2007-10-30T00:03:01  <dreimark> http://moinmo.in/TitleIndex shows pages below action/
2007-10-30T00:04:37  <dreimark> gn
2007-10-30T03:31:31  <PicoPlex> Hi - I see moinmo.in is read-only due to migration problems. I was wanting to record a bug I'm constantly getting "RuntimeError: maximum recursion depth exceeded in cmp" errors with a new build of 1.6.  FIrst got when reindexing with Xapian now get on searching and most page accesses - is there a problem with the latest code?
2007-10-30T03:31:57  <PicoPlex> (I have disabled Xapian and it makes no difference)
2007-10-30T03:48:53  <PicoPlex> Created a new wiki and problem has gone away - strange!
2007-10-30T08:46:13  <dreimark> moin
2007-10-30T08:46:22  <ThomasWaldmann> moin
2007-10-30T08:48:16  <ThomasWaldmann> PicoPlex: remember what you wanted to write until it is r/w again. you can save the traceback.html somewhere and give me the url.
2007-10-30T08:49:07  <dreimark> e.g. http://test.wikiwikiweb.de/
2007-10-30T12:37:22  * ThomasWaldmann repaired moinmaster's edit-logs
2007-10-30T13:19:26  <johill> were there any unconverted pages w/o #format?
2007-10-30T13:20:03  <johill> I have a wishlist item: I have a parser 'inline_latex' that inherits from 'wiki' and behaves almost identically, it'd be good to have a way to specify that such pages should also be converted
2007-10-30T13:24:40  <ThomasWaldmann> johill: i didnt find any yet
2007-10-30T13:25:19  <johill> I skimmed the log and dreimark seemed to think there were some
2007-10-30T13:25:27  <ThomasWaldmann> i fixed the 100000000 stuff manually and the macros/links in pre sections have to get fixed manually anyway.
2007-10-30T13:26:13  <ThomasWaldmann> johill: moinmo.in got a new converter output yesterday
2007-10-30T13:26:44  <ThomasWaldmann> johill: master.moinmo.in was running on 1.6 since a longer time and needs to get fixed manually
2007-10-30T13:28:34  <johill> yeah I figured master would have to be fixed
2007-10-30T13:29:34  <johill> I wish the docs were in the sources though
2007-10-30T13:29:42  <johill> maybe we can have that once we have backends
2007-10-30T13:29:54  <johill> we could write a simple backend that doesn't do page revisions and just keeps flat files
2007-10-30T13:31:19  <ThomasWaldmann> hm?
2007-10-30T13:32:25  <ThomasWaldmann> about other #format than wiki: this is an open problem anyway, esp. for rst
2007-10-30T13:33:07  <ThomasWaldmann> (there is no converter)
2007-10-30T13:33:27  <xorAxAx> johill: ping
2007-10-30T13:33:39  <xorAxAx> johill: ahead of you :)
2007-10-30T13:33:48  <johill> ahead of me? :)
2007-10-30T13:33:54  <johill> almost certainly good ;)
2007-10-30T13:34:04  <xorAxAx> i am ahead of you :)
2007-10-30T13:34:15  <johill> good :)
2007-10-30T13:34:25  <xorAxAx> at the same table ...
2007-10-30T13:34:35  <johill> have you written a backend already then?
2007-10-30T13:34:54  <xorAxAx> i mean, in front of, gah :)
2007-10-30T14:09:54  <CIA-32> moin: Johannes Berg <johannes AT sipsolutions DOT net> default * 2931:fff3e18fb5e4 1.7/MoinMoin/action/edit.py: remove LOL multiple buttons message
2007-10-30T14:12:12  <ThomasWaldmann> hehe
2007-10-30T14:12:37  <ThomasWaldmann> did it flood your log?
