2007-07-02T00:03:10 <zenhase> since i always got lost in .send_page even during such a simple action like show i started a wikipage to outline the codeflow for different actions
2007-07-02T00:12:17 <grzywacz> bbl...
2007-07-02T00:29:53 <ThomasWaldmann> moin
2007-07-02T00:54:30 <ThomasWaldmann> lanius: the get/set_lock api is a bit strange
2007-07-02T00:57:05 <ThomasWaldmann> get_lock could just return is_locked, lock_timestamp, lock_user (the last 2 being e.g. None if not is_locked)
2007-07-02T00:59:02 <ThomasWaldmann> set_lock would just get lock_state, lock_timestamp=None, lock_user=None as params
2007-07-02T00:59:19 <ThomasWaldmann> or even better: split that into set_lock and delete_lock
2007-07-02T01:01:11 <ThomasWaldmann> ehrm, oops, this is a property.
2007-07-02T01:06:27 <ThomasWaldmann> lanius: in your editlock emu, use a tuple of names (timestamp, ....) instead of line
2007-07-02T01:07:51 <ThomasWaldmann> lanius: + if DELETED in metadata and metadata[DELETED] == True: s/ == True//
2007-07-02T01:09:23 <ThomasWaldmann> btw, i guess the edit-lock file ends with a \n
2007-07-02T01:10:54 <ThomasWaldmann> don't use "string" as variable name. name clash with the string module. and it also is somewhat nothing-saying.
2007-07-02T01:12:22 <ThomasWaldmann> +LOCK_TIMESTAMP = "lock-timestamp" maybe it is better to use keys that would be valid python identifiers, so s/-/_/
2007-07-02T08:26:13 <ThomasWaldmann> moin
2007-07-02T08:36:58 <starshine> moin :)
2007-07-02T08:39:18 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2300:a607b0329922 1.7-maninfo-vpv/MoinMoin/_tests/test_wikiutil.py: add tests for parsing and making of query strings
2007-07-02T08:39:20 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2301:3318504c5dd5 1.7-maninfo-vpv/MoinMoin/ (_tests/test_wikiutil.py wikiutil.py): create tickets as fn(time, pagename, action, secret) - so they are not reusable, add tests for tickets
2007-07-02T08:39:21 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2302:463f3de3f981 1.7-maninfo-vpv/MoinMoin/ (_tests/test_wikiutil.py wikiutil.py): more wikiutil tests, minor bugfix
2007-07-02T08:39:23 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2303:e6792cdb3e83 1.7-maninfo-vpv/MoinMoin/_tests/test_Page.py: Page: added some tests
2007-07-02T08:39:24 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2304:9f9a79d5a8d9 1.7-maninfo-vpv/MoinMoin/ (8 files in 4 dirs): tests: move macro and parser tests to macro/_tests and parser/_tests
2007-07-02T08:39:27 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2305:1a2c91fa17a5 1.7-maninfo-vpv/MoinMoin/ (20 files in 11 dirs): tests: move them to /_tests/
2007-07-02T08:39:30 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2306:7f703eba4403 1.7-maninfo-vpv/MoinMoin/_tests/test_parser_text_moin_wiki.py: remove leftover test, see MoinMoin/parser/_tests/test_text_moin_wiki.py for current test
2007-07-02T08:39:34 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2307:2aa65d5835d4 1.7-maninfo-vpv/tests/rpctest.py: remove useless test
2007-07-02T08:39:39 <CIA-20> moin: Ville-Pekka Vainio <vpivaini AT cs DOT helsinki DOT fi> * 2308:626ab12c215f 1.7-maninfo-vpv/ (33 files in 16 dirs): merge from main
2007-07-02T08:39:48 <vpv> moin ;)
2007-07-02T09:09:45 <lanius> moin
2007-07-02T09:09:56 <lanius> ThomasWaldmann: what is best for the set_lock api
2007-07-02T09:10:06 <xorAxAx> moin
2007-07-02T09:11:02 <ThomasWaldmann> lanius: while reading the diff, i first didnt see it being a property (you dont have much choice if it is)
2007-07-02T09:12:06 <ThomasWaldmann> but if you dont use a prop, the interface would get more natural
2007-07-02T09:13:40 <CIA-20> moin: Heinrich Wendel <h_wendel@cojobo.net> * 2121:441a7adf2ee1 1.7-storage-hwendel/MoinMoin/storage/ (4 files in 2 dirs): fixes to the lock code
2007-07-02T10:28:50 <CIA-20> moin: Heinrich Wendel <h_wendel@cojobo.net> * 2122:17c1ae83139b 1.7-storage-hwendel/MoinMoin/ (3 files in 3 dirs): use the new locking code in PageLock
2007-07-02T10:33:32 <CIA-20> moin: Heinrich Wendel <h_wendel@cojobo.net> * 2123:72cafa7b7a8a 1.7-storage-hwendel/MoinMoin/PageEditor.py: fix self.owner_html
2007-07-02T12:24:40 <ThomasWaldmann> lanius: btw the long() timestamps were primarily for py 2.2 compatibility, as we require 2.3 now anyway, you don't need to keep the long() stuff
2007-07-02T12:25:54 <ThomasWaldmann> (otoh, you also don't need to remove it, we can do this later also)
2007-07-02T12:28:20 <grzywacz> moin
2007-07-02T12:52:04 <ThomasWaldmann> btw, 1.7 main branch has some more tests now, maybe all developers want to merge them soon
2007-07-02T12:59:05 <lanius> can you merge grzywacz fix for userform.py ?
2007-07-02T13:03:54 <lanius> ThomasWaldmann: http://hg.moinmo.in/moin/1.7-jabber-knowak/rev/8aad85519f76
2007-07-02T13:07:30 <johill> userform.py. hmm. what are you doing with it?
