1 2012-02-13T00:00:17 <dreimark> the former one was integrated in google docs
2 2012-02-13T00:00:28 <ThomasWaldmann> that syncing algorithm should maybe work for arbitrary stuff (byte sequences), i don't see why it should be limited to text
3 2012-02-13T00:00:55 <ThomasWaldmann> also, we need multiple synced stuff (data and metadata)
4 2012-02-13T00:01:29 <dreimark> the olpc xo has also such a sync mechanism. if we would have gci this could be a research task
5 2012-02-13T00:02:30 <dreimark> we need multiple syncs at the same time
6 2012-02-13T00:03:08 <ThomasWaldmann> that sync paper is beyond pre-university students
7 2012-02-13T00:06:17 * dreimark reads FAQ
8 2012-02-13T00:12:35 <dreimark> Markdown support like pagedown (+1 rspeer), but making something that's both Markdown and live-editable is going to be interesting from an interface point of view. Not impossible, but interesting.
9 2012-02-13T00:12:48 <dreimark> https://github.com/Pita/etherpad-lite/wiki/TODO
10 2012-02-13T00:13:08 <dreimark> that goes in at least example for wordproc.
11 2012-02-13T00:15:32 <dreimark> it makes not really sense to write that again
12 2012-02-13T00:17:28 <dreimark> *** officially announce EuroPython 2012, that will be held in Florence, Italy, on July 2nd-8th!
13 2012-02-13T00:18:00 <dreimark> gn
14 2012-02-13T00:21:39 <ThomasWaldmann> gn dreimark
15 2012-02-13T00:23:49 *** moin3
16 2012-02-13T00:23:53 *** moin3
17 2012-02-13T01:21:39 <ThomasWaldmann> http://opencoweb.org/ maybe interesting
18 2012-02-13T01:51:16 <ThomasWaldmann> https://github.com/phrearch/py-infinote
19 2012-02-13T02:23:43 *** helixfelix1
20 2012-02-13T02:25:47 *** helixfelix
21 2012-02-13T02:26:10 <ThomasWaldmann> http://en.wikipedia.org/wiki/Operational_transformation
22 2012-02-13T02:48:46 *** ashlurgen
23 2012-02-13T02:55:04 *** ashlurgen2
24 2012-02-13T03:06:22 *** yufra_
25 2012-02-13T03:31:21 <ThomasWaldmann> http://www.youtube.com/watch?v=84zqbXUQIHc
26 2012-02-13T03:56:39 *** yufra_
27 2012-02-13T03:58:30 *** sudo_dirk
28 2012-02-13T04:26:54 *** CIA-64
29 2012-02-13T04:27:29 *** CIA-41
30 2012-02-13T04:53:48 *** ashlurgen
31 2012-02-13T05:24:34 *** yufra_
32 2012-02-13T05:27:05 *** ashlurgen2
33 2012-02-13T06:05:39 *** sudo_dirk
34 2012-02-13T06:55:52 *** yufra_
35 2012-02-13T07:35:46 *** ashlurgen
36 2012-02-13T09:56:54 *** greg_f
37 2012-02-13T10:07:16 *** qxcv
38 2012-02-13T10:47:52 <dreimark> moin
39 2012-02-13T10:56:24 *** aa_
40 2012-02-13T10:56:26 *** yufra
41 2012-02-13T10:56:27 *** vpv
42 2012-02-13T10:56:28 *** xorAxAx
43 2012-02-13T10:59:52 *** aa_
44 2012-02-13T11:00:31 *** yufra
45 2012-02-13T11:00:31 *** vpv
46 2012-02-13T11:00:31 *** xorAxAx
47 2012-02-13T11:38:06 *** yufra
48 2012-02-13T11:38:06 *** vpv
49 2012-02-13T11:38:07 *** xorAxAx
50 2012-02-13T11:39:35 *** yufra
51 2012-02-13T11:39:36 *** vpv
52 2012-02-13T11:39:36 *** xorAxAx
53 2012-02-13T11:59:54 *** helixfelix1
54 2012-02-13T13:47:29 *** dave_largo
55 2012-02-13T14:43:49 *** dave_largo
56 2012-02-13T15:06:50 *** qxcv
