Team Chat Logs

2006 3
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

April 10, 2006

[00:04:59] * truijllo has joined #trac
[00:05:01] <truijllo> hi all!
[00:07:41] <alect> hey
[00:11:35] * cmlenz has joined #trac
[00:12:01] <alect> greetings chris
[00:21:04] * wnorrix has quit IRC
[00:54:55] * peppelorum has quit IRC
[01:00:26] * locus has joined #trac
[01:06:03] <locus> can anyone tell me how to disable some mainnav link icons? afaik there is no such option, do I have to edit/hack a .py script for that?
[01:07:13] <alect> like what?
[01:09:54] <locus> timeline, roadmap and tickets
[01:13:58] <alect> easiest way is probably to remove the permissions for those components
[01:14:18] <alect> otherwise, yes, manual .py fondling required
[01:14:39] * hlb has quit IRC
[01:23:03] * fimion has quit IRC
[01:33:09] * truijllo has quit IRC
[01:40:42] <locus> alect, I've just seen that there's a [component] section in trac.ini that lets you disable/remove any components..fyi
[01:53:28] <alect> yes i am aware of it
[02:04:16] * ValiSystM has joined #trac
[02:06:08] * s0undt3ch has quit IRC
[02:06:42] * MasterC has joined #trac
[02:08:11] <locus> ah, thats probably what you meant with "easiest way is probably to remove the permissions for those components" ;)
[02:11:42] * spike_ has joined #trac
[02:11:45] * spike_ is now known as spike
[02:27:40] * cmlenz has quit IRC
[02:33:27] * eblot_ has joined #trac
[02:57:10] * cmlenz has joined #trac
[02:59:50] * eblot has quit IRC
[03:17:23] * tuxipuxi has joined #trac
[03:33:08] * idnar has quit IRC
[03:33:11] * idnar_ has joined #trac
[03:39:21] * idnar_ is now known as idnar
[03:42:37] * imajes has joined #trac
[03:50:41] <sam`> hi
[03:50:46] <sam`> trac-admin hotcopy returns
[03:50:50] <sam`> Command failed: unicode() argument 2 must be string, not None
[03:50:55] <sam`> (r3116)
[03:56:25] * locus has quit IRC
[04:20:51] * eblot_ has quit IRC
[04:28:42] * converter has joined #trac
[05:02:44] * gluegadget has joined #trac
[05:04:45] <gluegadget> I'm writing an IRC bot for our project that uses trac, you think may it be useful for you?!
[05:05:54] <gluegadget> works fine and now in early stage supports: !roadmap, !ticket and !changeset
[05:09:08] * wereMouse|sleep is now known as wereHamster
[05:13:21] * MasterC has quit IRC
[05:21:26] * otaku42 has joined #trac
[05:37:59] <alect> we have a supybot in #trac
[05:38:05] <alect> @faq faq
[05:38:05] <evil_twin> alect: 'faq' is http://projects.edgewall.com/trac/wiki/TracFaq <-- Read this
[05:40:29] * converter has quit IRC
[05:44:05] * agile has quit IRC
[06:09:16] * imajes_ has joined #trac
[06:09:43] <gluegadget> lemme to show it to you
[06:11:19] * JawsBot has joined #trac
[06:11:32] <gluegadget> !changeset 3100
[06:11:43] <JawsBot> Timestamp: 12/31/05 21:28:40 | Author:pablo
[06:11:43] <JawsBot> Message: Hrrr... fixed a bug
[06:12:09] <gluegadget> !ticket 405
[06:12:15] <JawsBot> Status: new
[06:12:15] <JawsBot> As all of us know some languages like Farsi and Arabic are RTL (Right To Left), and this makes the Jaws localization a bit harder than just translating the strings.We've a quick but dirty solution for this problem that makes the process more pain, and that's to write another template for all gadgets and put them in a theme, so I think we should to make it built-in and put the value in the Jaws Settings
[06:12:33] <gluegadget> !roadmap 0.7.0
[06:12:42] <JawsBot> Complete: 6% | Active: 47 | Closed: 3
[06:14:50] <gluegadget> and now lemmet to make it to work for your project's trac
[06:15:19] * cprov has joined #trac
[06:17:08] * JawsBot has quit IRC
[06:17:15] * JawsBot has joined #trac
[06:17:32] <gluegadget> !ticket 3011
[06:17:42] <JawsBot> Status: assigned
[06:18:36] <gluegadget> !changeset 3018
[06:18:36] <JawsBot> Timestamp: | Author:
[06:18:38] <JawsBot> Message:
[06:18:55] <gluegadget> !roadmap 0.9.5
[06:18:57] <JawsBot> Complete: 14% | Active: | Closed:
[06:21:17] * JawsBot has quit IRC
[06:24:04] <gluegadget> alect, any idea ?
