1 2011-03-20T00:01:22 <CIA-58> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 134:901bd4e1861e 2.0/MoinMoin/ (31 files in 5 dirs):
2 2011-03-20T00:01:22 <CIA-58> import absolute_import, division from __future__ in every module
3 2011-03-20T00:01:22 <CIA-58> add it also to the _template.py
4 2011-03-20T00:01:22 <CIA-58> remove with_statement future import (we require 2.6 and 2.6 has it by default)
5 2011-03-20T00:01:22 <CIA-58> fix one place with divisions, so they are integer divisions
6 2011-03-20T00:37:23 *** Mengyun
7 2011-03-20T00:38:38 *** Mengyun
8 2011-03-20T00:44:12 *** m4k3r
9 2011-03-20T00:56:27 *** m4k3r
10 2011-03-20T00:58:02 *** RogerHaase
11 2011-03-20T01:47:32 *** MattMaker
12 2011-03-20T01:49:37 *** MattMaker
13 2011-03-20T02:15:08 * ThomasWaldmann plays with 2to3
14 2011-03-20T02:32:01 <CIA-58> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 135:75312c9b94e3 2.0/MoinMoin/ (32 files in 13 dirs): use except...as syntax, use raise Exception(...) syntax, fix some imports
15 2011-03-20T02:42:23 <CIA-58> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 136:13bce99cba0b 2.0/MoinMoin/ (15 files in 7 dirs): use sorted(), use while True
16 2011-03-20T03:49:50 *** kasun
17 2011-03-20T04:01:01 *** Mengyun
18 2011-03-20T04:27:31 <MattMaker> I think the fs2 was not being aggressive enough in closing connections. this patch makes the tests work again with fs2: http://paste.pocoo.org/show/356437/
19 2011-03-20T04:30:49 <MattMaker> reference: http://www.sqlalchemy.org/trac/browser/lib/sqlalchemy/dialects/sqlite/pysqlite.py#L96
20 2011-03-20T04:31:50 <MattMaker> and http://www.sqlalchemy.org/docs/core/pooling.html
21 2011-03-20T05:10:04 *** sinha_
22 2011-03-20T05:11:32 *** kasun
23 2011-03-20T05:41:01 *** sl33k_
24 2011-03-20T05:43:35 *** sl33k_
25 2011-03-20T05:45:51 *** sinha_
26 2011-03-20T05:59:00 *** kasun
27 2011-03-20T06:04:33 *** kasun
28 2011-03-20T06:16:42 *** kasun
29 2011-03-20T06:37:46 *** kasun
30 2011-03-20T06:55:16 *** sl33k__
31 2011-03-20T06:56:37 *** sl33k_
32 2011-03-20T06:58:08 *** m4k3r
33 2011-03-20T07:16:23 *** m4k3r
34 2011-03-20T07:26:49 *** m4k3r
35 2011-03-20T10:12:59 <dreimark> ThomasWaldmann: i guess we have a problem with xapian
36 2011-03-20T10:13:09 <dreimark> in 1.9.3.x
37 2011-03-20T10:13:19 <dreimark> MoinMoin/script/moin.py --config-dir=`pwd` index build --mode=rebuild
38 2011-03-20T10:13:42 <xorAxAx> MattMaker: but doesnt it slow down the backend?
39 2011-03-20T10:14:11 <xorAxAx> OTOH, connecting to sqlite should be pretty fast and cheap
40 2011-03-20T10:14:33 <dreimark> ThomasWaldmann: i get always 2011-03-20 10:13:56,710 WARNING MoinMoin.search:52 Slow moin search is used because the Xapian index does not exist. You should create it using the moin index build command
41 2011-03-20T10:15:04 *** kasun
42 2011-03-20T11:17:32 <dreimark> bbl
43 2011-03-20T11:34:31 *** greg_f
44 2011-03-20T11:40:51 *** Marchael
45 2011-03-20T11:48:52 *** sl33k___
46 2011-03-20T11:49:08 *** sl33k__
47 2011-03-20T11:49:11 *** sl33k___
48 2011-03-20T12:09:51 *** ThomasWaldmann
49 2011-03-20T12:09:57 *** ThomasWaldmann
50 2011-03-20T12:09:58 *** ThomasWaldmann
51 2011-03-20T12:22:25 *** m4k3r
52 2011-03-20T12:24:15 *** m4k3r
53 2011-03-20T12:28:29 <ThomasWaldmann> MattMaker: hmm, wouldn't it be better to solve the connection teardown problem in general?
54 2011-03-20T13:22:55 *** ErSame
55 2011-03-20T13:27:55 *** m4k3r
56 2011-03-20T13:28:56 <MattMaker> ThomasWaldmann, xorAxAx: this solution is appropriate for sqlite to disk file, so I suggest it is correct for fs2.py.
57 But yes, a different solution should be investigated for other backend types, in case they have a hidden problem with connection pooling.
58 2011-03-20T13:29:53 *** kasun
59 2011-03-20T13:33:38 <ThomasWaldmann> MattMaker: as far as i could see, the problem is that the connection(s) do not get teared down
60 2011-03-20T13:33:50 <ThomasWaldmann> there is no code in moin that does it
61 2011-03-20T13:34:04 <ThomasWaldmann> and it seems to not happen automatically
62 2011-03-20T13:34:13 <ThomasWaldmann> and i didn't find docs about it
63 2011-03-20T13:34:15 <MattMaker> also, I think this issue was caused by having several dozen moin flasks created by the testing suite
64 2011-03-20T13:34:47 <ThomasWaldmann> sure, but there should be a way to close that stuff
65 2011-03-20T13:34:56 *** ErSame
66 2011-03-20T13:34:57 <MattMaker> I found docs that suggest the default is for all the instances to share the default, and the default keeps a bunch of connections all for up to 30 seconds
67 2011-03-20T13:35:19 <MattMaker> so yes, that could be experimented with explicitly
68 2011-03-20T13:35:50 *** m4k3r
69 2011-03-20T13:36:24 <ThomasWaldmann> if the app isn't needed any more, all connections belonging to it should be closed
70 2011-03-20T13:36:52 <MattMaker> they are... after 30 seconds
71 2011-03-20T13:36:58 <MattMaker> but wait
72 2011-03-20T13:37:01 <MattMaker> "By default, the connect() method will return the same connection that is already checked out in the current thread. This allows a particular connection to be used in a given thread without needing to pass it around between functions. To disable this behavior, specify use_threadlocal=False to the manage() function."
73 2011-03-20T13:37:10 <MattMaker> testing is all in one thread
74 2011-03-20T13:37:34 <MattMaker> so during a 30 second window you run out of open files
75 2011-03-20T13:38:12 <MattMaker> I ran the test suite under python -m trace... which made it go so slow that it was okay again :)
76 2011-03-20T13:39:03 <MattMaker> but there will still be a problem of how many moins that testing opens side by side under a single thread
77 2011-03-20T13:39:27 <MattMaker> there are some ways maybe to tweak that in sqlalchemy
78 2011-03-20T13:39:46 <MattMaker> but maybe the single thread multiple moins is a problem with having valid tests?
79 2011-03-20T13:40:25 <MattMaker> "The Engine returned by the create_engine() function in most cases has a QueuePool integrated, pre-configured with reasonable pooling defaults. If you’re reading this section to simply enable pooling- congratulations! You’re already done."
80 2011-03-20T13:41:03 <MattMaker> these "reasonable" defaults appears to be pool_size=5, max_overflow=10, timeout=30
81 2011-03-20T13:41:31 <xorAxAx> i thought it would be backend specific
82 2011-03-20T13:41:35 <xorAxAx> i.e. db specific
83 2011-03-20T13:41:41 <xorAxAx> but maybe they changed it
84 2011-03-20T13:43:23 <MattMaker> xorAxAx you may be right, it does claim "This is used for SQLite" under the docs for pool.SingletonThreadPool (pool size = 5)
85 2011-03-20T13:43:45 *** kasun
86 2011-03-20T13:44:06 <ThomasWaldmann> isn't there some method to close down stuff used by an engine?
