2007-06-25T08:35:16 <grzywacz> moin
2007-06-25T08:36:15 <ThomasWaldmann> moin
2007-06-25T09:00:28 <grzywacz> xorAxAx: are you already awake? :)
2007-06-25T09:06:49 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2267:d48a4d0a0c70 1.7-maninfo-vpv/ (MoinMoin/server/STANDALONE.py moin.py): pycallgraph support
2007-06-25T09:06:52 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2268:07fdbda38be2 1.7-maninfo-vpv/MoinMoin/ (Page.py config/__init__.py): move split_regex to config (so it is compiled only once)
2007-06-25T09:06:53 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2269:4197965b0820 1.7-maninfo-vpv/MoinMoin/config/__init__.py: fix: use re.UNICODE for split_regex
2007-06-25T09:06:54 <CIA-27> moin: Ville-Pekka Vainio <vpivaini AT cs DOT helsinki DOT fi> * 2270:3e69eb75f665 1.7-maninfo-vpv/ (4 files in 4 dirs): merge from main
2007-06-25T11:21:16 <xorAxAx> grzywacz: now i am :)
2007-06-25T11:21:26 <grzywacz> xorAxAx: ok
2007-06-25T11:21:59 <grzywacz> xorAxAx: I've realised that installing pyxmpp is a real pita, so I'm going to focus on MoinMoin these 2 days.
2007-06-25T11:22:17 <grzywacz> xorAxAx: the question is, should Moin always produce the nice, blue screen informing about error?
2007-06-25T11:22:32 <xorAxAx> grzywacz: yes
2007-06-25T11:22:50 <xorAxAx> anything else is a bug (or you disabled cgitb somehow)
2007-06-25T11:23:16 <grzywacz> xorAxAx: I haven't disabled anything, I think. I'll look into it first, then. Any directions?
2007-06-25T11:24:08 <xorAxAx> ask ThomasWaldmann, he said something about integrating it
2007-06-25T11:24:14 <xorAxAx> (the disabling option)
2007-06-25T11:24:46 <grzywacz> ThomasWaldmann: please comment on the problem above. :)
2007-06-25T12:05:22 <ThomasWaldmann> moin has config options to save the tracebacks to disk and not display them to the user (if that was the question)
2007-06-25T12:06:17 <grzywacz> Hm, then it's a bug, because sometimes the nice screen displayed, and sometimes not.
2007-06-25T12:07:27 <ThomasWaldmann> but?
2007-06-25T12:08:23 <grzywacz> But what?:)
2007-06-25T12:08:57 <ThomasWaldmann> what happens in the other case?
2007-06-25T12:09:19 <grzywacz> A plain python traceback is displayed.
2007-06-25T12:09:22 <grzywacz> Not the "blue page".
2007-06-25T12:09:45 <ThomasWaldmann> iirc, we have 2 handlers
2007-06-25T12:10:19 <grzywacz> For example exceptions thrown in DefaultConfig cause the ugly effect.
2007-06-25T12:10:20 <ThomasWaldmann> the pretty one is set when it can work, the simpler one is used before
2007-06-25T12:10:42 <grzywacz> Or when there's no accesible underlay.
2007-06-25T12:11:02 <ThomasWaldmann> (and maybe also if there is an exception in the pretty one, double fault)
2007-06-25T12:11:59 <grzywacz> hm
2007-06-25T12:12:23 <xorAxAx> yes, you have to look at the stacktrace very well
2007-06-25T12:15:27 <xorAxAx> grzywacz: http://hg.alexanderweb.de/moin-1.7-jabber-knowak/rev/741ab277616a -- self.notify_be_email = True -- s/be/by/
2007-06-25T12:16:32 <grzywacz> xorAxAx: woot, thanks
2007-06-25T12:17:36 <xorAxAx> ThomasWaldmann found it :)
2007-06-25T12:17:51 <grzywacz> weird typo
2007-06-25T12:18:26 <ThomasWaldmann> notify, be email! :)
2007-06-25T12:21:21 * ThomasWaldmann .oO(Romanes eunt domus)
2007-06-25T12:22:59 <grzywacz> 0,o
2007-06-25T12:25:08 <grzywacz> Argh, windows so fails for open source development. ;)
2007-06-25T12:26:10 <xorAxAx> i was using it for 2 years for moin development :)
2007-06-25T12:26:24 <grzywacz> Pain.
