Bug 111429 – [Qt][MiniBrowser] Improve the update of touch indicators

• Component: New Bugs
• Status: RESOLVED FIXED

[Andras Becsi]

[Qt][MiniBrowser] Improve the update of touch indicators

[Andras Becsi]

Created attachment 191473 [details]
Patch

[Jocelyn Turcotte]

Comment on attachment 191473 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=191473&action=review

> Tools/MiniBrowser/qt/BrowserWindow.cpp:122
> +    if (touchPoints.isEmpty()) {

Is it possible that we get a partial list of touchpoints where some IDs would be missing?

[Andras Becsi]

(In reply to comment #2)
> (From update of attachment 191473 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=191473&action=review
> 
> > Tools/MiniBrowser/qt/BrowserWindow.cpp:122
> > +    if (touchPoints.isEmpty()) {
> 
> Is it possible that we get a partial list of touchpoints where some IDs would be missing?

The problem is not missing IDs, it's just that not all events have released state.

We receive an empty list because of a released control key an this results in problems if you test gesture chains, since not all the states are released by the time the control key is released.

Maybe this makes more sense in context with:
https://bugs.webkit.org/show_bug.cgi?id=111430

[Andras Becsi]

(In reply to comment #2)
> (From update of attachment 191473 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=191473&action=review
> 
> > Tools/MiniBrowser/qt/BrowserWindow.cpp:122
> > +    if (touchPoints.isEmpty()) {
> 
> Is it possible that we get a partial list of touchpoints where some IDs would be missing?

To also answer your actual question :)
It looks like we always receive all the possible ID because touch points are updated for all combinations.
But I can verify that if you have concerns.

[Jocelyn Turcotte]

Comment on attachment 191473 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=191473&action=review

> Tools/MiniBrowser/qt/BrowserWindow.cpp:79
> +    m_activeMockComponents.reserve(3);

I think that at this level it won't have any noticeable benefit, if any.

>>>> Tools/MiniBrowser/qt/BrowserWindow.cpp:122
>>>> +    if (touchPoints.isEmpty()) {
>>> 
>>> Is it possible that we get a partial list of touchpoints where some IDs would be missing?
>> 
>> The problem is not missing IDs, it's just that not all events have released state.
>> 
>> We receive an empty list because of a released control key an this results in problems if you test gesture chains, since not all the states are released by the time the control key is released.
>> 
>> Maybe this makes more sense in context with:
>> https://bugs.webkit.org/show_bug.cgi?id=111430
> 
> To also answer your actual question :)
> It looks like we always receive all the possible ID because touch points are updated for all combinations.
> But I can verify that if you have concerns.

Ok I see what you mean, it's passed that way explicitely by MiniBrowserApplication::sendTouchEvent.
I was wondering if it was worth checking for missing ids in the list as a general solution instead of checking the only subcase that interests us: "isEmpty()".

An explicit way would be to have a clearVisualMockTouchPoints method that does this. Just an idea, both are fine to me so r=me.

[Andras Becsi]

Committed r144758: <http://trac.webkit.org/changeset/144758>