lp://qastaging/~jaytaoko/+junk/nuxex
- Get this branch:
- bzr branch lp://qastaging/~jaytaoko/+junk/nuxex
Branch information
- Owner:
- Jay Taoko
- Status:
- Experimental
Recent revisions
- 524. By Andrea Azzarone
-
* Added mouse button test: only mouse buttons 1, 2 and 3 should generate mouse down/up events (jaytaoko)
* Merged with Nux Trunk.
* Tries to fix:
[*] Bug #775872: Menubar - horizontal scroll over a menu opens it
[*] Bug #809717: Horizontal scroll on maximize/restore window control restores window.
[*] Bug #837004: Horizontally scrolling while highlighting an item in the dash or on an indicator will open the item.
[*] a not reported bug: horizontally scrolling on minimize window control (-) blows up unity (at least on my laptop).---
------- ------- -
- Preface:
---------- ------- --
In X11 the horizontal scroll and the vertical one are emulated using a wheel-to-button mapping. The wheel-to-button mapping always generates a press followed instantly by a release.
As you know currently nux doesn't support horizontal scroll.---
------- ------- -------
- What happens in nux
---------- ------- -------
So in the mouse_press function (GraphicsDisplayX11.cpp) and in the mouse_release one is setted to NUX_MOUSE_PRESSED and NUX_MOUSE_RELEASED not only when button1, button2 or button3 are pressed/released, but also when button4, button5, button6 and button7 are pressed/released. Regarding vertical scoll we have no problems because in MouseEventCycle function (WindowComposit
orX11.cpp) , NUX_MOUSE_WHEEL has an higher priority then NUX_MOUSE_PRESSED, so the mouse_wheel signal is emitted. As said in the preface nux doesn't support horizontal scroll. So NUX_MOUSE_WHEEL is not setted and in MouseEventCycle because NUX_MOUSE_PRESSED is setted, the mouse_click (or mouse_up) signal is emitted.
---
------- ------- ------- --
- Solution
---------- ------- ------- --
Don't set NUX_MOUSE_PRESSED (or NUX_MOUSE_RELEASED) if it's a scroll.---
------- ------- ------- ---
- P.S.
---------- ------- ------- ---
No regressions found during my manual test. I cannot imagine any kind of regression. Please let me know if something is wrong.
. Fixes: . Reviewed by Jay Taoko. - 523. By Jay Taoko
-
* Added automated tests from mouse events
* Fixed double-click event: bug #708020
* Fixed: added missing functions in Layout.cpp/.h. Fixes: 708020. Reviewed by Jay Taoko. - 522. By Jay Taoko
-
Many of the duplicates in bug #810182, report errors with loading of GLX.
extension "GLX" missing on display ":0".
Xlib: extension "GLX" missing on display ":0".
Xlib: extension "GLX" missing on display ":0".Many of the Xorg logs also report error in loading of the Nvidia or Nouveau driver. I will add a test to detect the presence of GLX on the system. If the system does not have GLX support, then unity_support_test will not continue further and the test will fail (but not crash).
. Fixes: 810182. Reviewed by Jay Taoko. - 521. By Didier Roche-Tolomelli
-
Add needed TODO notice. By Didier Roche, reviewed by Didier Roche. Fixes: .
- 519. By Jay Taoko
-
* Added requirement for XTest
* Fixed crash in GraphicsDisplay at program termination.
* Added Automated test Framework
* Added Testing framework
* Added empty program test
* Added Button test through XTest
* Added macros nuxOkMsg to NuxCore - 518. By Tim Penhey
-
Change the smart pointer's GetPointer method so there is only one that returns a non const T*.
- 517. By Brandon Schaefer
-
Fix uninitialised variables in GraphicsDisplay (X11 implementation).
Quicklists were not getting focus on login because of an uninitialized
variable (_global_pointer_ grab_active) . Now the class uses an
initialization list.
Branch metadata
- Branch format:
- Branch format 7
- Repository format:
- Bazaar repository format 2a (needs bzr 1.16 or later)