| [00:10:21] |
* |
Epcylon has joined #trac |
| [00:44:32] |
* |
MasterC has joined #trac |
| [00:45:15] |
* |
MasterC_ has joined #trac |
| [01:26:52] |
* |
moolight has quit IRC |
| [01:26:55] |
* |
moolight_xPHAD has joined #trac |
| [01:27:05] |
* |
moolight_xPHAD is now known as moolight |
| [01:31:20] |
* |
ralreegorganon has joined #trac |
| [01:36:17] |
* |
cmlenz has joined #trac |
| [01:38:46] |
* |
tuxipuxi has joined #trac |
| [01:38:46] |
* |
idostyle has quit IRC |
| [01:39:47] |
* |
Pyretic has left #trac |
| [01:42:16] |
* |
idostyle has joined #trac |
| [02:05:08] |
* |
MasterC_ has quit IRC |
| [02:15:34] |
* |
Hydra has quit IRC |
| [02:19:51] |
* |
Hydra has joined #trac |
| [03:19:28] |
* |
hpnadig has joined #trac |
| [03:28:55] |
* |
maxb has joined #trac |
| [03:41:56] |
<maxb> |
Is anyone doing anything about the severe flood of idiotic ticket-spam afflicting Trac's Trac? |
| [03:43:16] |
<maxb> |
Ideally, someone needs to delete a *lot* of comments from ticket 1809 |
| [03:43:21] |
<maxb> |
umm, 1890 |
| [03:43:24] |
* |
minsu_ju has joined #trac |
| [03:47:59] |
* |
raidman|Away is now known as raidman|Work |
| [03:57:22] |
* |
hpnadig has quit IRC |
| [03:58:25] |
* |
hpnadig has joined #trac |
| [04:08:58] |
* |
Blackhex has joined #trac |
| [04:09:34] |
* |
raidman|Work is now known as raidman|Away |
| [04:23:05] |
* |
angrymike has quit IRC |
| [04:38:02] |
<Blackhex> |
any trac tags developer here? |
| [04:41:02] |
* |
raidman|Away is now known as raidman|Work |
| [04:41:09] |
* |
prologic has joined #trac |
| [04:41:19] |
<Blackhex> |
prologic: hi |
| [04:41:46] |
<prologic> |
hi |
| [04:41:50] |
<prologic> |
not quite setup yet :) |
| [04:41:55] |
<prologic> |
back later (desktop died) |
| [04:42:18] |
* |
prologic has quit IRC |
| [04:42:54] |
* |
MasterC has quit IRC |
| [04:44:07] |
<Blackhex> |
or someone who knows TracTags code? |
| [04:52:01] |
* |
raidman|Work is now known as raidman|Away |
| [04:54:11] |
* |
dcab has joined #trac |
| [04:55:39] |
* |
moolight has quit IRC |
| [04:55:51] |
* |
moolight has joined #trac |
| [04:55:55] |
<dcab> |
hi all |
| [04:56:06] |
<dcab> |
i have installed trac 0.10dev |
| [04:56:33] |
<dcab> |
but code highliting in {{{ }}} tags is not working |
| [04:56:54] |
<dcab> |
for instance {{{ #!xml <x> </x> }}} |
| [05:08:12] |
* |
flosch has joined #trac |
| [05:10:27] |
* |
prologic has joined #trac |
| [05:10:35] |
* |
prologic has left #trac |
| [05:17:19] |
* |
raidman|Away is now known as raidman|Work |
| [05:19:55] |
* |
dcab_ has joined #trac |
| [05:21:17] |
* |
dcab has quit IRC |
| [05:21:22] |
* |
dcab_ is now known as dcab |
| [05:22:52] |
<dcab> |
I use clearsilver |
| [05:29:43] |
* |
moolight has quit IRC |
| [05:30:01] |
* |
moolight_yEWIW has joined #trac |
| [05:30:05] |
* |
moolight_yEWIW is now known as moolight |
| [05:37:48] |
* |
aspyrine has joined #trac |
| [05:37:52] |
* |
cmlenz has quit IRC |
| [05:37:53] |
<aspyrine> |
hi |
| [05:39:23] |
<aspyrine> |
I have an issue with repository zip download. If some directories are not readable by anonymous, downloading is not possible, even if unreadable directories are not under the downloadable directory |
| [05:40:53] |
* |
cmlenz has joined #trac |
| [05:42:32] |
* |
coderanger_ has joined #trac |
| [05:51:01] |
* |
moolight has quit IRC |
| [05:51:12] |
* |
moolight has joined #trac |
| [05:57:20] |
* |
ComcastOnline has joined #trac |
| [06:02:13] |
* |
dcab has quit IRC |
| [06:04:29] |
* |
dort has quit IRC |
| [06:05:09] |
* |
boorad has quit IRC |
| [06:06:09] |
* |
raidman|Work has left #trac |
