| *** pcardune_ has quit IRC | 00:51 | |
| *** FarcePest has quit IRC | 01:21 | |
| *** FarcePest has joined #schooltool | 01:21 | |
| *** ignas has joined #schooltool | 01:35 | |
| ignas | anyone alive ? | 01:46 |
|---|---|---|
| tvon | I'm here | 02:00 |
| ignas | what should be permissions on school calendar ? | 02:23 |
| ignas | i mean - now in a clean database - non users can't see the calendar | 02:24 |
| ignas | authenticated users have "View" | 02:24 |
| ignas | unauthenticated "viewcalendar" set | 02:24 |
| ignas | though if one upgrades from 0.10 | 02:24 |
| ignas | he gets "view" set for authenticated and not authenticated users | 02:25 |
| ignas | and "viewcalendar" set for unauthenticated | 02:25 |
| ignas | some mystic :/ - if users that have not logged in should not see the schoolwide calendar - then why they have viewcalendar set ? | 02:26 |
| ignas | and if they should see the schoolwide calendar - well they can't! | 02:26 |
| ignas | unless they have upgraded from an older SchoolTool\ | 02:27 |
| tvon | sorry, back | 02:28 |
| tvon | In a clean DB, unauthenticated users cannot see the site-wide calendar?? | 02:28 |
| ignas | yep | 02:28 |
| tvon | unauth should have viewCalendar | 02:28 |
| ignas | yep they do have it | 02:28 |
| ignas | though it doesn't help that much | 02:29 |
| ignas | if you don't have view permission | 02:29 |
| tvon | is this SB or ST? | 02:29 |
| ignas | ST\ | 02:29 |
| ignas | the release branch | 02:29 |
| tvon | trunk seems okay, unauth has view and viewcalendar with a fresh db | 02:34 |
| ignas | (<schoolbell.calendar.simple.ImmutableCalendar object at 0xb170316c>, 'expand', 'schoolbell.view') | 02:35 |
| ignas | maybe there was an issue with this ? | 02:35 |
| tvon | that was the other zcml issue marius mentioned | 02:35 |
| tvon | view vs viewCalendar | 02:35 |
| ignas | r4735: fix ExpandedCalendarEvent permission | 02:35 |
| ignas | i guess this one | 02:35 |
| tvon | filed as the same issue, yes | 02:36 |
| ignas | you should mark it to backport | 02:36 |
| tvon | ExpandedCalendarEvent and ImmutableCalendar both had the same permissions problem | 02:36 |
| ignas | because i guess it is not in the release branch yet | 02:36 |
| tvon | the ImmutableCalendar permissions were not fixed, just ExpandedCalendarEvent | 02:37 |
| ignas | well, if you know how to do it, could you please do it and maybe add it to the issue (or create another issue) | 02:39 |
| ignas | yet - i guess there was some revision that fixed it ... | 02:39 |
| tvon | How are you getting that error? | 02:39 |
| ignas | just going to localhost:7080 | 02:40 |
| tvon | thats it? | 02:40 |
| ignas | on a release branch | 02:40 |
| ignas | btw 4735 hasn't been backported yet :/ i just checked it | 02:42 |
| tvon | was just looking for that in my mail | 02:43 |
| tvon | adding a comment to the bug | 02:43 |
| *** bskahan has joined #schooltool | 02:54 | |
| ignas | hi | 02:58 |
| povbot | /svn/commits: * srichter committed revision 4752: | 03:07 |
| povbot | /svn/commits: Moved Course and Section implementations into their own package. | 03:07 |
| povbot | /svn/commits: It is getting there. The code looks already much more managable. However, the tests are generally in bad shape. The unittests use a lot of other classes that should be really stubs and we should put much more emphasis putting the tests into text files and making them more instructional. | 03:07 |
| ignas | tvon, are you tackling the ImmutableCalendar permissions or should i create an issue in the issue tracker ? | 03:12 |
| tvon | ignas: I'm working on it | 03:12 |
| tvon | wondering why it isn't happening in trunk | 03:13 |
| tvon | I guess the test could be a straight permissions check | 03:13 |
| ignas | well it's a functional one imho | 03:14 |
| ignas | so it should be easy to look at the schoolwide calendar without any authorization info | 03:15 |
| tvon | I agree, but It's not showing up in trunk | 03:15 |
| ignas | if zcml is not right - it will surely fail | 03:15 |
| ignas | well if it doesnot then you should diff the trunk's zcml file with the release zcml file | 03:16 |
| ignas | :/ i don't have my laptop at the moment, i had a script that would display all the revisions of some file, so one could easily grep for the change | 03:17 |
| *** bskahan has quit IRC | 03:20 | |
| tvon | nothing seems relevant in app/configure.zcml | 03:21 |
| ignas | yep | 03:21 |
| ignas | weird | 03:21 |
| ignas | what are permisions set on the calendar ? | 03:21 |
| * ignas is coin Zope3 so could run the trunk | 03:22 | |
