MoinMoin: MoinMoinChat/Logs/moin-dev/2007-08-18

2007-08-18T00:06:44  <CIA-14> moin: Marian Neagul marian@info.uvt.ro main * 2883:8674156336b8 1.7-classify-mneagul/MoinMoin/ (PageEditor.py events/__init__.py): Add EventResult for changing page save behaviour
2007-08-18T00:07:01  <neagulm> xorAxAx, please see 8674156336b8
2007-08-18T00:07:30  <neagulm> xorAxAx, is this the way you suggested ?
2007-08-18T00:08:27  <neagulm> xorAxAx, this would be enough for removing my edit.py code
2007-08-18T00:10:54  <xorAxAx> hmm, thats not composable
2007-08-18T00:11:12  <xorAxAx> if you have  2 changeinfos, both overwrite each other
2007-08-18T00:11:36  <xorAxAx> or if there is an extra kwarg even
2007-08-18T00:11:55  <grzywacz> xorAxAx: since "result" forms don't seem to be supported (args!) I'll use a normal one to report search results. oh well.
2007-08-18T00:12:07  <grzywacz> s/args/argh/
2007-08-18T00:13:22  <xorAxAx> as long as its usable, its fine :)
2007-08-18T00:14:09  <neagulm> for the notify event it is easy to implement something composable but for "extra" I don't see how
2007-08-18T00:14:31  <neagulm> s/event/property
2007-08-18T00:15:52  <grzywacz> xorAxAx: will be usable in a moment
2007-08-18T00:17:24  <xorAxAx> neagulm: how about concatenating?
2007-08-18T00:17:39  <xorAxAx> or at least raising an exception
2007-08-18T00:17:48  <xorAxAx> simply overwriting is bad
2007-08-18T00:18:42  <dreimark> hi neagulm
2007-08-18T00:19:45  <neagulm> moin dreimark
2007-08-18T00:20:20  <dreimark> neagulm: seems you don't have a merge to a fix of johill related to the Avtion macro
2007-08-18T00:21:02  <dreimark> If you try to create a homepage by clicking on your username in the testwiki you see it missing
2007-08-18T00:22:25  <neagulm> dreimark, I see
2007-08-18T00:25:54  <johill> huh
2007-08-18T00:25:54  <johill>   File "/usr/local/src/moin/MoinMoin/macro/__init__.py", line 385, in macro_Mail
2007-08-18T00:25:58  <johill> To
2007-08-18T00:26:00  <johill>     result += self.formatter.code(1) + \
2007-08-18T00:27:17  <xorAxAx> hmm?
2007-08-18T00:27:28  <johill> result referenced before assignment
2007-08-18T00:28:34  <johill> obviously
2007-08-18T00:28:48  <johill> but maybe my code is just too old now
2007-08-18T00:28:49  <grzywacz> xorAxAx: http://spoko.math.uni.lodz.pl/~grzywacz/forms/search.png
2007-08-18T00:30:39  <dreimark> johill: looks buggy text default is text=u''
2007-08-18T00:31:55  <xorAxAx> grzywacz: hmm
2007-08-18T00:32:05  <xorAxAx> grzywacz: in a normal message, the links would be clickable
2007-08-18T00:32:11  <xorAxAx> --> better usability
2007-08-18T00:33:07  <grzywacz> xorAxAx: can't be done, I'm afraid
2007-08-18T00:33:10  <johill> dreimark: I think the bug is a missing result=''
2007-08-18T00:33:35  <grzywacz> xorAxAx: at least they are easy to copy&paste
2007-08-18T00:33:38  <johill> but I can't look now
2007-08-18T00:33:40  <xorAxAx> grzywacz: P
2007-08-18T00:33:42  <johill> am off
2007-08-18T00:33:44  <grzywacz> xorAxAx: with oob they will be clickable, tho
2007-08-18T00:33:44  <xorAxAx> s/P/?/
2007-08-18T00:33:54  <grzywacz> xorAxAx: ?
2007-08-18T00:34:04  <xorAxAx> grzywacz: if you dont use forms, they would be clickable as well, nor?
2007-08-18T00:34:09  <grzywacz> xorAxAx: I can't make them clickable, it's a client issue
2007-08-18T00:34:13  <xorAxAx> grzywacz: that means "replace the P by a question mark"
2007-08-18T00:34:21  <grzywacz> I know what that means. :P
2007-08-18T00:34:33  <xorAxAx> grzywacz: you can send them in a normal message and have 90% of the clients show the clickable
2007-08-18T00:34:40  <grzywacz> xorAxAx: sure
2007-08-18T00:34:43  <xorAxAx> its just that i dont seethe point using a form here
2007-08-18T00:34:47  <grzywacz> xorAxAx: there's a text-based search as well
2007-08-18T00:34:59  <dreimark> johill: before the if construct it should be initialized
2007-08-18T00:35:15  <grzywacz> xorAxAx: the point is it may be more intuitive to perform a form search as a response to some other action
2007-08-18T00:35:29  <xorAxAx> hmm?
2007-08-18T00:35:38  <xorAxAx> why?
2007-08-18T00:35:49  <grzywacz> xorAxAx: for example, you get a pange change notification, choose a "Perform a search option", you get a search form, fill a field or two, submit, get a result
2007-08-18T00:35:54  <grzywacz> Then it's all consistent.
2007-08-18T00:36:30  <xorAxAx> yes, cant you show a message in the same window?
2007-08-18T00:36:35  <xorAxAx> or will a client open another one?
2007-08-18T00:36:49  <grzywacz> Psi ignores message data when form is present.
2007-08-18T00:37:07  <grzywacz> I'm not sure if that's a correct behaviour.
2007-08-18T00:38:16  <grzywacz> xorAxAx: I should probably use "fixed" fields, not text-single anyway, because they're not supposed to be edited in any way
2007-08-18T00:38:25  <CIA-14> moin: Marian Neagul marian@info.uvt.ro main * 2884:2f4cdabd953e 1.7-classify-mneagul/MoinMoin/PageEditor.py: Compose PageChangedEvent disable requests and allow a single event handler to change the extra field
2007-08-18T00:38:50  <grzywacz> xorAxAx: but you can't select text in Psi if a field is fixed, so that would completly kill usability with, like, the most popular jabber client?
2007-08-18T00:38:58  <xorAxAx> grzywacz: and if you dont send a form at all, you will have a new window?
2007-08-18T00:39:34  <grzywacz> xorAxAx: in Psi? Submitting closes a form windows, so yes.
2007-08-18T00:39:45  <grzywacz> There will be a new message to be opened in a new window, if there's none
2007-08-18T00:40:02  <grzywacz> Or a window with a "next" button if some message from that jid is being displayed
2007-08-18T00:40:15  <johill> uh, my error log is huge
2007-08-18T00:40:19  <johill> I need to go through
2007-08-18T00:40:21  <grzywacz> But then, it's relying on Psi's behaviour. Not good.
