MoinMoin: MoinMoinChat/Logs/moin-dev/2009-06-06

2009-06-06T00:04:26  <CIA-19> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 5898:48ef381a7108 2.0-storage/MoinMoin/storage/backends/flatfile.py: storage: flatfile backend: more cleanup, small fixes
2009-06-06T00:04:28  <CIA-19> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 5899:c9494c600cc3 2.0-storage/MoinMoin/user.py: user.py: fix undefined _
2009-06-06T00:29:45  <CIA-19> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 5900:3757c6434dfc 2.0-storage/MoinMoin/wikiutil.py: wikiutil: remove deprecated functions
2009-06-06T01:44:46  <CIA-19> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 5901:b1449cfbe785 2.0-storage/MoinMoin/storage/ (4 files in 2 dirs): storage: rename fs17 backend to fs19
2009-06-06T01:57:33  *** grzywacz has quit IRC
2009-06-06T02:06:34  <CIA-19> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 5902:3ab0c9c36128 2.0-storage/MoinMoin/storage/backends/fs19.py: storage: fs19 cosmetic changes
2009-06-06T03:03:30  *** rkerr has quit IRC
2009-06-06T04:54:51  <CIA-19> Alexandre Martani <amartani AT gmail DOT com> default * 5896:76fd4e2d8df8 2.0-storage-editor-amartani/MoinMoin/ (4 files in 4 dirs): Real-time editor: display users editing the page
2009-06-06T04:55:03  <CIA-19> Alexandre Martani <amartani AT gmail DOT com> default * 5923:2fb258c4e6a0 2.0-storage-editor-amartani/MoinMoin/ (21 files in 9 dirs): Merge with 2.0-storage
2009-06-06T06:12:53  *** amartani has quit IRC
2009-06-06T06:30:18  *** rkerr has joined #moin-dev
2009-06-06T06:37:41  *** rkerr has quit IRC
2009-06-06T08:01:22  *** rkerr has joined #moin-dev
2009-06-06T08:20:36  *** rkerr has quit IRC
2009-06-06T10:35:37  *** grzywacz has joined #moin-dev
2009-06-06T11:23:57  <ThomasWaldmann> moin
2009-06-06T12:19:14  <dreimark> moin
2009-06-06T15:12:24  * ThomasWaldmann plays with the new moinmo.in server
2009-06-06T15:25:11  * ThomasWaldmann runs badblocks
2009-06-06T16:32:42  <dimazest> dreimark: old code has this line if has_member(groupname, username) and has_member('AutoAdminGroup', groupname):
2009-06-06T16:33:01  <dimazest> which i changed to the if username in groups[groupname] and groupname in groups['AutoAdminGroup']:
2009-06-06T16:35:08  <dimazest> but i think if username in groups['AutoAdminGroup']: should be enought
2009-06-06T16:49:42  *** rkerr has joined #moin-dev
2009-06-06T17:15:03  *** amartani has joined #moin-dev
2009-06-06T17:20:50  <ThomasWaldmann> dimazest: the last one is enough if recursive groups work
2009-06-06T17:21:24  <ThomasWaldmann> you would just check for AutoAdminGroup membership then
2009-06-06T17:33:41  * ThomasWaldmann reads groups/*
2009-06-06T17:34:21  <dimazest> ThomasWaldmann: that groups are little bit outdated
2009-06-06T17:34:43  <dimazest> i removed too much from the wikidicts
2009-06-06T17:42:03  *** amartani has quit IRC
2009-06-06T17:42:21  *** amartani has joined #moin-dev
2009-06-06T17:44:47  <ThomasWaldmann> dimazest: btw, where are the results of the diff we gave you?
2009-06-06T17:47:13  <ThomasWaldmann> dimazest: backends/__init__.py needs a docstring, see MoinMoin/_template.py
2009-06-06T17:54:11  <ThomasWaldmann> dimazest: didn't we already have a "config groups" backend? where is it?
2009-06-06T17:56:59  <ThomasWaldmann> dimazest: why do the wiki group tests have become_trusted?
