2007-07-15T00:03:00 <dreimark> gn
2007-07-15T00:04:05 <grzywacz> Hm.
2007-07-15T00:04:08 <grzywacz> Sure, I can try.
2007-07-15T00:04:44 <grzywacz> But how could that be a broken merge?
2007-07-15T00:04:59 <CIA-27> moin: Ville-Pekka Vainio <vpivaini AT cs DOT helsinki DOT fi> * 2458:4053a2593aac 1.7-maninfo-vpv/MoinMoin/script/xmlrpc/manimport.py: ismanfile() didn't recognize all man file names, fix that. also fix one weirdness in directory handling
2007-07-15T00:05:02 <CIA-27> moin: Ville-Pekka Vainio <vpivaini AT cs DOT helsinki DOT fi> * 2459:eab986c3fed3 1.7-maninfo-vpv/MoinMoin/script/xmlrpc/manimport.py: typo in last commit
2007-07-15T00:05:06 <CIA-27> moin: Ville-Pekka Vainio <vpivaini AT cs DOT helsinki DOT fi> * 2460:24ae558ec884 1.7-maninfo-vpv/MoinMoin/script/xmlrpc/manimport.py: handle man page inclusion
2007-07-15T00:05:12 <CIA-27> moin: Ville-Pekka Vainio <vpivaini AT cs DOT helsinki DOT fi> * 2461:3f45fb8b53c2 1.7-maninfo-vpv/MoinMoin/script/xmlrpc/manimport.py: handle 'clear' doclifter errors, some other small fixes
2007-07-15T00:07:06 <xorAxAx> grzywacz: well, it would contain stuff that is not replayable on the ancestors, no idea how to produce thta
2007-07-15T00:07:12 <xorAxAx> grzywacz: but #mercurial will know
2007-07-15T00:08:42 <grzywacz> Oh.
2007-07-15T00:08:45 <grzywacz> s/h/k/
2007-07-15T00:09:10 <xorAxAx> but kow?
2007-07-15T00:09:21 <grzywacz> "Ok."
2007-07-15T00:09:22 <grzywacz> :P
2007-07-15T00:09:25 <xorAxAx> ah
2007-07-15T00:09:29 <xorAxAx> %-)
2007-07-15T00:09:30 <grzywacz> :D
2007-07-15T00:09:50 <grzywacz> Ok, I'll get an old rev and update to the next merge points.
2007-07-15T00:09:55 <grzywacz> Maybe I'll find the faulty one.
2007-07-15T00:22:18 <grzywacz> I think this one might have broken stuff:
2007-07-15T00:22:20 <grzywacz> http://hg.alexanderweb.de/moin-1.7-main/rev/01f05e74aa9cc477d610eaf432a119e74df94afc
2007-07-15T00:22:44 <grzywacz> It fixes messages.py, that has been removed in my branch long before, but not merged to main/Thomas' branch.
2007-07-15T00:23:14 <grzywacz> That's my little deduction from branchview, now I have to test it. ;>
2007-07-15T00:23:23 <xorAxAx> yeah, but the question is why a simple pulling operation into a repo with one head of a repo with one head gives conflicts that need to be merged
2007-07-15T00:28:12 <ThomasWaldmann> tests on 1.7-classify completely fail if you don't have bayes stuff installed, this should be improved
2007-07-15T00:30:14 <grzywacz> Lol, found it. Pulling one revision after next introduces a conflict. O_o;
2007-07-15T00:46:02 <xorAxAx> ThomasWaldmann: can you reconstruct the tip of your local repo before pulling?
2007-07-15T00:48:19 <ThomasWaldmann> it was just a clean old rev, but i dont remember which
2007-07-15T00:49:29 <xorAxAx> yeah, i reproduced it
2007-07-15T00:55:18 <xorAxAx> grzywacz: now you may monitor this issue and throw that command line at various peeps until somebody figures out the reason :)
2007-07-15T00:55:30 <grzywacz> ;-)
2007-07-15T00:58:16 <grzywacz> xorAxAx, I don't know who is who on #mercurial, so I can't really target this cmd properly ;(
2007-07-15T00:58:28 <xorAxAx> me neither
2007-07-15T00:58:32 <xorAxAx> thomasah is nice, though
2007-07-15T00:58:50 <xorAxAx> he has a family and is german so he is not online currently
2007-07-15T00:59:12 <grzywacz> He is, but prolly away. ;>
2007-07-15T01:08:05 <xorAxAx> grzywacz: asak is very good at debugging such stuff as wel
2007-07-15T01:08:21 <grzywacz> ok
2007-07-15T01:08:57 <xorAxAx> brendan as well :)
2007-07-15T01:10:31 <ThomasWaldmann> http://hg.moinmo.in/moin/1.7-jabber-knowak/rev/aad0a118e609 ?
2007-07-15T01:10:53 <grzywacz> -_-;
2007-07-15T01:11:12 <grzywacz> No comment.
2007-07-15T01:36:45 <grzywacz> ...
2007-07-15T01:37:01 <xorAxAx> hmm? :)
2007-07-15T01:37:28 <grzywacz> I spent like 30 minutes debugging why I don't get error messages from my bot when I request an invalid page.
2007-07-15T01:37:44 <grzywacz> And I've just realised that I'm DnD and it's queueing them politely not to disturb me.
2007-07-15T01:37:45 <grzywacz> -_-;
2007-07-15T01:38:11 <xorAxAx> umm, b0rked by design :)
2007-07-15T01:38:24 <xorAxAx> it should only queue asnyc responses
2007-07-15T01:38:30 <grzywacz> So true.
2007-07-15T01:38:53 <xorAxAx> :)
2007-07-15T01:39:12 <grzywacz> Simple fix coming.
2007-07-15T01:43:15 <grzywacz> Fixed.
2007-07-15T01:43:19 <grzywacz> Funny bug, though. ;)
2007-07-15T05:14:40 <CIA-27> moin: Karol 'grzywacz' Nowak <grzywacz@sul.uni.lodz.pl> * 2383:4ac1c1635aba 1.7-jabber-knowak/MoinMoin/events/xapian_index.py: Remove a stray debugging print.
2007-07-15T05:14:42 <CIA-27> moin: Karol 'grzywacz' Nowak <grzywacz@sul.uni.lodz.pl> * 2384:ddb4c626062d 1.7-jabber-knowak/jabberbot/ (commands.py xmlrpcbot.py xmppbot.py):
2007-07-15T05:14:42 <CIA-27> moin: Big checkpoint commit. Details below:
2007-07-15T05:14:44 <CIA-27> moin: - queue only async notifications received from wiki
2007-07-15T05:14:46 <CIA-27> moin: - discover if client supports data forms when contact comes online
2007-07-15T05:14:48 <CIA-27> moin: - template for search form, beginning of search command
2007-07-15T05:14:50 <CIA-27> moin: - all commands are now case-insensitive
2007-07-15T05:14:52 <CIA-27> moin: - other minor bug fixes
2007-07-15T05:15:03 <CIA-27> moin: Karol 'grzywacz' Nowak <grzywacz@sul.uni.lodz.pl> * 2385:6b70cafe0967 1.7-jabber-knowak/ (4 files in 2 dirs): Merge with devel.
2007-07-15T05:46:42 <CIA-27> moin: Karol 'grzywacz' Nowak <grzywacz@sul.uni.lodz.pl> * 2386:ae44e3266486 1.7-jabber-knowak/jabberbot/ (commands.py xmppbot.py): s/search/searchform/, add a Search command (does nothing)
2007-07-15T05:46:43 <CIA-27> moin: Karol 'grzywacz' Nowak <grzywacz@sul.uni.lodz.pl> * 2387:0f9b7b20fb9c 1.7-jabber-knowak/jabberbot/ (commands.py xmppbot.py): Whitespace fixes.
2007-07-15T05:46:47 <CIA-27> moin: Karol 'grzywacz' Nowak <grzywacz@sul.uni.lodz.pl> * 2388:75817589c437 1.7-jabber-knowak/jabberbot/ (commands.py xmppbot.py): Merge devel.
2007-07-15T08:49:35 <johill> ThomasWaldmann: reason is that it breaks underlay badly to allow ['...'] quoting
2007-07-15T08:49:48 <johill> ThomasWaldmann: for example help on configuration: ['text_html'], help on userpreferences, ...
2007-07-15T10:30:13 <johill> ThomasWaldmann: see http://test.wikiwikiweb.de/HelpOnConfiguration/UserPreferences for example
2007-07-15T10:37:33 <johill> if you feel that ['..'] and wiki:'..' formatting should be allowed then lets document it, add test cases, revert my changeset and adjust the link converter script to modify it to [``' in existing cases
2007-07-15T10:51:59 <dreimark> Example for escaping " is "``Escaping``"
2007-07-15T10:55:20 <johill> hm?
2007-07-15T10:55:43 <johill> no, I think the idea was that you could write ['link-with-"-quote']
2007-07-15T11:05:04 <dreimark> ok, but then someone later will ask for a mixup of ' and " and then we will have another nesting problem
2007-07-15T11:15:55 <johill> precisely
2007-07-15T11:16:20 <johill> hence, I don't see what this buys us (and consequently removed it because it makes trouble)
2007-07-15T11:19:53 <xorAxAx> dreimark: thats an invalid reason to have it like it was because it still doesnt allow you to mix them
2007-07-15T11:20:12 <xorAxAx> you need to be able to mix freely if you want to introduce such quirks otherwise you missed the point
2007-07-15T11:21:20 <xorAxAx> dreimark: how did you redo your file moves in the repo?
2007-07-15T11:21:31 <xorAxAx> dreimark: manually or via import?
