MoinMoin: MoinMoinChat/Logs/moin-dev/2009-07-05

2009-07-05T02:23:55  *** amartani has joined #moin-dev
2009-07-05T06:40:58  <CIA-38> Chintalagiri Shashank <shashank.chintalagiri@gmail.com> default * 4753:b15d2cf6f596 1.9-adminui-cshashank/MoinMoin/script/install/installer.py: installer skeleton
2009-07-05T06:41:00  <CIA-38> Chintalagiri Shashank <shashank.chintalagiri@gmail.com> default * 4754:43358814fb7d 1.9-adminui-cshashank/MoinMoin/script/install/ (defaults.py installer.py): PEP8 compliance changes
2009-07-05T06:41:03  <CIA-38> Chintalagiri Shashank <shashank.chintalagiri@gmail.com> default * 4761:de1a44455121 1.9-adminui-cshashank/MoinMoin/ (6 files in 5 dirs): Merged with main
2009-07-05T08:25:13  *** amartani has quit IRC
2009-07-05T10:14:50  <dreimark> moin
2009-07-05T10:16:12  <dreimark> that paragraph commenting feature of http://hgbook.red-bean.com is nice
2009-07-05T10:28:49  *** grzywacz has joined #moin-dev
2009-07-05T12:01:24  <devilsadvocate> did something happen recently to the pep8 tests? :/
2009-07-05T12:02:17  <dreimark> devilsadvocate: what do you mean?
2009-07-05T12:03:00  <devilsadvocate> trying to run "py.test tes_sourcecode.py", it isnt checking the whole tree for me as it used to. Its checking just 8 files (and failing in 3 of those)
2009-07-05T12:03:28  <devilsadvocate> i did a merge with 1.9 main today. i hope i didnt break the merge again :\
2009-07-05T12:03:32  <dreimark> well it knows new files
2009-07-05T12:03:52  <devilsadvocate> 8 files. it used to do 400 something
2009-07-05T12:04:19  <dreimark> tests finished: 5 passed, 3 failed in 1.71 seconds
2009-07-05T12:04:19  * devilsadvocate tries to run it on a 1.9 clone
2009-07-05T12:04:27  <devilsadvocate> yep
2009-07-05T12:04:29  <dreimark> in your repo
2009-07-05T12:04:39  <dreimark> you should run the tests before commitment
2009-07-05T12:05:04  <devilsadvocate> i did :\
2009-07-05T12:05:49  <devilsadvocate> the commit today was just a merge before i started putting my code in (i've been working outside the repo, outside pep8, to get some things straight)
2009-07-05T12:06:28  <dreimark> ok, they ar not yours, I guess they are left over from the sprint in birmingham
2009-07-05T12:07:20  <devilsadvocate> why is it failing though? is it designed to exit after 3 files fail? 1.9 main gives me 3 failed, 3 passed
2009-07-05T12:07:57  <dreimark> it depends on the time see line 61 of the test script
2009-07-05T12:08:03  <dreimark> #RECENTLY = 0 # check everything!
2009-07-05T12:08:39  <devilsadvocate> ah
2009-07-05T12:10:48  <devilsadvocate> right. so that should mean my files are being tested.
2009-07-05T12:49:14  <dennda> ThomasWaldmann: seen my requests/questions?
2009-07-05T12:54:34  <dreimark> dennda: he is travelling today
2009-07-05T13:08:35  <dennda> ah
2009-07-05T13:41:03  <dennda> TheSheep: I think you broke 2.0-storage
2009-07-05T13:41:39  <dennda> TheSheep: http://paste.pocoo.org/show/126750/
2009-07-05T13:43:42  <dennda> fix that please
2009-07-05T13:47:48  <dreimark> has anyone an idea if this results of
2009-07-05T13:48:03  <dreimark> py.test --boxed are realistic?
2009-07-05T13:48:37  <dreimark> xorAxAx: ThomasWaldmann ^^^
2009-07-05T13:48:49  <dreimark> also why it seems to keep the state of the brokeness
2009-07-05T13:49:03  <dreimark> tests finished: 1722 passed, 23 failed, 75 skipped in 28.89 seconds
2009-07-05T13:49:17  <xorAxAx> why shouldnt they be realistic?
2009-07-05T13:49:54  <dreimark> I found some discussions that some people think --boxed has as bug
2009-07-05T13:53:20  <dreimark> e.g. test_wikisync.py[1] F test_wsgiapp.py[7] ....FFF  test_search.py[12] .....FFF.F.F
2009-07-05T13:53:47  <xorAxAx> well, look at the failures
2009-07-05T13:53:51  <xorAxAx> to decide
2009-07-05T14:08:36  <dreimark> looks like that in --boxed mode page deletion isn't working
2009-07-05T14:08:49  <dreimark> AccessDenied: You are not allowed to delete this page!