87 2011-03-20T13:44:29 <MattMaker> in theory, dispose() does it
88 2011-03-20T13:45:22 <MattMaker> but it's already too late for that to help when running tests, because having a pool means you'll have all those open files (pointing at the same file, but it still counts)
89 2011-03-20T13:46:05 <MattMaker> it happens automatically when each test is complete
90 2011-03-20T13:46:25 <MattMaker> but there are a bunch of moins (flasks?) all open at the same time
91 2011-03-20T13:46:31 <MattMaker> while the test is running
92 2011-03-20T13:47:26 <MattMaker> all the notes I just said are pulled from http://www.sqlalchemy.org/docs/core/pooling.html
93 2011-03-20T13:48:04 <MattMaker> and the advocacy of the NullPool solution for fs2 is in the source: http://www.sqlalchemy.org/trac/browser/lib/sqlalchemy/dialects/sqlite/pysqlite.py#L96
94 2011-03-20T13:51:44 <ThomasWaldmann> ok, I'll try the NullPool
95 2011-03-20T13:51:58 <MattMaker> separate from this, whatever pytest is doing that creates all these apps at once, then closes them later all at once, might be a problem
96 2011-03-20T13:52:37 <ThomasWaldmann> see conftest.py
97 2011-03-20T13:53:50 <MattMaker> I put some logging in conftest.py to try to understand, but I don't quite "get it" yet :)
98 2011-03-20T13:54:05 <ThomasWaldmann> it's a mess
99 2011-03-20T13:54:25 <ThomasWaldmann> if we only had test classes with tests in methods, it could be cleaner
100 2011-03-20T13:54:46 <ThomasWaldmann> but there are some test functions on module level
101 2011-03-20T13:55:14 <ThomasWaldmann> so we have to support both somehow right now (or at least check)
102 2011-03-20T13:55:43 <ThomasWaldmann> because of that, app creation etc. is nested weirdly
103 2011-03-20T13:56:34 <MattMaker> I have considered trying to help with the tests -- it seems to me that when I submit a change, I should submit a test for it also -- but I don't know how to do it until I understand conftest.py, unless I were to start over with something new for testing
104 2011-03-20T13:58:34 <MattMaker> one thing I was hacking at at one point, I wanted to see a unique id for each of the nested app creations... so I tried os.pid or something like that... and they were all the same one. I suppose they were the thread of pytest itself?
105 2011-03-20T14:00:18 <ThomasWaldmann> py.test is usually just one process (afaik)
106 2011-03-20T14:01:07 <MattMaker> ThomasWaldmann I see what you mean about supporting both types of tests, but what should be done? can the tests be rewritten, are there too many for that to be a happy plan?
107 2011-03-20T14:02:11 <ThomasWaldmann> MattMaker: can you compare the patch in the pastebin against a local created patch file?
108 2011-03-20T14:02:26 <ThomasWaldmann> patch unexpectedly ends in middle of line
109 2011-03-20T14:02:35 <ThomasWaldmann> this happens if one wgets that url and applies the patch
110 2011-03-20T14:02:51 <MattMaker> ah, I meant only to show it, sorry
111 2011-03-20T14:02:56 <MattMaker> just those 2 lines
112 2011-03-20T14:03:08 <ThomasWaldmann> so it would be interesting if the pastebin kills the final \n or if it never was there
113 2011-03-20T14:03:19 <MattMaker> remind me the proper command for patch?
114 2011-03-20T14:03:37 <ThomasWaldmann> patch -p1 --dry-run < wgettedfile
115 2011-03-20T14:05:28 <MattMaker> and to create it, hg diff MoinMoin/storage/backends/fs2.py >patch.txt ? or am I forgetting a detail
116 2011-03-20T14:06:35 <ThomasWaldmann> if you did not change anything else hg diff >patch.txt
117 2011-03-20T14:07:54 *** ErSame
118 2011-03-20T14:08:51 <MattMaker> oh interesting, my commandline pipe-to-clipboard fails because I am user "moin" in the commandline and "trip" in the graphical part of the OS
119 2011-03-20T14:09:38 <MattMaker> here is when I copy paste the whole thing: http://paste.pocoo.org/show/356558/
120 2011-03-20T14:10:31 *** sinha_
121 2011-03-20T14:10:38 <ThomasWaldmann> just compare wgetted file to locally created file
122 2011-03-20T14:10:51 <ThomasWaldmann> if length is same, the problem is local
123 2011-03-20T14:11:12 <ThomasWaldmann> if wgetted is one less than local, it is the pastebin service
124 2011-03-20T14:12:02 <MattMaker> the wgetted pastebin one says no newline at end of file
125 2011-03-20T14:13:36 <MattMaker> it may be because of the content-type
126 2011-03-20T14:13:59 <MattMaker> wget is getting it as text/html
127 2011-03-20T14:14:31 <ThomasWaldmann> you wgetted the raw url?
128 2011-03-20T14:15:14 <ThomasWaldmann> hm, wgetted or wgot? :)
129 2011-03-20T14:16:33 <MattMaker> hehehe
130 2011-03-20T14:17:00 <MattMaker> yes, the raw url
131 2011-03-20T14:17:45 <MattMaker> maybe not conjugate it as a verb, let it stay a token :)
132 2011-03-20T14:17:54 <ThomasWaldmann> ok, so is the wgetted file length different from your locally created file?
133 2011-03-20T14:19:00 <MattMaker> yes, by one
134 2011-03-20T14:19:18 <ThomasWaldmann> ok, so pastebin to blame
135 2011-03-20T14:19:19 <MattMaker> the newline at EOF I presume
136 2011-03-20T14:20:37 <MattMaker> wget -S to see headers
137 2011-03-20T14:20:48 <MattMaker> "Length: unspecified [text/plain]"
138 2011-03-20T14:21:46 *** ErSame
139 2011-03-20T14:22:27 <ThomasWaldmann> MattMaker: ok, NullPool patch fixes fs2 file leak
140 2011-03-20T14:23:06 <ThomasWaldmann> we still have a similar leak somewhere else btw. but it only leaks ~150 files so no showstopper right now
141 2011-03-20T14:23:17 <MattMaker> excellent! hopefully the change harms nothing
142 2011-03-20T14:23:30 <ThomasWaldmann> if you run the storage tests, you'll see
143 2011-03-20T14:23:47 <ThomasWaldmann> lsof -p <pid of py.test process> is nice to watch it
144 2011-03-20T14:23:48 <MattMaker> I will take a look for related leaks while it's fresh on my mind
145 2011-03-20T14:24:46 <MattMaker> and ... was it a leak, or was it rather that each instance had [quantity X] of files open, and there were dozens of instances (apps?)
146 2011-03-20T14:25:19 <ThomasWaldmann> i didn't verify that, but i suspect the latter
147 2011-03-20T14:25:31 <MattMaker> I like lsof, but I keep forgetting its name, I'll make a note of that
148 2011-03-20T14:25:56 <ThomasWaldmann> LiSt Open Files
149 2011-03-20T14:26:00 <MattMaker> I also made use of FsEventer.App on OS X, to watch the same graphically
150 2011-03-20T14:27:35 <MattMaker> hmm, so I haven't looked at how having multiple backends running affects things
151 2011-03-20T14:28:11 <MattMaker> but if more than one is using sqlalchemy and sqlite, in the same thread, it might be something to study
152 2011-03-20T14:28:41 <MattMaker> and the standard arrangement creates three backends, the main one, user, and trash
153 2011-03-20T14:28:58 <MattMaker> trash apparently is not indexed :)
154 2011-03-20T14:31:58 * ThomasWaldmann reruns tests. didn't have fs2 configured.
155 2011-03-20T14:32:43 <MattMaker> I am thinking that pytest will make some tests harder to pass, because of it launching things repatedly under its same pid... files that would have been destroyed, might linger, thinking their owner is still alive
156 2011-03-20T14:33:07 <MattMaker> maybe that's good though, having harder tests :)
157 2011-03-20T14:33:38 <ThomasWaldmann> wsgi process is also long running, but maybe not with that many apps
158 2011-03-20T14:34:26 *** RogerHaase
159 2011-03-20T14:37:21 <ThomasWaldmann> hi roger
160 2011-03-20T14:37:26 <sinha_> ThomasWaldmann: Hi , I was going through moin code structure at http://moinmo.in/MoinDev/CodeStructure ..there is said that first file which is called is moin.cgi..but i am not finding it here in moin2
161 2011-03-20T14:37:45 <ThomasWaldmann> i've seen you did some css work, just ping me if it is ok to pull the changes
162 2011-03-20T14:37:54 <MattMaker> if there are multiple tests for the Indexing backend, and there are lots of instances open, I wonder if they could end up fighting one another over the contents of the sqlite files
163 2011-03-20T14:38:32 <ThomasWaldmann> sinha_: that is for 1.9 and does not apply to moin2
164 2011-03-20T14:39:40 <sinha_> ThomasWaldmann : any it any other place where 2.0 is mentioned ? because i want to get some idea of request processing before getting started
165 2011-03-20T14:40:02 <ThomasWaldmann> sinha_: see MoinMoin2.0 page, there is a link to new docs
166 2011-03-20T14:40:30 <ThomasWaldmann> also, you can always just look into the code. start in MoinMoin.app.
167 2011-03-20T14:41:01 <ThomasWaldmann> as moin is a flask app, flask docs might be also useful to understand the framework.
