| [00:01:40] |
* |
hpnadig has quit IRC |
| [00:02:29] |
<pacopablo> |
well, it's bed time for me. |
| [00:02:34] |
<pacopablo> |
don't have too much fun! :) |
| [00:02:35] |
<pacopablo> |
later |
| [00:02:49] |
<cboos> |
bye |
| [00:06:48] |
<imajes> |
cboos: that patch fixed it |
| [00:06:49] |
<matt_good> |
cboos: I think that's really overcomplicating the situation |
| [00:06:52] |
<imajes> |
thanks |
| [00:07:03] |
<matt_good> |
cboos: re the pooling |
| [00:07:49] |
<matt_good> |
either adding it to the request object, or using something like paste.registry should be simpler, and portable to other uses |
| [00:08:26] |
<matt_good> |
directly adding a bunch of attributes to the request isn't great, but maybe we should have some standard for dealing with request-local objects |
| [00:10:47] |
<matt_good> |
cboos: I need to get some sleep, but I'll follow up with this on the dev list tomorrow |
| [00:13:48] |
* |
cmlenz has joined #trac |
| [00:18:19] |
* |
otaku42_away is now known as otaku42 |
| [00:20:44] |
<cboos> |
matt_good: I've got the patch already, I think it's quite clean |
| [00:21:08] |
<cboos> |
(it does mostly everything in the RepositoryManager) |
| [00:22:04] |
<cboos> |
cmlenz: hi, nice work on the error.html page ;) |
| [00:22:27] |
<cboos> |
does this mean Trac should rush to web 2.0 (or rather invent web 3.0 ;) )? |
| [00:22:37] |
<cmlenz> |
heh |
| [00:22:55] |
<cboos> |
btw, I noticed a typo (or what looks like a typo) |
| [00:22:55] |
<cmlenz> |
just because of a very tiny bit of animation? |
| [00:22:59] |
<matt_good> |
cboos: you're welcome to paste it, but I don't really see the point in a "pool" when all you need is to make the object local to the request |
| [00:23:20] |
<cboos> |
- /*]]>/**/</script> |
| [00:23:22] |
<cboos> |
+ /*]]>*/</script> |
| [00:23:28] |
<matt_good> |
er, not here |
| [00:23:32] |
<cmlenz> |
cboos: oh, right |
| [00:23:48] |
<cboos> |
matt_good: it's simpler than a pool |
| [00:24:08] |
<cboos> |
just a cache per thread, let me paste the patch ... |
| [00:24:50] |
<cboos> |
http://pastie.caboo.se/18286 |
| [00:25:25] |
<cboos> |
we already have the shutdown(tid) in place, so I reused it |
| [00:27:22] |
<cmlenz> |
that's basically a threading.local(), no |
| [00:27:24] |
<cmlenz> |
? |
| [00:27:37] |
<asmodai> |
cmlenz / cboos / matt_good : morning |
| [00:27:45] |
<matt_good> |
cmlenz: yeah, basically |
| [00:28:56] |
<matt_good> |
apparently there was an issue before with performance when using "env.get_repository()" repeatedly, which I just reintroduced |
| [00:29:28] |
<matt_good> |
which, we shouldn't need to recreate repository objects within the same request |
| [00:30:04] |
<cmlenz> |
yeah |
| [00:30:29] |
<matt_good> |
so we need to make the repository instance accessible across the life of the request |
| [00:31:08] |
* |
moraes has quit IRC |
| [00:32:08] |
<cmlenz> |
yeah, for that the patch looks good I think |
| [00:32:15] |
<cboos> |
cmlenz: yeah, hopefully we switch to 2.4 rather sooner than later |
| [00:32:28] |
<matt_good> |
threading.local isn't in 2.3? |
| [00:32:34] |
<cboos> |
so that we have real thread local storage |
| [00:32:40] |
<cboos> |
no, not AFAIK |
| [00:32:49] |
<cboos> |
it's a 2.4 thing |
| [00:33:01] |
<cmlenz> |
yup |
| [00:33:37] |
<cmlenz> |
we could copy over one of the various backports of threadlocal into trac.util, but that's not really worth it IMHO |
| [00:33:49] |
<cmlenz> |
adds more fragile code |
| [00:34:37] |
<cboos> |
so I'll stress that code a bit more and commit the patch as a follow-up to r3971, ok? |
| [00:34:45] |
<matt_good> |
well, I believe that the patch won't work if someone uses different repositories within the same thread |