2007-10-30T14:29:08  <johill> not exactly flood, but it seemed pretty useless noise in it, yeah
2007-10-30T14:29:38  <CIA-32> moin: Johannes Berg <johannes AT sipsolutions DOT net> default * 2932:27d192ebd485 1.7/MoinMoin/request/__init__.py: give traces for duplicate headers
2007-10-30T14:33:57  <CIA-32> moin: Johannes Berg <johannes AT sipsolutions DOT net> default * 2933:5ad34b3541e6 1.7/MoinMoin/request/__init__.py: only set Expires: header once for cache disabling
2007-10-30T14:35:03  <johill> how easy it gets to fix the duplicate header bugs once you know where they're coming from...
2007-10-30T14:40:35  <CIA-32> moin: Johannes Berg <johannes AT sipsolutions DOT net> default * 2934:2521eb511f77 1.7/MoinMoin/request/ (__init__.py _tests/test_request.py): log duplicate header emissions with tracebacks
2007-10-30T14:42:22  <ThomasWaldmann> johill: :)
2007-10-30T14:43:14  <johill> seriously :)
2007-10-30T14:44:41  * johill goes back to watching logs
2007-10-30T14:51:22  <johill> at least no more duplicate headers now
2007-10-30T14:57:23  <CIA-32> moin: Johannes Berg <johannes AT sipsolutions DOT net> default * 2935:3670e156ec1a 1.7/MoinMoin/request/ (__init__.py _tests/test_request.py): request: use sent_headers rather than sent_headers_trace
2007-10-30T14:59:01  <ThomasWaldmann> johill: after you have tested that enough (esp. if editing still has one no-cache header set), a backport to 1.6 would be nice
2007-10-30T14:59:56  <johill> editing?
2007-10-30T15:00:09  <johill> I didn't change behaviour, only logging
2007-10-30T15:00:34  <ThomasWaldmann> (14:33) <    CIA-32> moin: Johannes Berg <johannes AT sipsolutions DOT net> default * 2933:5ad34b3541e6 1.7/MoinMoin/request/__init__.py: only set  Expires: header once for cache disabling
2007-10-30T15:01:08  <johill> ah
2007-10-30T15:01:20  <johill> well it's already set in the first round of disableHttpCaching
2007-10-30T15:01:30  <johill> so it can't be missing really
2007-10-30T15:01:54  <johill> heh
2007-10-30T15:02:02  <johill> google doesn't like redirects in a sitemap
2007-10-30T15:02:28  <johill> ThomasWaldmann: backport just the fix?
2007-10-30T15:03:00  <johill> ?action=edit definitely has Expires: in the header
2007-10-30T15:04:07  <ThomasWaldmann> johill: yes, after you are very sure about it :)
2007-10-30T15:05:25  <johill> I am :)
2007-10-30T15:05:35  <CIA-32> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> default * 2286:cf95506b82f4 1.6/MoinMoin/i18n/tools/ (markup15to16.py wiki2po.py): i18n tools: fix small bugs
2007-10-30T15:05:38  <CIA-32> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> default * 2287:0c329e7bc571 1.6/MoinMoin/i18n/id.MoinMoin.po: added indonesian (id) i18n
2007-10-30T15:05:39  <CIA-32> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> default * 2288:cd9f0aa1f945 1.6/MoinMoin/i18n/ (40 files): updated i18n
2007-10-30T15:05:55  <xorAxAx> johill: you can only be sure if there are unit tests! :)
2007-10-30T15:07:06  <johill> uh
2007-10-30T15:07:10  <johill> 1.6 already has that fix
2007-10-30T15:07:11  <johill> grrrrr
2007-10-30T15:07:37  <johill> xorAxAx: yeah sadly, moin's code is in a state where too many comments are just wrong and abstractions are not honoured
2007-10-30T15:08:07  <xorAxAx> well, you can always go more into the direction of blackbox testing if this an issue (use paste fixture etc.)
2007-10-30T15:12:55  <johill> no I rather mean that the issue is that with moin you can never be sure where else an object's instance variable is used
2007-10-30T15:13:03  <johill> because such interfaces aren't proplery defined
2007-10-30T15:14:18  <xorAxAx> ?