2007-08-18T00:40:23  <xorAxAx> so where is the usability issue if you just send an ordinary response instead of a data form?
2007-08-18T00:40:27  <johill> and fix all the bugs
2007-08-18T00:40:38  <grzywacz> xorAxAx: style of the interface changes
2007-08-18T00:40:42  <xorAxAx> wellll, which other data form clients exist?
2007-08-18T00:40:44  <grzywacz> xorAxAx: and you get dropped to text-based one
2007-08-18T00:40:54  <grzywacz> xorAxAx: you have to know text commands to proceed
2007-08-18T00:40:55  <xorAxAx> grzywacz: thats completly minor compared to the issue of non-clickable links
2007-08-18T00:41:08  <grzywacz> xorAxAx: clickable-links are client specific as well
2007-08-18T00:41:20  <xorAxAx> show me any client that doesnt offer them :)
2007-08-18T00:41:27  <grzywacz> xorAxAx: imcom
2007-08-18T00:41:29  <grzywacz> :P
2007-08-18T00:41:31  <xorAxAx> argh
2007-08-18T00:41:39  * xorAxAx slaps grzywacz with a terminal emulator
2007-08-18T00:41:43  * grzywacz dances
2007-08-18T00:41:46  <xorAxAx> :)
2007-08-18T00:41:50  <grzywacz> Let me try with tkabber
2007-08-18T00:42:04  <xorAxAx> show me a client with data form support that is not called psi :)
2007-08-18T00:42:10  <grzywacz> xorAxAx: tkabber
2007-08-18T00:42:11  <grzywacz> :P
2007-08-18T00:43:21  <xorAxAx> any other?
2007-08-18T00:43:32  <grzywacz> xorAxAx: not that I'm aware of
2007-08-18T00:43:53  <xorAxAx> well, then you can have a look at the usability of the click path in that client as well
2007-08-18T00:44:28  <grzywacz> xorAxAx: can't really atm, because I rely on service discovery and this has been fixed in tkabber just recently after my bug report
2007-08-18T00:44:35  * grzywacz is rolling on the floor & crying
2007-08-18T00:44:36  <grzywacz> ;]
2007-08-18T00:44:55  <grzywacz> Let's see if they have nightlies (lol)
2007-08-18T00:45:44  <xorAxAx> haha, a flag called extra_dirty
2007-08-18T00:45:49  <grzywacz> rotfl
2007-08-18T00:45:54  <|Kev|> moin :)
2007-08-18T00:47:43  <xorAxAx> <message>hi</message>
2007-08-18T00:48:19  <|Kev|> 'lo
2007-08-18T00:48:41  <grzywacz> hi ;)
2007-08-18T00:49:35  <xorAxAx> didnt you want to ask |Kev| why psi doesnt show messages in case of a data form is present? :)
2007-08-18T00:49:47  <grzywacz> xorAxAx: we're talking on jabber :)
2007-08-18T00:50:23  <xorAxAx> and what did he say? :)
2007-08-18T00:50:36  <grzywacz> Nothing, I've just pasted him our merry chat. :P
2007-08-18T00:53:20  <|Kev|> would you rather I responded in here or xmpp?
2007-08-18T00:53:44  <xorAxAx> i am curious as well
2007-08-18T00:53:57  <grzywacz> |Kev|: here :)
2007-08-18T00:54:02  <|Kev|> I've only skimmed the epic paste
2007-08-18T00:54:29  <|Kev|> but my first impression is that the results of a search may as well be presented in a plain message as in a dataform
2007-08-18T00:54:42  <dreimark> neagulm: I do miss similiar pages listed currently
2007-08-18T00:54:55  <grzywacz> |Kev|: but not with two at the same time...
2007-08-18T00:55:12  <|Kev|> not with two /what/ at the same time? :)
2007-08-18T00:55:32  <grzywacz> Not both in a message and in a data form. That's a silly assumption, of course.
2007-08-18T00:55:48  <grzywacz> disregard
2007-08-18T00:56:30  <xorAxAx> well, my point is that first of all, the canonical way is totally unsusablly (fixed fields) and the current workaround is still unusable (editable and needs to be copied)
2007-08-18T00:59:49  <|Kev|> using the form for the result of a query doesn't seem necessary to me
2007-08-18T01:00:04  <xorAxAx> me neither
2007-08-18T01:00:10  <|Kev|> what's the motivation for doing it that way?
2007-08-18T01:00:20  <xorAxAx> didnt get his reason yet :)
2007-08-18T01:00:29  <xorAxAx> well, he can offer actions from the form
2007-08-18T01:00:33  <grzywacz> |Kev|: it fits into a "command flow", when you navigate through a sequence of forms
2007-08-18T01:00:43  <xorAxAx> launching a form from a chat window is not so easier
2007-08-18T01:00:49  <grzywacz> |Kev|: being dropped to text interface is nasty
2007-08-18T01:00:50  <|Kev|> right, if there are additional actions it makes sense
2007-08-18T01:00:52  <xorAxAx> but i think that clicking a link is more important
2007-08-18T01:01:03  <xorAxAx> than being able to navigate via an action
2007-08-18T01:01:31  <xorAxAx> esp. because digesting the search results is the primary usecase - and not activating most actions per minutes
2007-08-18T01:01:37  <grzywacz> I think fixed fields should be selectable and clickable when displaying URLs. It's a client issue, of course. And maybe a matter of an official recommendation.
2007-08-18T01:02:05  <|Kev|> fixed fields being selectable sounds sensible, yes
2007-08-18T01:02:12  <grzywacz> xorAxAx: you may want to get an another search form right after getting a result
2007-08-18T01:02:16  <|Kev|> whether they should be linkified is less clear
2007-08-18T01:02:48  <grzywacz> |Kev|: only when they contain a URL. Message text is meant to be opaque, yet clients make links clickable.
2007-08-18T01:02:51  <|Kev|> xorAxAx: I do agree in principle with grzywacz; if there's going to be a continued dialog between the client and moin then keeping it within data forms makes an amount of sense
2007-08-18T01:03:09  <xorAxAx> |Kev|: primarily i want to use the search results :)
2007-08-18T01:03:16  <xorAxAx> and thats a client <-> browser dialog
2007-08-18T01:03:32  <|Kev|> yes, there has to be a way to use the results, or having the additonal searches is pretty useless :)
2007-08-18T01:03:41  <grzywacz> xorAxAx: yes, but I'd rather get it right in general sense and leave fixing interface to Psi, than break the flow
2007-08-18T01:03:58  <xorAxAx> grzywacz: thats too naive IMHO :)
2007-08-18T01:04:12  <grzywacz> It may be suboptimal atm, but things can be fixed in a relatively short amount of time.
