**** BEGIN LOGGING AT Tue Jan 9 14:15:59 2007 Jan 09 14:15:59 --> You are now talking on #gtk-devel Jan 09 14:46:52 --> ebassi (~ebassi@86.149.197.190) has joined #gtk-devel Jan 09 14:58:00 i am here already Jan 09 14:58:18 survived the beer during the apple announcement Jan 09 15:01:07 good thing Jan 09 15:01:18 we don't quite have quorum yet Jan 09 15:01:24 nope Jan 09 15:01:37 lets wait a little bit for rambokid etc Jan 09 15:01:45 I have one hour time Jan 09 15:02:22 --> chpe (~chpe@pD954FFE0.dip0.t-ipconnect.de) has joined #gtk-devel Jan 09 15:02:34 did you see rambo's tooltip reply? Jan 09 15:03:06 I just spotted it, but haven't read yet Jan 09 15:03:14 * mclasen shakes his head some more about GTKVTS Jan 09 15:03:31 heh Jan 09 15:03:43 every single one of the menu tests sets a tearoff title and asserts that it is non-NULL Jan 09 15:03:46 wtf ? Jan 09 15:03:50 mclasen: I heard there seem to be issues with drag and drop in 2.10.7 Jan 09 15:03:55 wouldn't it be enough to do that in one test ? Jan 09 15:03:58 mclasen: did you change any dnd code by chance? Jan 09 15:04:07 I saw that bug, haven't investigated yet Jan 09 15:04:21 mclasen: I never had a really high "dunk" of the gtkvts people Jan 09 15:04:27 if that's proper english ... Jan 09 15:05:31 so you are back to working on tooltips ? Jan 09 15:06:02 i will probably have some time to polish it more and put it in svn Jan 09 15:06:11 will read tim's reply in detail tomorrow Jan 09 15:06:16 great Jan 09 15:06:35 if you have any input on the tooltips, please reply as soon as you have time :) Jan 09 15:06:46 I'll read/reply tonight Jan 09 15:07:33 great Jan 09 15:07:43 I want this api in 2.12, and I want it to be great Jan 09 15:07:59 I am working on improving treeview in my free time as much as I can ... Jan 09 15:08:16 you probably saw the recent scrolling bug fixes (which didn't go into 2.10.x btw) Jan 09 15:08:24 yes Jan 09 15:08:44 clearly the new tooltips will be much more visible/high impact than any amount of treeview fixes... Jan 09 15:08:53 yes Jan 09 15:09:07 but I am very happy with the scrolling bug testsuite -- scrolling bugs have been chasing me since 2.2 Jan 09 15:09:17 qn Jan 09 15:09:18 did you commit the testsuite too ? Jan 09 15:09:27 no, it's on the url I pointed at in my blog Jan 09 15:09:40 I want to integrate the testsuite with gtk+ once gtk+ gets a proper testsuite setup Jan 09 15:10:20 I thought some people were working on that Jan 09 15:10:31 but I haven't seen anything concrete coming out of that Jan 09 15:11:17 and when I am ever done with treeview for 2.x, I want to get back to combobox, entrycompletion maintenance ... Jan 09 15:11:26 I feel I should be responsible for those since I actually wrote them ... Jan 09 15:11:49 by now, they have beome at least as much my responsibility than yours... Jan 09 15:12:10 I knk Jan 09 15:12:11 anyway, did you see any other dnd problems with 2.10.7 apart from the thunderbird bug ? Jan 09 15:12:12 --> rambokid (~timj@d003075.adsl.hansenet.de) has joined #gtk-devel Jan 09 15:12:16 I know, just fixed a bunch of stuff in it Jan 09 15:12:20 (thanks for that btw) Jan 09 15:12:41 mclasen: I just heard a friend complaining that there is some dnd brokenness in 2.10.7 Jan 09 15:12:53 welcome rambokid Jan 09 15:12:58 --> fer (~fherrera@a88-115-27-99.elisa-laajakaista.fi) has joined #gtk-devel Jan 09 15:13:04 hello fernando Jan 09 15:13:09 hmm, the only possibly related change I see in 2.10 is #122688 Jan 09 15:13:22 hey rambokid Jan 09 15:13:28 hello Jan 09 15:13:29 --> tristan (~trisan@mail.touchtunes.com) has joined #gtk-devel Jan 09 15:13:31 mclasen: I didn't commit any treeview dnd change, so I don't see how treeview could be the issue here Jan 09 15:13:48 kris: I'll look at it tonight Jan 09 15:14:04 should we officially start the meeting by collecting agenda items ? Jan 09 15:14:12 mclasen: okay, if it really is a tree view issue, please forward Jan 09 15:14:16 sure Jan 09 15:14:24 mclasen: but I will be a little less responsive in the coming 2.5 weeks Jan 09 15:14:36 mclasen: (exam at the end of Jan, which I really need to pass :) Jan 09 15:14:44 understood Jan 09 15:15:42 hey everyone Jan 09 15:15:48 so, what I want do discuss today is how we get 2.12 development back on track Jan 09 15:15:49 no festival in the