2007-07-15T11:21:33 <dreimark> xorAxAx: it needs to be different solved
2007-07-15T11:21:58 <xorAxAx> yes
2007-07-15T11:22:27 <dreimark> xorAxAx: on 1.6 I did the whole changeset again by hand using hg mv no import
2007-07-15T11:23:00 <dreimark> I did an export of that new changeset
2007-07-15T11:23:34 <dreimark> and an import of that to 1.7 which doesn't work for me
2007-07-15T11:24:20 <dreimark> but I have only 0.9.3 may be it gives better results with 0.9.4
2007-07-15T11:26:39 <dreimark> I have tried with a repo without the commits of grzywacz of the wrong done changes
2007-07-15T11:27:23 <johill> funky. http://timhatch.com/projects/pybraces/
2007-07-15T11:28:28 <xorAxAx> its not the first braces language for the python vm, johill
2007-07-15T11:28:42 <xorAxAx> the vm-greek wrote one before
2007-07-15T11:28:43 <xorAxAx> NIH
2007-07-15T11:29:05 <johill> still a cute idea to implement it as an encoding filter
2007-07-15T11:33:41 <dreimark> I prefer that idea too something like wikiutil.link_escape
2007-07-15T11:35:52 <xorAxAx> johill: yes :)
2007-07-15T11:36:03 <xorAxAx> johill: the greek wrote a whole new compiler
2007-07-15T11:41:05 <xorAxAx> hehe, the pybraces guy held a lightning talk
2007-07-15T11:42:56 <johill> dreimark: what idea?
2007-07-15T11:48:07 <dreimark> johill: doesn't find it I was sure ThomasWaldmann has written something about we need to escape a " or a '
2007-07-15T11:49:11 <johill> hm
2007-07-15T11:49:33 <dreimark> or we have no escape char something like this
2007-07-15T11:49:36 <johill> I can see why we want to be able to have a link with a " in it
2007-07-15T11:49:45 <johill> but we can easily achieve that by doubling it
2007-07-15T11:50:04 <johill> or something else anyway
2007-07-15T11:50:11 <johill> I don't see a point in allowing ' and " but not mixed
2007-07-15T11:50:37 <johill> and then inventing something special for mixed yet again at some point
2007-07-15T11:50:49 <xorAxAx> yes
2007-07-15T11:50:58 <xorAxAx> Fabi: dont forget the mentor survey :)
2007-07-15T11:51:16 <Fabi> when does it end
2007-07-15T11:51:42 <xorAxAx> monday
2007-07-15T11:51:50 <grzywacz> Not sunday?
2007-07-15T11:51:56 <grzywacz> Moin.
2007-07-15T11:52:04 <xorAxAx> no idea
2007-07-15T11:52:08 <xorAxAx> moin grzywacz
2007-07-15T11:52:24 <xorAxAx> grzywacz: see the trouble we summoned upon our neighbourchannel #mercurial
2007-07-15T11:52:33 * grzywacz looks
2007-07-15T11:53:32 <zenhase> moin
2007-07-15T11:54:33 <dreimark> looks too
2007-07-15T11:55:50 <grzywacz> xorAxAx, they seem quite busy getting this solved :)
2007-07-15T11:55:51 * johill resists the temptation :P
2007-07-15T11:56:00 <xorAxAx> grzywacz: yes :)
2007-07-15T11:56:18 <xorAxAx> the only problem is that there are two kinds of people
2007-07-15T11:56:28 <ThomasWaldmann> moin
2007-07-15T11:56:33 <xorAxAx> those who can reproduce it using the sequences in the bug report and those who need my command line
2007-07-15T11:57:07 <grzywacz> 8)
2007-07-15T11:57:11 <grzywacz> hi Thomas
2007-07-15T12:04:04 <dreimark> xorAxAx: is the mercurial channel logged too?
2007-07-15T12:04:14 <xorAxAx> no idea
2007-07-15T12:06:29 <ThomasWaldmann> johill: if it should stay like it is, we need to have an escaping mech for " (and fixed unit tests)
2007-07-15T12:07:03 <xorAxAx> well, we need one for mixed stuff otherwise as well :)
2007-07-15T12:07:03 <johill> I don't disagree, but I don't think ' is a good mechanism as it breaks existing wiki pages and doesn't allow mixing
2007-07-15T12:07:24 <johill> I suppose we could just do it like pascal and double the ""
2007-07-15T12:07:29 <ThomasWaldmann> xorAxAx: sure, but in less cases than now
2007-07-15T12:07:42 <xorAxAx> ThomasWaldmann: currently it is broken
2007-07-15T12:07:54 <xorAxAx> ThomasWaldmann: because ['text_html'] was not considered as a link
2007-07-15T12:08:04 <xorAxAx> even though it should be according to your rules
2007-07-15T12:08:40 <johill> does anyone have any objections to just doubling the quote?
2007-07-15T12:09:32 <xorAxAx> no, thats the sanest way
2007-07-15T12:09:34 <johill> it would only break things like ["page""text with quotes around"]
2007-07-15T12:09:42 <ThomasWaldmann> maybe rather \" like python/C does it
2007-07-15T12:10:00 <johill> which you can write almost equivalently as "["page"text with quotes around]" or ["page" "text with quotes around"]
2007-07-15T12:10:06 <johill> ThomasWaldmann: but then you need to escape \ again
2007-07-15T12:10:11 <johill> etc etc
2007-07-15T12:10:20 <johill> and only programmers think that's natural
2007-07-15T12:10:22 <ThomasWaldmann> johill: yeah, \\
2007-07-15T12:10:39 <johill> yeah but it complicates the RE
2007-07-15T12:10:49 <ThomasWaldmann> that's why I wanted to avoid this for most cases just using different quoting chars
2007-07-15T12:10:50 <johill> because then \\" is not a qwuoted quote
2007-07-15T12:11:27 <johill> yeah but that obviously breaks
2007-07-15T12:11:44 <johill> and doubling isn't really hard, and allows everything to be quoted
2007-07-15T12:13:07 <ThomasWaldmann> we might use more quoting some time
2007-07-15T12:13:45 <ThomasWaldmann> e.g. some people asked for a way to break a line without really breaking it, like \
2007-07-15T12:13:48 <ThomasWaldmann> and so on
2007-07-15T12:13:57 <johill> inside a link? insane
2007-07-15T12:14:06 <johill> and even so
2007-07-15T12:14:12 <ThomasWaldmann> no, not in a link, in general
2007-07-15T12:14:12 <johill> that should be done before all the other markup is processed
2007-07-15T12:14:32 <ThomasWaldmann> but you can't use " for that
2007-07-15T12:14:54 <johill> I don't understand
2007-07-15T12:14:58 <ThomasWaldmann> so \ looks more universally usable
2007-07-15T12:15:04 <johill> if you use \ for that it'll be even worse
2007-07-15T12:15:13 <johill> because then you have to write \\" inside a link to get "
2007-07-15T12:15:19 <johill> nah
2007-07-15T12:15:22 <johill> it all doesn't make sense
2007-07-15T12:15:30 <johill> I'd to the line continuation as a simple preprocdessor
2007-07-15T12:15:37 <ThomasWaldmann> no, \" -> "
2007-07-15T12:15:39 <johill> that removes '\\n'
2007-07-15T12:15:55 <johill> /I'd to/I'd do/
2007-07-15T12:16:44 <ThomasWaldmann> it's just about not using too many different escaping methods at different places
2007-07-15T12:17:24 <johill> then why did you add ' quoting? ;)
2007-07-15T12:17:36 <johill> kidding
2007-07-15T12:17:44 <johill> anyway, I don't really care much
2007-07-15T12:17:44 <ThomasWaldmann> that's quoting, not escaping :-P
2007-07-15T12:17:53 <johill> but I do know that \ escaping will complicate the RE a lot
2007-07-15T12:18:04 <ThomasWaldmann> it was the same idea as regexes use
2007-07-15T12:18:19 <ThomasWaldmann> you can write /x/y/ but also #x#y#
2007-07-15T12:18:25 <johill> and then if you do that all the sudden people will expect \<eol> to work
2007-07-15T12:18:41 <johill> that's just perl and sed not generic REs but anyway :)
2007-07-15T12:21:04 <johill> I just don't see the benefit of adding that complexity
2007-07-15T12:21:28 <johill> and using an arbitrary quoting char is totally bogus since that'd mean you can't ever use a bracket :)
2007-07-15T12:22:39 <johill> ultimately it doesn't matter to me whether we use \ or double the "
2007-07-15T12:22:49 <johill> I just don't see the benefit of adding that complexity into this code
2007-07-15T12:23:12 <ThomasWaldmann> this is why i just used ' and " (similar to what python does)
2007-07-15T12:23:24 <johill> I know but that obviously doesn't work
2007-07-15T12:23:36 <johill> breaks many existing pages
2007-07-15T12:23:47 <ThomasWaldmann> yeah, because \" and \' is missing :D
2007-07-15T12:23:56 <johill> no
2007-07-15T12:24:02 <johill> because it makes [' into link markup when it wasn't
2007-07-15T12:24:40 <johill> escaping wouldn't help at all
2007-07-15T12:24:47 <ThomasWaldmann> only if you change the parser's big ugly regex
2007-07-15T12:25:10 <johill> ?