2007-07-02T13:08:18 <xorAxAx> lanius: you can do as well
2007-07-02T13:08:25 <xorAxAx> lanius: just pull from his branch
2007-07-02T13:08:29 <xorAxAx> it looks all very stable
2007-07-02T13:17:46 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2274:d782b0db4757 1.7-jabber-knowak/MoinMoin/ (_tests/test_wikidicts.py wikidicts.py): wikidicts: more comments, minor cleanup, converted tests to py.test
2007-07-02T13:17:49 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2275:0131f0aabb18 1.7-jabber-knowak/MoinMoin/_tests/ (test_packages.py test_wikisync.py): merged main
2007-07-02T13:17:49 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2276:d48a4d0a0c70 1.7-jabber-knowak/ (MoinMoin/server/STANDALONE.py moin.py): pycallgraph support
2007-07-02T13:17:51 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2277:07fdbda38be2 1.7-jabber-knowak/MoinMoin/ (Page.py config/__init__.py): move split_regex to config (so it is compiled only once)
2007-07-02T13:17:52 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2278:4197965b0820 1.7-jabber-knowak/MoinMoin/config/__init__.py: fix: use re.UNICODE for split_regex
2007-07-02T13:17:54 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2279:ece659592dd9 1.7-jabber-knowak/MoinMoin/ (_tests/test_wikidicts.py action/SyncPages.py wikidicts.py): wikidicts: refactoring, more tests
2007-07-02T13:17:57 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2280:2c8f25796120 1.7-jabber-knowak/ (4 files in 4 dirs): merged main
2007-07-02T13:18:00 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2281:564dee5eb254 1.7-jabber-knowak/MoinMoin/ (request/__init__.py wikidicts.py events/wikidictsnotify.py): use event system to trigger wikidicts cache update
2007-07-02T13:18:07 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2282:b74a4230a123 1.7-jabber-knowak/MoinMoin/events/ (wikidictsrescan.py wikidictsnotify.py): renamed wikidictsnotify > wikidictsrescan
2007-07-02T13:18:10 <CIA-20> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 2283:691b6a650d94 1.7-jabber-knowak/MoinMoin/_tests/parser/test_text_moin_wiki.py:
2007-07-02T13:18:13 <CIA-20> moin: test_text_moin_wiki: moved to _tests/parser, renamed from test_parser_text_moin_wiki, added test for colorized
2007-07-02T13:18:16 <CIA-20> moin: parser python and nesting {{{ }}}
2007-07-02T13:18:18 <CIA-20> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 2284:6327a199c666 1.7-jabber-knowak/MoinMoin/parser/text_moin_wiki.py:
2007-07-02T13:18:21 <CIA-20> moin: text_moin_wiki:fixed bug for Colorized parsers and nesting {{{ }}} (FeatureRequests/ParsersCanNotContain{{{}}})
2007-07-02T13:18:24 <CIA-20> moin: , should be added to 1.6 too
2007-07-02T13:18:26 <CIA-20> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 2285:9c66c14f6b62 1.7-jabber-knowak/MoinMoin/ (2 files in 2 dirs): text_moin_wiki: completing nesting for colorized parsers and some tests
2007-07-02T13:18:35 <CIA-20> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 2286:709eea9354c7 1.7-jabber-knowak/MoinMoin/parser/text_moin_wiki.py: text_moin_wiki: some small PEP8 changes, that version should go into 1.6 too
2007-07-02T13:18:38 <CIA-20> moin: Alexander Schremmer <alex AT alexanderweb DOT de> * 2287:c9a4fbbb2b2e 1.7-jabber-knowak/MoinMoin/ (2 files in 2 dirs): Merged main.
2007-07-02T13:18:41 <CIA-20> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 2288:7e04a576e8c7 1.7-jabber-knowak/MoinMoin/_tests/parser/test_text_moin_wiki.py: test_text_moin_wiki: small PEP8 changes
2007-07-02T13:18:46 <CIA-20> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 2289:ec8d3c66ad0c 1.7-jabber-knowak/MoinMoin/_tests/parser/test_text_moin_wiki.py: test_text_moin_wiki: typo fixed
2007-07-02T13:18:49 <CIA-20> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 2290:c1682ea7ed8c 1.7-jabber-knowak/MoinMoin/ (2 files in 2 dirs): text_moin_wiki: fixed an other nesting {{{ }}} problem and added a new test
2007-07-02T13:18:54 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2291:b0031a34f96d 1.7-jabber-knowak/MoinMoin/conftest.py: fixed some typos
2007-07-02T13:18:57 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2292:a607b0329922 1.7-jabber-knowak/MoinMoin/_tests/test_wikiutil.py: add tests for parsing and making of query strings
2007-07-02T13:19:00 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2293:3318504c5dd5 1.7-jabber-knowak/MoinMoin/ (_tests/test_wikiutil.py wikiutil.py): create tickets as fn(time, pagename, action, secret) - so they are not reusable, add tests for tickets
2007-07-02T13:19:06 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2294:463f3de3f981 1.7-jabber-knowak/MoinMoin/ (_tests/test_wikiutil.py wikiutil.py): more wikiutil tests, minor bugfix
2007-07-02T13:19:09 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2295:e6792cdb3e83 1.7-jabber-knowak/MoinMoin/_tests/test_Page.py: Page: added some tests
2007-07-02T13:19:12 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2296:9f9a79d5a8d9 1.7-jabber-knowak/MoinMoin/ (8 files in 4 dirs): tests: move macro and parser tests to macro/_tests and parser/_tests
2007-07-02T13:19:17 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2297:1a2c91fa17a5 1.7-jabber-knowak/MoinMoin/ (20 files in 11 dirs): tests: move them to /_tests/
2007-07-02T13:19:20 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2298:7f703eba4403 1.7-jabber-knowak/MoinMoin/_tests/test_parser_text_moin_wiki.py: remove leftover test, see MoinMoin/parser/_tests/test_text_moin_wiki.py for current test
2007-07-02T13:19:24 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2299:2aa65d5835d4 1.7-jabber-knowak/tests/rpctest.py: remove useless test
2007-07-02T13:19:29 <CIA-20> moin: Karol 'grzywacz' Nowak <grzywacz@sul.uni.lodz.pl> * 2300:529b17ec4396 1.7-jabber-knowak/ (43 files in 22 dirs): Merge with main.