[06:25:39] * imajes has quit IRC
[06:28:01] * imajes_ is now known as imajes
[06:28:30] * agile has joined #trac
[06:31:22] * gromgull has joined #trac
[06:31:52] <gromgull> I just updated to 0.9.4, and now all my wiki pages have all their html escaped...
[06:31:55] <gromgull> like here: http://gnowsistest.opendfki.de/wiki
[06:31:58] <gromgull> any ideas?
[06:33:12] * honda_S2000 has joined #trac
[06:38:29] <ValiSystM> whohw gromgull ... and how does it look like when you edit the page ?
[06:39:13] <gromgull> it's fine when i edit
[06:39:19] <gromgull> i.e. i get normal wiki markup
[06:39:30] <gromgull> preview also breaks though
[06:39:46] <gromgull> this was an upgrade from 0.9.2
[06:40:22] <honda_S2000> hi all i'm working with adminconsoleprovider patch but i don't know how to install
[06:40:26] <honda_S2000> can anyone help
[06:40:50] <honda_S2000> sorry i'm a rookie in this things
[06:42:21] * gromgull the only reason we upgraded was because the timeline page didn't work. And ironically it still doesn't, we get: File "/usr/lib/python2.3/sre.py", line 143, in sub
[06:42:21] * gromgull return _compile(pattern, 0).sub(repl, string, count)
[06:42:21] * gromgull RuntimeError: maximum recursion limit exceeded
[06:42:54] * gromgull has quit IRC
[06:43:04] <honda_S2000> doesn't anyone know how to apply this patch?
[06:43:48] * gromgull has joined #trac
[06:45:06] <gromgull> damn irc client
[06:45:23] <alect> gromgull: i'd try rming /usr/lib/python2.x/site-packages/trac
[06:45:29] <alect> nuke the whole thing and reinstall
[06:45:34] * mgalvin has joined #trac
[06:46:13] <alect> glue: that is pretty cool
[06:46:44] * fromvega has joined #trac
[06:46:46] <fromvega> hello
[06:48:07] <fromvega> I can not login on the Trac system using f****** Internet Explorer
[06:48:14] <fromvega> could you help-me?
[06:51:23] <fromvega> hello
[06:51:28] <gromgull> cool - nuking the python dir worked!
[06:51:29] <gromgull> thanks
[06:53:27] <gluegadget> alect, you think may it be useful for you or being mentioned some-where in track-hacks ?
[06:54:08] <alect> yeah, you could put it in as a hack under scripts perhaps
[06:54:24] <alect> maybe i should add a misc category
[06:54:41] <fromvega> I can not login into trac system using IE. Could you help-me?
[06:59:43] <gromgull> yes, use firefox :)
[06:59:47] <gromgull> (sorry)
[07:05:11] * danbeck has joined #trac
[07:06:10] * muness has joined #trac
[07:06:19] <muness> alect: there?
[07:06:43] <alect> hey muness
[07:06:45] <muness> hey
[07:06:57] <muness> sorted() is still causing problems on 2.3. :(
[07:07:00] * boorad has joined #trac
[07:07:07] <alect> howsat?
[07:07:11] <muness> I wanted to put out a release with your fixes, etc...
[07:07:18] <muness> macros.py:206, 209
[07:07:28] <muness> if there are no tickets tagged, they blow up.
[07:07:35] <muness> (209 blows up)
[07:07:41] <muness> if there are tagged tickets, 206 does.
