IRC log of #schooltool for Wednesday, 2005-08-10

*** gintas has quit IRC00:06
povbot/svn/commits: * srichter committed revision 4649:00:06
povbot/svn/commits: Let's rid ourselves from duplicate templates.00:06
srichtertvon: he is not here01:08
srichternot in my channel list at least01:08
tvonsrichter: yeah, I noticed after I said something01:08
srichtertvon: we now have only one that is shared between ST/SB in my refactoring branch01:09
tvonsrichter: nice01:09
tvonIt's hard to provide a regeression test for something when you can't reproduce it01:11
tvonbut I *think* know what the problem was01:11
tvonah, perhaps...01:12
tvonsrichter: is there a concept of roles in z3?01:22
srichterthe default security policy implements them01:23
povbot/svn/commits: * tvon committed revision 4650:01:42
povbot/svn/commits: Fix for issue328, please backport.01:42
*** munkee has quit IRC04:06
*** munkee has joined #schooltool04:14
povbot/svn/commits: * tvon committed revision 4651:04:59
povbot/svn/commits: broken ftest04:59
povbot/svn/commits: * tvon committed revision 4652:05:09
povbot/svn/commits: working on 329.  KHTML has no love for self-closing divs.05:09
povbot/svn/commits: * tvon committed revision 4653:05:26
povbot/svn/commits: the rest of the layout problem in issue329.  please backport this and r465205:26
povbot/svn/commits: * bskahan committed revision 4654:06:37
povbot/svn/commits: generation script to make schooltool's application calendar public and set security policy to the 0.11 defaults.06:37
povbot/svn/commits: I'm not sure if this is a good idea for back porting because the change in security policy - from unset on containers, to denyAccess - may result in unexpected behavior in existing sites.06:37
povbot/svn/commits: It's probably the correct way for SchoolTool to set access from now on though.06:37
povbot/svn/commits: Dragons:  The old is now  This shouldn't hurt people using release tarballs, but it may play havoc with devlopers svn sandboxes, sorry about that.06:37
povbot/svn/commits: * bskahan committed revision 4655:06:43
povbot/svn/commits: correct a docstring06:43
*** tvon has quit IRC06:52
*** tvon has joined #schooltool07:36
*** tvon has quit IRC08:13
*** tvon has joined #schooltool08:25
*** mgedmin has joined #schooltool10:48
*** ignas has joined #schooltool10:59
*** Aiste has joined #schooltool11:01
*** ignas has quit IRC11:01
*** ignas has joined #schooltool11:01
*** jonesieboy has joined #schooltool11:41
jonesieboyHi all - anyone around to offer a bit of help/reassurance to someone looking at the possibility of developing a component to plug into SchoolTool?11:44
jonesieboyGuess not :)  I'll be back!11:48
*** jonesieboy has left #schooltool11:48
*** jinty has joined #schooltool12:12
*** jinty has quit IRC12:23
*** jinty has joined #schooltool12:23
povbot/svn/commits: * jinty committed revision 4656:13:02
povbot/svn/commits: Port 4571, 4576 to the schoolbell release branch (
povbot/svn/commits: * jinty committed revision 4657:13:05
povbot/svn/commits: Back-port 4573 and 4578 to the schooltool release branch.13:05
povbot/svn/commits: * jinty committed revision 4658:13:11
povbot/svn/commits: merged 4631:4655 from debian-packaging into debian-packaging-nozope313:11
ignasany CanDo folks here ?13:41
*** SteveA has quit IRC13:48
*** SteveA has joined #schooltool13:51
*** bskahan has joined #schooltool14:11
* jinty gets really irritated with having to download zope for about the 5th time today and goes to kill the schooltool external14:30
povbot/svn/commits: * jinty committed revision 4659:14:35
povbot/svn/commits: Kill the schooltool zope external.14:35
*** ignas has quit IRC14:36
povbot/svn/commits: * jinty committed revision 4660:14:50
povbot/svn/commits: Remove Zope from the schoolbell tarball and update the README (sync from schoolbell).14:50
*** bskahan has quit IRC15:16
*** bskahan has joined #schooltool15:49
povbot/svn/commits: * jinty committed revision 4661:15:53
povbot/svn/commits: Add compatibility for building current release tarballs to the non-zope3 packaging stuff. tested as much as I can without having the rc3 tarballs.15:53
*** ignas has joined #schooltool16:29
*** srichter has quit IRC16:30
*** srichter has joined #schooltool16:34
*** strichter has joined #schooltool16:35
*** bskahan has quit IRC16:43
*** bskahan has joined #schooltool17:02
*** strichter has quit IRC17:12
*** jonesieboy has joined #schooltool17:53
*** jonesieboy has quit IRC17:56
*** th1a has joined #schooltool17:57
th1abskahan:  ayt?17:57
bskahanhi tom17:57
th1aHey.  Wasn't stopping resources from being added to groups an Etria story at some point?17:58