| [06:06:13] |
<uber`xxiiv> |
maxb, i dont think you can delete ticket comments yet |
| [06:09:48] |
<maxb> |
There is a plugin to do it. |
| [06:10:32] |
<ComcastOnline> |
Hello all. Anyone here have any experience with a little problem with Trac on Solaris where the Subversion interface can't seem to determine that the path you gave it is actually a repository? |
| [06:11:59] |
* |
hpnadig has quit IRC |
| [06:12:23] |
<ComcastOnline> |
I'm using Trac 0.9.6 with Subversion 1.3.2 on Solaris 10.2, and the problem seems to be around line 190 in svn_fs.py |
| [06:12:33] |
<maxb> |
ComcastOnline: Posting *exact* error might enable someone to help. |
| [06:12:54] |
<coderanger_> |
uber`xxiiv: TicketDelete on trac-hacks |
| [06:13:02] |
<ComcastOnline> |
specifically the line "self.path = repos.svn_repos_find_root_path(path, self.pool())" |
| [06:13:21] |
<ComcastOnline> |
Ah - ok - one sec and I'll get it.. |
| [06:14:24] |
<ComcastOnline> |
The error in the Trac browser window is "[my repos path] does not appear to be a Subversion repository. |
| [06:15:11] |
<uber`xxiiv> |
that deletes the whole ticket though |
| [06:15:17] |
<coderanger_> |
uber`xxiiv: It does both |
| [06:15:20] |
<ComcastOnline> |
That line in the svn_fs.py is generating NULL in process, but if I run it from the python command line, it seems to work |
| [06:15:27] |
<uber`xxiiv> |
ohh LOL |
| [06:15:33] |
<coderanger_> |
uber`xxiiv: Read the page first ;-) |
| [06:15:41] |
<uber`xxiiv> |
DOH ! |
| [06:16:28] |
<uber`xxiiv> |
btw -- i tried importing the sqlite dump into mysql and mysql doesnt like the dump import |
| [06:17:10] |
<aspyrine> |
I have an issue with repository zip download. If some directories are not readable by anonymous, downloading is not possible, even if unreadable directories are not under the downloadable directory |
| [06:22:21] |
<coderanger_> |
aspyrine: Did you file a bug? |
| [06:22:45] |
<aspyrine> |
nope, I looked for one, tried to find a fix |
| [06:29:35] |
* |
mitsuhiko has joined #trac |
| [06:35:59] |
<Blackhex> |
coderanger: hi, I have question about TracTags |
| [06:38:07] |
* |
agile has quit IRC |
| [06:39:56] |
<coderanger_> |
Blackhex: shoot |
| [06:40:22] |
<Blackhex> |
coderanger_: I was studying WikiTags component code but I can't figure out how are tags appended when new wiki page is created. |
| [06:40:39] |
<Blackhex> |
ther is only chache invalidating |
| [06:41:28] |
<coderanger_> |
Hmm, I don't know off hand, I would have to look at the code |
| [06:42:01] |
<Blackhex> |
ok, if you don't have mucht time right now I can wait |
| [06:42:19] |
<coderanger_> |
the site is taking a while to lad |
| [06:42:33] |
<ComcastOnline> |
Ok - lets try this - when I run Trac using tracd, I get the following error when I hit "browse source": |
| [06:42:45] |
<Blackhex> |
:-) |
| [06:43:57] |
<ComcastOnline> |
Exception exceptions.AttributeError: "'SubversionRepository' object has no attribute 'log'" in <bound method SubversionRepository.__del__ of <trac.versioncontrol.svn_fs.SubversionRepository object at 0x8361b6c>> ignored |
| [06:46:56] |
* |
danbeck has joined #trac |
| [06:49:01] |
<exarkun> |
I just installed the ticket deletion plugin, but it isn't doing anything. I submit the form it presents with a ticket number and the same page reloads with nothing changed, except the text field is empty again. |
| [06:53:04] |
<mitsuhiko> |
exarkun: i had the same problem |
| [06:53:23] |
<mitsuhiko> |
i upgraded to a newer version of the trac trunk and it worked |
| [06:55:24] |
<exarkun> |
I'm using trac 0.9.something, not the dev branch |
