Team Chat Logs

July 7, 2010

2010 6
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 31  
<< <   > >>

[00:48:35.023058]<evil_twin>t.e.o: Ticket #9484 ([PATCH] textarea formatting problems in TracNotification) updated - <http://trac.edgewall.org/ticket/9484#comment:6> - rblank
[01:30:57.783224]<mahound>hasienda: hey, thanks for the hint on the commit hook thing :)
[01:31:46.717125]<hasienda>mahound: did you manage to patch it meanwhile?
[01:36:17.404415]<hasienda>mahound: if so, I'd recommend to publish it with a ticket at trac.edgewall.org to fix it in trunk
[02:01:14.148579]<mahound>hasienda: not yet, still some things to decide
[02:02:37.812395]<hasienda>mahound: ok, np. But since it seemed like a perfectly valid improvement it would be worth sharing a solution anytime.
[02:06:05.169797]<hasienda>mahound: I'd assist in programming up to my still limitted Python knowledge, just nudge me later here,
[02:07:10.798685]<hasienda>mahound: but I'm currently unable to test because my Trac development environment lacks a repo configuration
[02:09:28.199808]<hasienda>mahound: By now I use Trac of other project management, not strictly related to coding, so doesn't need a repo.
[02:50:45.216684]<mahound>hasienda: well, no problem with programming, i'm quite comfortable with python
[02:50:55.279564]<mahound>hasienda: however, i'm still not sure how to solve it
[02:51:14.341471]<mahound>as in spite of being the same "patch", the two changesets are actually different
[02:51:18.226507]<mahound>they have different ids
[02:51:54.094092]<mahound>so, finding a criterion for telling if two changesets are identical... well, it can be very VC-dependent
[02:54:20.705474]<mahound>hasienda: actually, i'm not sure this is the method we want to change... i would instead add some kind of config option that makes _update_tickets ignore tickets that are already closed
[02:56:14.124075]<mahound>still, "references" will be valid even for tickets that are already closed :/
[02:56:19.387680]<mahound>it's not an easy fix
[02:56:36.322841]<hasienda>mahound: yeah, the duplicate check could be the point where to hook into existing code, but certainly not along the lines of that check
[02:57:28.640112]<hasienda>mahound: but as there is a easy way to set ticket to "closed" it should be to check, if it's already closed, right?
[02:58:26.609146]<mahound>hasienda: the problem is that there are other actions that won't close the ticket, but just add a comment
[02:58:39.796802]<mahound>and i guess they should remain valid even after the ticket is closed
[02:59:59.681691]<hasienda>mahound: so just check, if current attempted action is "close" and skip, if attempting a comment
[03:00:44.453228]<mahound>hasienda: yes, that one is obvious - if the ticket is closed, you can't close it again
[03:01:02.727143]<mahound>the problem is if there's a duplicate commit message, with "references #1234"
[03:01:41.451278]<mahound>even if the ticket is closed, i'm not sure if we should ignore it
[03:02:08.334903]<hasienda>mahound: define "duplicate commit message", please
[03:04:15.819308]<mahound>hasienda: you add some commit "references #123" message to branch_a, and then merge branch_a with branch_b
[03:04:28.376839]<mahound>the commit will be sent twice to trac
[03:17:58.364933]<get>hi all
[03:19:15.758357]<get>ive installed Accessmacro to protect some wikipages to be shown only by some members, is there a way to create usergroups and allow them to see wikipages?
[03:19:35.566622]<hasienda>mahound: I thought the duplicate check would cope with this already, no?
[03:23:47.187125]<hasienda>get: sure, you need to be a user with TRAC_ADMIN or at least PERMISSION_ADMIN priviledges, so you see and you're allowed to adjust permissions from web admin, see: http://trac.edgewall.org/wiki/TracPermissions#TracPermissions
[03:24:27.711258]<get>thx will check that out
[03:26:58.770620]<get>but, what i need is, for example, make a new group called "clients_admins" and those clients with permission clients_admins can see pages in wiki with {{{ #!access
[03:27:13.751951]<get>#allow(client_admins)
[03:27:26.908531]<get>what at the moment doesnt work
[03:38:56.749285]<evil_twin>t.e.o: Ticket #8875 (trac.wsgi does not work with multiple projects) updated - <http://trac.edgewall.org/ticket/8875#comment:6> - Lele Gaifax
[04:24:04.561903]<evil_twin>t.e.o: Ticket #8875 (trac.wsgi does not work with multiple projects) reopened - <http://trac.edgewall.org/ticket/8875#comment:7> - rblank
[04:34:06.918381]<evil_twin>t.e.o: Changeset [9947]: 0.12-stable: Follow-up to [8958], finally fix the wrong `environ` key. ... - <http://trac.edgewall.org/changeset/9947> - rblank
[04:34:11.917503]<evil_twin>t.e.o: Ticket #8875 (trac.wsgi does not work with multiple projects) closed - <http://trac.edgewall.org/ticket/8875#comment:8> - rblank
[04:47:09.433531]<mahound>hasienda: i think the duplicate chack only checks the revision number, no?
