I initially used this to correctly handle ranges when their respective containers
are mutated, to get weak references of Range objects. I now realise that the weak
references should be handled at a lower-level, closer to the JS-managed object.
Very basic touch events and touch scrolling
This implements just enough of [Touch Events](http://w3c.github.io/touch-events/) to enable scrolling on Android without regressing basic single-touch interaction like clicking on links.
Dragging a page will scroll it, unless the page calls `preventDefault` on the "touchstart" event.
Does **not** yet support pinch zooming or other multi-touch gestures or events.
Includes a `-Z convert-mouse-to-touch` command line flag for testing on non-touch platforms. This is also enabled by default on Android because Glutin currently translates touch input to mouse events on Android.
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/7204)
<!-- Reviewable:end -->
This is enabled by default on Android, because Glutin currently sends mouse
events instead of touch events on Android. It's also useful for testing on
non-touch platforms.
Code on the script thread can only observe self.layout_join_port being Some()
between the time it is set in force_reflow, and the join_layout call later in
that function, and no significant code is called in that code.
Since these functions do nothing useful if layout_join_port is None, there is
no point in keeping them.
Code on the script thread can only observe self.layout_join_port being Some()
between the time it is set in force_reflow, and the join_layout call later in
that function, and no significant code is called in that code.
This implies that layout_is_idle will always return true, so there is no
reason for the function to exist.
Code on the script thread can only observe self.layout_join_port being Some()
between the time it is set in force_reflow, and the join_layout call later in
that function, and no significant code is called in that code.
As join_layout does nothing when called if self.layout_join_port is None,
these two calls are useless.
add (un)premultiply tables for canvas {Get,Put}ImageData operations
Pretty straightforward use of lookup tables to replace a bunch of expensive float operations.
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/8086)
<!-- Reviewable:end -->
display input caret for textarea. fixes#7758
This adds the input caret for textareas. Although, it does not handle multiline textareas correctly. The caret gets displayed for each line.
I'll look into that but that will take more time. Some feedback on this small patch would be appreciated though.
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/7761)
<!-- Reviewable:end -->
Canvas is currently given a layer at the stacking context level.
Instead it's DisplayItem should be given a layer directly. This fixes
painting order issues where canvases are painted on top of other
positioned content that is later in tree order. It always simplifies
the code a bit.
Remove HAS_DIRTY_SIBLINGS.
This isn't doing anything right now, and we're not even setting it properly
in dirty_impl the |dirty_subtree(self)| was causing us to hit the skip case
for step 3.
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/8042)
<!-- Reviewable:end -->
This isn't doing anything right now, and we're not even setting it properly
in dirty_impl the |dirty_subtree(self)| was causing us to hit the skip case
for step 3.
Fix char types in script binding codegen
Use `libc::c_char` instead of `i8` for character data since that's more
portable. (Some architectures, e.g. AArch64, have unsigned characters,
i.e. `u8`.)
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/8132)
<!-- Reviewable:end -->
Ordering guarantees for timers
This is an rough solution to the issue described in #3396. XHRs still do their own thing and an overall clean up is in order. Before I do that, though, I'd really like someone to sign off on the overall idea.
There's one major difference to what jdm layed out #3396: The timers remain with the window/worker and only the earliest expiring one is coordinated with the dedicated timer thread.
That means both the timer thread and the window/worker have to keep track of which timer expires next, which feels a bit wonky. However, the upshot is that there's no need for communication with the timer thread when a pipeline is frozen, thawed or dropped.
Most relvant parts are
- the [`TimerScheduler`](6f5f661958 (diff-74137a6f50ab38e7a1e4d16920a66ce7R73)), which is the new per-constellation timer task and
- the [`ActiveTimers`](6f5f661958 (diff-86707d952414a2860b78bcf6c1db8e2eR34)) which is what's left on the window/worker side.
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/7450)
<!-- Reviewable:end -->
Use `libc::c_char` instead of `i8` for character data since that's more
portable. (Some architectures, e.g. AArch64, have unsigned characters,
i.e. `u8`.)
