[Webkit-unassigned] [Bug 23452] Add XHTMLMP support to Webkit

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=23452





------- Comment #16 from yichao.yin at torchmobile.com.cn  2009-02-11 19:53 PDT -------
(In reply to comment #12)
> (From update of attachment 27559 [review])
> > +#if ENABLE(XHTMLMP)
> > +    ASSERT(settings());
> > +    m_shouldProcessNoScriptElement = !settings()->isJavaScriptEnabled();
> > +#endif
> I don't think this assertion is helpful. We don't want to assert ever pointer
> just before we dereference it.
> But I also don't know if settings() is guaranteed to be non-zero here or not.
> Is it guaranteed?

Seems no guarantee for it, so that we'd better add null-check for it and
initialize it by assignment.


> > +#if PLATFORM(QT) && ENABLE(XHTMLMP)
> > +    // FIXME: Qt is flawed
> > +    // This hacking is for fixing one bug of javasrcipt in XHTMLMP document, 
> > +    // i.e: if alert() method is contained directly in <script> element, 
> > +    // it will cause QtLauncher crash when frame loader execute it.
> > +    if (m_isRunningScript && m_frame->document() && m_frame->document()->isXHTMLMPDocument())
> > +        return;
> > +#endif
> We need to figure out the real problem and fix this in a better way.

I will investigate it further.

All of the other issues in comments#12, I will fix them in the upcoming patch.


-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.