summer for me if I don't pass, so :) Jan 09 15:16:05 and the recent threading discussion Jan 09 15:16:06 --> tko (~tk@jumbo180.adsl.netsonic.fi) has joined #gtk-devel Jan 09 15:16:24 mclasen: maybe we should announce on #gnome-hackers too? Jan 09 15:16:29 or maybe we should just start Jan 09 15:16:37 kris: announce what ? Jan 09 15:16:41 the meeting? Jan 09 15:16:47 oh, I'd say just start Jan 09 15:16:57 its 3:15 already, I have to leave at 4 Jan 09 15:16:59 alrighty Jan 09 15:17:27 --> juanjo (~juanjo@cm0007.red.mundo-r.com) has joined #gtk-devel Jan 09 15:18:07 did you change the agenda since your last mail? Jan 09 15:18:18 I put the threading issue on it Jan 09 15:18:25 --> lewing (~lewing@cpe-70-112-186-182.austin.res.rr.com) has joined #gtk-devel Jan 09 15:18:44 ok Jan 09 15:18:53 so, 2.12 development first Jan 09 15:19:26 I expect that I'll be able to devote ~2 days per week to GTK+ in the near term Jan 09 15:19:44 and now that the big 2.10 bug push is over, I'll invest that time into 2.12 features Jan 09 15:20:23 --> mitch (~mitch@91.64.139.140) has joined #gtk-devel Jan 09 15:20:27 the things I personally want to look at first is the regexp code and the gtkbuilder stuff Jan 09 15:20:42 --> pbor|out (~urk@host181-85-dynamic.0-87-r.retail.telecomitalia.it) has joined #gtk-devel Jan 09 15:21:14 what do others want to get in 2.12 ? kris talked about tooltips already Jan 09 15:21:25 how is the gtkbuilder status? Jan 09 15:21:27 I 100% want tooltips in 2.12 Jan 09 15:21:37 fer: I just mentioned it Jan 09 15:21:40 does it only need review? or there is some more coding needed after tristan patches Jan 09 15:21:54 what's the schedule for 2.12, roughly? Jan 09 15:22:05 fer: I haven't really looked at the latest patches, so I can't say right now Jan 09 15:22:23 tko: I'd say that depends on what people want to get in and what amount of work it needs Jan 09 15:22:25 I just loaded the notes of our meeting at guadec here Jan 09 15:22:29 I can say that documentation is lacking Jan 09 15:22:37 and - the parser sucks Jan 09 15:22:39 well Jan 09 15:22:41 btw http://www.gtk.org/plan/ is slightly out of day Jan 09 15:22:41 we've been talking about a canvas, but I guess we won't be able to get that in 2.12 Jan 09 15:22:47 since we are already too late for gnome 2.18, we have a little bit of flexibility Jan 09 15:22:50 tko: I've been wanting to update that for months ... Jan 09 15:23:03 tko: rather, it is totally out of date Jan 09 15:23:04 "sucks" means that parse errors are not reported in a friendly way Jan 09 15:23:09 tko: yesterday I thought of proposing to update that page after today's meeting :) Jan 09 15:23:21 good proposal Jan 09 15:23:55 mclasen: I take it you are logging today's meeting? so we can upload that too Jan 09 15:24:07 mclasen, if its possible and fits in the schedule - I'd like to work on extending GtkAction to work better with builder etc. Jan 09 15:24:21 but better focus on builder first I guess. Jan 09 15:24:27 kris: yea, I'm logging Jan 09 15:24:37 maybe it's a good idea to post a summary on the GtkBuilder status to gtk-devel-list? Jan 09 15:25:16 kris: sounds like a good idea; maybe post status summaries of all the big items we decide to work on for 2.12 Jan 09 15:25:30 that would be regex, gtkbuilder, tooltips so far Jan 09 15:25:58 kris: did we have other things on the agenda back at guadec ? Jan 09 15:26:07 yes, but the person responsible for each of those feature should post such a mail Jan 09 15:26:25 I can post a mail on tooltips after I read tim's mail and maybe even committed it to svn Jan 09 15:26:29 who will post for GtkBuilder? Jan 09 15:26:43 jdahlin is not here ... Jan 09 15:26:51 another item was the canvas, as I just mentioned; but I don't think that will make it for 2.12 Jan 09 15:26:55 I can post my thoughts - not sure exactly whats expected Jan 09 15:26:56 (some time ago there has also been some work on gspell) Jan 09 15:27:16 another item -- Introspection Jan 09 15:27:22 what's going on with that at the moment? Jan 09 15:27:27 tristan: the kind of things you mentioned earlier sounded like good stuff, e.g. needs better error checking, could work better with gtkaction, etc Jan 09 15:27:50 tristan: I would say something like "how far are we? what's still needed?" Jan 09 15:27:51 kris: nothing atm. I