| [00:05:12] |
* |
agile has quit IRC |
| [00:05:37] |
<pacopablo> |
evening |
| [00:28:10] |
* |
hpnadig has quit IRC |
| [00:39:06] |
* |
agile has joined #trac |
| [00:48:48] |
* |
pombreda has joined #trac |
| [00:53:50] |
* |
Guardian has joined #trac |
| [00:53:51] |
* |
qk has joined #trac |
| [00:55:12] |
* |
pombred1 has joined #trac |
| [01:00:45] |
* |
divya has quit IRC |
| [01:00:52] |
* |
dh has joined #trac |
| [01:01:10] |
<dh> |
does anyone know what AttributeError: 'trac.web.auth.LoginModule' object has no attribute 'ignore_case' means |
| [01:01:34] |
* |
m_g has joined #trac |
| [01:02:16] |
* |
pombreda has quit IRC |
| [01:05:11] |
<webar7> |
coderanger, tnx |
| [01:31:36] |
<dh> |
\join #svn |
| [01:58:01] |
* |
qk has quit IRC |
| [02:03:25] |
* |
rsalveti1 has quit IRC |
| [02:05:18] |
* |
rsalveti has joined #trac |
| [02:26:30] |
* |
hpnadig has joined #trac |
| [02:56:36] |
* |
pombreda has joined #trac |
| [02:59:04] |
* |
pombred1 has quit IRC |
| [03:15:52] |
* |
pygi has joined #trac |
| [03:20:20] |
* |
mario_ has joined #trac |
| [03:20:50] |
* |
mario_ has quit IRC |
| [03:21:37] |
* |
pygi has quit IRC |
| [03:34:19] |
* |
hpnadig has quit IRC |
| [03:42:40] |
* |
Narada has left #trac |
| [03:51:02] |
* |
pombreda has quit IRC |
| [04:10:37] |
* |
avar has quit IRC |
| [04:11:16] |
* |
avar has joined #trac |
| [04:13:51] |
* |
tuxipuxi_ has joined #trac |
| [04:21:20] |
* |
m_g has quit IRC |
| [04:28:56] |
* |
mitsuhiko has quit IRC |
| [04:29:31] |
* |
mitsuhiko has joined #trac |
| [04:30:42] |
<dh> |
does anyone know what AttributeError: 'trac.web.auth.LoginModule' object has no attribute 'ignore_case' means |
| [04:40:36] |
* |
rene_est has joined #trac |
| [04:40:41] |
<rene_est> |
'module' object has no attribute 'AF_UNIX' |
| [04:40:55] |
<rene_est> |
i get this error everywhere on trac |
| [04:41:25] |
<rene_est> |
i did something with logging |
| [04:42:12] |
<rene_est> |
how can i change logging type back to "debug" or something what it was |
| [04:42:18] |
* |
tuxipuxi_ has quit IRC |
| [04:43:07] |
* |
m_g has joined #trac |
| [04:55:10] |
<rene_est> |
it can be changed in trac.ini |
| [04:55:13] |
* |
rene_est has left #trac |
| [05:03:33] |
* |
bryanl has quit IRC |
| [05:11:10] |
* |
bryanl has joined #trac |
| [05:25:16] |
* |
bryanl has quit IRC |
| [05:31:44] |
* |
agile has quit IRC |
| [05:58:36] |
* |
bryanl has joined #trac |
| [06:05:12] |
* |
agile has joined #trac |
| [06:12:00] |
* |
pisi has joined #trac |
| [06:13:18] |
<pisi> |
Hi. Is this normal: http://martin.paljak.pri.ee/trac.png |
| [06:13:35] |
<pisi> |
or is there an existing bug against trunk for this ? |
| [06:14:14] |
<coderanger> |
Is this on trunk? |
| [06:14:44] |
<pisi> |
yes. It has been for a while. |
| [06:16:15] |
<coderanger> |
Probably a pygments issue |
| [06:17:32] |
<pisi> |
is there something i could try to get it fixed ? |
| [06:17:45] |
<pisi> |
uff.. installed Pygments (what was not installed before) and now the display goes CRAZY. |
| [06:17:59] |
<coderanger> |
I think there are some tickets open about it |
| [06:19:47] |
<pisi> |
http://trac.edgewall.org/ticket/4246 I should probly apply the patches. |
| [06:20:41] |
<coderanger> |
Those are already merged in |
| [06:24:31] |
<stepz> |
speaking of patches, I don't want to seem pushy, but does anyone know what's the status with http://trac.edgewall.org/ticket/4245 |
| [06:25:30] |
<pisi> |