| * tvon wishes he had a point-and-click way to diff specific branch files | 03:23 | |
| ignas | well they are not that different | 03:23 |
| ignas | yhe only dfifference is the change you made\ | 03:23 |
| ignas | sorry dor so many typos, i am using a backup laptop | 03:24 |
| tvon | ah, heh. Something wrong with your normal laptop? | 03:24 |
| ignas | so - no think light, different keyboard, and only a candle as a lightsource | 03:24 |
| ignas | my old laptop is geting his backlight fixed | 03:24 |
| tvon | ahh | 03:25 |
| tvon | I love the think light | 03:25 |
| ignas | Unauthenticated users have a view permission in a clean Data.fs | 03:26 |
| ignas | when using trunk | 03:26 |
| ignas | \so permisions are as borken as they are in the release branch, though we can't see that because unauthenticated users got more permissions than they need ... | 03:26 |
| tvon | ahh | 03:27 |
| tvon | there it is | 03:31 |
| tvon | SchoolBell sets viewCalendar on app.calendar, SchoolTool sets the permissions on app | 03:31 |
| tvon | which is right? | 03:31 |
| tvon | Should the permissions be set on app or app.calendar? | 03:32 |
| *** eldafar has quit IRC | 03:35 | |
| ignas | sorry 3:39 am | 03:38 |
| ignas | brains barely working | 03:38 |
| tvon | no problem | 03:41 |
| povbot | /svn/commits: * ignas committed revision 4753: | 03:42 |
| povbot | /svn/commits: A script for testing whether evolution scripts are migrating data properly. Just launch and go to localhost:7080, then do some extensive clicking to see whether your evolution scripts broke anything or not. | 03:42 |
| ignas | yippie | 03:42 |
| ignas | and we wanted to release yesterday | 03:47 |
| tvon | yeah | 03:47 |
| tvon | r4579 is what changed the permissions on ST 'app' but it seems it was never backported... | 03:49 |
| tvon | (which is why trunk isn't getting the error) | 03:49 |
| tvon | I'm not sure it should be backported though | 03:49 |
| ignas | gintas will be online 8:00 GMT if that helps you | 03:51 |
| tvon | it does, I'll ask him about it | 03:51 |
| ignas | at least that's what he told me ;) | 03:51 |
| tvon | maybe I'll just leave a long comment on the bug and talk to him in my AM | 03:53 |
| *** bskahan has joined #schooltool | 04:31 | |
| bskahan | tvon: does it make sense to set 'viewCalendar' on the actual calendar object? | 04:33 |
| tvon | bskahan: I think so but I'm not sure if its 'right' | 04:33 |
| bskahan | I was thinking it didn't make sense | 04:33 |
| bskahan | since 'view' means "you can look at this context", and 'viewCalendar' means "you can look at the calendar this object owns" | 04:34 |
| tvon | do you know if the evolve script for those permissions changes was backported? | 04:34 |
| bskahan | yes, I believe gintas backported it | 04:35 |
| tvon | the permissions changes themselves never were | 04:35 |
| bskahan | ah | 04:35 |
| *** povbot has joined #schooltool | 04:53 | |
| tvon | hrm, front page re-direction to /calendar is ftestsed in the branch.. | 05:06 |
| povbot | /svn/commits: * srichter committed revision 4754: | 05:47 |
| povbot | /svn/commits: Added new test-setup framework. | 05:47 |
| povbot | /svn/commits: I think these very few lines of code have a lot of potential! I already converted the SchoolBell application setup and sitification to use the new test-setup registry, and it works great. The registry could also be used to setup implemented interfaces on components (usually done in ZCML), adapter setups, like notes and preferences and so on. If you want to know more about | 05:47 |
| povbot | /svn/commits: the test-setup registry read schoolbell.app.testing/README.txt. | 05:47 |
| *** ignas has quit IRC | 06:24 | |
| *** gintas has joined #schooltool | 11:04 | |
| *** th1a has joined #schooltool | 11:31 | |
| povbot | /svn/commits: * gintas committed revision 4755: | 12:12 |
| povbot | /svn/commits: Backported rev 4735 (issue 345). | 12:12 |
| *** jinty has joined #schooltool | 12:18 | |
| povbot | /svn/commits: * gintas committed revision 4756: | 12:18 |
| povbot | /svn/commits: I forgot to allow the title attribute in SchoolTool when I made the same change in SchoolBell. This is part of a temporary workaround; this problem should be fixed on trunk in a different, proper way. | 12:18 |
| jinty | gintas: re issue 334, I posted a more or less complete list of patches to be back-ported. I can't evaluate it, so I hand it to you. | 12:32 |
| povbot | /svn/commits: * gintas committed revision 4757: | 12:38 |
| povbot | /svn/commits: Fix for issue 351. | 12:38 |
