1 2011-12-06T00:11:46 <ThomasWaldmann> hi yufra
2 2011-12-06T00:12:07 <yufra> hello
3 2011-12-06T00:12:31 <ThomasWaldmann> we recently did a rewrite of the storage
4 2011-12-06T00:13:20 <ThomasWaldmann> so all is much simpler now, but no dvcs
5 2011-12-06T00:14:27 <yufra> What sort of confidence do you have that storage is now stable? And by stable I mean "not going to change much".
6 2011-12-06T00:14:51 <ThomasWaldmann> rather high. it nice and simple now.
7 2011-12-06T00:15:30 <yufra> great! I'll have a look at the code over the next week and probably stop by with questions periodically then
8 2011-12-06T00:15:59 <yufra> do you have any preferences/requests in regards to DVCS?
9 2011-12-06T00:16:27 <yufra> I am simply interested in easy synchronization of wikis, so the particular DVCS isn't too important to me
10 2011-12-06T00:16:35 <yufra> I have personal experience with Mercurial and Git, thoguh
11 2011-12-06T00:17:13 <ThomasWaldmann> the sync task maybe does not require new backend.
12 2011-12-06T00:17:40 <yufra> I have been away too long... sync task?
13 2011-12-06T00:17:48 <ThomasWaldmann> one reason why we did not recreate the hg backend was because we ( ronny and me ) felt somehow it was an abuse
14 2011-12-06T00:18:23 <ThomasWaldmann> the current storage revisions have revids, parent[s]
15 2011-12-06T00:18:23 <yufra> care explaining further?
16 2011-12-06T00:18:40 <yufra> right, so basically a VCS already
17 2011-12-06T00:18:51 <ThomasWaldmann> so if one compares what revids are "here" and which are "there", you see what's missing
18 2011-12-06T00:19:09 <ronny> a normal versioned filesystem is simply unfit for a typical wiki
19 2011-12-06T00:19:44 <ronny> since a wiki works in terms of pages, not in terms of atomic filesystem revisions
20 2011-12-06T00:19:52 <yufra> ronny, I won't agree or disagree since I don't do a lot of wiki coding (mostly wiki using)
21 2011-12-06T00:19:57 <ronny> it neither scales well, nor gives a good way to store the metadata
22 2011-12-06T00:20:10 <yufra> my desire was simply easy synchronization, but if there is an alternative built-in now then that is moot :-)
23 2011-12-06T00:21:36 <ThomasWaldmann> yufra: it would be very welcome if you tried working on sync (nobody else does yet). it is not a basic feature, but it would be nice to have at some time.
24 2011-12-06T00:22:30 <ThomasWaldmann> there is currently some sort of index/storage sync ("index update") going on internally when you update the index
25 2011-12-06T00:22:39 <yufra> sure, I don't mind taking a crack at it. Can you point me in the right direction to start, and what its current status is?
26 2011-12-06T00:22:42 <ThomasWaldmann> it checks storage contents vs. index contents there
27 2011-12-06T00:23:03 <ThomasWaldmann> maybe it could do the basic work for 2 wikis in a similar way
28 2011-12-06T00:23:33 <yufra> how do you envision a wiki sync working?
29 2011-12-06T00:23:38 <yufra> from a user perspective
30 2011-12-06T00:24:36 <ThomasWaldmann> well, there might be 2 heads for some items after pull
31 2011-12-06T00:24:50 *** raignarok
32 2011-12-06T00:25:29 <ThomasWaldmann> for some stuff it might be possible to merge automatically (simple text items), but for others a user will have to do it manually
33 2011-12-06T00:25:48 <yufra> sure, but even the pull operation... how do you see that working?