| [00:35:04] |
<cboos> |
matt_good: sure, but that's not the case right now ;) |
| [00:35:05] |
<matt_good> |
they'll just get back the first one repeatedly |
| [00:35:28] |
<matt_good> |
cboos: not in Trac itself, but Bitten might, or some other script/plugin |
| [00:35:45] |
<cmlenz> |
I think the API is limited to just one repos right now in any case |
| [00:35:49] |
<cboos> |
(I'm about to work on multi-repository support, and sure at that time I'll adapt the code) |
| [00:36:04] |
<cmlenz> |
self.repository_dir is an Option |
| [00:36:16] |
<cboos> |
cmlenz: yeah, right you can't even specify more than one rpos right now |
| [00:36:46] |
<matt_good> |
cmlenz: ok, that's a Component subclass, I'm just looking at the patch ATM |
| [00:37:08] |
<cmlenz> |
matt_good: to use multiple repositories, RepositoryManager is the wrong layer... you'd just use the SubversionRepository API directly |
| [00:37:41] |
<matt_good> |
ok, I was missing the context of that class based on what was in the patch |
| [00:37:57] |
* |
stevegt has quit IRC |
| [00:38:40] |
<cboos> |
actually, in the VcRefactoring, I want to turn the repository in a full fledged trac object (er, resource that is ;) ) |
| [00:38:59] |
<cboos> |
so you can add/remove them at will in the WebAdmin, |
| [00:39:28] |
<cboos> |
list them in a virtual "root" folder, attach a wiki description to each, that kind of thing... |
| [00:40:26] |
<cboos> |
but I haven't had time to start this, though the data model begins to be clearer |
| [00:40:47] |
* |
x^k has quit IRC |
| [00:41:18] |
<matt_good> |
cmlenz: is that a problem for Bitten, will it reuse the same RepositoryCache? if so it will need some tweaking to be able to resync with the repository, which is currently only done once per cache-object |
| [00:42:05] |
<cmlenz> |
hmm |
| [00:43:06] |
<angrymike> |
any chance to see the Ticket System w/ webadmin able to add custom ticket entries ? |
| [00:43:20] |
<cboos> |
matt_good: well, you could alwyas clear the 'synced' flag, if you really want to ensure a test for resync is attempted ... |
| [00:43:35] |
<cmlenz> |
matt_good: bitten calls sync() unconditionally, does that always sync anymore? |
| [00:44:02] |
<cmlenz> |
doesn't, even |
| [00:44:29] |
<cboos> |
cmlenz: you're right, sync() doesn't check the synced flag |
| [00:44:43] |
<cmlenz> |
http://bitten.cmlenz.net/browser/trunk/bitten/queue.py#L166 |
| [00:44:45] |
<cboos> |
only the get_changeset methods |
| [00:44:50] |
<cmlenz> |
yeah |
| [00:45:06] |
<cboos> |
so it's not an issue |
| [00:45:08] |
<matt_good> |
ok |
| [00:45:32] |
* |
Guardian has quit IRC |
| [00:45:52] |
<cboos> |
speaking about stress testing ... nobody has seen a problem in the recent trunk w.r.t concurrent requests? |
| [00:46:41] |
<cboos> |
(I'm getting crashes in pysqlite, but so far, that's only on a windows 64 machine with Python 2.5) |
| [00:49:11] |
<matt_good> |
I guess the thread-local will work for now, though maybe after switching to setuptools we could add paste.registry to the WSGI stack so there'd be a nice general way of dealing with request-local variables |
| [00:50:31] |
* |
tkp has joined #trac |
| [00:56:06] |
<cmlenz> |
matt_good: I think we should be safe as the component is bound to an environment, thus the thread-locals are too... why would we need stacking? |
| [00:58:24] |
<matt_good> |
cmlenz: stacking may not be necessary, but it'd just provide a common way that any Trac component or plugin could add a request-local without copying that same code |
| [00:58:32] |
<kemik> |
how do i increase the maximum attachment size? |
| [00:58:45] |
<matt_good> |
kemik: trac.ini |
| [00:59:00] |
