Team Chat Logs

2006 5
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    

June 15, 2006

[00:02:50] <Blackhex> only note (sorry if I'm wrong): PDF export plugin is still in hack-requests poll
[00:10:26] * hlb is now known as keroro
[00:13:42] * keroro is now known as hlb
[00:13:47] * lightcap has quit IRC
[00:17:44] * maxb has quit IRC
[00:19:49] * maxb has joined #trac
[00:33:58] <alect> fatal: workflow will be integrated in 0.11
[00:34:19] <alect> api is subject to change however
[00:34:45] <alect> blackhex: i think that export is for an export of many pages, not just one...
[00:36:21] * tomster has joined #trac
[00:38:17] * Guard][an has joined #trac
[00:38:31] * Guard][an has left #trac
[00:45:13] * peppelorum has joined #trac
[00:46:16] * phoenixz has quit IRC
[00:52:11] * klando has joined #trac
[01:02:57] * cboos has joined #trac
[01:06:18] * Blackhex has quit IRC
[01:10:11] <asmodai> and back
[01:22:38] <asmodai> mmm
[01:22:38] <asmodai> (For the record, on some programming languages like Python, you can
[01:22:38] <asmodai> use UTF-32 internally - for Python, it is a compilation option, the
[01:22:38] <asmodai> default is UTF-16.)
[01:23:16] <LarstiQ> ucs2 and ucs4
[01:26:12] <asmodai> Those are synonyms :)
[01:30:30] <LarstiQ> not entirely no
[01:31:04] * Hydra^ has joined #trac
[01:40:26] <asmodai> Nearly identical then.
[01:42:10] * Blackhex has joined #trac
[01:43:48] <LarstiQ> asmodai: fair enough :)
[01:45:09] * Hydra_ has quit IRC
[01:45:10] * Hydra^ is now known as Hydra_
[01:47:12] * prisoner has quit IRC
[01:52:41] <asmodai> LarstiQ: They're near identical. Mmm, UCS2 has issues with glyphs higher than U+FFFF and UTF-16 always uses a BOM? Mmm, need to read up again.
[01:54:43] <LarstiQ> asmodai: http://www.unicode.org/faq/basic_q.html#23
[01:55:24] <asmodai> Ah, righto.
[01:55:36] <asmodai> ta.
[01:55:38] <asmodai> So much info.
[01:56:09] * polarfox has joined #trac
[01:57:08] <asmodai> LarstiQ: Slowly trying to specialise myself in cultural backgrounds, g11n, l10n, i18n, typography etc. But the field is quite vast.
[01:59:37] <LarstiQ> asmodai: indeed
[02:00:09] * LarstiQ finds input methods to be the greatest pain
[02:00:31] <LarstiQ> although getting total unicode support right isn't easy either
[02:01:38] <asmodai> LarstiQ: I stopped using Unix on my desktop since I need input methods and even projects like uim are built up from a bunch of subprojects.
[02:01:46] <asmodai> I applaud their goal, I doubt the methods they use.
[02:01:58] <asmodai> So I am stuck on Windows now. And for IME it ain't that bad to use.
[02:02:09] <cboos> alect: hi -- just in case, to prevent duplicated work, I'm currently trying to port the TOC macro to 0.10 --
[02:02:10] <asmodai> Not to mention so many fonts are not updated past Unicode 2.x
[02:02:18] <asmodai> boosje!
[02:02:27] <cboos> hello
[02:02:33] <asmodai> How's tricks?
[02:03:43] <alect> cboos: righto
[02:04:13] * jrydberg has joined #trac
[02:08:36] <alect> it was actually working until the header changes
[02:09:23] <LarstiQ> asmodai: heh, your 'je' gave you away ;)
[02:09:46] <cboos> yep, but looking at the code, it was time to port it to the trunk api anyway ;)
[02:10:01] <cboos> besides, the latest change will make it's implementation slimmer
[02:10:14] <cboos> s/it's/its/ damn syntax
[02:11:26] <alect> what's changed in trunks api re the formatter?
[02:11:46] <alect> are you looking at the toc plugin? or the old macro?
[02:12:02] <cboos> TocMacro ... I think that's the plugin
[02:12:08] <cboos> is there another one?
[02:12:14] <cboos> which one should I look at?