| gintas | jinty, thanks | 13:10 |
| gintas | I'll look at it soon, I'm now working on the undeterministic tests | 13:10 |
| povbot | /svn/commits: * gintas committed revision 4758: | 13:26 |
| povbot | /svn/commits: Updated PersonTimetableSetupView to have a more deterministic sort order. Hopefully this will fix the unit test failure on Windows. | 13:26 |
| *** admp has joined #schooltool | 13:32 | |
| *** admp has quit IRC | 13:33 | |
| povbot | /svn/commits: * gintas committed revision 4759: | 13:35 |
| povbot | /svn/commits: Backported rev 4758 (fix for test failures on Windows). | 13:35 |
| povbot | /svn/commits: * jinty committed revision 4760: | 13:41 |
| povbot | /svn/commits: Fix the schoolbell part of issue 231. | 13:41 |
| *** bskahan has joined #schooltool | 13:43 | |
| povbot | /svn/commits: * gintas committed revision 4761: | 13:46 |
| povbot | /svn/commits: Oops, apparently I broke a functional test and somehow missed it. | 13:46 |
| povbot | /svn/commits: * jinty committed revision 4762: | 13:47 |
| povbot | /svn/commits: Fix the schooltool part of issue 231. | 13:47 |
| povbot | /svn/commits: * gintas committed revision 4763: | 13:50 |
| povbot | /svn/commits: Unwedged tests. | 13:50 |
| gintas | jinty, should 4760 & 4762 (copyright issues) be backported? | 13:54 |
| jinty | gintas: they depend on a lot of other patches. I was thinking of proposing them for x.y.1. | 14:03 |
| jinty | if you want, I can compile a list for review... | 14:03 |
| jinty | but not today, I need to start doing my paid work | 14:04 |
| povbot | /svn/commits: * gintas committed revision 4764: | 14:05 |
| povbot | /svn/commits: Backported rev 4579 (set view and viewCalendar permissions on SchooltoolApplication) | 14:05 |
| povbot | /svn/commits: I'm lucky Tom Hoffman noticed this. | 14:05 |
| th1a | Our users are lucky I noticed it ;-) | 14:06 |
| *** jinty has quit IRC | 14:06 | |
| *** tiredbones has joined #schooltool | 14:14 | |
| srichter | th1a: how was London? | 14:38 |
| th1a | Good. I put in a good word for you with Mark. | 14:38 |
| srichter | Great! :-) | 14:38 |
| srichter | thanks | 14:38 |
| srichter | did you meet him? | 14:39 |
| th1a | We were both at the Canonical offices all day. I mostly talked to Helen King from the Foundation. | 14:39 |
| th1a | She's going to a meeting in San Francisco hosted by a foundation there that is interested in open source SIS's. | 14:39 |
| srichter | I am really interested in hearing about the outcome there | 14:40 |
| th1a | Indeed. | 14:40 |
| srichter | did you see the adjusted proposal I sent you? | 14:41 |
| srichter | btw, I made some really good progress and the code is much cleaner now | 14:41 |
| srichter | I still have to work on the tests a bit | 14:41 |
| th1a | It looks like you've been moving a lot of code around :-) | 14:42 |
| srichter | yes | 14:43 |
| srichter | but more importantly I removed a lot of stuff too | 14:43 |
| srichter | I think that all code duplications are gone now | 14:43 |
| th1a | Right. | 14:44 |
| srichter | I wrote a testing setup framework (... well, aboiut 20 lines of code ;-) that seems very promising; I am still waiting for Marius' comments | 14:45 |
| th1a | It is a subject near and dear to his heart. | 14:45 |
| srichter | right, I think I have implemented about 50% of his dream ;-) | 14:46 |
| povbot | /svn/commits: * gintas committed revision 4765: | 14:48 |
| povbot | /svn/commits: Merging part of Tom's batching update: 4695:4702, 4707:4709, 4712 | 14:48 |
| povbot | /svn/commits: To be honest, this one is ugly. I decided to backport it because most of the changes were to page templates, which possibly should have landed in views because of their complexity, but they appeared to me rather declarative, i.e., the type of code that either works or it doesn't. I couldn't pinpoint any serious bugs in the patch, and I could not break it with manual testing. | 14:48 |
| povbot | /svn/commits: There are also some functional tests too, that's a plus. | 14:48 |
| povbot | /svn/commits: * gintas committed revision 4766: | 14:55 |
| povbot | /svn/commits: Backported the other part of batching improvements: revs 4710, 4713. | 14:55 |
| povbot | /svn/commits: This should close issue334. | 14:55 |
| povbot | /svn/commits: * srichter committed revision 4767: | 15:00 |
| povbot | /svn/commits: This is for Marius: Some syntactic sugar to implement setupThis() and setupThat() functions. This implements about 50% of Marius' testing framework dream. Currently we do not have to worry about teardowns, since all the registrations are done with the component architecture, which is torn down correctly anyways. | 15:00 |
