Merge lp://qastaging/~vanvugt/compiz-core/fix-925293.2 into lp://qastaging/compiz-core
Proposed by
Daniel van Vugt
| Status: | Merged |
|---|---|
| Merged at revision: | 3006 |
| Proposed branch: | lp://qastaging/~vanvugt/compiz-core/fix-925293.2 |
| Merge into: | lp://qastaging/compiz-core |
| Diff against target: |
133 lines (+28/-31) 3 files modified
src/event.cpp (+22/-29) src/privatescreen.h (+1/-1) src/screen.cpp (+5/-1) |
| To merge this branch: | bzr merge lp://qastaging/~vanvugt/compiz-core/fix-925293.2 |
| Related bugs: |
| Reviewer | Review Type | Date Requested | Status |
|---|---|---|---|
| Sam Spilsbury | Approve | ||
|
Review via email:
|
|||
This proposal supersedes a proposal from 2012-02-17.
Description of the change
Improved the fix for LP: #925293 (add tap support):
* Add correct tap detection for non-modifier-only key combinations.
* Avoid getting "stuck" with modifiers apparently locked down.
* Ungrab the keyboard on key releases, not presses.
* ReplayKeyboard of release events as well as presses (previously only
presses, which could cause release events to queue up and block the
server).
My test case which previously caused modifiers to get "stuck" now works. The test case was "Add correct tap detection for non-modifier-only key combinations", which caused the modifiers-stuck problem in previous revisions.
To post a comment you must log in.
Revision history for this message
| Sam Spilsbury (smspillaz) wrote | # |
Revision history for this message
| Sam Spilsbury (smspillaz) wrote | # |
Revision history for this message
| Daniel van Vugt (vanvugt) wrote | # |
Revision history for this message
| Sam Spilsbury (smspillaz) wrote | # |
Revision history for this message
| Daniel van Vugt (vanvugt) wrote | # |
Preview Diff
[H/L] Next/Prev Comment, [J/K] Next/Prev File, [N/P] Next/Prev Hunk
> XAllowEvents (priv->dpy, ReplayKeyboard, event->xkey.time);
Do we need to do that if an action event was handled ? I would think that AsyncKeyboard makes more sense here.