2007-07-02T13:19:32 <CIA-20> moin: Karol 'grzywacz' Nowak <grzywacz@sul.uni.lodz.pl> * 2301:abcb89a89ad2 1.7-jabber-knowak/ (9 files in 6 dirs): Merge with changes I made when away.
2007-07-02T13:20:18 <lanius> xorAxAx: i could, but it would be nice to have it in main
2007-07-02T13:20:50 <xorAxAx> lanius: it will go there shortly
2007-07-02T13:22:41 <lanius> sounds good
2007-07-02T13:22:57 <xorAxAx> shortly as in "in the next days"
2007-07-02T13:23:11 <xorAxAx> the point is that you dont have any disadvantages if you pull it yourself
2007-07-02T13:23:31 <lanius> but i don't need to merge either
2007-07-02T13:23:43 <xorAxAx> ?
2007-07-02T13:23:59 <lanius> i do not really need the changes from main, so i can wait until a more stable state
2007-07-02T13:24:01 <xorAxAx> the amount of necessary merging doesnt increase
2007-07-02T13:27:00 <ThomasWaldmann> lanius: there are also some new tests for Page and wikiutil, they maybe would also be useful in your branch
2007-07-02T13:32:05 <lanius> ok
2007-07-02T13:34:16 <ThomasWaldmann> grzywacz: for what exactly is that notify flag? should it send a PageChangedEvent EVER?
2007-07-02T13:34:46 <xorAxAx> ThomasWaldmann: hmm?
2007-07-02T13:35:06 <xorAxAx> in case of a simple page event, it will send an event
2007-07-02T13:35:11 <xorAxAx> s/event/edit/
2007-07-02T13:35:14 <ThomasWaldmann> saveText()
2007-07-02T13:35:42 <xorAxAx> as it was before ...
2007-07-02T13:36:57 <ThomasWaldmann> ah, you only suppress triggering double event sending?
2007-07-02T13:37:38 <xorAxAx> yes
2007-07-02T13:37:41 <xorAxAx> thats the idea
2007-07-02T13:38:12 <xorAxAx> we thought about a global ignore flag, but concluded that it is not clean enough
2007-07-02T13:41:33 <ThomasWaldmann> (if we use events for moin internals, it is important that they really get triggered, see events/wikidictsrescan.py :)
2007-07-02T13:42:39 <xorAxAx> yes
2007-07-02T13:42:51 <ThomasWaldmann> a similar thing could be done for xapian index update, pagelinks and formatter caches maybe
2007-07-02T13:43:08 <xorAxAx> formatter caches are synchronous
2007-07-02T13:43:19 <ThomasWaldmann> (i didnt do it yet to not produce conflicts within Page.py)
2007-07-02T13:43:22 <xorAxAx> and pagelinks depend on formatter internals
2007-07-02T13:43:25 <xorAxAx> but xapian makes sense
2007-07-02T13:45:32 <ThomasWaldmann> - server = xmlrpclib.ServerProxy('http://127.0.0.1/cgi-bin/wiki.bat/?action=xmlrpc') <- from the test I removed :D
2007-07-02T13:45:59 <xorAxAx> %-)
2007-07-02T14:01:34 <vpv> FYI, I'm currently getting to know Actions, there's some talk around whether I should use CVS or RPMs as man page sources, so until that's decided I'll try to work on sisterdiff.
2007-07-02T14:03:40 <ThomasWaldmann> vpv: btw, maybe try to keep the fedora specific stuff separate (or anything that e.g. Debian would maybe do different)
2007-07-02T14:05:05 <xorAxAx> vpv said that he planned to do that when the prototype is finished
2007-07-02T14:05:15 <vpv> that's the idea how to eventually do it. right now it's quite Fedora specific, but I am thinking about using super-/subclasses or something to handle that
2007-07-02T14:05:50 <vpv> after the man pages are found, the process is similar in any distro, but I'll need to separate the searching part somehow
2007-07-02T14:18:59 <grzywacz> ThomasWaldmann, as xorAxAx explained. :)
2007-07-02T14:19:37 <grzywacz> :D lol @ - server :D
2007-07-02T14:20:00 <xorAxAx> jürgen obviously used windows :)
2007-07-02T14:20:27 <ThomasWaldmann> the moin 0.1 release had some win32 in the filename
2007-07-02T14:27:32 <ThomasWaldmann> lanius: your storage tests use tarfile from stdlib. this is broken for some python versions, please use the fixed one from MoinMoin.support.
2007-07-02T14:41:04 <ThomasWaldmann> lanius: TODO copy edit log in external.copy_item means what? Isn't the "edit-log" contained in the metadata of the revisions?
2007-07-02T14:47:34 <lanius> yes, but not yet
2007-07-02T14:47:52 <ThomasWaldmann> copy_item does not close oldrev.data. maybe better close it explicitely - as you do open all revs in a loop, you maybe don't want to wait for the gc to do it for some thousand revs in some bad case.
2007-07-02T14:48:03 <lanius> right
2007-07-02T14:54:22 <ThomasWaldmann> new_item: you changed it to return an Item (but don't do anything else), but at some places you don't use the return value, but expect it to appear in some dict
2007-07-02T14:54:35 <ThomasWaldmann> can you please check this?
2007-07-02T14:57:53 <ThomasWaldmann> hmm, due to the implementation of __getitem__, this will work, but it will create the Item object again
2007-07-02T15:00:55 <ThomasWaldmann> btw, revision-number is written without -
2007-07-02T15:02:37 <ThomasWaldmann> if LOCK_TIMESTAMP and LOCK_USER in self.metadata:
2007-07-02T15:02:59 <ThomasWaldmann> nice idea, but I don't think Python will understand that correctly :)
2007-07-02T15:06:20 <ThomasWaldmann> lanius: before checking the len(lock) == 2, you should check type(lock) is tuple. Or you will never see that error msg in the else branch if someone just gives lock = True.