| povbot | /svn/commits: For reference, here is Marius' dream: ------------------------------------- | 15:00 |
| povbot | /svn/commits: I'm thinking this will make unit testing painful, soon. | 15:00 |
| povbot | /svn/commits: Do you think it is possible to come up with some sort of a common schooltool.testing.setup module with factory functions that call some set of agreed-upon subscribers? ?\194?\160Then in all the tests you could just do | 15:00 |
| povbot | /svn/commits: ?\194?\160 ?\194?\160 >>> app = setup.createApplication() | 15:00 |
| povbot | /svn/commits: It might also have some setup functions for registering common adapters etc. | 15:00 |
| povbot | /svn/commits: Sometimes I dream about unit test setup functions with automatic dependency resolution and registration of tearDown methods... | 15:00 |
| povbot | /svn/commits: ?\194?\160 ?\194?\160 >>> setUpThis() ?\194?\160 ?\194?\160 >>> setUpThat() ?\194?\160 ?\194?\160 >>> substituteUtility(IFoo, UtilityStub) | 15:00 |
| povbot | /svn/commits: ?\194?\160 (setUpThis calls placeless setup and traversal setup; setUpThat calls ?\194?\160 placeless setup and annotation setup; all setup functions are NOPs ?\194?\160 when invoked more than once for a single test case; all setup ?\194?\160 functions register tearDown handlers somewhere; substituteUtility ?\194?\160 calls provideUtility and registers a tearDown handler | 15:00 |
| povbot | /svn/commits: that ?\194?\160 automatically restores the old utility) | 15:00 |
| povbot | /svn/commits: ?\194?\160 ?\194?\160 >>> code that uses this and that | 15:00 |
| povbot | /svn/commits: ?\194?\160 (tearDown handlers are automatically called at the end of the test) | 15:00 |
| povbot | /svn/commits: It would probably involve some module global that holds the current test context with all these teardown handler registrations. ?\194?\160I am content to accept that all schooltool's doctest would need to specify some setUp/tearDown arguments to DocTestSuite/DocFileSuite... | 15:00 |
| povbot | /svn/commits: End of dream. | 15:00 |
| tiredbones | I'm a new lurker, the reason for this noob question, Can postgreSQL be used with schoolbell? | 15:00 |
| th1a | Hi tiredbones. | 15:01 |
| th1a | Not really. | 15:01 |
| th1a | It is possible. | 15:01 |
| th1a | There are Zope 3 adapters to postgres. | 15:01 |
| th1a | What would you need to do? | 15:01 |
| th1a | That is, it is possible, but probably not something you want to do, except in a simple case. | 15:02 |
| th1a | Unless you're a Zope 3 guru, I suppose. | 15:03 |
| tiredbones | I never used ZOPE and I understand it has a high learning curve. | 15:03 |
| srichter | tiredbones: Zope 3's learning curve is much lower | 15:03 |
| th1a | Zope 3 is more Pythonic. | 15:03 |
| *** jonesieboy has joined #schooltool | 15:04 | |
| srichter | here is what would be involved: | 15:04 |
| srichter | - let's say you want your users to be populated from SQL | 15:04 |
| srichter | - you would need to reimplement the Person and PersonContainer class | 15:05 |
| srichter | - Reimplement a couple views (not much) that refer to those classes | 15:05 |
| srichter | th1a: btw, if we want to support this use case better, we could do a lot by abstracting the code a bit more | 15:06 |
| th1a | More abstraction! The answer to everyone's problems. | 15:07 |
| srichter | right now, we often use the content component classes directly (i.e. CSV import), and it could be done with factories | 15:07 |
| tiredbones | I also understand that object DB are slower that RDBM.Do you folks find this so with ZOPB? | 15:08 |
| srichter | th1a: actually, it is one of those cases where Zope 3 provides the necessary facilities already and ST just does not use them | 15:08 |
| th1a | Ah. | 15:08 |
| srichter | tiredbones: usually databases for schools are not large, so you should have no problem | 15:08 |
| th1a | We aren't targeting big institutions at this point. | 15:09 |
| srichter | tiredbones: big sites, like boston.com, run on ZODB, so I think it will be fine for schools | 15:09 |
| th1a | There are other strategies for increasing performance. Caching, etc. | 15:09 |
| srichter | right, caching is one of those things that boston,com does very effectively | 15:10 |
| tiredbones | Well, I guess I'll download ZOPE 3 see what it looks like. | 15:10 |
| srichter | tiredbones: are you a Python developer? | 15:11 |
| tiredbones | At what point in time do you folks thing schoolbell will be able to do the moodle stuff. | 15:11 |