<kemik> |
matt_good: thx |
| [01:00:20] |
<matt_good> |
wtf happened to vim.org? |
| [01:01:37] |
<cmlenz> |
matt_good: but using threading.local on a Component seems a lot simpler |
| [01:01:50] |
* |
Guardian has joined #trac |
| [01:02:12] |
<cmlenz> |
paste.registry seems to be about globals, and the component system already addresses that |
| [01:02:25] |
<cmlenz> |
multiple app instance isolation in one process, that is |
| [01:05:11] |
<matt_good> |
supposing we could just use a threading.local, it'd still need hooked into the request lifecycle somehow since you want to create the Repository at the beginning and remove it at the end |
| [01:06:23] |
<matt_good> |
thus the shutdown(tid) method, which isn't something plugins can just extend |
| [01:09:55] |
<cmlenz> |
do we actually shutdown at the end of every request? |
| [01:10:03] |
<cmlenz> |
I don't think so |
| [01:13:05] |
<matt_good> |
cmlenz: it's part of dispatch_request |
| [01:13:43] |
<cmlenz> |
but only if wsgi.multithread is false |
| [01:13:54] |
<cmlenz> |
that should probably check wsgi.run_once |
| [01:14:31] |
* |
Getty has quit IRC |
| [01:15:20] |
* |
Getty has joined #trac |
| [01:15:51] |
<cboos> |
matt_good: we could eventually make the plugin be notified when shutdown() is called |
| [01:16:48] |
<cmlenz> |
when an environment is shutdown, in theory it'd be correct to delete all component instances... maybe that'd be enough of a notification? |
| [01:17:02] |
<cboos> |
;) |
| [01:17:17] |
<cboos> |
I was talking about the shutdonw(tid) actually |
| [01:17:42] |
<cmlenz> |
and I wasn't? :-P |
| [01:18:10] |
<cboos> |
I think you were talking about shutdown() (without "tid") |
| [01:18:22] |
<cboos> |
shutdown() => server exits |
| [01:18:42] |
<matt_good> |
cmlenz: the env objects are cached, so I believe the component objects are kept in memory too |
| [01:18:44] |
<cboos> |
shutdown(tid) => request handled by thread tid is finished |
| [01:19:17] |
<matt_good> |
cmlenz: but yes for the RepositoryManager to function correctly that should be wsgi.run_once |
| [01:19:21] |
<cboos> |
(which is even no the same as thread tid finished, think thread pool, like in twisted) |
| [01:19:52] |
<cboos> |
ok, I'm not really familiar with wsgi.run_once, care to explain? |
| [01:20:24] |
<matt_good> |
cboos: CGI is probably the only case it's True |
| [01:20:40] |
<matt_good> |
the process will terminate at the end of the request |
| [01:21:07] |
<cmlenz> |
"This value should evaluate true if the server or gateway expects (but does not guarantee!) that the application will only be invoked this one time during the life of its containing process. Normally, this will only be true for a gateway based on CGI (or something similar)." |
| [01:21:34] |
<matt_good> |
but you might have a forked FastCGI which is not "multithreaded", though the process will be woken to handle many different requests |
| [01:23:19] |
<cboos> |
IIUC, you say that we should always call shutdown(tid), except when we are sure the process is going away (i.e. wsgi.run_once==true), right? |
| [01:24:33] |
<cboos> |
yet there's the "(but does not guarantee!)" part which makes me think we should rather always call shutdown(tid) ;) |
| [01:25:27] |
<matt_good> |
yeah, I don't think it hurts to always call shutdown |
| [01:25:32] |
* |
flox has joined #trac |
| [01:25:46] |
<matt_good> |
(i.e. only hurts CGI, which sucks anyways) |
| [01:28:51] |
<matt_good> |
ok, now I *really* need to get to sleep |
| [01:28:52] |
<matt_good> |
later |
| [01:28:59] |
<cboos> |
;) |
| [01:29:05] |
<cboos> |
good night |
| [01:37:07] |
* |
MrDebout has quit IRC |
| [01:44:22] |
* |
stevegt has joined #trac |