168 2011-03-20T14:41:45 <ThomasWaldmann> MattMaker: there should not be tests running in parallel
169 2011-03-20T14:42:34 <sinha_> ThomasWaldmann : okay .
170 2011-03-20T14:46:26 <CIA-58> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 137:989ada5c5cbb 2.0/MoinMoin/storage/backends/fs2.py:
171 2011-03-20T14:46:26 <CIA-58> fs2: use NullPool class for sqlite db connection pool, fix #4 - thanks to Matt Maker
172 2011-03-20T14:46:26 <CIA-58> http://www.sqlalchemy.org/trac/browser/lib/sqlalchemy/dialects/sqlite/pysqlite.py#L96
173 2011-03-20T14:46:26 <CIA-58> http://www.sqlalchemy.org/docs/core/pooling.html
174 2011-03-20T14:46:54 * ThomasWaldmann likes the magic bug closing :)
175 2011-03-20T14:47:00 <MattMaker> :)
176 2011-03-20T14:47:31 <MattMaker> because of mentioning the #4 in the commit? I like that also
177 2011-03-20T14:47:53 <ThomasWaldmann> yes "fix #4" triggers closing of the bug #4
178 2011-03-20T14:49:42 <MattMaker> I see it. That's some excellent magic
179 2011-03-20T14:49:59 <ThomasWaldmann> RogerHaase: btw, while you're working on css, could you have a look at that: https://bitbucket.org/thomaswaldmann/moin-2.0/issue/5/characters-show-up-on-printouts
180 2011-03-20T14:50:22 <ThomasWaldmann> looks like easy to fix with some css
181 2011-03-20T14:51:09 <RogerHaase> ThomasWaldmann: Already fixed in my repo, I noted that a few days ago.
182 2011-03-20T14:51:33 <RogerHaase> You may have been sleeping and missed the message.
183 2011-03-20T14:53:24 <ThomasWaldmann> oh, cool :)
184 2011-03-20T14:53:40 <ThomasWaldmann> did you recently merged main repo into yours?
185 2011-03-20T14:53:58 <ThomasWaldmann> i did some minor changes to css to have icons in item lists
186 2011-03-20T14:54:34 <ThomasWaldmann> before that it always appended (text/x.moin.wiki) after the item name, but that was too much bloat
187 2011-03-20T14:55:03 <ThomasWaldmann> so i am just setting class attr now, so we can show some icon
188 2011-03-20T14:55:20 <RogerHaase> Yes, I merged and pushed about 10 minutes ago, but will do again since you had more updates.
189 2011-03-20T14:55:20 <ThomasWaldmann> (not many usable icons there currently, though)
190 2011-03-20T14:58:16 <ThomasWaldmann> RogerHaase: is it in a good state right now to pull?
191 2011-03-20T14:58:51 <RogerHaase> give me a minute for a quick test and another push.
192 2011-03-20T14:59:28 <sinha_> ThomasWaldmann : Before getting started do i really need to go through the details of Flask,werkzeug, jinja2,etc or i just startover working on any feature will get acquaintained by working on them
193 2011-03-20T15:00:20 <ThomasWaldmann> well, both might work, depends on the feature
194 2011-03-20T15:00:58 <RogerHaase> ThomasWaldmann: Ready
195 2011-03-20T15:01:14 <sinha_> ThomasWaldmann : okay then i will start with second one.I dont want to take much time to start working on moin :)
196 2011-03-20T15:01:35 <sinha_> ThomasWaldmann : will read stuff where i am not getting something.
197 2011-03-20T15:02:16 <ThomasWaldmann> you also can ask here if you're stuck - after doing some quick own research, but before wasting too much time
198 2011-03-20T15:09:20 <CIA-58> Roger Haase <crosseyedpenguin@yahoo.com> default * 138:eebbee623c44 2.0/MoinMoin/themes/modernized/static/css/common.css: add "clear: both" to block elements with borders to prevent borders from overlaying floated elements like TOC
199 2011-03-20T15:09:21 <CIA-58> Roger Haase <crosseyedpenguin@yahoo.com> default * 139:748102c6dac9 2.0/MoinMoin/templates/diff.html: correct spelling of between
200 2011-03-20T15:09:22 <CIA-58> Roger Haase <crosseyedpenguin@yahoo.com> default * 140:1e0f4393aa58 2.0/MoinMoin/templates/layout.html: add missing moin- prefix
201 2011-03-20T15:09:22 <CIA-58> Roger Haase <crosseyedpenguin@yahoo.com> default * 141:97f598c58283 2.0/MoinMoin/ (5 files in 2 dirs): Combine print.css, projection.css and screen.css into common.css.
202 2011-03-20T15:09:23 <CIA-58> Roger Haase <crosseyedpenguin@yahoo.com> default * 142:b1b9e7ffe2d2 2.0/MoinMoin/themes/modernized/static/css/common.css: reduce hyperlink clutter on printed output; color printed hyperlinks black; eliminate box-shadow on print
203 2011-03-20T15:09:24 <CIA-58> Roger Haase <crosseyedpenguin@yahoo.com> default * 143:ff9fb651f472 2.0/MoinMoin/themes/modernized/static/css/common.css: no functional changes; more consistent spacing and indentation; fewer lines
204 2011-03-20T15:09:26 <CIA-58> Roger Haase <crosseyedpenguin@yahoo.com> default * 144:bf0586f3b72c 2.0/MoinMoin/ (7 files in 2 dirs): merge
205 2011-03-20T15:09:27 <CIA-58> Roger Haase <crosseyedpenguin@yahoo.com> default * 145:eade7d9989a1 2.0/MoinMoin/ (7 files in 2 dirs): merge
206 2011-03-20T15:10:46 <ThomasWaldmann> very nice, 3 css files less :)
207 2011-03-20T15:11:54 <ThomasWaldmann> RogerHaase: https://bitbucket.org/thomaswaldmann/moin-2.0/issue/5/characters-show-up-on-printouts you want to close that one?
208 2011-03-20T15:13:32 <ThomasWaldmann> btw, if you fix a bug and commit a changeset with the fix, just talk about "fix #5" in the commit comment
209 2011-03-20T15:14:08 <ThomasWaldmann> it'll be magically closed then (I guess after I pull/push)
210 2011-03-20T15:14:52 <RogerHaase> OK, will do. When I made the change, I had not yet discovered that we had a bug list.
211 2011-03-20T15:15:29 <ThomasWaldmann> yes, it is rather new. since we have the small moin-2.0 repo.
212 2011-03-20T15:16:04 *** sinha_
213 2011-03-20T15:18:10 <ThomasWaldmann> eSyr: shall we live with the simple rst link scheme fix I did for the javascript: stuff or do you have something pending for it?
214 2011-03-20T15:30:53 <ThomasWaldmann> RogerHaase: do you have an idea how we could improve the editbar while still keeping it modernized-like?
215 2011-03-20T15:31:50 <ThomasWaldmann> a major issue i see there right now is that at least one often used thing ("global history") is currently hidden in the menu
216 2011-03-20T15:33:56 <ThomasWaldmann> also, sometimes the editbar jumps (likely because of that js hack)
217 2011-03-20T15:39:26 <ThomasWaldmann> shall we get rid of the menu and do it in the same way as with >>> in the navibar?
218 2011-03-20T15:41:19 <ThomasWaldmann> hmm, i tried making the browser window less wide and it seem to scale as well as one could expect
219 2011-03-20T15:41:58 <ThomasWaldmann> only one cosmetical issue: if editbar wraps, only first line is indented correctly, 2nd line is a bit too far to the left
220 2011-03-20T15:42:46 <ThomasWaldmann> any comments?
221 2011-03-20T15:44:29 <ThomasWaldmann> (after unifying all into editbar items, we could split it again, making groups of item-related views, global views)
222 2011-03-20T15:44:43 *** kasun
223 2011-03-20T15:47:22 <ThomasWaldmann> TheSheep: what do you think? it is a bit post-modernized then already, but still rather similar.
224 2011-03-20T15:47:51 *** cdom
225 2011-03-20T15:56:07 * MattMaker tries to imagine a wiki that represents Postmodernism
226 2011-03-20T15:59:25 * ThomasWaldmann tries it, if noone keeps me back :)
227 2011-03-20T16:00:44 <RogerHaase> ThomasWaldmann: I always disliked the editbar more actions drop down box. One idea was to make editbar actions into two lists and hide the second list with CSS. Clicking on "more" would unhide the second list and wrap to a second line.
228 2011-03-20T16:01:34 <ThomasWaldmann> yes, that's quite what I mean - plus the grouping
229 2011-03-20T16:02:09 <ThomasWaldmann> RogerHaase: can you try fixing the left border issue when it wraps?