[07:07:43] <alect> aha
[07:07:49] <alect> that's what these guys are reporting
[07:07:55] <alect> on the ml
[07:08:04] <muness> I didn't see those yet.
[07:08:13] <muness> the key=str hack works.
[07:08:17] <muness> should I just apply that?
[07:08:29] <alect> no, that won't sort the tickets correctly
[07:09:36] <alect> i'll force it to use that sorted on 2.4 for me
[07:09:56] <alect> hey, have you tried the expression filter yet?
[07:10:19] <muness> no.
[07:10:24] <muness> very excited about that. :)
[07:10:33] <alect> yeah! me too :)
[07:11:22] <muness> I was going to suggest that we add "implicit" tags to indicate whether an object was a wiki, a ticket, etc...
[07:11:39] <muness> it looks like you did something like that, right? At least for expressions?
[07:11:52] <alect> hmm, and ditch namespaces?
[07:12:02] <muness> not necessarily.
[07:12:10] <muness> just to simplify queries.
[07:12:12] <alect> actually no, although now that you mention it the examples seem to suggest that
[07:12:23] <muness> I see...
[07:12:28] <muness> what do you think?
[07:12:38] <alect> it currently still uses tagspaces=(wiki,ticket) style
[07:12:43] <muness> ah
[07:12:56] <alect> it's not a bad idea...could ditch tagspaces altogether then
[07:13:05] <alect> hmm
[07:13:06] <muness> So, each tagprovider can register it's implicit tag.
[07:13:07] <alect> although maybe not
[07:13:28] <alect> because a wiki page could be explicitly tagged with "ticket"
[07:13:32] <muness> yep.
[07:13:39] <muness> I consider that a good thing. :)
[07:13:49] <alect> it could just use the tagspace as the default
[07:13:58] <muness> ah.
[07:13:59] <alect> default tag that is
[07:14:04] <muness> makes sense.
[07:14:39] <alect> one issue is that if you get_tags() from an object and it includes the implicit tag
[07:14:49] <muness> why not?
[07:14:57] <alect> do you strip it before applying it when you set_tags()
[07:14:58] <alect> ?
[07:15:01] <muness> a user can always trip themselves up if they are persistent.
[07:15:07] <alect> that might confuse users
[07:15:34] <muness> strip it if it's already implicitly defined, sure.
[07:15:52] <muness> otherwise, I'd say we keep it, although we could constrain it.
[07:16:09] <muness> i.e. not allow wiki entries to be tagged ticket.
[07:16:25] <muness> it's just a thought.
[07:16:30] <alect> that seems a bit too authoritarian i think
[07:16:47] <muness> i agree. I'd rather let them do weird things if they want to.
[07:16:47] * peppelorum has joined #trac
[07:16:54] <alect> hehe
[07:16:59] <alect> users :)
[07:17:07] <alect> hmm, i need a fresh trac env
[07:18:39] <muness> i think we get a lot of power if we allow the implicit tagging type of thing. like a blog entry may later come from a different source without having to modify the query code.
[07:18:56] <muness> or folks could define their tickets as wiki entries if they're so inclined.
[07:19:20] <alect> yes, it's an interesting idea
[07:19:25] <alect> i like it
[07:19:39] <alect> my only concern is if they remove the implicit tag, it will reappear magically
[07:19:42] <alect> which might confuse them
[07:19:47] <muness> i see.
[07:19:53] <alect> or annoy them
[07:20:01] <muness> maybe we don't show it in the edit box?
[07:20:14] <alect> yeah, that could work
[07:20:25] <muness> and then we hide tagspaces from them.
[07:20:36] <muness> that should make them less confused.
[07:20:44] * fromvega has quit IRC
[07:21:01] <alect> yeah
[07:21:18] <alect> maybe we should just get rid of tagspaces
[07:21:24] <alect> if we go the implicit tag route
[07:21:31] <alect> there's no real need for them
[07:22:09] * whitelynx|firest has joined #trac
[07:22:32] <alect> aha!