| [06:59:00] |
<coderanger_> |
exarkun: Full-ticket deletion or change deletion? |
| [06:59:08] |
<exarkun> |
Oh |
| [06:59:18] |
<exarkun> |
I didn't realize there was a difference. |
| [06:59:42] |
<coderanger_> |
Blackhex: http://muness.textdriven.com/trac/browser/tags-plugin/trunk/tractags/web_ui.py#L33 |
| [06:59:48] |
<exarkun> |
I downloaded the thing on this page. http://www.trac-hacks.org/wiki/TicketDeletePlugin |
| [06:59:53] |
<exarkun> |
I want to delete an entire ticket. |
| [06:59:55] |
<exarkun> |
Is that not the right thing? |
| [07:00:11] |
<coderanger_> |
exarkun: There should be an entry under web-admin labeled "Delete" |
| [07:01:08] |
<exarkun> |
ah |
| [07:02:04] |
<exarkun> |
That's it, thanks. |
| [07:02:29] |
<coderanger_> |
What were you trying before OOC? |
| [07:03:19] |
<exarkun> |
"Delete Changes" |
| [07:03:20] |
<Blackhex> |
coderanger_: thx, and then what is this class http://muness.textdriven.com/trac/browser/tags-plugin/trunk/tractags/wiki.py#L35 for? |
| [07:03:40] |
<exarkun> |
But the ticket had nothing that qualified as a change, I suppose |
| [07:04:01] |
<exarkun> |
Might be nice to have some feedback to that effect on the post response page |
| [07:04:12] |
<coderanger_> |
exarkun: Yeah, the whole UI needs massive work |
| [07:04:24] |
<coderanger_> |
exarkun: The ticket delete procedure is just annoying |
| [07:04:45] |
<coderanger_> |
exarkun: When I first wrote it no one thought it would really be needed (in the pre-spam days) |
| [07:08:12] |
<ComcastOnline> |
Anyone have any idea about the Solaris-repos browsing problem? |
| [07:08:14] |
<coderanger_> |
Blackhex: For keeping the link cache in sync |
| [07:09:17] |
* |
boorad has joined #trac |
| [07:09:19] |
<Blackhex> |
coderanger_: so do I have to implement ITaggingSystemProvider interface when I don't need cache updating? |
| [07:10:18] |
<coderanger_> |
Having never done so, I couldn't say |
| [07:10:39] |
* |
agile has joined #trac |
| [07:14:43] |
* |
uber`xxiiv has quit IRC |
| [07:17:44] |
* |
milk-it has joined #trac |
| [07:22:00] |
* |
The_Tick has quit IRC |
| [07:22:37] |
<Blackhex> |
ok, thx |
| [07:58:09] |
* |
phoenixz has joined #trac |
| [07:58:10] |
* |
Semhirage has quit IRC |
| [07:58:41] |
* |
matt-tvm has joined #trac |
| [08:08:02] |
<matt-tvm> |
Hi all; I have a quick question regarding the "Browse Source" link in the Trac wiki that I'm sure there's a savvy way to solve, but I'm at a bit of a loss. Should I ask here, or on the mailing list? (What's considered good etiquette/process?) |
| [08:09:44] |
<exarkun> |
ask here, if no one knows, ask on the mailing list. or ask on the mailing list, if no one knows, ask here. ;) |
| [08:10:02] |
<exarkun> |
(don't ask if you can ask though ;) |
| [08:10:19] |
<matt-tvm> |
Right. :) |
| [08:11:18] |
* |
pygi has joined #trac |
| [08:11:20] |
<pygi> |
http://libburn.pykix.org/browser |
| [08:11:29] |
<pygi> |
why it looks for revision 3 when there is only rev1? |
| [08:11:58] |
<matt-tvm> |
We have a project that has an HTTPS (developer) and HTTP (public) interface. Access is controlled via authz. One subdirectory is currently available for public consumption; we'd like the "Browse Source" link in Trac to point to that subdirectory when accessed via a non-SSL link, but to the top of the repos when accessed via SSL... |
| [08:12:10] |
<coderanger_> |
pygi: did you mess with your repo? |
| [08:12:18] |
<matt-tvm> |
When I ask it, actually, it sounds like I'm looking for magic... now that I put it into words. |
| [08:12:22] |
<pygi> |
coderanger, yes ^_^ |