230 2011-03-20T16:02:25 <RogerHaase> But that requires javascript. But a default with no javascript browsers could be to show all actions. That could cause the editbar to jump when the page is rendered. Would have to try.
231 2011-03-20T16:02:38 <RogerHaase> Will work on left border.
232 2011-03-20T16:04:10 <MattMaker> I do feel strongly that things should still be possible when javascript is disabled or filtered by the user's security
233 2011-03-20T16:04:43 <ThomasWaldmann> yes
234 2011-03-20T16:05:07 * MattMaker wonders about making tests for that
235 2011-03-20T16:05:34 <MattMaker> I guess it's possible for tests to get *too* specific
236 2011-03-20T16:06:44 <MattMaker> ThomasWaldmann we should think about what happens when the user adds plugins, they will expect those to appear somewhere, as they do now in the dropdown
237 2011-03-20T16:11:06 <RogerHaase> MattMaker: the plugins would appear in the second list, no different from any other item.
238 2011-03-20T16:11:35 <MattMaker> ok
239 2011-03-20T16:13:04 *** z^ck
240 2011-03-20T16:14:45 * ThomasWaldmann tweaks editbar.html
241 2011-03-20T16:19:47 <MattMaker> bbl
242 2011-03-20T16:24:00 *** Marchael
243 2011-03-20T16:32:17 <ThomasWaldmann> RogerHaase: indentation is also wrong in 3rd line. so basically in any > first.
244 2011-03-20T16:34:33 *** m4k3r
245 2011-03-20T16:34:57 *** m4k3r
246 2011-03-20T16:40:54 *** Mengyun
247 2011-03-20T16:42:14 <ThomasWaldmann> RogerHaase: the lower copy of editbar seems to have a bit less upper and lower padding, text content is closer to border
248 2011-03-20T16:43:53 <ThomasWaldmann> also, the list entries are vertically closer to each other when wrapping occurs
249 2011-03-20T16:46:14 <RogerHaase> The indent is a Firefox + Opera problem, IE, Iron, and Safari are OK per my testing.
250 2011-03-20T16:49:28 <ThomasWaldmann> but it works with the upper editbar
251 2011-03-20T16:51:41 <ThomasWaldmann> (yes, ff4 here)
252 2011-03-20T16:53:50 *** Mengyun
253 2011-03-20T17:01:24 <CIA-58> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 146:751ebb4ff42b 2.0/MoinMoin/ (5 files in 4 dirs): (log message trimmed)
254 2011-03-20T17:01:24 <CIA-58> modernized theme: killed "More" menu, put everything into editbar (for now)
255 2011-03-20T17:01:24 <CIA-58> this is just a first step and there needs to be more work done to make this
256 2011-03-20T17:01:24 <CIA-58> more pretty / more usable.
257 2011-03-20T17:01:24 <CIA-58> killed the actionsMenuInit js code (was a bit unpretty anyway, caused jumping
258 2011-03-20T17:01:24 <CIA-58> of editbar)
259 2011-03-20T17:01:25 <CIA-58> killed actions_menu_options method from the theme support (not needed any
260 2011-03-20T17:01:46 <ThomasWaldmann> RogerHaase: ^^
261 2011-03-20T17:12:13 <RogerHaase> Now IE, Iron, and Safari show problem when second word of link wraps (Firefox and Opera OK there). Will work on this later today/tonight.
262 2011-03-20T17:17:14 <ThomasWaldmann> ok
263 2011-03-20T17:17:28 * ThomasWaldmann does more tweaking so reordering stuff is possible more easily
264 2011-03-20T17:23:09 <RogerHaase> Another CSS issue is hyperlink mouseover effect -- right now, different links turn red, green, or stay the same color with underlines. Any objection to making these all the same?
265 2011-03-20T17:24:48 <RogerHaase> But for now, leave the navibar tabs as is.
266 2011-03-20T17:35:23 <ThomasWaldmann> well, it should reflect whether it was already visited and whether it exists
267 2011-03-20T17:44:10 <RogerHaase> Indication of visited and exists makes sense in page content and TOC, but not needed/wanted/current in page trail, editbar, navibar, etc.
268 2011-03-20T17:56:38 <ThomasWaldmann> hmm
269 2011-03-20T17:57:09 <ThomasWaldmann> i also thought about giving up the exists classification at some places, but due to performance reasons
270 2011-03-20T17:57:37 <ThomasWaldmann> (like in trail)
271 2011-03-20T17:58:32 <ThomasWaldmann> so i guess we have to look at each thing individually and decide
272 2011-03-20T18:11:30 <MattMaker> it might be interesting to set up some sort of analytics on the test wiki, to see where people click on it. Actually, setting up formal A/B testing with volunteers who are given goals to achieve and then you watch what they do with the UI, could that be a good project for some GSOC (?) participants?
273 2011-03-20T18:12:20 <MattMaker> it wouldn't even require learning the codebase, just setting up the experiments
274 2011-03-20T18:12:41 <MattMaker> (and it's a marketable skill :) )
275 2011-03-20T18:12:51 *** kasun
276 2011-03-20T18:15:25 <RogerHaase> ThomasWaldmann: Has there been any thought to saving client-side state in a cookie or local storage? One use would be if user shows/hides part of editbar; every subsequent page would be displayed the same way until user changes his choice.
277 2011-03-20T18:20:14 <ThomasWaldmann> MattMaker: it can be part of a coding project to implement something that makes sense :)
278 2011-03-20T18:21:39 <ThomasWaldmann> RogerHaase: well, for logged in users, we have a session, but this needs to be done on the server then (the session data is stored client side in the end, though)
279 2011-03-20T18:22:27 <ThomasWaldmann> btw, there is currently something going on at the EU lawmakers about "tracking users" with cookies
280 2011-03-20T18:22:48 <ThomasWaldmann> it will have to be seen how paranoid this will get
281 2011-03-20T18:25:43 <RogerHaase> Using local storage would be better, but some browsers (maybe all) ask for permission in a rather scary way. I have only seen the message once and I don't remember the site.
282 2011-03-20T18:28:51 <RogerHaase> If we leave editbar as is (maybe change the order of items), then an on-load JS could hide all but the first 5 items and add a "more" to the list. Clicking on "more" displays the entire editbar list and adds a "less" at the end of the list. Clicking on less, etc.
283 2011-03-20T18:29:54 <RogerHaase> Last preference of user is saved in a cookie/local storage and subsequent pages are displayed the same.
284 2011-03-20T18:33:06 <RogerHaase> The info to be saved is only used client-side, if in a cookie, the server-side would ignore it.
285 2011-03-20T18:34:02 <MattMaker> actually I have been annoyed when a page that doesn't exist is in my trail, like for example I misspelled something, it becomes something to accidentally click on or be confused by later.
286 2011-03-20T18:35:33 <RogerHaase> good point.
287 2011-03-20T18:48:41 <MattMaker> is http://moinmo.in/CodingStyle up-to-date?
288 2011-03-20T18:50:11 <MattMaker> and also http://moinmo.in/MoinDev/CodingStandards
289 2011-03-20T19:01:19 *** sinha_
290 2011-03-20T19:13:18 *** greg_f
291 2011-03-20T19:22:16 * dreimark from the ML We have temporarily disabled the creation of new requests and invites
292 2011-03-20T19:22:19 <dreimark> in preparation of the launch of the new UI for Melange later this
293 2011-03-20T19:22:22 <dreimark> week.
294 2011-03-20T19:22:31 <dreimark> re
295 2011-03-20T19:28:53 <ThomasWaldmann> omg
296 2011-03-20T19:28:55 <CIA-58> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 147:c7e83ed2ee4b 2.0/MoinMoin/ (5 files in 3 dirs): (log message trimmed)
297 2011-03-20T19:28:55 <CIA-58> edit_bar, actions_excluded - use endpoint names as entries
298 2011-03-20T19:28:55 <CIA-58> edit_bar now has tuples like (endpointname, label) so we can easier build the
299 2011-03-20T19:28:55 <CIA-58> edit_bar html in the template. order is now important, the edit_bar entries
300 2011-03-20T19:28:55 <CIA-58> will be rendered in the order given.