34 2011-12-06T00:26:22 <ThomasWaldmann> yeah, we need some sort of rpc for that, we don't have that yet
35 2011-12-06T00:26:24 <yufra> For example, the usage case I am thinking of is having a wiki on my laptop for access during travel, and then another wiki on my desktop for when I am at work
36 2011-12-06T00:26:38 <ThomasWaldmann> 1.9 use xmlrpc, for 2 maybe rather jsonrpc
37 2011-12-06T00:27:14 <yufra> gotcha
38 2011-12-06T00:27:16 *** qxcv
39 2011-12-06T00:28:01 <yufra> Well this is an important feature for me so I will try to first familiarize myself with the new storage code
40 2011-12-06T00:28:17 <ThomasWaldmann> wait a sec please
41 2011-12-06T00:28:21 <yufra> sure
42 2011-12-06T00:30:13 <ThomasWaldmann> docs/devel/development.rst has a short section about storage
43 2011-12-06T00:30:36 <ThomasWaldmann> also there are some docstrings in storage/* that are useful
44 2011-12-06T00:31:12 <ThomasWaldmann> the most power comes from indexing module, everything else is rather simple
45 2011-12-06T00:32:16 <yufra> and indexing is in the middleware, right?
46 2011-12-06T00:33:04 <yufra> ah, reading the indexing docstring now... I am getting the basic idea
47 2011-12-06T00:33:05 <ThomasWaldmann> yes
48 2011-12-06T00:37:41 <yufra> I'll stay in the channel and check back in a day or two
49 2011-12-06T00:38:15 <ThomasWaldmann> ok.
50 2011-12-06T00:38:43 <yufra> oh, actually... what is the difference between stores and backends?
51 2011-12-06T00:38:44 <ThomasWaldmann> btw, sync in moin2 will be for all types of items, automatically (1.9 was just pages, not attachments)
52 2011-12-06T00:39:07 <ThomasWaldmann> backend is talking to a real storage and does simply key/value
53 2011-12-06T00:39:12 <ThomasWaldmann> ehrm
54 2011-12-06T00:39:15 <yufra> yup, I like the change to everything is treated as an item
55 2011-12-06T00:39:16 <ThomasWaldmann> s/backend/store
56 2011-12-06T00:39:40 <ThomasWaldmann> a backend is on top of 2 stores, one for metadata one for data
57 2011-12-06T00:39:50 <yufra> ok
58 2011-12-06T00:40:01 <ThomasWaldmann> alternatively, there is this fileserver backend
59 2011-12-06T00:40:21 <ThomasWaldmann> which just uses files in the host filesystem. and makes up metadata as good as it can.
60 2011-12-06T00:41:41 <yufra> so stores is the default backend?
61 2011-12-06T00:42:28 <yufra> or at least suggested backend>?
62 2011-12-06T00:42:53 <ThomasWaldmann> yes, fileserver is readonly
63 2011-12-06T00:43:11 <ThomasWaldmann> it is just for "ftp like" application
64 2011-12-06T00:43:57 <ThomasWaldmann> and you see, stores is also very simple
65 2011-12-06T00:44:20 <yufra> I do indeed, it looks quite nice
66 2011-12-06T00:44:22 <ThomasWaldmann> it just binds together the data and the meta for one revision using meta[DATAID]
67 2011-12-06T00:47:36 <ronny> ThomasWaldmann: i think the data storace c an basically be 1:1 synced, the metadata store needs some extra logic for purges
68 2011-12-06T00:48:30 <ThomasWaldmann> purges?
