1 2011-08-19T00:00:29 <dreimark> and at the current place it has no effect for "New Item"
2 2011-08-19T00:00:49 <sinha> hmm, yes
3 2011-08-19T00:01:33 <sinha> dreimark: i guess we need some kind of sidebar to put these things ( i am not sure if i could do that before SOC ends)
4 2011-08-19T00:01:59 <dreimark> momentanly i think we should have it in a working state
5 2011-08-19T00:02:16 <dreimark> so we can merge it
6 2011-08-19T00:03:15 <dreimark> we are interested to get new developer all the time, so if you want to continue after soc this is very welcome
7 2011-08-19T00:04:14 <dreimark> sinha: does it make sense to have a select [] left of Actions
8 2011-08-19T00:04:27 <dreimark> if we have only this a new menu is not necessary
9 2011-08-19T00:05:35 <sinha> you mean put that checkbox separately ?
10 2011-08-19T00:06:04 <dreimark> yes, with a label
11 2011-08-19T00:08:22 <sinha> dreimark: wont it be better to put "New item" as separate ?
12 2011-08-19T00:08:28 <dreimark> no
13 2011-08-19T00:08:37 <dreimark> we need anyway the label
14 2011-08-19T00:09:13 <dreimark> if i ask a non programmer user he did not click on that checkbox
15 2011-08-19T00:11:02 *** grad
16 2011-08-19T00:11:50 <sinha> hmm, i see, so just a link as "select all" or checkbox also ?
17 2011-08-19T00:12:26 <dreimark> checkbox too, because this lets simple explain the checkboxes left of items
18 2011-08-19T00:16:56 * dreimark needs some sleep
19 2011-08-19T00:16:58 <dreimark> gn
20 2011-08-19T00:21:43 <sinha> gn
21 2011-08-19T00:21:46 *** sinha
22 2011-08-19T00:25:46 *** raignarok
23 2011-08-19T00:52:44 *** MattMaker
24 2011-08-19T05:30:39 *** Marchael
25 2011-08-19T05:32:43 <Marchael> moin
26 2011-08-19T06:24:58 *** Marchael
27 2011-08-19T06:55:36 *** Marchael
28 2011-08-19T07:07:57 *** Marchael1
29 2011-08-19T07:10:12 *** Marchael
30 2011-08-19T07:16:43 *** Marchael1
31 2011-08-19T07:21:02 *** Marchael
32 2011-08-19T07:21:41 *** pkumar
33 2011-08-19T07:22:54 *** Marchael
34 2011-08-19T07:29:10 *** pkumar
35 2011-08-19T08:02:26 *** Marchael
36 2011-08-19T08:02:54 <Marchael> moin
37 2011-08-19T08:16:54 *** Marchael
38 2011-08-19T08:18:10 *** Marchael
39 2011-08-19T08:44:13 <dreimark> moin
40 2011-08-19T09:07:13 <Marchael> ThomasWaldmann: pkumar found bug in renaming item, only last revision was renamed. revisions in all index stays as is.
41 2011-08-19T09:17:48 *** raignarok
42 2011-08-19T09:31:48 *** raignarok
43 2011-08-19T09:53:23 *** mkerrin
44 2011-08-19T09:57:16 <ThomasWaldmann> moin
45 2011-08-19T09:58:12 <ThomasWaldmann> Marchael: well, fix it :)
46 2011-08-19T09:59:02 <Marchael> ThomasWaldmann: ItemIndex() haven't rename() method, how could i do it?
47 2011-08-19T10:00:26 <ThomasWaldmann> you file a bug about it, exactly describing the issue
48 2011-08-19T10:00:43 <ThomasWaldmann> then you first finish what you have already begun and close THESE bugs
49 2011-08-19T10:04:33 <Marchael> ThomasWaldmann: could you review please http://codereview.appspot.com/4896049/ ps 8
50 2011-08-19T10:13:59 <Marchael> ThomasWaldmann: thx, I make s/delete/destroy/ but after may I commit?
51 2011-08-19T10:16:24 <ThomasWaldmann> yes
52 2011-08-19T10:17:44 <CIA-114> Michael Mayorov <marchael@kb.csu.ru> * 2151ef2c8425 r446 moin-2.0/MoinMoin/ (_tests/wikiconfig.py storage/_tests/test_indexing.py): Added tests for checking runtime indexing
53 2011-08-19T10:19:43 <Marchael> ThomasWaldmann: ok, could you point me to code with renaming. Yes, grep is my friend.
