**** BEGIN LOGGING AT Tue Mar 14 15:56:12 2006 Mar 14 15:56:12 --> You are now talking on #gtk-devel Mar 14 15:56:12 --- Topic for #gtk-devel is Gtk+ meeting channel, for general discussions visit #gtk+ Mar 14 15:56:12 --- Topic for #gtk-devel set by rkaway at Tue Mar 7 15:18:04 2006 Mar 14 15:58:31 --> rambokid (~timj@c135160.adsl.hansenet.de) has joined #gtk-devel Mar 14 15:59:47 --> simosx (Simos@134.83.157.131) has joined #gtk-devel Mar 14 16:00:36 gah, time already Mar 14 16:00:45 Current UTC (or GMT/Zulu)-time used: Tuesday, March 14, 2006 at 21:00:30 Mar 14 16:00:47 yes Mar 14 16:02:41 ok, first thing I wanted to do is propagate the status report I got from J5 on printing earlier today Mar 14 16:03:06 he said that the dialog is mostly done now (except for beautification/higification/etc) Mar 14 16:03:34 alexl is currently looking into how to best handle page setup Mar 14 16:03:37 mm beatification Mar 14 16:03:46 and J5 is starting on turning the backends into modules Mar 14 16:04:20 they are still working in libegg, but at some point we will have to switch to a gtk branch or so Mar 14 16:05:25 so, if people want to give some feedback on the current state of things, that would be great Mar 14 16:05:47 I'll do another review of the current libegg/print-operation code later this week Mar 14 16:07:47 cool Mar 14 16:08:59 kris: do you have code for your tooltips proposal yet ? Mar 14 16:09:31 mclasen: no Mar 14 16:09:44 there's a "semipatch" for the old proposal from owen written by garnacho Mar 14 16:10:04 I plan to write tooltips code once we agree that the current proposal is a step in the right direction Mar 14 16:10:11 and close that the API we want to end up with Mar 14 16:10:14 s/that/to/ Mar 14 16:10:56 it certainly looks like a step in the right direction to me (and its relatively close to the older proposals) Mar 14 16:11:35 ok Mar 14 16:11:47 I'll start writing tooltips code soon then ;) Mar 14 16:11:59 I'm currently working on processing federico's comments on the async filechooser branch Mar 14 16:12:51 kris: I guess there are 2 or 3 testcases that the new tooltips api should pass Mar 14 16:13:01 a) set tooltips on treeview cells/iconview cells Mar 14 16:13:09 b) set tooltips on tags in text views Mar 14 16:13:22 c) rich text (including images) Mar 14 16:13:29 that should all work Mar 14 16:13:37 if you can do those three nicely, the new api should be good Mar 14 16:13:38 you can even pack a tree view in a tooltip Mar 14 16:14:04 of course, you _need_ to have a recursive tooltip test... Mar 14 16:14:06 * jrb htinks we're going to need a suggested style guide for tooltips Mar 14 16:15:57 kris: can tooltip windows be shaped, btw ? Mar 14 16:16:05 * mclasen fears people Mar 14 16:16:10 s creativity Mar 14 16:16:17 mclasen: I was thinking of allowing that Mar 14 16:16:28 or make the tooltips window themable or something Mar 14 16:17:03 (it's mentioned in the new proposal under "random thoughts") Mar 14 16:18:23 its probably fine to allow it Mar 14 16:23:03 --> diego (~diego@wnn0533.wireless.dtu.dk) has joined #gtk-devel Mar 14 16:25:26 <-- diego (~diego@wnn0533.wireless.dtu.dk) has left #gtk-devel Mar 14 16:25:29 ok Mar 14 16:25:30 any more? Mar 14 16:25:38 mclasen: are you looking into recent files already? Mar 14 16:26:04 kris: not quite; I may start in the evening Mar 14 16:27:06 ok Mar 14 16:27:39 kris: I did look over some of the other loose ends we had in bugzilla though, and committed some small things (If you want blog stuff...) Mar 14 16:27:57 i saw a patch to make all dialogs hig compliant Mar 14 16:28:00 or closer to that Mar 14 16:28:18 yes Mar 14 16:28:55 mclasen: d) support tooltips on arbitrary widget regions (canvas items) Mar 14 16:29:19 rambokid: similar to my b), but yes Mar 14 16:29:57 kris: I'm about to commit a patch to make the gif loader only load the first frame if you don't ask for an animation Mar 14 16:30:05 oh Mar 14 16:30:07 sweet Mar 14 16:30:13 kris: makes a big difference if you have multi-MB byzanz screencasts lying around... Mar 14 16:30:35 e) support dynamically changing tooltips (without the user code constantly changing it); i.e. think mouse position indicators Mar 14 16:31:16 * mclasen sees a nice testtooltip.c coming out of this list... Mar 14 16:32:42 i would apprechiate comments on #334446 actually. Mar 14 16:33:29 I commented in the bug earlier today Mar 