| [08:12:30] |
<coderanger_> |
matt-tvm: You cannot do that with Trac easily |
| [08:12:41] |
<coderanger_> |
pygi: trac-admin /path/to/env resync |
| [08:13:43] |
<coderanger_> |
matt-tvm: With authz you can control access, but you would still have it rooted at the same place |
| [08:13:49] |
<pygi> |
coderanger, thanks ^_^ |
| [08:14:19] |
* |
chandlerc has quit IRC |
| [08:14:19] |
<coderanger_> |
matt-tvm: You could have two different tracs, but then it gets complex |
| [08:14:26] |
* |
aspyrine has left #trac |
| [08:14:28] |
<matt-tvm> |
Aye. |
| [08:17:04] |
<matt-tvm> |
Many thanks; I'll do some brainstorming and come up with something. Cheers. |
| [08:18:00] |
* |
matt-tvm has quit IRC |
| [08:22:13] |
* |
klasstek has joined #trac |
| [08:26:34] |
* |
bowen has joined #trac |
| [08:28:00] |
* |
bowen has quit IRC |
| [08:28:25] |
* |
ged has quit IRC |
| [08:28:28] |
* |
madduck has left #trac |
| [08:32:00] |
* |
Riccc has joined #trac |
| [08:33:33] |
* |
chjunior has joined #trac |
| [08:48:55] |
* |
Plouf has quit IRC |
| [08:56:01] |
* |
chandlerc has joined #trac |
| [09:00:32] |
<pacopablo> |
morning all |
| [09:01:02] |
<coderanger_> |
Alo |
| [09:03:33] |
* |
The_Tick has joined #trac |
| [09:04:30] |
* |
chandler has joined #trac |
| [09:05:04] |
<The_Tick> |
hey cmlenz |
| [09:05:06] |
<The_Tick> |
you around? |
| [09:05:11] |
<cmlenz> |
yo |
| [09:05:16] |
<The_Tick> |
hey, chandler runs a pastebot |
| [09:05:18] |
<The_Tick> |
called lisppaste |
| [09:05:26] |
<The_Tick> |
he can put it in here if you'd like |
| [09:05:35] |
<The_Tick> |
I've asked him to if you guys are alright with it |
| [09:05:43] |
<coderanger_> |
The_Tick: We already have a bot |
| [09:05:48] |
<cmlenz> |
sure, that'd be lovely |
| [09:05:53] |
<The_Tick> |
oh, which one? |
| [09:05:59] |
<coderanger_> |
The_Tick: evil_twin |
| [09:06:02] |
<The_Tick> |
evil_twin: url |
| [09:06:02] |
<evil_twin> |
The_Tick: Error: "url" is not a valid command. |
| [09:06:03] |
<cmlenz> |
coderanger_: but not a pastebot |
| [09:06:15] |
<chandler> |
you folks want python colorization by default? |
| [09:06:18] |
<The_Tick> |
chandler: they could use it I believe :) |
| [09:06:28] |
<cmlenz> |
yeah |
| [09:06:54] |
<coderanger_> |
cmlenz: I would think supybot has a plugin |
| [09:07:20] |
<cmlenz> |
dunno |
| [09:07:33] |
<The_Tick> |
we use lisppaste effectively in every channel I'm in |
| [09:08:10] |
<The_Tick> |
I could show you guys how it works in #growl if you like |
| [09:08:13] |
<chandler> |
cmlenz: how about I add the bot, and if you want it taken out just /msg me |
| [09:08:27] |
<chandler> |
is that OK? |
| [09:08:30] |
<cmlenz> |
chandler: sure :-) |
| [09:08:48] |
* |
lisppaste5 has joined #trac |
| [09:08:51] |
<chandler> |
lisppaste5: url? |
| [09:08:52] |
<lisppaste5> |
To use the lisppaste bot, visit http://paste.lisp.org/new/trac and enter your paste. |
| [09:09:44] |
<lisppaste5> |
cmlenz pasted "Test" at http://paste.lisp.org/display/23789 |
| [09:09:54] |
<cmlenz> |
ah, nice |
| [09:09:55] |
<coderanger_> |
Odd, it seems supybot doesn't have a plugin for this |
| [09:10:15] |
<pacopablo> |
damn supybot! |
| [09:10:30] |
<pacopablo> |
looks like you need to start coding, coderanger_ :) |
| [09:10:42] |
* |
pacopablo hates it when boxes run out of disk space |
| [09:11:16] |
<cmlenz> |
thanks chandler! |
| [09:11:21] |
<coderanger_> |
pacopablo: Something like that |
| [09:12:07] |
* |
rputzler has joined #trac |
| [09:12:22] |
<chandler> |
cmlenz: no problem |
| [09:12:26] |
* |
chandler has left #trac |
| [09:14:29] |
* |
xjjk has joined #trac |