[04:59:13.980807]<evil_twin>t.e.o: Ticket #9111 (TimeoutError: Unable to get database connection within 20 seconds) updated - <http://trac.edgewall.org/ticket/9111#comment:13> - cboos
[05:04:13.077043]<evil_twin>t.e.o: t9111-create-close-outside-cs.2.diff attached to Ticket #9111 - <http://trac.edgewall.org/attachment/ticket/9111/t9111-create-close-outside-cs.2.diff> - cboos
[05:39:18.245691]<evil_twin>t.e.o: Ticket #9111 (TimeoutError: Unable to get database connection within 20 seconds) updated - <http://trac.edgewall.org/ticket/9111#comment:14> - cboos
[05:44:18.265676]<evil_twin>t.e.o: Ticket #9439 (KeyError: 'trac/locale') updated - <http://trac.edgewall.org/ticket/9439#comment:18> - anonymous
[06:11:18.178008]<ultramage>hello, I'm using latest trunk, and when updating an environment from yesterday to today, the revisions in the timeline got mixed up
[06:11:47.535331]<ultramage>actually, r7269 says 6.7.2010, r7279 says 5.7.2010...
[06:12:32.448655]<ultramage>damn.
[06:14:27.390854]<evil_twin>t.e.o: 0.10.4/TracOnUbuntuHardy edited - <http://trac.edgewall.org/wiki/0.10.4/TracOnUbuntuHardy?version=3> - anonymous
[06:14:32.390545]<evil_twin>t.e.o: 0.11 edited - <http://trac.edgewall.org/wiki/0.11?version=3> - anonymous
[06:18:30.211424]<ultramage>I don't get it... my mirror synced using a wrong time
[06:19:24.294295]<evil_twin>t.e.o: TracDev/ToDo edited - <http://trac.edgewall.org/wiki/TracDev/ToDo?version=102> - anonymous
[06:19:29.292297]<evil_twin>t.e.o: WikiContext edited - <http://trac.edgewall.org/wiki/WikiContext?version=7> - anonymous
[06:19:34.292220]<evil_twin>t.e.o: TracFaq edited - <http://trac.edgewall.org/wiki/TracFaq?version=318> - anonymous
[06:19:39.292010]<evil_twin>t.e.o: 0.10/WikiMacros edited - <http://trac.edgewall.org/wiki/0.10/WikiMacros?version=2> - anonymous
[06:19:44.291961]<evil_twin>t.e.o: TracGuide edited - <http://trac.edgewall.org/wiki/TracGuide?version=61> - anonymous
[06:19:49.290869]<evil_twin>t.e.o: 0.10/TracPlugins edited - <http://trac.edgewall.org/wiki/0.10/TracPlugins?version=2> - anonymous
[06:19:54.291308]<evil_twin>t.e.o: 0.10/TracUpgrade edited - <http://trac.edgewall.org/wiki/0.10/TracUpgrade?version=3> - anonymous
[06:24:25.117437]<evil_twin>t.e.o: 0.10/TracPermissions edited - <http://trac.edgewall.org/wiki/0.10/TracPermissions?version=2> - anonymous
[06:31:26.360382]<pipern>spams :-(
[06:40:53.608436]*retracile spouts something inconsequential about mornings.
[06:53:20.524041]<ultramage>I'm running trac-admin resync to see if the weird timestamp issue goes away
[06:55:13.502272]<retracile>ultramage: what does svn log tell you the timestamps are supposed to be?
[06:56:11.507097]<ultramage>they're perfectly fine
[06:56:12.515109]<ultramage>._.
[06:57:09.592516]<ultramage>the r7269 revision correctly says 4th july instead of 6th july (that trac sees)
[07:08:11.641804]<retracile>ultramage: btw, the resync command can accept a single revision to resync
[07:19:24.734025]<ultramage>;o
[07:19:43.244211]<ultramage>thanks for the info... I did a full resync (since there were multiple revisions messed up like that), and now everything's okay
[07:19:57.417352]<retracile>Ah, good.