54 2011-08-19T10:20:49 <ThomasWaldmann> renaming isn't easy and maybe some other code is broken there, too
55 2011-08-19T10:21:27 <ThomasWaldmann> so maybe rather work on that other bug
56 2011-08-19T10:21:34 <Marchael> i guess that better add rename() method to ItemIndex, and call it instead of calling from somewhere else
57 2011-08-19T10:21:57 <Marchael> like rename(name, oldname, rev_no)
58 2011-08-19T10:22:11 <Marchael> because renaming process for all revs isn't trivial
59 2011-08-19T10:22:29 <ThomasWaldmann> i am not sure it can be fixed until the other code is also fixed
60 2011-08-19T10:22:40 *** raignarok
61 2011-08-19T10:24:12 <Marchael> hm, then what I should do now? write docs?
62 2011-08-19T10:25:16 <ThomasWaldmann> https://bitbucket.org/marchael/moin-2.0/issue/27/has_item-using-whoosh-patch-breaks-unit you see that some stuff is rather broken with rename
63 2011-08-19T10:25:22 *** mkerrin
64 2011-08-19T10:26:31 <ThomasWaldmann> Marchael: https://bitbucket.org/marchael/moin-2.0/issue/23/ how about that one?
65 2011-08-19T10:29:30 <ThomasWaldmann> Marchael: you didn't react to all my review comments btw for the indexing tests
66 2011-08-19T10:29:47 <ThomasWaldmann> i.e. it still does rmtree in setup AND teardown
67 2011-08-19T10:30:41 <ThomasWaldmann> also that rmtree shouldn't be implemented there
68 2011-08-19T10:33:09 <Marchael> what can I use instead of rmtree?
69 2011-08-19T10:33:34 <dreimark> re
70 2011-08-19T10:33:50 <ThomasWaldmann> the point is that this should be done by ItemIndex (or even whoosh), not by the tests
71 2011-08-19T10:34:21 <ThomasWaldmann> just imagine that whoosh uses some other storage, not in the filesystem in that index_dir
72 2011-08-19T10:34:26 <Marchael> ItemIndex has close method
73 2011-08-19T10:34:35 <Marchael> but it doesn't remove index
74 2011-08-19T10:37:01 *** mkerrin
75 2011-08-19T10:37:51 <ThomasWaldmann> def test_create_rev(self):
76 2011-08-19T10:37:59 <Marchael> ok, I'll do it
77 2011-08-19T10:38:01 <ThomasWaldmann> that could be rather called update_item
78 2011-08-19T10:38:26 <ThomasWaldmann> because it checks if an update works ok
79 2011-08-19T10:40:18 <Marchael> but why teardown is bad? I use it for cleaning old stuff
80 2011-08-19T10:40:43 <ThomasWaldmann> well, it is mostly duplicate
81 2011-08-19T10:41:35 <ThomasWaldmann> except for the last test method called. but then, if you ALWAYS make sure you have a clean index when starting a test, even that is not needed
82 2011-08-19T10:42:37 <ThomasWaldmann> but if you like to leave it cleanly, you can keep that, but move that rmtree to ItemIndex
83 2011-08-19T10:43:29 <ThomasWaldmann> (in the end, after refactoring all the stuff enough, we may have to talk to mchaput whether we move some generally useful code to whoosh)
84 2011-08-19T10:45:06 <Marchael> so I should create ItemIndex.clean() method and move rmtree there, right?
85 2011-08-19T10:48:49 <ThomasWaldmann> you could even call it .remove() to make more clear what it does
86 2011-08-19T10:48:58 <Marchael> ok
87 2011-08-19T10:50:23 <Marchael> ThomasWaldmann: could I implement rename() also?