bskahanyes, there's no more members-resources17:59
th1aWas that in this contract?18:00
bskahanI think so, let me look18:00
th1aWe need a generations script to boot them out as well.18:00
bskahanok, that was discussed and voted down18:00
th1aIt seems like John Baillie's bug might be related to having resources in groups.18:01
bskahanok, I'll do a generation to remove them18:01
bskahanit shouldn't be difficult18:01
th1aOK.  Should be simple enough.18:01
bskahanwe talked about the generation initially and the conclusion was to not change existing groups on people18:02
bskahanbut I think that was just a path of least resistence decision18:02
th1aOK.  I guess I forgot our conclusion.18:02
th1aIt probably didn't seem to make much difference one way or another.18:02
bskahanmgedmin: in case people using 0.10 had set up groups with resources allready18:02
mgedminI'm not saying that is bad, but I'd feel better if the database state after an upgrade can always be exactly recreated by starting from a fresh DB and doing operations TTW or via REST18:03
mgedminit's just a feeling I have18:03
bskahanthis gets into my issues with the ACL generation script as well18:04
bskahanin place generations are great, but there's no way to tell people about what changed18:04
bskahanso when john baillie (for example) start wondering where all his projectors went ...18:05
srichterusing REST to recreate the previous DB state is impossible18:05
srichterthink about logging operations18:05
th1aWell, perhaps we could wait to see if resources in groups is actually the source of the problem.18:06
th1aI guess I could try to replicate it locally.18:06
bskahanth1a: I'm reasonably certain that having a resource in a group isn't the direct cause of the problem, it may be the root cause of the permissions error but we probably can fix a permissions error there without actually removing resources18:08
bskahanI'll experiment with it now18:08
*** gintas has joined #schooltool18:09
th1abskahan:  OK.  Sounds prudent.18:12
gintastvon, your change in rev 4650 is not well-tested18:15
gintasthat revision looks a bit messy as well, so I'd like to have reliable unit tests before I backport it18:18
gintasbskahan, can you deal with 332?18:19
bskahangintas: sure, which behavior do you think is correct?18:20
bskahanth1a: can you look at my comment on 332?18:20
th1aI'd say a) is fine.18:21
povbot/svn/commits: * gintas committed revision 4662:18:22
povbot/svn/commits: Backported revisions 4652, 4653 (should fix issue 329).18:22
bskahanth1a: on issue 333 (john's issue)18:23
bskahanI can't recreate it18:23
gintasand whoever disallowed resources as members of groups needs to write a generation script18:23
gintasor maybe not18:24
bskahangintas: if you were using 0.10 and had spent time creating groups of resources and did an apt-get upgrade to find those groups empty, would you be happy?18:24
th1aOK.  Let's wait and see what the problem is.18:24
bskahanI'd like to boot them in a generation because I think in the long run it will be a source of hard to track down bugs18:25
gintasthis does not look like a critical bug, I think we can release rc3 without a fix18:25
bskahanI've created (in 0.10) a group with 1 person and 1 resource.  added an event to the group, subscribed the user to the groups calendar18:26
bskahanafter upgrade it continues to work as expected18:26
gintastvon, are you there?18:26
gintasI need you to finish up 32818:26
th1aMaybe I'll see if John will send me his Data.fs.18:26
bskahanth1a: that would be good, if he CC's it to me I'll start working on it18:27
bskahanor to the list18:27
th1aOK.  I'll send him an email in a second.18:27
bskahanask him to send the _pre-upgrade_ data.fs18:28
bskahanif possible18:28
bskahangintas: is 328 the only thing stopping RC3?18:29
gintasI'd like a fix for 332 too, it does not look that hard18:33
gintasbut I definitely don't want to ship rc3 without a fix for 32818:34
gintasif Tom does not turn up I may decide to release with the fix that he's committed, after testing it manually thoroughly18:35
gintashe will still owe me some unit tests though18:35
gintasbskahan, can you fix 332 during the next hour or so?18:36
gintasif not, I'll just release without it, it's not that big of a deal (and the fix should not affect much anyway, so it's not risky)18:37
bskahanworking on it now18:37
gintashow long do you think it will take?18:39
gintasby the way, are you solving the general problem or just fixing the persons importer?18:40
gintasI think that probably the same problem is in the groups & resources importer, isn't it?18:41
bskahangintas: I think so18:42
