Bug 186320 – attenuationFactor should be in range [0,1]

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[Jeremy Jones]

attenuationFactor should be in range [0,1]

[Jeremy Jones]

rdar://problem/40821456

[Jeremy Jones]

Created attachment 341992 [details]
Patch

[Jon Lee]

Comment on attachment 341992 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=341992&action=review

> Source/WebKit/UIProcess/ios/fullscreen/FullscreenTouchSecheuristic.cpp:81
> +    return std::max(std::min(normalizedTimeDelta * m_weight, 1.0), 0.0);

So the weight factor is incorrect when you pass some threshold, but with the clamping, how long must I wait in order for the old influence to be at 0, and basically we start from scratch? Before, the old influence would end up having a negative impact.

[Jer Noble]

Comment on attachment 341992 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=341992&action=review

>> Source/WebKit/UIProcess/ios/fullscreen/FullscreenTouchSecheuristic.cpp:81
>> +    return std::max(std::min(normalizedTimeDelta * m_weight, 1.0), 0.0);
> 
> So the weight factor is incorrect when you pass some threshold, but with the clamping, how long must I wait in order for the old influence to be at 0, and basically we start from scratch? Before, the old influence would end up having a negative impact.

We set the ramp down speed to 1_s and the weight to 0.1, so it would take 10 seconds until the old influence had negative impact under the old code.

[WebKit Commit Bot]

Comment on attachment 341992 [details]
Patch

Clearing flags on attachment: 341992

Committed r232545: <https://trac.webkit.org/changeset/232545>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.