| [09:14:54] |
<De_mon> |
err... what permission is required to create a custom query? |
| [09:15:43] |
<De_mon> |
ticket_admin seems to be the only thing that works |
| [09:21:14] |
* |
tharvey has joined #trac |
| [09:21:40] |
<tharvey> |
has anyone done any work on mods to the wiki button-bar? |
| [09:23:00] |
<coderanger_> |
tharvey: Mods? |
| [09:23:51] |
<tharvey> |
added any features to the buttonbar? |
| [09:24:33] |
<coderanger_> |
You cant |
| [09:24:58] |
<tharvey> |
why not? |
| [09:25:22] |
<coderanger_> |
Its created using javascript, and that javascript is hardcoded |
| [09:25:32] |
<tharvey> |
I see |
| [09:25:38] |
<tharvey> |
thanks |
| [09:25:44] |
<coderanger_> |
I mean if you start hacking the templates you could |
| [09:25:53] |
<coderanger_> |
but not in any kind of modular way |
| [09:25:56] |
<tharvey> |
right |
| [09:26:00] |
<De_mon> |
ah hah, theres the answer. |
| [09:28:10] |
* |
imajes has quit IRC |
| [09:28:31] |
<tharvey> |
how do you suggest working in a fashion where a bug is fixed by a developer, then assigned to a tester to be verified. I did not see a way to add/modify the resolutions (fixed,invalid,wontfix,duplicate,worksforme) to add a 'resolved' and 'verified'. Do you suggest simply moving a bug to 'fixed' and assigning it to a tester and they will reopen if necessary? |
| [09:28:50] |
<coderanger_> |
tharvey: Look at the workflow brach |
| [09:29:34] |
* |
imajes has joined #trac |
| [09:30:52] |
<tharvey> |
thanks... I'm reading http://trac.edgewall.org/wiki/WorkFlow now |
| [09:34:37] |
<tharvey> |
how does the 'sandbox' tree in svn differ from the 'branches' tree? why wouldn't workflow be a branch? |
| [09:35:23] |
<coderanger_> |
Its about the same thing |
| [09:35:28] |
<coderanger_> |
just different words |
| [09:36:27] |
<tharvey> |
so will trac someday encorporate the features of workflow or is that an ongoing discussion? |
| [09:37:17] |
<coderanger_> |
Yes, it will be merged after 0.10 is out |
| [09:37:35] |
<coderanger_> |
along with the security and time zone sandboxes for starters |
| [09:38:11] |
<pacopablo> |
and markup and then sometime SQLAlchemy |
| [09:39:09] |
<tharvey> |
nice... I like the features. I'm sure you hate this question, but what kind of rough timeframe (months?) are we looking at for 0.10? according to the roadmap its 94% complete with 17 active tickets |
| [09:39:23] |
<coderanger_> |
Real Soon Now (TM) |
| [09:39:29] |
<tharvey> |
heh |
| [09:40:05] |
<pacopablo> |
here is my completely unreliable guess: |
| [09:40:16] |
<pacopablo> |
beta next week, full release and of August |
| [09:40:24] |
<pacopablo> |
s/and/end |
| [09:40:39] |
<tharvey> |
cool... thanks |
| [09:41:10] |
<tharvey> |
do milestones fall off the roadmap after they are met? |
| [09:41:20] |
<pacopablo> |
they become hidden |
| [09:41:28] |
<tharvey> |
ah, I see the checkbox |
| [09:41:29] |
<pacopablo> |
there should be a link to show old milestones |
| [09:42:23] |
<tharvey> |
how is a milestone met? doesn't appear to be when it reaches 100% - does it go off of due date? |
| [09:42:58] |
<pacopablo> |
yes |
| [09:44:40] |
<coderanger_> |
No, you mark it as finshed |
| [09:44:56] |
<coderanger_> |
You can have an overdue milestone |
| [09:47:03] |
* |
flosch has quit IRC |
| [09:49:16] |
<tharvey> |
ah, I see where you can edit the milestone and mark it as completed... thanks |
| [09:53:24] |
* |
flosch has joined #trac |
| [09:53:31] |
<tharvey> |
anyone know if you can move your project from trac-trunc to trac-sandbox-workflow? |
| [09:55:01] |
<tharvey> |