http://trac.edgewall.org/ticket/4337 mentions the problem. Genshi 0.3.6 fixed the problem. |
| [06:30:00] |
<coderanger> |
stepz: cmlenz is gone for a while, so he will likely not look at it until he is back |
| [06:36:16] |
* |
d0rt has joined #trac |
| [06:37:30] |
<stepz> |
ok, thank you |
| [06:39:10] |
* |
Miek has quit IRC |
| [06:40:52] |
* |
xenxox has joined #trac |
| [06:46:45] |
* |
danbeck has joined #trac |
| [06:55:37] |
* |
tpollari has joined #trac |
| [06:57:27] |
* |
jcrawford has joined #trac |
| [06:57:58] |
<jcrawford> |
the company i work for is looking to use trac however our code is proprietary, is there a way to not allow just anyone to view the trac install? |
| [06:59:31] |
* |
retracile has joined #trac |
| [07:04:35] |
<dh> |
does anyone know why i would get AttributeError: 'trac.web.auth.LoginModule' object has no attribute 'ignore_case' following installing svn trac |
| [07:06:29] |
* |
blinx has quit IRC |
| [07:07:56] |
* |
bergheim has joined #trac |
| [07:09:40] |
<tpollari> |
jcrawford: check the website -- http://trac.edgewall.org/wiki/TracPermissions and http://trac.edgewall.org/wiki/FineGrainedPermissions |
| [07:10:40] |
<tpollari> |
jcrawford: if you're using trac with apache, you can use basic authentication |
| [07:14:58] |
* |
Idle0ne- has joined #trac |
| [07:15:08] |
* |
romanoffi has joined #trac |
| [07:17:00] |
* |
tsb has quit IRC |
| [07:17:40] |
* |
blinx has joined #trac |
| [07:17:49] |
* |
klasstek has joined #trac |
| [07:20:52] |
<coderanger> |
dh: Are you trying to use any plugins? |
| [07:23:27] |
* |
jcrawford has quit IRC |
| [07:23:43] |
<dh> |
nope |
| [07:24:03] |
<dh> |
the installation was done with a prewritten script, http://wiki.dreamhost.com/index.php/DreamTracInstall |
| [07:24:07] |
<coderanger> |
Are you insalling 0.11dev over an earlier version? |
| [07:25:19] |
<dh> |
it actually does use 2 by the looks of it, webadmin and accountmanager |
| [07:25:44] |
<dh> |
its 0.9.6 |
| [07:26:10] |
* |
Miek has joined #trac |
| [07:27:05] |
<dh> |
http://dev.arandomurl.com/ is the url |
| [07:28:02] |
* |
Miek has quit IRC |
| [07:28:18] |
* |
Miek has joined #trac |
| [07:28:25] |
<coderanger> |
Ahh, thats not svn Trac |
| [07:28:46] |
<coderanger> |
You are using the wrong version of accountmanager |
| [07:28:58] |
* |
ktne has joined #trac |
| [07:28:58] |
<coderanger> |
that version is for 0.10.x |
| [07:29:02] |
<coderanger> |
which is the current stable |
| [07:29:48] |
<dh> |
http://trac-hacks.org/svn/accountmanagerplugin/0.9 ? |
| [07:32:00] |
<coderanger> |
Is there a reason you are not using 0.10? |
| [07:33:35] |
* |
bergheim has quit IRC |
| [07:33:53] |
* |
bergheim has joined #trac |
| [07:34:46] |
<dh> |
its a prewritten script that had worked for dreamhost customers, ive gone throught the nightmare of installing trac manually, wouldnt want to try on a shared server, so hoped this would be simpler |
| [07:35:09] |
<The_Tick> |
dreamhost trac is easy :D |
| [07:35:31] |
<dh> |
the tick, any ideas how to solve http://dev.arandomurl.com/ then? |
| [07:35:52] |
<The_Tick> |
wtf, weird |
| [07:35:57] |
<coderanger> |
dh: One sec |
| [07:36:04] |
<coderanger> |
dh: It was a bad backport |
| [07:36:07] |
<The_Tick> |
http://natmaster.com/articles/installing_trac.php is mostly what I did |
| [07:36:28] |
<The_Tick> |
I've almost modified the dh wiki to reflect his howto |
| [07:36:43] |
* |
pombreda has joined #trac |
| [07:38:19] |
<coderanger> |
dh: New version is in flight |