2009-07-05T14:30:06  <dreimark> .oO at least some of the tests are broken
2009-07-05T14:31:22  * dreimark fixes 1.9 tests
2009-07-05T16:22:53  <CIA-38> Reimar Bauer <rb.proj AT googlemail DOT com> default * 4752:e6190a499a7e 1.9/MoinMoin/ (3 files in 3 dirs): fixed permissions for nuke_page by adding become_trusted()
2009-07-05T16:25:02  * dreimark wonders why --boxed was needed to figure this
2009-07-05T17:14:33  <dreimark> with boxed I have 10 failing tests in test_security
2009-07-05T17:22:27  *** amartani has joined #moin-dev
2009-07-05T17:28:53  *** amartani has quit IRC
2009-07-05T17:37:41  *** amartani has joined #moin-dev
2009-07-05T17:43:06  *** czarny has joined #moin-dev
2009-07-05T17:45:14  *** czarny has quit IRC
2009-07-05T17:48:48  *** amartani has quit IRC
2009-07-05T17:49:01  *** amartani has joined #moin-dev
2009-07-05T17:57:18  <amartani> meeting soon?
2009-07-05T17:57:54  <dreimark> yes
2009-07-05T18:02:29  <dreimark> amartani: devilsadvocate  dennda dimazest xorAxAx ThomasWaldmann waldi meeting!
2009-07-05T18:03:16  <dreimark> amartani: waldi do you want to start ?
2009-07-05T18:04:27  <amartani> ok.
2009-07-05T18:04:27  <amartani> LAST: Bugfixing, improving handling of editors, faster saving.
2009-07-05T18:04:29  <amartani> NEXT: Preview on save, support for choosing editing system.
2009-07-05T18:04:31  <amartani> BLOQUERS: none.
2009-07-05T18:05:02  <dreimark> waldi: do you be with us?
2009-07-05T18:05:32  <dreimark> amartani: I have filed two issues
2009-07-05T18:06:03  <dreimark> http://moinmo.in/RealTimeEditing2009/2009-07-03
2009-07-05T18:06:35  <amartani> dreimark: yes, I've read it. This is happening on original mobwrite too, so I have to talk to its devs about it.
2009-07-05T18:06:35  <dreimark> we can talk on that later
2009-07-05T18:06:43  <dreimark> ok
2009-07-05T18:07:06  <dreimark> next is devilsadvocate (if you are with us)
2009-07-05T18:09:13  <dreimark> amartani: thanks for reporting
2009-07-05T18:09:43  <dreimark> who else is here dennda ? dimazest ? devilsadvocate ?
2009-07-05T18:10:22  <dreimark> ( dimazest is probably on a long trip from bermingham back to home )
2009-07-05T18:13:20  <dreimark> ok since noone else is alive currently we postpone missing student reports
2009-07-05T18:14:19  <dreimark> amartani: you posted last week about: NEXT: Make it possible to choose between regular editor and mobwrite.
2009-07-05T18:14:23  <dreimark> Is this finished?
2009-07-05T18:15:41  <amartani> no, I have been fixing bugs and improving things a bit before doing this
2009-07-05T18:16:22  <amartani> and also I am not sure how should I do this in a clean way using the Items idea..
2009-07-05T18:17:45  <dreimark> please add a wiki page about your ideas or what you need.
2009-07-05T18:20:02  <amartani> dreimark: sure. was there any plan on how the graphical editor will be integrated to 2.0?
2009-07-05T18:21:04  <dreimark> if it can't be combined with mobwrite (in time and easy) we need to add it similiar to the old approach
2009-07-05T18:22:05  <dreimark> for now  an easy ui could be "You can either create the item from scratch or select a template. [[mobwrite]] [[Simple]] [[GUI]]"
2009-07-05T18:38:18  <dennda> sorry for being late
2009-07-05T18:39:20  <dreimark> dennda: then please continue. ThomasWaldmann will read it later after he is back at home
2009-07-05T18:41:53  <dennda> LAST: data_backend and user_backend are no more. A wiki admin now only specifies storage = get_enduser_backend(...) and the correct backends are created automatically. Made Items aware of their full name.
2009-07-05T18:42:36  <dennda> NEXT: Finnish routerbackend, finnish acl integration
2009-07-05T18:43:32  <dennda> BLOCKERS: TheSheep broke my repo!!! No other blockers except it would be nice if we had gotten more input for the acl stuff. Oh and I asked a few questions that weren't yet answered. No real blockers though (except that sheepish sabotage)
2009-07-05T18:43:53  <xorAxAx> dennda: your tasks in LAST sound like a one day job, right?