2007-08-18T01:04:18  <xorAxAx> and psi wont make it usable if you cant convince people like |Kev|
2007-08-18T01:04:22  <xorAxAx> 01:02:17 < |Kev|> whether they should be linkified is less clear
2007-08-18T01:04:23  <grzywacz> Psi can be fixed before moin 1.7 is released :P
2007-08-18T01:04:50  <grzywacz> xorAxAx: I've already responded to that
2007-08-18T01:05:03  <grzywacz> Message text is opaque to client.
2007-08-18T01:05:17  <CIA-14> moin: Marian Neagul marian@info.uvt.ro main * 2885:890aeabf3ed9 1.7-classify-mneagul/MoinMoin/classifier/recommendation_system.py: Add some docstrings and fix sanity check
2007-08-18T01:05:20  <neagulm> dreimark, fixed
2007-08-18T01:05:21  <grzywacz> Making links clickable is entirely a usability decission of the interface programmer.
2007-08-18T01:05:28  <xorAxAx> grzywacz: nevertheless he isnt convinced yet :-)
2007-08-18T01:05:34  <grzywacz> And that's the case with forms as well.
2007-08-18T01:05:39  <grzywacz> xorAxAx: *yet* ;)
2007-08-18T01:05:47  <xorAxAx> well, but you have to judge in which way its most usable
2007-08-18T01:06:01  <xorAxAx> not in which way it could be made most usable if programmers could read your mind
2007-08-18T01:06:11  <xorAxAx> or used moin all day long with this feature
2007-08-18T01:06:16  <grzywacz> xorAxAx: imagine, if fixed fields with URLs were clickable and selectable, there would be no issue
2007-08-18T01:06:27  <xorAxAx> yes, kind of dreaming :)
2007-08-18T01:06:34  <grzywacz> eh...
2007-08-18T01:06:46  <|Kev|> if Psi displayed both the form and the message, presumably this would be moot
2007-08-18T01:06:53  <xorAxAx> yes
2007-08-18T01:07:04  <xorAxAx> grzywacz: and i am rather curious how tkabber behaves here
2007-08-18T01:07:29  <grzywacz> xorAxAx: tkabber shows message and form in separate windows, afaik
2007-08-18T01:07:42  <xorAxAx> and how does the form look like?
2007-08-18T01:07:42  <grzywacz> Unfortunately there are no guildelines.
2007-08-18T01:08:01  <grzywacz> xorAxAx: I'll prepare a screenshot in a minute
2007-08-18T01:08:11  <xorAxAx> sounds like a serious issue ... not suggesting usability guidelines in a form spec :)
2007-08-18T01:08:26  <grzywacz> Yes.
2007-08-18T01:09:00  <|Kev|> I can't stay up any longer, I'm afraid, I'm falling asleep
2007-08-18T01:09:04  <xorAxAx> |Kev|: is anybody using them at all besides moin? :)
2007-08-18T01:09:05  <grzywacz> :(
2007-08-18T01:09:06  <|Kev|> shall we continue this tomorrow or such?
2007-08-18T01:09:08  <xorAxAx> oh, then good night
2007-08-18T01:09:10  <|Kev|> xorAxAx: forms?
2007-08-18T01:09:13  <xorAxAx> yes
2007-08-18T01:09:18  <grzywacz> |Kev|: tomorrow please, if you can
2007-08-18T01:09:20  <|Kev|> yes, they're used in adhoc amongst other things
2007-08-18T01:09:27  <xorAxAx> what is adhoc?
2007-08-18T01:09:30  <|Kev|> which is used for controlling entities
2007-08-18T01:09:38  <|Kev|> so you can remote control clients, for example
2007-08-18T01:09:43  <|Kev|> and you can administer servers with it
2007-08-18T01:09:47  <grzywacz> These are simple usage patterns, though.
2007-08-18T01:10:36  <|Kev|> well, I'm still here for a moment then
2007-08-18T01:10:44  <|Kev|> but I'm very tired
2007-08-18T01:10:53  <|Kev|> why do you ask about other usage?
2007-08-18T01:11:07  <|Kev|> in fact, I'm mentoring a gsoc student with a forms project
2007-08-18T01:11:24  <grzywacz> |Kev|: what project is that?
2007-08-18T01:11:26  <neagulm> dreimark, i'm going to sleep.
2007-08-18T01:11:28  <grzywacz> I'd like to take a brief look.
2007-08-18T01:11:29  <|Kev|> (writing a framework for wysiwyg form creation, and a bot for presenting these to users
2007-08-18T01:11:34  <grzywacz> Ah, this one.
2007-08-18T01:11:36  <neagulm> dreimark, see you tomorrow
2007-08-18T01:11:41  <neagulm> good night all
2007-08-18T01:11:46  <dreimark> neagulm: wait a moment please Ia mstaring currently
2007-08-18T01:12:00  <neagulm> dreimark, ok
2007-08-18T01:12:33  <grzywacz> xorAxAx: regarding tkabber, the form looks esentially the same
2007-08-18T01:12:54  <grzywacz> xorAxAx: buth both the form and message text are displayed
2007-08-18T01:12:56  <grzywacz> *but
2007-08-18T01:13:25  <grzywacz> xorAxAx: fixed fields are not selectable either
2007-08-18T01:14:06  <dreimark> neagulm: I have redone all the steps MarianNeagul/RecommendationSystem
2007-08-18T01:14:13  <dreimark> and the result is broken
2007-08-18T01:14:23  <neagulm> what happens ?
2007-08-18T01:15:56  <dreimark> the output of wikiclusters is seperated from the similiar pages
2007-08-18T01:16:17  <|Kev|> grzywacz, xorAxAx: is there anything I can do for you tonight, or can this wait until tomorrow?
2007-08-18T01:16:17  <dreimark> neagulm: the box is below an itemlist
2007-08-18T01:17:16  <dreimark> neagulm: all similiar pages lists below a page are empty
2007-08-18T01:17:18  <xorAxAx> |Kev|: we can defer it
2007-08-18T01:17:18  <grzywacz> |Kev|: let's wait, good night :)
2007-08-18T01:17:22  <|Kev|> nn
2007-08-18T01:17:47  <grzywacz> xorAxAx: so...