2007-07-15T12:25:27 <ThomasWaldmann> it shouldnt recognize that if you don't
2007-07-15T12:25:58 <johill> ah but then we can emit markup that isn't recognised ;)
2007-07-15T12:27:51 <ThomasWaldmann> anyway, I am not a fan of doing it the pascal way because pascal isnt known by too many people and it is also a special hack to only solve this single problem (while me might also want to solve 1 or 2 more of this kind and then would need other methods for solving them, making the syntax more complicated as needed)
2007-07-15T12:28:17 <johill> I just said "pascal way" because it's a good description. the help wouldn't say that ;)
2007-07-15T12:28:26 <johill> and I don't see what other problems of this kind we'd get
2007-07-15T12:28:41 <johill> ".." would be able to contain everything except unprintable characters
2007-07-15T12:28:57 <ThomasWaldmann> line continuations
2007-07-15T12:29:09 <johill> but line continuations are a special thing above the syntax
2007-07-15T12:29:19 <johill> they have to be done before parsing anyway
2007-07-15T12:29:27 <johill> otherwise *nothing* will possibly work right
2007-07-15T12:29:31 <johill> think
2007-07-15T12:29:38 <johill> == heading \
2007-07-15T12:29:39 <johill> ==
2007-07-15T12:29:52 <ThomasWaldmann> (there might be even open feature requests for them)
2007-07-15T12:30:06 <johill> so if you want to add line continuations you really need to do that as a preprocessor step
2007-07-15T12:30:14 <ThomasWaldmann> yeah, sure, might need other processing, but for the user it is a similar thing
2007-07-15T12:30:15 <johill> and then it doesn't interfere with linking markup at all
2007-07-15T12:30:47 <johill> in fact it gets more complicated
2007-07-15T12:30:55 <johill> think ["abc\\
2007-07-15T12:31:02 <johill> "def"]
2007-07-15T12:31:11 <ThomasWaldmann> "if I need the next char specially treated, I use \"
2007-07-15T12:31:47 <ThomasWaldmann> johill: nobody would write that :)
2007-07-15T12:31:49 <johill> imho you're thinking as a programmer again
2007-07-15T12:31:59 <johill> nobody thinks of \n as a character
2007-07-15T12:32:52 <johill> imho it's far easier to teach people to use "" because then they don't have to worry about \ as well
2007-07-15T12:33:03 <johill> i.e. unless they need it they don't have to remember anything even if the use \
2007-07-15T12:33:08 <dreimark> we should be carefull with adding to much programming language features to the wiki markup
2007-07-15T12:33:21 <johill> and people who do need line continuations can be easily told to use \<eol>
2007-07-15T12:33:44 <johill> only a programmer thinks of these as unified
2007-07-15T12:34:13 <ThomasWaldmann> dreimark: if you just want to do what is currently possible, you dont need to use this
2007-07-15T12:35:15 <dreimark> but some of the requests people have is to have multiline macro calls, which will be solved automaticly if parsers and macro calls unified
2007-07-15T12:36:25 <ThomasWaldmann> that's not in the near future
2007-07-15T12:37:18 <ThomasWaldmann> btw, as far as the link conversion stuff is concerned, we are talking about 1.6, not 1.7
2007-07-15T12:38:08 <johill> ThomasWaldmann: but if you want to do what is currently possible you *still* have to remember that \ is special
2007-07-15T12:38:25 <xorAxAx> yes, that will break a lot of links
2007-07-15T12:38:44 <dreimark> I would prefer not to change it's behaviour for 1.6 because we have not much tests for the wiki parser currently
2007-07-15T12:38:54 <xorAxAx> dreimark: such tests are easy to write
2007-07-15T12:39:30 <ThomasWaldmann> dreimark: we need quoting in 1.6 because of blanks now just being blanks
2007-07-15T12:39:48 <ThomasWaldmann> (same for umlauts and other strange stuff)
2007-07-15T12:39:54 <dreimark> ah ok
2007-07-15T12:40:32 <johill> we already have " quoting anyway
2007-07-15T12:40:33 <ThomasWaldmann> no attachment:some%20text.txt but attachment:"some text.txt"
2007-07-15T12:40:53 <ThomasWaldmann> brb
2007-07-15T12:41:17 <johill> restating: imho "" is easier because it's a totally special case and influences nothing else (unlike \)
2007-07-15T12:41:34 <johill> and \<eol> is also special and influences nothing else unless you need it
2007-07-15T12:41:47 <johill> hence you only need to remember these two things if you actually need them
2007-07-15T12:42:08 <johill> with \ escaping you *always* need to remember that \ is special
2007-07-15T12:42:19 <johill> which is unnatural to non-programmers
2007-07-15T12:47:37 <xorAxAx> grzywacz: and i suggest you to tackle i18n soon
2007-07-15T12:47:51 <xorAxAx> because thats often an architectural problem
2007-07-15T12:52:18 <xorAxAx> grzywacz: + @param commands_in: an inpu command queue (from xmpp)
2007-07-15T12:57:13 <johill> anyway to skip the pep8 test?
2007-07-15T12:58:22 <xorAxAx> i suggest moving it somewhere else
2007-07-15T12:58:29 <ThomasWaldmann> johill: maybe I'll modify it soon to only look at stuff recently changed
2007-07-15T12:59:12 <xorAxAx> ThomasWaldmann: the collection is the slow thing, you seem to traverse too many folders
2007-07-15T12:59:36 <xorAxAx> johill: you can use -k to sleect only certain tests
2007-07-15T12:59:50 <johill> I can just give the test file too
2007-07-15T12:59:52 <ThomasWaldmann> it looks at everything *.py
2007-07-15T13:00:46 <xorAxAx> ThomasWaldmann: it traverses all folders
2007-07-15T13:00:51 <xorAxAx> too many ...
2007-07-15T13:01:00 <xorAxAx> johill: yes
2007-07-15T13:01:21 <xorAxAx> we dont have any source code in the underlay folder
2007-07-15T13:02:51 <ThomasWaldmann> we have
2007-07-15T13:03:03 <xorAxAx> but nothing that we want to check
2007-07-15T13:03:09 <xorAxAx> *sigh*
2007-07-15T13:03:11 <ThomasWaldmann> but I am looking at how to make it faster, right now
2007-07-15T13:03:25 <xorAxAx> also benchmark with cold cache
2007-07-15T13:03:43 <johill> /proc/sys/vm/drop_caches helps with that
2007-07-15T13:20:09 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 2114:be9403b04443 1.6/wiki/server/ (moin.cgi moin.fcg): fixing missing changes to example scripts for server and request renaming
2007-07-15T13:24:13 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2376:d1b05e0eb6ea 1.7/MoinMoin/_tests/test_sourcecode.py: test_sourcecode: save some ms by some more excludes
2007-07-15T13:38:41 <grzywacz> xorAxAx, hmmm, fine.
2007-07-15T13:40:40 <johill> hm
2007-07-15T13:40:45 <johill> can't make the RE work
2007-07-15T13:47:08 <grzywacz> xorAxAx, i18n isn't a simple task indeed.
2007-07-15T13:47:21 <grzywacz> xorAxAx, for example determining what language to use in communication with the bot
2007-07-15T13:47:57 <grzywacz> xorAxAx, using lang attribute would be the best, protocol-wise, but if someone's client doesn't support it for some obscure reasons, it *might* be good to get this info from wiki. Maybe.
2007-07-15T13:47:58 <xorAxAx> well
2007-07-15T13:48:06 <grzywacz> This would, of course, increase traffic and load. ;S
2007-07-15T13:48:22 <xorAxAx> thats the minor issue
2007-07-15T13:48:37 <xorAxAx> you have 2 sources of info: the lang attrib and the user profile
2007-07-15T13:48:38 <grzywacz> So, "i18n no matter what"? 8)
2007-07-15T13:48:50 <xorAxAx> the next question is - how do you load the strings? via a package dependency or via xmlrpc?
2007-07-15T13:49:17 <grzywacz> Strings?
2007-07-15T13:49:29 <grzywacz> I think the whole translations stuff could happen in the bot itself.
2007-07-15T13:49:32 <grzywacz> Or even: should.
2007-07-15T13:49:42 <grzywacz> Making queries to wiki for each command/response/other message would be awful.
2007-07-15T13:50:03 <xorAxAx> well
2007-07-15T13:50:10 <xorAxAx> you could load the whole i18n bundle in one go
2007-07-15T13:50:21 <xorAxAx> so you want to duplicate the gettext infrastructure in the bot?
2007-07-15T13:50:22 <grzywacz> It's only a matter of asking for language settings when first communication occurs...
2007-07-15T13:50:44 <xorAxAx> hmm
2007-07-15T13:50:46 <grzywacz> xorAxAx, I planned to do so, yes.
2007-07-15T13:50:58 <grzywacz> But I'm open to other proposals. :)
2007-07-15T13:51:21 <xorAxAx> then you need to get the translators to translate 2 files as well
2007-07-15T13:51:28 <grzywacz> Well, yes.
2007-07-15T13:51:31 <xorAxAx> and duplicate everything in the master wiki etc. :)
2007-07-15T13:51:38 <grzywacz> Duplicate?
2007-07-15T13:51:46 <xorAxAx> well, the structure, not the contents
2007-07-15T13:51:51 <grzywacz> Yes.
2007-07-15T13:51:51 <xorAxAx> the question is - will anybody use the bot without moin?
2007-07-15T13:51:57 <grzywacz> Maybe?
2007-07-15T13:52:04 <grzywacz> xmlrpc interface used so far is "standard".
2007-07-15T13:52:07 <xorAxAx> is it usable without moin currently? :)
2007-07-15T13:52:37 <grzywacz> Well, getting auth tokens doesn't work without moin, for there's some custom code.
2007-07-15T13:52:45 <grzywacz> But.. well. I think it should work. ;)
2007-07-15T13:52:56 <xorAxAx> there is only 1 other wiki that implements wikirpc
2007-07-15T13:53:04 <grzywacz> %)
2007-07-15T13:53:05 <xorAxAx> the one that wrote that "standard"
2007-07-15T13:53:58 <grzywacz> So you suggest getting all translated strings with a special xmlrpc call, and using them as needed?