2007-06-25T12:26:30 <grzywacz> I'm not going to install cygwin here just to have ssh.
2007-06-25T12:26:30 <grzywacz> ;]
2007-06-25T12:27:01 <xorAxAx> you can use putty and plink
2007-06-25T12:31:11 <grzywacz> I can, how about mercurial? ;]
2007-06-25T12:32:50 <xorAxAx> also
2007-06-25T12:53:36 <grzywacz> argh
2007-06-25T12:54:25 <grzywacz> Forgot to identify myself.
2007-06-25T12:55:17 <grzywacz> ThomasWaldmann: please take a look at your priv ;)
2007-06-25T12:55:23 <xorAxAx> hopefully you didnt push :)
2007-06-25T12:55:34 <xorAxAx> in that case, you can revert your commit locally
2007-06-25T12:55:46 <grzywacz> No, I mean identify on irc.
2007-06-25T12:55:50 <xorAxAx> ah
2007-06-25T12:56:05 <grzywacz> Probably Thomas doesn't receive messages from unidentified users.
2007-06-25T12:56:06 <xorAxAx> yeah, tom didnt disable the censorship mode
2007-06-25T12:56:24 <grzywacz> That explains the lag with setting my temp ssh key. ;)
2007-06-25T12:56:53 <grzywacz> hm, lol
2007-06-25T12:57:01 <grzywacz> hg breaks with paths containing spaces...
2007-06-25T12:57:02 <grzywacz> -_-
2007-06-25T12:57:57 <xorAxAx> it shouldnt
2007-06-25T12:58:05 <xorAxAx> why do you think so?
2007-06-25T12:58:11 <grzywacz> "hg outgoing" fails.
2007-06-25T12:58:26 <xorAxAx> how?
2007-06-25T12:58:41 <grzywacz> "C:\Documents" is not a known command
2007-06-25T12:58:51 <grzywacz> abort: no suitable response from remote hg!
2007-06-25T12:59:10 <grzywacz> (It's "C:\Documents and settings" of course)
2007-06-25T12:59:47 <xorAxAx> well, then your config is broken, not hg :)
2007-06-25T12:59:48 <grzywacz> I'll try download the binary boundle with latest fixes.
2007-06-25T12:59:57 <xorAxAx> the one where you specified plink
2007-06-25T13:00:43 <grzywacz> Very cryptic eror for "outgoing" failing because of plink setting.
2007-06-25T13:00:49 <xorAxAx> i dont think so
2007-06-25T13:01:21 <grzywacz> Why would outgoing use plink settings?
2007-06-25T13:01:54 <grzywacz> ssh="C:\Documents and Settings\Grzywacz\Pulpit\plink.exe" -ssh -i "C:\Documents and settings\Grzywacz\privkey"
2007-06-25T13:02:01 <grzywacz> hm
2007-06-25T13:02:17 <grzywacz> Anything wrong with this?
2007-06-25T13:04:42 <grzywacz> This bug is even described on mercurial wiki.
2007-06-25T13:05:12 <xorAxAx> hmm
2007-06-25T13:05:25 <xorAxAx> then its a bug, yes
2007-06-25T13:14:18 <grzywacz> Waiting for Tomas to set my key, so I can test if this setup works. ;)
2007-06-25T13:31:37 <grzywacz> xorAxAx: writing unit tests, can I also somehow obtain a page object from outside like with request?
2007-06-25T13:31:52 <grzywacz> or should I instatiate one in the test code?
2007-06-25T13:33:23 <xorAxAx> you should instantiate one
2007-06-25T13:39:58 <grzywacz> Oh noes, I need python 2.4 for py.test...
2007-06-25T13:43:53 <xorAxAx> ?
2007-06-25T13:43:57 <xorAxAx> do you have 2.3?
2007-06-25T13:44:06 <grzywacz> yes
2007-06-25T13:44:35 <xorAxAx> on windows, installing 2 pythons is easy
2007-06-25T13:44:58 <grzywacz> Yes, but I feel that half of my time here will be spent on installing stuff.
2007-06-25T13:44:59 <grzywacz> ;S
2007-06-25T13:46:30 <xorAxAx> installing python is less than 5 minutes
2007-06-25T13:46:46 <grzywacz> Unit testing works, yay.