[07:20:16.461822]<retracile>though I have to wonder how it got messed up in the first place..
[07:22:34.471872]<ultramage>yep
[07:22:51.614596]<ultramage>maybe trunk latest has a glitch somewhere
[07:23:23.142376]<ultramage>Revision: 9931
[07:23:30.575196]<ultramage>Last Changed Date: 2010-06-29 05:29:40 +0200 (Tue, 29 Jun 2010)
[07:23:36.822288]<ultramage>(ok, so from last week :)
[07:32:37.689405]<get>back
[07:32:41.029634]<get>hi again
[08:14:46.363832]<evil_twin>t.e.o: Ticket #9488 (ReferenceError: weakly-referenced object no longer exists) created - <http://trac.edgewall.org/ticket/9488> - jinleijp@…
[08:16:33.522858]<get>whats the correct way for creating global permissions? like REPORT_VIEW, or WIKI_VIEW ?
[08:19:44.680038]<evil_twin>t.e.o: Ticket #9488 (ReferenceError: weakly-referenced object no longer exists) closed - <http://trac.edgewall.org/ticket/9488#comment:1> - cboos
[08:19:49.679037]<evil_twin>t.e.o: TracGuide edited - <http://trac.edgewall.org/wiki/TracGuide?version=61> - anonymous
[08:19:54.838576]<retracile>get: What do you want to use these new permissions for?
[08:20:41.258723]<get>to use restrict users to only some wikipages, read and/or read+write
[08:21:39.752350]<retracile>get: using an existing plugin?
[08:22:04.850795]<get>yes
[08:22:14.428496]<retracile>get: which one?
[08:22:18.008634]<get>restricted area
[08:22:37.191902]<get>but it only permits 1 permission
[08:22:41.705139]<get>i need some of them
[08:22:47.683552]<get>for each of our costumers
[08:23:33.054651]<get>so i can "jail" them to see only those wiki pages and/or edit, where i have asigned them
[08:23:50.602528]<get>i will prevent our costumers to see our own internal wikipages
[08:24:08.911715]<retracile>get: Look at the extrapermissionsprovider.py in the sample-plugins/permissions directory of the Trac source.
[08:24:19.960323]<get>mmm
[08:24:28.201144]<get>any manual online for?
[08:24:38.557378]<retracile>get: that will let you add arbitrary permissions to the system.
[08:24:58.376921]<retracile>get: docs are in the file
[08:25:52.169161]<get>will check that
[08:32:51.401931]<get>retracile: !!!!!
[08:32:59.160524]<get>many thx!
[08:33:04.522649]<get>thats just the stuff
[08:35:04.715125]<retracile>great :)
[08:41:07.663303]<get>retracile: and this extraperm
[08:41:15.024162]<get>i have to compile or what?
[08:41:24.519933]<get>or how to integrate it?
[08:41:55.794891]<retracile>get: you should be able to toss the .py into your plugins directory in your trac environment (and maybe restart the Trac server)
[08:42:21.868713]<get>ah, move that to the plugin dir?
[09:00:29.139875]<wild_oscar>isn't it possible to use < > operators in tracQuery? like id>100?
[10:37:42.711362]<hasienda> mahound: are you still there? I'm back now, and I'll check the code now to find out about your last question.
[10:51:08.293502]<hasienda>mahound: looks like it will consider not only the revision no. cite: "cset_id = (changeset.rev, changeset.message, changeset.author, changeset.date)", but the following line looks not too promizing: "if cset_id != self._last_cset_id", this should only check the very last message, right?
[10:51:29.741526]<hasienda>mahound: As I understood you, the duplicates arise from kind of replaying old revisions, so these older ones would certainly go undetected.
[10:52:13.709787]<hasienda>mahound: But I have to question you about the whole issue a little more now. Do you speak from live, own experience?
[10:55:07.710393]<hasienda>mahound: If so, what kind of repo did you use? I ask because from my small Mercurial experience this shouldn't happen as you describe.
[10:55:12.993581]<evil_twin>t.e.o: Ticket #206 (rss output for <title> needs to be encapsulated CDATA) updated - <http://trac.edgewall.org/ticket/206#comment:4> - anonymous
[10:55:17.767504]<hasienda>mahound: A merge will require a new revision and this is just that, a new revision. I can't see, where it should trigger a comment to a ticket mentioned in any earlier revision.?
