://trac.edgewall.org/ticket/5835>
[02:54:35] * Bazy has quit IRC
[03:05:21] * qasim has joined #trac
[03:06:35] <qasim> is it possible to change the "available projects" page? I need to categorise it but could not find how to do that
[03:13:42] * moolight has quit IRC
[03:16:39] <evil_twin> New news from t.e.o: Ticket #5836 (defect closed): Dropdown editor <http://trac.edgewall.org/ticket/5836#comment:1> || Ticket #5835 (defect closed): Error pages <http://trac.edgewall.org/ticket/5835#comment:2>
[03:21:54] * Bazy has joined #trac
[03:33:54] <qasim> is it possible to change the "available projects" page? I need to categorise the projects that are listed in it but could not find how to do that
[03:41:07] * mOolight has joined #trac
[03:46:49] <evil_twin> New news from t.e.o: Ticket #5838 (task created): first ticket <http://trac.edgewall.org/ticket/5838> || Ticket #5837 (defect closed): Using different file-names <http://trac.edgewall.org/ticket/5837#comment:1>
[03:53:01] * jae has joined #trac
[03:54:27] <TinoW> qasim: which plugin/macro you are referring to?
[04:01:28] * mbuf has joined #trac
[04:16:55] <evil_twin> New news from t.e.o: Ticket #5838 (task closed): first ticket <http://trac.edgewall.org/ticket/5838#comment:1>
[04:24:57] <qasim> TinoW: I am not referring to any plugin or macro, I just want to know that is it possible to change the "Available Projects" default page and if yes then what is the process of doing that
[04:30:48] * nelhage has quit IRC
[04:35:08] * coling has quit IRC
[04:36:57] * mbuf has quit IRC
[04:39:34] * fwierzbicki has quit IRC
[04:46:30] * kopernikus has joined #trac
[04:50:50] <mitsuhiko> what's the correct way to render a codeblock with pygments in trac0.11?
[04:50:56] <mitsuhiko> {{{#!pycon doesn't work
[05:01:29] * jpatota_ is now known as jpatota
[05:02:52] * coling has joined #trac
[05:07:35] * thatch has joined #trac
[05:19:47] * stevegt_ has joined #trac
[05:21:09] * fwierzbicki has joined #trac
[05:22:13] * fwierzbicki has quit IRC
[05:22:32] * fwierzbicki has joined #trac
[05:28:46] * Esaj has joined #trac
[05:35:14] * stevegt1 has quit IRC
[05:47:35] <evil_twin> New news from t.e.o: Ticket #5839 (enhancement closed): tichet custom fields depending on ticket type <http://trac.edgewall.org/ticket/5839#comment:1> || Ticket #5839 (enhancement created): tichet custom fields depending on ticket type <http://trac.edgewall.org/ticket/5839>
[05:57:49] * qasim has left #trac
[06:00:11] * NeRMe has joined #trac
[06:13:43] * erf has joined #trac
[06:13:46] <erf> Hey
[06:16:06] <erf> I'm having issues with trac, it just produces an internal server error in apache, does anyone have any ideas?
[06:18:32] * coling_ has joined #trac
[06:19:02] * nurv has joined #trac
[06:21:10] * coling has quit IRC
[06:22:23] * coling_ is now known as coling
[06:36:37] * d0rt has joined #trac
[06:36:49] * danbeck has joined #trac
[06:45:40] * Cheez has joined #trac
[06:52:32] <ricky> erf: Does the error log contain anything useful?
[06:54:00] <Cheez> Hi guys, I've got Trac 0.10.2 running using mod_python (python 2.4) on apache 2.2.4 and subversion 1.4.3, and im finding i have to restart apache before new commits to the repository will show up in the source browser. trac-admin's resync command has no effect on it. is that normal or have i screwed something up somewhere?
[06:54:36] * nurv has quit IRC
[07:08:51] * retracile has joined #trac
[07:09:14] * retracile mutters something unflattering about mornings.