57 2012-02-13T15:11:23 *** jek
58 2012-02-13T15:18:51 *** dave_largo
59 2012-02-13T15:57:46 *** yufra_
60 2012-02-13T16:16:12 *** RogerHaase
61 2012-02-13T16:23:27 *** chuck1
62 2012-02-13T16:25:00 <chuck1> Hello! Which opportunities are there to share authentication with other web applicatons (eg squirrlemail, oder java based 'JabberChat Candy', or phpBB forum)
63 2012-02-13T16:25:41 <ThomasWaldmann> hi chuck1
64 2012-02-13T16:26:46 <ThomasWaldmann> in general, if you directly want to share (kind of peer-to-peer), you need to know the cookie format of one system and make the other system trust and reuse information from that cookie
65 2012-02-13T16:27:31 <ThomasWaldmann> the alternative is to use some centralized thing, like ldap, kerberos, openid, etc.
66 2012-02-13T16:28:17 <ThomasWaldmann> you need to differentiate whether you want single-sign-on or just want to use same username and password (but still log on into each system separately)
67 2012-02-13T16:29:50 <chuck1> yes, the better thing would be a single-sign-on of course but only with ldap this will not work, or? Is this possible with apache&ldap ?
68 2012-02-13T16:30:19 <chuck1> I also mean only logins hosted by apache
69 2012-02-13T16:31:38 *** yaMatt
70 2012-02-13T16:32:25 <yaMatt> hi guys, I'm writing a couchdb store for moin2.0 and I'm having difficulties with the tests
71 2012-02-13T16:32:55 <yaMatt> I'm not so familiar with pytest so it's more than likely I'm missing something but I get an error here: http://hg.moinmo.in/moin/2.0/file/bf3ef935aced/MoinMoin/storage/stores/_tests/conftest.py#l96
72 2012-02-13T16:33:08 <ThomasWaldmann> chuck1: this is rather a apache question. as far as moin concerned i can say that moin can reuse stuff done by apache using the GivenAuth auth method of moin.
73 2012-02-13T16:33:30 <yaMatt> it looks like it's pulling out the param from the store but most stores don't have it
74 2012-02-13T16:33:35 <ThomasWaldmann> ronny: ^^ see what yamatt said
75 2012-02-13T16:34:43 <yaMatt> cheers :)
76 2012-02-13T16:37:19 <ThomasWaldmann> it seems to relate to line 57
77 2012-02-13T16:37:41 <ronny> oO
78 2012-02-13T16:38:12 <ThomasWaldmann> too much magic for the non-pytest-magicians
79 2012-02-13T16:38:21 <yaMatt> heheh
80 2012-02-13T16:38:46 <ronny> yaMatt: http://hg.moinmo.in/moin/2.0/file/bf3ef935aced/MoinMoin/storage/stores/_tests/conftest.py#l52 seys param up
81 2012-02-13T16:38:55 <ronny> the funcarg uses it to instanciate stuff
82 2012-02-13T16:39:47 <yaMatt> so it's pulling out the variables from the modules and trying to match them?
83 2012-02-13T16:44:03 <ronny> yaMatt: not sure what you mean
84 2012-02-13T16:45:25 <ronny> yaMatt: basically it iterates over stores, anc which classes to use, and generates them as parameters for a test
85 2012-02-13T16:45:47 <ronny> ThomasWaldmann: this would be much simpler with the more recent pytest i think
86 2012-02-13T16:45:53 <ronny> (parametrize is awesome)
87 2012-02-13T16:46:26 <ThomasWaldmann> which release do you mean?
