[08:48] <davmor2> Morning All
[09:09] <ara> morning
[09:09] <eeejay> morning ara
[09:09] <ara> hey eeejay
[09:12] <ara> eeejay: I may have some time this morning to work on the pidgin test
[09:15] <eeejay> ara: excellent
[09:16] <eeejay> ara: do you mind if i got ahead and merge dx to trunk?
[09:16] <ara> eeejay: go ahead, it is only new classes and new scripts. no "real" merging :)
[09:16] <eeejay> ara: exactly!
[09:17] <eeejay> ara: but as for pidgin, i think we might need to add the possibility to have params on the suite level, not just the case level
[09:17] <ara> eeejay: what for?
[09:17] <eeejay> ara: for example: <suite blah blah><params><protocol>MSN</protocol>...
[09:18] <eeejay> ara: because the open() function in pidgin signs in the user and the buddy
[09:19] <ara> eeejay: mmm, true. maybe it shouldn't. maybe it should just open the application (no signing on) and then use a login method at the testcase level
[09:19] <eeejay> ara: the problem would be that you sign out/in on every case
[09:20] <eeejay> ara: my internet here is too slow for that :)
[09:20] <ara> eeejay: hehehe
[09:20] <ara> eeejay: yes, that's true. we might need that then. also, it is more flexible. if you came up with a case, some others might come up in the future
[09:21] <eeejay> ara: the other option is to have a very basic subclass, PidginMsn.protocol = "MSN"
[09:23] <ara> eeejay: I think having parameters on the testsuite level is better option
[09:24] <eeejay> ara: yup, so that is less trivial, i'll do it now in my branch. it should take a few minutes
[10:07] <davmor2> eeejay: I just read your blog post is the change in backend so apps that don't support at-spi work?
[10:36] <eeejay> davmor2: hah, nope.
[10:37] <davmor2> eeejay: What is the benefit of it then?
[10:39] <eeejay> davmor2: sounder design, and it will be a hopefully automatic transition when GNOME 3.0 comes in
[10:40] <eeejay> davmor2: ldtp uses some deprecated libraries, and it has grown a lot over the years
[10:41] <davmor2> eeejay: Ah okay.