2007-07-15T13:56:03 <xorAxAx> yes
2007-07-15T13:56:42 <xorAxAx> if anybody chose to use the bot for something else, he could easily reuse the moin po files
2007-07-15T13:57:04 <xorAxAx> and integrate the necessary po loading etc.
2007-07-15T13:58:49 <johill> hrm
2007-07-15T14:00:41 <johill> ahh
2007-07-15T14:05:17 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2377:656685f94105 1.7/MoinMoin/_tests/test_sourcecode.py: test_sourcecode: only test recently changed stuff, use a single os.stat call's results
2007-07-15T14:09:21 <ThomasWaldmann> (there are some hints in the test about how to proceed with a fresh clone)
2007-07-15T14:12:18 <johill> ugh
2007-07-15T14:12:22 <johill> what crap
2007-07-15T14:12:46 <xorAxAx> hm?
2007-07-15T14:12:53 <johill> formatter.interwikilink
2007-07-15T14:12:58 <johill> re-parsers the already parsed link
2007-07-15T14:13:10 <johill> so wikiutil.split_wiki is invoked 4 times
2007-07-15T14:13:14 <johill> with the wrong arguments twice
2007-07-15T14:13:25 <johill> s/wrong arguments/already parsed name/
2007-07-15T14:14:04 <johill> text_html.py:
2007-07-15T14:14:06 <johill> wikitag, wikiurl, wikitail, wikitag_bad = wikiutil.resolve_wiki(self.request, '%s:"%s"' % (interwiki, pagename))
2007-07-15T14:14:09 <johill> crap
2007-07-15T14:14:49 <xorAxAx> :-)
2007-07-15T14:14:57 <johill> see, no quoting there of pagename
2007-07-15T14:15:11 <johill> so with " in it no matter how it was quoted it had to be broken
2007-07-15T14:15:15 <johill> crap crap crap
2007-07-15T14:15:16 <johill> well
2007-07-15T14:15:26 <johill> I won't fix that quote-unquote-quote-unquote again mess
2007-07-15T14:15:30 <johill> I'll just make it work now
2007-07-15T14:15:35 <xorAxAx> try to write some tests
2007-07-15T14:15:53 <xorAxAx> maybe you can pass the result in some object
2007-07-15T14:15:58 <xorAxAx> to avoid reparsing
2007-07-15T14:16:07 <xorAxAx> some IWLink class
2007-07-15T14:16:08 <johill> I have a test that fails because of it not requoting
2007-07-15T14:16:23 <johill> but it's not really possible to test whether it quotes/unquotes again internally :)
2007-07-15T14:16:26 <xorAxAx> cool
2007-07-15T14:18:12 <johill> so as a fix for now I'll just have it quote it again properly
2007-07-15T14:19:39 <xorAxAx> are you using "" quoting now?
2007-07-15T14:19:43 <johill> yes
2007-07-15T14:20:05 <xorAxAx> heh, thats rather easy :)
2007-07-15T14:20:10 <johill> not
2007-07-15T14:20:13 <xorAxAx> why?
2007-07-15T14:20:21 <johill> I need to change the RE, get all the code fixed up, and add tests
2007-07-15T14:20:28 <xorAxAx> right
2007-07-15T14:20:31 <xorAxAx> i mean the quoting part
2007-07-15T14:20:37 <johill> oh yes
2007-07-15T14:20:40 <johill> but unquoting is harder
2007-07-15T14:20:54 <johill> not by much but still
2007-07-15T14:20:57 <xorAxAx> there is some general unquoter in MoinMoin.packages
2007-07-15T14:21:14 <xorAxAx> unpackLine, maybe you can use it somehow
2007-07-15T14:21:21 <johill> nah it hardcodes \\
2007-07-15T14:21:33 <johill> and can't deal with double-quoting
2007-07-15T14:21:38 <johill> it's not hard anyway
2007-07-15T14:21:40 <johill> I already have it working
2007-07-15T14:21:47 <johill> just trying to write a combined test
2007-07-15T14:22:50 * ThomasWaldmann just has seen hg's uppercase quoting :)
2007-07-15T14:22:58 <johill> uppercase? huh?
2007-07-15T14:23:07 <ThomasWaldmann> for win32
2007-07-15T14:23:12 <johill> oh
2007-07-15T14:23:43 <johill> http://johannes.sipsolutions.net/patches/moin/quoting-fixes.patch
2007-07-15T14:24:10 <xorAxAx> johill: it doesnt hardcode that :)
2007-07-15T14:24:34 <johill> xorAxAx: sure
2007-07-15T14:24:37 <johill> if escaped and char in ('\\', separator):
2007-07-15T14:24:41 <johill> escaped = (char == '\\')
2007-07-15T14:24:47 <xorAxAx> umm, ok
2007-07-15T14:24:49 <xorAxAx> :-)
2007-07-15T14:24:51 <johill> I don't want \ escaping
2007-07-15T14:25:01 <johill> anyway look at the patch
2007-07-15T14:25:32 <johill> passes all tests now
2007-07-15T14:25:53 <johill> obviously I had to change a few to make sense without ' quoting
2007-07-15T14:26:06 <johill> I also removed ' url quoting while at it
2007-07-15T14:27:41 <johill> refresh, I added more joint test
2007-07-15T14:34:40 <grzywacz> oops, looks like I'm going to have an unexpected family visit soon -_-;
2007-07-15T14:35:01 <ThomasWaldmann> johill: (u"Sarah O'Connor", u"\"Sarah O'Connor\""), # contains ', quote with "
2007-07-15T14:35:16 <ThomasWaldmann> if you ever quote with ", you can remove that comment
2007-07-15T14:36:32 <ThomasWaldmann> + for joint in ('', 'lala " lala ', 'lala "" lala '):
2007-07-15T14:36:42 <ThomasWaldmann> was this directly from the radio? :D
2007-07-15T14:37:12 <xorAxAx> it rather reminds of a some group sitting together with a joint :)
2007-07-15T14:37:53 <grzywacz> ;>
2007-07-15T14:39:53 <ThomasWaldmann> johill: ok, as we have no other code for that problem currently anyway, we'll take yours :)
2007-07-15T14:40:13 <ThomasWaldmann> please also backport it to 1.6 after having tested it enough
2007-07-15T14:44:14 <dreimark> bbl some hours,
2007-07-15T14:45:55 <johill> right
2007-07-15T14:46:44 <johill> I'll give it some more checks and then commit to both
2007-07-15T14:47:15 <xorAxAx> presumely with a lot of fun because of thomas love for running whitespace cleaners :)
2007-07-15T14:48:15 <grzywacz> ;-)
2007-07-15T15:31:50 <ThomasWaldmann> http://www.itwire.com.au/content/view/13517/53/
2007-07-15T15:40:55 <xorAxAx> neagulm: it looks like you didnt finish moving your code into the event file (This message has been postponed.)
2007-07-15T15:41:24 <CIA-27> moin: Marian Neagul marian@info.uvt.ro work * 2357:0fe910099e2c 1.7-classify-mneagul/MoinMoin/classifier/SpamBayes.py: Fix locking problem
2007-07-15T15:41:28 <CIA-27> moin: Marian Neagul marian@info.uvt.ro work * 2358:a8b4c0b3279d 1.7-classify-mneagul/ (MoinMoin/_tests/test_classifier.py wikiconfig.py): Disable SB-XMLRPC benchmark
2007-07-15T15:41:29 <CIA-27> moin: Marian Neagul marian@info.uvt.ro work * 2359:c8dda853e601 1.7-classify-mneagul/ (5 files in 3 dirs): Display some classifier information in SystemInfo macro.
2007-07-15T15:41:35 <CIA-27> moin: Marian Neagul marian@info.uvt.ro work * 2360:0a304ec8b45c 1.7-classify-mneagul/MoinMoin/classifier/SpamBayesRPC.py: SpamBayes RPC Fixes
2007-07-15T15:41:41 <CIA-27> moin: Marian Neagul marian@info.uvt.ro work * 2361:c62c8fbd2dcb 1.7-classify-mneagul/MoinMoin/classifier/SpamBayes.py: Check for correct tokenizer. Use the configured treshold for spam reporting.
2007-07-15T15:41:54 <CIA-27> moin: Marian Neagul marian@info.uvt.ro work * 2362:87c3954fdd94 1.7-classify-mneagul/MoinMoin/theme/__init__.py: Fix invalid classifier check.
2007-07-15T15:42:02 <CIA-27> moin: Marian Neagul marian@info.uvt.ro work * 2363:7cd4b2352a33 1.7-classify-mneagul/MoinMoin/classifier/SecurityPolicy.py: Fix classifier presence checking
2007-07-15T15:42:10 <CIA-27> moin: Marian Neagul marian@info.uvt.ro work * 2364:a083d1196b0a 1.7-classify-mneagul/MoinMoin/ (7 files in 4 dirs): Replace text_classifier with spam_classifier
2007-07-15T15:42:18 <CIA-27> moin: Marian Neagul marian@info.uvt.ro work * 2365:b54b2e62108c 1.7-classify-mneagul/.hgtags: Added tag Google Summer of Code Start. Classification project for changeset 8f130438d8e8
2007-07-15T15:42:26 <CIA-27> moin: Marian Neagul marian@info.uvt.ro work * 2366:c9970418e713 1.7-classify-mneagul/MoinMoin/theme/__init__.py: Replace text_classifier
2007-07-15T15:42:28 <xorAxAx> hi neagulm
2007-07-15T15:42:32 <xorAxAx> neagulm: did you see my message?