2009-06-06T17:58:48  <ThomasWaldmann> dimazest: can we please have a bit more realistic and easier to follow tests? I mean it is a bit unlikely that you have Admin, Editor and Apple in the same group.
2009-06-06T18:05:42  <dimazest> :) ok
2009-06-06T18:12:27  <ThomasWaldmann> i don't think ":) ok" is a sufficient answer.
2009-06-06T18:21:33  <dimazest> i'm working on it
2009-06-06T18:22:03  <ThomasWaldmann> and btw, there is no such thing as a "unicode string"
2009-06-06T18:22:54  <ThomasWaldmann> python (2.x) has 2 related data types: str and unicode
2009-06-06T18:23:38  <ThomasWaldmann> and str is usually called "a string"
2009-06-06T18:24:30  <ThomasWaldmann> (luckily, that'll change with Python 3)
2009-06-06T18:25:43  <ThomasWaldmann> so, if you mean unicode, just say unicode or unicode object, but not unicode string.
2009-06-06T18:39:15  <ThomasWaldmann> dimazest:      3.9 +   * request.dicts was removed, use request.groups instead.
2009-06-06T18:39:58  <ThomasWaldmann> you have seen that a Dict has a different functionality to a Group? And also different pagenames?
2009-06-06T18:52:29  <ThomasWaldmann> amartani: can you please add more comments to your code about what it is doing and how?
2009-06-06T18:55:07  <dimazest> ThomasWaldmann: that's why now i'm doing it in the different way, thats why, i asked if it is possible to change history by removing some changesets
2009-06-06T19:02:02  <ThomasWaldmann> you should tell to what of my 10 unanswered posts you are referring
2009-06-06T19:20:12  <ThomasWaldmann> dennda: save-hook: yes, that was used by antispam secpol. but I removed that code. not sure what else uses it, use grep. :)
2009-06-06T19:23:00  <dennda> save-hook?
2009-06-06T19:23:12  <ThomasWaldmann> dennda: btw, be more careful with ACLs. In 1.9, having #acl <and nothing behind> was something different than not having any #acl ...
2009-06-06T19:23:22  <ThomasWaldmann> dennda: SecPol.save
2009-06-06T19:23:30  <dennda> ah
2009-06-06T19:23:34  <dennda> referring to my comment
2009-06-06T19:24:02  <dennda> I'll take a look once I've finished my university stuff
2009-06-06T19:24:55  <ThomasWaldmann> dennda: http://hg.moinmo.in/moin/2.0-storage/rev/01ec0110f44f#l1.80 < sub-optimal
2009-06-06T19:32:19  <dennda> ThomasWaldmann: --v :-)
2009-06-06T19:37:00  <ThomasWaldmann> dennda: i guess you are beyond learning programming basics, so look yourself
2009-06-06T19:41:57  <dennda> ThomasWaldmann: do you mean the fact that I bound may to the seemingly internal _may method of the backend object?
2009-06-06T19:43:00  <ThomasWaldmann> no, but you don't need "may" if you raise
2009-06-06T19:44:31  <dennda> ah
2009-06-06T19:44:50  <dennda> Thought you were talking about something serious :D
2009-06-06T19:46:36  <ThomasWaldmann> i said "sub-optimal", not "wrong".
2009-06-06T19:48:16  <dennda> you're right of course
2009-06-06T19:48:51  * ThomasWaldmann checks the storage page
2009-06-06T19:49:52  <CIA-19> Christopher Denter <moin GUESSWHAT the DASH space DASH station ROUNDTHING com> default * 5903:437c5853ba68 2.0-storage/MoinMoin/security/__init__.py: storage: Don't unnecessarily assign may
2009-06-06T19:50:02  <ThomasWaldmann> please checkmark the stuff already done on Storage2009
2009-06-06T19:50:56  <dennda> aye
2009-06-06T19:55:44  <dennda> I don't see any storage related test failures other than the hg ones and perhaps TestUnsafeSync().testBasicTagThings
2009-06-06T19:55:47  <dennda> agree?
2009-06-06T19:58:37  <ThomasWaldmann> please check the default / inherited acl test cases if you see something missing respective to '' vs. None
2009-06-06T19:59:27  <ThomasWaldmann> is the difference clear from my recent comments?