| [02:18:54] |
<Fatal_2> |
hi guys.. i was wondering if there were any trac hacks or anything to be able to maybe put an icon next to the tickets that have file attachments |
| [02:21:13] |
<Fatal_2> |
(when you do a custom search / report ) |
| [02:21:48] |
<Fatal_2> |
and/or possibly being able to filter by (attachments/no attachments) in the custom search page |
| [02:44:30] |
* |
minsu_ju_ has joined #trac |
| [02:45:22] |
* |
minsu_ju has quit IRC |
| [02:52:39] |
* |
hiteck has joined #trac |
| [02:52:53] |
<hiteck> |
hi to all! |
| [02:54:19] |
* |
hiteck has quit IRC |
| [02:58:12] |
* |
blake_ has quit IRC |
| [03:24:13] |
* |
hpnadig has joined #trac |
| [03:25:23] |
* |
sgorilla78 has quit IRC |
| [03:29:58] |
* |
hpnadig has quit IRC |
| [03:43:44] |
* |
jesterKing is now known as amino |
| [03:44:03] |
* |
groton has joined #trac |
| [03:44:05] |
<groton> |
hello all |
| [03:44:43] |
<groton> |
which version of python do i need? i am seeing py2.3+, but as far i could remember the newer versions of python was not working with trac, but not sure... |
| [03:52:01] |
<neuralis> |
groton: 2.4 works fine, and 2.5 has a problem with clearsilver that might have actually been resolved by now (i haven't checked). |
| [03:52:17] |
<groton> |
ah, ok, so the problem was on a dependency, thanks |
| [03:54:41] |
* |
Guardian has quit IRC |
| [03:55:43] |
* |
number5 has quit IRC |
| [04:19:08] |
* |
blake_ has joined #trac |
| [04:22:30] |
<groton> |
i installed trac .10, restored a 9.9.5 trac hot copy, but i am getting an error (dunno which though :)) |
| [04:22:53] |
* |
matt_good has quit IRC |
| [04:23:18] |
* |
matt_good has joined #trac |
| [04:23:25] |
<groton> |
in the "Available project" page Ii get: "trac_root: Error " |
| [04:23:43] |
<groton> |
and my trac enviromnet is stored in c:\trac_root |
| [04:26:55] |
<groton> |
of course if i try to upgrade the environment, i get an error: C:\Python24\Scripts>..\python trac-admin c:\trac_root upgrade ----> Command failed: |
| [04:27:09] |
<groton> |
but the reason of the failing, well, it is unknown |
| [04:40:33] |
* |
amino is now known as jesterKing |
| [04:42:37] |
<asmodai> |
ok |
| [04:42:41] |
<asmodai> |
all validation errors gone |
| [04:42:51] |
* |
asmodai now is moving to the AttributeError range |
| [04:46:35] |
<cboos> |
cmlenz: http://trac.edgewall.org/changeset/3982 ;) |
| [04:58:40] |
<tsb> |
Is it possible to add aliases to wikipages? Or make entries case-insensitive? |
| [04:59:24] |
<tsb> |
I am using the "autowikifyplugin" to match words elsewhere, but I want "Something" (beginning of sentence) to be matched as well as "something". |
| [05:02:59] |
* |
Guardian has joined #trac |
| [05:06:25] |
<cmlenz> |
cboos: now you removed the label/field association :-) |
| [05:09:50] |
<cboos> |
oh, yes, forgot about that, wanted to add the for= attribute ... will do |
| [05:14:39] |
* |
ak|ra has left #trac |
| [05:15:31] |
* |
tsb has quit IRC |
| [05:15:32] |
<coderanger> |
cypromis: SYN |
| [05:22:42] |
* |
moonwatcher has joined #trac |
| [05:22:55] |
<moonwatcher> |
hello |
| [05:23:20] |
* |
omry has quit IRC |
| [05:35:15] |
* |
jesterKing is now known as amino |
| [05:38:04] |
<groton> |
anyone could help me please? |
| [05:51:08] |
* |
matt_good has quit IRC |
| [05:51:27] |
<groton> |
oh, i fixed it, i just forgot to install pysql :) |
| [05:52:19] |
* |
matt_good has joined #trac |
| [05:54:53] |
* |
tsb has joined #trac |
| [05:55:00] |
<cypromis> |
hmmm |
| [05:55:04] |
<cypromis> |
it seems to work now |
| [05:55:11] |
<cypromis> |
although i need to test the changtelog subscriptions |
| [05:58:05] |
* |
lovedaddy has joined #trac |
