1 2010-01-09T00:12:51 *** fucheeno
2 2010-01-09T02:40:14 <CIA-44> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 4446:0dd02729b3b5 1.8/MoinMoin/util/SubProcess.py: SubProcess: fix imports, cosmetic other changes
3 2010-01-09T02:41:31 <CIA-44> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 5407:a88ddb0f486b 1.9/MoinMoin/util/SubProcess.py: merged moin/1.8
4 2010-01-09T07:01:42 *** fucheeno
5 2010-01-09T07:03:48 *** fucheeno
6 2010-01-09T07:32:37 *** boca
7 2010-01-09T07:33:08 *** boca
8 2010-01-09T10:39:49 *** eisi
9 2010-01-09T11:46:42 <dreimark> ThomasWaldmann: txt2tags uses extension moin for moin markup
10 2010-01-09T12:22:52 *** eSyr
11 2010-01-09T12:27:45 <ThomasWaldmann> moin
12 2010-01-09T12:28:01 <ThomasWaldmann> dreimark: ok, then we use .moin also
13 2010-01-09T12:28:16 <eSyr> moin
14 2010-01-09T12:28:49 * eSyr is trying to add some information about revisions to diff output, like mw does
15 2010-01-09T12:33:00 <dreimark> eSyr: moin 2.0 ?
16 2010-01-09T12:33:44 <dreimark> btw. txt2tags isn't working very well i filed a bug report
17 2010-01-09T12:33:58 <eSyr> dreimark, no, just information about rev timestamp/author/etc
18 2010-01-09T12:34:17 <dreimark> ok
19 2010-01-09T12:35:14 <eSyr> (because it is rather unconmfortable to swich between info and diff to check out rev timestamp and so)
20 2010-01-09T12:55:50 *** grzywacz
21 2010-01-09T13:04:12 *** johill
22 2010-01-09T13:05:07 *** johill
23 2010-01-09T13:17:02 <ThomasWaldmann> http://moinmo.in/MoinMoinBugs/1.9CookiePathHandlingChanged < comments?
24 2010-01-09T13:28:40 <eSyr> ThomasWaldmann, i've just got 500 on trying to preview page on master19
25 2010-01-09T13:28:51 <eSyr> page SystemPagesInEnglishGroup
26 2010-01-09T13:29:59 * ThomasWaldmann looks
27 2010-01-09T13:30:15 <ThomasWaldmann> eSyr: btw, did you get some jabber msg from me?
28 2010-01-09T13:30:23 <eSyr> hm...
29 2010-01-09T13:30:28 <eSyr> i think, no.
30 2010-01-09T13:30:37 <eSyr> to which jid have you sent it?
31 2010-01-09T13:30:48 <ThomasWaldmann> esyr at jabber.org
32 2010-01-09T13:31:48 <eSyr> oh, yes, i've go it.
33 2010-01-09T13:47:25 <ThomasWaldmann> eSyr: SPIEG page on master19 works for me
34 2010-01-09T13:48:18 <eSyr> i've tried to add deprecation pi and preview it.
35 2010-01-09T13:53:32 * ThomasWaldmann tries
36 2010-01-09T13:59:40 <ThomasWaldmann> ok, works for me
37 2010-01-09T14:00:19 <ThomasWaldmann> hmm, edit/save works, but preview does not
38 2010-01-09T14:01:47 <ThomasWaldmann> eSyr: http://paste.pocoo.org/show/163628/
39 2010-01-09T14:02:32 <eSyr> hehe, i've just got the same error while trying to patch diff action.
40 2010-01-09T14:02:54 <eSyr> i think, it because of parsing of deprecate pi.
41 2010-01-09T14:07:40 <ThomasWaldmann> it is because it tries to use add_msg at a time when the msgs are already emitted
42 2010-01-09T14:08:45 <eSyr> aha.
43 2010-01-09T14:11:59 <ThomasWaldmann> looks like all add_msg calls should be reviewed
44 2010-01-09T14:13:10 <eSyr> ThomasWaldmann, http://paste.pocoo.org/show/163632/ — it's about diff action additional info
45 2010-01-09T14:30:14 <ThomasWaldmann> are those new i18n strings? did you try to reuse old ones?