2009-06-06T19:59:57  <dennda> The difference between default / inherited ACL tests?
2009-06-06T20:01:30  <ThomasWaldmann> no, between empty and no acl
2009-06-06T20:02:26  <dennda> I guess if there's no #acl, one might add one without being admin. If there is one, even if it's '', one might not change it unless he's ADMIN
2009-06-06T20:02:29  <dennda> Is that correct?
2009-06-06T20:02:50  <ThomasWaldmann> no
2009-06-06T20:03:25  <dennda> Then I need to check your comments.
2009-06-06T20:03:32  <ThomasWaldmann> without admin rights, you may not change acls. having none -> having one == change.
2009-06-06T20:03:36  <dennda> Still learning for one of my courses... :-)
2009-06-06T20:03:47  <dennda> ok
2009-06-06T20:04:18  <dennda> So you basically always need admin unless your revision's acl didn't change
2009-06-06T20:04:29  <dennda> (compared to the last rev's)
2009-06-06T20:04:46  <ThomasWaldmann> yes
2009-06-06T20:05:16  <ThomasWaldmann> but what i meant is:
2009-06-06T20:05:26  <ThomasWaldmann> #acl
2009-06-06T20:05:28  <ThomasWaldmann> vs.
2009-06-06T20:05:31  <ThomasWaldmann> <nothing>
2009-06-06T20:05:57  <ThomasWaldmann> (^^ example using 1.9 way)
2009-06-06T20:08:02  <dennda> Isn't that the same? What's the difference? #acl == '', <nothing> == None
2009-06-06T20:09:32  <ThomasWaldmann> the acl behaviour definition tells, that if there is no acl on a page, it uses default acl instead
2009-06-06T20:10:11  <ThomasWaldmann> but if you give "#acl", you have an acl. that acl does not give any permissions to anyone.
2009-06-06T20:10:43  <ThomasWaldmann> and that's likely different from the default acl
2009-06-06T20:11:09  <dennda> Do we want it that way?
2009-06-06T20:11:22  <dennda> i.e., keep the 1.9 behavior?
2009-06-06T20:11:22  <ThomasWaldmann> similar is true for hierarchic acl inheritance: it only inherits if there is nothing, if there is some acl, it just uses that.
2009-06-06T20:12:08  <ThomasWaldmann> how else would you remove default acl permissions from a page and give no permissions?
2009-06-06T20:13:19  <dennda> Convinced :-)
2009-06-06T20:15:15  <ThomasWaldmann> (btw, autoadmin SecPol still uses save)
2009-06-06T20:22:10  * ThomasWaldmann adds tests
2009-06-06T20:26:32  <ThomasWaldmann> -> 2 failures
2009-06-06T20:32:16  <CIA-19> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 5904:1f484a0b696d 2.0-storage/MoinMoin/security/_tests/test_security.py: acls: added more tests
2009-06-06T20:32:31  <ThomasWaldmann> dennda: ok, stuff to fix :)
2009-06-06T21:04:09  <CIA-19> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 5905:68b8376bf17d 2.0-storage/MoinMoin/ (4 files in 3 dirs): removed SecurityPolicy.save and its users (like autoadmin policy)
2009-06-06T21:11:40  <dimazest> strange, some tests fails time to time
2009-06-06T21:12:16  <dimazest> can it be because of the cache? and mtime granularity problems
2009-06-06T21:19:42  <ThomasWaldmann> dimazest: find out and get rid of the crap
2009-06-06T21:21:42  <dennda> I think johill once mentioned something like that
2009-06-06T21:22:24  <dennda> Something with a lock with some time.sleep() workaround or something similar if I'm not totally confusing things
2009-06-06T21:22:56  <dimazest> dennda: yeap, i want to try time.sleep()
2009-06-06T21:23:27  <dennda> (afair, he wasn't all that happy with that particular workaround)
2009-06-06T21:23:29  <dimazest> the only question, home many tests i should run to be sure that this workaround works
2009-06-06T21:27:49  <waldi> why workarounds? fix it
2009-06-06T21:29:28  <CIA-19> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 5907:d71299197349 2.0-storage/MoinMoin/_tests/test_wikiutil.py: fixed test to use split_interwiki
2009-06-06T21:57:10  <CIA-19> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 5908:9336a9c95c5f 2.0-storage/MoinMoin/security/__init__.py: ACLs: updated/fixed docstring
2009-06-06T22:14:42  <amartani> hi..