| [07:38:40] |
<dh> |
ive tried disabling the account manager in the trac.ini (I set acct_mgr.* = disabled and saved) but still the same error |
| [07:38:53] |
<coderanger> |
dh: Reup from svn |
| [07:39:19] |
<dh> |
reup? |
| [07:39:39] |
<coderanger> |
redownload and install the plugin |
| [07:41:44] |
* |
matt_good has joined #trac |
| [07:41:56] |
<coderanger> |
speak of the devil :P |
| [07:42:06] |
<dh> |
reinstall account manager from svn trunk? (0.9.6 trac install) |
| [07:42:15] |
<coderanger> |
dh: No, from the 0.9 folder |
| [07:43:48] |
<dh> |
cool, ill try that now, cheers |
| [07:48:35] |
* |
pombreda has quit IRC |
| [07:49:47] |
* |
pombreda has joined #trac |
| [07:49:54] |
<dh> |
same error http://dev.arandomurl.com/ |
| [07:50:22] |
<coderanger> |
That line is gone, so somehow it is using the old code still |
| [07:50:34] |
<coderanger> |
dif you kill the FCGI daemon process? |
| [07:51:17] |
<dh> |
nup, that was my next question, if i need to restart it or what |
| [07:51:26] |
<coderanger> |
yeah |
| [07:52:43] |
<dh> |
ive always used cgi, how do you restart fgci ? |
| [07:53:10] |
<coderanger> |
killall trac.fcgi |
| [07:54:02] |
<dh> |
wow, thanks, that was great |
| [07:55:00] |
<coderanger> |
yay for one line fixes |
| [07:55:41] |
<coderanger> |
I would highly recommend upgrading to 0.10 at some point though |
| [07:55:51] |
<coderanger> |
0.9.6 is pretty old |
| [07:56:23] |
<retracile> |
(and when he says 0.10, he means 0.10.3 ;) ) |
| [07:56:45] |
<coderanger> |
yes, the 0.10 branch |
| [07:58:04] |
<retracile> |
(0.10.1 and 0.10.2 had some... issues.) |
| [07:58:35] |
* |
francois-simond has joined #trac |
| [07:59:40] |
<dh> |
I always had problems with installing 0.10, partially because the server I installed it on was suse9.2, which is pretty old itself |
| [08:00:30] |
<dh> |
but I probably would be able to manually install trac on a shared server, so I looked for tutorials / etc |
| [08:00:39] |
<dh> |
*wouldnt |
| [08:01:43] |
<coderanger> |
http://e.cactuswax.net/blog/articles/2006/07/installing-trac-on-dreamhost.html |
| [08:01:50] |
<coderanger> |
I would highly recommend that guide |
| [08:02:39] |
* |
pombreda has quit IRC |
| [08:03:21] |
<dh> |
thanks |
| [08:03:30] |
* |
pombreda has joined #trac |
| [08:04:09] |
* |
Tetraboy has quit IRC |
| [08:13:13] |
* |
Tetraboy has joined #trac |
| [08:15:45] |
* |
gezick has joined #trac |
| [08:19:32] |
* |
gezick has quit IRC |
| [08:19:51] |
<boorad> |
pacopablo: you around? |
| [08:19:52] |
* |
Miek has quit IRC |
| [08:20:14] |
<boorad> |
coderanger: or you - I'm puzzled about a SQL call being made often |
| [08:20:22] |
<boorad> |
INSERT INTO system (name,value) VALUES ('repository_dir','svn:0e526d9c-4fd9-0310-bfa2-c33b56aefcee:/var/svn/dwt') |
| [08:25:08] |
* |
tpollari has quit IRC |
| [08:25:40] |
* |
tpollari has joined #trac |
| [08:27:09] |
* |
Miek has joined #trac |
| [08:28:55] |
* |
danbeck has quit IRC |
| [08:30:19] |
* |
pombreda has quit IRC |
| [08:30:38] |
* |
converter has joined #trac |
| [08:32:27] |
* |
pombreda has joined #trac |
| [08:35:43] |
* |
romanoffi has left #trac |
| [08:37:02] |
<dh> |
can you inset a password into an svn co url, like svn co svn+ssh://user:password@host.com |
| [08:37:31] |
<dh> |
*insert |
| [08:37:44] |
<avar> |
dh: you want to use ssh keys |
| [08:37:55] |
<avar> |
dh: man 1 ssh-keygen |
| [08:38:07] |
<dh> |
cheers |
| [08:38:31] |
* |
webar7 has quit IRC |
| [08:43:37] |
<coderanger> |