46 2010-01-09T14:32:41 <ThomasWaldmann> and if you are emitting raw html, you have to be careful to escape stuff, so no xss is possible
47 2010-01-09T14:34:07 <ThomasWaldmann> eSyr:
48 2010-01-09T14:35:10 <eSyr> hm, strings are mostly borrowed from info actions.
49 2010-01-09T14:35:20 <eSyr> a'll try to reuse axisting strings.
50 2010-01-09T14:35:25 <eSyr> *i'll
51 2010-01-09T14:47:03 <eSyr> well, i've found all i18n strings, except for revision number.
52 2010-01-09T14:49:48 <eSyr> but i'm not sure that rev num is important, because it's already mentioned.
53 2010-01-09T14:53:50 <eSyr> ThomasWaldmann, http://paste.pocoo.org/show/163645/
54 2010-01-09T14:58:46 <eSyr> can i push it?
55 2010-01-09T14:59:29 <dreimark> any idea for an easier way to get a valid session id for editmoin than
56 2010-01-09T14:59:32 <dreimark> http://paste.pocoo.org/show/163648/
57 2010-01-09T15:02:42 <dreimark> eSyr: is rev1_comment already escaped?
58 2010-01-09T15:03:08 <dreimark> same question for rev2_comment
59 2010-01-09T15:03:32 * dreimark should read the whole patch
60 2010-01-09T15:03:58 <dreimark> see it now
61 2010-01-09T15:05:58 *** eisi
62 2010-01-09T15:16:00 <eSyr> well, i think no, because in info action wikiutil.escape(comment) is used, so i've used the same
63 2010-01-09T15:21:01 <dreimark> eSyr: my fault i haven't read it completly
64 2010-01-09T15:32:32 <dreimark> ok, solved editmoin problem
65 2010-01-09T16:17:07 <eSyr> hm, may be we should just ignore add_msg calls after send_title?
66 2010-01-09T16:35:58 <ThomasWaldmann> eSyr: check if author name needs escaping
67 2010-01-09T16:37:29 <eSyr> ThomasWaldmann, it generates sanitized code (with title for nonregistered users and link to homepage for registered)
68 2010-01-09T16:39:18 <eSyr> ThomasWaldmann, also, about deprecation page faults: http://www.moinmo.in/MoinMoinBugs/InternalServerErrorOnActionCallingWithContentOfDeprecatedPage
69 2010-01-09T16:46:25 <eSyr> ThomasWaldmann, http://paste.pocoo.org/show/163670/ (some minor changes to i18n strings)
70 2010-01-09T16:49:38 <ThomasWaldmann> eSyr: code looks ok
71 2010-01-09T16:49:56 <ThomasWaldmann> how is presentation, do you have a demo somewhere?
72 2010-01-09T16:50:02 <eSyr> hm...
73 2010-01-09T16:52:10 <ThomasWaldmann> what I am asking myself is whether the left and right part of that table is as broad as the respective content below
74 2010-01-09T16:52:15 <eSyr> http://esyr.org:8080/ — it is sshtunneled to my local moin-hg instance
75 2010-01-09T16:55:20 <ThomasWaldmann> http://esyr.org:8080/lol?action=diff&rev2=4&rev1=3
76 2010-01-09T16:55:55 <eSyr> yes, it is.
77 2010-01-09T16:56:15 <ThomasWaldmann> maybe it would be prettier if it was a single table, but that's maybe hard to implement with the other code
78 2010-01-09T16:57:40 <ThomasWaldmann> http://esyr.org:8080/lol?action=diff&rev2=5&rev1=4 (no it is not :)
79 2010-01-09T16:58:09 <eSyr> well, i think, the other solution is to add css rule like table.diff > tr > td:first-child { width:50%; }
80 2010-01-09T17:01:32 <ThomasWaldmann> what does that do?