| tiredbones | srichter, I dable in it. | 15:12 |
| srichter | you should be a book on Zope 3 (preferably mine ;-) | 15:14 |
| tiredbones | srichter, what the name of book and how thick is it? | 15:15 |
| srichter | Zope 3 Developer's Handbook by SAMS | 15:16 |
| *** mgedmin has joined #schooltool | 15:16 | |
| srichter | It's about 400 pages | 15:16 |
| th1a | tiredbones: We have funding to do student information system functionality. We don't have funding to do Moodle-type elearning stuff. | 15:16 |
| srichter | Part III and IV are most interesting to you right now | 15:17 |
| th1a | However, combining SchoolTool with Zope 3 content management as that emerges should make it relatively easy to create a cool learning management system (a la Moodle). | 15:17 |
| srichter | part I and II are introductory material to get you in the Zope 3 mindset | 15:17 |
| *** tvon has quit IRC | 15:18 | |
| tiredbones | srichter, at 400 pages it doesn't sound like a door stopper. I'll take a look at. | 15:18 |
| srichter | the chapters are all fairly small recipes, so you can read as much as you need without reading everything | 15:18 |
| srichter | it is basically a learn-it-yourself book for Zope 3 (I hate dry text books) | 15:19 |
| srichter | mgedmin: did you see the testing framework checkins? I think it is decent first attempt to implement your dream ;-) | 15:20 |
| tiredbones | I just downloaded schoolbell, do I need schooltools also? | 15:20 |
| srichter | if you want schooltool functionality, yes | 15:21 |
| srichter | in fact, then you can just download schooltool | 15:21 |
| tiredbones | Well, I'll go back to lurker now. | 15:21 |
| tiredbones | s/lurker/lurking/ | 15:22 |
| povbot | /svn/commits: * gintas committed revision 4768: | 15:22 |
| povbot | /svn/commits: Cosmetic fix. | 15:22 |
| th1a | tiredbones: SchoolBell is for small organizations that aren't schools. If you're a school, definitely use SchoolTool. | 15:23 |
| mgedmin | srichter, not yet | 15:25 |
| mgedmin | people, please do not write just "fixed issue 123512365213" in a commit message -- it is PAIN to figure out later | 15:25 |
| mgedmin | tell what the issue was about, in a couple of words | 15:25 |
| mgedmin | gintas, regarding this | 15:26 |
| mgedmin | "Backported rev 4579 (set view and viewCalendar permissions on SchooltoolApplication)" | 15:26 |
| mgedmin | is this the same issue that ignas and tvon discussed late at night? | 15:26 |
| mgedmin | unauthenticated users unable to view the school-wide calendar? | 15:26 |
| mgedmin | no, it doesn't look that way... | 15:28 |
| mgedmin | <tvon> SchoolBell sets viewCalendar on app.calendar, SchoolTool sets the permissions on app | 15:29 |
| mgedmin | <tvon> which is right? | 15:30 |
| mgedmin | <tvon> Should the permissions be set on app or app.calendar? | 15:30 |
| mgedmin | tvon: when you set the permission on 'app', it means the user can view ALL calendars -- permission grants are inherited | 15:30 |
| mgedmin | when you set the permission on 'app.calendar', it means the user can view just the school-wide calendar | 15:30 |
| mgedmin | so it depends on your goals | 15:31 |
| povbot | /svn/commits: * gintas committed revision 4769: | 15:32 |
| povbot | /svn/commits: Cosmetic fixes. | 15:32 |
| * mgedmin going to the office | 15:33 | |
| *** mgedmin has quit IRC | 15:33 | |
| *** tvon has joined #schooltool | 15:37 | |
| tvon | I think the issue325 changes need to be backed out before the release | 15:45 |
| tvon | The ImmutableCalendar permissions bug in issue345 will fix it | 15:45 |
| tvon | this would mean backing out the evolve3 <-> evolve4 juggling | 15:45 |
| *** tvon has quit IRC | 15:48 | |
| *** tvon has joined #schooltool | 15:48 | |
| *** tvon has quit IRC | 15:56 | |
| *** tvon has joined #schooltool | 15:56 | |
| *** tvon has quit IRC | 16:03 | |
| *** tvon has joined #schooltool | 16:03 | |
| *** jonesieboy has left #schooltool | 16:03 | |
| *** mgedmin has joined #schooltool | 16:06 | |
| bskahan | tvon: why does 325 need to be reverted? | 16:07 |
| povbot | /svn/commits: * gintas committed revision 4770: | 16:08 |
| povbot | /svn/commits: Partial fix for issue 352. The PDF views need to be updated as well. | 16:08 |
| tvon | It's excessive, we should just be setting viewCalendar on the calendar | 16:09 |
| tvon | mgedmin: ayt? | 16:09 |
| mgedmin | tvon, yep | 16:09 |
| tvon | 4579 is the same thing we were talking about last night | 16:10 |
| tvon | ignas and I that is | 16:10 |
| mgedmin | ok | 16:10 |