69 2011-12-06T00:48:42 <ronny> ThomasWaldmann: when the content of a rev is killed
70 2011-12-06T00:49:36 <ronny> to be n a better posiion it probably makes sense to find all metadata items one needs to transfer, then collect the assigned data items
71 2011-12-06T00:51:25 <ThomasWaldmann> yup
72 2011-12-06T00:52:10 <ThomasWaldmann> i guess the most difficult won't be the pull, but the merge ui / workflow
73 2011-12-06T00:59:03 <ThomasWaldmann> yufra: btw, just that you have the full picture: there is currently a "namespaces" branch in the repo
74 2011-12-06T00:59:28 <ThomasWaldmann> it still needs more work before merge, but what it'll give us is:
75 2011-12-06T00:59:37 <ThomasWaldmann> multiple names for a revision
76 2011-12-06T00:59:52 <ThomasWaldmann> namespaces instead of fstab-like mounting
77 2011-12-06T01:00:08 <yufra> oh! ok I can see how that could be useful
78 2011-12-06T01:00:17 <ThomasWaldmann> new sight on delete, rename, alias
79 2011-12-06T07:51:08 <dreimark> moin
80 2011-12-06T10:17:29 *** MattMaker
81 2011-12-06T10:18:20 *** greg_f
82 2011-12-06T10:27:02 *** MattMaker
83 2011-12-06T10:43:55 *** mkerrin
84 2011-12-06T10:44:23 *** mkerrin
85 2011-12-06T10:49:48 *** mkerrin
86 2011-12-06T10:58:30 <ThomasWaldmann> moin
87 2011-12-06T11:44:15 <ThomasWaldmann> qxcv: are you still working on the subitems task?
88 2011-12-06T11:46:42 <ThomasWaldmann> qxcv: btw, commit comments as "Fix some CSS" are rather useless (rather describe what you fixed)
89 2011-12-06T11:47:46 * ThomasWaldmann reviews qxcv's changesets
90 2011-12-06T11:55:16 <qxcv> ThomasWaldmann: I'm waiting for it to be reviewed atm
91 2011-12-06T12:03:38 <ThomasWaldmann> #moin-page:after { content: " "; visibility: hidden; clear: both; display: block; height: 0; }
92 2011-12-06T12:03:49 <ThomasWaldmann> content?
93 2011-12-06T12:04:15 <qxcv> sorry, I should remove that - it was for the inline sidebar
94 2011-12-06T12:07:51 *** mkerrin
95 2011-12-06T12:09:25 <qxcv> ThomasWaldmann: are you sure you're reading the current tip in my moin fork?
96 2011-12-06T12:09:30 <qxcv> that css shouldn't be in there
97 2011-12-06T12:10:32 <ThomasWaldmann> qxcv: https://bitbucket.org/qxcv/moin-2.0/changeset/305fd62da8bf 550+ looks strange, see that }
98 2011-12-06T12:10:48 <ThomasWaldmann> i am reading diffs
99 2011-12-06T12:10:56 <qxcv> oic
100 2011-12-06T12:12:05 <qxcv> yes, that was a css rule that had been put across multiple lines at the same indent level with a comment in between
101 2011-12-06T12:12:14 <qxcv> noticed that a few minutes ago and fixed it up :)
102 2011-12-06T12:13:09 <ThomasWaldmann> btw, when using css tricks, a comment with a url that describes the trick would be nice
103 2011-12-06T12:14:58 <ThomasWaldmann> does giving the width as 140px make sense? shouldn't that be em or %?
104 2011-12-06T12:15:43 <qxcv> I just used px because that's what everything else uses
105 2011-12-06T12:16:39 <qxcv> also em is problematic because it changes rather unpredictably due to typefaces
106 2011-12-06T12:17:09 <qxcv> I can use em or % if you want, though
107 2011-12-06T12:18:58 <ThomasWaldmann> well, em is expected to change with typefaces :)
108 2011-12-06T12:19:34 <ThomasWaldmann> but when saying that a sidebar should be n pixels wide, you maybe rather expect that it should be able to display a specific amount of text
109 2011-12-06T12:19:56 <qxcv> that's a good point, I'll change it now
110 2011-12-06T12:20:28 <ThomasWaldmann> (that's different from saying "o i just want a little border", where maybe 2px is also fine)
111 2011-12-06T12:20:59 <ThomasWaldmann> btw, take all what I say about css with a grain of salt, i am no css expert.
