[webkit-qt] API review, comments needed!

[Joseph Ligman]

QWebFrame::scrollRecursively's implicit dependency on the last mouse  
move event creates an error in my client code. I was planning on  
fixing it. I will fix the documentation and add the other tests as well.



On Mar 5, 2010, at 10:16 AM, Simon Hausmann wrote:

> Hi,
>
> we just went through the diff of the header files between Qt 4.6's  
> WebKit and
> the trunk. It's a relatively small diff, but nevertheless taking  
> another look
> at it revealed a few interesting issues. Here are the comments and  
> questions
> we collected. Please comment :)
>
>
> QWebFrame::scrollRecursively:
>
>    * Implicit dependency on last mouse move event is bad. The  
> position should
> become an explicit function parameter.
>
>    * The use of "recursion" in the name is confusing, it suggests  
> recursive
> behaviour to child frames. However the behaviour is to continue  
> scrolling the
> parent frames.
>
>    * The documentation does not explain what the return value is.
>
>    * The auto test only checks successful scrolling. It never tests  
> the case
> of the function
>      returning false.
>
> QWebSettings:
>
>    * The following hunk causes binary compatibility issues:
>
>         LocalContentCanAccessRemoteUrls,
> -        DnsPrefetchEnabled
> +        LocalContentCanAccessFileUrls,
> +        DnsPrefetchEnabled,
>
>      because the values of the enum changed.
>
>   * The default values of the settings are not documented!
>
>   * The documentation for the XSSAuditorEnabled feature is missing a  
> pointer
> to a more detailed
>     explanation of the WebKit Cross Site Auditing feature.
>
>   * Should it be XSSAudit_ing_Enabled?
>
>
>
> Simon
> _______________________________________________
> webkit-qt mailing list
> webkit-qt at lists.webkit.org
> http://lists.webkit.org/mailman/listinfo.cgi/webkit-qt