| mgedmin | I looked at the checkin | 16:11 |
| mgedmin | it grants permissions on the app, and then denies them for subcontainers | 16:11 |
| tvon | right | 16:11 |
| mgedmin | I found it a bit unclean | 16:11 |
| mgedmin | why do we need 'view' permission on the app just to be able to see its calendar? | 16:11 |
| mgedmin | is that just for the title? | 16:11 |
| mgedmin | or something else as well? | 16:11 |
| bskahan | mgedmin: can you have viewCalendar without also having 'view'? | 16:12 |
| mgedmin | yes, that's what it was for! | 16:12 |
| mgedmin | if I want you to be able to see my calendar, but not my home phone number, I grant you 'viewCalendar', but not 'view' | 16:12 |
| bskahan | ok | 16:13 |
| mgedmin | at least I think that was the original intent | 16:13 |
| tvon | so we should just be setting them directly on the calendar then? | 16:13 |
| *** tvon has quit IRC | 16:13 | |
| *** tvon has joined #schooltool | 16:13 | |
| bskahan | tvon: if I follow correctly now, no. We should be able to grant 'viewCalendar' on the app | 16:14 |
| mgedmin | let me think for a moment... | 16:14 |
| tvon | bskahan: but then we need to go around and remove the permission on everything else | 16:14 |
| th1a | My problem with that patch is that we aren't also giving the same permissions to authenticated users, which is confusing. | 16:14 |
| mgedmin | it would be better if we could grant 'viewCalendar' on app.calendar and avoid the inheritance mess | 16:14 |
| bskahan | +1 | 16:14 |
| th1a | +1 | 16:15 |
| tvon | then we need to pull out 4579 and the generation that goes with it | 16:15 |
| bskahan | I think the current behavior of explicitly denyToPrincipal on the containers is the correct behavior | 16:15 |
| mgedmin | let's not reshuffle everything once again this close to a release! | 16:16 |
| mgedmin | (are we close to a release?) | 16:16 |
| bskahan | even if the change to the app permisions change | 16:16 |
| th1a | I'd really like to get this out today. | 16:16 |
| bskahan | mgedmin: by having the deny set on the containers of interest, we avoid having permission on app inherit down | 16:17 |
| bskahan | we only have inheritance complications if we leave permissions unset | 16:18 |
| bskahan | which in the future we probably should avoid like the plague | 16:18 |
| povbot | /svn/commits: * gintas committed revision 4771: | 16:19 |
| povbot | /svn/commits: Remaining part of issue 352 - the PDF view (resource calendar is not accessible when it contains an event from an inaccessible calendar). | 16:19 |
| tvon | permissions are not unset | 16:19 |
| bskahan | on the containers, for many principals there's neither a deny or a grant, so the containers inherit from above | 16:20 |
| bskahan | which could get an admin in trouble if they don't realize that the permissions they set higher will inherit down | 16:21 |
| *** tvon has quit IRC | 16:24 | |
| *** tvon has joined #schooltool | 16:24 | |
| tvon | I'm moving to better wifi, brb | 16:26 |
| mgedmin | bskahan, you have a point there | 16:26 |
| * tvon thinks if we want to do that we should have event subscribers for the containers | 16:27 | |
| tvon | brb | 16:27 |
| *** tvon has left #schooltool | 16:27 | |
| *** tvon has quit IRC | 16:27 | |
| bskahan | since the admin could get [fired|heaps of trouble] for that sort of accident, unless there's a strong reason to revert it ... | 16:31 |
| povbot | /svn/commits: * gintas committed revision 4772: | 16:44 |
| povbot | /svn/commits: Backported fix for issue 352 - the PDF view (resource calendar is not accessible when it contains an event from an inaccessible calendar). | 16:44 |
| *** tvon has joined #schooltool | 16:44 | |
| povbot | /svn/commits: * srichter committed revision 4773: | 16:49 |
| povbot | /svn/commits: Converted tests to new setup framework. | 16:49 |
| povbot | /svn/commits: * gintas committed revision 4774: | 16:55 |
| povbot | /svn/commits: Backported rev 4716 (missing tests in rev 4712). | 16:55 |
| povbot | /svn/commits: * gintas committed revision 4775: | 16:56 |
| povbot | /svn/commits: Fixed failing functional test. | 16:56 |
| bskahan | I'm trying to fix the access to the top level calendar for authenticated users | 16:58 |
| bskahan | schoolbell.app.security.personPermissionsSubscriber sets the default permissions for new users | 16:59 |
| povbot | /svn/commits: * gintas committed revision 4776: | 16:59 |
| povbot | /svn/commits: Backported revision 4706: modified the section membership subscriber to grant/unset permissions on 'view' and 'viewCalendar'. | 16:59 |