I see the 'Upgrade Notes' section in http://trac.edgewall.org/wiki/NewWorkflow but I'm not sure if that page is up to date |
| [09:56:27] |
<pacopablo> |
tharvey: depends on how new of a trunk |
| [09:56:55] |
<pacopablo> |
if it's after the last merge to workflow, then you might have problems, especially if a db upgrade was involved on the trunk side |
| [09:57:18] |
<pacopablo> |
if it's any version before the last merge to workflow, you shouldn't have any problem |
| [09:57:38] |
<tharvey> |
ok, and how would I see when merges from sandbox/workflow to trunk took place? (by looking at the svn logs?) |
| [09:57:44] |
* |
dcab has joined #trac |
| [09:57:59] |
<pacopablo> |
alect: puts it in his commit message, so yes |
| [09:58:46] |
<pacopablo> |
http://trac.edgewall.org/changeset/3378 |
| [10:00:13] |
<dcab> |
how to uninstall SilverCity and switch to Enscript? |
| [10:00:49] |
<coderanger_> |
dcab: Depends on your system |
| [10:00:57] |
<pacopablo> |
dcab: depends entirely on your distro and how you installed it in the first place |
| [10:01:05] |
<dcab> |
windows + apache + trac 0.10dev |
| [10:01:18] |
<pacopablo> |
Add/Remove Porgrams |
| [10:01:28] |
<pacopablo> |
er, you get teh idea |
| [10:01:57] |
<dcab> |
pacopable: but after uninstall I get errors that SilverCity is not installed |
| [10:02:12] |
<pacopablo> |
cus it's not |
| [10:02:24] |
<dcab> |
yes |
| [10:02:26] |
<pacopablo> |
though if it's not, it should simply fall back to enscript |
| [10:03:11] |
<dcab> |
but it still try to use SilverCity |
| [10:03:22] |
<coderanger_> |
I dont know how well enscript works on windows |
| [10:03:35] |
<pacopablo> |
I believe that trac will always try to use silvercity |
| [10:03:43] |
<pacopablo> |
and then if it can't, fall back to enscript |
| [10:03:47] |
<dcab> |
it work fine with my previos install |
| [10:05:40] |
* |
otavio has left #trac |
| [10:05:48] |
<dcab> |
i have 0.95 without SilverCity (with enscript) and it colorize my {{{#!xml }}} sections. But 0.10dev with SilverCity doesn't %( |
| [10:06:27] |
* |
lisppaste5 has quit IRC |
| [10:08:41] |
* |
lisppaste5 has joined #trac |
| [10:19:06] |
* |
hpnadig has joined #trac |
| [10:21:37] |
* |
spity has joined #trac |
| [10:21:39] |
<spity> |
hi |
| [10:22:12] |
<coderanger_> |
hello |
| [10:22:18] |
<spity> |
is it possible to migrate from sqlite backend to postgresql backend? |
| [10:22:59] |
<pacopablo> |
yes, though not easily |
| [10:23:13] |
<pacopablo> |
there is no script or automated process for doing so |
| [10:23:35] |
<pacopablo> |
one would need to dump the sqlite db and then try to impor it into postgresql |
| [10:23:56] |
<pacopablo> |
however, there are slight incompatabilities between the SQL that sqlite dumps and what postgresql accepts. |
| [10:24:41] |
<spity> |
aha, i thought so |
| [10:24:51] |
<spity> |
i support SQL types and quoting will be problem |
| [10:25:19] |
<spity> |
does trac use some advanced functionality of sqlite? (if sqlite does support some :) |
| [10:26:18] |
<coderanger_> |
No |
| [10:26:32] |
<spity> |
ok |
| [10:26:35] |
<spity> |
thanks |
| [10:27:20] |
<coderanger_> |
If you know a bit of python it may be easier to do the move in code |
| [10:27:26] |
<coderanger_> |
instead of just dumping things |
| [10:27:29] |
<pygi> |
weee, python ^_^ |
| [10:29:14] |
* |
ralreegorganon has quit IRC |
| [10:30:09] |
* |
ralreegorganon has joined #trac |
| [10:32:04] |
* |
ralreegorganon has joined #trac |
| [10:33:12] |
* |
ralreegorganon has joined #trac |
| [10:34:28] |
* |
ralreegorganon has joined #trac |
| [10:40:22] |
* |
tharvey has quit IRC |