boorad: type:uuid:folder |
| [08:43:57] |
<coderanger> |
its used to check if a full resync is needed |
| [08:43:58] |
<boorad> |
coderanger: ok, fine. but this kept the PG connection idle |
| [08:44:11] |
<coderanger> |
boorad: Yeah, not sure why it would continually insert |
| [08:44:18] |
<boorad> |
so would it be a permanent row in the table, then? |
| [08:44:24] |
<boorad> |
b/c in this one env, it's not there. |
| [08:44:29] |
<coderanger> |
it should be |
| [08:44:49] |
<boorad> |
despite being attempted to be inserted a ton from 10:30 to 11:00 am today |
| [08:46:18] |
<boorad> |
coderanger: pg trac db's should be UTF-8, no? |
| [08:46:25] |
<coderanger> |
should be |
| [08:46:34] |
* |
Idle0ne- is now known as jcrawford |
| [08:47:50] |
<boorad> |
coderanger: I was able to insert that row in psql command-line util |
| [08:48:12] |
<boorad> |
but we don't need semicolons to end the SQL stmt in py code, right? |
| [08:48:19] |
<coderanger> |
correct |
| [08:48:27] |
<boorad> |
psycopg2 is doing that |
| [08:48:29] |
<boorad> |
hmm, wtf? |
| [08:50:01] |
* |
danbeck has joined #trac |
| [08:52:51] |
<boorad> |
coderanger: a db.commit() after the insert in cache.py, maybe? |
| [08:53:13] |
<coderanger> |
I should only insert if it has changed though |
| [08:53:33] |
<boorad> |
well, there was no row in there for repository_dir |
| [08:53:38] |
<boorad> |
just for db version |
| [08:54:06] |
<boorad> |
and the inserts weren't taking. |
| [08:54:29] |
<boorad> |
there is a db.commit() in the sync() method, but lower, and in a bit on an if/while block |
| [08:56:44] |
<coderanger> |
I may need to start postgres issues |
| [08:57:02] |
* |
pombreda has quit IRC |
| [08:57:13] |
<coderanger> |
I am helping the CS dept here do a Trac+SVN system for all students |
| [08:57:19] |
<coderanger> |
and they are pure Postgres |
| [08:57:50] |
* |
pombreda has joined #trac |
| [09:00:48] |
<boorad> |
coderanger: you may need to start postgres issues? |
| [09:00:54] |
<boorad> |
looking at? |
| [09:00:58] |
<coderanger> |
yeah |
| [09:01:01] |
<coderanger> |
me type good :) |
| [09:01:20] |
<boorad> |
pacopablo and I would be happy to help, I'm sure. |
| [09:01:48] |
<boorad> |
which psycpg2 are you using? |
| [09:01:51] |
<coderanger> |
heh, more importantly I need to finish TracForge to automate the setup process for new sites |
| [09:02:05] |
<coderanger> |
I'm not using PG on here, just sqlite |
| [09:02:10] |
<boorad> |
facist. |
| [09:02:25] |
<boorad> |
I've seen the "database locked" error too many times ;) |
| [09:02:32] |
* |
_shawn has quit IRC |
| [09:02:35] |
<coderanger> |
surprisingly that hasn't been an issue |
| [09:02:55] |
<boorad> |
it would go away as I cleared out the session table |
| [09:03:02] |
<coderanger> |
then again I had a huge deadlock problem, so I HUP apache every 15 minutes |
| [09:03:28] |
<coderanger> |
never did track it down |
| [09:03:30] |
<boorad> |
and another weird thing about sqlite, is, it would only do that database locked stack trace for certain users - not me, but another guy, same env, same time |
| [09:03:44] |
<boorad> |
I guess I got the good mod_python proc then :) |
| [09:04:23] |
<coderanger> |
I had thought it was something in tracforge, but even with very careful mutexing it still happened |
| [09:04:44] |
<boorad> |
yep, there's some gremlins in Trac - gotta be. |
| [09:05:30] |
* |
lisppaste5 has quit IRC |
| [09:09:41] |
* |
lisppaste5 has joined #trac |
| [09:10:04] |