14 16:33:56 <-- jrb has quit (Leaving) Mar 14 16:34:07 hm, yeah. just saw that. didn't show up in my inbox though, i need to check my bugzilla preferences... Mar 14 16:36:13 rambokid: I think it is probably the right thing to do most of the time; I can't really think of a situation where you want to keep scrolling Mar 14 16:41:13 ok, I need to get going. bye Mar 14 16:41:17 --- You are now known as mclasen_away Mar 14 16:42:08 <-- johannes has quit (... und tschüß) Mar 14 16:44:39 there is a font bug that appers to come from gtk+ et al. with some insight, I am happy to bugzilla this. Mar 14 16:44:58 compare http://master.math.upatras.gr/~kit/flight_5/Screenshot-Add_to_Panel.png and http://master.math.upatras.gr/~kit/flight_5/Screenshot-Add_to_Panel-1.png Mar 14 16:45:53 in the second screenshot, we selected an icon, and some of the text got shifted slightly to the right. Mar 14 16:47:07 the "o" in pilot is different Mar 14 16:47:57 everything else shifted to compensate for that Mar 14 16:50:17 it looks like the same shifting happens independently on the fourth and the fifth columns. Mar 14 16:51:16 indeed the o in Pilot is bigger (elsewhere is ok), but why do the adjacent columns shift as well? Mar 14 16:51:25 and a mu (i think that is a mu) Mar 14 16:51:29 I think that the shifting is happening on the bitmap level. Mar 14 16:53:06 indeed, it's a mu. in the screenshot that shows the "proper" version, both o and mu (the second mu in the word) look unnatural. Mar 14 16:53:56 and, i do not know the name of the character that looks like a three in the lower right selected text, but it seems to be embolding Mar 14 16:54:13 simosx: it looks like more work to make those changes than not to Mar 14 16:54:16 that's the ksi Mar 14 16:55:03 i think (probably) anyone who finished their math studies in united states now knows where mine left off ;) Mar 14 16:55:12 this is not the task though.... Mar 14 16:57:24 I feel that when one selects an icon, the icon gets highlighted, and for this process the text gets shifted (as bitmap) to the right. The offsets of this shifting appear screwed up, hence the problem. Mar 14 16:58:00 I am inclined to bugzilla it as a gtk+ bug. Mar 14 17:00:14 except the characters are changing Mar 14 17:00:56 the o in "pilot" is not the same o that is used in "connect" in the first screenshot Mar 14 17:01:07 in the second, it is the same character Mar 14 17:02:05 and the text is shifting because there are different characters being used. highlighting doesn't seem to be the cause. Mar 14 17:02:52 well, maybe it causes the use of a different character but the highlighting didn't cause the different o or mu Mar 14 17:03:18 the cause of the different ksi, even Mar 14 17:04:01 however, i should just hush about it now. i am looking at the logic of the problem and not the code. Mar 14 17:09:55 I had a look at bugzilla, looks like http://bugzilla.gnome.org/show_bug.cgi?id=329547 (generic) and http://bugzilla.gnome.org/show_bug.cgi?id=327985 (another occurence). Mar 14 17:13:56 except that ksi is bold in one and not in the other Mar 14 17:14:03 that would not be a rounding error Mar 14 17:15:23 if it were adding the pixel to the center. but it is not. Mar 14 17:15:36 maybe i am making rules that don't exist though. Mar 14 17:16:48 the ksi situation appears to be, a rectangle is copied one pixel to the right, therefore, the ksi "spine" appears thicker, if it makes sense.. :) Mar 14 17:17:37 like copying a rectangle and pasting one pixel to the right, so you mix old and pasted rectangle in one thing. Mar 14 17:21:02 --> AME (~chatzilla@portal.extratech.com) has joined #gtk-devel Mar 14 17:21:27 <-- AME (~chatzilla@portal.extratech.com) has left #gtk-devel Mar 14 17:23:40 it just seems weird to me that it did not do the same thing with the o and with the mu Mar 14 17:38:29 <-- simosx has quit (Ping timeout: 600 seconds) Mar 14 18:15:54 <-- ifoox has quit (Leaving) Mar 14 19:21:00 --- rambokid is now known as rkaway Mar 14 22:30:53 <-- halfline has quit (ircd.gimp.org irc.gimp.ca) Mar 14 22:30:53 <-- rkaway has quit (ircd.gimp.org irc.gimp.ca) Mar 14 22:35:35 --> rkaway (~timj@c135160.adsl.hansenet.de) has joined #gtk-devel Mar 14 22:35:35 --> halfline (~rstrode@nat-pool-rdu.redhat.com) has joined #gtk-devel Mar 15 00:32:28 --- Disconnected (Connection timed out). **** ENDING LOGGING AT Wed Mar 15 00:32:28 2006