| bskahan | but I need to get the PermissionManager from the app, which works fine normally | 16:59 |
| bskahan | app_permissions = IPrincipalPermissionManager(getSchoolBellApplication()) | 16:59 |
| bskahan | except on bootstrap, it can't get the app when it adds the manager | 17:00 |
| gintas | bskahan, I will be tagging a release within an hour probably | 17:00 |
| gintas | if you want to get some changes in, you'll have to hurry (unless the changes are important, then I'll wait) | 17:00 |
| bskahan | gintas: I think this one is important | 17:01 |
| bskahan | but it's quick | 17:01 |
| bskahan | can I try/except on the ValueError in the above case? | 17:01 |
| mgedmin | bskahan, you could use setSite in your bootstrap code | 17:01 |
| tvon | bskahan: you have to call setSite before the event is fired in the bootstrap... I believe | 17:01 |
| bskahan | thanks | 17:02 |
| mgedmin | those of you who get functional test failures in schooltool, svn up Zope 3 | 17:21 |
| mgedmin | I wish I knew how to tell buildbot to do that | 17:21 |
| mgedmin | previously the dreaded svn:externals took care of everything | 17:21 |
| bskahan | gintas: I turns out its not as quick as I though | 17:23 |
| bskahan | t | 17:23 |
| gintas | bskahan, how much time do you need? | 17:25 |
| bskahan | gintas: I'm not going to have enough time | 17:26 |
| bskahan | If I get it done before you release, good. But I don't want to hold you up at this point | 17:27 |
| gintas | bskahan, how long? | 17:28 |
| bskahan | gintas: I'm going to have to stop | 17:40 |
| bskahan | can't get the site setup to work properly in all tests and I'm out of time | 17:40 |
| gintas | OK | 17:41 |
| povbot | /svn/commits: * gintas committed revision 4777: | 17:42 |
| povbot | /svn/commits: Updated Debian packaging to make /var/lib/school* not world-readable. | 17:42 |
| *** jinty has joined #schooltool | 17:45 | |
| gintas | hi jinty | 17:45 |
| gintas | I'm about to tag the release | 17:45 |
| jinty | cool | 17:45 |
| jinty | can I just have a look what's been done first? | 17:46 |
| gintas | sure | 17:46 |
| gintas | I still have to update some docs, but that shouldn't take long | 17:47 |
| povbot | /svn/commits: * gintas committed revision 4778: | 17:47 |
| povbot | /svn/commits: Bumping version number. | 17:47 |
| povbot | /svn/commits: * gintas committed revision 4779: | 17:47 |
| povbot | /svn/commits: * gintas committed revision 4780: | 17:57 |
| povbot | /svn/commits: Removed a fair chunk of RELEASE.txt. | 17:57 |
| povbot | /svn/commits: * gintas committed revision 4781: | 17:58 |
| povbot | /svn/commits: Removed a fair chunk from RELEASE.txt. | 17:58 |
| mgedmin | interesting... who ate the checkin message for rev 4779? | 17:59 |
| mgedmin | povbot, apparently | 17:59 |
| povbot | mgedmin: Error: "apparently" is not a valid command. | 17:59 |
| mgedmin | and where is povbuildbot? | 18:00 |
| th1a | sleeping, apparently. | 18:00 |
| gintas | povbot, self-destruct | 18:00 |
| povbot | gintas: Error: "self-destruct" is not a valid command. | 18:00 |
| jinty | gintas: no-issues, though I will re-implement the debian stuff. | 18:01 |
| jinty | gintas: I want to back-port the "Installing with distutils" warning | 18:02 |
| gintas | jinty, that's fine, I'm trying to get the job done without diverging from what we have right now | 18:02 |
| gintas | jinty, go ahead | 18:02 |
| *** bskahan has quit IRC | 18:04 | |
| povbot | /svn/commits: * jinty committed revision 4782: | 18:05 |
| povbot | /svn/commits: Back-port the distutils warning. | 18:05 |
| povbot | /svn/commits: * jinty committed revision 4783: | 18:06 |
| povbot | /svn/commits: Typo. | 18:07 |
| povbot | /svn/commits: * jinty committed revision 4784: | 18:10 |
| povbot | /svn/commits: Back-port schoolbell distutils warning. | 18:10 |
| jinty | gintas: done. | 18:10 |
| *** jinty has quit IRC | 18:10 | |
| povbot | /svn/commits: * gintas committed revision 4785: | 18:13 |
| povbot | /svn/commits: I was a bit too brutish in the last commit. | 18:13 |
| povbot | /svn/commits: * gintas committed revision 4786: | 18:15 |
| povbot | /svn/commits: Added some more info to RELEASE.txt. | 18:15 |
| mgedmin | a bee tries to prevent the schooltool release by scaring gintas away from his laptop :-) | 18:24 |
| tvon | haha | 18:28 |
| povbot | /svn/commits: * gintas committed revision 4787: | 18:29 |
| povbot | /svn/commits: Note about security. | 18:29 |
| povbot | /svn/commits: * gintas committed revision 4788: | 18:29 |
| povbot | /svn/commits: Security note. | 18:29 |