| [05:59:07] |
* |
milk-it has joined #trac |
| [06:07:26] |
* |
Fastly has joined #trac |
| [06:08:27] |
<Fastly> |
hi i've installed trac and getting a seg fault when i try to load trac in the browser. i'm using the tracd server. what is the best way for me to diagnose the problem (e.g. find out more information about what is causing the seg fault) so that I can post a bug or a message in the trac forum. thanks! |
| [06:08:40] |
<Fastly> |
? ... that was a question ;-) |
| [06:09:26] |
<coderanger> |
Fastly: Can you run python -c "from svn import fs" |
| [06:09:42] |
<coderanger> |
and do other pages work |
| [06:10:56] |
<Fastly> |
yes.. that works |
| [06:11:43] |
<lovedaddy> |
quick question - you know when you allocate a ticket, is there a way to make it so the assigned field is a drop down box with a list of users? |
| [06:11:50] |
<coderanger> |
how about python -c "from pysqlite2 import test; test.test()" |
| [06:12:04] |
<coderanger> |
lovedaddy: restrict_owner in trac.ini |
| [06:13:05] |
<lovedaddy> |
dang it, trac is great, cheers! |
| [06:13:18] |
<Fastly> |
coderanger, that causes an error |
| [06:13:29] |
<coderanger> |
Fastly: Whats the error? |
| [06:13:53] |
<Fastly> |
http://pastebin.ca/208146 |
| [06:14:24] |
<coderanger> |
Fastly: are you using a debianish distro? |
| [06:14:31] |
<Fastly> |
RHEL3 |
| [06:14:45] |
<coderanger> |
hrmm, looks like sqlite3.x isn't installed |
| [06:15:00] |
<coderanger> |
not sure what RPM thats in |
| [06:15:51] |
<Fastly> |
if i just type 'sqlite' it goes into a sqlite prompt and says it's SQLite version 3.3.8 |
| [06:16:24] |
<coderanger> |
thats weird, the sqlite binary should always be 2.x |
| [06:16:32] |
<coderanger> |
the 3.x binary is "sqlite3" |
| [06:16:42] |
<Fastly> |
weird |
| [06:16:48] |
<coderanger> |
sounds like RH is screwing with things :-/ |
| [06:17:18] |
<Fatal_2> |
is there any trac hacks or anything to be able to maybe put an icon next to the tickets that have file attachments? |
| [06:17:43] |
<Fatal_2> |
when u do a report/custom search |
| [06:18:51] |
<Fastly> |
should i try and install the sqlite RPM? |
| [06:18:55] |
<Fastly> |
sqlite 2... |
| [06:19:14] |
<coderanger> |
Fatal_2: Not in 0.10, unless you do silly things |
| [06:19:33] |
<coderanger> |
Fastly: No, you want to use sqlite3 |
| [06:20:08] |
<Fastly> |
so the fact that sqlite points to 3 shouldn't be a problem, right? |
| [06:20:45] |
<coderanger> |
Fastly: Not itself, but it indicates that something may be amiss with file naming or something |
| [06:22:56] |
<coderanger> |
Fastly: http://rpm.involution.com/sqlite3/ |
| [06:23:10] |
<Fatal_2> |
coderanger: ok.. how about just as a simple 'yes'/'no' column in a custom report.. somehow join with the attachments table.. if there are any rows for this ticket, 'yes' otherwise 'no' |
| [06:23:11] |
<coderanger> |
Theres a somewhat recent RPM |
| [06:23:29] |
<coderanger> |
Fatal_2: With enough SQL, probably |
| [06:23:48] |
<coderanger> |
Join to the attachment table and use a CASE |
| [06:23:50] |
<Fastly> |
i'm running EL3, not 4... will it still work ok? |
| [06:24:12] |
<coderanger> |
Fastly: No idea, its because of things like this that I dumped RH after 9 |
| [06:25:06] |
<Fatal_2> |
coderanger: cool.. that would be enoough.. just to save having to look at each ticket manually... coz if we have any sql scripts or whatever to complete a ticket, we attach them to the ticket.. and when we promote a milestone into a new environment we need to ggrab all of the attached sql files and run them |
| [06:25:10] |
<Fastly> |
yeah, i agree that RH sucks balls... this is my rackspace hosting server.. i wouldn't dream of using it otherwise |