* |
ajt has joined #trac |
| [09:10:22] |
<ajt> |
Is the xml-rpc trac plugin come with 0.10.2? |
| [09:10:26] |
<ajt> |
errr does the |
| [09:10:33] |
<coderanger> |
no, it is a plugin |
| [09:10:52] |
<ajt> |
thanks |
| [09:11:04] |
<ajt> |
coderanger: you ever use it with mylar? |
| [09:11:09] |
<coderanger> |
Nope |
| [09:11:26] |
<coderanger> |
I hate Eclipse |
| [09:11:36] |
<coderanger> |
TextMate ftw |
| [09:12:48] |
<ajt> |
I heard textmate is sexy :) |
| [09:12:50] |
<mitsuhiko> |
UBUNTU!!!!111!1!!11 |
| [09:12:55] |
<mitsuhiko> |
+gvim :) |
| [09:13:09] |
<ajt> |
haah, anyway didn't mean to start a flamewar :) |
| [09:13:20] |
<ajt> |
Just wondering if installing the xml-rpc module is worth the effort |
| [09:13:24] |
<ajt> |
I don't want to break anything |
| [09:13:27] |
<coderanger> |
mitsuhiko: Did you see the Satanic Edition? ;-) |
| [09:13:33] |
<mitsuhiko> |
coderanger: sure :D |
| [09:13:36] |
<coderanger> |
ajt: Its pretty easy to install |
| [09:14:31] |
<ajt> |
coderanger: I'm curious about this line "If you have the AccountManagerPlugin enabled and you followed their advise/example to disable the standard login module" and use HttpAuthPlugin |
| [09:14:33] |
<jcrawford> |
the company i work for is looking to use trac however our code is proprietary, is there a way to not allow just anyone to view the trac install? |
| [09:14:37] |
<ajt> |
Will it work with the DB auth plugin |
| [09:14:43] |
<coderanger> |
ajt: No |
| [09:15:01] |
<coderanger> |
jcrawford: TracPermissions |
| [09:15:03] |
<ajt> |
jcrawford: yeah disable all permissions for the anonymouse user |
| [09:15:14] |
<ajt> |
coderanger: thanks you jsut saved me a headache :) |
| [09:15:51] |
<coderanger> |
ajt: HttpAuth could probably be modified to support it, but right now it depends on acct_mgr |
| [09:16:00] |
* |
wkornewald has joined #trac |
| [09:16:15] |
<wkornewald> |
hi, any Trac developer here? |
| [09:16:16] |
<ajt> |
that's cool I'm not going to piss off the project admin by breaking trac :) |
| [09:16:41] |
<coderanger> |
wkornewald: Depending on your question, I might be able to help |
| [09:17:27] |
<wkornewald> |
I'm not sure whether I should add a new permission 'TRAC_SEE_EMAIL_ADDRESS' (or something like that) to allow for viewing email addresses or whether it's okay to only give TRAC_ADMIN this permission |
| [09:17:39] |
<wkornewald> |
or TICKET_ADMIN or TICKET_MODIFY |
| [09:18:10] |
<wkornewald> |
or whether I should not add any permission |
| [09:18:13] |
<coderanger> |
probably EMAIL_VIEW |
| [09:18:14] |
<wkornewald> |
http://trac.edgewall.org/ticket/153 |
| [09:18:35] |
<coderanger> |
permissions should be of the form subject_action |
| [09:18:47] |
<wkornewald> |
coderanger: what would be the best place for this permission? |
| [09:18:58] |
<coderanger> |
wkornewald: What do you mean |
| [09:19:03] |
<wkornewald> |
which source file? |
| [09:19:19] |
<coderanger> |
Depends on how you are doing the patch |
| [09:19:25] |
<wkornewald> |
correctly ;) |
| [09:19:27] |
<ajt> |
coderanger: thanks again |
| [09:19:27] |
* |
ajt has left #trac |
| [09:20:06] |
<coderanger> |
wkornewald: The way you are doing it is "wrong" :P but its because to do this correctly requires some big changes |
| [09:20:12] |
<wkornewald> |
I just want to add a new permission. since it's global (it will also affect wikis, etc.) I think it should be put in some global palce |
| [09:20:14] |
<coderanger> |
users need to become first-class objects |