basically dropped it, and nobody picked it up yet Jan 09 15:27:56 ok Jan 09 15:28:19 then we have multi-line labels / libsexy cherry picking Jan 09 15:28:20 mclasen, kris; ok great - I'll post my comments Jan 09 15:28:34 that's something what could happen -- if there is somebody to pick up the ball here Jan 09 15:28:44 and post status/ideas to gtk-devel-list too .... Jan 09 15:29:03 there was talk about behdad looking at it, maybe. but so far, he hasn't afaik Jan 09 15:29:42 okay Jan 09 15:29:44 and IIRC ebassi is here Jan 09 15:29:49 (at least he tab completes :) Jan 09 15:30:02 yep, and I even greeted everyone ;-) Jan 09 15:30:04 I have an item on recent file code updates: GtkRenctAction Jan 09 15:30:12 GtkFileChooser integration Jan 09 15:30:17 anything going on in that area? Jan 09 15:30:27 --> garnachAFK (~carlos@84.77.168.153) has joined #gtk-devel Jan 09 15:30:36 kris: I resumed my work on GtkRecentAction to get the ball rolling again Jan 09 15:30:40 nice Jan 09 15:30:45 <-- juanjo has quit (Connection reset by peer) Jan 09 15:30:57 --> juanjo (~juanjo@cm0007.red.mundo-r.com) has joined #gtk-devel Jan 09 15:31:12 so, based on these items, can we come up with a rough schedule for 2.12 ? Jan 09 15:31:13 the UIManager bit is still relying on the hacky approach by mclasen, but I'm trying to clean that up Jan 09 15:31:29 mclasen: I also have international calender support Jan 09 15:31:48 I think that is a behdad item too? Jan 09 15:31:51 hmm, yes Jan 09 15:32:12 I think roozbeh or one of his coworkers had glib patches Jan 09 15:32:16 okay, if we drop canvas and introspection; which seems likely Jan 09 15:32:28 I guess the timeschedule more or less depends o=n GtkBuilder Jan 09 15:32:33 since that looks like that biggest item to me Jan 09 15:32:36 kris: what's missing for introspection? Jan 09 15:32:53 matthias knows most about introspection I would guess Jan 09 15:33:17 --> mbarnes (~mbarnes@nat-pool-bos.redhat.com) has joined #gtk-devel Jan 09 15:34:04 ebassi: it needs someone to drive it to completion. I'd say it is 75% there Jan 09 15:34:12 the big missing piece is a header parser Jan 09 15:34:17 and general robustness Jan 09 15:34:37 another thing for a 2.12 time schedule: the next cairo release? Jan 09 15:34:57 thats expected to happen before the end of january Jan 09 15:35:08 but I don't think we depend on any new cairo api, or do we ? Jan 09 15:35:41 no, but it's on my guadec list :) Jan 09 15:35:45 but it looks like that wont be a problem Jan 09 15:36:31 ok Jan 09 15:36:37 will gvfs git in the 2.12 schedule? Jan 09 15:36:41 no Jan 09 15:36:48 ok Jan 09 15:37:03 alex said that it won't be ready before gnome 2.20 Jan 09 15:37:31 I think that we should aim to include that early in the next glib devel cycle Jan 09 15:39:06 if it's ready and getting there I fully agree Jan 09 15:39:49 as for the gtk+ 2.12 schedule I am really afraid we will have to start looking at march/april this year Jan 09 15:39:57 (at least) Jan 09 15:40:24 I was just about to propose something like mid to end of March as target date Jan 09 15:40:33 --> bratsche (~cody@adsl-68-94-31-189.dsl.rcsntx.swbell.net) has joined #gtk-devel Jan 09 15:40:45 if we aim for getting the feature merge-ready by mid-February Jan 09 15:41:04 I am not sure if med-Feb is doable ... Jan 09 15:41:10 ok Jan 09 15:41:47 does end of Feb sound more doable for landing all 2.12 features ? Jan 09 15:42:14 --> jpe (~jpe@c-71-232-26-186.hsd1.ma.comcast.net) has joined #gtk-devel Jan 09 15:42:27 end of Feb for feature, end of March for release might be nice -- but maybe we should wait for the various status reports on gtk-devel-list first? Jan 09 15:43:13 yes, lets keep that schedule tentative until we know where the features stand Jan 09 15:43:30 ok, I have about 15 min left, and I would like to discuss threading before I leave Jan 09 15:44:12 Currently every single application I run complains that threads are initialized too late... Jan 09 15:46:09 rambokid: is your verdict on late thread initialization in gslice final ? Jan 09 15:46:40 I really think we cannot start demanding that every application calls g_thread_init() just because some library underneath might use threads behind its back Jan 09 15:47:03 i think the most practical approach is to allow late g_thread_init in