2007-07-15T15:42:34 <CIA-27> moin: Marian Neagul marian@info.uvt.ro work * 2367:45e373088ec2 1.7-classify-mneagul/MoinMoin/classifier/SpamBayes.py: Small cleanup
2007-07-15T15:42:37 <neagulm> hello xorAxAx
2007-07-15T15:42:40 <neagulm> xorAxAx, yes now
2007-07-15T15:42:42 <CIA-27> moin: Marian Neagul marian@info.uvt.ro work * 2368:88eed66799bc 1.7-classify-mneagul/MoinMoin/macro/SystemInfo.py: Fix classifier presence checking
2007-07-15T15:42:50 <CIA-27> moin: Marian Neagul marian@info.uvt.ro work * 2369:a038e3acbe83 1.7-classify-mneagul/docs/CLASSIFIER: Initial documentation
2007-07-15T15:42:51 <neagulm> xorAxAx, you mean the rename page code ?
2007-07-15T15:43:00 <xorAxAx> i mean that your event handler is empty
2007-07-15T15:43:44 <neagulm> xorAxAx, I need to do some checking when a page get's renamed but i didn't find any event for this
2007-07-15T15:43:55 <xorAxAx> http://hg.alexanderweb.de/moin-1.7-classify-mneagul/file/25e327b42998/MoinMoin/events/classifiernotify.py
2007-07-15T15:44:09 <xorAxAx> there are such events nowadays
2007-07-15T15:44:32 <xorAxAx> the goal should be that your branch doesnt have any modifications in the code paths related to page editing
2007-07-15T15:44:38 <xorAxAx> except for the event handler
2007-07-15T15:46:24 <neagulm> xorAxAx, I have removed all changes to Page, PageEditor, and the RenamePage action...
2007-07-15T15:46:33 <ThomasWaldmann> neagulm: + msg = _("Sorry, your page can't be saved because no clasifier is enabled")
2007-07-15T15:46:40 <ThomasWaldmann> +s
2007-07-15T15:47:32 <xorAxAx> neagulm: then how does the classification work?
2007-07-15T15:47:36 <xorAxAx> neagulm: the event handler is empty ...
2007-07-15T15:47:43 <ThomasWaldmann> neagulm: + if score > request.cfg.classify_spamtreshold:
2007-07-15T15:47:51 <ThomasWaldmann> +h
2007-07-15T15:49:01 <neagulm> ThomasWaldmann, +s, +h ? sorry but i think I don't understand
2007-07-15T15:49:01 <ThomasWaldmann> neagulm: + if category == self.spam_page:
2007-07-15T15:49:01 <ThomasWaldmann> + is_spam = True else:
2007-07-15T15:49:01 <ThomasWaldmann> - is_spam=False
2007-07-15T15:49:12 <ThomasWaldmann> typos, those letters are missing
2007-07-15T15:49:58 <ThomasWaldmann> and you can optimize that if/else to read is_spam = category == self.spam_page
2007-07-15T15:51:26 <neagulm> xorAxAx, The classifier is query'ed only by the theme to display the classification or by the SecurityPolicy. The larning part is handled by a separate action: togglespamstatus.py
2007-07-15T15:52:31 <xorAxAx> neagulm: ??? so spam isnt detected automatically?
2007-07-15T15:54:24 <neagulm> xorAxAx, it is, via the theme or security policy. I can move the whole functionality to any part of MoinMoin but in my opinion this is the right approach.
2007-07-15T15:54:42 <neagulm> xorAxAx, You think that Page should be aware of the spam status?
2007-07-15T15:54:48 <xorAxAx> the security policy?
2007-07-15T15:54:48 <xorAxAx> argh
2007-07-15T15:54:54 <xorAxAx> please put that into the event handler
2007-07-15T15:55:02 <xorAxAx> hmm, antispam should go there as well
2007-07-15T15:55:02 <xorAxAx> grzywacz:
2007-07-15T15:55:08 <xorAxAx> neagulm: no!
2007-07-15T15:55:29 <neagulm> xorAxAx, I thought so :)
2007-07-15T15:56:09 <xorAxAx> so, have you had a look at the goal of multiple categories and which classifier to use/modify?
2007-07-15T15:56:54 <ThomasWaldmann> xorAxAx: antispam needs to intercept before the page was saved
2007-07-15T15:58:08 <neagulm> xorAxAx, yep. For the particular case of language classification we can use DivMod's Reverend (I will maintain it or even rewrite it). For generic text classification we need a interface to something more powerfull, Orange might be a good choice (http://www.ailab.si/orange/)
2007-07-15T15:59:27 <neagulm> xorAxAx, The goals might be: automatic tagging/grouping, Hierarchical grouping, and I'm thinking about other uses
2007-07-15T15:59:37 <neagulm> brb
2007-07-15T16:00:01 <xorAxAx> ThomasWaldmann: hmm, indeed
2007-07-15T16:00:16 <xorAxAx> so we need a pagepresaveent
2007-07-15T16:00:20 <xorAxAx> event
2007-07-15T16:00:26 <xorAxAx> grzywacz:
2007-07-15T16:00:44 <xorAxAx> neagulm: yes, when will you start? :)
2007-07-15T16:08:55 <xorAxAx> neagulm: why are you using the securitypolicy? are you rejecting suspected spam edits?
2007-07-15T16:09:09 <grzywacz> xorAxAx
2007-07-15T16:09:10 <xorAxAx> neagulm: how should the wiki learn in such a world where it only sees ham?
2007-07-15T16:09:22 <xorAxAx> grzywacz: can you add that event? :)
2007-07-15T16:10:09 * grzywacz reads backlog
2007-07-15T16:10:59 <grzywacz> What event? :s
2007-07-15T16:11:39 <xorAxAx> pagepresaveent
2007-07-15T16:11:51 <xorAxAx> that can be used to abort page saves
2007-07-15T16:11:57 <grzywacz> Hm.
2007-07-15T16:12:39 <grzywacz> Give me a few minutes. ;)
2007-07-15T16:25:29 <neagulm> xorAxAx, related to the securitypolicy: it should be enabled only after serious training not before.
2007-07-15T16:25:54 <neagulm> xorAxAx, I will start working for n-way classification after i get green light from dreimark
2007-07-15T16:26:33 <neagulm> xorAxAx, I need to leave for a few minutes...
2007-07-15T16:27:05 <xorAxAx> well ...
2007-07-15T16:27:19 <johill> I don't see anything further wrong with the quoting now
2007-07-15T16:27:19 <xorAxAx> how should you be able to train it if there is no spam in the wiki
2007-07-15T16:27:22 <johill> will update docs
2007-07-15T16:27:27 <xorAxAx> thats broken by design IMHO
2007-07-15T16:27:40 <xorAxAx> it should not depend on remote corpii
2007-07-15T16:28:34 <grzywacz> xorAxAx, argh, family, afk ;S
2007-07-15T16:29:37 <neagulm> xorAxAx, the user will train when it sees spam... I'm not really an advocate of using a SecurityPolicy for statistical classifiers
2007-07-15T16:30:13 <xorAxAx> neagulm: i mean the other direction
2007-07-15T16:30:20 <xorAxAx> neagulm: imagine the system rejects some edit
2007-07-15T16:30:27 <xorAxAx> you cannot learn it as ham
2007-07-15T16:32:10 <neagulm> xorAxAx, that's way I suggested to not use SecurityPolicy. Of course this can be overcome by disabling the policy enforcing for admin users, or for users that are allowed to train the classifier
2007-07-15T16:32:52 <xorAxAx> neagulm: the admin users? the admin users dont want to edit and they cant see the edit at all
2007-07-15T16:33:25 <xorAxAx> thats orthogonal to the implementation issue, i am talking about the design and usability
2007-07-15T16:34:32 <xorAxAx> it might e.g. make sense to reject suspected spam only for users that are known for less than n days
2007-07-15T16:34:54 <xorAxAx> everything else has to into the system and has to have the chance to be relearned as ham
2007-07-15T16:35:57 <CIA-27> moin: Johannes Berg <johannes AT sipsolutions DOT net> * 2378:a7f4b02f5fdb 1.7/MoinMoin/ (5 files in 5 dirs):
2007-07-15T16:35:57 <CIA-27> moin: fix the quoting mess
2007-07-15T16:35:57 <CIA-27> moin: * double double-quotes (") to get a double quote inside quotes
2007-07-15T16:35:57 <CIA-27> moin: * quote things correctly when needed (not just "%s")
2007-07-15T16:35:59 <CIA-27> moin: * add and fix unit tests for this quoting
2007-07-15T16:36:01 <CIA-27> moin: * formatter must quote the name before it can pass it to resolve_wiki
2007-07-15T16:38:02 <johill> hm
2007-07-15T16:38:07 <johill> are there no parser tests in 1.6?