2007-08-18T01:18:14  <xorAxAx> feel free to defer it as well, i am sure you have other stuff to do :)
2007-08-18T01:18:19  <grzywacz> xorAxAx: the basic issue is that there are no usability/interface guidelines in xmpp extensions :)
2007-08-18T01:18:46  <grzywacz> xorAxAx: yes, I'm going to leave it as is for now and implement the missing actions
2007-08-18T01:19:16  <xorAxAx> good night
2007-08-18T01:21:10  <grzywacz> night xorAxAx
2007-08-18T01:21:52  <neagulm> dreimark, I found the problem... the sanity checks are insane :)
2007-08-18T01:22:31  <dreimark> neagulm: ok, you can fix it after some sleep I just wanted to report it
2007-08-18T01:22:46  <neagulm> :)
2007-08-18T01:23:05  <dreimark> good night neagulm
2007-08-18T01:23:14  <neagulm> good night dreimark
2007-08-18T01:23:33  <dreimark> good night xorAxAx
2007-08-18T01:23:57  <neagulm> good night xorAxAx
2007-08-18T01:26:55  <grzywacz> dreimark: night
2007-08-18T01:27:08  <grzywacz> neagulm: night as well :)
2007-08-18T01:27:27  <neagulm> grzywacz, good night :)
2007-08-18T01:53:00  <grzywacz> http://successfulsoftware.net/2007/08/16/the-software-awards-scam/
2007-08-18T01:59:04  <dreimark> grzywacz: no linux support
2007-08-18T01:59:22  <dreimark> so its not perfect, hehe
2007-08-18T01:59:42  <dreimark> I try to sleep, good night
2007-08-18T02:11:25  <grzywacz> dreimark: of course it does support linux. It doesn't run on it as *well* as on windows. :P
2007-08-18T02:11:35  <grzywacz> *does
2007-08-18T02:38:48  <CIA-14> moin: Karol Nowak <grzywacz@sul.uni.lodz.pl> * 2694:724b6b93f2f9 1.7-jabber-knowak/jabberbot/xmppbot.py: Show search results with a normal form. Make search available from forms.
2007-08-18T02:38:49  <CIA-14> moin: Karol Nowak <grzywacz@sul.uni.lodz.pl> * 2695:89c40244fa6a 1.7-jabber-knowak/jabberbot/xmppbot.py: Don't add the warning to messages with data forms. It can be shown with the form.
2007-08-18T02:43:39  <grzywacz> Crap, I forgot my passwd for moinmo.in -_-;
2007-08-18T03:59:06  <CIA-14> moin: Karol Nowak <grzywacz@sul.uni.lodz.pl> * 2696:80cf17344f98 1.7-jabber-knowak/ (3 files in 2 dirs): Reverts from forms should work, but I didn't do much testing.
2007-08-18T03:59:11  <grzywacz> night
2007-08-18T06:15:45  *** irc.freenode.net sets mode: +o ChanServ
2007-08-18T07:38:08  <ThomasWaldmann> moin
2007-08-18T10:59:54  <grzywacz> moin ThomasWaldmann
2007-08-18T11:44:34  <|Kev|> moin grzywacz
2007-08-18T12:00:40  <grzywacz> hi |Kev|
2007-08-18T12:06:24  <DrFrasierCrane> hey
2007-08-18T12:06:32  <ThomasWaldmann> ho
2007-08-18T12:06:34  <DrFrasierCrane> now tell me if I understand this correctly
2007-08-18T12:08:21  <DrFrasierCrane> in Newton a want to create a view containing wiki contents in GTK Tree control and I want it to be structured. I wonder if it makes sense to base this structure on categories, I mean if I wanted to make tree-like category structure I would just make link in category B to category A and then category B becomes a child of A right?
2007-08-18T12:09:39  <ThomasWaldmann> categories are sets
2007-08-18T12:09:48  <ThomasWaldmann> and look at the "explorer" theme
2007-08-18T12:10:14  <xorAxAx> also note that this is off-topic here :)
2007-08-18T12:10:38  <DrFrasierCrane> hmm ok
2007-08-18T12:10:44  <DrFrasierCrane> I have autojoin on moin-dev sorry
2007-08-18T12:10:55  <DrFrasierCrane> I'll remove it and join only if I have something related
2007-08-18T12:11:12  <grzywacz> -_-;
2007-08-18T12:12:13  <xorAxAx> well
2007-08-18T12:12:21  <xorAxAx> he should have used #moin
2007-08-18T12:13:25  <xorAxAx> if it hadnt been the second time i told him about this, i would have been more detailed again
2007-08-18T12:14:32  <grzywacz> That was a strange reaction...
2007-08-18T12:14:58  <grzywacz> |Kev|: please ping us when you have time to continue the discussion we started yestarday
2007-08-18T12:15:30  <|Kev|> pong
2007-08-18T12:15:39  <grzywacz> That was fast. ;)
2007-08-18T12:16:38  <xorAxAx> grzywacz: can i see the tkabber screenshot?
2007-08-18T12:16:52  <grzywacz> xorAxAx: hm, wait
2007-08-18T12:38:45  <grzywacz> http://spoko.math.uni.lodz.pl/~grzywacz/forms/tkabber1.png
2007-08-18T12:38:51  <grzywacz> http://spoko.math.uni.lodz.pl/~grzywacz/forms/tkabber2.png
2007-08-18T12:39:50  <xorAxAx> grzywacz: hmm, how do you think about sending 1 message and 1 form?
2007-08-18T12:40:08  <grzywacz> xorAxAx: total mess :)
2007-08-18T12:40:45  <|Kev|> I'm off again now
2007-08-18T12:40:45  <|Kev|> bbl
2007-08-18T12:40:57  <grzywacz> heh
2007-08-18T12:42:27  <xorAxAx> grzywacz: why?
2007-08-18T12:42:41  <xorAxAx> a form only to allow to initiate the actions
2007-08-18T12:42:45  <xorAxAx> the results as a message
2007-08-18T12:49:14  * xorAxAx is gone jogging because of a time out :)
2007-08-18T12:56:42  <grzywacz> xorAxAx: ...
2007-08-18T13:56:37  <xorAxAx> re
2007-08-18T15:13:51  <CIA-14> moin: Marian Neagul marian@info.uvt.ro main * 2886:0758a5e8b77c 1.7-classify-mneagul/MoinMoin/classifier/recommendation_system.py: Fix (again) sanity check
2007-08-18T15:13:56  <CIA-14> moin: Marian Neagul marian@info.uvt.ro main * 2887:175d13c45fe3 1.7-classify-mneagul/MoinMoin/action/edit.py: Remove edit.py changes
2007-08-18T15:54:49  <dreimark> bbl
2007-08-18T17:36:13  * ThomasWaldmann finds that current 1.6 wiki parser is parsing links differently than 1.5.8
2007-08-18T17:45:19  <grzywacz> ...
2007-08-18T17:45:22  <grzywacz> so?:P
2007-08-18T17:49:35  <ThomasWaldmann> so i have to rewrite the converter using the 1.5.8 parser code...