88 2011-08-19T10:50:31 <Marchael> I mean rename for given revision
89 2011-08-19T10:55:23 *** greg_f
90 2011-08-19T10:56:44 <ThomasWaldmann> read what i already said
91 2011-08-19T10:57:30 <ThomasWaldmann> if you're finished with cleaning up your tests, see issue tracker, there's something new to implement
92 2011-08-19T10:57:38 <ThomasWaldmann> bbl
93 2011-08-19T13:52:29 *** sinha
94 2011-08-19T14:00:55 <sinha> dreimark: how about this http://moinmo.in/AkashSinha/Gsoc2011Diary/2011-08-18?action=AttachFile&do=get&target=filter1.png , when "More" is clicked helper-text is shown and otherwise they remain hidden
95 2011-08-19T14:07:33 *** raignarok_
96 2011-08-19T14:11:25 *** Marchael
97 2011-08-19T14:11:25 *** raignarok
98 2011-08-19T14:11:25 *** ronny
99 2011-08-19T14:12:53 *** ronny
100 2011-08-19T14:17:51 *** raignarok__
101 2011-08-19T14:20:50 *** raignarok_
102 2011-08-19T14:22:11 *** raignarok__
103 2011-08-19T14:29:11 *** Marchael
104 2011-08-19T14:38:45 *** raignarok
105 2011-08-19T15:15:24 <ThomasWaldmann> sinha: maybe binary file and unknown should be in same category
106 2011-08-19T15:16:24 <ThomasWaldmann> because application/octet-stream ("binary") basically means "we do not know how to handle this" which is about the same as for unknown types
107 2011-08-19T15:54:43 <sinha> ThomasWaldmann: to do that, i have to remove that entry from the pre-defined contenttype_groups ?
108 2011-08-19T15:55:15 <ThomasWaldmann> we need it there :)
109 2011-08-19T15:55:42 <ThomasWaldmann> because if you upload some stuff, you need the option to tell it is application/octet-stream
110 2011-08-19T15:56:23 <ThomasWaldmann> if it is a problem you can also leave it like it is, it is just a bit strange
111 2011-08-19T15:57:48 * ThomasWaldmann just viewed a local history view of an item with 3428 revisions. that took a while. ;)
112 2011-08-19T15:59:23 <sinha> yes, some time is getting involved in grouping of revisions, editors etc
113 2011-08-19T15:59:53 <ThomasWaldmann> for that you paging code will also help :)
114 2011-08-19T16:00:11 <sinha> yeah maybe, keep the number of revision less
115 2011-08-19T16:16:20 *** mkerrin
116 2011-08-19T16:25:43 <ThomasWaldmann> hmm, global index also takes ages
117 2011-08-19T16:32:37 <sinha> ThomasWaldmann: global index has those contenttype filtering thing
118 2011-08-19T16:33:20 <sinha> and because of these unknown types, i had to add some extra checks
119 2011-08-19T16:34:48 <ThomasWaldmann> sinha: dreimark: btw, in general, how is the ready-to-merge-state of your repo?
120 2011-08-19T16:35:45 <sinha> ThomasWaldmann: just has to resolve 1-2 bug
121 2011-08-19T16:35:48 <sinha> maybe by tomorrow
122 2011-08-19T16:36:04 <ThomasWaldmann> ok
123 2011-08-19T16:38:22 <ThomasWaldmann> sinha: btw, did you change anything in (local) history rendering?
124 2011-08-19T16:43:16 *** MattMaker
125 2011-08-19T16:48:01 <sinha> rendering ?
126 2011-08-19T16:48:13 <ThomasWaldmann> html template
127 2011-08-19T16:48:27 *** Marchael
128 2011-08-19T16:48:29 <sinha> ThomasWaldmann: in global history, yes i did
129 2011-08-19T16:49:38 <ThomasWaldmann> ok, maybe i don't touch it before the merge. i just found it looks a bit crappy (in main repo).
130 2011-08-19T17:04:30 *** pkumar
131 2011-08-19T17:26:54 <pkumar> ThomasWaldmann : hi, please look into http://paste.pocoo.org/show/460824/. there in line 11 we have empty value, this creates an error on item.commit. I'm not much sure why it behaves like that?
132 2011-08-19T17:28:03 <pkumar> the test pass on commenting/removing/replacing u'' with some non empty value
133 2011-08-19T17:31:01 <ThomasWaldmann> pkumar: looks like a whoosh bug
134 2011-08-19T17:36:20 <pkumar> ThomasWaldmann : yeah, do not create an empty item name
135 2011-08-19T17:37:15 <pkumar> ThomasWaldmann : so ,all the tests seems happy to me, will be happier once clean up work gets over in index dir
136 2011-08-19T17:38:09 <pkumar> index dir contents some how seem to tamper with backend.history incorrectly
137 2011-08-19T17:38:13 *** TheSheep
138 2011-08-19T17:38:13 <pkumar> ThomasWaldmann : ^
139 2011-08-19T17:38:32 <pkumar> somehow*
140 2011-08-19T17:40:08 <pkumar> ThomasWaldmann : when the tests are run, some index files are created in index dir. at what stage are they created exactly? in backend initialization?