2007-06-25T13:46:47 <xorAxAx> keppy said the same and didnt install moin locally in his 3 months
2007-06-25T13:46:58 <xorAxAx> :-)
2007-06-25T13:47:07 <grzywacz> Please don't compare me with him. ;-)
2007-06-25T13:47:15 <xorAxAx> i am sorry :)
2007-06-25T13:47:37 <grzywacz> It's just that I miss the ease of use of apt/emerge. And I miss having all the stuff installed and ready. ;)
2007-06-25T15:01:00 <grzywacz> oh my, .hgignore doesn't work
2007-06-25T15:01:02 <grzywacz> ;<
2007-06-25T15:13:44 <ThomasWaldmann> re
2007-06-25T15:29:19 <ThomasWaldmann> lanius: backends.py ca line 145 there are some blanks too much
2007-06-25T15:31:04 <ThomasWaldmann> lanius: maybe we better use lowercase metadata keys instead of Size and Deleted. currently it makes no difference, but we might use an editor to edit metadata - and just using lower() is easier than producing some specific casing. and we also don't want to use lower() on both sides of a comparison all the time.
2007-06-25T15:33:45 <xorAxAx> ThomasWaldmann: the editor should show some localised user readable name
2007-06-25T15:33:54 <xorAxAx> so the internal key doesnt matter IMHO
2007-06-25T15:34:45 * ThomasWaldmann was talking of some easy to implement editor, just editing raw text, not a form
2007-06-25T15:35:32 <xorAxAx> even if he uses them lower cased, the user has to respect case if you want to collate the keys case-insensitively
2007-06-25T15:35:53 <ThomasWaldmann> lanius: better do from MoinMoin import x ; x.fn() than from MoinMoin.x import fn ; fn()
2007-06-25T15:36:36 <ThomasWaldmann> ?
2007-06-25T15:38:45 <ThomasWaldmann> dreimark: neagulm should please use # Comment instead of #Comment
2007-06-25T15:39:35 <ThomasWaldmann> xorAxAx: what's the problem of lower(key) for all user input keys?
2007-06-25T15:47:18 <ThomasWaldmann> grzywacz: btw, only 587km from lodz to vilnius :)
2007-06-25T16:09:58 <xorAxAx> ThomasWaldmann: at which point?
2007-06-25T16:20:52 <ThomasWaldmann> user edits some metadata in plain text editor (same as text edit now, just without the data), we go over it and lower() all keys
2007-06-25T16:22:55 <xorAxAx> ThomasWaldmann: you can upper() them as well :)
2007-06-25T16:23:13 <xorAxAx> but IMHO the metadata system should handle that
2007-06-25T16:23:17 <xorAxAx> not the layers above
2007-06-25T16:23:27 <ThomasWaldmann> yeah, but you can't make_them_as_wanted()
2007-06-25T16:23:41 <xorAxAx> and the user will see lowered keys on the next edit
2007-06-25T16:23:53 <xorAxAx> but nevertheless the lower layers need to handle it
2007-06-25T16:24:00 <xorAxAx> because the key is shared on many layers
2007-06-25T16:24:31 <ThomasWaldmann> my only point is that mixed cased keys are harder-than-necessary to handle
2007-06-25T16:24:55 <ThomasWaldmann> no matter where you handle them
2007-06-25T16:27:46 * ThomasWaldmann refactors wikidicts + tests of it
2007-06-25T16:30:29 <xorAxAx> using the event framework? :)
2007-06-25T16:30:38 <xorAxAx> that would allow moin to avoid polling
2007-06-25T16:36:11 <ThomasWaldmann> currently rather the dict/group classes
2007-06-25T16:37:34 <ThomasWaldmann> and event cant be used for that anyway
2007-06-25T16:38:07 <xorAxAx> why not?
2007-06-25T16:38:13 <xorAxAx> i mean for scandicts :)
2007-06-25T16:38:25 <xorAxAx> (which has been broken for 3 years :))
2007-06-25T16:38:27 <ThomasWaldmann> because it only works in same process
2007-06-25T16:39:11 <xorAxAx> the pickle would be updated via an event handler
2007-06-25T16:39:22 <xorAxAx> and every process checks the pickle regularly
2007-06-25T16:39:36 <xorAxAx> easy, fast, working
2007-06-25T16:40:47 <ThomasWaldmann> ah, ok, but it has still to poll the pickle
2007-06-25T16:41:32 <xorAxAx> yes, a stat every x seconds
2007-06-25T16:41:38 <xorAxAx> thats a timestamp in memory
2007-06-25T16:42:29 <ThomasWaldmann> ok, i added a todo :)
2007-06-25T16:44:31 <xorAxAx> (and dont forget locking :))
2007-06-25T16:44:35 <xorAxAx> for the pickle file
2007-06-25T16:44:43 <xorAxAx> but caching does that already
2007-06-25T16:49:50 <ThomasWaldmann> what if 2 group page update events happen at about the same time?