[10:57:37.852109]<hasienda>mahound: input to commit updater should be just the commit message of the revision after the merge, and that doesn't need to contain anything from previous revisions. - But maybe I overlook something obvious due to my limited knowledge.
[11:00:13.971323]<evil_twin>t.e.o: Ticket #9489 (Admin-Login window is not acceptiong alphebets) created - <http://trac.edgewall.org/ticket/9489> - anonymous
[11:10:15.925618]<evil_twin>t.e.o: Ticket #9489 (Admin-Login window is not acceptiong alphebets) updated - <http://trac.edgewall.org/ticket/9489#comment:1> - anonymous
[11:30:18.864603]<evil_twin>t.e.o: GameReferences created - <http://trac.edgewall.org/wiki/GameReferences?version=1> - bmoussavi@…
[11:30:23.864317]<evil_twin>t.e.o: GameReferences edited - <http://trac.edgewall.org/wiki/GameReferences?version=2> - bmoussavi@…
[11:40:20.832177]<evil_twin>t.e.o: AnimalCrossing created - <http://trac.edgewall.org/wiki/AnimalCrossing?version=1> - bmoussavi@…
[11:40:25.830266]<evil_twin>t.e.o: GameReferences edited - <http://trac.edgewall.org/wiki/GameReferences?version=3> - bmoussavi@…
[11:45:21.814984]<evil_twin>t.e.o: AnimalCrossingWildWorld created - <http://trac.edgewall.org/wiki/AnimalCrossingWildWorld?version=1> - anonymous
[11:50:27.027183]<evil_twin>t.e.o: AnimalCrossingWildWorld edited - <http://trac.edgewall.org/wiki/AnimalCrossingWildWorld?version=2> - bmoussavi@…
[11:50:32.027008]<evil_twin>t.e.o: AnimalCrossingWildWorld edited - <http://trac.edgewall.org/wiki/AnimalCrossingWildWorld?version=3> - bmoussavi@…
[12:45:30.080972]<retracile>*sigh* It's always fun when someone gets on teo thinking they're on their own Trac instance....
[12:47:27.083233]<retracile>ok, those 3 pages are cleaned up.
[12:52:54.906955]<srl295>t.e.o isn't my own trac/wiki host? </sarcasm>
[12:59:06.502142]<retracile>srl295: well, for _some_ of us it is ;)
[13:31:02.549744]<exarkun>svn post-commit hook to add comments to trac tickets seems really unreliable
[13:31:19.614471]<exarkun>when I was using SQLite it was unreliable and failed with an obvious cause, like a SQLite lock error
[13:31:30.091120]<exarkun>since I switched to postgres, it's unreliable and fails with no obvious cause
[13:31:35.738395]<exarkun>anyone seen anything like that?
[13:41:03.229740]*srl295 is very happy to have completed trac 0.12 + bitten upgrade from 0.10: http://bugs.icu-project.org/trac
[13:43:26.323608]<srl295>exarkun> no, but i changed my httpd setup to be mpm_worker to reduce lock errors within http, and instead of using a hook to add comments, I wrote a plugin that just searches (a cached database of) the changeset<->ticket mappings
[13:44:24.664630]<exarkun>plugin sounds like a more reliable way to do it, I suppose... except since I have no idea why it fails now, maybe it wouldn't actually be more reliable.
[13:55:44.515842]<evil_twin>t.e.o: Ticket #9489 (Admin-Login window is not acceptiong alphebets) closed - <http://trac.edgewall.org/ticket/9489#comment:2> - rblank
[13:55:49.515421]<evil_twin>t.e.o: BadContent edited - <http://trac.edgewall.org/wiki/BadContent?version=129> - rblank
[14:07:22.765622]<ngirard>Hi all. I need to build a query matching tickets whose keywords contain *both* A and B. How can I achieve this ?
[14:09:23.650290]<ngirard>unless i'm mistake, TicketQuery(keywords=~A&keywords=~B) performs an OR
[14:09:30.827255]<ngirard>while i need an AND
[14:25:50.383932]<evil_twin>t.e.o: Ticket #7975 (Viewing changesets is very slow) updated - <http://trac.edgewall.org/ticket/7975#comment:9> - myers@…
[17:16:21.653170]<evil_twin>t.e.o: Ticket #2099 (Custom ticket fields were deleted from trac.ini, but still appear in view) updated - <http://trac.edgewall.org/ticket/2099#comment:12> - anonymous
[21:47:08.708078]<evil_twin>t.e.o: Ticket #9490 (Python and subversion bindings) created - <http://trac.edgewall.org/ticket/9490> - alex_m80@…