301 2011-03-20T19:28:55 <CIA-58> actions_excluded now uses the same endpointnames as edit_bar
302 2011-03-20T19:28:56 <CIA-58> editbar.html - much less duplication now
303 2011-03-20T19:37:34 * ThomasWaldmann updated CodingStyle
304 2011-03-20T19:41:14 <dreimark> btw. sinha_ textch is missing for +copy item if you want to implement that, feel free
305 2011-03-20T19:41:52 <sinha_> dreimark: please give me some more details or any reference
306 2011-03-20T19:42:17 <dreimark> if you start the recent moin-2 code you see it on the edit bar
307 2011-03-20T19:43:05 <dreimark> http://static.moinmo.in/m2/html/admin/configure.html?highlight=textcha
308 2011-03-20T19:43:16 <dreimark> for configuring a textcha
309 2011-03-20T19:44:24 *** m4k3r
310 2011-03-20T19:44:57 *** m4k3r
311 2011-03-20T19:45:00 <dreimark> sinha_: search the old related 2.0-dev repo
312 2011-03-20T19:45:20 <dreimark> there was a gci task about implementing textcha to various views
313 2011-03-20T19:45:29 <dreimark> copy was left over or was missed
314 2011-03-20T19:46:45 <dreimark> http://hg.moinmo.in/moin/2.0-dev just search on that online
315 2011-03-20T19:47:34 <sinha_> dreimark: okay i am doing that..
316 2011-03-20T19:47:35 *** cdom
317 2011-03-20T19:47:42 <dreimark> :)
318 2011-03-20T19:48:01 *** cdom
319 2011-03-20T19:49:25 <ThomasWaldmann> rename edit_bar > item_bar, actions_excluded > endpoints_excluded ?
320 2011-03-20T19:50:09 <ThomasWaldmann> or item_views?
321 2011-03-20T19:50:28 <dreimark> better
322 2011-03-20T19:50:45 <dreimark> item_bar sounds strange and did not describe it
323 2011-03-20T19:50:58 <dreimark> or only views?
324 2011-03-20T19:51:34 <ThomasWaldmann> maybe the global stuff should not be in there at all, only the stuff that is related to the item
325 2011-03-20T19:53:53 <dreimark> Local Site Map and Global* and Administration
326 2011-03-20T19:55:12 <RogerHaase> I like item_views. Maybe leave Admin&docs in list and move other global items to Admin&docs menu?
327 2011-03-20T19:55:18 <dreimark> we have to give some id or name to those views which should stay. so that plugin views will move to the correct place
328 2011-03-20T19:55:57 <ThomasWaldmann> RogerHaase: global history is too important to be hidden deep in admin index
329 2011-03-20T19:56:56 <ThomasWaldmann> but i am thinking about putting it back into navibar somehow, where it used to be
330 2011-03-20T19:58:01 <RogerHaase> I was thinking user could add it to his navibar, if he visited that page often.
331 2011-03-20T19:58:53 <ThomasWaldmann> well, yes, but still: too important to hide
332 2011-03-20T19:59:32 <ThomasWaldmann> i have visited some "wikis" in the past and was always slightly p*ssed if they hid recentchanges so it was not easy to find
333 2011-03-20T20:00:15 <ThomasWaldmann> also, the docs/help, the global index, the tags should be also rather easy to find
334 2011-03-20T20:00:43 *** Dae-ekleN
335 2011-03-20T20:01:36 <RogerHaase> agree, those would be better placed on the navibar.
336 2011-03-20T20:02:22 <ThomasWaldmann> btw, for the navibar, we have still a problem to solve: all that link parsing / split_navi_link stuff
337 2011-03-20T20:02:39 <ThomasWaldmann> plus user profile migration from 1.9 in case we change it
338 2011-03-20T20:05:49 <Dae-ekleN> hello, i'm interested in user interface task for GSOC. i have a few questions about it, can i ask them now?
339 2011-03-20T20:07:05 *** kasun
340 2011-03-20T20:07:06 <ThomasWaldmann> sure
341 2011-03-20T20:07:14 <CIA-58> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 148:972943980b7b 2.0/MoinMoin/ (4 files in 2 dirs):
342 2011-03-20T20:07:14 <CIA-58> rename cfg attrs: edit_bar > item_views, actions_excluded > endpoints_excluded
343 2011-03-20T20:07:14 <CIA-58> item_views is not just about editing, it has all sort of views related to
344 2011-03-20T20:07:14 <CIA-58> the item (it currently has even some global stuff, but that shall change)
345 2011-03-20T20:07:14 <CIA-58> we don't have "actions" any more (that's a kind of moin 1.9 plugin) and the
346 2011-03-20T20:07:15 <CIA-58> exclude list now has endpoint names in it
347 2011-03-20T20:08:29 <Dae-ekleN> how much important for you is experience with js and jinja2?
348 2011-03-20T20:09:39 <ThomasWaldmann> well, experience with j2 helps, but it isn't rocket science, so you are maybe able to learn it quickly
349 2011-03-20T20:09:41 <Dae-ekleN> how i cab see in requirements list you first mentioned about artistic skills, design, ui, css,..
350 2011-03-20T20:10:14 <ThomasWaldmann> experience with js is maybe more important, as it is harder to get into it
351 2011-03-20T20:10:32 <ThomasWaldmann> but it depends on the amount of js
352 2011-03-20T20:11:36 <ThomasWaldmann> UI design skills, taste, html, css is likely most important
353 2011-03-20T20:11:54 <Dae-ekleN> i have mentioned above qualities, i also have 8 months of thml+css+design working experience
354 2011-03-20T20:12:21 <Dae-ekleN> i also love programming, specially python and php
355 2011-03-20T20:12:49 <dreimark> how can you love php if you do python?
356 2011-03-20T20:12:51 <ThomasWaldmann> python experience of course is also helpful, in case you need to extend some theme code, which is rather likely
357 2011-03-20T20:13:10 <Dae-ekleN> the question is how many chances i have to be accepted and what should i do? do you have any tasks?
358 2011-03-20T20:13:48 <ThomasWaldmann> well, you should show that you can do what you propose as project
359 2011-03-20T20:14:05 <ThomasWaldmann> give links to own stuff in your application
360 2011-03-20T20:14:18 <ThomasWaldmann> get moin2 and play with it, fix some stuff
361 2011-03-20T20:14:59 <Dae-ekleN> what do you mean by own stuff?
362 2011-03-20T20:15:26 <dreimark> well you may be have contributed to some project
363 2011-03-20T20:15:29 <ThomasWaldmann> well, e.g. if you designed some web site or contributed something to an OSS project
364 2011-03-20T20:16:19 <Dae-ekleN> i see, i have such. thank you
365 2011-03-20T20:16:34 <ThomasWaldmann> most important is maybe to get into moin2, so that you can develop own ideas about how a theme should look like
366 2011-03-20T20:17:37 <dreimark> http://moinmo.in/MoinMoin2.0
367 2011-03-20T20:17:43 <ThomasWaldmann> and if you're not much into js yet, maybe read some stuff about it (and about jquery)
368 2011-03-20T20:18:39 <dreimark> besides that it is coke http://www.romancortes.com/blog/pure-css-coke-can/
369 2011-03-20T20:18:48 <Dae-ekleN> i'm also familiarized with it, but have no previous experience
370 2011-03-20T20:19:57 <ThomasWaldmann> do you know some other templating language for html?
371 2011-03-20T20:23:55 <Dae-ekleN> i don't
372 2011-03-20T20:24:43 <ThomasWaldmann> ok. just have a look at the templates we have already. if you know python, you should feel at home with jinja2.
373 2011-03-20T20:25:17 <Dae-ekleN> ok, thanks
374 2011-03-20T20:27:43 <dreimark> which nick has MarkusMajer ?
375 2011-03-20T20:27:59 <Dae-ekleN> sorry, missed it: after familiarising with python i also understood that php is not so prfect
376 2011-03-20T20:30:21 <ThomasWaldmann> dreimark: mpathy
377 2011-03-20T20:30:42 <dreimark> ok
378 2011-03-20T20:33:07 <CIA-58> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 149:601b8be0ced6 2.0/MoinMoin/ (6 files in 3 dirs): rename editbar to itemviews (template, css names)
379 2011-03-20T20:42:24 <ThomasWaldmann> for text in self.cfg.navi_bar:
380 2011-03-20T20:42:24 <ThomasWaldmann> url, link_text, title = self.split_navilink(text)
381 2011-03-20T20:42:24 <ThomasWaldmann> items.append(('wikilink', url, link_text, title))
382 2011-03-20T20:42:43 <ThomasWaldmann> this is how the configured navi_bar is processed right now
383 2011-03-20T20:43:20 <ThomasWaldmann> how about just configuring a list of tuples (css_class, url, link_text, title)?
384 2011-03-20T20:43:48 <ThomasWaldmann> that's quicker and more flexible. but also maybe a bit harder.
385 2011-03-20T20:44:30 *** cdom
386 2011-03-20T20:45:55 <ThomasWaldmann> the urls could be created using url_for(...)