[02:12:22] <alect> there was a pure macro version
[02:12:27] <alect> not a plugin
[02:12:41] <cboos> ah no, that's not this... I'm hacking the tocmacro/0.9/macro.py file
[02:15:38] <cboos> he, what's that: # Bail out if we are in a no-float zone
[02:15:46] <cboos> if 'macro_no_float' in req.hdf:
[02:15:49] <cboos> ?
[02:16:11] <alect> it's a hack so that other plugins can tell macros not to render
[02:16:29] <cboos> what plugin sets this?
[02:16:31] <alect> eg. blogshow, so TOC's don't show up on the blog overview
[02:16:54] <asmodai> LarstiQ: *grin*
[02:16:58] <alect> if formatters had a context associated with them it wouldn't be a problem :)
[02:17:16] <asmodai> LarstiQ: Only for Belgians, Dutchies and people acquainted with Dutch.
[02:17:44] <LarstiQ> asmodai: daar zijn er nogal veel van ;)
[02:18:07] <asmodai> Like my Japanese girlfriend, calls herself aitje ;)
[02:19:09] <cboos> alect: you mean, the formatter having some kind of user properties?
[02:19:52] <alect> yeah
[02:19:56] * peppelorum has quit IRC
[02:19:57] * tuxipuxi has joined #trac
[02:20:33] <alect> ticket, wiki page, etc.
[02:20:57] <alect> hmm perhaps even a stack of contexts
[02:21:20] <cboos> yeah, ok, that would be the main object to which the text being formatted belongs
[02:22:10] <cboos> but also, for this "no float zone" stuff, this could be "if 'macro_no_float' in formatter.properties'' ...
[02:22:39] <alect> that exists already?
[02:22:50] <cboos> no, just ideas for 0.11 ;)
[02:22:56] <alect> aha
[02:25:19] <alect> a stack of context could be damn cool
[02:25:34] <cboos> example?
[02:25:59] <cboos> (as I'm currently thinking context == object to which the text being formatted belongs to)
[02:26:01] <alect> WikiPage -> BlogShow -> WikiPage -> TOC
[02:26:38] <alect> i think of it more along the lines of "the context in which the object is rendered"
[02:26:44] <cboos> ah, those would be kind of rendering contexts ...
[02:26:46] <cboos> right
[02:27:30] <alect> .properties would be good too
[02:28:18] <cboos> but then, we'd better come with a scheme that doesn't for each renderer to know about every other renderer name...
[02:29:13] <alect> yeah
[02:29:16] <alect> tracobjects!!!
[02:29:18] <alect> :)
[02:30:06] <cboos> but actually I think that those are two distinct notions of context ... the rendering context, as you said, which would be a stack
[02:30:27] <cboos> and the "information context", which would be the tracobject to which the text belongs
[02:30:38] <cboos> formatter.object simply
[02:32:21] <alect> hmm well that would have to be a stack too probably
[02:32:51] <cboos> like?
[02:33:05] <alect> eg. if WikiStart has [[WikiInclude(SandBox)]]
[02:33:13] <alect> WikiStart -> SandBox
[02:34:05] * cboos tries to follow...
[02:34:11] <alect> ok..
[02:34:55] <alect> anything in WikiStart will have the WiiPage('WikiStart') object as the context?
[02:36:09] <cboos> when rendering the content of that page, that object will be associated to the renderer/formatter
[02:36:29] <alect> right
[02:37:03] <alect> so when WikiInclude renders SandBox, the object will be WikiPage('SandBox')
[02:37:25] <cboos> but if in the course of doing so, you fetch some other object and (sub)render that object, then you'd associate that object to the new formatter... ok there's a kind of notion of stack, but maybe a stack of active renders, then
[02:37:49] <alect> yeah, exactly
[02:39:55] <alect> we are squishing these 0.10 bugs at a pretty good rate i think
[02:41:00] <cboos> well, not that bad, yes, but unfortunately I'm back at work next week... but maybe there won't be much to do, so I can still find some Trac time ;)
[02:42:18] <alect> heh
[02:42:34] <alect> yes, i will be away next week, then starting a new job
[02:42:45] <alect> so i won't be able to do anything for some time
[02:42:53] <cboos> you made it at googel ?
[02:43:16] <cboos> (mangled on purpose, in case they're listening...)
[02:43:36] <asmodai> lol
[02:43:57] <alect> haha
[02:43:59] <alect> yeah i did
[02:44:09] <cboos> great ;)
[02:44:11] <asmodai> I am trying to get deeper into the trac codebase
[02:44:14] <asmodai> just takes a while to grasp it all
[02:44:31] <alect> yeah pretty awesome :). they're flying me to mountain view for a wekk too!!
