Possible fix for #74616

[Germain Garand]

Le Mardi 17 Février 2004 00:39, Dirk Mueller a écrit :
> On Monday 16 February 2004 22:08, rwlbuis at xs4all.nl wrote:
> > BTW this patch fixes the bug on my system.
> > Comments?
>
> I think calling parseAttribute for the first case (m_attrId != 0) should be
> enough. Otherwise I think its fine.
>
>

mmh not sure... if an attribute has been requested by script before, then it 
will have a full implementation (m_attrId==0 flag).

I think just sticking an 
m_element->parseAttribute(&m_attrs[i]);
after the if block would be fine (we are in an assert(m_element) context).

It's also faster to use this form of parseAttribute (the other one just build 
a temporary :)