| l3on | larsu, can you please take a look at https://code.launchpad.net/~l3on/ubuntu-themes/csd-fix-resize-and-shadow/+merge/252992 . | 15:38 |
|---|---|---|
| l3on | Thanks in advance! | 15:38 |
| Trevinho | l3on: we're unfortunately having troubles in supporting shadows in compiz side :( | 17:10 |
| l3on | Trevinho, that's a patch for the theme, nothing more. What about compiz? | 17:11 |
| Trevinho | l3on: I know, but the fact is when enabling shadows, gtk windows defines a gtk_frame_extent, that compiz should ignore... but doing that is currently breaking some other stuff | 17:12 |
| l3on | like resizing ? | 17:12 |
| larsu | l3on: ya, this depends on compiz, which is why we don't have this yet | 18:35 |
| larsu | l3on: I also don't like the padding (and it shouldn't be in the same commit) | 18:35 |
| larsu | l3on: btw, this branch contains some headerbar fixes if you're interested: lp:~larsu/ubuntu-themes/titlebars | 18:39 |
| larsu | it makes the decoration buttons look more like unity's | 18:39 |
| l3on | larsu, Trevinho ok I got it ... we need Compiz as "shadow maker". That's clear... Anyway, such kind of changes (the ones I proposed) could be a temporary workaround... no? | 20:26 |
| l3on | larsu, ah.. interesting branch, the icons are too big... aren't they ? | 20:26 |
| larsu | l3on: by one pixel... | 20:28 |
| l3on | larsu, I think that part of code should be in the same file of unity's one. I mean, instead of having a "apps/unity.css", it may have sense have a "windows.css" | 20:30 |
| l3on | larsu, about padding: I found nice the padding added to nautilus titlebar.. buttons are bigger, it's true.. and (from my point of view) they are so because of the importance ... if a button is in HeaderBar you may assume it has a "primary function" in the app... | 20:32 |
| l3on | that's way they are put in the HeaderBar. (imho) | 20:32 |
| larsu | l3on: I put it in unity.css because it already existed | 20:32 |
| larsu | unity uses little padding almost everywhere | 20:33 |
| larsu | I don't think adding a lot of padding to those buttons makes sense | 20:33 |
| l3on | In my opinion, this: http://i.imgur.com/0WYFu0H.png | 20:39 |
| l3on | is better than this: http://i.imgur.com/csPHkRG.png | 20:39 |
| l3on | (larsu unity.css is untouched, you used gtk-widgets.css.. that's why I said that) | 20:41 |
| larsu | oh, indeed | 20:41 |
| larsu | meh :) | 20:41 |
| larsu | I like the one with more padding as well, but I feel that it's not consistent with the rest of the theme | 20:42 |
| l3on | Not really .. padding comes from nautilus bar :) | 20:43 |
| l3on | larsu, see this http://i.imgur.com/uIoxm10.png | 20:44 |
| larsu | that's a toolbar, though | 20:45 |
| larsu | not sure - we can add it if people like it | 20:45 |
| l3on | ok... | 20:45 |
| larsu | but as I said, please don't put it in the same merge request so that we can have some discussion about it | 20:45 |
| * larsu has a feeling people will hate us if we widen their title bars | 20:46 | |
| larsu | also, all of this is irelevant until compiz gets fixes | 20:46 |
| larsu | *fixed | 20:46 |
| larsu | sadly :( | 20:46 |
| l3on | so, with compiz "fixed" every windows will be decorated via compiz itself ? | 20:46 |
| larsu | that would be the fallback if implementing GTK_FRAME_EXTENTS fails | 20:47 |
| larsu | not sure what the status is on that | 20:47 |
| larsu | if we get frame extents, we can draw the shadows client side and resizing at the window border will still work | 20:48 |
| l3on | Just for the log: GTK_FRAME_EXTENTS has been dropped since GTK 3.13.1 | 20:49 |
| larsu | l3on: really? In favor of what? | 20:50 |
| larsu | gdk_x11_window_set_shadow_width() still sets it... | 20:52 |
| l3on | larsu, I think now GTK checks only for gdk_screen_is_composited and has alpha (rgba). In 3.12 the library was sending a Xatom named _GTK_FRAME_EXTENTS to check the support .. | 20:57 |
| l3on | larsu, now look at gtkwindow.c/gtk_window_supports_csd | 20:58 |
| l3on | so, in compiz you may not want to catch the GTK_FRAME_EXTENTS atom... | 20:59 |
| larsu | but how does it find out the actual extents of the window if we draw a shadow client side? | 20:59 |
| larsu | and like I said, it still sets the atom unconditionally on x11 | 21:00 |
| l3on | gdk_x11_window_set_shadow_width is used only in a deprecated (since 3.12) function ... so we can asssume it deprecated as well | 21:02 |
| larsu | l3on: no, it's the vfunc for gdk_window_set_shadow_width() | 21:04 |
| larsu | which is not deprecated | 21:04 |
| larsu | and the atom is still set on my machine | 21:05 |
| larsu | for windows with csd, of course | 21:05 |
| l3on | you're right. | 21:10 |
| l3on | the GTK says this: http://paste.ubuntu.com/10606460/ | 21:11 |
| l3on | I misunderstood it. | 21:11 |
| larsu | ah, makes sense | 21:12 |
| larsu | no worrie | 21:12 |
| l3on | larsu, ok, padding removed. I will prepare another merge request | 21:17 |
| l3on | gtg | 21:19 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