81 2010-01-09T17:02:38 <ThomasWaldmann> (what I mean is that same alignment / common middle border for both left and right column rows would be nice - for info header AND content diff display)
82 2010-01-09T17:03:57 <eSyr> yes, i understoo.
83 2010-01-09T17:04:16 <ThomasWaldmann> an easy way to achieve might be to sneak that metadata into the content string given to diff_html.diff :D
84 2010-01-09T17:04:57 <ThomasWaldmann> of course it would highlight changes there then also X)
85 2010-01-09T17:07:00 <eSyr> afk, bbl.
86 2010-01-09T17:38:30 * dreimark has submitted a patch to editmoin for 1.8 and 1.9
87 2010-01-09T17:45:48 <ThomasWaldmann> is 1.8 different to 1.7?
88 2010-01-09T17:46:09 <dreimark> i have not tested 1.7
89 2010-01-09T17:46:21 * dreimark looks
90 2010-01-09T17:46:40 <ThomasWaldmann> so what was the last version support by editmoin before your patch?
91 2010-01-09T17:46:49 <dreimark> 1.6
92 2010-01-09T17:47:04 <dreimark> but without session authentication
93 2010-01-09T17:49:24 <dreimark> 1,7 works too
94 2010-01-09T19:01:39 <ThomasWaldmann> btw, mitsuhiko on #pocoo suggested we do always use path=/ in our cookies and differentiate them via the cookie name
95 2010-01-09T19:02:22 <ThomasWaldmann> while that solves the kind of "cookie confusion" we currently have sometimes,
96 2010-01-09T19:02:52 <ThomasWaldmann> we have to be careful not to introduce other confusions by name collisions
97 2010-01-09T19:04:03 *** eSyr
98 2010-01-09T19:15:33 *** eSyr
99 2010-01-09T19:20:04 <dreimark> then we might also differentiate the used port
100 2010-01-09T19:32:47 <ThomasWaldmann> how does that work right now?
101 2010-01-09T19:38:45 <dreimark> usuall i have to kill the cookies
102 2010-01-09T19:39:32 <dreimark> bbl
103 2010-01-09T19:43:20 <ThomasWaldmann> does that mean that running multiple moins at same domain/path, but different ports does not work currently?
104 2010-01-09T19:43:48 <ThomasWaldmann> (if you do not want them to share the session)
105 2010-01-09T19:51:27 <dreimark> ThomasWaldmann: hmm, i check this.
106 2010-01-09T19:54:43 <dreimark> ThomasWaldmann: moin-1.9$ MoinMoin/script/moin.py server standalone --port=8082
107 2010-01-09T19:55:10 <dreimark> abd in a different shell moin-1.9$ MoinMoin/script/moin.py server standalone --po081
108 2010-01-09T19:56:06 <dreimark> you share the session cookie
109 2010-01-09T19:57:10 * dreimark is logged in in both
110 2010-01-09T19:57:33 <dreimark> and has created the cookie for port 8080
111 2010-01-09T19:59:33 <dreimark> logging out from port 8081 ends the session on 8082 too
112 2010-01-09T20:00:46 <dreimark> s/--po081/--port8081
113 2010-01-09T20:08:01 <ThomasWaldmann> ok
114 2010-01-09T21:08:30 <dreimark> always if I have a valid session id on 8080 / url I am not able to get a session on 80:/url for more than one request
115 2010-01-09T21:09:41 <dreimark> if i log out on the 8080 / wiki I am logged in into the 80:/url1 80:/url2 by the next request
116 2010-01-09T21:10:39 <ThomasWaldmann> i have a patch to test soon
117 2010-01-09T21:11:08 <dreimark> :)
118 2010-01-09T21:56:50 <ThomasWaldmann> http://pastebin.com/f31beadab < please review/comment
119 2010-01-09T21:57:25 <ronny> ThomasWaldmann: oh man, please use paste.pocoo.org :P
120 2010-01-09T21:57:48 <ThomasWaldmann> i usually do, except when using pastebinit :)
121 2010-01-09T21:57:49 <ronny> or even better, the hg paste extension by georg
122 2010-01-09T22:00:57 <TheSheep> hey guys, I made a visualisation of the repo history: http://www.sheep.art.pl/misc/moinmoin.ogv
123 2010-01-09T22:01:48 <dreimark> ronny: please add a hint at http://moinmo.in/MoinDev
124 2010-01-09T22:01:58 <ThomasWaldmann> we need some task for radomir, he seems to be bored :D
125 2010-01-09T22:02:07 * TheSheep runs
126 2010-01-09T22:03:30 <ronny> ThomasWaldmann: what tool did you use?