| povbot | /svn/commits: * gintas committed revision 4789: | 18:30 |
| povbot | /svn/commits: Backported security warning. | 18:30 |
| povbot | /svn/commits: * gintas committed revision 4790: | 18:30 |
| *** gintas has quit IRC | 18:32 | |
| povbot | /svn/commits: * gintas committed revision 4791: | 18:46 |
| povbot | /svn/commits: Tagging SchoolBell 1.2. | 18:46 |
| povbot | /svn/commits: * gintas committed revision 4792: | 18:47 |
| povbot | /svn/commits: Tagging SchoolTool 0.11. | 18:47 |
| povbot | /svn/commits: * gintas committed revision 4793: | 18:48 |
| povbot | /svn/commits: Bound ST 0.11 to SB 1.2. | 18:48 |
| mgedmin | there's one last chance to fix http://issues.schooltool.org/issue233, if somebody bothers to merge the relevant zope 3 bugfix into Zope-3.1 branch | 18:49 |
| mgedmin | (zope 3 issue 410, bugfix in revision 30458.) | 18:50 |
| *** gintas has joined #schooltool | 18:50 | |
| mgedmin | gintas: | 18:50 |
| mgedmin | <mgedmin> there's one last chance to fix http://issues.schooltool.org/issue233, if somebody bothers to merge the relevant zope 3 bugfix into Zope-3.1 branch | 18:50 |
| mgedmin | <mgedmin> (zope 3 issue 410, bugfix in revision 30458.) | 18:50 |
| mgedmin | you made the fix in zope 3 trunk | 18:51 |
| gintas | hmm | 18:55 |
| gintas | the version of Zope3 we're releasing with is a specific Zope3 revision | 18:55 |
| gintas | not the branch | 18:55 |
| gintas | of course we can change that | 18:55 |
| mgedmin | ah | 18:57 |
| mgedmin | I don't know how important issue 233 is | 18:57 |
| mgedmin | probably not very | 18:57 |
| srichter | well, I think it is a terrible idea to release based on a specific revision | 18:59 |
| gintas | just yesterday some change in Zope 3.1 broke SB/ST trunk | 19:00 |
| gintas | I don't want to keep track of such changes and possibly introduce subtle changes, in say, release candidates | 19:00 |
| mgedmin | gintas, so if I check out the release branch, and type 'make', I will get a specific Zope 3 revision? | 19:03 |
| mgedmin | I think not | 19:03 |
| gintas | right, you will get Zope-3.1 | 19:05 |
| gintas | I guess I should have changed that part in the Makefile | 19:05 |
| povbot | /svn/commits: * gintas committed revision 4794: | 19:08 |
| povbot | /svn/commits: Use x.y.z for versioning. | 19:08 |
| povbot | /svn/commits: * gintas committed revision 4795: | 19:09 |
| povbot | /svn/commits: Use x.y.z versioning scheme. | 19:09 |
| gintas | oh, btw, the current release is bound to Zope 3 trunk | 19:10 |
| gintas | which means that the fix will be included | 19:10 |
| gintas | so there's no problem at all | 19:10 |
| mgedmin | I suddenly do not want to know anything | 19:17 |
| tiredbones | srichter, I thought you would like to know I order your book. Where do I pick up my commission? | 20:05 |
| srichter | good! :-) | 20:07 |
| *** gintas has quit IRC | 20:25 | |
| * mgedmin wishes srichter made smaller checkins | 20:29 | |
| mgedmin | e.g. one for adding the new testing setup infrastructure, then another that converts tests to it | 20:29 |
| mgedmin | now I have a 1200 line diff (described as "these very few lines of code have a lot of potential!") that does one and some of the other | 20:30 |
| mgedmin | otoh that diff splits nicely, along file boundaries | 20:39 |
| mgedmin | it's just that this wasn't immediatelly apparent | 20:39 |
| *** pcardune has joined #schooltool | 20:48 | |
| mgedmin | checkouts of various schooltool/bell branches ate all my disk space! | 21:18 |
| *** th1a has quit IRC | 21:20 | |
| srichter | mgedmin: I am trying to keep checkins as small as possible, but I tend not to checkin new code without using it; I just want to test it for while myself. | 21:25 |
| mgedmin | which is a good thing | 21:25 |
| srichter | btw, I think it might be better to just merge the trunk into the branch and then copy the branch over to the trunk | 21:25 |
| mgedmin | why? | 21:26 |
| srichter | I wonder how hard/easy it is to merge the branch by revisions to the trunk | 21:26 |
| mgedmin | probably hard | 21:27 |
| mgedmin | :( | 21:27 |
| srichter | yes | 21:27 |
| srichter | that's what I remember from my blow-services work | 21:28 |
| srichter | Once I completed the merge between SB & ST I will try to merge some trunk revisions to the branch in the hope it will be easier to merge | 21:28 |
| *** pcardune has left #schooltool | 21:37 | |
| *** mgedmin has quit IRC | 22:29 | |
| *** dwoo has joined #schooltool | 22:47 | |
| *** dwoo has quit IRC | 23:15 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!