[07:17:54] * mnemoc has left #trac
[07:25:01] * idnar has quit IRC
[07:25:03] * idnar_ has joined #trac
[07:35:38] * asmodai pats retracile
[07:35:52] <retracile> how goes it, asmodai?
[07:37:19] <retracile> thatch: looks like [5891] could use a testcase; I'll see what I can put together.
[07:39:12] <asmodai> Pretty decent.
[07:39:19] <asmodai> Getting my pointer stuff straight in my head again
[07:39:55] <asmodai> Only took 5 mins this time.
[07:40:03] <asmodai> So annoying when in Python you're used to
[07:40:08] <asmodai> blah = ""
[07:40:16] <asmodai> and later just: blah = "a new text"
[07:40:26] <asmodai> and in C you're doing char buffer[100];
[07:40:30] <asmodai> strcpy(...)
[07:41:06] <retracile> asmodai: or even malloc() and friends...
[07:41:09] * asmodai nods
[07:41:23] <asmodai> Oh well
[07:41:32] <retracile> pointers are fun. function pointers are even more fun. ;)
[07:41:47] <asmodai> Pointers do rock yes
[07:43:36] <retracile> Hrm.... when I left, I thought all these testcases were passing...
[07:43:39] * retracile is puzzled
[07:50:48] <TinoW> asmodai: you can use PyStringObject in C ;-)
[07:57:30] * Esaj_ has joined #trac
[07:57:49] * Esaj has quit IRC
[08:16:52] * brosner has joined #trac
[08:18:05] * Esaj_ has quit IRC
[08:20:08] <thatch> Cheez: that's not particularly normal. Especially the 'resync does nothing' issue.
[08:20:58] <thatch> Cheez: can you try tracd to see if you have the same issue, so we can rule out Apache?
[08:22:54] * mcr has joined #trac
[08:23:22] <mcr> I'm looking for advice on how to total one column as part of a report.
[08:24:31] <mcr> do I need to hack the report code, is there a plugin I've missed, or is there something you can do in the sql?
[08:25:35] * rfrenz has joined #trac
[08:25:45] <retracile> mcr: Doesn't sql have a SUM() function?
[08:31:42] * rfrenz has quit IRC
[08:34:27] <pacopablo> morning
[08:36:47] <retracile> hey pacopablo
[08:37:39] <pacopablo> how's the little one?
[08:38:04] <retracile> doing well. We're not getting much sleep, but that's to be expected.
[08:38:43] <pacopablo> yup, comes with the territory
[08:39:08] <retracile> Yeah.
[08:40:09] <mcr> retracile, yes, of course I can SUM() things.
[08:40:15] <mcr> but, I want to both show them, and SUM() them.
[08:40:57] <mcr> Are you suggesting that I can somehow synthesize one additional row in the results a give it the title TOTAL?
[08:41:03] <retracile> mcr: Ah, I see. Um, I'd have to play with it to figure that out, sorry.
[08:41:16] <mcr> note that I'd like to ideally sum this per milestone too :-)
[08:41:25] <mcr> for the reports that organize by milestone.
[08:41:53] <mcr> I don't see how I can do this from SQL only. It seems that I'd need to modify the report code.
[08:44:50] * idnar_ is now known as idnar
[08:48:47] <evil_twin> New news from t.e.o: Changeset [5893]: Testing: improve robustness of the user additions <http://trac.edgewall.org/changeset/5893>
[08:51:13] <Cheez> thatch will do so now
[08:51:15] <mcr> it would be cool if someone did a video on how to use the functional testing features in trac.
[08:52:51] <Cheez> hmm,
[08:52:55] <Cheez> need to find something to commit now
[08:53:59] <Cheez> thatch its fine in tracd yeah, must be something apache or mod_python is doing
[08:54:15] * cbalan has left #trac
[09:02:46] * bewst has joined #trac
[09:03:13] <bewst> coderanger: TracHome widgets can't be dragged right now?