127 2010-01-09T22:03:37 <ronny> eh TheSheep
128 2010-01-09T22:04:05 <TheSheep> ronny: gource
129 2010-01-09T22:04:10 <ThomasWaldmann> crazy stuff :)
130 2010-01-09T22:04:24 * TheSheep watches little men pissing at files
131 2010-01-09T22:04:41 <ronny> dreimark: where exactly should i add that note
132 2010-01-09T22:05:02 <dreimark> i should have called it
133 2010-01-09T22:05:19 <dreimark> from the big screen
134 2010-01-09T22:06:19 <ronny> dreimark: i suppose the mercurial guide will be a good place
135 2010-01-09T22:06:23 <dreimark> ronny DeveloperTools
136 2010-01-09T22:06:31 <dreimark> yes or mercurial
137 2010-01-09T22:07:00 <dreimark> last is better
138 2010-01-09T22:08:04 * ThomasWaldmann mirrors that at static.moinmo.in
139 2010-01-09T22:08:10 <dreimark> TheSheep: how does moin-2 looks like =
140 2010-01-09T22:08:11 <ronny> added it at the bottom of the mercurial guide
141 2010-01-09T22:08:57 <TheSheep> dreimark: try it, you just run taht program in the repo: http://code.google.com/p/gource/
142 2010-01-09T22:09:08 <TheSheep> (there is even a deb)
143 2010-01-09T22:11:43 <dreimark> ok
144 2010-01-09T22:12:33 <dreimark> ThomasWaldmann: you rename MOIN_SESSION to SESSION? that makes my editmoin patch invalid
145 2010-01-09T22:12:45 <dreimark> but that doesn't matter
146 2010-01-09T22:13:01 <TheSheep> you know what's cool? if you use mercurial backends, you can use that tool for visualizing your wiki :)
147 2010-01-09T22:13:11 <dreimark> I just see that MOIN_SESSION does not conflict with other SESSION files
148 2010-01-09T22:14:12 <dreimark> TheSheep: wow great
149 2010-01-09T22:14:23 <ThomasWaldmann> TheSheep: if you could make a wiki page about that visualization, we could link it from frontpage
150 2010-01-09T22:15:22 <TheSheep> ThomasWaldmann: ok, but I need different music then, that one is (c)
151 2010-01-09T22:15:29 <ThomasWaldmann> TheSheep: is that moin/1.9?
152 2010-01-09T22:15:33 <TheSheep> yes
153 2010-01-09T22:15:47 <ThomasWaldmann> oh, yeah, please no (c) stuff :)
154 2010-01-09T22:16:22 <TheSheep> yeah
155 2010-01-09T22:19:33 <ThomasWaldmann> dreimark: it is MOIN_SESSION_x now, and x depends on configuration:
156 2010-01-09T22:20:11 <ThomasWaldmann> if cookie_name is None (default), then it'll look like MOIN_SESSION_wikiconfig or MOIN_SESSION_mywiki
157 2010-01-09T22:21:23 <ThomasWaldmann> if cookie_name is 'urlmagic', it'll make something up based on the URL, like MOIN_SESSION_8080_ROOT_wiki1 if the url is http://something:8080/wiki1/*
158 2010-01-09T22:22:07 <ThomasWaldmann> if cookie_name is "somethingelse", then it will be MOIN_SESSION_somethingelse
159 2010-01-09T22:26:31 <dreimark> ok that can be retrieved or i need to invent a new config option for editmoin
160 2010-01-09T22:26:39 <ThomasWaldmann> (and the cookies will always contain path=/ )
161 2010-01-09T22:28:23 <ThomasWaldmann> i have no idea what editmoin does currently
162 2010-01-09T22:28:50 <ThomasWaldmann> if you log in to moin, you'll get such a cookie, of course
163 2010-01-09T22:31:46 <dreimark> http://launchpadlibrarian.net/37648869/editmoin.patch
164 2010-01-09T22:34:20 <ThomasWaldmann> that'ld need some changes then
165 2010-01-09T22:34:29 <dreimark> http://bazaar.launchpad.net/%7Eniemeyer/editmoin/trunk/annotate/head%3A/editmoin#L296
166 2010-01-09T22:34:58 <dreimark> it uses then addheader to append the session name=value
167 2010-01-09T22:35:37 <dreimark> i'll add a new patch for 1.9.1 then
168 2010-01-09T22:37:25 <ThomasWaldmann> ronny: TheSheep: waldi: xorAxAx: can you have a look at the session cookie patch?