141 2011-08-19T17:40:55 <pkumar> or is it item creation?
142 2011-08-19T18:15:33 <pkumar> brb
143 2011-08-19T18:18:52 *** sinha1
144 2011-08-19T18:19:28 *** sinha
145 2011-08-19T18:23:49 *** Marchael
146 2011-08-19T18:41:40 <ronny> sup
147 2011-08-19T19:20:00 <pkumar> Marchael : hi, can you please tell me, when exactly those index files are created in index directory. is it backend initialization?
148 2011-08-19T19:21:14 <Marchael> at moin start up
149 2011-08-19T19:21:33 <Marchael> may be at init_backend() stage
150 2011-08-19T19:28:39 <pkumar> k, you are currently working on test cleanup due to index files in index directory right?
151 2011-08-19T19:28:45 <pkumar> Marchael : ^
152 2011-08-19T19:32:51 <pkumar> because I'm having some issues if I cleanup by deleting all files from the directory at teardown
153 2011-08-19T19:33:22 <ThomasWaldmann> pkumar: the index is updated when items are saved into the backend
154 2011-08-19T19:33:44 <ThomasWaldmann> so, e.g. if items are loaded from xml, it will save them to backend and create/update the index
155 2011-08-19T19:34:24 <ThomasWaldmann> if load_xml is not used, moin will open the index sooner or later. if there is no index, it will autocreate it.
156 2011-08-19T19:46:38 <pkumar> ThomasWaldmann: say an instance here: a test from test_backend, http://paste.pocoo.org/show/460893/ now at the beginning of the test I delete all the contents from index dir. should the test raise an error on item.commit?
157 2011-08-19T19:47:19 <pkumar> or it should create an index file there?
158 2011-08-19T19:55:32 * ThomasWaldmann must go, but will be back asap
159 2011-08-19T19:57:00 <pkumar> ThomasWaldmann : if I do something like above. it gives this result : http://paste.pocoo.org/show/460899/
160 2011-08-19T19:57:26 <pkumar> the test run was test_backend_sql
161 2011-08-19T20:25:05 <ThomasWaldmann> pkumar: no, that pastebin contents should not crash
162 2011-08-19T20:25:37 *** greg_f
163 2011-08-19T20:25:45 <ThomasWaldmann> but if you kill the index after it created an empty index, it might get into trouble
164 2011-08-19T20:26:18 <ThomasWaldmann> no index != empty index
165 2011-08-19T20:29:32 <pkumar> ThomasWaldmann : at the staring of the tests _all_revisions_index_0 and _latest_revisions_index_0 were created on init backend. now I deleted them before the given test in pastebin and got the traceback like : http://paste.pocoo.org/show/460899/
166 2011-08-19T20:29:45 <pkumar> so thats not a bug right?
167 2011-08-19T20:31:24 <pkumar> shouldn't return SQLAlchemyBackend(verbose=True) in test
168 2011-08-19T20:31:48 <pkumar> _backend_sqla create similar empty files there?
169 2011-08-19T20:32:49 <pkumar> the index dir was emptied before the statement return SQLAlchemyBackend(verbose=True)
170 2011-08-19T20:33:43 <Marchael> ThomasWaldmann: I made some refactoring in test http://codereview.appspot.com/4923042
171 2011-08-19T20:33:51 <Marchael> I also removed rmtree
172 2011-08-19T20:36:06 *** ThomasWaldman2
173 2011-08-19T20:40:02 <ThomasWaldman2> pkumar: don't delete the files
174 2011-08-19T20:40:42 *** brunomartin
175 2011-08-19T20:40:43 *** ThomasWaldmann
176 2011-08-19T20:40:48 <pkumar> ThomasWaldman2 : k
177 2011-08-19T20:46:46 *** pkumar
178 2011-08-19T20:48:11 *** ThomasWaldman2
179 2011-08-19T20:51:13 <ThomasWaldmann> Marchael: see most recent bug in my tracker
180 2011-08-19T20:53:52 <Marchael> hm, and what about review?