[09:04:47] * rfrenz has joined #trac
[09:05:24] <Cheez> http://trac.cheezyblog.net:9001/cbnet/browser/trunk tracd, showing the latest revision (104), http://trac.cheezyblog.net/browser/trunk through apache, not showing it, same environment, vhost config is http://pastebin.ca/650010 nothing in the error log. :/
[09:07:12] <rfrenz> Hi, does anyone have a complete list of dependencies for Silvercity (for syntax highlighting C++)? I'm having trouble building it on Ubuntu 7.04.
[09:08:49] * Epcylon_ has joined #trac
[09:09:13] * Epcylon has quit IRC
[09:09:19] * Epcylon_ is now known as Epcylon
[09:36:33] <coderanger> bewst: If you have javascript on, they can be
[09:36:53] <coderanger> rfrenz: Just use pygments, its pure python
[09:37:14] <asmodai> Pygments, good. Silvercity, bad.
[09:37:16] <rfrenz> thanks coderanger, I'm trying that now
[09:37:20] <rfrenz> I agree so far
[09:37:40] <rfrenz> I was initially scared off by the need for a plugin, since I don't feel like breaking my Trac install
[09:37:52] <rfrenz> I'm unsing 0.10.3
[09:38:16] * pacopablo hasn't had any issues with pygments and 0.10.3
[09:38:55] <mitsuhiko> trac bad, doesn't want to highlight text/x-python-doctest
[09:38:56] <asmodai> pacopablo: still not solved my windows permission problem :(
[09:39:08] <mitsuhiko> guess trac doesn't forward all pygments mimetypes
[09:39:43] <asmodai> mitsuhiko: I found that I sometimes needed to force a cache clear
[09:39:55] <rfrenz> I'm not saying pygments breaks trac, just that I'm afraid I will by installing TracPygments plugin
[09:40:11] <mitsuhiko> rfrenz: trac 0.10 or 0.11?
[09:40:18] <rfrenz> 0.10.3
[09:40:29] <mitsuhiko> then the plugin won't break it :)
[09:40:31] * lakin has joined #trac
[09:40:38] <mitsuhiko> unless you have pygments installed
[09:40:38] <rfrenz> well that's good
[09:40:48] <rfrenz> I do have pygments installed
[09:40:48] <mitsuhiko> *have not
[09:41:23] <rfrenz> I've installed the plugin, I'm now tracking down an error with the python egg cache
[09:42:21] <rfrenz> I add [components] tracpygments.* = enabled to trac.ini
[09:42:26] <rfrenz> then restart apache, and get:
[09:43:00] <rfrenz> The following error occurred while trying to extract file(s) to the Python egg
[09:43:02] <rfrenz> cache:
[09:43:04] <rfrenz> [Errno 13] Permission denied: '/var/www/.python-eggs'
[09:43:38] <coderanger> @wiki TracPlugins
[09:43:39] <evil_twin> coderanger: http://trac.edgewall.org/wiki/TracPlugins
[09:43:53] <coderanger> Set the PYTHON_EGG_CACHE variable to something useful (like /tmp)
[09:44:42] <rfrenz> that's a system variable? or apache?
[09:46:29] * pombreda has quit IRC
[09:48:53] <pacopablo> apache
[09:49:14] <evil_twin> New news from t.e.o: Changeset [5895]: Testing: Merge through r5892 from trunk; all tests PASS <http://trac.edgewall.org/changeset/5895> || Changeset [5894]: Testing: Add a FAILING regression test for #5819 <http://trac.edgewall.org/changeset/5894>
[09:50:02] <retracile> Ok, this is... odd. On the timeline, I seem to be unable to disable 'Wiki changes'. Does anyone else see that?
[09:50:47] <bewst> coderanger: "If you have javascript on, they can be." Nope. Maybe because the RSS reader is continuously "Loading..."
[09:51:23] <coderanger> bewst: Yeah, that widget triggers an error sometimes for some reason. Its not seeing its config, even though its hard coded.
[09:51:27] <coderanger> bewst: Just remove it
[09:52:38] <retracile> hm... t.e.o won't let me enable wiki changes...