112 2011-12-06T12:34:59 * ThomasWaldmann runs qxcv's stuff
113 2011-12-06T12:39:46 <ThomasWaldmann> qxcv: what do you think about that: instead of writing "ThisItem /" above it (which is not very useful, except that it somehow [together with the indentation of items below it] indicates that we are seeing subitems)
114 2011-12-06T12:51:10 <ThomasWaldmann> instead of that, just writing "Subitems" (like a headline) and not using that | SubItemName but just Itemname?
115 2011-12-06T12:51:34 <ThomasWaldmann> ehrm, just the SubItemNames I mean
116 2011-12-06T12:52:18 <ThomasWaldmann> that would save a little width (which i noticed is too small for my test subitem names)
117 2011-12-06T12:52:42 <qxcv> okay, that's probably more helpful than just the item name
118 2011-12-06T12:53:18 <ThomasWaldmann> also make it a little more wide maybe
119 2011-12-06T12:53:32 <ThomasWaldmann> there can't be automatic width adjustment, right?
120 2011-12-06T12:53:53 <qxcv> as in the user gets to adjust it?
121 2011-12-06T12:54:16 <qxcv> oh, as in it scales to the width of the text?
122 2011-12-06T12:55:14 <ThomasWaldmann> the latter
123 2011-12-06T12:56:43 <qxcv> yes, but it can look a bit funny when you have big differences in width
124 2011-12-06T12:57:12 <qxcv> perhaps I'll give it a minimum and a maximum width, and let it move in between the two automatically?
125 2011-12-06T12:57:23 <qxcv> so huge item names don't break things on small screens
126 2011-12-06T12:57:27 <ThomasWaldmann> hm, right
127 2011-12-06T12:57:33 <ThomasWaldmann> how would that work?
128 2011-12-06T12:59:21 <qxcv> the sidebar will always be at least x-units wide, or at most y-units wide
129 2011-12-06T12:59:42 <qxcv> and the browser will choose a value between the two depending on the width of the text
130 2011-12-06T13:02:07 <qxcv> it would probably be easier just to make the bar a constant width relative to the text, turn on word wrapping and add a margin above and below each element to separate them
131 2011-12-06T13:07:32 <ThomasWaldmann> maybe, can you try?
132 2011-12-06T13:12:52 <qxcv> here's what the word wrapping looks like: http://ompldr.org/vYm0xMQ (slightly larger sidebar there, too)
133 2011-12-06T13:16:43 <qxcv> here's what it looks like when resized automatically up to a max width of 30em: http://ompldr.org/vYm0xMg
134 2011-12-06T13:18:19 <qxcv> a matter of personal taste, I guess
135 2011-12-06T13:22:09 * ThomasWaldmann looks
136 2011-12-06T13:22:22 <ThomasWaldmann> first one :)
137 2011-12-06T13:23:40 <qxcv> sweet, I'll commit that and change the header to read "subitems"
138 2011-12-06T13:27:10 <qxcv> done, https://bitbucket.org/qxcv/moin-2.0/changeset/35d2c958b4e5
139 2011-12-06T13:30:22 <ThomasWaldmann> great :)
140 2011-12-06T13:30:57 <ThomasWaldmann> qxcv: you want to work on the followup now?
141 2011-12-06T13:31:33 <qxcv> ok, sounds good
142 2011-12-06T13:31:53 <ThomasWaldmann> there is a task with similar title for modify view
143 2011-12-06T13:32:52 <ThomasWaldmann> i guess we'll have to work on the general design/layout anyway later
144 2011-12-06T13:33:38 *** Sysi
145 2011-12-06T13:34:05 <qxcv> will moin-2 use the modernised theme when it is released?
146 2011-12-06T13:37:17 *** dave_largo
147 2011-12-06T13:38:45 <CIA-59> Sam Toyer <samATqxcvDOTnet> default * 1123:ebc4f8213e74 2.0/MoinMoin/ (4 files in 3 dirs): Add a flat subitem index for show item view.