2007-07-15T16:38:19 <xorAxAx> they are in broken i guess
2007-07-15T16:38:24 <johill> oh
2007-07-15T16:38:42 <xorAxAx> the old test framework doesnt support skipping so thomas realised enterprise grade module-wise skipping hehre
2007-07-15T16:38:49 <johill> heh
2007-07-15T16:44:50 <CIA-27> moin: Johannes Berg <johannes AT sipsolutions DOT net> * 2115:77cce7920124 1.6/MoinMoin/ (formatter/text_html.py parser/text_moin_wiki.py wikiutil.py): backport quoting fixes, this makes HelpOnConfiguration et al. show correctly again
2007-07-15T16:45:13 <johill> also updated the documentation
2007-07-15T16:46:53 <johill> on moinmaster that is, obviously it doesn't properly show it yet
2007-07-15T16:47:27 <xorAxAx> ThomasWaldmann: it makes sense to move moinmaster to 1.6 in these days
2007-07-15T16:48:28 <ThomasWaldmann> xorAxAx: you mean if the mig script was finished :-P
2007-07-15T16:48:40 <johill> no need on moinmaster ;)
2007-07-15T16:48:43 <xorAxAx> yep
2007-07-15T16:48:46 <johill> I ran mine on all of underlay and it made no changes
2007-07-15T16:49:27 <ThomasWaldmann> i tried it quite a while ago and there definitely were some problems
2007-07-15T16:50:58 <xorAxAx> so it should be tried again and problems should be documented
2007-07-15T16:51:41 * ThomasWaldmann does tax stuff now ;(
2007-07-15T16:54:40 <johill> just updated http://johannes.sipsolutions.net/files/linkconv.py.txt, I had forgotten _remark_repl
2007-07-15T16:54:56 <johill> not that it should have been present in 1.5 wikis afaict
2007-07-15T16:55:51 <xorAxAx> johill: you didnt checkin the conversion script yet, did you?
2007-07-15T16:56:16 <johill> no
2007-07-15T16:56:22 <johill> hm
2007-07-15T16:56:32 <johill> is that remark markup documented anywhere?
2007-07-15T16:57:08 <johill> it seems fairly broken
2007-07-15T16:57:18 <xorAxAx> annoate blames whom? :)
2007-07-15T16:57:40 <johill> changeset: 1786:2c9853dc11f8
2007-07-15T16:57:41 <johill> user: Thomas Waldmann <tw AT waldmann-edv DOT de>
2007-07-15T16:57:41 <johill> date: Thu Feb 22 12:04:46 2007 +0100
2007-07-15T16:57:41 <johill> summary: new wiki markup for /* inline comments */, support toggling visibility of them
2007-07-15T16:58:30 <johill> http://test.wikiwikiweb.de/remark
2007-07-15T16:58:55 <johill> another one of those programmer features in wiki markup
2007-07-15T16:58:56 <johill> crap
2007-07-15T16:59:10 <johill> maybe I should write my own wiki markup and just stop using this
2007-07-15T16:59:13 * xorAxAx doesnt like it either
2007-07-15T16:59:25 <xorAxAx> johill: thesheep started a markup parser
2007-07-15T16:59:37 <johill> like a real one? interesting
2007-07-15T16:59:37 <xorAxAx> thats fairly compatible and much cleaner
2007-07-15T17:01:05 <johill> wtf
2007-07-15T17:01:15 <johill> why does the remark test page behave differently in preview?
2007-07-15T17:01:18 <xorAxAx> neagulm: ARGH
2007-07-15T17:01:19 <xorAxAx> + except:
2007-07-15T17:01:19 <xorAxAx> + pass
2007-07-15T17:01:31 <johill> ah no
2007-07-15T17:02:29 <xorAxAx> neagulm: never ever suppress exceptions
2007-07-15T17:04:08 <johill> urks
2007-07-15T17:04:10 <johill> http://test.wikiwikiweb.de/remark2
2007-07-15T17:04:14 <johill> compare to the source
2007-07-15T17:04:22 <johill> how obvious behaviour ;)
2007-07-15T17:04:48 <johill> text_moin_programmer_markup.py ;)
2007-07-15T17:05:07 <xorAxAx> and we already have a sane comment markup
2007-07-15T17:06:17 <johill> well, whatever. unless we remove it we'll have to adjust the converter to insert `` inside /*
2007-07-15T17:06:49 * xorAxAx would like to see it removed
2007-07-15T17:06:54 * johill too
2007-07-15T17:07:03 <xorAxAx> and replaced by #-comments (the toggle-functionality)
2007-07-15T17:07:20 <johill> + # XXX we don't really enforce the correct sequence /* ... */ here
2007-07-15T17:07:33 <johill> yeah. duh
2007-07-15T17:07:53 <johill> # comments?
2007-07-15T17:07:57 <johill> toggle functionality?
2007-07-15T17:08:18 <johill> ah you want to be able to view # comments without going to raw?
2007-07-15T17:08:30 <xorAxAx> yes, thats the new functionality using the old syntax
2007-07-15T17:08:40 <xorAxAx> it just doesnt support in-p comments
2007-07-15T17:08:43 <johill> yeah, lines should be combined then I guess
2007-07-15T17:08:59 <johill> big deal, just put it on top then
2007-07-15T17:09:02 <johill> or below
2007-07-15T17:09:14 <xorAxAx> hmm, you can even have in-p comments using that, yes
2007-07-15T17:09:17 <johill> besides this can't support it
2007-07-15T17:09:31 <johill> oh sorry I thought of pre
2007-07-15T17:12:08 <johill> ah well
2007-07-15T17:12:27 <johill> no more time for today
2007-07-15T18:26:54 <ThomasWaldmann> 120 f = open('/home/mvirkkil/log.txt', "w")
2007-07-15T18:26:57 <ThomasWaldmann> hmpf
2007-07-15T18:50:06 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1775:9ae9f564380e 1.6-docbook-mvirkill/MoinMoin/ (PageEditor.py PageGraphicalEditor.py): fix backtrace on editor preview, refactor code to use 'request', not 'self.request'
2007-07-15T18:50:09 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1776:70e505db8728 1.6-docbook-mvirkill/MoinMoin/mail/mailimport.py: mailimport: fix backtrace for empty target addr realname, loop in target addrs until some realname was found
2007-07-15T18:50:17 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1777:7c58e8af1a97 1.6-docbook-mvirkill/MoinMoin/mail/mailimport.py: mailimport: fix wrong comparison value
2007-07-15T18:50:23 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1778:35ebda626b4b 1.6-docbook-mvirkill/MoinMoin/Page.py: getPageBasePath: add default value for use_underlay and improve docstring
2007-07-15T18:50:41 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1779:66f2ea899da7 1.6-docbook-mvirkill/MoinMoin/ (action/fckdialog.py request/__init__.py): fckdialog action: don't call emit_http_headers twice, fix wrong attribute name in emit_http_headers
2007-07-15T18:50:52 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1780:c54be3afb830 1.6-docbook-mvirkill/MoinMoin/converter/text_html_text_moin_wiki.py: gui editor converter: indenting fixes, code simplification
2007-07-15T18:51:02 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1781:3c6f59cafbbe 1.6-docbook-mvirkill/MoinMoin/ (Page.py action/fckdialog.py request/__init__.py): merged main
2007-07-15T18:51:08 <CIA-27> moin: Reimar Bauer < rb.proj@googlemail.com > * 1782:790c1e655a5b 1.6-docbook-mvirkill/MoinMoin/macro/ImageLink.py: ImageLink gots missing feature of interwikilink added
2007-07-15T18:51:25 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1783:0803e5da055d 1.6-docbook-mvirkill/MoinMoin/filter/EXIF.py: EXIF.py: optional parameter 'name' for process_file and dump_IFD. If used the reading of EXIF data ends after that name. some PEP8 changes
2007-07-15T18:51:32 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1784:fe19c9ac4b82 1.6-docbook-mvirkill/MoinMoin/ (2 files in 2 dirs): text_gedit: Feature scalable images for gui added
2007-07-15T18:51:42 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1785:66b17318081d 1.6-docbook-mvirkill/docs/CHANGES: updated CHANGES for ImageLink / gui resizing images stuff
2007-07-15T18:51:54 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1786:a31498666626 1.6-docbook-mvirkill/MoinMoin/formatter/text_gedit.py: text_gedit: minor refactoring
2007-07-15T18:52:04 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1787:dd2bdae20379 1.6-docbook-mvirkill/MoinMoin/converter/text_html_text_moin_wiki.py: text_html_text_moin_wiki: minor refactoring
2007-07-15T18:52:16 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1788:ac830832ca26 1.6-docbook-mvirkill/MoinMoin/converter/text_html_text_moin_wiki.py: ImageLink conversion: optimized code
2007-07-15T18:52:22 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1789:2f7793521ccd 1.6-docbook-mvirkill/MoinMoin/userform.py: userform: SendMyPassword hint added
2007-07-15T18:52:32 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1790:2077ed6974d7 1.6-docbook-mvirkill/MoinMoin/macro/Login.py: Login macro: gets feature sendmail form
2007-07-15T18:52:44 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1791:1e699d639b78 1.6-docbook-mvirkill/MoinMoin/macro/Login.py: Login macro: sentence for wrong used parameter added
2007-07-15T18:52:54 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1792:c0dd392bed62 1.6-docbook-mvirkill/docs/CHANGES: updated CHANGES for recovering a lost password
2007-07-15T18:53:06 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1793:1f18d08ec3ba 1.6-docbook-mvirkill/ (5 files in 5 dirs): better display of renamed pages on info/RC, show diff-link for deleted pages on RC
2007-07-15T18:53:12 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1794:8f944671df08 1.6-docbook-mvirkill/ (MoinMoin/macro/Login.py MoinMoin/userform.py docs/CHANGES): merged main
2007-07-15T18:53:24 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1795:8d2035dfb153 1.6-docbook-mvirkill/MoinMoin/ (macro/RecentChanges.py theme/__init__.py): use separate conflict icon (TODO: add a png for it), minor cleanups
2007-07-15T18:53:35 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1796:9a3290aeff36 1.6-docbook-mvirkill/MoinMoin/ (Page.py PageEditor.py): fixed cached inconsistency caused by DeletePage