2007-07-02T15:07:40 <lanius> ThomasWaldmann: as far as i tried it does
2007-07-02T15:07:46 <lanius> understand the and semantic
2007-07-02T15:09:27 <ThomasWaldmann> >>> l = [0,1,2,3]
2007-07-02T15:09:28 <ThomasWaldmann> >>> 0 and 1 in l
2007-07-02T15:09:32 <ThomasWaldmann> 0
2007-07-02T15:10:12 <lanius> hm, okey ;)
2007-07-02T15:10:30 <ThomasWaldmann> >>> 0 in l and 1 in l
2007-07-02T15:10:30 <ThomasWaldmann> True
2007-07-02T15:11:24 <ThomasWaldmann> iirc, the only magic thing for "in" is that "not in" is also understood as a token
2007-07-02T15:12:38 <lanius> ThomasWaldmann: that's the difference between new_item and new_revision
2007-07-02T15:12:38 <ThomasWaldmann> ehrm, "operator"
2007-07-02T15:13:14 <ThomasWaldmann> what? :)
2007-07-02T15:13:32 <lanius> new_item creates an item and returns it
2007-07-02T15:13:47 <lanius> new_revision creates a revision, caches it and returns it
2007-07-02T15:14:05 <lanius> caches it in a dict
2007-07-02T15:14:11 <ThomasWaldmann> but you don't use what new_item returns
2007-07-02T15:14:30 <lanius> ah right, i should do that
2007-07-02T15:15:25 <ThomasWaldmann> do you code in C or other non-python languages currently ? ;)
2007-07-02T15:15:31 <lanius> java
2007-07-02T15:15:44 * xorAxAx is also doing java :)
2007-07-02T15:15:44 <ThomasWaldmann> raise BackendError(_("Copy failed because name and newname are equal."));
2007-07-02T15:15:58 <ThomasWaldmann> s/;//
2007-07-02T15:16:03 <lanius> ^^
2007-07-02T15:16:15 <xorAxAx> you could write a unittest that checks for ; at the line end
2007-07-02T15:16:20 <ThomasWaldmann> (also at some other places)
2007-07-02T15:16:20 <xorAxAx> pypy has one that checks for tabs
2007-07-02T15:16:26 <xorAxAx> you could copy it and add a ;-check
2007-07-02T15:16:32 <xorAxAx> its in pypy.tool.*
2007-07-02T15:16:33 <ThomasWaldmann> or trailing blanks at EOL
2007-07-02T15:16:51 <xorAxAx> yes, that as well
2007-07-02T15:17:16 <ThomasWaldmann> xorAxAx: i guess I'll do that in 1.7 main
2007-07-02T15:17:52 * ThomasWaldmann .oO(that will be the most failing test :D )
2007-07-02T15:19:41 <dreimark> currently I reconsider all my parser changes for nestings (I hope I get it sane)
2007-07-02T15:20:25 <ThomasWaldmann> hehe
2007-07-02T15:20:54 <ThomasWaldmann> dreimark: btw, I moved your tests to testedpackage/_tests/
2007-07-02T15:21:02 <grzywacz> Hm, weird test, but I like the idea. ;D
2007-07-02T15:21:16 <dreimark> I have seen that its a better place
2007-07-02T15:24:50 <xorAxAx> ThomasWaldmann: https://codespeak.net/pypy/dist/pypy/tool/test/test_tab.py
2007-07-02T15:24:58 <xorAxAx> mit licensed, copyright armin rigo
2007-07-02T15:32:07 <grzywacz> It should use regexp.
2007-07-02T15:32:16 <grzywacz> Tabs in strings should cause asserts. :P
2007-07-02T15:32:20 <grzywacz> *shouldn't
2007-07-02T15:37:34 <xorAxAx> well
2007-07-02T15:37:37 <xorAxAx> i disagree :)
2007-07-02T15:39:15 <ThomasWaldmann> xorAxAx: it misses (c) info. you know for sure it was AR?
2007-07-02T15:39:33 <ThomasWaldmann> > AssertionError: '/parser/text_cplusplus.py' contains tabs!
2007-07-02T15:40:06 <xorAxAx> ThomasWaldmann: yes
2007-07-02T15:40:15 <xorAxAx> ThomasWaldmann: looked into the svn history
2007-07-02T15:40:32 <ThomasWaldmann> ok
2007-07-02T15:41:07 <ThomasWaldmann> (c) 2007?