2009-07-05T18:45:22  <dennda> well I also fixed all storage related tests and some other fixes but yes, you're right
2009-07-05T18:46:26  <dennda> I didn't have that much time this week due to a difficult exam for which I needed to prepare
2009-07-05T18:46:33  <dennda> next week will be better though
2009-07-05T18:48:28  <dreimark> dennda: this is reflected in missing diary pages
2009-07-05T18:48:58  <dreimark> can you just add some lines there also when you have to do some exams
2009-07-05T18:49:12  <dennda> yes, sorry
2009-07-05T18:50:19  <dreimark> xorAxAx: dennda anything else to say ?
2009-07-05T18:50:28  <dennda> No
2009-07-05T18:51:41  <dennda> is it ok if I backout TheSheep's changeset until he has time to fix it?
2009-07-05T18:51:53  <dreimark> ok thanks for reporting.
2009-07-05T18:52:20  <xorAxAx> dennda: yes, thats fine
2009-07-05T18:52:21  <dreimark> dennda: yes, please give him also a note by mail/wiki
2009-07-05T18:52:53  <dennda> i pinged him already, he'll notice
2009-07-05T18:52:54  <dreimark> he stays next week in birmingham for vacation.
2009-07-05T18:53:16  <dreimark> devilsadvocate: ping
2009-07-05T19:00:01  <dreimark> the meeting is finished.  thank you for your attendance
2009-07-05T19:00:32  <dreimark> dimazest: devilsadvocate please post your lines after you are back
2009-07-05T19:02:37  <xorAxAx> devilsadvocate: can you explain me why your projects page has been unrelated to your soc task since april?
2009-07-05T19:34:38  <amartani> is it easy to execute wikiserver on windows?
2009-07-05T19:37:57  <dreimark> amartani: yes
2009-07-05T19:38:06  <dreimark> you need only python installed
2009-07-05T19:39:45  <dreimark> me uses sometimes http://www.voidspace.org.uk/python/movpy/distributions.html
2009-07-05T19:41:19  <dreimark> http://code.google.com/p/movable-python/downloads/list
2009-07-05T19:42:01  <amartani> dreimark: thanks!
2009-07-05T19:52:59  *** czarny has joined #moin-dev
2009-07-05T20:40:41  <ThomasWaldmann> re
2009-07-05T21:00:54  <ThomasWaldmann> amartani: http://hg.moinmo.in/moin/2.0-storage-editor-amartani/rev/e332596d29f9 <- why?
2009-07-05T21:04:06  <amartani> in mobwrite docs, it says to use names only with certain characters, which doesn't include, ex. spaces (although it seems to work, but can stop working later).
2009-07-05T21:05:05  <amartani> and I had just before worked on a bug with ":" character, so I though it was better to avoid other cases like this..
2009-07-05T21:05:51  <amartani> also, caching does this
2009-07-05T21:06:06  <ThomasWaldmann> ok
2009-07-05T21:06:21  <ThomasWaldmann> see my comment on the wiki btw
2009-07-05T21:16:23  <ThomasWaldmann> amartani: http://hg.moinmo.in/moin/2.0-storage-editor-amartani/rev/410fb7b2fa83 why dict() and not cleaned_editors?
2009-07-05T21:16:49  <ThomasWaldmann> and why do you use dict() and not {}
2009-07-05T21:17:05  <amartani> ThomasWaldmann: Idea is good. But "share this" seems like it will open some "buddy list" to choose someone to share..
2009-07-05T21:18:24  <ThomasWaldmann> [ ] collaborate
2009-07-05T21:19:45  *** czarny has quit IRC
2009-07-05T21:20:51  <amartani> ThomasWaldmann: well, I just thought to be sure to be a clean dict. and I think an empty dict as {} is ugly.
2009-07-05T21:21:11  <amartani> ThomasWaldmann: sorry, I have to go now, will come back later
2009-07-05T21:21:14  *** amartani has quit IRC
2009-07-05T21:21:31  *** devilsadvocate has quit IRC
2009-07-05T21:21:57  *** devilsadvocate has joined #moin-dev
2009-07-05T21:28:44  <ThomasWaldmann> dimazest: http://hg.moinmo.in/moin/1.9-groups-dmilajevs/rev/fc0c70fe84f3 you review the stuff that was commented out?