2007-07-15T18:53:47 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1797:10c60d1131dd 1.6-docbook-mvirkill/MoinMoin/packages.py: packages:AddAttachment, DelAttachment, RenamePage added
2007-07-15T18:53:57 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1798:1d0663c1c44f 1.6-docbook-mvirkill/MoinMoin/action/PackagePages.py: PackagePages:added storing of attachments, form pagename text field gots current page name
2007-07-15T18:54:11 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1799:3244350dc5ad 1.6-docbook-mvirkill/MoinMoin/parser/text_moin_wiki.py: just using WikiName: as abbreviation for WikiName: gives too many false positives
2007-07-15T18:54:24 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1800:8a0c277a20f7 1.6-docbook-mvirkill/MoinMoin/ (action/PackagePages.py packages.py): merged main
2007-07-15T18:54:31 <CIA-27> (186 lines omitted)
2007-07-15T18:54:40 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1977:3d2dbe3e8a4c 1.6-docbook-mvirkill/MoinMoin/theme/__init__.py: shouldShowPageinfo: CopyPage added
2007-07-15T18:54:52 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1978:b5e94fec8669 1.6-docbook-mvirkill/MoinMoin/ (caching.py util/filesys.py): caching: mkstemp uses 0600 to create files, we need chmod to change them to our preferred file mode
2007-07-15T18:55:05 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1979:55617b3b1ac9 1.6-docbook-mvirkill/MoinMoin/ (PageEditor.py action/edit.py theme/__init__.py): merged main
2007-07-15T18:55:17 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1980:89fffad859b5 1.6-docbook-mvirkill/MoinMoin/action/SpellCheck.py: SpellCheck: check if user may write to request.cfg.page_local_spelling_words
2007-07-15T18:55:27 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1981:86819d96cb01 1.6-docbook-mvirkill/MoinMoin/ (caching.py util/filesys.py): SpellCheck:check write access to request.cfg.page_local_spelling_words
2007-07-15T18:55:41 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1982:6628902acfe7 1.6-docbook-mvirkill/MoinMoin/macro/SystemInfo.py: SystemInfo: minor refactoring
2007-07-15T18:55:47 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1983:0a6764ab0c16 1.6-docbook-mvirkill/MoinMoin/action/SpellCheck.py: merged main
2007-07-15T18:56:04 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1984:82667e81b928 1.6-docbook-mvirkill/MoinMoin/action/AttachFile.py: Fix crash in AttachFile when called by xmlrpc and request.page is None
2007-07-15T18:56:12 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1985:60a6dd88624a 1.6-docbook-mvirkill/MoinMoin/action/fullsearch.py: escape search terms before output
2007-07-15T18:56:20 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1986:4c424f7b5d55 1.6-docbook-mvirkill/MoinMoin/Page.py: fix double content div for MissingPage
2007-07-15T18:56:34 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1987:3799c27e664a 1.6-docbook-mvirkill/MoinMoin/Page.py: Page.getPageLinks() - minor refactoring
2007-07-15T18:56:44 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1988:8bd4b581cc60 1.6-docbook-mvirkill/ (9 files in 5 dirs): make converter pluggable, dynamically load correct converter, move loading of parser to send_page_content
2007-07-15T18:56:53 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1989:be140a589f11 1.6-docbook-mvirkill/MoinMoin/ (6 files in 2 dirs): improved some module docstrings
2007-07-15T18:57:03 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1990:5c3dffe2abf1 1.6-docbook-mvirkill/MoinMoin/ (6 files in 4 dirs): new function AttachFile.exists to reduce code duplication
2007-07-15T18:57:17 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1991:0b2ad3099fcf 1.6-docbook-mvirkill/MoinMoin/action/ (AttachFile.py info.py): AttachFile.size to reduce code duplication, fixed docstring
2007-07-15T18:57:27 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1992:901ff624d942 1.6-docbook-mvirkill/MoinMoin/action/subscribe.py: subscribe action: give msg if page subscription failed
2007-07-15T18:57:35 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1993:6769679b26f6 1.6-docbook-mvirkill/MoinMoin/action/quicklink.py: quicklink action: give msg if quicklink toggle failed
2007-07-15T18:57:50 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1994:eb68960b1dfb 1.6-docbook-mvirkill/MoinMoin/ (8 files in 7 dirs): remove unused imports
2007-07-15T18:57:56 <CIA-27> moin: Franz Pletz <fpletz AT franz-pletz DOT org> * 1995:ed85c389031d 1.6-docbook-mvirkill/MoinMoin/search/Xapian.py: xapian.Query needs to be imported
2007-07-15T18:58:08 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1996:d41e6ef31c0e 1.6-docbook-mvirkill/MoinMoin/PageEditor.py: PageEditor.copyPage: write access checked on old page instead of the copied one
2007-07-15T18:58:20 <CIA-27> moin: Reimar Bauer <rb.proj AT googlemail DOT com> * 1997:9a25a7ae3514 1.6-docbook-mvirkill/MoinMoin/ (userform.py wikiutil.py): clean_comment added to some inputs
2007-07-15T18:58:32 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 1998:bb2e053067fb 1.6-docbook-mvirkill/MoinMoin/ (247 files in 33 dirs): fixing copyright headers: remove umlauts (encoding troubles), make epydoc compatible, reformat
2007-07-15T18:58:42 <CIA-27> (220 lines omitted)
2007-07-15T19:02:28 * ThomasWaldmann sings "spam, spam, spam"
2007-07-15T19:55:27 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2197:6fa44eb2856f 1.6-docbook-mvirkill/MoinMoin/ (6 files in 4 dirs): PEP8 / whitespace fixes
2007-07-15T20:48:15 <CIA-27> moin: Ville-Pekka Vainio <vpivaini AT cs DOT helsinki DOT fi> * 2462:73fcab6240b8 1.7-maninfo-vpv/MoinMoin/xmlrpc/__init__.py: Fix another xmlrpc/__init__.py self bug
2007-07-15T21:26:17 <starshine> 16
2007-07-15T21:29:16 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2198:03f520b60f60 1.6-docbook-mvirkill/docs/CHANGES.mvirkkil: improved CHANGES.mvirkkil
2007-07-15T21:54:44 <vpv> damn, I thought my last commit would have fixed this, but no, I still occasionally get "global name \'_cleanup_stale_tokens\' is not defined"
2007-07-15T21:56:35 <vpv> it's line 622 in MoinMoin/xmlrpc/__init__.py so I guess this error happens pretty randomly ;)
2007-07-15T22:01:13 <xorAxAx> dreimark: neagulm is waiting for you to ack the implementation of multi-category classification (This message has been postponed.)
2007-07-15T22:01:35 <dreimark> moin
2007-07-15T22:04:20 <dreimark> xorAxAx: ?
2007-07-15T22:05:49 <dreimark> xorAxAx: neagulm, I like to have multi-category classification implemented, but first I want to see the documentation of the spambayes implementation
2007-07-15T22:06:27 <dreimark> there should be one nearly finished before the next is done
2007-07-15T22:07:58 <dreimark> neagulm: and I guess it is easier now because you have implemented spambayes. But where is the for saturday promised description?
2007-07-15T22:11:33 <dreimark> neagulm, ok I found one at docs/CLASSIFIER
2007-07-15T22:16:18 <grzywacz> Oh good, children going to sleep now. -_-;
2007-07-15T22:17:03 <dreimark> hmm?
2007-07-15T22:19:32 <grzywacz> I have a sudden family visit with my very young cousins.
2007-07-15T22:19:37 <grzywacz> Spent half of the day playing with them. ;S
2007-07-15T22:23:28 <dreimark> oh, i was just visiting someone of my family in the hospital as everyday of last week.
2007-07-15T22:27:51 <xorAxAx> dreimark: makes sense
2007-07-15T22:28:01 <xorAxAx> dreimark: he is a bit inactive currently
2007-07-15T22:29:22 <dreimark> xorAxAx: documenting is not finished now http://hg.moinmo.in/moin/1.7-classify-mneagul/file/a038e3acbe83/docs/CLASSIFIER
2007-07-15T22:30:12 <grzywacz> Oook, silent at last. Now, where did I finish...
2007-07-15T22:30:20 <dreimark> I do want his current code merged into 1.7 before he starts with http://hg.moinmo.in/moin/1.7-classify-mneagul/file/a038e3acbe83/docs/CLASSIFIER
2007-07-15T22:30:41 <xorAxAx> dreimark: well, i have some objections against merging
2007-07-15T22:30:44 <xorAxAx> i described them today
2007-07-15T22:30:59 <dreimark> ok I read them in the logs tommorrow
2007-07-15T22:31:03 <xorAxAx> :-)
2007-07-15T22:31:11 <xorAxAx> agile development :)
2007-07-15T22:31:17 <grzywacz> ?:)
2007-07-15T22:31:39 <xorAxAx> grzywacz: thomas uses the old log exporter that doesnt instantly publish the irc logs
2007-07-15T22:31:58 <xorAxAx> so dreimark has to wait for the next day to see the log
2007-07-15T22:32:50 <grzywacz> Oh. Ok.
2007-07-15T22:32:51 <dreimark> grzywacz: I have requested for that because it is much easier if you are in the office and like to go home wihout missing parts of the discussion, or if you are offline for some time
2007-07-15T22:33:03 <grzywacz> Well, logging is good.
2007-07-15T22:34:14 <grzywacz> xorAxAx, http://rafb.net/p/mA6QGs94.html
2007-07-15T22:34:15 <grzywacz> Like that?
2007-07-15T22:37:34 <xorAxAx> grzywacz: maybe its sensible to factor out the isinstance call into a should_abort function ... and the event needs the new page text and some new revision number (have a look at the antispam handler what it exactly needs)
2007-07-15T22:38:04 <grzywacz> xorAxAx, I guess I have to clone neaglum's repo first?