| [06:25:26] |
<Fatal_2> |
so i might check that out when i get to work in the morning |
| [06:25:54] |
<Fastly> |
actually i compiled sqlite3 from source because i couldn't find an RPM |
| [06:26:02] |
<Fastly> |
and it wasn't in my up2date |
| [06:26:22] |
<coderanger> |
Fastly: Did you compile pysqlite yourself? |
| [06:27:14] |
<Fastly> |
yes |
| [06:27:21] |
<coderanger> |
weird |
| [06:27:26] |
<Fastly> |
pysqlite-2.3.2 |
| [06:28:17] |
* |
sobersabre has joined #trac |
| [06:28:21] |
<coderanger> |
I would try rebuilding pysqlite and see if you can force it to find the library |
| [06:28:30] |
<sobersabre> |
hi, i have trac not able to show the svn repository |
| [06:28:52] |
<sobersabre> |
I see that in /usr/lib/python2.4/site-packages I have all the required modules, |
| [06:28:59] |
<sobersabre> |
but trac uses python2.3 |
| [06:29:07] |
<sobersabre> |
how can I make it use python2.4 |
| [06:29:08] |
<sobersabre> |
? |
| [06:29:16] |
<coderanger> |
sobersabre: Install the modules for 2.4 |
| [06:29:30] |
<sobersabre> |
coderanger They are installed for 2.4, |
| [06:29:32] |
<coderanger> |
Trac will work with either 2.3 or 2.4 (or 2.5 for current trunk) |
| [06:29:46] |
<sobersabre> |
coderanger, fact is that it tries using 2.3 |
| [06:30:07] |
<sobersabre> |
the files it fails to load are from /some/where/python2.3/balabla |
| [06:30:12] |
<coderanger> |
sobersabre: That just means that you have "python" as a symlink to "python2.3" |
| [06:30:41] |
<sobersabre> |
coderanger ... wrong. |
| [06:30:43] |
<sobersabre> |
:( |
| [06:30:50] |
<sobersabre> |
the link is towards 2.4 |
| [06:30:52] |
<coderanger> |
If you run is with "python2.4" it will force it |
| [06:31:07] |
<sobersabre> |
how do I do this ? |
| [06:31:22] |
<coderanger> |
How are you running Trac? |
| [06:31:29] |
<Fastly> |
hmm... just tried rebuilding pysqlite and got the same error that i posted in pastebin |
| [06:31:43] |
<coderanger> |
mod_python uses the Python it is compiled against |
| [06:31:47] |
<sobersabre> |
coderanger: apache2, mod_python |
| [06:31:57] |
<sobersabre> |
shall I reinstall mod_python then ? |
| [06:32:03] |
<coderanger> |
Other methods use the shebang at top of the file |
| [06:32:12] |
<coderanger> |
sobersabre: You would need to rebuild it |
| [06:32:18] |
<sobersabre> |
ok... |
| [06:32:20] |
<sobersabre> |
I'll try. |
| [06:32:33] |
<coderanger> |
Fastly: Ticket on pysqlite.org + Ask the ML? |
| [06:32:49] |
<Fastly> |
i extracted pysqlite tarball again and did "python2.5 setup.py build" then "python2.5 setup.py install" |
| [06:33:04] |
<coderanger> |
Fastly: Ahh, you are using 2.5 |
| [06:33:10] |
<Fastly> |
yes |
| [06:33:23] |
<coderanger> |
Fastly: 2.5 doesn't need pysqlite, its now a standard module |
| [06:33:31] |
* |
eblot_ has joined #trac |
| [06:33:34] |
<jave> |
is it possible to change the wiki syntax used in trac? I would like it to be compatible with emacs muse |
| [06:33:40] |
* |
chandlerc has quit IRC |
| [06:33:42] |
<sobersabre> |
coderanger , any ideas how do I rebuild mod_python for python2.4 on debian ? |
| [06:33:45] |
<coderanger> |
Fastly: And you will need the current trunk to use it with Trac (and ClearSilver won't work) |
| [06:34:08] |
<coderanger> |
sobersabre: I think debian unstable has 2.4 related stuffs somewhere |
| [06:34:10] |
<Fastly> |
i couldn't find a RHEL3 python RPM greater than 2.2.3 so compiled it from source and symlinked python to it |
| [06:34:36] |
<Fastly> |
current trunk of python...? |
| [06:34:41] |
<coderanger> |
jave: You can add new syntax through plugins |
| [06:34:45] |
<coderanger> |
Fastly: No, of Trac |
| [06:34:53] |
<jave> |
coderanger: ok cool |
| [06:35:14] |
