[webkit-reviews] review granted: [Bug 3677] Safari fail to access a second time an element whose content was dynamically modified. : [Attachment 2616] Address comments

[bugzilla-request-daemon at opendarwin.org]

Darin Adler <darin at apple.com> has granted Anders Carlsson <andersca at mac.com>'s
request for review:
Bug 3677: Safari fail to access a second time an element whose content was
dynamically modified.
http://bugzilla.opendarwin.org/show_bug.cgi?id=3677

Attachment 2616: Address comments
http://bugzilla.opendarwin.org/attachment.cgi?id=2616&action=edit

------- Additional Comments from Darin Adler <darin at apple.com>
The code in DocumentImpl::getElementById says "(char *)idCount - 1" in a place
where it should say "(char *)(idCount - 1)". That will work, but only because
pointer arithmetic and integer arithmetic happen to be the same in this case.



I'm going to mark this r=me, although I think we should make that change.

I wondered at first whether m_accessKeyDictValid needed to be set in the
getElementById function. Then I realized that I don't understand the code
setting m_accessKeyDictValid at all. Why on earth does the presence or absence
of an ID have anything to do with the validity of the access key dictionary?

Clearly this patch makes things no worse in that respect, but I'm pretty sure
this points to a bug in the access key support.