2007-07-15T22:38:19 <grzywacz> Well, web view.
2007-07-15T22:38:21 <xorAxAx> no
2007-07-15T22:38:27 <xorAxAx> antispam is in moin since 1.2
2007-07-15T22:38:46 <xorAxAx> moinmoin.security.antispam or something like that
2007-07-15T22:38:48 <grzywacz> Oh, so it's not new code.
2007-07-15T22:38:53 <grzywacz> Ok, checking.
2007-07-15T22:42:40 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2199:abe48ca50f8f 1.6-docbook-mvirkill/docs/CHANGES.mvirkkil: improved CHANGES.mvirkkil
2007-07-15T22:45:47 <grzywacz> xorAxAx, revision can be read from PageEditor object directly, so the "rev" parameter in SecurityPolicy's save() is rather redundant.
2007-07-15T22:46:48 <xorAxAx> hmm, yeah, the question is if the event class should be coupled to the dirty pageeditor class
2007-07-15T22:47:05 <xorAxAx> how about passing the revno explicitly ...
2007-07-15T22:47:10 <xorAxAx> hmm
2007-07-15T22:47:20 <xorAxAx> how aobut other events, do you pass names or objects?
2007-07-15T22:47:25 <grzywacz> objects
2007-07-15T22:47:32 <xorAxAx> hrm
2007-07-15T22:47:39 <grzywacz> It certainly reduces overhead.
2007-07-15T22:47:45 <xorAxAx> sure
2007-07-15T22:47:56 <xorAxAx> but this is in the refactoring state, you know :)
2007-07-15T22:48:00 <grzywacz> If you need a page object you already have, it's easier to pass it along rather than its name only to instantiate the page again.
2007-07-15T22:48:10 <grzywacz> Well, if it is, then it can be changed. 8)
2007-07-15T22:48:17 <xorAxAx> well, ok, pass the object, we can fix it later-on
2007-07-15T22:48:20 <grzywacz> ack
2007-07-15T22:48:20 <xorAxAx> but write a test please
2007-07-15T22:48:33 <grzywacz> To test if there's an object passed?
2007-07-15T22:48:37 <xorAxAx> this is easily testable - just register a handler and call the pageditor
2007-07-15T22:48:47 <xorAxAx> to see if you can abort edits using this new system
2007-07-15T22:48:53 <grzywacz> Hrmpf.
2007-07-15T22:49:01 <xorAxAx> 5 lines of code :)
2007-07-15T22:49:17 * grzywacz wonders how to register a handler dynamically %)
2007-07-15T22:49:48 <grzywacz> Ah, config.
2007-07-15T22:50:24 <grzywacz> Ok, I'll write a test.
2007-07-15T22:51:13 <xorAxAx> its so useful to try to get this TDD paradigm into the blood :)
2007-07-15T22:51:26 <xorAxAx> esp. in such dirty areas (pageeditor legacy code etc.)
2007-07-15T22:51:54 <CIA-27> moin: Ville-Pekka Vainio <vpivaini AT cs DOT helsinki DOT fi> * 2463:2bf595cec211 1.7-maninfo-vpv/MoinMoin/script/xmlrpc/manimport.py: Improved doclifter error handling, group multicalls to 10 man pages per call, some other small fixes. Prints a lot of debug stuff now, needs serious cleanups but works with Fedora's updates-testing repository
2007-07-15T22:51:59 <grzywacz> TDD?
2007-07-15T22:52:12 <grzywacz> ah
2007-07-15T22:52:14 <grzywacz> x)
2007-07-15T22:52:48 <xorAxAx> test driven development
2007-07-15T22:53:58 <grzywacz> I know. :P
2007-07-15T22:57:07 <grzywacz> xorAxAx, should a message be shown if page has not been saved due to Abort returned by an event handler?
2007-07-15T22:57:34 <xorAxAx> grzywacz: yes, the event handler should be able to specify it
2007-07-15T22:57:46 <xorAxAx> i.e. its an attrib of the abort instance
2007-07-15T22:57:56 <grzywacz> Right.
2007-07-15T23:07:01 <dreimark> I have seen we have time till monday 19:00 UTC for midterm survey
2007-07-15T23:09:23 <xorAxAx> yes
2007-07-15T23:10:15 * grzywacz is easily influenced
2007-07-15T23:10:49 <grzywacz> My friend showed me a piece of java code yestarday and I've already managed to write "instanceof" instead of "isinstance" %)
2007-07-15T23:13:37 <vpv> it's going to take probably about an hour to import all of Fedora's updates-testing man page to the wiki, and that's the smallest repo there is...
2007-07-15T23:14:04 <grzywacz> But it's a one-time action, no?
2007-07-15T23:14:45 <vpv> hopefully, but it seems I'll really need to implement a good updates-handling thingie too
2007-07-15T23:15:30 <vpv> so that people could do just the smallest possible amount of imports
2007-07-15T23:20:48 <dreimark> gn
2007-07-15T23:22:19 <grzywacz> night dreimark
2007-07-15T23:33:46 <grzywacz> xorAxAx, should a few aborts be returned, do we need all error messages, or is the first one sufficient? ;-)
2007-07-15T23:34:19 <grzywacz> Concatenating them seems ugly. ;S
2007-07-15T23:34:56 <xorAxAx> concat them :)
2007-07-15T23:35:23 <grzywacz> Normal concat or with some fugly <br>s?
2007-07-15T23:35:32 <xorAxAx> normal
2007-07-15T23:35:37 <xorAxAx> no
2007-07-15T23:35:43 <xorAxAx> with fugly <br>s!
2007-07-15T23:35:46 <xorAxAx> <evil laughter>
2007-07-15T23:35:53 <grzywacz> _-_;
2007-07-15T23:35:57 <xorAxAx> :-)
2007-07-15T23:36:04 <grzywacz> zenhase would get angry, you know.
2007-07-15T23:36:04 <grzywacz> :P
2007-07-15T23:36:26 <xorAxAx> why?
2007-07-15T23:36:46 <grzywacz> Spitting html all around doesn't look like a good thing to do in terms of templating.
2007-07-15T23:36:47 <grzywacz> ;]
2007-07-15T23:37:11 <xorAxAx> ah, let me think
2007-07-15T23:37:14 <xorAxAx> welll
2007-07-15T23:37:30 <xorAxAx> we dont care here because this is content
2007-07-15T23:37:39 <xorAxAx> but indeed
2007-07-15T23:37:50 <xorAxAx> we would need to make msg a list everywhere
2007-07-15T23:37:54 <xorAxAx> and pass it into the context
2007-07-15T23:38:14 <xorAxAx> so concat it and make some #XXX in the list (do the concat in the PE)
2007-07-15T23:38:19 <xorAxAx> s/list/file/
2007-07-15T23:38:24 <grzywacz> mhm, ok
2007-07-15T23:39:57 <grzywacz> Ok, I hope I got it right. Pushing soon.
2007-07-15T23:46:26 <CIA-27> moin: Karol 'grzywacz' Nowak <grzywacz@sul.uni.lodz.pl> * 2389:cf85b6916d17 1.7-jabber-knowak/MoinMoin/ (PageEditor.py events/__init__.py): Add a PagePreSave event; abort page save if handlers return Abort.
2007-07-15T23:46:28 <CIA-27> moin: Karol 'grzywacz' Nowak <grzywacz@sul.uni.lodz.pl> * 2390:a10ae31a6133 1.7-jabber-knowak/MoinMoin/_tests/test_PageEditor.py: Test if page save really fails when PreSave event handlers return Abort.
2007-07-15T23:46:30 <CIA-27> moin: Karol 'grzywacz' Nowak <grzywacz@sul.uni.lodz.pl> * 2391:f9d087b59fd4 1.7-jabber-knowak/MoinMoin/ (PageEditor.py _tests/test_PageEditor.py events/__init__.py): Merge devel.
2007-07-15T23:46:36 <CIA-27> moin: Karol 'grzywacz' Nowak <grzywacz@sul.uni.lodz.pl> * 2392:245181ec9fe2 1.7-jabber-knowak/MoinMoin/ (PageEditor.py _tests/test_PageEditor.py events/__init__.py): Whitespace fixes.
2007-07-15T23:50:36 <vpv> the importing took 78 minutes, but at least it works...
2007-07-15T23:51:16 <xorAxAx> is it cpu bound?
2007-07-15T23:53:18 <xorAxAx> vpv:
2007-07-15T23:54:51 <vpv> I don't think it is, the cpu is even idle at some points
2007-07-15T23:55:21 <vpv> frankly I have no idea where the time is spent, but it seems to be something in the xmlrpc stuff
2007-07-15T23:55:32 <grzywacz> sourge protect?
2007-07-15T23:55:40 <grzywacz> *surge
2007-07-15T23:56:40 <xorAxAx> vpv: how many man pages?
2007-07-15T23:56:55 <xorAxAx> yes, maybe you are hitting the surge protection
2007-07-15T23:57:26 <vpv> there were 78 packages with man pages on them, so maybe 150 pages or something
2007-07-15T23:57:39 <grzywacz> Hm, not so many...
2007-07-15T23:58:31 <vpv> should the surge protection give an error message or raise an exception?
2007-07-15T23:59:38 <xorAxAx> no idea
2007-07-15T23:59:48 <xorAxAx> well, it makes sense if you find out where the time is spent
2007-07-15T23:59:54 <xorAxAx> feel free to use a profiler MoinMoin: MoinMoinChat/Logs/moin-dev/2007-07-15 (last edited 2007-10-29 19:10:02 by localhost)