[02:44:34] <alect> GOOGLE HQ!
[02:44:47] <cboos> asmodai: start at [T1], and watch it evolve ;)
[02:44:54] <asmodai> if you encounter Ben Sussman, say hi ;)
[02:45:05] <asmodai> cboos: T1 being?
[02:45:11] <cboos> trac:changeset:1
[02:45:13] <asmodai> ah
[02:45:17] <alect> heh, ok
[02:45:22] * asmodai proves the point.
[02:45:35] <alect> i want to find guido and tell him that python rocks :)
[02:45:44] <asmodai> I am happy Guido relented on lambda
[02:45:54] <alect> hell yeah
[02:46:07] <asmodai> Yeah Python rocks. It makes sense to me.
[02:46:10] <asmodai> Perl never did.
[02:46:25] <asmodai> I even did Ruby before I went with Python
[02:46:28] <alect> perl makes me ill
[02:46:42] <alect> ruby also makes me ill
[02:46:46] <asmodai> The perl side of things has been quiet.
[02:46:55] <asmodai> I can remember a few years ago how perl was hyped.
[02:50:42] * jrydberg has quit IRC
[03:05:03] * peppelorum has joined #trac
[03:07:04] * phoenixz has joined #trac
[03:08:33] <cboos> imajes: are you there?
[03:09:15] <imajes> nope :)
[03:09:26] <imajes> cboos: how can i help
[03:09:47] <cboos> have you upgraded yet?
[03:09:58] <cboos> the error seems to be still there...
[03:10:01] <imajes> no, i haven't compiled in yet
[03:10:05] <imajes> 101 things to do last night
[03:10:51] <cboos> no problem, I was just wondering if my fix didn't work ...
[03:11:48] <imajes> sure... i'll apply it later and get back to you
[03:11:57] <imajes> now, however, time for more sleep after being woken up!
[03:12:07] <imajes> (not by you - phone call)
[03:17:40] * Riccc`plula has joined #trac
[03:32:58] * chandlerc has quit IRC
[03:34:00] * chandlerc has joined #trac
[03:35:15] * Riccc has quit IRC
[03:47:24] * chandlerc has quit IRC
[03:48:20] * chandlerc has joined #trac
[03:53:38] * chandlerc has quit IRC
[03:54:36] * chandlerc has joined #trac
[04:06:52] <maxb> alect: I'm fixing TocMacro to work with trunk - would you prefer a patch which is cleanest, but drops 0.9 support, or a patch which duplicates code in 0.10, but works with 0.9 too?
[04:07:33] <cboos> maxb: ugh, did that this morning...
[04:07:57] * chandlerc has quit IRC
[04:08:11] <maxb> cboos: But not committed yet?
[04:08:13] <cboos> maxb: have a look at #441 on trac hacks
[04:08:54] * chandlerc has joined #trac
[04:09:08] <cboos> ... that would be alect
[04:10:00] <cboos> I've attached the fixed file on #441, though. You could test that and enhance it further
[04:10:34] <cboos> btw, that file also works as a standalone plugin (simply drop it in the plugins folder of your env)
[04:12:45] <maxb> Ugh. Reporter of #441 didn't check for existing tickets. I already had it filed as #439
[04:30:01] * erral has joined #trac
[04:30:08] * erral has left #trac
[04:53:44] * matt_good has quit IRC
[04:54:09] * matt_good has joined #trac
[04:55:25] * vmx has quit IRC
[05:04:28] * vmx has joined #trac
[05:17:23] <Blackhex> alect: ok, I have bugfixes patch for doxygen plugin, tested with tracd and apache+mod_python, should work now, now only left to wait :-)
[05:22:54] <maxb> Help, I'm confused! How can req.hdf['wiki.action'] cause a KeyError, yet accessing the same url with ?hdfdump=1 show wiki.action *is* present in the HDF !?!