2007-07-02T15:45:38 <xorAxAx> 2006
2007-07-02T15:52:53 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2273:80c083a0e4a7 1.7/MoinMoin/_tests/test_tab.py: tests: test for source files containing tabs, thanks to Armin Rigo
2007-07-02T15:53:35 <CIA-20> moin: Heinrich Wendel <h_wendel@cojobo.net> * 2124:a5f065786ff4 1.7-storage-hwendel/MoinMoin/storage/ (external.py fs_moin16.py interfaces.py): minor changes
2007-07-02T15:57:39 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2274:01f6172dabd0 1.7/MoinMoin/ (6 files in 3 dirs): remove tabs from source files, remove excludes from test_tab
2007-07-02T16:06:59 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2275:023e0f0d264e 1.7/MoinMoin/ (3 files in 2 dirs): tests: rename test_tab to test_sourcecode, also test for sourcefiles not ending with a newline char
2007-07-02T16:11:57 <CIA-20> moin: Ville-Pekka Vainio <vpivaini AT cs DOT helsinki DOT fi> * 2309:cc6bebeed861 1.7-maninfo-vpv/MoinMoin/script/xmlrpc/manimport.py: fix an if that's implicit anyway
2007-07-02T16:11:58 <CIA-20> moin: Ville-Pekka Vainio <vpivaini AT cs DOT helsinki DOT fi> * 2310:8410ccb95d2a 1.7-maninfo-vpv/MoinMoin/action/SisterDiff.py: Add initial implementation of SisterDiff
2007-07-02T16:13:38 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2311:80c083a0e4a7 1.7-maninfo-vpv/MoinMoin/_tests/test_tab.py: tests: test for source files containing tabs, thanks to Armin Rigo
2007-07-02T16:13:39 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2312:01f6172dabd0 1.7-maninfo-vpv/MoinMoin/ (6 files in 3 dirs): remove tabs from source files, remove excludes from test_tab
2007-07-02T16:13:42 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2313:023e0f0d264e 1.7-maninfo-vpv/MoinMoin/ (3 files in 2 dirs): tests: rename test_tab to test_sourcecode, also test for sourcefiles not ending with a newline char
2007-07-02T16:13:43 <CIA-20> moin: Ville-Pekka Vainio <vpivaini AT cs DOT helsinki DOT fi> * 2314:8696cb29389a 1.7-maninfo-vpv/MoinMoin/ (7 files in 4 dirs): merge from main
2007-07-02T16:23:16 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2276:e56af6f9b2cc 1.7/MoinMoin/_tests/test_sourcecode.py: tests: test for crlf in Python sourcecode
2007-07-02T16:23:18 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2277:5638ffa2fb38 1.7/ (4 files in 4 dirs): replace crlf in python src files by lf
2007-07-02T16:23:54 <ThomasWaldmann> vpv: your branch is outdated :D
2007-07-02T16:26:23 <vpv> lol, I'll let it be until tomorrow :P
2007-07-02T16:32:32 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2278:9df10b5358b5 1.7/MoinMoin/_tests/test_sourcecode.py: test_sourcecode: add test for trailing spaces (currently disabled), other minor changes
2007-07-02T16:33:06 <ThomasWaldmann> vpv: just joking, maybe only merge if you've done some own changesets :)
2007-07-02T16:39:22 <vpv> me too, but I like to merge often, that's how I found Karol's xmlrpc bug ;)
2007-07-02T16:58:21 * ThomasWaldmann solves the trailing space issue :)
2007-07-02T17:06:49 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2279:938d6761519f 1.7/MoinMoin/_tests/test_sourcecode.py: test_sourcecode: ignore/test/fix modes for trailing spaces, fixed file mode to use binary
2007-07-02T18:35:57 * ThomasWaldmann tries to integrate pep8.py with py.test
2007-07-02T18:44:01 <CIA-20> moin: Alexander Schremmer <alex AT alexanderweb DOT de> * 2280:27cd29db1b96 1.7/MoinMoin/conftest.py: Fix conftest to allow generative tests.
2007-07-02T18:44:06 <xorAxAx> ThomasWaldmann:
2007-07-02T19:00:40 <ThomasWaldmann> ah, great :)
2007-07-02T19:07:44 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2281:a6a4eaa948aa 1.7/MoinMoin/_tests/ (test_sourcecode.py pep8.py): test_sourcecode: added PEP8 checker
2007-07-02T19:07:45 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2282:bb9853b194f7 1.7/MoinMoin/conftest.py: merge main
2007-07-02T19:33:50 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2283:959d0815e149 1.7/MoinMoin/_tests/ (pep8.py test_sourcecode.py): pep8: don't count stuff we ignore, added comments about pep8 test usage
2007-07-02T19:39:47 <ThomasWaldmann> xorAxAx: generative tests don't like me
2007-07-02T19:40:20 <xorAxAx> add a failing test :)
2007-07-02T19:40:58 <ThomasWaldmann> simple stuff now works (as the example code), but if I yield a test from a recursive fn inside a test function, it doesnt do the test
2007-07-02T19:41:42 <ThomasWaldmann> it doesnt fail, it simply does nothing
2007-07-02T19:41:57 <xorAxAx> then add a test that does nothing :)
2007-07-02T19:43:02 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2284:71f7bde1077b 1.7/MoinMoin/_tests/test_sourcecode.py: test_sourcecode: trying to use generative tests (doesn't work yet)
2007-07-02T19:46:18 <neagulm> hello dreimark_
2007-07-02T19:48:30 <dreimark_> hi neagulm
2007-07-02T19:48:36 <xorAxAx> ThomasWaldmann: well
2007-07-02T19:48:54 <xorAxAx> ThomasWaldmann: you have to modify walk("") into for _ in walk(""): yield _
2007-07-02T19:49:28 <xorAxAx> thats how generators work
2007-07-02T19:49:28 <xorAxAx> you have to modify every walk call like that
2007-07-02T19:50:39 * ThomasWaldmann tries
2007-07-02T19:54:29 * ThomasWaldmann runs 3600 generative tests....................
2007-07-02T19:54:30 <ThomasWaldmann> 3522 passed, 160 failed
2007-07-02T19:54:48 <dreimark_> he
2007-07-02T19:57:01 <ThomasWaldmann> xorAxAx: stdout is lost for any failing test except the first failing one
2007-07-02T19:57:24 <xorAxAx> that sounds like a pylib bug :)
2007-07-02T19:57:29 <xorAxAx> the channel is #pylib btw
2007-07-02T20:13:32 <ThomasWaldmann> not much traffic there :)
2007-07-02T20:39:00 <ThomasWaldmann> the stdout problem was no pylib / py.test problem
2007-07-02T20:44:26 <xorAxAx> hmm
2007-07-02T20:46:06 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2285:751e149a9a0a 1.7/MoinMoin/_tests/test_sourcecode.py: test_sourcecode: fix generative test
2007-07-02T20:47:41 <ThomasWaldmann> tests finished: 3542 passed, 140 failed
2007-07-02T20:48:20 <ThomasWaldmann> looks like we have some PEP8 issues.
2007-07-02T20:51:54 <dreimark_> hi skipm, neagulm
2007-07-02T20:52:07 <neagulm> hello skipm, dreimark
2007-07-02T20:52:12 <skipm> dreimark, neagulm: hi
2007-07-02T20:54:12 <dreimark_> how are you skipm, neagulm ?
2007-07-02T20:54:38 <skipm> dreinmark: fine.
2007-07-02T20:54:59 <neagulm> skipm, I have played around with the SB tokenizer... :)
2007-07-02T20:55:04 <skipm> dreimark, neagulm: have either of you looked at SpamBayes 1.1a4?