| [09:20:39] |
<wkornewald> |
coderanger: what do you mean? |
| [09:21:14] |
<coderanger> |
wkornewald: There should be a User object, and the obfuscation should happen there, this also ties in the user metadata stuff |
| [09:21:31] |
<coderanger> |
but thats not easily doable right now |
| [09:21:46] |
<wkornewald> |
that would require DB schema modifications |
| [09:21:54] |
<coderanger> |
No it wouldnt |
| [09:22:01] |
<coderanger> |
but thats besides the point |
| [09:22:20] |
<wkornewald> |
currently, the owner and the reporter are stored as text directly in the ticket. I want to obfuscate them. how can I do this with the User object? |
| [09:22:56] |
<wkornewald> |
if it were a User ID instead of text I could do this, but I don't know how else I can do it |
| [09:23:10] |
<coderanger> |
when building the output it should do something of the form email = User(self.env, username).printable_email |
| [09:23:30] |
<coderanger> |
the user model object can obfuscate it there |
| [09:23:38] |
<wkornewald> |
but which username should it use for anonymous users? |
| [09:23:38] |
<coderanger> |
but I'm not suggesting you do that :P |
| [09:23:46] |
<coderanger> |
their session id |
| [09:24:40] |
<coderanger> |
I would put the permission in trac.perm |
| [09:24:42] |
* |
dh has quit IRC |
| [09:24:43] |
<wkornewald> |
and how do I know the session ID if the "reporter" field just contains a text like "Waldemar Kornewald <wkornewald@....net>"? |
| [09:26:09] |
<coderanger> |
wkornewald: You can make it backwards compatible with the old form, and just use the session ids from that point on |
| [09:26:53] |
<wkornewald> |
okay, but then I'm still missing the information on whether the session id describes a user ID or the actual username... |
| [09:27:53] |
<coderanger> |
if name is a user: username elif: name is 32 character alnum string and is a valid session ID: session else: old style name |
| [09:27:59] |
<wkornewald> |
this would at least require an "authenticated" field |
| [09:28:25] |
<wkornewald> |
and if I name my session like a username I have a problem ;) |
| [09:28:37] |
<coderanger> |
you can't do that |
| [09:28:42] |
<coderanger> |
it checks :P |
| [09:28:55] |
<wkornewald> |
ah, that's still not nice, though ;)) |
| [09:29:29] |
<coderanger> |
its better than scattering metadata logic all over the place |
| [09:29:46] |
<wkornewald> |
indeed |
| [09:30:23] |
<coderanger> |
all that stuff is on the agenda for 0.12 I think |
| [09:30:57] |
<wkornewald> |
but I think Trac needs a real solution, here |
| [09:31:04] |
<coderanger> |
for what? |
| [09:31:09] |
<wkornewald> |
for user authentication |
| [09:31:13] |
<wkornewald> |
and user management |
| [09:31:20] |
<coderanger> |
yes, thats on the list for 0.12 |
| [09:31:23] |
* |
Guardian has quit IRC |
| [09:31:26] |
<coderanger> |
user handling in general |
| [09:32:38] |
<wkornewald> |
hmm, maybe we should not allow for changing the session ID into a string of len != 32? |
| [09:33:18] |
<coderanger> |
its still easy enough to check if a given string is a known session ID |
| [09:33:24] |
<wkornewald> |
or maybe we should just use a prefix like "session:<session-id>" to indicate that the reporter used a session... |
| [09:34:23] |
<wkornewald> |
but then, this information could disappear...maybe rather indicate authenticated users? "user:<username>" (and disallow session names starting with "user:")? |
| [09:34:38] |
<coderanger> |
none of this is needed |