2007-06-25T16:51:43 <ThomasWaldmann> that might lead to an incomplete cache
2007-06-25T16:52:02 <xorAxAx> no
2007-06-25T16:52:12 <xorAxAx> well, then locking is broken
2007-06-25T16:52:18 <xorAxAx> the locking must ensure that this doesnt happen
2007-06-25T16:52:37 <xorAxAx> before reading the cache, the writer has to acquire a writelock
2007-06-25T16:53:41 <ThomasWaldmann> ok, then we need to do this additionally to what caching does internally
2007-06-25T16:54:27 <xorAxAx> hmm, no, you would use caching without the locking
2007-06-25T16:54:31 <CIA-27> moin: Marian Neagul marian@info.uvt.ro * 2295:3509dc595baa 1.7-classify-mneagul/MoinMoin/classifier/ (ReverendBayes.py SpamBayes.py __init__.py): Tokenizer refactoring.
2007-06-25T16:54:56 <xorAxAx> maybe you have read problems sometimes then, hmm
2007-06-25T16:55:06 <xorAxAx> the issue is that locking in the caching module isnt very sensible
2007-06-25T16:55:14 <xorAxAx> the caching module and the write lock must use the same lock
2007-06-25T16:55:27 <xorAxAx> so either caching exports locking primitives or it has to be disabled at all
2007-06-25T17:02:33 <ThomasWaldmann> caching could have some lock-read-callbacktomodify-write-unlock functionality
2007-06-25T17:03:41 <xorAxAx> hmm, yes
2007-06-25T17:04:02 <xorAxAx> not sure if that makes it simpler than the classic pattern but might be the case
2007-06-25T18:36:34 <ThomasWaldmann> http://code.google.com/p/yaki/issues/detail?id=8&can=2&q= deja vu
2007-06-25T18:37:39 <dreimark> bbl
2007-06-25T18:44:27 <ThomasWaldmann> http://python.net/~gherman/Contact.html another kind of email spam protection
2007-06-25T18:47:23 <lanius> ThomasWaldmann: some spaces too much?
2007-06-25T18:49:48 <ThomasWaldmann> yeah, in function calls or headers, you don't put blanks around =
2007-06-25T18:50:48 <ThomasWaldmann> btw, if you want to see some really bad formatted code, look at the yaki source
2007-06-25T18:51:01 <lanius> ah there, have to look into my pylint cfg, it should detect such things
2007-06-25T18:53:52 <ThomasWaldmann> but the other comments were more important :)
2007-06-25T18:54:25 <lanius> yes for the lowercase
2007-06-25T18:54:31 <lanius> for the imports you mean wikiQuote?
2007-06-25T18:55:53 <ThomasWaldmann> yes, but quite some others also
2007-06-25T18:56:31 <ThomasWaldmann> if you refactor code, it can be rather hard to search for stuff not using that module.name notation, but just name
2007-06-25T18:56:58 <ThomasWaldmann> esp. if name is a rather common thing, also used at other places
2007-06-25T18:58:45 <ThomasWaldmann> e.g. currently i am refactoring wikidicts module. if it is used like wikidicts.Group, a search for wikidicts will find it. if it is just Group, I will have to do a second search for that. and get quite some hits in the comments, etc.