2007-07-02T20:55:04 <lanius> ThomasWaldmann: xorAxAx: how did we say to store the global edit log again
2007-07-02T20:55:12 <xorAxAx> lanius: i have no idea
2007-07-02T20:56:05 <neagulm> skipm, dreimark not yet. I have 2-3 days until I will finish with my diploma exam. and after that I will play with it.
2007-07-02T20:56:27 <neagulm> skipm, dreimark I have used MoinMoin's wiki parser to extract tokens.
2007-07-02T20:58:03 <skipm> neagulm, dreimark: I'm still unclear how you're doing things. There shouldn't be any need to tokenize things. SpamBayes can do that for you.
2007-07-02T20:58:35 <dreimark_> skipm: how does it do it?+
2007-07-02T20:58:51 <neagulm> skipm, and the questions is: was this the right approach? I adopted it because the wiki markup is more simple then HTML and I can extract a lot more data then the SB tokenizer.
2007-07-02T20:59:31 <skipm> dreimark: well, it basically splits the message into words - but it also generates lots of synthetic "words" based upon various analyses including analyzing ip addresses, resolving hostnames, etc
2007-07-02T20:59:49 <neagulm> skipm, and of course the system will be able to use SB's tokenizer without change.
2007-07-02T21:00:36 <neagulm> skipm, the same data is extracted by me to, but with taking into account wiki markup.
2007-07-02T21:01:13 <skipm> Look at 1.1a4. I give an API called score() which takes a dictionary (the form submission contents) and a (possibly empty) list of user-provided tokens. I think all you need to do is add stuff to the token list (is the user logged in? for how long? etc)
2007-07-02T21:02:57 <ThomasWaldmann> lanius: iirc, my suggestion was to use metadata of a special item
2007-07-02T21:03:15 <lanius> ThomasWaldmann: that would be the virtual / item then
2007-07-02T21:03:36 <lanius> ThomasWaldmann: where we could store wiki wide metadata in revision -1 as well
2007-07-02T21:03:46 <neagulm> skipm, dreimark that is ok, I will write a module for that. I want to have an extensible system in order to provide the user with the choice of using several classifiers
2007-07-02T21:05:10 <ThomasWaldmann> lanius: yeah
2007-07-02T21:05:20 <neagulm> skipm, In the current state We have 2 classifiers: SB for spam detection (with built-in or SB tokenizer) and another (bundled) classifier for language classification
2007-07-02T21:05:24 <skipm> neagulm: you're writing a spam classifier from scratch?
2007-07-02T21:06:55 <neagulm> skipm, no :)) But I want to extract the last bit of efficiency from the classifiers
2007-07-02T21:07:09 <skipm> neagulm, dreimark: i've written an an auditor (I think that's the right term) for Roundup which uses the SpamBayes server
2007-07-02T21:07:19 <skipm> neagulm: efficiency in what way?
2007-07-02T21:09:26 <neagulm> skipm, dreimark classification. Accuracy. etc I'm testing because of my plans of implementing support for n-way classification
2007-07-02T21:09:53 <neagulm> skipm, You don't agree with having a MM tokenizer ? Use the one provided with SB ?
2007-07-02T21:11:31 <skipm> neagulm: this is too hard for me to carry on in irc - maybe we should exchange a couple emails on the subject. i must admit I don't understand why there's such a strong focus on the tokenization aspect
2007-07-02T21:15:48 <dreimark_> skipm: one point could be that you can compare independent from builtin tokenisation
2007-07-02T21:16:44 <dreimark_> but you both are the experts if you like to exchange your mind by mail its ok if we got later on a summary
2007-07-02T21:16:59 <skipm> dreimark: compare what? tokenization is little more than pipe.read().split() followed by generation of a number of synthetic tokens as I indicated earlier
2007-07-02T21:17:20 <neagulm> skipm, ok. I will write the interface to SB 1.1a4 and to some benchmarks related to the tokenization issue. I agree that this might not be really important but I think that some testing is good
2007-07-02T21:21:35 <neagulm> dreimark, Feature extraction is only a part of the classifier. My opinion is that the more usefull information we extract and provide to the classifier the more accurate will be (if the classifier considers that this features are relevant).
2007-07-02T21:22:26 <dreimark_> skipm: neagulm did you ever tried with rtl text does this matter?
2007-07-02T21:22:30 <skipm> dreimark, neaculm: my Roundup auditor is at http://www.webfast.com/~skip/spamcheck.py
2007-07-02T21:23:04 <skipm> dreimark, neagulm: sorry, I can't spell/type today...
2007-07-02T21:24:16 <dreimark_> skipm neagulm I have learned to use autocomplete, (tab key here)
2007-07-02T21:24:30 <dreimark_> tip from xorAxAx some time ago
2007-07-02T21:25:24 <neagulm> dreimark_, it does not really matter as long as we can extract features
2007-07-02T21:25:37 <neagulm> skipm, ok
2007-07-02T21:29:44 <dreimark_> skipm: neagulm ./spambayes/tokenizer.py I like this description :)
2007-07-02T21:32:07 <dreimark_> skipm: neagulm How big should "a word" be? wiki words for pages could be very long (longer as 12chars) should wiki words separated, splitted in single words?
2007-07-02T21:32:21 <skipm> hang on - discssions here...
2007-07-02T21:34:35 <neagulm> dreimark_, one approach is to provide both: the long version and the building parts. Another would be to use the whole word or to split it
2007-07-02T21:34:52 <skipm> dreimark: you will have to test optimal word length. lots of testing went into getting SB where it is today. Little, if anything, is just a guess about what would or would not work
2007-07-02T21:35:51 <skipm> dreimark: yes, perhaps wiki names should be split, but you can sublass SB's tokenizer and do that step, leaving the rest to the default
2007-07-02T21:37:04 <skipm> neagulm, dreimark: hang on - discussions here (we
2007-07-02T21:37:29 <skipm> neagulm, dreimark: are building R and having problems...)