148 2011-12-06T13:38:45 <CIA-59> Sam Toyer <samATqxcvDOTnet> default * 1124:f106f50a5d3c 2.0/MoinMoin/templates/show.html: Make the subitem bar only show when there are subitems
149 2011-12-06T13:38:46 <CIA-59> Sam Toyer <samATqxcvDOTnet> default * 1125:b185c8303215 2.0/docs/devel/ (support.rst translate.rst): Merge in moin2 repository
150 2011-12-06T13:38:47 <CIA-59> Sam Toyer <samATqxcvDOTnet> default * 1126:211e5d99c842 2.0/MoinMoin/ (templates/show.html themes/__init__.py): Remove the location breadcrumbs, make the subitem menu patch a bit more clear.
151 2011-12-06T13:38:47 <CIA-59> Sam Toyer <samATqxcvDOTnet> default * 1127:d8f41116f3e2 2.0/MoinMoin/themes/modernized/static/css/common.css: Fix some CSS
152 2011-12-06T13:38:48 <CIA-59> Sam Toyer <samATqxcvDOTnet> default * 1128:aec416a9ec05 2.0/MoinMoin/ (2 files in 2 dirs): Move the bar to the bottom
153 2011-12-06T13:38:49 <CIA-59> Sam Toyer <samATqxcvDOTnet> default * 1129:305fd62da8bf 2.0/MoinMoin/ (3 files in 2 dirs): Move subitems to a sidebar.
154 2011-12-06T13:38:50 <CIA-59> Sam Toyer <samATqxcvDOTnet> default * 1130:e92a157518e2 2.0/MoinMoin/ (2 files in 2 dirs): Move into a sidebar using CSS2.1 table display trickery.
155 2011-12-06T13:38:50 <CIA-59> Sam Toyer <samATqxcvDOTnet> default * 1131:13c89a119841 2.0/MoinMoin/themes/modernized/static/css/common.css: Fix sidebar width issues (table-layout)
156 2011-12-06T13:38:52 <CIA-59> Sam Toyer <samATqxcvDOTnet> default * 1132:4f360308c740 2.0/MoinMoin/themes/modernized/static/css/common.css: Size tweaks
157 2011-12-06T13:38:52 <CIA-59> Sam Toyer <samATqxcvDOTnet> default * 1133:7f3c15bb0858 2.0/MoinMoin/themes/modernized/static/css/common.css: Fix a typo in one of the common.css blocks
158 2011-12-06T13:38:52 <CIA-59> Sam Toyer <samATqxcvDOTnet> default * 1134:f57df5ee7a61 2.0/MoinMoin/themes/modernized/static/css/common.css: s/px/em for sidebar width
159 2011-12-06T13:38:53 <CIA-59> Sam Toyer <samATqxcvDOTnet> default * 1135:35d2c958b4e5 2.0/MoinMoin/ (templates/show.html themes/modernized/static/css/common.css): Subitem bar: change header to "subitems", wrap links
160 2011-12-06T13:38:54 <CIA-59> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 1136:59cb25f4ef37 2.0/MoinMoin/ (4 files in 3 dirs): merged
161 2011-12-06T13:40:34 <ThomasWaldmann> qxcv: well, currently there is nothing else
162 2011-12-06T13:40:58 <ThomasWaldmann> now and then i think about a sidebar theme and i recently did some experiments
163 2011-12-06T13:41:12 <ThomasWaldmann> but I always feel I am not the right one to do such stuff :)
164 2011-12-06T13:42:11 <ThomasWaldmann> maybe we should have 2 themes: one similar to the one in moin 1.x (modernized), one with sidebar(s)
165 2011-12-06T13:43:03 <qxcv> I'm guessing that it's hard to port moin-1 themes to moin-2?