| [10:48:21] |
* |
Semhirage has joined #trac |
| [10:51:39] |
* |
milk-it has quit IRC |
| [10:53:06] |
* |
hpnadig is now known as hpnadig_dinner |
| [10:53:38] |
<spity> |
coderanger_: good idea! |
| [10:53:40] |
<spity> |
thanks |
| [10:57:15] |
* |
pygi has quit IRC |
| [11:08:26] |
<ComcastOnline> |
Can someone tell me why, in 0.9.6, svn_fs.py is broken on Solaris unless you add "self.log = log" and then log a message? |
| [11:09:32] |
<ComcastOnline> |
Otherwise, you get errors about log not being an attribute of SubversionRepository and that the path you've pointed to isn't a valid repository? |
| [11:09:56] |
<coderanger_> |
Sounds like a bug in the bindings maybe? |
| [11:10:42] |
<ComcastOnline> |
Adding "self.log = log" fixes the 'SubversionRepository' object has no attribute 'log' problem... |
| [11:11:20] |
* |
MasterC has joined #trac |
| [11:11:45] |
<ComcastOnline> |
... but until you add a line that uses the logger object, the "browse source" button gives you the error that your repos isn't valid.. |
| [11:13:40] |
<ComcastOnline> |
coderanger - what bindings? This all happens at line 177 in the svn_fs.py file |
| [11:14:06] |
<coderanger_> |
There are some weird race conditions in the svn SWIG code |
| [11:14:28] |
<ComcastOnline> |
Hmm... and here I was thinking it couldn't possibly be a race condition... |
| [11:15:34] |
<ComcastOnline> |
So just adding a log message give the svn code time to find your repository as valid? That's nuts... |
| [11:16:33] |
<coderanger_> |
Possibly |
| [11:16:41] |
<coderanger_> |
What version of subversion are you using? |
| [11:17:01] |
<ComcastOnline> |
I'm using the Blastwave packages for svn 1.3.2 and Trac 0.9.6 on solaris 10.2 |
| [11:17:34] |
<coderanger_> |
I would say to try 1.2, its somewhat more stable |
| [11:17:43] |
* |
cmlenz has quit IRC |
| [11:17:58] |
<ComcastOnline> |
The bindings are using the right version of subversion - I checkec the SVN_VER major and minor |
| [11:19:19] |
<ComcastOnline> |
Should probably document this somewhere that it's probably the timings - I'm counting at lest three tickets about this problem on trac's site... |
| [11:19:24] |
* |
PingYeh is now known as ping_away |
| [11:19:35] |
<ComcastOnline> |
...and none of them finish with a definitive fix or answer |
| [11:20:00] |
<ComcastOnline> |
One of the tickes even suggests recompiling your iconv libs.... |
| [11:30:28] |
* |
hpnadig_dinner is now known as hpnadig |
| [11:30:57] |
* |
dcab has quit IRC |
| [11:31:16] |
* |
hpnadig has joined #trac |
| [11:52:46] |
* |
cmlenz has joined #trac |
| [12:10:35] |
* |
MasterC has quit IRC |
| [12:12:47] |
<De_mon> |
max_preview_size = 26214400 |
| [12:13:21] |
<De_mon> |
but I still get the 'HTML preview not available.' for a 262k update.patch attachment |
| [12:15:58] |
* |
spity has left #trac |
| [12:17:06] |
<De_mon> |
giving it a .txt extension does give it a html preview? huh.. |
| [12:29:46] |
* |
The_Tick has left #trac |
| [12:33:16] |
* |
cboos has joined #trac |
| [12:33:57] |
<cboos> |
hi |
| [12:34:34] |
<cboos> |
cmlenz, are you there? |
| [12:34:49] |
<cmlenz> |
yeah |
| [12:34:52] |
<cmlenz> |
hi |
| [12:35:09] |
<cboos> |
I'd like to hear your opinion on #2580, |
| [12:35:41] |
<cboos> |
the fact that casting to unicode an Element instance yields a Markup instance, |
| [12:35:54] |
<cboos> |
which then doesn't get HTML escaped... |
| [12:36:18] |
<cboos> |
Looking at the Markup code, it seems that a cast to unicode yields a proper unicode object there. |
| [12:36:53] |
<cboos> |
So for me, it looks like we should do that also in trac.util.markup.Fragment.__unicode__ |