2007-07-02T21:37:49 <skipm> dreimark: thanks for the TAB trick...
2007-07-02T21:38:26 <dreimark_> :)
2007-07-02T21:41:05 <lanius> ThomasWaldmann: what about the event-log
2007-07-02T21:46:31 <ThomasWaldmann> add a todo for later, there are more important things :)
2007-07-02T21:50:05 <dreimark_> skipm: neagulm and spain does have one letter words e.g. http://moinmoin.wikiwikiweb.de/FeatureRequests/OneLetterWordInWikiWords
2007-07-02T21:51:34 <dreimark_> is it necessary to have language dependent optimisation ?
2007-07-02T21:55:02 <skipm> dreimark: one-letter words are okay - shorter than (by default) three-letter words are ignored
2007-07-02T21:55:30 <dreimark_> ahh ok
2007-07-02T21:56:02 <lanius> ThomasWaldmann: a page local event-log is nowhere used, is it?
2007-07-02T21:58:21 <lanius> xorAxAx: can you take a look at wikisync.py and eliminate the use of getPagePath?
2007-07-02T21:59:01 <xorAxAx> lanius: hmm
2007-07-02T21:59:16 <xorAxAx> lanius: it only uses it for the tags, right?
2007-07-02T21:59:35 <lanius> tags and cache, either elimnate or tell me what i have to add to the storage layer
2007-07-02T21:59:43 <lanius> s/cache/lock_dir/
2007-07-02T22:00:07 <xorAxAx> well, the tags are modifiable meta-data
2007-07-02T22:00:17 <xorAxAx> you need a metadata key for synctags
2007-07-02T22:00:23 <xorAxAx> but the system is flexible
2007-07-02T22:00:28 <xorAxAx> you can store them like you like
2007-07-02T22:00:33 <xorAxAx> see the abstract base class
2007-07-02T22:01:00 <xorAxAx> e.g. you could keep the old class for the fs16 and use a meta-data based for anything else
2007-07-02T22:01:25 <lanius> xorAxAx: i don't think i need to implement that on the storage layer then since you can put any metadata key you want into an item
2007-07-02T22:01:34 <xorAxAx> lanius: yes
2007-07-02T22:01:44 <xorAxAx> i just need to be able to modify it without creating a new revision
2007-07-02T22:01:53 <lanius> that is possible
2007-07-02T22:02:08 <lanius> use pageobj._item and do what you want ;)
2007-07-02T22:03:45 <xorAxAx> ugh
2007-07-02T22:03:54 <xorAxAx> why does it have an underscore if i have to use it ...
2007-07-02T22:04:27 <xorAxAx> lanius: can you document somewhere how i can modify the metadata without creating a new reivison?
2007-07-02T22:04:40 <lanius> xorAxAx: you can even use ItemCollection[page_name] directly
2007-07-02T22:04:55 <lanius> xorAxAx: if you use the external interface instead of Page.py nothing is created automatically
2007-07-02T22:05:44 <xorAxAx> lanius: ok
2007-07-02T22:05:55 <xorAxAx> neverless i want a short writetup :)
2007-07-02T22:07:43 <lanius> k
2007-07-02T22:08:03 <lanius> ThomasWaldmann: i can't find a tarfile lib in moin
2007-07-02T22:11:30 <dreimark_> lanius: from MoinMoin.support import tarfile
2007-07-02T22:15:46 <lanius> xorAxAx: http://moinmoin.wikiwikiweb.de/StorageRefactoring/SOC2007#head-498b776143077058af4f5df28ae5b70c11323304
2007-07-02T22:15:49 <lanius> is that sufficient?
2007-07-02T22:17:08 <xorAxAx> lanius: well, i am wondering how that ensures transactional safety :)
2007-07-02T22:18:18 <lanius> dreimark_: ah have to merge main first
2007-07-02T22:18:20 <lanius> xorAxAx: not yet
2007-07-02T22:19:04 <CIA-20> moin: Heinrich Wendel <h_wendel@cojobo.net> * 2125:c6716efbad22 1.7-storage-hwendel/MoinMoin/action/AttachFile.py: removed unused code
2007-07-02T22:19:05 <CIA-20> moin: Heinrich Wendel <h_wendel@cojobo.net> * 2126:049864ddf737 1.7-storage-hwendel/MoinMoin/action/PackagePages.py: remove getPagePath where possible
2007-07-02T22:19:55 <lanius> xorAxAx: but it can be implemented without any change to your code
2007-07-02T22:20:16 <lanius> xorAxAx: when getting the item a lock will be done which will be released with save
2007-07-02T22:20:28 <lanius> xorAxAx: any other write operation has to wait for the lock
2007-07-02T22:21:49 <xorAxAx> lanius: and what if i dont modify anything?
2007-07-02T22:22:17 <lanius> then i have to think about it again
2007-07-02T22:22:22 <xorAxAx> and how does that fix the dirty write problem again ...what was our solution ...
2007-07-02T22:22:34 <lanius> not fix it was our solution
2007-07-02T22:24:55 <skipm> dreimark, neagulm : are we finished for today?
2007-07-02T22:25:55 <dreimark_> neagulm: any more questions?
2007-07-02T22:26:00 <xorAxAx> lanius: and how can i lock single pages for writes?
2007-07-02T22:28:11 <lanius> xorAxAx: that's something i have to implement yet
2007-07-02T22:28:53 <dreimark_> skipm: I believe neagulm is a bit busy with his diploma exam. too.
2007-07-02T22:29:27 <xorAxAx> lanius: i think thats the prerequisite i need before spending time on the code
2007-07-02T22:29:55 <neagulm> dreimark_, skipm :) I was reading
2007-07-02T22:30:16 <dreimark_> hehe
2007-07-02T22:30:41 <neagulm> skipm, a last question: we want to use 2 separate classifiers for pages and attachments. What do you think?