387 2011-03-20T20:47:45 <Dae-ekleN> i'll try to do smth with it, thank you
388 2011-03-20T20:53:27 <ThomasWaldmann> hmm, url_for doesn't work in config
389 2011-03-20T20:53:31 <ThomasWaldmann> too early
390 2011-03-20T20:57:18 *** cdom
391 2011-03-20T20:59:05 <ThomasWaldmann> ok, but we can give endpoint name and args dict, so that would work
392 2011-03-20T21:01:42 <ThomasWaldmann> ('wikilink', 'frontend.global_history', {}, 'History', ''),
393 2011-03-20T21:05:26 <ThomasWaldmann> hmm, that just works for endpoints of course, not for arbitrary urls
394 2011-03-20T21:10:38 <ThomasWaldmann> but we could always have and endpoint that just does a redirect to some arbitrary url given as url arg
395 2011-03-20T21:10:44 <ThomasWaldmann> an*
396 2011-03-20T21:22:39 <Dae-ekleN> i'm trying to get general idea about how code works now
397 2011-03-20T21:22:53 <MattMaker> ugh, so when I want to merge my clone of a repo (pull in new stuff from the original), what is the most sensible way to do it, and how can I remember how to do it?
398 2011-03-20T21:23:30 <MattMaker> once I figure it out I'm going to make it an alias or something
399 2011-03-20T21:24:51 <ThomasWaldmann> Dae-ekleN: there is a rough description in the moin2 docs, linked from MoinMoin2.0 page
400 2011-03-20T21:25:28 <dreimark> http://paste.pocoo.org/show/356792/
401 2011-03-20T21:25:42 <Dae-ekleN> i found it, thanks
402 2011-03-20T21:25:44 <ThomasWaldmann> MattMaker: you pull it into your (local) repo, merge if necessary, push it to your remote repo
403 2011-03-20T21:25:59 <dreimark> that solves half of the middle mouse insert into the search field
404 2011-03-20T21:26:32 <ThomasWaldmann> if you did not publish your changes yet, you can also hg rebase instead of hg merge - but you must be rather sure about that :)
405 2011-03-20T21:26:48 <dreimark> pasting content using the middle mouse button works with that
406 2011-03-20T21:27:01 <MattMaker> ThomasWaldmann is there a way to do it on BitBucket directly?
407 2011-03-20T21:27:13 <ThomasWaldmann> you don't need
408 2011-03-20T21:27:26 <ThomasWaldmann> just use your local one and push afterwards
409 2011-03-20T21:27:29 <MattMaker> well, I have some changes I'm not ready to commit to my local
410 2011-03-20T21:27:50 <MattMaker> but I want the new stuff pulled across
411 2011-03-20T21:28:10 <MattMaker> and doing a new checkout just for that purpose seems a bit much
412 2011-03-20T21:28:24 <ThomasWaldmann> you could hg diff >mychanges ; hg revert --all ; hg pull ... ; ..... ; patch -p1 <mychanges
413 2011-03-20T21:28:56 <MattMaker> ah, thanks, I think that will work
414 2011-03-20T21:29:05 <xorAxAx> hg shelve
415 2011-03-20T21:29:08 <xorAxAx> does that
416 2011-03-20T21:29:11 <MattMaker> ooh
417 2011-03-20T21:29:59 <MattMaker> is that an extension?
418 2011-03-20T21:30:06 * ThomasWaldmann uses a risc command set :P
419 2011-03-20T21:30:22 <xorAxAx> MattMaker: yes
420 2011-03-20T21:30:37 <xorAxAx> but its shipped, you only need to enable it
421 2011-03-20T21:31:01 <dreimark> http://mercurial.selenic.com/wiki/RecordExtension
422 2011-03-20T21:31:03 <MattMaker> probably easier for me to remember than the diff/patch
423 2011-03-20T21:32:09 <MattMaker> dreimark I definitely have too many changes for RecordExtension to work well
424 2011-03-20T21:35:03 <MattMaker> http://mercurial.selenic.com/wiki/TransplantExtension , too. so many ways to do things, so many ways to get a repo tangled up
425 2011-03-20T21:36:53 <MattMaker> xorAxAx http://mercurial.selenic.com/wiki/ShelveExtension says it's not shipped
426 2011-03-20T21:37:03 <xorAxAx> MattMaker: oh, ok
427 2011-03-20T21:37:26 <xorAxAx> i have so many exts enabled, i dont really remember where they come from :)
428 2011-03-20T21:40:25 <MattMaker> and there are like three more of these
429 2011-03-20T21:40:33 <MattMaker> all doing the same thing a little differently
430 2011-03-20T21:42:17 <ThomasWaldmann> it is maybe too easy to write hg extensions :P+
431 2011-03-20T21:47:19 <sinha_> ThomasWaldmann: i want to implement textcha for copy thing ,but i am not seeing any textcha in my local copy...where can i see it ?
432 2011-03-20T21:48:21 <ThomasWaldmann> you need to enable them in your wiki config
433 2011-03-20T21:48:32 <ThomasWaldmann> see the hints at the end of wikiconfig.py
434 2011-03-20T21:50:10 <sinha_> ThomasWaldmann: i have to add a item for that ..something like TEXTCHA='true' ?
435 2011-03-20T21:50:27 <sinha_> ThomasWaldmann: in the wikiconfig.py
436 2011-03-20T21:50:43 <ThomasWaldmann> http://paste.pocoo.org/show/356806/
437 2011-03-20T21:52:06 *** m4k3r
438 2011-03-20T21:56:04 <sinha_> ThomasWaldmann: I have added these lines..and restarted the server but still doing rename/delete isnt asking for captcha
439 2011-03-20T22:01:33 <sinha_> dreimark: i have tried to enable textcha in wikiconfig.py file but i am not seeing it anywhere on the pages..when i do rename/delete ? what could be problem ?
440 2011-03-20T22:04:49 <dreimark> sinha_: can you pastebin your changes please
441 2011-03-20T22:05:44 <sinha_> dreimark: added these lines in wikiconfig.py http://paste.pocoo.org/show/356806/ suggested by ThomasWaldmann
442 2011-03-20T22:06:35 <xorAxAx> sinha_: with the indentation?
443 2011-03-20T22:06:54 <CIA-58> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 150:d3deda1731fb 2.0/ (4 files in 4 dirs):
444 2011-03-20T22:06:54 <CIA-58> configure navi_bar using tuples
445 2011-03-20T22:06:54 <CIA-58> saves us from first having to parse to stuff - faster
446 2011-03-20T22:06:54 <CIA-58> it is also more flexible as one can easily link to any endpoint with any params
447 2011-03-20T22:06:54 <CIA-58> move the global views from item_views to default navi_bar
448 2011-03-20T22:06:55 <CIA-58> remove now wrong navi_bar docs
449 2011-03-20T22:07:13 <sinha_> xorAxAx: yes , with 4 spaces
450 2011-03-20T22:08:24 <dreimark> sinha_: and where ?
451 2011-03-20T22:08:33 <dreimark> can you show the whole file
452 2011-03-20T22:09:03 <dreimark> or a diff, because that shows some surrounding lines
453 2011-03-20T22:11:48 <dreimark> was that at the end added ?
454 2011-03-20T22:11:58 <sinha_> dreimark: http://paste.pocoo.org/show/356822/ here is the file
455 2011-03-20T22:12:21 <sinha_> dreimark: added after SECRET_KEY
456 2011-03-20T22:12:29 <dreimark> sinha_: class Config(DefaultConfig):
457 2011-03-20T22:12:38 <dreimark> add it to the class
458 2011-03-20T22:13:02 <sinha_> dreimark: okay,adding..
459 2011-03-20T22:13:43 <dreimark> and then if it works undo it and read 99
460 2011-03-20T22:13:47 <dreimark> and do it that way
461 2011-03-20T22:14:09 <dreimark> read 96
462 2011-03-20T22:15:31 <sinha_> dreimark: yes will add to separate config file
463 2011-03-20T22:16:47 <CIA-58> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 151:704a9b888fe9 2.0/MoinMoin/translations/ (MoinMoin.pot de/LC_MESSAGES/messages.po): update i18n
464 2011-03-20T22:16:59 <sinha_> dreimark: okay working ! now where is the copy option ?
465 2011-03-20T22:18:08 <dreimark> you see the wiki ?
466 2011-03-20T22:19:29 <sinha_> dreimark: that repo one ?
467 2011-03-20T22:19:58 <dreimark> may be i don't get the question
468 2011-03-20T22:20:08 <dreimark> do you see it on your local wiki now?
469 2011-03-20T22:20:14 <dreimark> is the wiki started?
470 2011-03-20T22:20:34 <dreimark> or do you ask where the code of it is located?