[09:53:55] <rfrenz> got it...code highlighting working!
[09:53:57] <rfrenz> thanks all
[09:53:58] <retracile> and a 0.10.3.1 environement won't let me turn off wiki changes in the timeline either. Surely that's not intentional?
[09:54:22] * rfrenz has quit IRC
[09:56:10] * lakin_ has joined #trac
[09:56:14] <coderanger> On the timeline? Works okay for me.
[09:56:53] <retracile> coderanger: yeah, on the timeline.
[09:57:04] <coderanger> http://trac.edgewall.org/timeline?from=08%2F08%2F2007&daysback=14&milestone=on&ticket=on&ticket_details=on&changeset=on&wiki=on&update=Update
[09:57:13] <coderanger> Does that show wiki changes?
[09:57:17] <retracile> yes
[09:57:30] <retracile> ... and I can't get the checkbox to uncheck
[09:58:33] <retracile> Oh goody.... looks like it may be konqueror-specific. :(
[09:58:41] <coderanger> Works in Safari3
[09:58:50] <retracile> works in firefox
[09:59:04] <coderanger> And this is why they are giving up on KHTML ;-)
[09:59:16] <retracile> *sigh*
[09:59:39] * retracile grumbles
[10:00:12] <retracile> ?!? it works in a different konqueror window.... wha?!?!
[10:04:02] * coling has quit IRC
[10:05:11] * bewst has left #trac
[10:12:22] * lakin has quit IRC
[10:24:43] <mitsuhiko> cmlenz_: do you plan to port genshi over to parser (from compiler)?
[10:25:14] <cmlenz_> not sure what you mean
[10:25:37] <cmlenz_> why would I do that?
[10:25:38] <mitsuhiko> cmlenz_: the python compiler package will be deprecated in python 2.6
[10:25:52] <cmlenz_> bah
[10:25:55] <cmlenz_> in favor of?
[10:26:01] <mitsuhiko> cmlenz_: compile() with the ast flag set
[10:26:04] <mitsuhiko> which requires python2.5
[10:26:14] <mitsuhiko> alternative method is the parser module
[10:27:11] <cmlenz_> well, what I need is parsing into AST, transforming the AST, and generating bytecode from the transformed AST
[10:27:16] <mOolight> does the parser work with 2.4?
[10:27:22] <cmlenz_> I guess the first two steps would work with compile() and the new AST
[10:27:30] <cmlenz_> how does the third work?
[10:27:39] <mitsuhiko> cmlenz_: but then you would have to drop python2.3 and python2.4 support
[10:27:44] <mitsuhiko> i guess parser is the best idea
[10:27:58] <cmlenz_> parser is awful, it's too low-level
[10:28:29] <mitsuhiko> i know. for jinja i now wrote a standalone parser
[10:28:56] <mOolight> ninja?
[10:29:04] <mOolight> Jinja Template engine
[10:29:14] <mOolight> nite
[10:29:15] * mOolight has quit IRC
[10:29:31] <mitsuhiko> cmlenz_: >>> compile("1 + 1", "<foo>", "exec", 1024)
[10:38:16] * AlexTrofast has joined #trac
[10:40:47] <AlexTrofast> I had a question about the SQL Queries for viewing tickets
[10:42:37] <coderanger> AlexTrofast: Just ask it
[10:43:04] <AlexTrofast> Oh okay so some of the queries look like this in ORDER BY (status = 'assigned') right?
[10:43:12] <AlexTrofast> But I didn't find that in any standard
[10:43:24] <AlexTrofast> Wouldn't it be better to write them liek CASE WHEN status = 'assigned' THEN 1 ELSE 2 END?
[10:43:33] <AlexTrofast> That's a standard and would be more transportable
[10:44:00] <coderanger> What DB does that not run on?
[10:44:17] * Pectabyte has joined #trac
[10:44:17] <AlexTrofast> Ingres
[10:44:18] <coderanger> It should work as its just an expression
[10:44:22] <Pectabyte> yo guys
[10:44:25] <Pectabyte> hows it going?