2007-08-18T18:20:01  <CIA-14> moin: Marian Neagul marian@info.uvt.ro main * 2888:91ca8299970e 1.7-classify-mneagul/MoinMoin/ (PageEditor.py events/__init__.py): Make PageEditor.saveText send all kwarg data and correctly update some fields
2007-08-18T18:20:02  <neagulm> hello
2007-08-18T18:20:03  <CIA-14> moin: Marian Neagul marian@info.uvt.ro main * 2889:c213eed0e0b4 1.7-classify-mneagul/MoinMoin/events/classifiernotify.py: Instructs PageEditor to change edit-log based on classification result
2007-08-18T18:20:04  <CIA-14> moin: Marian Neagul marian@info.uvt.ro main * 2890:7975da5230ad 1.7-classify-mneagul/MoinMoin/theme/__init__.py: Correctly check if recommendation system is enabled
2007-08-18T18:20:05  <CIA-14> moin: Marian Neagul marian@info.uvt.ro main * 2891:8c73c575ec30 1.7-classify-mneagul/wiki/config/more_samples/wikiconfig_recommendation.py: Sample recommendation system wikiconfig
2007-08-18T18:20:06  <CIA-14> moin: Marian Neagul marian@info.uvt.ro main * 2892:132306d6e425 1.7-classify-mneagul/MoinMoin/events/ (spamfilter.py classifiernotify.py): Rename bayesian spam filter notification script
2007-08-18T18:20:09  <CIA-14> moin: Marian Neagul marian@info.uvt.ro main * 2893:96ff9cf30200 1.7-classify-mneagul/MoinMoin/events/recommender.py: Notification script for recommender training on page save
2007-08-18T18:27:24  <ThomasWaldmann> moin neagulm
2007-08-18T18:27:38  <neagulm> hello ThomasWaldmann
2007-08-18T18:27:44  <neagulm> hello dreimark
2007-08-18T18:27:56  <dreimark> hi neagulm
2007-08-18T18:28:36  <dreimark> neagulm: macro Action for Homepage cretion isn't fixed in current ?
2007-08-18T18:29:14  <dreimark> neagulm: and outpu from wikiclusters is wrong rendered
2007-08-18T18:30:28  <neagulm> dreimark, a simple diff between my MM/macro/Action.py and the one from 1.7 shows no differences
2007-08-18T18:30:48  <dreimark> lets see if there is a change in the wiki page
2007-08-18T18:31:47  <dreimark> neagulm: wikiclusters is probably right I was wondering about an empty Similiar Pages
2007-08-18T18:32:28  <neagulm> dreimark, I excluded the current page from recommendations
2007-08-18T18:35:20  <neagulm> dreimark, the recommendations displayed might differ from the one shown in WikiClusters because the cache is updated when classification differs from the cached one
2007-08-18T18:37:48  <dreimark> neagulm, ThomasWaldmann, johill Action(Edit ... is broken for 1.7 on user homepage creation
2007-08-18T18:38:04  <dreimark> error is: [[Action: Cannot have arguments without name following named arguments]]
2007-08-18T18:39:42  <dreimark> neagulm: http://localhost:8081/RecentChanges?action=login -> gives 'NoneType' object has no attribute 'div'
2007-08-18T18:43:31  <ThomasWaldmann> sounds like johill wants to help <g>
2007-08-18T18:46:33  * ThomasWaldmann is busy on converter stuff
2007-08-18T18:55:15  <neagulm> dreimark, in what situation does not a Page instance have a formatter attribute ?
2007-08-18T18:55:33  <neagulm> dreimark, to be correct: page.formatter == None
2007-08-18T19:00:34  <dreimark> neagulm: I think that happens if its not a real page
2007-08-18T19:01:02  <neagulm> dreimark, ok
2007-08-18T19:01:11  <zenhase> re ... i am back from an extended offline period
2007-08-18T19:01:32  <dreimark> wb zenhase, how are you doing?
2007-08-18T19:01:37  <CIA-14> moin: Marian Neagul marian@info.uvt.ro main * 2894:ffec3bd1158b 1.7-classify-mneagul/MoinMoin/classifier/recommendation_system.py: Fix special page error
2007-08-18T19:02:13  <CIA-14> moin: Florian Krupicka <florian.krupicka@googlemail.com> * 2736:2b8e34dd9467 1.7-template-fkrupicka/ (47 files in 25 dirs): merged main
2007-08-18T19:02:14  <CIA-14> moin: Florian Krupicka <florian.krupicka@googlemail.com> * 2737:8e19839f5cb9 1.7-template-fkrupicka/MoinMoin/templating/ (chrome.py helpers.py): Some more pydoc for the various methods
2007-08-18T19:02:15  <CIA-14> moin: Florian Krupicka <florian.krupicka@googlemail.com> * 2738:f40a4358197e 1.7-template-fkrupicka/MoinMoin/templating/theme.py: Use the chrome methods for adding the rel-links in <head>
2007-08-18T19:02:16  <CIA-14> moin: Florian Krupicka <florian.krupicka@googlemail.com> * 2739:cf4925d2776b 1.7-template-fkrupicka/MoinMoin/templating/_base.html: Removed explicit link-rel-tags, reworked the comments a bit
2007-08-18T19:02:18  <CIA-14> moin: Florian Krupicka <florian.krupicka@googlemail.com> * 2740:8345f188ea50 1.7-template-fkrupicka/docs/CHANGES.genshitheme: Added documentation in docs/CHANGES.genshitheme
2007-08-18T19:02:21  <CIA-14> moin: Florian Krupicka <florian.krupicka@googlemail.com> * 2741:1c99a8acdd38 1.7-template-fkrupicka/MoinMoin/templating/chrome.py: And another test_sourcecode fix (missed a param doc)
2007-08-18T19:02:24  <CIA-14> moin: Florian Krupicka <florian.krupicka@googlemail.com> * 2742:52f94cd6c7cf 1.7-template-fkrupicka/ (130 files in 7 dirs): Made genshitest-theme into the example-theme, added some comments in the template too
2007-08-18T19:03:42  <zenhase> dreimark: regarding the finishing touches to the project or regarding my person?
2007-08-18T19:04:21  <dreimark> zenhase: your person in first place
2007-08-18T19:06:56  <zenhase> i am processing some recent events from the last week
2007-08-18T19:07:10  <zenhase> but in general i am ok :)
2007-08-18T19:08:03  <ThomasWaldmann> wb zenhase :)
2007-08-18T19:09:25  <zenhase> hi thomas
2007-08-18T19:09:49  <dreimark> zenhase: :)
2007-08-18T19:10:12  <dreimark> neagulm: what are your current plans till Monday 7pm?
2007-08-18T19:10:18  <zenhase> i would like you to comment on the CHANGES.genshitheme and example.html
2007-08-18T19:11:35  <neagulm> dreimark, write tests for the recommendation system, update the tests for the spam filter and add docstring where needed
2007-08-18T19:11:38  <zenhase> when you write code/software you tend to lose the ability to distinguish what could be non-obvious and needs to be documented
2007-08-18T19:12:59  <neagulm> and... maybe rewrite the ART implementation... :)
2007-08-18T19:13:20  <dreimark> neagulm: ok, what is with the users recommendations?