| [09:34:47] |
<coderanger> |
just look if the string exists as an sid |
| [09:35:20] |
<wkornewald> |
but what if someone else created a session with a username that you use later on? this couldn't be checked in some environments |
| [09:35:26] |
<wkornewald> |
(single-sign-on, especially) |
| [09:35:33] |
<wkornewald> |
(LDAP) |
| [09:35:36] |
<wkornewald> |
(...) |
| [09:36:16] |
<coderanger> |
then reset the session ID for the old user and rename anything of theirs |
| [09:36:39] |
<wkornewald> |
not very polite... |
| [09:36:50] |
<wkornewald> |
just deleting your session... |
| [09:36:51] |
* |
stepz is now known as stepz|afw |
| [09:36:55] |
<coderanger> |
also not very likely |
| [09:37:09] |
<coderanger> |
unless you were an anonymous user, and then got an account |
| [09:37:21] |
<coderanger> |
in which case you probably want everything left in your name |
| [09:37:32] |
<coderanger> |
so make the renaming optional I suppose |
| [09:37:46] |
<wkornewald> |
nah, it should be done without adding options... |
| [09:38:29] |
<coderanger> |
thats impossible without a major DB change though |
| [09:38:36] |
<coderanger> |
which isnt a good idea |
| [09:38:47] |
<wkornewald> |
the problem is that you'll take over the bugs of the old session |
| [09:38:48] |
<retracile> |
could someone create a session with a conflicting username with the purpose of causing problems for the original user? |
| [09:38:59] |
<coderanger> |
retracile: No |
| [09:39:24] |
<coderanger> |
retracile: An anon user can't change their sid to be the same as anyone else, including normal users |
| [09:39:45] |
<coderanger> |
wkornewald: Not if you choose "rename old session" when adding the new user |
| [09:39:47] |
<wkornewald> |
another problem is that if we use session IDs instead of free-form text fields we must keep all sessions in the DB (deleting will make the data unusable) |
| [09:40:11] |
<wkornewald> |
coderanger: you can't really choose this when users are coming from LDAP, etc. |
| [09:40:12] |
<coderanger> |
No you don't |
| [09:40:31] |
<coderanger> |
wkornewald: The session ID just gets fed into a metadata provider |
| [09:40:41] |
<coderanger> |
it can then choose what to do with it |
| [09:41:48] |
<wkornewald> |
what if the session already existed and you now create an LDAP user with the same name? ah, well, you could also rename the session ID in all tickets... |
| [09:42:04] |
<coderanger> |
wkornewald: That is what I am describing |
| [09:42:11] |
<wkornewald> |
I didn't realize that :) |
| [09:43:00] |
<wkornewald> |
and before deleting an old session we just replace the ticket fields with the user's name and email? |
| [09:44:55] |
<coderanger> |
wkornewald: No reason to not leave the session there then |
| [09:45:18] |
<wkornewald> |
coderanger: then, the DB will blow up with spammers and users just visiting the Trac site |
| [09:45:34] |
<wkornewald> |
hmm, well, we could only delete when the user didn't create anything |
| [09:45:51] |
<coderanger> |
wkornewald: Well just have it not delete session that own any objects |
| [09:46:47] |
<wkornewald> |
I think this is too much for my small IUserStore patch :) I hope that you'll still commit it (when it's finished) |
| [09:47:05] |
<coderanger> |
like I said, that would be how to do it "right" |
| [09:47:11] |
<coderanger> |
but I'm not suggesting you try to |
| [09:47:43] |
<wkornewald> |
I'd love to, but then I must set my priorities on Haiku and not Trac |
| [09:48:05] |