[07:22:33] <alect> replicated
[07:23:39] <muness> i see what you mean... yeah, one less concept to deal with is usually a good thing.
[07:24:02] <alect> try this:
[07:24:03] <alect> def sorted(iterable, key=None):
[07:24:03] <alect> lst = key and [(key(i), i) for i in iterable] or list(iterable)
[07:24:03] <alect> lst.sort()
[07:24:03] <alect> return lst
[07:24:14] <alect> ah
[07:24:18] <alect> scratch that
[07:25:37] <alect> we don't even use key=
[07:26:12] <muness> yeah, we never do.
[07:26:40] <alect> so we can just use this:
[07:26:40] <alect> def sorted(iterable):
[07:26:40] <alect> lst = list(iterable)
[07:26:41] <alect> lst.sort()
[07:26:41] <alect> return lst
[07:26:58] <exarkun> sort takes a key argument
[07:27:13] <alect> under 2.3?
[07:27:16] <exarkun> pft sucks to 2.3
[07:27:26] <exarkun> ;)
[07:27:27] <alect> yeah
[07:27:31] <alect> all this shit is for 2.3 users
[07:27:33] <alect> very annoying
[07:27:47] <muness> i hear you.
[07:28:04] <muness> but we're stuck with 2.3 until svn python bindings come out for 2.4.
[07:28:10] <alect> yeah :(
[07:28:14] <muness> us poor windows users. :(
[07:28:22] <alect> so that should fix the bug. i'll commit the fix
[07:28:26] * s0undt3ch has joined #trac
[07:28:36] <alect> hmm, should we perhaps have a "devel" branch?
[07:28:45] <alect> and roll back the expression stuff
[07:29:02] * gromgull has left #trac
[07:29:13] <muness> it's alright to leave the sneak preview in.
[07:29:15] <muness> ;)
[07:29:16] <alect> what i'm thinking is have a "stable" branch, which gets tagged for releases
[07:29:27] <alect> and contains just bugfixes
[07:29:33] <alect> so 0.3.1 will be bugfixes
[07:29:46] <alect> and "trunk" is destined for 0.4
[07:29:50] <alect> ?
[07:29:57] <muness> if we have something major, like maybe the get rid of tagspace, sure.
[07:30:10] <muness> but I don't mind rolling out the expression stuff now.
[07:30:15] <alect> yeah that was pretty much what i was thinking of
[07:30:18] <alect> ok
[07:30:50] <muness> I'll copy 0.3 to branches. we can use that for fixes from now on.
[07:31:02] <muness> trunk can be experimental after this release.
[07:31:08] <alect> sounds good
[07:31:41] <alect> amazing how many people don't follow the instructions
[07:31:44] <muness> hehe
[07:31:50] <alect> it's unreal
[07:32:00] <alect> you see the guy with the . instead of _?
[07:32:04] <muness> that's why I want to give them a really simple instruction page and eggs to download.
[07:32:09] <alect> yeah
[07:32:12] <muness> yes, I saw that.
[07:32:25] <muness> but I am really surprised at the number of people using this now.
[07:32:34] <alect> yeah it's good to see :)
[07:32:47] <muness> I wonder what %age of trac users do.
[07:33:28] <muness> and here it was the result of a long night of drinking and futurama . :)
[07:33:33] <alect> hehe
[07:33:34] <muness> originally anyway...
[07:33:41] <alect> futurama rules
[07:34:01] <muness> yeah. they should let him do that and kill the simpsons already.
[07:37:38] <muness> you me to commit that new sorted?
[07:37:42] <muness> it works for me.
[07:37:50] <alect> yeah, go for it
[07:38:03] <alect> i have some expr bugfixes that i need to clean up before i can commit
[07:38:43] * boorad has quit IRC
[07:39:55] <muness> k
[07:40:12] <muness> go ahead and mark trunk as 0.4? or not yet?
[07:42:27] <alect> wait until i've pushed this expr fix...it breaks if you have tags like "alec@foo.com"
[07:42:37] <alect> which you can get a lot of from the ticket system
[07:43:09] <muness> :)
[07:43:15] <muness> that's so cool.