2007-08-18T19:13:20  <ThomasWaldmann> $ hg up
2007-08-18T19:13:23  <ThomasWaldmann> abort: untracked local file 'MoinMoin/_tests/test_sourcecode.py' differs from remote version
2007-08-18T19:13:26  <ThomasWaldmann> wtf?
2007-08-18T19:13:35  <zenhase> o_O
2007-08-18T19:13:47  <grzywacz> b0rk3d
2007-08-18T19:13:50  <ThomasWaldmann> (happens after hg pull ; hg update in templating branch)
2007-08-18T19:13:59  <zenhase> omg
2007-08-18T19:14:25  <zenhase> what happened there
2007-08-18T19:14:34  <grzywacz> zenhase: are you sure that 52f94cd6c7cf had to touch 130 files?
2007-08-18T19:14:35  <grzywacz> ;)
2007-08-18T19:15:08  <zenhase> grzywacz: hg mv of a dir with css and img-files :/
2007-08-18T19:15:17  <dreimark> neagulm:if that is not ready till monday add a ToDo and use the same for anonymous.
2007-08-18T19:15:36  * ThomasWaldmann makes a new clone
2007-08-18T19:15:44  <neagulm> dreimark, if I think better I will leave the ART rewrite for the next week... after i read again the original article
2007-08-18T19:16:27  <dreimark> neagulm: ok, testing and fixing will be enough till monday
2007-08-18T19:16:36  <zenhase> hmm, is there something like 'blame' in hg?
2007-08-18T19:16:42  <dreimark> neagulm: and extended documentation
2007-08-18T19:16:44  <xorAxAx> yes
2007-08-18T19:16:47  <xorAxAx> zenhase: hg annotate
2007-08-18T19:17:05  <zenhase> ah ok
2007-08-18T19:18:14  <neagulm> dreimark, I hope I will finish the user part until Monday but I'm not really sure about the correct use case... It really depends on what the wiki owner wants to accomplish...
2007-08-18T19:20:41  <dreimark> neagulm: A difference to anonymous are acl protected pages. If a wiki has a public and a restricted part
2007-08-18T19:20:49  <dreimark> this will make the difference
2007-08-18T19:20:49  <xorAxAx> discussing use cases three days before end rocks! :)
2007-08-18T19:21:07  <grzywacz> xorAxAx: ekhem
2007-08-18T19:21:07  <grzywacz> ;]
2007-08-18T19:22:41  <ThomasWaldmann> zenhase: needs to be indented so it is compatible with CHANGES
2007-08-18T19:23:08  <ThomasWaldmann> zenhase: also maybe add some samples to the howto
2007-08-18T19:23:35  <ThomasWaldmann> zenhase: the known main issues looks far too empty :)
2007-08-18T19:24:30  <zenhase> yeah, i wanted to skip through all my notes again to produce those :)
2007-08-18T19:24:33  <ThomasWaldmann> (maybe divide it into two parts: known main issues of the genshitheme stuff and known main issues why you couldn't do more)
2007-08-18T19:24:53  <zenhase> that sounds reasonable
2007-08-18T19:26:00  <dreimark> xorAxAx: use case are ever discussed
2007-08-18T19:27:13  <neagulm> xorAxAx, I had something in mind before starting and now I have totally different ideas...
2007-08-18T19:28:06  <xorAxAx> neagulm: hmm, happens sometimes
2007-08-18T19:28:23  <xorAxAx> neagulm: but the world wont stop spinning, even after 20th, you may commit stuff :)
2007-08-18T19:29:47  <neagulm> xorAxAx, dreimark Currently I'm thinking about using an different approach similar to the ones used on social networking sites.
2007-08-18T19:30:27  <neagulm> xorAxAx, dreimark it could be combined with the content based recommender
2007-08-18T19:30:41  <dreimark> neagulm: example
2007-08-18T19:30:54  <xorAxAx> neagulm: different approach for what?
2007-08-18T19:31:04  <neagulm> xorAxAx, recommendations
2007-08-18T19:31:59  <xorAxAx> present the current implementation and your new idea
2007-08-18T19:32:22  <neagulm> dreimark, a clustering algorithm could be used in computing "user clusters" and extract recommendations based on this clusters and the content based recommendation
2007-08-18T19:32:47  <neagulm> xorAxAx, the current recommendation system is a pure content based recommender
2007-08-18T19:32:53  <xorAxAx> yes
2007-08-18T19:33:23  <xorAxAx> web 2.0 sites use tag clouds which simply do scaling of words by occurence of user suggestions
2007-08-18T19:35:19  <neagulm> xorAxAx, recommendations could be made not only from content but combined with user behavior (this is the place where tags appear). It can be based on similarity between users
2007-08-18T19:35:39  <grzywacz> Neverending projects. 8)
2007-08-18T19:35:39  <neagulm> xorAxAx, clustering algorithms are perfect for the type of applications you say
2007-08-18T19:35:52  <xorAxAx> neagulm: why?
2007-08-18T19:35:57  <neagulm> xorAxAx, http://labs.google.com/papers/orkut-kdd2005.html
2007-08-18T19:36:53  <xorAxAx> sounds nice
2007-08-18T19:36:58  <xorAxAx> too busy to read it right now, though :)
2007-08-18T19:37:53  <neagulm> xorAxAx, clustering algorithm allow you to find groups of users, and you can do stuff based on this groups (page recommendations, product recommendation, marketing, and so on)
2007-08-18T19:38:07  <xorAxAx> yeah
2007-08-18T19:38:17  <xorAxAx> well, do you want to cluster users or pages?
2007-08-18T19:38:19  <xorAxAx> in our case
2007-08-18T19:38:52  <neagulm> both
2007-08-18T19:39:02  <xorAxAx> well, how do you want to cluster users?
2007-08-18T19:39:05  <xorAxAx> they are nearly stateless
2007-08-18T19:39:21  <xorAxAx> and a wikihomepage doesnt provide any valuable content
2007-08-18T19:40:23  <neagulm> pages visited, pages modified (obviously a user modifies pages that are of interest for him), pages deleted
2007-08-18T19:40:55  <grzywacz> neagulm: I'm not sure, some users may be doing trivial maintenance changes
2007-08-18T19:41:07  <neagulm> whether a user subscribed a page, etc I think that we could find lots of examples
2007-08-18T19:45:30  <ThomasWaldmann> zenhase: if you merge main again, the 2 failing tests will be skipped
2007-08-18T19:48:41  <dreimark> neagulm:I have no idea how the result could be shown. it taste something similiar to PageSize or the gdchart which
2007-08-18T19:48:48  <dreimark> shows how the wiki is accesed
2007-08-18T19:49:02  <neagulm> grzywacz, research stuff... it may all be useless... but I'm pretty sure that it will work, I have "that feeling" :)
2007-08-18T19:49:23  <zenhase> ThomasWaldmann: the jabberbot-tests?
