Bug 124396 – Few webpages flicker(shift in offset) while scrolling

• Component: WebKit2
• Status: RESOLVED DUPLICATE

[Praveen Jadhav]

In a few webpages where content scale factor is not equal to 1, UI position of content(visible rectangle offset) in webpage is saved after scaling. So, the display offset and content position as saved in WebView are not equal. Whenever there is a request to update viewsize port, the values stored as content position are passed for displaying without rescaling it back. This changes display offset and hence flickers.

[Praveen Jadhav]

Created attachment 217019 [details]
P

[Praveen Jadhav]

Created attachment 217021 [details]
Patch

Patch to recalculate WebPage viewsize port with content scalefactor.

[Gyuyoung Kim]

CC'ing Noam. This patch needs to be reviewed by Noam.

[Noam Rosenthal]

Marcelo, does this seem right? It seems sensible to me.

[Sergio Correia (qrwteyrutiyoup)]

Pra(In reply to comment #0)
> In a few webpages where content scale factor is not equal to 1, UI position of content(visible rectangle offset) in webpage is saved after scaling. So, the display offset and content position as saved in WebView are not equal. Whenever there is a request to update viewsize port, the values stored as content position are passed for displaying without rescaling it back. This changes display offset and hence flickers.

Would you have any website to give as an example, that demonstrate this problem?

[Praveen Jadhav]

(In reply to comment #5)
> Pra(In reply to comment #0)
> > In a few webpages where content scale factor is not equal to 1, UI position of content(visible rectangle offset) in webpage is saved after scaling. So, the display offset and content position as saved in WebView are not equal. Whenever there is a request to update viewsize port, the values stored as content position are passed for displaying without rescaling it back. This changes display offset and hence flickers.
> Would you have any website to give as an example, that demonstrate this problem?

I was going through http://www.rediff.com. While scrolling, we can observe blinking effect mentioned in this bug. I have tested this patch on EFL port and it seems to have fixed the problem.

[Sergio Correia (qrwteyrutiyoup)]

(In reply to comment #6)
> (In reply to comment #5)
> > Pra(In reply to comment #0)
> > > In a few webpages where content scale factor is not equal to 1, UI position of content(visible rectangle offset) in webpage is saved after scaling. So, the display offset and content position as saved in WebView are not equal. Whenever there is a request to update viewsize port, the values stored as content position are passed for displaying without rescaling it back. This changes display offset and hence flickers.
> > Would you have any website to give as an example, that demonstrate this problem?
> 
> I was going through http://www.rediff.com. While scrolling, we can observe blinking effect mentioned in this bug. I have tested this patch on EFL port and it seems to have fixed the problem.

Thanks. I am going to test it on EFL and Nix Minibrowser (as we also use Coordinated Graphics). I have added EunMi Lee, since his patch at https://bugs.webkit.org/show_bug.cgi?id=118548 is almost identical to this one. He might have insights on this issue.

[Gyuyoung Kim]

(In reply to comment #7)

> Thanks. I am going to test it on EFL and Nix Minibrowser (as we also use Coordinated Graphics). I have added EunMi Lee, since his patch at https://bugs.webkit.org/show_bug.cgi?id=118548 is almost identical to this one. He might have insights on this issue.

Yes, this patch seems same one with Eunmi's patch in Bug 118548. Eunmi, what do you think ?

[Eunmi Lee]

(In reply to comment #8)
> (In reply to comment #7)
> 
> > Thanks. I am going to test it on EFL and Nix Minibrowser (as we also use Coordinated Graphics). I have added EunMi Lee, since his patch at https://bugs.webkit.org/show_bug.cgi?id=118548 is almost identical to this one. He might have insights on this issue.
> 
> Yes, this patch seems same one with Eunmi's patch in Bug 118548. Eunmi, what do you think ?

Yes, this patch is same with Bug 118548.
The m_contentPosition(= contentPosition()) is scaled value with contentsScaleFactor, so we have to be careful to use it.

The transformToScene() also should be fixed with same reason, so I think it is better to merge Bug 118548.

Noam, would you review Bug 118548? I've fixed for your comments.

[Praveen Jadhav]

Lets follow up this issue in https://bugs.webkit.org/show_bug.cgi?id=118548. Patch in Bug 118548 consider more scenarios than that proposed in this bug.

*** This bug has been marked as a duplicate of bug 118548 ***

[Csaba Osztrogonác]

Comment on attachment 217021 [details]
Patch

Cleared review? from attachment 217021 [details] so that this bug does not appear in http://webkit.org/pending-review.  If you would like this patch reviewed, please attach it to a new bug (or re-open this bug before marking it for review again).