2009-07-05T21:33:58  <ThomasWaldmann> dimazest: http://hg.moinmo.in/moin/1.9-groups-dmilajevs/rev/cf3c81f67862#l2.30
2009-07-05T21:34:07  <ThomasWaldmann> http://hg.moinmo.in/moin/1.9-groups-dmilajevs/rev/cf3c81f67862#l2.32
2009-07-05T21:34:24  <ThomasWaldmann> http://hg.moinmo.in/moin/1.9-groups-dmilajevs/rev/cf3c81f67862#l2.22
2009-07-05T21:35:09  <ThomasWaldmann> s/group/dict/
2009-07-05T21:35:42  <ThomasWaldmann> http://hg.moinmo.in/moin/1.9-groups-dmilajevs/rev/cf3c81f67862#l2.6 < that docstring should be a bit more verbose (in all composite backends)
2009-07-05T21:56:31  <dennda> ThomasWaldmann: welcome back :-)
2009-07-05T21:59:16  <dennda> ThomasWaldmann: Please take a look at the questions I put at 22:44 and 01:41
2009-07-05T21:59:30  <dennda> when you find the time
2009-07-05T22:10:53  <ThomasWaldmann> dennda: i am working on some bigger related change. aside from that, you think/test on your own, right?
2009-07-05T22:11:47  <ThomasWaldmann> (that was about the backend cloning stuff)
2009-07-05T22:12:27  <dennda> what bigger related change?
2009-07-05T22:17:09  <ThomasWaldmann> making the tests work better, deserialize the needed pages from xml, minimize the needed pages
2009-07-05T22:17:18  <ThomasWaldmann> dennda: simultaneaosly < typo
2009-07-05T22:17:29  <dreimark> wb ThomasWaldmann
2009-07-05T22:19:44  <ThomasWaldmann> dennda: simultaneosly < typo
2009-07-05T22:19:58  <dennda> ok i guess I will look that word up :-)
2009-07-05T22:24:08  <ThomasWaldmann> dennda: test_history: i don't get what you mean
2009-07-05T22:25:02  <dennda> yes that's a little complicated
2009-07-05T22:25:16  <dennda> the test checks if history provides a backends rev in the proper order
2009-07-05T22:25:34  <dennda> they are ordered by their unix timestamp
2009-07-05T22:25:43  <dennda> which is based on seconds since the epoch
2009-07-05T22:25:57  <dennda> now the test creates the items, all in less than a second
2009-07-05T22:26:09  <dennda> that means several items have the same timestamp
2009-07-05T22:26:48  <ThomasWaldmann> and?
2009-07-05T22:26:50  <dennda> usually that is not a problem (at least in memory backend), since the items are internally stored in a datastructure that preserves order (completely unrelated to the timestamp) and hence the order is kept
2009-07-05T22:27:08  <dennda> for the routerbackend, where several such backends are in use, that is not the case, however, so that test failed
2009-07-05T22:27:34  <dennda> and I don't see a way to properly "fix" that
2009-07-05T22:28:38  <dennda> i dont even think we need to
2009-07-05T22:29:58  <ThomasWaldmann> i just wanted to ask about what the problem is, in practice
2009-07-05T22:30:15  <dennda> well, i *think* there is none
2009-07-05T22:32:16  <ThomasWaldmann> aside from that, why is the timestamp a int?
2009-07-05T22:32:51  <dennda> what else should it be?
2009-07-05T22:33:00  <dennda> if we are talking about unix timestamps
2009-07-05T22:40:25  <ThomasWaldmann> print time.time()
2009-07-05T22:44:55  <dennda> oh
2009-07-05T22:45:05  <dennda> hm, no idea
2009-07-05T22:53:10  <ThomasWaldmann> well, you have designed the storage api, so get an idea :)
2009-07-05T22:54:08  <dennda> well I at least didn't know that it's more granular than seconds
2009-07-05T22:54:30  <dennda> so we could use those more precise representation as well
2009-07-05T22:54:51  <dennda> that would not solve the testcase, though
2009-07-05T22:56:53  <ThomasWaldmann> why?
2009-07-05T23:00:19  <dennda> because then the testcase would always fail if more than one item is created in less than 0.01 seconds. that is a) dependent on the hardware you're running the test on and b) not really easily reproducable
2009-07-05T23:03:38  <ThomasWaldmann> yeah, but you don't need to sleep 1s then
2009-07-05T23:04:54  <dennda> yes :-)
2009-07-05T23:05:01  <dennda> you need to sleep 0.01s though
2009-07-05T23:05:09  <dennda> which would speed up tests at least
2009-07-05T23:13:11  <dennda> ThomasWaldmann: I guess we won't get any further input on the acl privilege side of things, unfortunately
2009-07-05T23:53:18  *** amartani has joined #moin-dev