[05:25:19] * matt_good has quit IRC
[05:25:41] * matt_good has joined #trac
[05:47:12] * coderanger_ has joined #trac
[05:50:58] <alect> evening
[05:53:01] <coderanger_> alo
[05:53:18] <coderanger_> lots of fun bugs for Toc I see
[05:53:38] <alect> yeah :)
[05:53:48] <alect> and fixes as well, which is handy ;)
[05:54:49] <coderanger_> heh, I need work to stop taking up all my time so I can get back to Trac hacking
[05:57:45] * Blackhex has quit IRC
[06:02:44] * klasstek has joined #trac
[06:02:55] * Blackhex has joined #trac
[06:13:14] <maxb> OK.... I've figured it out... the wiki.action HDF var isn't getting populated until *after* wikimacros have been rendered :-(
[06:14:30] <asmodai> from #Django:
[06:14:30] <asmodai> << BlueAidan_work>> Have I said I love trac?
[06:23:30] * prologic_ has joined #trac
[06:24:13] * agile has quit IRC
[06:25:19] * prologic has quit IRC
[06:28:18] <alect> hehe
[06:28:53] * sudoer has quit IRC
[06:30:10] <maxb> Are there any conventions for marking trac tickets as containing a patch awaiting review?
[06:30:37] <alect> [patch] as the summary prefix i've seen before
[06:30:59] * mgalvin has joined #trac
[06:31:28] <cboos> maxb: I use 'review' as a keyword
[06:31:59] <maxb> Should I do one or both to #1890?
[06:32:38] <cboos> keyword is enough, the [patch] stuff is from pre-TicketTypes days ;)
[06:32:50] <alect> 0ld sk00l
[06:33:32] <coderanger_> cboos: I saw you bumped the extras patch to 0.11
[06:34:06] <cboos> well, yes, it seems cmlenz is the most qualifeid to integrate this
[06:34:18] <cboos> and by the time the worldcup is over, 0.10 will be out ;)
[06:34:29] <alect> heh
[06:34:43] <coderanger_> Heh
[06:35:05] <coderanger_> The only problem is there is currently no way to give optional dependencies without breaking the load ordering
[06:35:30] <coderanger_> so anyone using easy_install may get more than they bargain for
[06:36:00] <alect> i don't quite understand what extras do for you
[06:36:26] <alect> ah unless it drags stuff in as a dep if it exists
[06:36:39] * alenoosh has joined #trac
[06:36:42] <asmodai> cboos: I find [patch] useful for report/1
[06:36:46] <coderanger_> Optional dependencies, and you can mark entry points as only to be loaded if a certain optional set is loadable
[06:36:48] <alect> it's primarily for enforcing order
[06:36:52] <alect> ?
[06:36:57] <alect> ah that is cool
[06:36:58] <coderanger_> For Trac thats the big thing
[06:37:13] <alect> how does one specify the optional entry point?
[06:37:36] <coderanger_> throw [name_of_extra] after the module name
[06:37:50] <alect> very cool
[06:38:00] <coderanger_> it only works in >=0.6b1 though, as PJE had to fix some stuff in the parser
[06:38:40] <coderanger_> The optional loading you can do in your code with try/excepts, so thats not as important to me as the ordering
[06:38:58] <alect> yeah that's why i was wondering
[06:39:01] <coderanger_> Its just a handy by-product ;-)
[06:39:42] <asmodai> so cmlenz is watching soccer eh? ;)
[06:39:54] <alect> that's our suspicion
[06:40:02] <asmodai> Jeez, I'm European and I don't!
[06:40:19] <alect> cboos: i might be unavailable for > 1 month rather than just a couple of weeks
[06:40:20] <asmodai> ;)
[06:40:31] <asmodai> alect: noooooooooooooo
[06:40:43] <alect> i know!
[06:40:48] <asmodai> lol
[06:40:54] <asmodai> the roundup URL got jinked
[06:40:57] <asmodai> http://wiki.python.org/moin/CallForTrackers
[06:41:07] <alect> hopefully i'll be able to irc from google
[06:41:24] <asmodai> I think ben (svn developer) does it as well.
[06:41:32] <coderanger_> alect: 20% time :P
[06:41:37] <asmodai> coderanger_: that at least :D
[06:41:45] * asmodai asks the typical question
[06:41:50] <alect> ok, although i'll be visiting so i may not have a desk
[06:41:52] <asmodai> alect: So, whatcha gonna do @ google?
[06:42:12] <coderanger_> Playing with big expensive toys
[06:42:41] <alect> i have no idea :)
[06:42:45] <alect> i'm in the sysadmin team though
[06:42:51] <klando> uuhh
[06:42:55] <alect> but they do everything from kernel dev through to system builds
[06:42:58] <alect> so....