181 2011-08-19T20:57:27 <ThomasWaldmann> give urls
182 2011-08-19T20:57:27 <Marchael> am
183 2011-08-19T20:57:27 <Marchael> http://codereview.appspot.com/4923042
184 2011-08-19T21:00:17 <ThomasWaldmann> that is functionally not equivalent to what you had before. explain why it is correct.
185 2011-08-19T21:01:58 <ThomasWaldmann> and aside from that, it is still at the wrong place. for same reason I gave before you moved it.
186 2011-08-19T21:02:50 <Marchael> hm, but you pointed me to create remove() in ItemIndex
187 2011-08-19T21:03:03 <Marchael> I do it, but just without rmtree
188 2011-08-19T21:03:27 <ThomasWaldmann> sorry, i likely confused ItemIndex with WhooshIndex
189 2011-08-19T21:03:51 <ThomasWaldmann> but you did not think about reasons, otherwise you'ld found the right place anyway
190 2011-08-19T21:04:00 <ThomasWaldmann> +have
191 2011-08-19T21:04:49 <Marchael> I thought that is rather strange to move remove() in ItemIndex becuase that class shouldn't manipulate index files
192 2011-08-19T21:19:22 <ThomasWaldmann> and about other strangeness: calling a method create_index in a method remove_index
193 2011-08-19T21:19:46 <Marchael> what strange?
194 2011-08-19T21:20:08 <Marchael> create_index just purge old index
195 2011-08-19T21:28:42 <ThomasWaldmann> you initially used rmtree, that removes all files with nothing left
196 2011-08-19T21:28:59 <ThomasWaldmann> create_index creates index files (with no contents)
197 2011-08-19T21:29:24 <ThomasWaldmann> so think about what makes sense
198 2011-08-19T21:29:27 <Marchael> mm, yes. But you told me to find another way(not rmtree) to clean index
199 2011-08-19T21:29:59 <ThomasWaldmann> no, i said you implemented that at wrong place (and that you redundantly call it twice most the time)
200 2011-08-19T21:30:35 <Marchael> I call it before and after test because I don't want to leave junk
201 2011-08-19T21:31:53 <ThomasWaldmann> yes, but you leave an empty index then
202 2011-08-19T21:32:27 <Marchael> hm, but is it bad? It's empty anyway
203 2011-08-19T21:32:57 <ThomasWaldmann> it's not much better than with 2 items in it
204 2011-08-19T21:33:08 <ThomasWaldmann> so basically pointless
205 2011-08-19T21:33:49 <Marchael> ok, then how I can remove indexes without rmtree?
206 2011-08-19T21:34:35 <ThomasWaldmann> (21:27) < ThomasWal> so think about what makes sense
207 2011-08-19T21:34:36 <ThomasWaldmann> (21:27) < Marchael> mm, yes. But you told me to find another way(not rmtree) to clean index
208 2011-08-19T21:34:38 <ThomasWaldmann> (21:28) < ThomasWal> no, i said you implemented that at wrong place (and that you redundantly call it twice most the time)
209 2011-08-19T21:36:23 <Marchael> ok
210 2011-08-19T21:36:51 *** brunomartin
211 2011-08-19T21:53:54 <Marchael> ThomasWaldmann: http://codereview.appspot.com/4923042 ps 2
212 2011-08-19T21:58:37 <ThomasWaldmann> now we find the architectural issues :)
213 2011-08-19T21:59:47 <Marchael> mm?
214 2011-08-19T22:00:46 <Marchael> oops
215 2011-08-19T22:10:59 <Marchael> ThomasWaldmann: ps 3
216 2011-08-19T22:14:33 * ThomasWaldmann has one :D
217 2011-08-19T22:16:22 <ThomasWaldmann> you realized what the problem with the previous one was?
218 2011-08-19T22:17:01 <Marchael> yeah, that stuff gives traceback
219 2011-08-19T22:17:08 <Marchael> :)
220 2011-08-19T22:17:26 <ThomasWaldmann> that might be a symptom, but what is the problem?
221 2011-08-19T22:17:46 <Marchael> this was incorrect because I tried remove index before writing doc there
222 2011-08-19T22:18:17 <ThomasWaldmann> and you maybe even killed files that are open?
