[Wayland-bugs] [Bug 744932] Wrong (ultra tiny/small) cursor size on HiDPI screen

[mutter (GNOME Bugzilla)]

https://bugzilla.gnome.org/show_bug.cgi?id=744932

--- Comment #68 from Jonas Ådahl <jadahl at gmail.com> ---
(In reply to Jasper St. Pierre from comment #62)
> Review of attachment 307718 [details] [review]:
> 
> Nah. core/ is for core window management logic. Just leave it in backends/,
> please.

Well its not part of the backend either, but fine.

(In reply to Jasper St. Pierre from comment #64)
> Review of attachment 307715 [details] [review]:
> 
> Actually, it says "this causes the X11 case to always generate the texture,
> will be fixed in a following commit", but I don't see any place where you
> actually fix this...

It happens in the huge (second last) patch. The texture is loaded from
"meta_cursor_sprite_realize_texture() which is called by the renderer when it
knows it needs it.

(In reply to Jasper St. Pierre from comment #61)
> Review of attachment 307717 [details] [review]:
> 
> Not at all a fan of this. Instead of a GObject, perhaps just use a regular
> struct of vfuncs?

In the second last patch I put state in the cursor role object and I also
g_signal_connect_object it. The signal I suppose can be connected to the
surface object, but then I'd still have to have a data type that has instance
variables, vfuncs and a destroy vfunc, which means I'd end up with special
purpose GObject-mini type. Just using GObject felt more reasonable because of
that.

I have also had ideas about moving out all the window logic from
MetaWaylandSurface to a window role or similar, and then it might be nice to
actually be a GObject.

(In reply to Jasper St. Pierre from comment #57)
> Review of attachment 307719 [details] [review]:
> 
> Considering I originally wrote the X11 cursor renderer with the goal of
> nested in mind, I'm not exactly sure what the point is here.

With cursors having different themes depending on what (nested) output it is
on, it felt more close to reality to just always draw it as a texture instead
of hiding and showing the server cursor. It felt cleaner to just have a
separate renderer for the nested case since it'd share none of the
functionality.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/wayland-bugs/attachments/20150721/4e0ccdd5/attachment.html>