<Fastly> |
i'm running 0.10 |
| [06:35:28] |
<coderanger> |
Fastly: Yeah, that won't run under 2.5 |
| [06:35:33] |
<Fastly> |
ahhh |
| [06:35:42] |
<Fatal_2> |
jave: check out www.trac-hacks.org for some extra macros that people have created.. |
| [06:35:43] |
<jave> |
coderanger: but I dont want to add, I want to change the syntax, like using * for h1 instead of = h1 = |
| [06:36:03] |
<sobersabre> |
coderanger, how can I check which python is mod_python built against ? |
| [06:36:05] |
<coderanger> |
2.5+Trac is still somewhat painful because ClearSilver won't compile against it |
| [06:36:05] |
<Fastly> |
i'll try installing from the SVN |
| [06:36:19] |
<coderanger> |
sobersabre: No idea, my package manager handles that for me |
| [06:36:20] |
<Fastly> |
would you recommend downgrading python? |
| [06:36:29] |
<coderanger> |
Fastly: For now, probably |
| [06:36:35] |
<Fastly> |
cool, thanks |
| [06:36:42] |
<coderanger> |
Fastly: Unless you actually need something from 2.5 |
| [06:37:00] |
<Fastly> |
2.4.4? |
| [06:37:16] |
<coderanger> |
jave: Its easier to just add new syntax than remove the old |
| [06:37:42] |
<jave> |
coderanger: ok I see. |
| [06:37:42] |
<jave> |
|
| [06:37:51] |
<coderanger> |
Fastly: Looks like thats not released yet, I use 2.4.3 on here |
| [06:37:59] |
<alect> |
what the hell |
| [06:38:06] |
<alect> |
i have an email from rms |
| [06:38:15] |
<coderanger> |
ahh, it was released last week |
| [06:38:16] |
<Fastly> |
http://www.python.org/download/ lists it as a standard release.. |
| [06:38:21] |
<Fastly> |
k |
| [06:38:38] |
<coderanger> |
Fastly: On the page it says its an RC1, but it shouldnt really matter |
| [06:38:40] |
<coderanger> |
alect: ? |
| [06:39:06] |
<alect> |
about supporting trac in order to resist jira |
| [06:39:11] |
<Fatal_2> |
you know you've made it when rms graces you with an email :) |
| [06:39:24] |
<alect> |
hehe |
| [06:39:44] |
<coderanger> |
alect: Because clearly those JIRA bastards are evil and whatnot |
| [06:40:10] |
<coderanger> |
I hear they get together with the Launchpad people and kick puppies |
| [06:40:41] |
* |
Guard has joined #trac |
| [06:40:46] |
<alect> |
well, he does have a point i think |
| [06:41:07] |
<asmodai> |
alect! |
| [06:41:12] |
<alect> |
hey hey |
| [06:41:18] |
<asmodai> |
alect: How's it going? |
| [06:41:28] |
<coderanger> |
alect: Did you see that amusing page edit on t-h? |
| [06:41:30] |
<alect> |
not too bad dude. damn hot here atm |
| [06:41:33] |
<asmodai> |
email from RMS? Ye flippin' gods. |
| [06:41:38] |
<asmodai> |
alect: Want some of our cold? |
| [06:41:40] |
<alect> |
coderanger: nope, missed it |
| [06:41:43] |
<coderanger> |
alect: http://trac-hacks.org/wiki/implementation/management/priv/bugreporter/thankyou.php |
| [06:41:44] |
<alect> |
asmodai: please!! |
| [06:41:51] |
<asmodai> |
alect: Oh wait, spring/summer on your side ain't it? |
| [06:41:51] |
<jave> |
incidentally I read rms wouldnt be adverse for a bugtracker for emacs now |
| [06:42:01] |
<sobersabre> |
coderanger ... i managed: my package manager ALSO handles these things: there is libapache2-mod-python2.3 or python2.4 |
| [06:42:05] |
<alect> |
coderanger: hahaha :) |
| [06:42:25] |
<coderanger> |
alect: Someone hasn't heard of wikis clearly ;-) |
| [06:43:00] |
<sobersabre> |
OK, stage one done. |
| [06:43:16] |
<sobersabre> |
is there a way to smoothly move from sqlite to postgres ? |
| [06:43:29] |
<coderanger> |
sobersabre: There is a script for it on trac-hacks |
| [06:44:25] |
<sobersabre> |
cool. |
| [06:44:29] |
<sobersabre> |
thanks. |
| [06:44:46] |
* |
mad_lord has joined #trac |
| [06:45:19] |
<mad_lord> |
hi |
| [06:45:45] |