[06:43:20] <alect> i wonder why somebody removed the roudnup url
[06:43:23] <alect> weird
[06:43:29] * asmodai slips alect a BSD cd
[06:43:36] <alect> hah!
[06:43:59] <asmodai> T ALECT B FIXINK LINUKS WIF BSD K THX
[06:44:16] <alect> hehe rofl
[06:44:45] <asmodai> I think you're person #4 or 5 I know who now works for google
[06:44:55] <coderanger_> I would imagine they are trying setup an example tracker
[06:44:57] <alect> you will be assimilated
[06:44:59] <alect> we are the borg
[06:45:25] <asmodai> alect: If I could work on g11n/l10n/i18n, culture, languages, typography, I'd be happy.
[06:45:33] <asmodai> mmm, epydoc looks ok
[06:46:04] <Blackhex> asmodai: I started working on wiki i18n macro
[06:46:12] <alect> epydoc looks cool
[06:46:30] <alect> if it had extensible file traversal code,wecould use it for pydocplugin
[06:46:33] <asmodai> Blackhex: Ah cool
[06:47:02] <Blackhex> asmodai: already have stolen country flags from KDE :-)
[06:47:27] <alect> actually i'm thinking of a different one
[06:48:03] <alect> PUDGE
[06:48:11] <alect> http://pudge.lesscode.org/
[06:48:15] <alect> that one is nice
[06:49:00] <asmodai> Blackhex: What's your aim?
[06:49:12] <asmodai> alect: done with epydoc: http://djangoapi.quamquam.org/trunk/
[06:50:26] <alect> ooh nice
[06:50:38] <alect> bit overwhelming
[06:51:17] <asmodai> true
[06:51:37] <Blackhex> asmodai: that sould be macro [[Translation(en:WikiStart, cz:HlavniStanka]] which displays absolute possitioned icons of english and czech flags, first it detects language from request header and redirect to apropriate page or when user click on apropriate flags it will use that lagnuage as default
[06:52:05] <alect> blackhex: wouldn't that be better off in a plugin rather than a macro?
[06:52:18] <alect> it could add the flags to metanav or mainnav
[06:52:30] <alect> have a table of wiki page mappings
[06:52:59] <alect> and lookup and redirect in the nav handler
[06:53:10] <Blackhex> alect: by macro i mean plguin that implements IWikiMacroProvider interface
[06:53:55] <alect> yeah cool, but wouldn't it be better to not have to modify the page to insert the macro?
[06:54:00] <asmodai> Blackhex: There's a problem with flags
[06:54:12] * danbeck has joined #trac
[06:54:14] <Blackhex> asmodai: there are not free?
[06:54:15] <asmodai> Blackhex: e.g. for English you probably used the Union Jack?
[06:54:41] <asmodai> i.e. The United Kingdom's flag
[06:55:32] <asmodai> Blackhex: see http://www.cs.tut.fi/~jkorpela/flags.html
[06:55:33] <Blackhex> us - USA flag gb - UK flag au - Austalian flag ie
[06:55:42] <asmodai> Explains it quite nice why flags are problematic.
[06:56:02] * alenoosh has quit IRC
[06:56:58] * agile has joined #trac
[06:58:02] <Blackhex> asmodai: ok look of icons could be configurable
[06:58:45] <alect> flags are a lot more compact than country names imo
[06:59:07] <Blackhex> alect: i thinks so it is more straightforward
[06:59:09] <alect> and the iso codes like "en" may be known by users
[06:59:17] <alect> may NOT be know...
[07:00:34] <Blackhex> and someone could't notice that there is something to change language when there will be only country codes
[07:00:56] <alect> yeah
[07:01:09] <Blackhex> but if you like it you could pick display style it in configuration
[07:01:28] <alect> i suspect most people from colonies will know the flag of their colonisers (eg. english -> australia/america, spanish ... other places :))
[07:03:22] * vmx has quit IRC
[07:04:33] <Blackhex> alect: back to your suggest: there could be icons in mainnav or ... and table which defines what page belongs to what language but i think that macro specification of connected pages should be easy-to-use
[07:04:43] * vmx has joined #trac
[07:05:18] <Blackhex> you simply put that macro on page beggining and dont have to care about other administration
[07:05:39] <coderanger_> Blackhex: This sounds like a good place to use the pre-request handlers ;-)
[07:05:50] <Blackhex> coderanger_: :-)
[07:06:23] <Blackhex> coderanger_: should be faster than redirect :-)
[07:07:13] <alect> fair enough, but it just feels a bit clunky to have to edit an existing page to define the translations for it
[07:07:14] * alect shrugs
[07:08:08] <alect> TAGS!!!