169 2010-01-09T22:37:35 <ThomasWaldmann> http://pastebin.com/f31beadab
170 2010-01-09T22:38:41 <ThomasWaldmann> the problems that patch tries to solve are:
171 2010-01-09T22:38:53 * TheSheep looks
172 2010-01-09T22:39:18 <ThomasWaldmann> cookies do not store the port, thus current moin confuses wikis running at same netloc / same path, but different port
173 2010-01-09T22:40:15 <ThomasWaldmann> and if there are cookies at path=/ and path=/mywiki, moin also gets confused because it gets the / cookie, not the /mywiki one
174 2010-01-09T22:40:34 <ThomasWaldmann> s/cookies at/cookies with/
175 2010-01-09T22:41:40 <TheSheep> ThomasWaldmann: line 52, s/usemagic/urlmagic/
176 2010-01-09T22:44:42 <ThomasWaldmann> yeah, thanks
177 2010-01-09T22:47:47 <TheSheep> looks good
178 2010-01-09T22:48:09 <TheSheep> a little disappointing that the mechanism they included in browsers for that doesn't really work
179 2010-01-09T22:49:17 <ThomasWaldmann> yeah, cookie somehow look like some quick-hacked fixes of a quick-hack
180 2010-01-09T22:49:37 <dreimark> or b00rked
181 2010-01-09T22:50:41 <ThomasWaldmann> is something missing? path and port is enough?
182 2010-01-09T22:50:43 * dreimark hopes that also fixes the problems i can' reproduce
183 2010-01-09T22:51:21 <dreimark> it is stored to the host so i think that is enough
184 2010-01-09T22:52:59 <TheSheep> ThomasWaldmann: prorocol?
185 2010-01-09T22:53:02 <TheSheep> protocol
186 2010-01-09T22:53:18 <TheSheep> but only if the port is missing
187 2010-01-09T22:53:27 <TheSheep> hmm
188 2010-01-09T22:53:38 <ThomasWaldmann> i first had that, but i thought it is redundant
189 2010-01-09T22:53:54 <ThomasWaldmann> because it will tell 80 and 443 anyway?
190 2010-01-09T22:54:11 <TheSheep> I'm not sure there is always SERVER_PORT in environ, let me look into pep 333
191 2010-01-09T22:54:32 * ThomasWaldmann isn't sure either
192 2010-01-09T22:54:46 <ThomasWaldmann> we also need to think about proxying
193 2010-01-09T22:54:50 <TheSheep> SERVER_NAME and SERVER_PORT can never be empty strings, and so are always required.
194 2010-01-09T22:55:03 <TheSheep> phew
195 2010-01-09T23:02:09 <ThomasWaldmann> proxying, anyone?
196 2010-01-09T23:02:50 <TheSheep> it shouldn't affect cookies, I think
197 2010-01-09T23:03:19 <TheSheep> and you can always force that cookie name for fancy setups
198 2010-01-09T23:03:40 <eSyr> Hello, sorry for being away so much time.