<wkornewald> |
I hope that my patch will be helpful, at least |
| [09:48:31] |
<wkornewald> |
what is missing to convince you to commit it? :) an admin interface? |
| [09:48:55] |
<wkornewald> |
(and of course fixing and adapting to my email obfuscation patch) |
| [09:49:02] |
<coderanger> |
I'm not a trac dev |
| [09:49:15] |
<coderanger> |
I just spend too much time writting plugins |
| [09:49:17] |
<coderanger> |
:) |
| [09:49:20] |
<wkornewald> |
hehe :) |
| [09:49:39] |
<wkornewald> |
and what would you say is missing form your non-trac-dev perspective? :) |
| [09:50:52] |
<wkornewald> |
what is this __all__ stuff in the .py files? |
| [09:51:16] |
<coderanger> |
it looks like your patch just hides addresses |
| [09:51:31] |
<coderanger> |
__all__ is the symbols that get imported what you do from <blah> import * |
| [09:51:44] |
<wkornewald> |
coderanger: I'm working on a new patch that still shows them if you have EMAIL_VIEW permission |
| [09:52:01] |
<wkornewald> |
coderanger: also, I was talking about the IUserStore patch |
| [09:52:22] |
<coderanger> |
right, but I would say you should obfuscate somewhat like google does, to avoid confusing people |
| [09:52:29] |
<coderanger> |
wkornewald: That stuff should be left until 0.12 |
| [09:52:46] |
<wkornewald> |
coderanger: it obfuscates such that someone@somewhere.com becomes someone@... |
| [09:52:57] |
<wkornewald> |
coderanger: I find it important, so I want to do it now |
| [09:53:13] |
<coderanger> |
wkornewald: Thats fine, but it shouldn't get merged in just because of that :P |
| [09:53:40] |
<wkornewald> |
hmm, maybe you're right and 0.11 should just hide the addresses? |
| [09:53:45] |
<coderanger> |
I am very much hoping that 0.11 will be feature-frozen very soon |
| [09:54:06] |
<coderanger> |
wkornewald: For obfuscation that sounds okay |
| [09:54:42] |
<wkornewald> |
you mean, just hiding them? |
| [09:54:55] |
<coderanger> |
I mean what you do now |
| [09:57:39] |
<wkornewald> |
ah, ok |
| [09:57:55] |
<wkornewald> |
but I think that this will be wasted work because 0.12 will do this better... |
| [09:58:24] |
<coderanger> |
possibly, but that doesn't mean its a bad idea :P |
| [09:58:38] |
<coderanger> |
some duplication of effort is par for the course |
| [09:59:25] |
<wkornewald> |
should I create a new class just for the EMAIL_VIEW permission? is there no easier way? |
| [10:01:10] |
<wkornewald> |
ah, found it |
| [10:01:51] |
<wkornewald> |
PermissionSystem, is that okay? |
| [10:04:22] |
<wkornewald> |
no, probably not... |
| [10:04:51] |
<wkornewald> |
I'll try it. bye |
| [10:05:03] |
* |
wkornewald has quit IRC |
| [10:07:37] |
* |
_shawn has joined #trac |
| [10:20:15] |
* |
pombreda has quit IRC |
| [10:20:40] |
<m_g> |
does anybody here use trac gantt? i need the necessary configuration entry and the tracgantt site is down |
| [10:20:56] |
* |
pombreda has joined #trac |
| [10:21:54] |
* |
klasstek has quit IRC |
| [10:25:15] |
* |
tpollari has quit IRC |
| [10:25:45] |
* |
tpollari has joined #trac |
| [10:30:24] |
<m_g> |
got it |
| [10:36:17] |
* |
hotte has joined #trac |
| [10:52:13] |
<m_g> |
hmm gantt seems to be incompatible with 0.10.3 |
| [10:52:57] |
* |
jborg has joined #trac |
| [10:53:54] |
* |
bmcquay has joined #trac |
| [10:53:59] |
<bmcquay> |
aloha |
| [10:54:32] |
<bmcquay> |
i want to do an alter table on my trac db but it looks like the db version is old and doesn't yet support alter table |
| [10:55:23] |
< |