2007-10-30T15:14:21  <xorAxAx> unittests help here
2007-10-30T15:14:40  <xorAxAx> they will fail if somebody is (illegaly) use such var
2007-10-30T15:16:54  <johill> depends
2007-10-30T15:17:04  <johill> the request tests for example actually use such a var to reset the request
2007-10-30T15:17:38  <CIA-32> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> default * 2289:933647e940bd 1.6/MoinMoin/i18n/tools/ (3 files): update text for mailing language maintainers, add todo to the code
2007-10-30T15:26:05  <johill> empty log so far, goodie
2007-10-30T15:59:24  <CIA-32> moin: Johannes Berg <johannes AT sipsolutions DOT net> default * 2936:c71beaa6a978 1.7/MoinMoin/action/fullsearch.py: remove another LOL log
2007-10-30T16:04:19  <johill> aahhrrgg
2007-10-30T16:05:09  <johill> why don't we use the argument parser stuff for format_args?
2007-10-30T16:20:15  <ThomasWaldmann> maybe because this was a quick hack for 1.6 and that code was not there at that time?
2007-10-30T16:20:52  <ThomasWaldmann> (meanwhile the stuff is in wikiutil)
2007-10-30T16:24:25  <johill> xorAxAx: how about ranges like svn or so do: "1.5:2.0,2.5:3.0" or "1.6:"
2007-10-30T16:25:14  <ThomasWaldmann> johill: btw, the logging is currently at debug level
2007-10-30T16:25:30  <johill> hmm
2007-10-30T16:25:36  <johill> well actually, fcgi doesn't have all that logging yet
2007-10-30T16:25:38  <ThomasWaldmann> (by default at least)
2007-10-30T16:25:51  <johill> ThomasWaldmann: is it possible to have page caches outside of the page dir?
2007-10-30T16:26:03  <ThomasWaldmann> btw, i changed the fcgi stuff recently
2007-10-30T16:26:22  <ThomasWaldmann> sure, under data/cache
2007-10-30T16:26:38  <ThomasWaldmann> ehrm, not sure about PAGE caches
2007-10-30T16:27:29  <ThomasWaldmann> (btw, that is for storage refactoring branch)
2007-10-30T16:27:53  <johill> I guess..
2007-10-30T16:28:16  <ThomasWaldmann> do you use the new fcgi stuff?
2007-10-30T16:29:39  <johill> I guess not
2007-10-30T16:32:14  <ThomasWaldmann> btw, we (==not me alone :) should review 1.6 and 1.7, I guess there is quite some stuff that needs forward/backporting
2007-10-30T16:32:34  <ThomasWaldmann> (after finishing the converter trouble)
2007-10-30T16:33:22  <johill> yeah quite possible
2007-10-30T16:34:13  <johill> that there's lots to port, I mean
2007-10-30T16:53:43  <johill> http://johannes.sipsolutions.net/files/version.py.txt
2007-10-30T16:53:53  <johill> xorAxAx: how about that?
2007-10-30T17:10:55  <xorAxAx> johill: i dont see the unitest :)
2007-10-30T17:12:29  <xorAxAx> and it doesnt handle the discussed #format issue, but besides that its fine
2007-10-30T17:13:14  <ThomasWaldmann> do we have any converter-related unresolved major issues left on moinmo.in?
2007-10-30T17:40:45  <ThomasWaldmann> anybody?
2007-10-30T20:11:52  <dreimark> bb
2007-10-30T20:13:02  <ThomasWaldmann> wb dreimark :)
2007-10-30T20:13:08  <dreimark> ThomasWaldmann: that page were inside {{{ an url was changed to [[url
2007-10-30T20:13:12  <dreimark> is that fixed too
2007-10-30T20:13:25  <dreimark> ?