gintasso, will your fix apply for resources and groups too?18:42
gintashow long do you think it will take for you to finish the fix?18:46
gintasjinty, are you there?18:51
*** Aiste has quit IRC18:56
gintasbskahan, will you land the fix in the next 15 minutes? otherwise I'm releasing rc3 now19:06
*** th1a has quit IRC19:08
tvongintas: 328 is fixed as far as I know19:14
*** alga has joined #SchoolTool19:15
gintastvon, are you sure it is tested well?19:16
gintasI added a 'b0rk' into a random place and all the tests passed19:16
gintasreplace the 'print' you had in there with an 'assert False'19:17
gintassee my reply to rev 465019:18
gintashappens to me too occasionally19:19
gintasanyway, the fix seems to be working19:19
gintasI backported the fix19:20
gintasyou can check in the missing test later19:20
povbot/svn/commits: * gintas committed revision 4663:19:22
povbot/svn/commits: Backported revision 4650 (without the spurious 'print' statement).  This is not well tested, but seemed to work well enough for me.19:22
povbot/svn/commits: By the way, a functional test broke because the order of overlaid calendars was switched around.  I'm not sure why this happened, it could be one of those undeterministic order bugs.19:22
*** thisfred has joined #schooltool19:24
bskahangintas: sorry about that, got pulled away by a phone call.  I have a fix for 332 but I'm having trouble reproducing it in the tests to get a regression test done19:26
gintasthat's odd19:26
gintasthe bug should be easy to reproduce19:26
gintasI cloned one line in src/schoolbell/app/browser/ftests/app.txt and got a failure19:27
gintasline 80019:27
bskahangintas: working on unit tests19:28
bskahanone sec19:28
bskahanthink I know what it is19:28
gintasbskahan, how's it going?19:37
bskahanfine, there was no PersonCSVImporterView test19:38
gintasI was planning to do the release before going home19:39
gintaslooks like I'll just pack it up and upload after I get back, so you don't need to hurry then19:39
* bskahan nods19:43
bskahansorry it took so long19:43
gintasnah, it's fine19:43
gintasI got all the packing stuff automated, so I can build packages with a single keypress19:44
gintasso there's not much of a difference when I get to build the packages19:44
gintasI'd just like to upload rc3 today19:44
bskahanI'll be done shortly I hope19:45
gintaswell, I still have to write the announcement, create the directories in plone, etc.19:45
* bskahan nods19:45
gintasI'll be heading home then, I should be back online in an hour19:45
gintashopefully you'll be done by then19:45
*** gintas has quit IRC19:46
bskahan        for dry_run in [True, False]:19:50
bskahan            for rowdata in rows:19:50
bskahan                self.createAndAdd(rowdata, dry_run)19:50
tvonruns through twice, first one is a dry run19:57
tvon(if "wtf" is what you meant)19:57
bskahanstarted like that20:01
bskahanfigured out why now20:01
jintygintas: I'm here now20:03
bskahantvon: do you know why the assert dry_run, ("...") is in BaseCSVimporter?20:04
jintydrat, didn' t realise he left already20:04
* tvon looks20:04
*** mgedmin has quit IRC20:05
tvonbskahan: that came from the old csvimporter20:05
bskahanok, it doesn't get raised in any test so I'm removing it20:06
tvonit will be raised if there are errors and it is not a dry run20:08
tvonI guess if it could be tested it would be accounted for and not necessary20:09
povbot/svn/commits: * bskahan committed revision 4664:20:16
povbot/svn/commits: fix for issue332 - persons/@@person-csvimport.html error This should be backported.20:16
*** ignas has quit IRC20:41
*** thisfred has quit IRC20:56
*** alga has quit IRC21:00
*** bskahan has quit IRC21:04
jintygintas: I have to go and have supper. But let me know your questions and I will try resolve them later tonight or tomorrow morning,21:25
*** gintas has joined #schooltool21:48
*** bskahan has joined #schooltool21:57
povbot/svn/commits: * gintas committed revision 4665:22:46
povbot/svn/commits: Backported rev 4664, added missing i18n:translate in
povbot/svn/commits: * gintas committed revision 4666:22:57
povbot/svn/commits: Tagging SchoolBell 1.2rc3.22:57
povbot/svn/commits: * gintas committed revision 4667:22:58
povbot/svn/commits: Tagging SchoolTool 0.11rc3.22:58
povbot/svn/commits: * gintas committed revision 4668:23:05
povbot/svn/commits: Updated Czech translation.23:05
povbot/svn/commits: * gintas committed revision 4669:23:16
povbot/svn/commits: Bumped version number for rc3.23:16
povbot/svn/commits: * gintas committed revision 4670:23:18
povbot/svn/commits: Bumped version, bound ST rc3 to SB rc3.23:18

Generated by 2.15.1 by Marius Gedminas - find it at!