199 2010-01-09T23:03:52 <ThomasWaldmann> wb eSyr
200 2010-01-09T23:04:40 <TheSheep> eSyr: honestly, apologies are not enough, if it continues like that we won't pay you! ;)
201 2010-01-09T23:04:51 <TheSheep> no, wait
202 2010-01-09T23:05:07 <ThomasWaldmann> TheSheep: yeah, one even can force the path still (not sure whether it can make sense)
203 2010-01-09T23:05:54 <ThomasWaldmann> eSyr: http://www.sheep.art.pl/misc/moinmoin.ogv < made by TheSheep
204 2010-01-09T23:06:09 <TheSheep> actually made by autmated script
205 2010-01-09T23:06:56 <TheSheep> ThomasWaldmann: I also tried hg convert on 1.9 repo and successfully removed some files with it
206 2010-01-09T23:07:09 <TheSheep> ThomasWaldmann: removed as in obliterating their whole history
207 2010-01-09T23:07:30 <eSyr> oh, gource
208 2010-01-09T23:07:34 <TheSheep> not 100% sure if the resulting repo is perfectly correct though
209 2010-01-09T23:08:03 <ThomasWaldmann> well, that change cs hashes
210 2010-01-09T23:08:08 <TheSheep> yes
211 2010-01-09T23:08:45 <ThomasWaldmann> but I think we'll do something like that after hg supports partial history or how it was called (ronny mentioned they are working on something)
212 2010-01-09T23:09:12 <ronny> ThomasWaldmann: shallow/narrow history
213 2010-01-09T23:09:58 <ronny> ThomasWaldmann: my hope is that they will implement it at the february sprint
214 2010-01-09T23:10:04 <ThomasWaldmann> (to create moin/2.0 repo)
215 2010-01-09T23:10:24 <eSyr> (one of main choices in gource videos is right soundtrack, yeah)
216 2010-01-09T23:17:10 <eSyr> hm, strange parameters for gource visualisation.
217 2010-01-09T23:18:23 <eSyr> ThomasWaldmann, what do you think about add_msg patch?
218 2010-01-09T23:19:16 <ThomasWaldmann> i added a comment there
219 2010-01-09T23:19:57 <ThomasWaldmann> if we can't/won't fix those problems, it is pointless to raise an exception, just to catch and suppress it
220 2010-01-09T23:20:22 <eSyr> oh, i see.
221 2010-01-09T23:20:32 <eSyr> well, ok.
222 2010-01-09T23:26:15 <eSyr> I've added version of patch with logging :)
223 2010-01-09T23:35:10 <ThomasWaldmann> eSyr: yeah, please commit that one. even better would be to have a traceback there, so one sees where the wrong call comes from.
224 2010-01-09T23:35:34 <eSyr> hm.
225 2010-01-09T23:35:38 <eSyr> ok.
226 2010-01-09T23:35:47 <ThomasWaldmann> ok, so everyone is happy with the cookie changeset?
227 2010-01-09T23:36:15 <eSyr> ThomasWaldmann, about diffs: i think, adding td { width:50% } is ok for most cases (when words are not so long that the break layout), so separate table will be ok for most cases.
228 2010-01-09T23:37:16 <ThomasWaldmann> isn't the diff below automatically adjusting to content width?
229 2010-01-09T23:38:07 <eSyr> so i propose to make it width:50% :)
230 2010-01-09T23:38:41 <eSyr> (for each column)
231 2010-01-09T23:39:39 <ThomasWaldmann> well, it'll get overridden by browser anyway I guess (if it needs to)
232 2010-01-09T23:40:24 <ThomasWaldmann> so the point is not so much whether it is 50% everywhere, but rather that top part is divided in same proportion as lower part
233 2010-01-09T23:40:51 <ThomasWaldmann> but I guess that'ld require same table
234 2010-01-09T23:41:05 <ThomasWaldmann> maybe look if that is easily possible
235 2010-01-09T23:41:21 <eSyr> ok, i'll try.
236 2010-01-09T23:44:44 <eSyr> maybe, adding support of custom header in diff_html.diff solves problem?
237 2010-01-09T23:55:16 <ThomasWaldmann> exactly :)