166 2011-12-06T13:49:23 <ThomasWaldmann> is like a rewrite
167 2011-12-06T13:49:45 <ThomasWaldmann> but maybe we should have a clean start there also
168 2011-12-06T13:59:29 *** dave_largo
169 2011-12-06T14:01:42 <ThomasWaldmann> qxcv: btw, the task says "link/transclude markup", but for the first iteration I'ld suggest just inserting the subitem name
170 2011-12-06T14:02:10 <ThomasWaldmann> also, try to reuse / not duplicate code
171 2011-12-06T14:02:50 <ThomasWaldmann> (moin support many different markups)
172 2011-12-06T14:03:14 <qxcv> is ckeditor support required?
173 2011-12-06T14:03:42 <ThomasWaldmann> not initially
174 2011-12-06T14:04:20 <ThomasWaldmann> for js please use jquery where appropriate
175 2011-12-06T14:04:48 <qxcv> ok
176 2011-12-06T14:41:58 *** qxcv
177 2011-12-06T14:52:19 *** dave_largo
178 2011-12-06T15:39:42 <ThomasWaldmann> dreimark: http://www.google-melange.com/gci/task/view/google/gci2011/7172302
179 2011-12-06T15:49:30 <dreimark> fsck, now it is good grammar but somehow broken. I should have tried to apply it.
180 2011-12-06T15:50:10 <dreimark> and I missed for the reviews cr already. I try at home too to get it imported.
181 2011-12-06T15:52:57 <ThomasWaldmann> dreimark: well, as i said, there is no conflict (AFAIK) from our side, so jordon should fix that
182 2011-12-06T16:00:20 *** dave_largo
183 2011-12-06T16:04:07 *** RogerHaase
184 2011-12-06T16:04:27 *** MattMaker
185 2011-12-06T16:36:09 *** mkerrin
186 2011-12-06T17:27:12 *** cdbs
187 2011-12-06T18:05:57 * ThomasWaldmann check's DJ_DENT's docs
188 2011-12-06T18:06:11 <ThomasWaldmann> checks* even
189 2011-12-06T18:11:22 *** dave_largo
190 2011-12-06T18:12:36 <CIA-59> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 1137:9b5e99f59e7b 2.0/docs/admin/configure.rst: docs: proofread configure.rst, thanks to DJ_DENT
191 2011-12-06T18:30:56 *** MattMaker
192 2011-12-06T18:50:17 *** mkerrin
193 2011-12-06T18:56:47 *** greg_f
194 2011-12-06T19:54:49 *** grzywacz
195 2011-12-06T19:54:49 *** grzywacz
196 2011-12-06T19:56:06 <dreimark> ThomasWaldmann: http://moinmo.in/MoinMoinBugs/TableOfContentsWithOtherDefaultMarkup ok, for 1.9?
197 2011-12-06T20:09:10 *** raignarok
198 2011-12-06T20:22:00 <dreimark> and I need the format info for our wiki2sphinx tools http://paste.pocoo.org/show/517192/
199 2011-12-06T20:37:41 *** RogerHaase
200 2011-12-06T20:50:40 <dreimark> bbl
201 2011-12-06T20:51:29 *** RogerHaase
202 2011-12-06T20:58:26 *** bgola
203 2011-12-06T21:30:44 <ThomasWaldmann> moin
204 2011-12-06T21:54:15 <ThomasWaldmann> dreimark: you tried it with the supported markups?
205 2011-12-06T22:01:49 *** yufra_
206 2011-12-06T22:18:25 *** dave_largo
207 2011-12-06T22:26:35 *** raignarok_
208 2011-12-06T22:26:36 *** raignarok
209 2011-12-06T22:32:44 *** raignarok_
210 2011-12-06T22:56:45 *** yufra_
211 2011-12-06T22:56:57 *** yufra_
212 2011-12-06T23:08:45 <dreimark> ThomasWaldmann: yes rst and wiki
213 2011-12-06T23:09:09 <dreimark> i need that because we convert everything which is not rst to pdf figures
214 2011-12-06T23:35:22 *** RogerHaase
215