88 2012-02-13T16:47:26 <yaMatt> ronny, so I need to set my arg names for the classes to match the right store names?
89 2012-02-13T16:47:38 *** chuck1
90 2012-02-13T16:48:21 <ronny> yaMatt: not sure what you mean, you just need to add yout store to the list, then it will run the tests
91 2012-02-13T16:49:06 <yaMatt> I added it, but it was causing errors
92 2012-02-13T16:49:27 <yaMatt> on line 96
93 2012-02-13T16:49:28 <ronny> can you paste your changes somewhere?
94 2012-02-13T16:49:46 <ThomasWaldmann> maybe put the whole new store stuff on codereview
95 2012-02-13T16:50:01 <yaMatt> sadly not right now, I just had a few minutes while at work, I can send it over later
96 2012-02-13T16:50:16 <ronny> yaMatt: unless you wrote a custom test, that lacks the right param names, that shouldnt happen
97 2012-02-13T16:50:23 <yaMatt> I was just wondering how concrete the tests were or if I was missing something
98 2012-02-13T16:50:48 <yaMatt> I used the sqlite one as my template, I don't think I changed too much
99 2012-02-13T16:51:39 <ronny> yaMatt: th decorators are essential
100 2012-02-13T16:51:57 <yaMatt> definitely didn't touch those
101 2012-02-13T16:52:25 <yaMatt> I added my store to the constructors list, but I've just noticed a STORES variable, I'll try adding it in there tonight
102 2012-02-13T16:54:00 <ronny> unless i see code, i cant tell what exactly is messed up
103 2012-02-13T16:54:34 <ronny> btw, what lib are you using to access couchdb?
104 2012-02-13T16:55:31 <yaMatt> sure
105 2012-02-13T16:55:41 <yaMatt> the python-couchdb module
106 2012-02-13T16:56:06 <ronny> ouch
107 2012-02-13T16:56:11 <yaMatt> how comes?
108 2012-02-13T16:56:11 <ronny> that lib is kind of bad/dead
109 2012-02-13T16:56:24 <ronny> couchdbkit works better
110 2012-02-13T16:57:00 <yaMatt> it changes so much, I know that python-couchdb is in most repos
111 2012-02-13T17:28:46 *** yufra_
112 2012-02-13T17:32:43 *** yufra_
113 2012-02-13T18:13:43 <RogerHaase> ThomasWaldmann: a fix for issue 80, where transclusion loop is only detected if the starting page is part of the loop: http://codereview.appspot.com/5651088/
114 2012-02-13T18:16:49 <ThomasWaldmann> hi RogerHaase
115 2012-02-13T18:16:52 * ThomasWaldmann looks
116 2012-02-13T18:24:48 <ThomasWaldmann> RogerHaase: done. no idea about your question on the left side.
117 2012-02-13T18:28:49 <RogerHaase> my comment is wrong: page_href_new != page_href is NEVER (!always) true. So the stack has the starting page followed by a bunch of None entgries.
118 2012-02-13T18:29:55 <RogerHaase> Thus, if the first page is not part of the loop, the loop is not detected.
119 2012-02-13T18:30:26 <RogerHaase> entgries/entries
120 2012-02-13T18:34:47 <ThomasWaldmann> ah
121 2012-02-13T18:36:55 <RogerHaase> The resulting error message looks like: Transclusion loop error via: /page2, /page3, /page4, and /page2
122 2012-02-13T18:37:47 *** greg_f
123 2012-02-13T18:37:49 <ThomasWaldmann> can you upload a changeset with fixed indentation?
124 2012-02-13T18:38:18 <ThomasWaldmann> also maybe re-check the logic at that place
125 2012-02-13T18:41:23 <RogerHaase> The indentation mess seems to be a codereview non-feature: http://paste.pocoo.org/show/550371/
126 2012-02-13T18:47:42 <ThomasWaldmann> i don't quite understand what underlying issue the change is fixing
127 2012-02-13T18:47:59 <ThomasWaldmann> why is the "if" you removed never triggering?