2007-07-02T22:31:39 <lanius> xorAxAx: ok
2007-07-02T22:31:41 <neagulm> skipm, dreimark_ Currently we can extract data from: opendocument, pdf, etc. using the available MM filters
2007-07-02T22:32:53 <skipm> neagulm: what's the motivation for two classifiers? in my experience with wiki spam it is generally spam in the page or spam in the attachment (usually an html file). either will be full of clues for a spam checker to munch on.
2007-07-02T22:33:17 <neagulm> skipm, dreimark xorAxAx sugested to use a separate classifier for attachments in order to avoid problems with unrevisioned attachments
2007-07-02T22:33:44 <xorAxAx> neagulm: no, just a different corpus
2007-07-02T22:33:44 <skipm> neagulm: not sure I understand the connection
2007-07-02T22:34:32 <skipm> xorAxAx: or the motivation for different corpus - that's just going to be extra complication
2007-07-02T22:34:55 <xorAxAx> skipm: well, he is chaining attachments to pages if they are linked from the page
2007-07-02T22:35:20 <skipm> xorAxAx: sure - they'll be related won't they?
2007-07-02T22:35:23 <xorAxAx> skipm: thats a large design issue that i was refering to - i dont care about the actual solution of fixing the resulting problems
2007-07-02T22:35:35 <xorAxAx> skipm: the issue is updating the corpus if an attachment changes
2007-07-02T22:35:43 <xorAxAx> he doesnt have a reverse mapping of attachment -> page
2007-07-02T22:35:55 <xorAxAx> and he didnt handle attachments refered from other pages
2007-07-02T22:36:15 <skipm> xorAxAx: i feel like i'm operating with incomplete inputs again...
2007-07-02T22:36:27 <xorAxAx> well
2007-07-02T22:36:36 <xorAxAx> imagine sombody deletes an attachment
2007-07-02T22:36:43 <xorAxAx> he cant know on which pages it was linked
2007-07-02T22:36:47 <xorAxAx> without checking every page
2007-07-02T22:37:00 <skipm> i don't understand why we care about the relationship of an attachment to a page - either it's spam or it's not
2007-07-02T22:37:13 <xorAxAx> skipm: me neither
2007-07-02T22:37:27 <skipm> so two use two corpora?
2007-07-02T22:37:40 <skipm> eh,,, so *why* use two corpora?
2007-07-02T22:37:41 <xorAxAx> not necessarily, one would be fine ... the problem was the connection
2007-07-02T22:37:49 <xorAxAx> that he implemented
2007-07-02T22:37:57 <dreimark_> skipm: xorAxAx how does retrain work if the database is lost and the files are gone
2007-07-02T22:38:20 <dreimark_> where it is based on
2007-07-02T22:38:35 <neagulm> skipm, the problem is with retraining
2007-07-02T22:38:54 <skipm> neagulm: in sb you generally just retrain from scratch. it only takes a couple of each kind of submission for it to start classifying reasonably well
2007-07-02T22:39:28 <skipm> but i thought attachments were stored in subdirectories of their pages
2007-07-02T22:39:49 <dreimark_> skipm: neagulm retraining from scratch should be ommited
2007-07-02T22:40:31 <dreimark_> omitted
2007-07-02T22:41:10 <skipm> dreimark: how do you handle the initial case? That's effectively retraining from scratch
2007-07-02T22:42:10 <neagulm> skipm, The users start with an untrained classifier and do Train-On-Error.
2007-07-02T22:42:22 <dreimark_> the first is ok, but we have to think on updates too. so while we collectspam and ham pages we could use them to retrain
2007-07-02T22:42:36 <dreimark_> and if the database is lost too
2007-07-02T22:45:29 <dreimark_> skipm: currently we have only one version of an attachment, that will change during develop. if 1.7
2007-07-02T22:46:12 <skipm> dreimark: yes, but shouldn't you know if the new attachment is ham or spam before you overwrite its predecessor?
2007-07-02T22:48:02 <neagulm> skipm, dreimark_ Until now we did not impose a policy related to spam. We just warn the user, that will be the default behavior with attachments to.
2007-07-02T22:48:31 <neagulm> skipm, dreimark brb
2007-07-02T22:49:19 <skipm> neagulm: you can't overwrite an existing attachment without first deleting it, so I suppose that's fine (you've lost the "hammy" attachment before you've even seen the "spammy
2007-07-02T22:49:22 <skipm> one
2007-07-02T22:50:34 <ThomasWaldmann> lanius: no, no local event log, but the idea is good :)
2007-07-02T22:51:10 <dreimark_> skipm: sure, the splitting of categorisation is our current idea to solve the problem retraining with revisions of pages while we don't be able to do it with revisions of attachments
2007-07-02T22:58:12 <neagulm> dreimark_, skipm sorry but i will have to leave. Skipm i will get back to you with an e-mail about the tokenization, attachments, and the SB 1.1a4 interface. I expect to be able to do so after Friday.
2007-07-02T22:58:44 <skipm> neagulm: no problem. i have to take off soon as well - good luck with your exams
2007-07-02T22:58:59 <dreimark_> neagulm: from me too
2007-07-02T22:59:07 <neagulm> skipm, thank you... i will need it :D
2007-07-02T22:59:10 <dreimark_> neagulm: good night
2007-07-02T22:59:27 <neagulm> dreimark_, skipm good night and thank you again
2007-07-02T23:06:08 <dreimark_> gn
2007-07-02T23:33:30 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2286:01f05e74aa9c 1.7/ (212 files in 55 dirs): Big PEP8 and whitespace cleanup
2007-07-02T23:37:34 <xorAxAx> looks like people will want to merge as soon as possible
2007-07-02T23:38:43 <ThomasWaldmann> yes, and run py.test afterwards and fix their own stuff
2007-07-02T23:41:43 * ThomasWaldmann removed the PEP8 errors that triggered too often
2007-07-02T23:49:39 <CIA-20> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2287:0005f996ff92 1.7/MoinMoin/wikidicts.py: wikidicts: add comment about (empty) dict entry syntax MoinMoin: MoinMoinChat/Logs/moin-dev/2007-07-02 (last edited 2007-10-29 19:09:21 by localhost)