<mad_lord> |
i'm newbie in trac, so i have this bug: |
| [06:45:59] |
<mad_lord> |
Repository checkins event provider (ChangesetModule) failed: TracError: Unsupported version control system "svn" |
| [06:46:18] |
<mad_lord> |
this means some error in configuration? |
| [06:46:19] |
<coderanger> |
mad_lord: Run this `python -c "from svn import fs"` |
| [06:46:36] |
<cmlenz> |
cboos: is there a good reason why SubversionAuthorizer is hardcoded in the AnyDiffModule? |
| [06:46:46] |
<mad_lord> |
Traceback (most recent call last): |
| [06:46:47] |
<mad_lord> |
File "<string>", line 1, in ? |
| [06:46:47] |
<mad_lord> |
ImportError: No module named svn |
| [06:46:49] |
<mad_lord> |
hmm |
| [06:46:58] |
<coderanger> |
mad_lord: Bindings aint' installed |
| [06:47:07] |
<mad_lord> |
hmm, nice, thank you coderanger |
| [06:47:13] |
<coderanger> |
@wiki TracSubversion |
| [06:47:13] |
<evil_twin> |
coderanger: http://trac.edgewall.org/wiki/TracSubversion |
| [06:47:20] |
<mad_lord> |
i'll see this now |
| [06:47:21] |
<cboos> |
no I don't think so, if it's like that, it's certainly I didn't find a workaround |
| [06:47:55] |
* |
eblot has quit IRC |
| [06:48:04] |
<cmlenz> |
cboos: we should be able to do something like repos.authz, no? |
| [06:48:32] |
<cboos> |
yes |
| [06:49:03] |
<cboos> |
(I don't remember that part of the code, let me look) |
| [06:50:44] |
<cboos> |
yeah, sure, we should use repos.authz there |
| [06:56:17] |
<tsb> |
Hrm, any easy way to make the wikientries case-insensitive? i.e /wiki/bleh would be the same as /wiki/Bleh ? |
| [06:56:41] |
<tsb> |
all I can find is "ignore_auth_case", but that's not for this. |
| [06:56:56] |
<coderanger> |
tsb: Its doable with a plugin |
| [06:57:25] |
<tsb> |
right, I've looked around at hacks though |
| [06:58:01] |
<coderanger> |
tsb: One second |
| [06:59:45] |
<tsb> |
coderanger: actaully, what I really want is the autowikify plugin to be case-insensitive I guess.. I want the glossary to match both capitalized and normal words in it. Right now I have to dublicate for both |
| [07:01:01] |
* |
Moisei has joined #trac |
| [07:01:08] |
<coderanger> |
tsb: For that you would need alter that plugin |
| [07:01:16] |
<tsb> |
figured so, right. |
| [07:03:25] |
* |
Guardian has quit IRC |
| [07:03:55] |
<coderanger> |
probably just need to make one of the regexs (?i) |
| [07:06:54] |
<Moisei> |
hi all |
| [07:07:23] |
<Moisei> |
i am wondering where trac keeps the wiki pages ? is there some location on the disk or it is a part of the trac.db? |
| [07:08:09] |
* |
xjjk has quit IRC |
| [07:08:18] |
<coderanger> |
Moisei: In the db |
| [07:08:22] |
<coderanger> |
Moisei: The wiki table |
| [07:09:33] |
<groton> |
i just tryed to set up authentication in trac0.10, but i failed. For example, when i click on Login i get: "assert req.remote_user, 'Authentication information not available.'" The followins is what i added to httpd.conf of apache http://www.rafb.net/paste/results/s7LNv790.html |
| [07:10:11] |
* |
milk-it has quit IRC |
| [07:14:43] |
* |
danbeck has joined #trac |
| [07:18:38] |
<coderanger> |
groton: One of those is capitalized, one isnt |
| [07:19:05] |
<groton> |
i think case sensitiveness does not matter there, aint it |
| [07:19:21] |
<coderanger> |
It should, URLs are case sensitive |
| [07:19:36] |
<coderanger> |
Basically everything in computers is (except for path names under win32) |
| [07:19:36] |
<groton> |
really? |
| [07:20:06] |
<groton> |
the hostname part it is never, at least :) |
| [07:20:29] |
<coderanger> |
As evidenced by the fact that I am making a plugin right now to make the wiki case insensitive, yep ;-) |
| [07:20:48] |
<groton> |
<