general, but it'll take significant effort to a) implement this, and b) maintain it (or validate the current code) Jan 09 15:47:31 so it'll probably take some time until we can declare this "perfectly supported" Jan 09 15:47:45 I thought that it mostly worked fine so far - with the exception of gslice Jan 09 15:48:16 I mean, file chooser backends have been doing that for quite some time without adverse effects Jan 09 15:52:38 rambokid: I haven't looked in detail at the patch that mmeeks posted. Is that a good starting point for the required work ? Jan 09 15:55:25 we should probably continue that particular discussion on the mailing list Jan 09 15:55:49 before I run off, I'll probably have to do another round of stable releases soon (like early next week) Jan 09 15:55:55 okay Jan 09 15:56:05 to fix some textview api screwup Jan 09 15:56:18 if there is some filechooser stuff to fix, we should try that soon Jan 09 15:56:19 and the glib-links-against-pthread messup Jan 09 15:56:30 huh... which textview issue? Jan 09 15:56:39 * pbor|out is interested in those... Jan 09 15:56:44 i saw a mail from federico somewhere where he claims he has zero time for gtk+ Jan 09 15:56:48 pbor|out: I changed an internal api and did't update all callers... Jan 09 15:56:49 ... Jan 09 15:57:00 pbor|out: line numbers crash i guess ... Jan 09 15:57:16 mclasen: ok... haven't seen gedit reports yet (fortunately) :) Jan 09 15:57:16 with federico being unavailable, file chooser fixes would most likely come from you, if at all... Jan 09 15:57:29 pbor: muntyan sent a mail about it Jan 09 15:57:41 * pbor|out looks Jan 09 15:57:50 mclasen: maybe, maybe not. i just need the time to sit down and think about the right implementation, regardless of whether i have to implement it myself or not ;) ...and need to take the time to inspect the other thread_init related code portions, like owen did some time ago. Jan 09 15:57:57 mclasen: yes, but I want at least a second look from somebody before I commit Jan 09 15:58:23 kris: sure, I can always have a look Jan 09 15:58:32 mclasen: that'd be cool Jan 09 15:58:41 even if I am not as likely to catch filechooser specific things as federico... Jan 09 15:58:49 when I was sleeping last night I suddely got ideas on fixing confirm-overwrite ... Jan 09 15:58:55 mclasen: me neither ... Jan 09 15:59:22 mclasen: the file chooser is a pretty complicated thing, we are both not able to catch all specific issues .... Jan 09 15:59:40 ok, quick thing before I have to take off: is this time ok for everybody ? Jan 09 15:59:48 or should we move to a different day/time for next week ? Jan 09 15:59:54 it's ok for me Jan 09 15:59:59 assuming we want to pick up the weekly schedule again Jan 09 16:00:13 this time should be good enough for now Jan 09 16:00:22 it would be really great to have a weekly meeting again Jan 09 16:00:38 indeed Jan 09 16:00:41 ok, lets target next tuesday again then Jan 09 16:00:45 I'll send out mail Jan 09 16:00:49 cool! Jan 09 16:00:52 and I'll post the logs in some form Jan 09 16:01:00 mclasen: or just mail me the log Jan 09 16:01:04 I can try updating the website .... Jan 09 16:01:09 thanks for showing up, everybody, see you soon Jan 09 16:01:12 Is it okay if I just hang out and read what you guys have to say? I'm mostly hacking on Win32 stuff in gtk, but I'm interested in other stuff. Jan 09 16:01:24 ...and btw. I won't be able to make the FOSDEM meeting Jan 09 16:01:27 bratsche: no problem at all, that's why it's all public :) Jan 09 16:01:37 but that shouldn't stop any gtk-related activities there Jan 09 16:01:39 bratsche: and if you have comments, please make them Jan 09 16:01:47 Just checking.. I didn't say anything when I joined, because I didn't want to disturb. Jan 09 16:01:49 * mclasen takes off Jan 09 16:02:05 talk to you later mclasen Jan 09 16:02:13 sure Jan 09 16:02:35 kris: I don't know anything about the threads stuff and all.. I'm trying to take on the easy things right now. :) Jan 09 16:03:09 I posted a patch last night for drawing focus on expanders with no labels, for example. That's the type of stuff I want to hack on right now. :) Jan 09 16:04:29 And Win32 stuff. Jan 09 16:05:17 mmmm expanders Jan 09 16:05:27 reminds me of my expander patch Jan 09 16:05:28 heh **** ENDING LOGGING AT Tue Jan 9 16:07:27 2007