128 2012-02-13T18:55:07 <RogerHaase> I will revert these changes and put my logging entries back in to show what the existing code does. Will take a while as I have an interruption.
129 2012-02-13T18:55:39 <ThomasWaldmann> page_href might be a suspect
130 2012-02-13T18:55:52 <ThomasWaldmann> i see it is assigned at multiple places
131 2012-02-13T19:36:40 *** yufra_
132 2012-02-13T19:38:04 <RogerHaase> ThomasWaldmann: the underlying issue is demonstrated here: http://test.moinmo.in/transloopA
133 2012-02-13T20:19:20 * ThomasWaldmann looks
134 2012-02-13T20:19:56 <ThomasWaldmann> The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application.
135 2012-02-13T20:59:34 <RogerHaase> ThomasWaldmann: I reverted include.py and added 4 logging.error messages: http://codereview.appspot.com/5660043/ . Here is a bit of the logged messages from running test_include.py: http://paste.pocoo.org/raw/550444/ . Scroll down and right to see only first page is in stack, all else is None.
136 2012-02-13T21:04:13 <ThomasWaldmann> waldi: can you have a look? ^^
137 2012-02-13T21:04:48 <RogerHaase> ThomasWaldmann: I did check page_href and renamed those inside the for loop, it had no effect on the looping error (but I agree that some instances of page_href should be renamed to make code more clear).
138 2012-02-13T21:07:18 <ThomasWaldmann> maybe after 119 log page_href_new
139 2012-02-13T21:15:13 <RogerHaase> ThomasWaldmann: http://paste.pocoo.org/show/550451/ and revised http://codereview.appspot.com/5660043/
140 2012-02-13T21:42:04 * ThomasWaldmann does some src archaeology :)
141 2012-02-13T21:59:35 <RogerHaase> Looks like "bug" is in original code and moved around in 9417 (04c7c099da1c). I see no other changes
142 2012-02-13T22:11:57 *** dave_largo
143 2012-02-13T22:18:09 <ThomasWaldmann> RogerHaase: i somehow suspect that page_href use in include is somehow incorrect and we in fact should have different name at some places
144 2012-02-13T22:19:04 <ThomasWaldmann> when page_href is given to recurse() i think it should mean "the current page", not "the included page"
145 2012-02-13T22:23:25 <RogerHaase> I changed page_href below ~226 for page, page_href in pages: to another name, but it had no effect on tranclusion loop. Where is "include"
146 2012-02-13T22:24:35 <ThomasWaldmann> include is the module
147 2012-02-13T22:25:18 <RogerHaase> duh :-)
148 2012-02-13T22:30:53 <RogerHaase> Do you agree the page_href within the for loop is misnamed? That is the only potential problem I see.
149 2012-02-13T22:33:40 <ThomasWaldmann> can you try renaming all page_href in the loop to something else except the page_href used in the recurse call?
150 2012-02-13T22:34:14 <ThomasWaldmann> i think that should fix the stack correctness
151 2012-02-13T22:35:54 <RogerHaase> OK, will do
152 2012-02-13T22:43:53 <RogerHaase> hey, looks like that may have fixed it :-)
153 2012-02-13T22:49:34 <ThomasWaldmann> then please choose reasonable names now :)
154 2012-02-13T22:53:41 <dreimark> re
155 2012-02-13T23:00:55 <RogerHaase> ThomasWaldmann: next try: http://codereview.appspot.com/5651088/
156 2012-02-13T23:08:14 * ThomasWaldmann looks
157 2012-02-13T23:24:27 <ThomasWaldmann> RogerHaase: done
158 2012-02-13T23:30:54 <RogerHaase> gtg, will address the changes later
159 2012-02-13T23:31:24 *** RogerHaase
160