2007-06-25T18:59:46 <lanius> yeah, but the wikiQuote things are quite obvious
2007-06-25T20:49:58 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2252:ece659592dd9 1.7/MoinMoin/ (_tests/test_wikidicts.py action/SyncPages.py wikidicts.py): wikidicts: refactoring, more tests
2007-06-25T20:50:00 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2253:2c8f25796120 1.7/ (4 files in 4 dirs): merged main
2007-06-25T21:16:37 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2271:ece659592dd9 1.7-maninfo-vpv/MoinMoin/ (_tests/test_wikidicts.py action/SyncPages.py wikidicts.py): wikidicts: refactoring, more tests
2007-06-25T21:16:40 <CIA-27> moin: Thomas Waldmann <tw AT waldmann-edv DOT de> * 2272:2c8f25796120 1.7-maninfo-vpv/ (4 files in 4 dirs): merged main
2007-06-25T21:16:41 <CIA-27> moin: Ville-Pekka Vainio <vpivaini AT cs DOT helsinki DOT fi> * 2273:95004c821dae 1.7-maninfo-vpv/MoinMoin/ (_tests/test_wikidicts.py action/SyncPages.py wikidicts.py): merge from main
2007-06-25T21:16:43 <CIA-27> moin: Ville-Pekka Vainio <vpivaini AT cs DOT helsinki DOT fi> * 2274:0b8ef94aa355 1.7-maninfo-vpv/MoinMoin/script/xmlrpc/manimport.py: Fix comments, apparently should be # text not #text
2007-06-25T21:29:10 <lanius> ThomasWaldmann: how can i avoid recursive imports?
2007-06-25T21:29:22 <lanius> ThomasWaldmann: the following problem
2007-06-25T21:29:42 <lanius> external.py which contains ItemCollection imports security.AccessControlList
2007-06-25T21:29:45 <lanius> security imports user
2007-06-25T21:29:49 <lanius> and user imports ItemCollection again
2007-06-25T21:29:56 <xorAxAx> what is external?
2007-06-25T21:30:20 <lanius> the new api ItemCollection/Item/Metadata/Data
2007-06-25T21:30:25 <xorAxAx> you can avoid them by deferring the import to the function itself or redesigning dependencies
2007-06-25T21:30:32 <xorAxAx> ?
2007-06-25T21:30:48 <xorAxAx> you mean external.py contains all of that?
2007-06-25T21:30:52 <lanius> yes
2007-06-25T21:32:31 <xorAxAx> then try moving imports into functions/methods
2007-06-25T21:48:08 <CIA-27> moin: Marian Neagul marian@info.uvt.ro * 2296:aab6fe285e9b 1.7-classify-mneagul/MoinMoin/classifier/ReverendBayes.py: New MoinMoin.parser based tokenizer.
2007-06-25T21:52:36 <neagulm> hello
2007-06-25T21:53:17 <ThomasWaldmann> hi neagulm
2007-06-25T21:53:27 <ThomasWaldmann> grzywacz: there is no rename event fired yet
2007-06-25T21:53:53 <xorAxAx> neagulm: hi, it would be nice if you could use the event framework for your stuff :)
2007-06-25T21:54:06 <xorAxAx> neagulm: that avoids changing the whole pageeditor etc. code
2007-06-25T21:55:13 <ThomasWaldmann> grzywacz: is it intended that delete fires changed and deleted events?
2007-06-25T21:55:50 <xorAxAx> ThomasWaldmann: thats what happens internally
2007-06-25T21:56:07 <xorAxAx> but he hasnt attempted to send just one event
2007-06-25T21:56:13 <xorAxAx> could be done, though
2007-06-25T21:56:24 <ThomasWaldmann> sure, but if you do expensive stuff, you might want to only do it once
2007-06-25T21:56:26 <neagulm> xorAxAx, ok, i will change the code to use the event system in the next push
2007-06-25T21:56:34 <xorAxAx> cool
2007-06-25T21:56:53 <dreimark> hi neagulm
2007-06-25T21:57:30 <ThomasWaldmann> esp. for rename this gets interesting as following the current scheme, it would fire 3 events (rename, 2x changed, maybe even 1x deleted?)
2007-06-25T22:00:30 <dreimark> neagulm: some PEP8 small bugs check wiki config for wissing blanks e.g. notification_server=None
2007-06-25T22:01:04 <dreimark> and comments should be # comment and not #comment e.g. #text_classifier
2007-06-25T22:03:34 <dreimark> neagulm: currently not all pages in HamPages are linked to theire wikipages
2007-06-25T22:03:48 <dreimark> especially if they are no WikiNames
2007-06-25T22:04:32 <dreimark> you should probably test for that and write them as ["example"]
2007-06-25T22:05:41 <dreimark> forget the last sentence that was an old one
2007-06-25T22:05:47 <neagulm> dreimark, I currently use wikiutil.pagelinkmarkup(page.page_name)
2007-06-25T22:06:01 <dreimark> its ok I was confused by an old entry
2007-06-25T22:06:43 <neagulm> dreimark, now I have a working Reverend classifier. We could use it for language detection and do the initial training with help pages :)
2007-06-25T22:07:03 <dreimark> very good
2007-06-25T22:07:42 <dreimark> the next would be to add some unit tests ?