2007-08-18T19:51:43  <grzywacz> neagulm: :)
2007-08-18T19:51:53  <xorAxAx> neagulm: yes, sounds nice but needs a lot of data that you would need to track
2007-08-18T19:52:01  <xorAxAx> or derive from existing logs
2007-08-18T19:52:13  <ThomasWaldmann> zenhase: yes
2007-08-18T19:52:13  <xorAxAx> neagulm: and sounds like a nice idea for the months after SOC :)
2007-08-18T19:54:16  <neagulm> xorAxAx, yes... I will be part in a team from my University that was awarded a research grant from EU in this field. A researcher that works in that project suggested me to use ART for recommendations
2007-08-18T19:55:32  <ThomasWaldmann> neagulm: sounds great :)
2007-08-18T19:58:15  <neagulm> ThomasWaldmann, the MoinMoin project is great, I learned lots of stuff: both MoinMoin internals and machine learning
2007-08-18T19:58:46  <neagulm> ThomasWaldmann, dreimark xorAxAx but for now I have to write docstrings and tests :D
2007-08-18T19:59:06  <dreimark> neagulm: good choice :)
2007-08-18T19:59:10  <neagulm> i'l leave the ideas for later
2007-08-18T19:59:22  <xorAxAx> hehe
2007-08-18T19:59:42  <xorAxAx> maybe you can suggest your team to allow you to use moin as a testbed for algorithms
2007-08-18T20:00:18  <grzywacz> That would be killing two birds with one stone. ;]
2007-08-18T20:02:01  <zenhase> that sounds harsh and cruel oO
2007-08-18T20:02:19  <dreimark> neagulm, xorAxAx : Is destinguishing user part from anonymous now wanted  (difference are acl pages)
2007-08-18T20:04:22  <xorAxAx> dreimark: hmm?
2007-08-18T20:04:36  <xorAxAx> zenhase: he didnt say rabbits :)
2007-08-18T20:04:44  <dreimark> hehe
2007-08-18T20:04:58  <xorAxAx> nor tree friends
2007-08-18T20:05:07  <zenhase> :o
2007-08-18T20:05:12  <zenhase> =:)
2007-08-18T20:06:00  <dreimark> neagulm: xorAxAx: the current version allows staring for recomendation for anonymous only
2007-08-18T20:06:06  <neagulm> dreimark, one use case would be that the recommender is initially available only to administrator for training. After the expected results are obtained it could be enabled for each user
2007-08-18T20:06:13  <xorAxAx> dreimark: staring?
2007-08-18T20:06:18  <neagulm> xorAxAx, :D
2007-08-18T20:06:57  <neagulm> xorAxAx, dreimark I will replace it with something like "Train" or "I like it!"
2007-08-18T20:08:21  <dreimark> neagulm: both are better
2007-08-18T20:08:50  <dreimark> mostly we use one word in the edit bar
2007-08-18T20:10:05  <neagulm> xorAxAx, dreimark administrators will have the choice of "Training on edit/Train by user/Train by Admin"
2007-08-18T20:10:23  <xorAxAx> sounds good
2007-08-18T20:10:38  <dreimark> ok
2007-08-18T20:16:36  <xorAxAx> lanius: so did you find the logs? (This message has been postponed.)
2007-08-18T20:17:21  <ThomasWaldmann> moin lanius :)
2007-08-18T20:17:49  <xorAxAx> btw, all students must deliver a main-merged branch
2007-08-18T20:17:53  <xorAxAx> otherwise they will be failed
2007-08-18T20:17:57  <xorAxAx> or something like that :)
2007-08-18T20:18:39  <lanius> hi
2007-08-18T20:18:49  <lanius> what is a main-merged-branch?
2007-08-18T20:19:14  <xorAxAx> a branch that contains the newest changesets from the main branch
2007-08-18T20:19:33  <zenhase> and if your off by even one changeset? :)
2007-08-18T20:19:34  <grzywacz> ah so
2007-08-18T20:19:36  <lanius> i see, should not be that difficult ;)
2007-08-18T20:19:47  <xorAxAx> lanius: you know, keppy last year ...
2007-08-18T20:19:51  <xorAxAx> i think he was a few months behind
2007-08-18T20:19:56  <xorAxAx> :-)
2007-08-18T20:19:57  <grzywacz> lol...
2007-08-18T20:21:22  <xorAxAx> zenhase: then we will deescalade it to the international court of justice in den haag
2007-08-18T20:21:31  <lanius> xorAxAx: :
2007-08-18T20:21:32  <lanius> 2007-06-29T23:48:19  <lanius> so we agree to drop the copyUnderlay stuff
2007-08-18T20:21:32  <lanius> 2007-06-29T23:48:26  <xorAxAx> yes
2007-08-18T20:22:01  <lanius> ThomasWaldmann: what issues are left? somehow my conection was lost yesterday and there is no log yet
2007-08-18T20:22:13  <xorAxAx> lanius: let me read up the context :)
2007-08-18T20:22:44  <lanius> http://moinmo.in/MoinMoinChat/Logs/moin-dev/2007-06-29
2007-08-18T20:23:15  <xorAxAx> lanius: yes, we concluded that "a script" is necessary
2007-08-18T20:23:22  <xorAxAx> so somebody needs to write "a script" :)
2007-08-18T20:23:44  <xorAxAx> and it would be nice if the underlay backend could enforce r-o semantics until its out of use
2007-08-18T20:24:05  <xorAxAx> if that is easily possible
2007-08-18T20:24:18  <xorAxAx> (like you can do an r-o mount of a fs)
2007-08-18T20:24:44  <lanius> the r-o does compete with the script idea
2007-08-18T20:25:10  <xorAxAx> hmm
2007-08-18T20:25:35  <xorAxAx> somehow, yes
2007-08-18T20:25:43  <xorAxAx> ok, its not very sensible
2007-08-18T20:26:04  <ThomasWaldmann> lanius: you have still todo in your changes file :)
2007-08-18T20:26:23  <xorAxAx> grzywacz: i fear you have to introduce exceptions
2007-08-18T20:26:44  <grzywacz> xorAxAx: yes?
2007-08-18T20:26:53  <xorAxAx> grzywacz: because returning a msg is ugly
2007-08-18T20:26:57  <grzywacz> ok
2007-08-18T20:27:02  <grzywacz> one moment
2007-08-18T20:27:27  <xorAxAx> grzywacz: and the cache clearing should happen in the pageeditor
2007-08-18T20:27:32  <lanius> ThomasWaldmann: yes one
2007-08-18T20:27:32  <xorAxAx> and the event as well
2007-08-18T20:27:41  <grzywacz> xorAxAx: ?