[07:08:08] <Blackhex> alect: ok, there could be that macro way with ie webadmin way together
[07:08:19] <Blackhex> or tags :-)
[07:08:23] <alect> :)
[07:09:16] <Blackhex> alect: don't know how tags works inside so can't imagige how could I do that
[07:09:33] <Blackhex> but I'll take a look
[07:10:41] <alect> i was being a bit silly, but you could tag the HlavniStanka page with WikiStart:cz
[07:10:43] <alect> or something
[07:10:48] * alect gesticulates wildly
[07:11:49] <Blackhex> I was thinking that this will be simple plugin but it is getting swell up :-)
[07:11:57] <alect> hehe
[07:12:12] <alect> your first thought was probably the right one ;)
[07:12:33] <Blackhex> you mean guestbook?
[07:12:46] <alect> nah, making it simple
[07:12:54] <Blackhex> ah, ho
[07:12:57] <Blackhex> so
[07:13:39] <Blackhex> so you think that all whoud be integrated in TagIt ?
[07:13:49] <Blackhex> macro
[07:14:09] <alect> TagIt is not used anymore, you just put your tags in the input tag field when editing the page
[07:15:17] <Blackhex> alect: ok I'll consider everything again and will get know tomorrow
[07:16:56] <Blackhex> but that floow in tag cloud really scares me :-)
[07:17:05] <Blackhex> s/flow/flood
[07:17:34] * tat_ has quit IRC
[07:18:17] <alect> you can configure it so it limits the tags shown :)
[07:18:26] <alect> eg. only show tags that have > 3 references
[07:31:04] <alect> i wish webadmin were going into 0.10
[07:31:29] <asmodai> alect: disabled by default I'd hope ;)
[07:31:49] <alect> hell no
[07:31:54] <alect> enabled
[07:31:57] <maxb> What's the best way to deal with very small patches (i.e. move 2 lines a bit earlier in a function)? ticket? email to trac-dev@..?
[07:31:59] <asmodai> Not a potential security issue from the box?
[07:32:11] <asmodai> maxb: tickets I'd reckon
[07:32:12] <alect> not if it needs TRAC_ADMIN, as it does
[07:32:33] <alect> maxb: what's the patch? the hdf stuff for macros?
[07:32:39] <alect> mail it to me and i'll check and commit
[07:32:48] <alect> alec@swapoff.org
[07:34:03] <alect> actually just pastebin it :)
[07:34:06] <maxb> Yes, assignment to the HDF too late for macros to see.
[07:37:06] <maxb> http://trac.pastebin.com/710775
[07:40:31] * ged has joined #trac
[07:40:58] <alect> maxb: action can be modified after assignment to the hdf
[07:41:07] <alect> eg. 109
[07:41:41] <maxb> 109?
[07:41:48] <alect> line
[07:42:33] <asmodai> alect: ok, perfect :D
[07:42:50] <maxb> Oh. Hmm.
[07:47:08] * matt_good has quit IRC
[07:48:36] <imajes> cboos: you haven't told me which section of the conf that abs_href lives in
[07:49:26] <cboos> imajes: that's [trac] base_url
[07:49:57] <imajes> i just added that - no luck?
[07:50:49] <cboos> hm... you mean that doesn't get propagated to the links in the RSS, or you simply still have the former error?
[07:51:15] <imajes> same error
[07:51:17] <cboos> ah still the error
[07:51:22] <cboos> I saw it just now
[07:51:28] <imajes> base_url is set
[07:51:31] <cboos> ... but you did upgrade, did you?
[07:51:35] <imajes> yes
[07:51:39] <cboos> to ... ?
[07:51:43] <imajes> latest svn
[07:51:50] <imajes> as of ~20 minutes ago
[07:52:15] <cboos> ok... now can I see the stack trace?
[07:52:26] <imajes> sure
[07:53:46] <imajes> nothing in access/error log- i'll turn on debugging
[07:54:26] <imajes> cboos: wait, what are the valid options for log_type ?
[07:54:35] <cboos> stderr
[07:54:36] <cboos> file
[07:54:38] <imajes> ah great
[07:54:56] <cboos> @logging
[07:54:56] <evil_twin> cboos: "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.
[07:55:34] * lightcap has joined #trac
[07:56:09]