2007-10-30T20:13:50  <ThomasWaldmann> no
2007-10-30T20:14:10  <ThomasWaldmann> (no idea why that happens)
2007-10-30T20:14:19  <ThomasWaldmann> all that {{{}}} stuff is strange
2007-10-30T20:14:24  <dreimark> yeah
2007-10-30T20:22:27  <CIA-32> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> default * 2290:2eb73b579a5a 1.6/MoinMoin/ (config/multiconfig.py i18n/dummy.py): fix small i18n issues
2007-10-30T20:22:28  <CIA-32> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> default * 2291:f9414749caf4 1.6/MoinMoin/i18n/ (40 files): updated i18n
2007-10-30T20:24:54  <dreimark> ThomasWaldmann: I wonder about a link to MoinMoinBugs/DeletePageUsesInsecure its now MoinMoinBugs/DeletePageUsesInsecureValidation
2007-10-30T20:25:21  <dreimark> Is it correct that I have to fix manually my links ?
2007-10-30T20:26:10  <ThomasWaldmann> what links?
2007-10-30T20:26:24  <ThomasWaldmann> url?
2007-10-30T20:26:33  <dreimark> the add links links in my personal tabs
2007-10-30T20:27:04  <ThomasWaldmann> if it got renamed by the migration, it should still match
2007-10-30T20:27:13  <ThomasWaldmann> if it got renamed before, it broke before
2007-10-30T20:27:40  <dreimark> ok
2007-10-30T20:27:58  <dreimark> if we rename in 1.6 a page does it change user tabs ?
2007-10-30T20:28:08  <ThomasWaldmann> no
2007-10-30T20:28:25  <dreimark> I'll add for that a FR later on
2007-10-30T20:29:10  * ThomasWaldmann is a bit reluctant with switching to rw until some people say it's ok now :)
2007-10-30T20:30:43  <dreimark> ThomasWaldmann: can you test if  the log entry of the 1.6 converter is always ge the last revison
2007-10-30T20:35:00  <ThomasWaldmann> dreimark: ?
2007-10-30T20:36:14  <dreimark> most pages I ve seen the changes of the converter increases the page content by some bytes if there are changes.
2007-10-30T20:36:21  <ThomasWaldmann> ge = greater-or-equal?
2007-10-30T20:36:29  <dreimark> yeah
2007-10-30T20:36:31  <ThomasWaldmann> and what should that be good for?
2007-10-30T20:36:55  <dreimark> to see if there some pages could be completly broken
2007-10-30T20:37:20  <ThomasWaldmann> completely broken means?
2007-10-30T20:37:59  <dreimark> some text is missing
2007-10-30T20:38:33  <ThomasWaldmann> i cant easily check that
2007-10-30T20:39:15  <ThomasWaldmann> and in some cases new markup is shorter than before
2007-10-30T20:40:22  <dreimark> indeed e.g. http://moinmo.in/AdamShand
2007-10-30T20:40:38  <dreimark> 3 bytes shorter :)
2007-10-30T20:44:59  <ThomasWaldmann> wiki:/Self: is not needed any more
2007-10-30T20:46:01  <dreimark> ThomasWaldmann: this action url for info should be changed into something not used on MM yet. e.g. process, instance
2007-10-30T20:48:28  <ThomasWaldmann> or axn :)
2007-10-30T20:49:24  <ThomasWaldmann> xschn?
2007-10-30T20:51:41  <ThomasWaldmann> 4c710n?
2007-10-30T20:54:10  <dreimark> i dont get the phonetics of the last one
2007-10-30T20:56:49  <ThomasWaldmann> http://www.brenz.net/l337Maker.asp
2007-10-30T20:57:51  <dreimark> cool
2007-10-30T21:09:47  <dreimark> ThomasWaldmann: I verified yesterday some more page as today and dont get more failures as the known {{{ }}} pages
2007-10-30T21:10:38  <dreimark> so I do vote for rw
2007-10-30T21:54:35  <dreimark> someone asked me if action move works well in mod_python by email. I dont have mod_python so I cant check.
2007-10-30T21:57:09  <dreimark> so have to leave,
2007-10-30T21:57:11  <dreimark> gn all
2007-10-30T22:11:04  <ThomasWaldmann> ok, moinmo.in is rw again

MoinMoin: MoinMoinChat/Logs/moin-dev/2007-10-30 (last edited 2007-12-22 22:45:02 by IrcLogImporter)