[07:43:20] <alect> yeah :)
[07:43:26] <alect> i like being able to use any ticket field
[07:43:29] <alect> that is very nice
[07:43:38] <alect> i haven't done custom fields yet though
[07:44:13] <muness> Do we need to worry about custom? Do you need that for trachacks?
[07:44:42] <alect> i think that can wait
[07:47:24] * s0undt3ch has quit IRC
[07:50:05] <muness> what's the syntax for customizing ticket fields to search on?
[07:50:05] <muness> [tags.ticket]
[07:50:05] <muness> fields=[component|severity|keywords|priority|owner|reporter|cc|version|milestone|status|resolution|summary|description]
[07:50:48] <alect> comma separated iirc
[07:50:55] <alect> eg. fields=component,severity,keywords
[07:50:56] <alect> etc.
[07:51:16] * tolsen has joined #trac
[07:53:45] <alect> [tags]
[07:53:45] <alect> expression_space = true
[07:53:53] <alect> that's to enable expression support under /tags
[07:54:02] <muness> got it.
[07:54:04] <alect> it *should* be backwards compatible
[07:56:52] <alect> okay, that commit should fix the expr problems
[07:59:51] <muness> I've updated the docs.
[08:00:04] <muness> I'll rebuild the eggs and send the announcement...
[08:00:15] <alect> okay
[08:00:16] <alect> 0.3.1?
[08:00:23] <muness> 0.3.1, yes.
[08:00:40] <muness> Would you check the docs?
[08:00:45] <muness> http://muness.textdriven.com/trac/wiki/tags/Setup
[08:01:48] <alect> yep, looks good to me
[08:02:01] <muness> cool.
[08:04:09] * klasstek has joined #trac
[08:05:15] * cprov is now known as cprov-lunch
[08:07:02] <alect> might get rid of that cache i think
[08:07:10] * s0undt3ch has joined #trac
[08:07:15] <alect> mmm, maybe
[08:08:08] <muness> causing trouble?
[08:08:45] <alect> only if you have, say, a page called "foo" for tag "foo", then delete the page
[08:09:00] <alect> any tags will still refer to the wiki page
[08:09:00] * chandlerc has joined #trac
[08:09:30] <alect> actually, likewise if you create a new page, it will still refer to the /tags/ instanced
[08:09:32] <alect> hmmmmmmmmmmmmmmmmmmmm
[08:09:41] <muness> maybe expire the cache?
[08:10:14] <alect> actually i should be able to use the wikichangelistener code
[08:10:23] <muness> but then we should put in something more flexible that we can use for all tag stuff.
[08:10:23] <alect> flush the cache when the wiki changes
[08:10:49] <muness> short term solution methinks.
[08:11:15] * tolsen has quit IRC
[08:11:17] <muness> but it should work fine.
[08:11:21] <alect> yeah
[08:14:00] <muness> coderanger: does the hackinstallplugin work with tractags? or is there anything I can do to enable that?
[08:15:25] <muness> alect: I was also wondering if there was a way to tag commit messages/changesets ?
[08:15:35] <alect> yes, i've pondered that myself
[08:16:23] <alect> actually, do you want to add this cache flushing to 0.3.1?
[08:16:38] <alect> it'll probably avoid some support queries :)
[08:16:56] <alect> ah, too late :)
[08:18:24] <muness> We can always post another release.
[08:18:46] <muness> I'll list it it the setup page as a possible issue.
[08:19:02] <alect> okay
[08:19:04] <alect> fix committed
[08:20:19] * ryepup__ has joined #trac
[08:23:54] * whitelynx|firest has quit IRC
[08:30:03] * ryepup__ has quit IRC
[08:30:52] * tolsen has joined #trac
[08:42:36] * l33tn3t has joined #trac
[08:49:14] <alect> cor, you know, you could even have virtual tag providers, for things like the timeline
[08:49:37] <alect> it could provide virtual tags for dates
[08:50:01] <alect> so if you searched for 20060401 it would show you all the timeline events from that day
[08:50:08] <alect> holy shit! that's awesome
[08:53:18] * cprov-lunch is now known as cprov
[08:53:28] * l33tn3t is now known as avanu
[08:53:57] * Eridius has quit IRC
[08:54:01] * avanu is now known as l33n3t
[08:54:15] * Eridius_ has joined #trac
[08:58:35] <alect> hmmm, except there's no way for one provider to contribute tags to another providers objects
[09:04:41] <pacopablo> slacker! fix it, now! :)
[09:04:50] <alect> api change!