471 2011-03-20T22:21:49 <sinha_> dreimark: NO..only this much option is coming # Show # Highlight # Meta # Modify # Download # History # Add Link # Comments # Index # Discussion and in that select box nothing is like copy
472 2011-03-20T22:22:02 <sinha_> dreimark: in that edit bar
473 2011-03-20T22:22:46 <sinha_> dreimark: and yes where is code located...is this will be helpful http://hg.moinmo.in/moin/2.0-dev/rev/b0c92a806602
474 2011-03-20T22:23:28 <dreimark> sinha_: this is the old repo where the gci student has worked on
475 2011-03-20T22:23:39 <dreimark> you have to look for similiar code for copy in 2.0
476 2011-03-20T22:24:26 <dreimark> if you look in MoinMoin/templates/
477 2011-03-20T22:24:33 <dreimark> you will find the copy template
478 2011-03-20T22:24:48 <ThomasWaldmann> btw, we need some piece of js now that collapses/expands the item_views. we have something similar for quicklinks, but the code is too specialized, so it only works for that afaics
479 2011-03-20T22:25:15 <dreimark> sinha_: and you will find the other places for copy too
480 2011-03-20T22:25:25 <dreimark> and you learn more if you look into it first
481 2011-03-20T22:26:08 <sinha_> dreimark :yes i am seeing this..but what i am asking is there is no copy option coming on the wiki pages ?
482 2011-03-20T22:26:53 <dreimark> it is on mine
483 2011-03-20T22:27:03 <dreimark> * Show
484 2011-03-20T22:27:03 <dreimark> * Highlight
485 2011-03-20T22:27:03 <dreimark> * Meta
486 2011-03-20T22:27:03 <dreimark> * Download
487 2011-03-20T22:27:03 <dreimark> * History
488 2011-03-20T22:27:06 <dreimark> * Comments
489 2011-03-20T22:27:08 <dreimark> * Index
490 2011-03-20T22:27:11 <dreimark> * Discussion
491 2011-03-20T22:27:13 <dreimark> * What refers here?
492 2011-03-20T22:27:16 <dreimark> * Local Site Map
493 2011-03-20T22:27:18 <dreimark> * Items with similar names
494 2011-03-20T22:27:21 <dreimark> * *
495 2011-03-20T22:27:23 <dreimark> * Modify
496 2011-03-20T22:27:26 <dreimark> * Copy
497 2011-03-20T22:27:28 <dreimark> * Rename
498 2011-03-20T22:27:31 <dreimark> * Delete
499 2011-03-20T22:27:33 <dreimark> * Destroy
500 2011-03-20T22:27:36 <dreimark> ops
501 2011-03-20T22:29:03 <xorAxAx> dreimark: you should have parted
502 2011-03-20T22:29:06 <xorAxAx> temporarily
503 2011-03-20T22:29:29 <sinha_> dreimark: i have all these but not the copy link
504 2011-03-20T22:30:30 <ThomasWaldmann> sinha_: if you repo is not current, it is excluded
505 2011-03-20T22:30:53 <ThomasWaldmann> maybe do a hg pull -u to have the latest code (i removed the "copy" exclude)
506 2011-03-20T22:31:18 <dreimark> xorAxAx: i forgot that it is a list
507 2011-03-20T22:31:19 <sinha_> ThomasWaldmann : okay..or any edition in edit_bar config file may also work ?
508 2011-03-20T22:31:30 <dreimark> sinha_: hg pull -u
509 2011-03-20T22:32:01 <dreimark> you should train to use these tools
510 2011-03-20T22:32:45 <sinha_> dreimark: Now a traceback error comes..due to change in user.py file
511 2011-03-20T22:33:21 <sinha_> dreimark ThomasWaldmann http://paste.pocoo.org/show/356841/
512 2011-03-20T22:34:10 <dreimark> sinha_: you didn't read completly all from line 96
513 2011-03-20T22:34:20 <dreimark> of wikiconfig.py
514 2011-03-20T22:34:32 <sinha_> dreimark: ah i was just doing that..
515 2011-03-20T22:35:45 <dreimark> xorAxAx: and i should have read as it tells 17lines ...
516 2011-03-20T22:36:06 <xorAxAx> dreimark: but as i said, you can still cancel the paste by parting
517 2011-03-20T22:36:30 <CIA-58> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 152:06223ed7d72b 2.0/MoinMoin/ (config/default.py templates/itemviews.html): do not show 'modify' in itemviews for nonexisting item, remove delimiter (looks ugly if there is nothing afterwards)
518 2011-03-20T22:45:31 <ThomasWaldmann> ok, votes for leftmost places in item_views (aka edit_bar)
519 2011-03-20T22:45:57 <ThomasWaldmann> Show - i think this must be leftmost due to logical reasons, although a bit redundant sometimes
520 2011-03-20T22:47:38 <ThomasWaldmann> so, 1. place = show
521 2011-03-20T22:48:55 <ThomasWaldmann> 2. place = download (because it is another fundamental way of "getting the item")
522 2011-03-20T22:50:03 <ThomasWaldmann> 3. place = supplementation items (a read only viewer might want to look at Discussion page)
523 2011-03-20T22:50:35 <ThomasWaldmann> 4. place = item history (read only viewer still)
524 2011-03-20T22:51:09 <ThomasWaldmann> 5. place = modify (rw user)
525 2011-03-20T22:52:53 <sinha_> dreimark: in this line " forms.render_textcha(gen, form) " form where does this 'form' variable is coming ?
526 2011-03-20T22:53:32 <sinha_> dreimark: *from where does this 'form' variable is coming ?
527 2011-03-20T22:53:38 <dreimark> sinha_: url
528 2011-03-20T22:53:52 <dreimark> http://hg.moinmo.in/moin/2.0 based on
529 2011-03-20T22:55:00 <dreimark> ThomasWaldmann: i don't like the supplementation there at all
530 2011-03-20T22:55:27 <dreimark> it is nothing more than a special subpage
531 2011-03-20T22:55:34 <dreimark> ehem item
532 2011-03-20T22:55:58 <ThomasWaldmann> well, if you do not use suppl subitems, it won't show up anyway
533 2011-03-20T22:56:25 <ThomasWaldmann> but if you use it for discussion subitem, should it be rather near the original item?
534 2011-03-20T22:56:47 <dreimark> it should be near to the item name
535 2011-03-20T22:57:04 <dreimark> or there should be something to go to other subitems
536 2011-03-20T22:57:45 <sinha_> dreimark: but in file copy.html it saying that form is undefined ?
537 2011-03-20T22:58:28 <dreimark> ThomasWaldmann: s/"What refers here?"/"referers"/ because it is the only one as a sentence
538 2011-03-20T22:58:55 <ThomasWaldmann> ok
539 2011-03-20T22:59:12 <dreimark> sinha_: look at render_template
540 2011-03-20T23:00:00 <dreimark> of that changeset http://hg.moinmo.in/moin/2.0-dev/rev/b0c92a806602 - i show it in that repo because we have rebooted the repo after that
541 2011-03-20T23:00:36 <dreimark> sinha_: by the render_template you can submit paramters to the template
542 2011-03-20T23:02:27 <dreimark> sinha_: if you look at http://hg.moinmo.in/moin/2.0/file/06223ed7d72b/MoinMoin/apps/frontend/views.py#l353 it is missing
543 2011-03-20T23:03:40 <MattMaker> "what refers here" is a rather unusual contortion in English, I am trying to think what else to suggest
544 2011-03-20T23:04:06 <dreimark> ThomasWaldmann: what other "Site Map" than Local can we have? otherwise we can shorten it
545 2011-03-20T23:06:08 <dreimark> "Items with similar names" is also too long
546 2011-03-20T23:06:41 <dreimark> we can have tooltips too. if it needs longer description
547 2011-03-20T23:06:51 <ThomasWaldmann> dreimark: it is called local sitemap because it shows the site from the view of the current item
548 2011-03-20T23:07:03 <ThomasWaldmann> we already have them :)
549 2011-03-20T23:07:52 <ThomasWaldmann> hmm, no, not there, but in navibar
550 2011-03-20T23:08:17 <dreimark> I assumed but it did not work
551 2011-03-20T23:08:23 <dreimark> ah yes
552 2011-03-20T23:08:48 <MattMaker> https://secure.wikimedia.org/wikipedia/en/wiki/Cross-reference <- for a plural noun describing a set, perhaps "cross-references", "xrefs", "references", or for a verb phrase, "find links to", "references", "links to this", "links to here" "find xrefs"
553 2011-03-20T23:09:07 <MattMaker> actually "What refers here?" is starting to sound better
554 2011-03-20T23:09:43 <CIA-58> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 153:12d0ff564156 2.0/MoinMoin/config/default.py: item_views reordered, first try
555 2011-03-20T23:10:18 <MattMaker> "lexicography", lol
556 2011-03-20T23:10:25 <ThomasWaldmann> MattMaker: it is not only links, but also transclusions (embeddings), thus the more general term "refer"
557 2011-03-20T23:10:47 <ThomasWaldmann> but Referrers sounds good to me also
558 2011-03-20T23:11:13 <MattMaker> what about "Backlinks"?