2007-06-25T22:08:43 <neagulm> yes.. They will contain some tests about the interfaces and some benchmarking... This would help spot incorrect tokenization of pages
2007-06-25T22:09:07 <dreimark> neagulm: is the comment input text alredy checked too?
2007-06-25T22:10:07 <dreimark> and ThomasWaldmann, lanius how is that data handled in the new storage system
2007-06-25T22:10:08 <neagulm> dreimark, yes
2007-06-25T22:10:22 <dreimark> :(
2007-06-25T22:10:26 <dreimark> :) :)
2007-06-25T22:10:33 <dreimark> sorry wrong key
2007-06-25T22:13:59 <dreimark> neagulm: there is a bug somewhere in the names on HamPages / SpamPages
2007-06-25T22:14:31 <dreimark> always the first one is no ["wikiname"] eg. Emm1
2007-06-25T22:16:01 <dreimark> testing more may be its a relice
2007-06-25T22:16:04 <dreimark> relict
2007-06-25T22:17:17 <neagulm> Reverend requires more training than SB until it has a decent behavior...
2007-06-25T22:17:47 <neagulm> We will have some performance measures after I implement the tests...
2007-06-25T22:18:32 <xorAxAx> why did you chose reverend instead of SB?
2007-06-25T22:18:43 <dreimark> if one adds pages onto the Ham or Spam Pages page will the new ones automaticly used to enter into the categorisation?
2007-06-25T22:20:19 <dreimark> so one can provide a list with ham or spam pages
2007-06-25T22:20:31 <neagulm> no, Only after retraining. Should I implement this type of behavior ?
2007-06-25T22:21:30 <neagulm> xorAxAx, I did not chose to use use Reverend instead SB :) Reverend is realy small and good for '''basic''' language classification
2007-06-25T22:21:40 <xorAxAx> ok
2007-06-25T22:22:03 <neagulm> xorAxAx, It could be used when the user does not want to use SB...
2007-06-25T22:22:46 <dreimark> I guess that is a nice feature, because many of us will start with an 'old' wiki and the pages there are Ham
2007-06-25T22:23:46 <dreimark> xorAxAx: implenting two tools makes it easier to setup the code to make the categorizer exchangeable
2007-06-25T22:23:51 <xorAxAx> yeah
2007-06-25T22:26:26 <neagulm> SB is far more better then Reverend (at least at the empirical level)... It would be the recommend classifier
2007-06-25T22:26:49 <xorAxAx> ok
2007-06-25T22:26:57 <xorAxAx> and its somehow maintained
2007-06-25T22:28:18 <dreimark> I prefer a working small package added as supplement to MMs code to define how others should behave to get easy implemented or to show them how it could be done. And then we can show with spambayes
2007-06-25T22:28:35 <dreimark> that it is possible to use a very advanced categorizer
2007-06-25T22:29:06 <xorAxAx> well, just to have some good API, you dont need to ship reverend, but having it pluggable sounds nice
2007-06-25T22:30:13 <dreimark> xorAxAx: you can use it with DE a lot of people i know they start with DE to test features
2007-06-25T22:32:04 <xorAxAx> yeah, of course
2007-06-25T22:34:16 <neagulm> dreimark, xorAxAx I will have to go offline for ~1hour.... there are 25 degrees outside and in the student dormitory room is really hot...
2007-06-25T22:34:21 <neagulm> bbl
2007-06-25T22:34:33 <dreimark> oj
2007-06-25T22:34:35 <dreimark> ok
2007-06-25T22:41:06 <xorAxAx> cold and rainy here
2007-06-25T22:42:48 <dreimark> here too, last year it was to warm :)
2007-06-25T22:46:02 <dreimark> donred spams MM by attachments
2007-06-25T23:02:12 <dreimark> have to leave, gn
2007-06-25T23:03:36 <ThomasWaldmann> gn dreimark
2007-06-25T23:03:45 <ThomasWaldmann> and game over for donred
2007-06-25T23:05:45 <dreimark> :)
MoinMoin: MoinMoinChat/Logs/moin-dev/2007-06-25 (last edited 2007-10-29 19:10:35 by localhost)