[09:04:53] <pacopablo> morning all
[09:05:00] <alect> hey :)
[09:05:54] <pacopablo> I'm thinking that I need to take a look at the tags expression syntax
[09:06:00] <pacopablo> could be usefull for blog
[09:06:58] <alect> yeah, likely
[09:08:27] <alect> i wish i'd used it from the beginning rather than the operation=... stuff
[09:10:33] * honda_S2000 has quit IRC
[09:12:55] * whitelynx|firest has joined #trac
[09:14:10] <pacopablo> any page that explains the expression syntax?
[09:14:45] <alect> http://muness.textdriven.com/trac/wiki/tags#ListTagged
[09:15:03] <pacopablo> thanks
[09:15:12] * coderanger_cs has joined #trac
[09:15:13] <Semhirage> just a quick question, did you guys create trac from scratch, or did you use and exisiting web framework for it?
[09:15:28] <alect> actually that quoting thing is a lie now, it automagically quotes anything that's not an operator
[09:15:40] <coderanger_cs> Morning all
[09:15:43] <alect> it was created from scratch afaik
[09:15:44] <pacopablo> Semhirage: I don't know the answer to that, but from looking at the code, my guess is that they created it from scratch
[09:15:46] <alect> hey coderanger
[09:15:53] <pacopablo> morning coderanger_cs
[09:15:54] <coderanger_cs> Semhirage: Trac uses a lot of things, but the core framework is new
[09:15:57] <Semhirage> ok, kinda figured
[09:16:06] <Semhirage> any comments on dJango?
[09:16:19] <alect> i like it
[09:16:32] <pacopablo> my biggest problem with it is it's ORM
[09:16:43] <pacopablo> no multi-column primary keys
[09:16:47] <pacopablo> that's just EVIL!
[09:16:54] <pacopablo> though SQLObject sucks too
[09:17:01] <alect> magic removal branch gets rid of a lot of the cruft
[09:17:10] * l33n3t has quit IRC
[09:17:14] <coderanger_cs> achemy looks very shiny though :)
[09:17:20] <coderanger_cs> *alchemy
[09:17:21] <alect> yeah it does
[09:17:29] <pacopablo> sqlalchemy has a lot of promise
[09:17:44] * fimii has joined #trac
[09:18:22] * whitelynx|firest has quit IRC
[09:18:25] <Semhirage> hrm, can i link sql-alchemy and django together?
[09:18:31] <coderanger_cs> doubtful
[09:18:43] <alect> if you want to experience pain beyond anything you've imagined
[09:18:43] <alect> sure
[09:18:45] <coderanger_cs> Other than ignoring the django ORM stuff totally
[09:18:53] <coderanger_cs> heh
[09:18:59] * whitelynx|firest has joined #trac
[09:19:02] <Semhirage> i don't know a whole lot of python and im trying to port a php/mysql based web app to python/postgres
[09:19:11] <Semhirage> sql-alchemy sounds perfect for the dba end of things
[09:19:14] <coderanger_cs> Thats the problem with any highly integrated system, you cant pick and choose
[09:19:19] <Semhirage> but i have no clue how to do python web work
[09:19:25] <coderanger_cs> use Trac!
[09:19:26] <coderanger_cs> :)
[09:19:33] <alect> i'd use django personally
[09:19:38] <Semhirage> i do use trac, for my game developement dev site
[09:19:40] <alect> the admin interface is gold
[09:20:17] <Semhirage> how heavy is django though, all im really doing is building a web based remote admin system for my dns server
[09:20:37] <pacopablo> django is fairly light, requirements wise at least
[09:20:42] <alect> you might consider web.py in that case
[09:20:55] <alect> http://webpy.org/
[09:20:56] <pacopablo