559 2011-03-20T23:11:20 <MattMaker> that's what the code calls them!
560 2011-03-20T23:11:53 <MattMaker> https://secure.wikimedia.org/wikipedia/en/wiki/Backlinks
561 2011-03-20T23:12:19 <MattMaker> Referrers might confuse some people since it has an HTTP header meaning
562 2011-03-20T23:12:31 <sinha_> dreimark : what is RevisionNumberMismatchError ?
563 2011-03-20T23:12:44 <sinha_> dreimark : How to resolve it
564 2011-03-20T23:13:29 <ThomasWaldmann> MattMaker: no, as the http header is spelled referer, haha
565 2011-03-20T23:13:56 <MattMaker> well, I thought they revised it so that both spellings were valid headers :)
566 2011-03-20T23:14:06 <ThomasWaldmann> oh, really?
567 2011-03-20T23:14:40 <ThomasWaldmann> but well, in a wiki sense, it is the equivalent meaning somehow. not for the current request of course.
568 2011-03-20T23:14:44 *** cdom
569 2011-03-20T23:15:38 <MattMaker> https://secure.wikimedia.org/wikipedia/en/wiki/Referer says kind of yes, depending where it is found
570 2011-03-20T23:19:30 <MattMaker> grr why is my merge "Unable to diff "
571 2011-03-20T23:20:40 <MattMaker> and why is there not an "hg undo"
572 2011-03-20T23:20:53 <MattMaker> surely that would be an obvious thing to have
573 2011-03-20T23:22:12 <RogerHaase> how about References?
574 2011-03-20T23:24:35 <dreimark> sinha_: where do you see it?
575 2011-03-20T23:24:56 <sinha_> ThomasWaldmann: while copying the item..
576 2011-03-20T23:25:01 *** ErSame
577 2011-03-20T23:25:09 <sinha_> dreimark: while copying the item
578 2011-03-20T23:25:17 <sinha_> dreimark: Textcha is working fine
579 2011-03-20T23:25:29 <sinha_> dreimark: something wrong with tem copy module
580 2011-03-20T23:25:58 <sinha_> dreimark: its saying " RevisionNumberMismatchError: You cannot create a revision with revno 0. Your revno must be greater than the item's last revision, which is 3. "
581 2011-03-20T23:26:42 <dreimark> sinha_: this is a different problem
582 2011-03-20T23:26:53 <dreimark> an intersting student can fix
583 2011-03-20T23:27:16 <sinha_> dreimark: tell me i would like to do this
584 2011-03-20T23:27:25 <dreimark> first do the first one
585 2011-03-20T23:27:44 <dreimark> because doing different things in one patch is bad working style
586 2011-03-20T23:27:47 <sinha_> dreimark :i guess thats done..its coming on copy page
587 2011-03-20T23:27:58 <dreimark> if you are ready show a paste bin
588 2011-03-20T23:28:05 *** RogerHaase
589 2011-03-20T23:28:09 <dreimark> with your diff
590 2011-03-20T23:28:21 <sinha_> dreimark: ohk
591 2011-03-20T23:28:24 <dreimark> or run py.test beforehand
592 2011-03-20T23:28:46 <dreimark> that will show pep8 failures too and we don't have to tell
593 2011-03-20T23:34:23 <sinha_> dreimark: this is the diff http://paste.pocoo.org/show/356877/ ... I did py.test MoinMoin/apps/frontend/views.py it didnt show any error..currently running py.test for all
594 2011-03-20T23:35:46 <sinha_> dreimark: i guess i changed some space/lines in other files also will rectify that if this is fine
595 2011-03-20T23:36:27 <CIA-58> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 154:877c264a766d 2.0/MoinMoin/ (config/default.py templates/itemviews.html): item_views: add support for titles
596 2011-03-20T23:36:46 <ThomasWaldmann> dreimark: check english version of it
597 2011-03-20T23:37:03 <sinha_> dreimark: py.test ended with this line " 51 failed, 2289 passed, 25 skipped, 4 error, 7 xpassed in 297.75 seconds "
598 2011-03-20T23:37:33 <dreimark> sinha_: remove all not needed changes and show a new pastebin
599 2011-03-20T23:37:46 <sinha_> dreimark: yes
600 2011-03-20T23:38:03 <dreimark> and if you yet not have configured hg you should do it
601 2011-03-20T23:38:24 <dreimark> later on we want to have an exported changeset
602 2011-03-20T23:38:28 <dreimark> with your name in it
603 2011-03-20T23:38:41 <sinha_> dreimark: okay will do it
604 2011-03-20T23:38:57 *** ErSame
605 2011-03-20T23:38:58 <dreimark> and look at indenting
606 2011-03-20T23:39:14 <dreimark> if you add a line keep the indenting
607 2011-03-20T23:39:17 <dreimark> e.g. line 13
608 2011-03-20T23:39:26 <dreimark> and also line 20
609 2011-03-20T23:39:49 <dreimark> and never never use tabs
610 2011-03-20T23:40:20 <dreimark> hmm, we need to check the pep8 checker
611 2011-03-20T23:42:44 <MattMaker> "Unable to diff" it still says. can I just delete the local copy of this file it is complaining about?
612 2011-03-20T23:45:19 <dreimark> MattMaker: i sometime do hg revert --all
613 2011-03-20T23:45:24 <dreimark> and then a hg pull -u
614 2011-03-20T23:45:35 <MattMaker> that failed due to "uncommitted" somethings
615 2011-03-20T23:45:47 <dreimark> the revert too?
616 2011-03-20T23:45:48 <MattMaker> finally, hg update --clean worked
617 2011-03-20T23:45:50 <MattMaker> yes
618 2011-03-20T23:46:00 <MattMaker> or rather, it wanted a specific revision for revert
619 2011-03-20T23:46:36 <CIA-58> Thomas Waldmann <tw AT waldmann-edv DOT de> default * 155:5b8c72be936d 2.0/MoinMoin/translations/ (MoinMoin.pot de/LC_MESSAGES/messages.po): updated i18n
620 2011-03-20T23:47:14 <MattMaker> wait, I did hg merge again and it fails again
621 2011-03-20T23:47:54 * ThomasWaldmann likes the menuless theme
622 2011-03-20T23:48:05 <ThomasWaldmann> finally a reason for a wide screen
623 2011-03-20T23:48:53 <ThomasWaldmann> MattMaker: before you push, make sure it doesn't push crap (hg out)
624 2011-03-20T23:48:57 <MattMaker> hmm, it actually does look like rebase is the thing that does what I am trying to do
625 2011-03-20T23:50:12 <dreimark> http://test.moinmo.in/MyMoinMoin
626 2011-03-20T23:50:39 <dreimark> the two item bars are not aligned the same
627 2011-03-20T23:52:09 <dreimark> ThomasWaldmann: because of the 3 and 4 selection/counting of our brain we should move important features also to the end
628 2011-03-20T23:52:26 <dreimark> (thesheep explained at EP)
629 2011-03-20T23:52:45 <MattMaker> ThomasWaldmann yes, hg out shows some crap
630 2011-03-20T23:53:23 <dreimark> Meta at the right
631 2011-03-20T23:54:05 <MattMaker> and again it's at "Unable to diff" "merging ... failed!"
632 2011-03-20T23:54:14 <sinha_> dreimark: this is the diff http://paste.pocoo.org/show/356883/ and also i have created a patch file
633 2011-03-20T23:54:16 <MattMaker> I'm glad I made a backup
634 2011-03-20T23:55:38 <ThomasWaldmann> MattMaker: you have a clean workdir when trying to merge?
635 2011-03-20T23:56:46 <MattMaker> I am trying to get your repo into mine so that it's up to date. I don't really care how, I just want to do it
636 2011-03-20T23:57:16 <MattMaker> it was a supposedly clean workdir, but it didn't like that it was a different repo
637 2011-03-20T23:57:38 <dreimark> Show Download History Discussion Index Comments Highlight Referrers Site Map Similar Highlight Copy Rename Delete Destroy Meta Modify
638 2011-03-20T23:58:02 <dreimark> Show Download History Discussion Index Comments Highlight Referrers Site Map Similar Copy Rename Delete Destroy Meta Modify
639 2011-03-20T23:59:02 <dreimark> sinha_: still tabs
640 2011-03-20T23:59:10 <dreimark> look at line 14
641 2011-03-20T23:59:28 <dreimark> line 21
642 2011-03-20T23:59:59 <sinha_> dreimark: oh..rectifying these