| [12:37:16] |
<cboos> |
... but as it's mainly your domain, I'm asking first ;) |
| [12:38:29] |
<cmlenz> |
okay, need to catch up on what's going on first :-P |
| [12:39:52] |
<cboos> |
easy, sometimes we produce HTML by returning Markup objects, sometimes by returning Element objects. No difference in the result, except when we try to prepare that for RSS, i.e. when we want to force HTML escaping of this HTML... |
| [12:40:54] |
<cboos> |
Then, doing to_unicode() on a Markup objects gives a unicode object, which will get HTML escaped, but doing this to_unicode() on a Element object will yield... a Markup object, which won't, hence the error the user saw in #2580. |
| [12:41:25] |
<cmlenz> |
what a mess :-P |
| [12:42:10] |
<cboos> |
Like I said above, I have the feeling that this wouldn't happen with "Markup", as the __unicode__ method there yields a unicode object, not a Markup object... |
| [12:42:36] |
<cboos> |
s/yields/returns/ to be precise |
| [12:43:28] |
<cboos> |
so I was wondering if there are places where we rely on casting an Element to unicode returning a Markup object... |
| [12:43:38] |
<cmlenz> |
yeah, same here |
| [12:44:10] |
<cmlenz> |
I'm pretty sure we do, actually |
| [12:45:12] |
<cmlenz> |
guess the only way is to try and test a lot :-/ |
| [12:45:42] |
<cmlenz> |
place that could rely on that behavior are most likely the wiki formatter and HDFWrapper, I think |
| [12:46:06] |
<cmlenz> |
I had to patch the wiki formatter somewhat on the markup sandbox branch |
| [12:46:13] |
<cboos> |
but I'm looking at Markup code, what I see is: |
| [12:47:30] |
<cboos> |
Fragment.__unicode__ returns unicode(Stream(self._generate)) (http://markup.edgewall.org/browser/trunk/markup/builder.py#L44) |
| [12:48:48] |
<cboos> |
and the Stream.__unicode__ is itself returning self.render(encoding=None), which itself returns u''.join(...) (http://markup.edgewall.org/browser/trunk/markup/core.py#L92) |
| [12:49:28] |
<cboos> |
... so it looks like it returns a unicode object, not a Markup object. |
| [12:49:45] |
<cmlenz> |
right |
| [12:50:47] |
<cboos> |
ok, so the right direction seems to be to have Fragment returning the unicode object, but now, IIUC, you said that this would introduce some issues (excuse me if I'm slow ;) ) |
| [12:51:10] |
<cmlenz> |
I'm not sure, really |
| [12:51:17] |
<cmlenz> |
I agree that would be the right thing to do |
| [12:51:37] |
<cmlenz> |
but I'd make sure to test some of the fragment-using wiki macros, the timeline, search, etc |
| [12:51:51] |
<cboos> |
Ok, so I'll experiment a bit with this approach and let you know |
| [12:52:00] |
<cmlenz> |
actually, the wiki formatter should be no problem IIUC |
| [12:52:11] |
<cmlenz> |
it wraps up anything in Markup() when it's done |
| [12:52:38] |
<cmlenz> |
btw, what do you think about making the wiki formatter generate Markup events in 0.11? |
| [12:52:49] |
<cboos> |
ha. |
| [12:52:53] |
<cmlenz> |
i.e. streaming out events instead of generating one big string |
| [12:53:14] |
<cboos> |
Yes, I think this makes a lot of sense. |
| [12:53:36] |
<cboos> |
There's a lot to be gained in having some structured output |
| [12:54:05] |
<cboos> |
i.e. see all the benefits of reusing the "link" elements... |
| [12:54:09] |
<cmlenz> |
right |
| [12:55:07] |
<cboos> |
the real question for me is should we go one step further and produce some kind of wiki document object model, from which we could generate HTML (by the way of the Elements), or do we consider the stream of Elements to be _the_ document model? |
| [12:56:0 |