2009-06-06T22:16:43  <dimazest> hg qap
2009-06-06T22:17:30  * dimazest missed a window
2009-06-06T22:17:37  <dimazest> amartani: hi
2009-06-06T22:32:44  <CIA-19> Dmitrijs Milajevs <dimazest@gmail.com> default * 4764:a7a763c042f0 1.9-groups-dmilajevs/ (58 files in 12 dirs): Merge with 1.9
2009-06-06T22:32:46  <CIA-19> Dmitrijs Milajevs <dimazest@gmail.com> default * 4765:f0c2ab8ef256 1.9-groups-dmilajevs/ (7 files in 7 dirs): Groups2009: Merge with the second head to get rid of it. No changes from df833ab68df2 are merged.
2009-06-06T22:32:49  <CIA-19> Dmitrijs Milajevs <dimazest@gmail.com> default * 4766:de7ca86a9f62 1.9-groups-dmilajevs/ (7 files in 6 dirs):
2009-06-06T22:32:49  <CIA-19> Groups2009: MoinMoin.groups.backends.wiki_group backend was added.
2009-06-06T22:32:49  <CIA-19> Cache management methods load_cache and update_cache for the GroupManager.
2009-06-06T22:32:50  <CIA-19> request.groups proviedes access to group definitions. request.dits provides acces to the http://moinmo.in/WikiDict
2009-06-06T22:32:53  <CIA-19> Dmitrijs Milajevs <dimazest@gmail.com> default * 4767:1ff6498db9ec 1.9-groups-dmilajevs/MoinMoin/_tests/ (wikiconfig.py wikiconfig_groups.py): Groups2009: wikiconfig_groups.py is used for the groups-related configuration instead of altering wikiconfig.py in _tests. wiki_group backend is used by default to access group definitions.
2009-06-06T22:33:03  <CIA-19> Dmitrijs Milajevs <dimazest@gmail.com> default * 4768:31e0bbcf4fa6 1.9-groups-dmilajevs/MoinMoin/groups/backends/ (_tests/test_wiki_group.py wiki_group.py):
2009-06-06T22:33:03  <CIA-19> Groups2009: Group objects of wiki_group backend expands themselves during the initialization.
2009-06-06T22:33:04  <CIA-19> Test for wiki_group are refactored to be more realistic and easier to maintain.
2009-06-06T23:04:18  <ThomasWaldmann> dimazest: didn't you already commit/push those changesets?
2009-06-06T23:05:50  <ThomasWaldmann> (4766 4767)
2009-06-06T23:08:09  <ThomasWaldmann> dimazest: TYPOS (even in the commit comments, sigh)
2009-06-06T23:10:08  <ThomasWaldmann> proviedes request.dits acces
2009-06-06T23:16:18  <CIA-19> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 5909:eff13fd7decc 2.0-storage/MoinMoin/config/multiconfig.py: multiconfig: remove gdchart config, use new style class
2009-06-06T23:28:16  *** moinBot has joined #moin-dev
2009-06-06T23:31:04  <CIA-19> Alexandre Martani <amartani AT gmail DOT com> default * 5924:22a89261d291 2.0-storage-editor-amartani/MoinMoin/action/mobwrite.py: Real-time editor: more comments, docstrings and small changes
2009-06-06T23:43:43  <ThomasWaldmann> dimazest:     1.90 +        for (group, members) in self.expanded_groups.iteritems():
2009-06-06T23:43:54  <ThomasWaldmann> no parens needed
2009-06-06T23:50:12  <ThomasWaldmann> xorAxAx: still alive?
2009-06-06T23:55:20  <ThomasWaldmann> dimazest: expansion code from parse_page looks like it duplicates code from _expand_group