[10:44:47] <Pectabyte> So i got a wierd problem with Radiant 1.5
[10:45:06] <AlexTrofast> Ingres2006 doesn't let that query go through
[10:45:08] <Pectabyte> When I try to compile a map in Quake 1 mode it says it can't find hqbsp
[10:45:21] <Pectabyte> where do i get hqbsp?
[10:45:37] <coderanger> Pectabyte: Wrong channel
[10:45:48] <Getty> totally
[10:46:05] <Pectabyte> This is the channel the radiant site gave me
[10:46:54] <coderanger> Pectabyte: They use Trac, as you might guess by the name and topic, this is a channel for Trac
[10:47:07] <Pectabyte> yup
[10:47:20] <AlexTrofast> Well, I just thought I'd propose that change since it would be more portable, I tried it out on SQLite and the new syntax works fine
[10:48:06] <Pectabyte> Im desperatly trying to get help with my issue. I cant find the solution anywhere and nobody has answers
[10:48:24] <pacopablo> Pectabyte: well, I doubt any of us have an answer either
[10:48:34] * pacopablo has never even heard of radiant
[10:49:04] * Pectabyte has quit IRC
[10:51:00] <coderanger> Hmm, I can't find ORDER BY clauses in the grammar for SELECT at all actually
[10:52:08] * coderanger digs out the ANSI standard
[10:52:12] * lakin_ has quit IRC
[10:53:49] <AlexTrofast> Heh
[10:54:41] * lakin has joined #trac
[10:56:42] <coderanger> Yeah, ORDER BY on a simple select doesn't seem to be in any standard
[10:56:54] <coderanger> Its supposed to be for cursors
[10:57:42] * Weedy has joined #trac
[10:58:38] <Weedy> any one using fcgid? it doesnt seem to make a trac "server" any more
[11:00:19] <AlexTrofast> Hmm so how do you tell which one works on most systems? :P
[11:00:41] <coderanger> AlexTrofast: Asking some of my local SQL gurus if they know where any of this is formalized
[11:01:01] <coderanger> If not, I guess we try both forms on sqlite/pg/my and see if they work
[11:01:22] <coderanger> This just looks like a convergent, non-standard extension
[11:01:24] <AlexTrofast> Well the person at Ingres most knowledgeable on SQL said he preferred the CASE version
[11:01:46] <AlexTrofast> Well I tried teh CASE on sqllite and that does work, I haven't tried on MySQL yet
[11:01:57] <coderanger> as long as that won't break our current engines, its all the same to me :)
[11:02:14] <AlexTrofast> Exactly
[11:04:04] <pacopablo> AlexTrofast: are you saying that ORDER BY doesn't work on Ingres?
[11:04:21] <AlexTrofast> ORDER BY (status = 'assigned') doesn't
[11:04:36] <pacopablo> ahh
[11:04:41] <AlexTrofast> BUT: ORDER BY (CASE WHEN status = 'assigned' THEN 1 ELSE 2 END) DESC does the same thing
[11:05:03] <pacopablo> k, that makes more sense
[11:05:15] <AlexTrofast> I tried this new syntax on SQLite and it seems to work as well
[11:05:22] <AlexTrofast> Haven't tried on the other two DBs
[11:05:47] <pacopablo> which report is that?