223 2011-08-19T22:18:18 <Marchael> so traceback appears because directory doesn't exist
224 2011-08-19T22:18:37 <Marchael> of course
225 2011-08-19T22:18:42 <Marchael> It was stupid
226 2011-08-19T22:19:09 <ThomasWaldmann> so now you are doing the same thing, just later? :)
227 2011-08-19T22:20:02 <Marchael> hm
228 2011-08-19T22:20:35 <ThomasWaldmann> also, you are calling a method of an internal object
229 2011-08-19T22:21:02 <Marchael> ps 4
230 2011-08-19T22:21:45 <Marchael> so now I closing files indexes before removes them
231 2011-08-19T22:28:47 <dreimark> moin
232 2011-08-19T22:29:29 <Marchael> ThomasWaldmann: I know, that crap, but please tell something
233 2011-08-19T22:33:58 <dreimark> sinha1: sounds good
234 2011-08-19T22:34:58 <dreimark> can you have fixed both issues today or in some hours?
235 2011-08-19T22:35:22 <dreimark> we should merge soon
236 2011-08-19T22:38:07 <ThomasWaldmann> hi dreimark
237 2011-08-19T22:40:41 <dreimark> ThomasWaldmann: sinhas has to do a small improvement then the code is ready to merge
238 2011-08-19T22:40:51 <dreimark> documentation needs to be done and merged later
239 2011-08-19T22:41:23 <dreimark> I think we should do the code merge before he is ready with documentation
240 2011-08-19T22:41:55 <dreimark> s/ready/finished/
241 2011-08-19T22:44:00 <dreimark> ThomasWaldmann: https://github.com/ggozad/jarn.xmpp.collaboration
242 2011-08-19T22:44:11 <dreimark> Jarn hat für Plone eine Realtime Collaboration Erweiterung geschrieben
243 2011-08-19T22:44:11 <dreimark> die auf XMPP basiert.
244 2011-08-19T22:44:20 <dreimark> http://vimeo.com/22507185
245 2011-08-19T22:46:47 <ThomasWaldmann> jarn.xmpp.collaboration relies on the wonderful Diff-Match-Patch from Neil Fraser at Google. It is distributed under the Apache License 2.0.
246 2011-08-19T22:47:04 <ThomasWaldmann> == mobwrite author
247 2011-08-19T22:48:06 <ThomasWaldmann> + zope2 stuff
248 2011-08-19T22:49:18 <dreimark> i know, we need a different solution but this one has gpl license in setup.py
249 2011-08-19T22:54:25 <ThomasWaldmann> it uses same AL2 stuff
250 2011-08-19T22:55:20 <dreimark> ok, then it did not help
251 2011-08-19T23:08:53 *** birkenfeld
252 2011-08-19T23:14:22 <Marchael> ThomasWaldmann: http://codereview.appspot.com/4923042/ ps 5
253 2011-08-19T23:14:37 <Marchael> It's unclena, but I'll make split later
254 2011-08-19T23:14:43 <Marchael> *unclean
255 2011-08-19T23:25:17 <ThomasWaldmann> done
256 2011-08-19T23:26:03 <Marchael> ok
257 2011-08-19T23:26:47 <Marchael> ThomasWaldmann: but on bb you pointed me to use > itemlinks:item_name OR itemtransclusions:item_name
258 2011-08-19T23:27:16 <Marchael> if I understand right direct access means using searcher.documents() for that
259 2011-08-19T23:27:31 <Marchael> but in that case we will search for a AND b
260 2011-08-19T23:29:39 <ThomasWaldmann> i didn't say that
261 2011-08-19T23:29:57 <ThomasWaldmann> i said you directly construct a query, not parse one from a string
262 2011-08-19T23:30:08 <Marchael> ok
263 2011-08-19T23:41:52 *** raignarok
264 2011-08-19T23:42:10 <Marchael> ThomasWaldmann: http://codereview.appspot.com/4923042/ ps 6
265 2011-08-19T23:44:12 <ThomasWaldmann> Marchael: yeah
266 2011-08-19T23:44:55 <Marchael> ok, so what will we do? Split this ps to 2 commits and close issues?
267 2011-08-19T23:45:15 <Marchael> #23 and #34
268 2011-08-19T23:45:43 <Marchael> or I could done something for #34?
269 2011-08-19T23:52:02 <Marchael> ThomasWaldmann: ^