2007-08-18T20:27:54  <xorAxAx> grzywacz: well, currently its done in the action - that doesnt make much sense to me
2007-08-18T20:28:05  <ThomasWaldmann> lanius: the second item under "other notes" is unclear / misses context
2007-08-18T20:28:10  <xorAxAx> of course in a handler it would make sense as well
2007-08-18T20:28:13  <xorAxAx> just not i nthe action :)
2007-08-18T20:28:19  <grzywacz> xorAxAx: ok, will move it to PageEditor, then to handler
2007-08-18T20:28:23  <grzywacz> To keep the order of things.
2007-08-18T20:28:26  <xorAxAx> hehe
2007-08-18T20:29:29  <ThomasWaldmann> lanius: your refer to HelpOnConfiguration/Storage, where is it?
2007-08-18T20:30:03  <xorAxAx> somewhere in his repo
2007-08-18T20:33:37  <grzywacz> erm...
2007-08-18T20:34:48  <xorAxAx> hmm? :)
2007-08-18T20:34:52  <grzywacz> PageEditor defines SaveError, but it's only raised by antispam.py
2007-08-18T20:35:42  <grzywacz> ah, inheritance
2007-08-18T20:35:42  <xorAxAx> :)
2007-08-18T20:35:48  <ThomasWaldmann> lanius: why is LocalEditLog post-SOC?
2007-08-18T20:36:12  <grzywacz> Because 20th is in 2 days? :P :D
2007-08-18T20:36:18  <ThomasWaldmann> hah!
2007-08-18T20:36:29  <xorAxAx> because youth hostels arent a nice athmosphere for coding
2007-08-18T20:36:58  <lanius> ThomasWaldmann: it is not
2007-08-18T20:37:08  <lanius> it is only the attachement part
2007-08-18T20:37:24  <lanius> which still uses add which can be removed after that
2007-08-18T20:38:03  <ThomasWaldmann> lanius: well, that was just quoting your CHANGES file :)
2007-08-18T20:38:23  <lanius> but imcomplete
2007-08-18T20:38:32  <lanius> remove logfile.editlog.LocalEditLog.add
2007-08-18T20:38:41  <ThomasWaldmann> so I suggest you remove it except from attachments stuff
2007-08-18T20:38:49  <lanius> yes
2007-08-18T20:44:54  <ThomasWaldmann> maybe we can resolve the attachments stuff by migrating it to items
2007-08-18T20:47:11  <dreimark> neagulm: please describe __artlock__ and __recommendationlock__ too
2007-08-18T20:48:03  <ThomasWaldmann> lanius: btw, we could remove LocalEditLog usage completely by directly using metadata
2007-08-18T20:50:34  <dreimark> neagulm: sample wikiconfig_recommendation.py has some lines with to much blanks
2007-08-18T20:50:59  <dreimark> e.g. recommender = Recommender("tokens.db", "features.db", "art.db",
2007-08-18T20:51:01  <neagulm> dreimark, it's fixed in my repo
2007-08-18T20:51:57  <dreimark> neagulm: ok
2007-08-18T20:52:21  <lanius> ThomasWaldmann: yes, but this was easier for now
2007-08-18T20:54:07  <ThomasWaldmann> sure, maybe you can find some stuff to factor out when looking at the places it is used
2007-08-18T20:54:32  <ThomasWaldmann> e.g. "find oldest item rev newer than <timestamp>"
2007-08-18T20:59:09  <ThomasWaldmann> or "find newest page not from <editor>"
2007-08-18T21:15:49  <neagulm> dreimark, the patch for SpamBayes 1.1a4 XML-RPC plugin was pushed in SpamBayes CVS HEAD
2007-08-18T21:39:08  <dreimark> neagulm: :)
2007-08-18T21:51:32  <dreimark> so patching isn't neccesarry very good
2007-08-18T21:59:51  <neagulm> dreimark, i need to leave for ~1h walk
2007-08-18T22:00:26  <dreimark> neagulm: ok
2007-08-18T22:56:34  <dreimark> grzywacz: looks like moin ... maint cleancache is broken because of ImportError: No module named events
2007-08-18T22:58:30  <grzywacz> dreimark: what?
2007-08-18T22:58:54  <dreimark> grzywacz: http://moinmaster.wikiwikiweb.de/HelpOnMoinCommand?highlight=%28moin%29#head-06459b039ad12246c812093b72eb7227d89b17b4
2007-08-18T22:59:36  <grzywacz> dreimark: ???
2007-08-18T23:00:10  <dreimark> try moin --config-dir=$config_dir --wiki-url=http://localhost  maint cleancache
2007-08-18T23:00:33  <dreimark> with the 1.7 wiki instance set config_dir to the dir of wikiconfig
2007-08-18T23:00:59  <dreimark> that command is to clean the cache
2007-08-18T23:01:37  <dreimark> grzywacz: stop may be i am wrong
2007-08-18T23:01:42  <ThomasWaldmann> dreimark: are you missing a events dir in your data_dir?
2007-08-18T23:01:44  <dreimark> looks like a wrong version of moin.py
2007-08-18T23:02:52  <dreimark> ThomasWaldmann: grzywacz   false alarm, sorry
2007-08-18T23:03:59  <dreimark> ThomasWaldmann: how do you handle several moin versions (moin.py from script is copied to bin)
2007-08-18T23:05:48  <ThomasWaldmann> use sys.path
2007-08-18T23:08:08  <dreimark> ThomasWaldmann:  ok, another question
2007-08-18T23:08:33  <dreimark> should cleancache clean cache from underlay too?
2007-08-18T23:09:46  <dreimark> johill: may be we should change underlay/pages/MissingHomePage to [[Action(edit&template=HomepageTemplate,text=Create my home page now!)]]
2007-08-18T23:09:53  <dreimark> adding text=
2007-08-18T23:10:35  <dreimark> after that change the cache of that page has to be manually purged
2007-08-18T23:33:00  <ThomasWaldmann> dreimark: how do you do i18n for HomepageTemplate?
2007-08-18T23:33:48  <ThomasWaldmann> hm, ignore
2007-08-18T23:53:49  <neagulm> back
2007-08-18T23:54:54  <dreimark> neagulm: when do you plan tomorrow to be online?
2007-08-18T23:55:42  <neagulm> dreimark, 10:00EET 23:59EET
2007-08-18T23:56:08  <dreimark> neagulm: sounds good
2007-08-18T23:56:30  <neagulm> dreimark, :)
2007-08-18T23:59:50  <grzywacz> hehe