[11:05:55] <AlexTrofast> Well that is just part of it
[11:06:03] <AlexTrofast> But I think four or five of the reports use syntax like that
[11:06:10] <AlexTrofast> So I had to change it to make it work with Ingres
[11:06:13] <AlexTrofast> To the CASE WHEN style
[11:06:19] <pacopablo> name one, and I'll check it on postgres
[11:07:00] <AlexTrofast> The "My Tickets" query
[11:07:10] <AlexTrofast> SELECT p.value AS __color__,
[11:07:10] <AlexTrofast> (CASE status WHEN 'assigned' THEN 'Assigned' ELSE 'Owned' END) AS __group__,
[11:07:10] <AlexTrofast> id AS ticket, summary, component, version, milestone,
[11:07:10] <AlexTrofast> t.type AS type, priority, time AS created,
[11:07:10] <AlexTrofast> changetime AS _changetime, description AS _description,
[11:07:11] <AlexTrofast> reporter AS _reporter
[11:07:13] <AlexTrofast> FROM ticket t
[11:07:15] <AlexTrofast> LEFT JOIN enum p ON p.name = t.priority AND p.type = 'priority'
[11:07:17] <AlexTrofast> WHERE t.status IN ('new', 'assigned', 'reopened') AND owner = $USER
[11:07:19] <AlexTrofast> ORDER BY (CASE WHEN status = 'assigned' THEN 1 ELSE 2 END), p.value, milestone, t.type, time
[11:07:21] <AlexTrofast> IS how I rewrote it
[11:07:23] <pacopablo> use pastebin
[11:07:29] <pacopablo> lisppaste5: url
[11:07:29] <lisppaste5> To use the lisppaste bot, visit http://paste.lisp.org/new/trac and enter your paste.
[11:08:25] <pacopablo> the CASE syntax works fine for PostgreSQL
[11:08:34] <AlexTrofast> Cool
[11:08:49] <AlexTrofast> The CASE syntax works on MySQL, I just tried the general idea on it, not the Trac though
[11:09:03] <pacopablo> I'd open a ticket for the change
[11:09:21] <AlexTrofast> Alright, thanks guys!
[11:09:27] <coderanger> Yeah, this looks like the right way to do it
[11:09:29] <coderanger> Thanky :)
[11:09:48] * coderanger mutters something about the ANSI guys smoking too much crack
[11:10:43] <AlexTrofast> LOL
[11:11:18] <AlexTrofast> In the ticket should I paste in the change code or just in general what should be done?
[11:12:18] <pacopablo> if you have a diff, then that's preferrable
[11:12:40] <pacopablo> otherwise, just explain what needs to be done and paste the snippet
[11:16:05] * tsb has quit IRC
[11:16:21] * columbo has quit IRC
[11:19:50] <evil_twin> New news from t.e.o: Ticket #5840 (enhancement created): Change to the ORDER BY clause in view ticket queries <http://trac.edgewall.org/ticket/5840>
[11:26:47] * cmlenz_ has quit IRC
[11:28:48] * element has joined #trac
[11:29:28] <element> coderanger: you around?
[11:30:36] * khundeen has joined #trac
[11:34:49] * Weedy has quit IRC
[11:34:57] * Weedy has joined #trac
[11:39:45] * element has left #trac
[11:46:55] * em-dash has joined #trac
[11:48:18] <khundeen> Hello can you tell me the code that calculate the age of file in browser page?
[11:49:20] * bronson has joined #trac
[11:58:06] * coderanger_ has joined #trac
[12:04:15] <retracile> AlexTrofast: We no longer use the "WHERE t.status IN ('new', 'assigned', 'reopened')"... we check that status <> 'closed'
[12:04:37] <retracile> AlexTrofast: That's necessary due to the new configurable workflow.
[12:04:48] <AlexTrofast> Alright, thanks
[12:05:43] <AlexTrofast> Which version is that in? I'm building on the 0.10.4 one
[12:05:56] <retracile> AlexTrofast: 0.11dev has the configurable workflow
[12:06:06] <AlexTrofast> Ahh I see
[12:07:16] <AlexTrofast> So all queries that have the status IN (...) should be status <> 'closed'?
[12:08:01] <retracile> AlexTrofast: In 0.11dev they should be. The default reports have that, and if you upgrade from 0.10.*, it'll make the change for you too.
[12:08:16] <retracile> For 0.10.*, it doesn't really matter.
[12:08:36] <AlexTrofast> Alright, so the custom changes I've made, how will they be affected if I upgrade?
[12:08:44] <retracile> (And initially, I misunderstood what you were changing in the ORDER BY thing.)
[12:09:28] <retracile> If you have status IN (...), it'll get replaced with the status <> 'closed' version instead. The CASE change you made shouldn't be affected.
[12:09:57] * cmlenz has joined #trac
[12:10:17] <AlexTrofast> Okay thanks :)
[12:12:08] <khundeen> retracile: have you use dateinfo() function?
[12:12:15] <mitsuhiko> cmlenz: you could also stay with compiler and ship the python2.4 module :)
[12:12:44] <retracile> khundeen: probably at some point... why?
[12:13:49] * coling has joined #trac
[12:15:55] <AlexTrofast> I made some changes to the schema to allow for Ingres to work, such as renaming the session table to t_session
[12:15:57] * stevegt_ has quit IRC
[12:16:08] <AlexTrofast> Do I need to write an dbxx.py in upgrades to account for that?
[12:17:06] * teefive has quit IRC
[12:17:08] * t5_ has joined #trac
[12:17:18] <coderanger_> Yes, and such things are very tricky to handle
[12:17:32] <coderanger_> session is an ingress keyword?
[12:18:15] <cmlenz> mitsuhiko: well, deprecated != gone
[12:18:30] <cmlenz> I'll check out whether I can get the same functionality with the new AST in 2.5
[12:18:43] <mitsuhiko> cmlenz: yes, but it will give you a stupid warning :)
[12:18:49] <cmlenz> the critical question is only if there's a way from AST->bytecode, but I suppose there is
[12:18:54] <cmlenz> well, you can disable warnings ;-)
[12:19:09] <AlexTrofast> coderanger: unfortunately it is
[12:19:17] <coderanger_> AlexTrofast: We have tried very hard to keep the schema human-readable, and fairly stable. There are a number of scripts and plugins that interact directly with the session table as it has no abstraction model
[12:19:29] <coderanger_> AlexTrofast: So that change would have to made on the fly in your DB backend
[12:20:05] <AlexTrofast> Hmmm
[12:20:11] <evil_twin> New news from t.e.o: TracTimeline edited by anonymous <http://trac.edgewall.org/wiki/TracTimeline>
[12:20:39] <AlexTrofast> So... changing the schema bad idea? :)
[12:20:50] <mcr> btw: I moved from sqlite to postgres (coderanger, your suggestion is working well. see post to -users), and I had to hack my reports a bit due to changes in the SQL.
[12:20:53] <coderanger_> That bit can't realistically be changed
[12:21:02] <mcr> I don't see how to do what I want, btw, with queries.
[12:21:30] <coderanger_> AlexTrofast: At least not without breaking a number of plugins
[12:22:07] <AlexTrofast> Ahh that's a shame, I'll see what I can do
[12:30:59] * khundeen has quit IRC
[12:37:23] * dgynn has joined #trac
[12:38:57] <Weedy> any one using fcgid? it doesnt seem to make a trac "server" any more
[12:43:35] * StPierre has joined #trac
[12:45:06] * Rica has quit IRC
[12:46:03] * pombreda has joined #trac
[12:50:27] <evil_twin> New news from t.e.o: Ticket #5841 (defect created): For RSS feed for source code, all <http://trac.edgewall.org/ticket/5841>
[12:51:05] * adj has joined #trac
[12:54:40] * pombred1 has joined #trac
[12:55:02] * brosner has quit IRC
[12:55:14] * pombreda has quit IRC
[12:59:31] <sadmac> I'm trying to do a screen scraping thing on a trac install, and it isn't setting the trac_auth cookie when I try to log in.
[13:02:45] <adj> are there any frontends that let you rapidly deploy a trac site?
[13:03:02] <adj> so my programmers can just click "new project" fill in a form and hit "go" ?
[13:03:13] * fwierzbicki has quit IRC
[13:11:09] * stevegt_ has joined #trac
[13:24:03] * d0rt has quit IRC
[13:31:12] * jae_ has joined #trac
[13:31:56] * _shawn has quit IRC
[13:32:44] * AlexTrofast has quit IRC