Re[2]: [Gnash-dev] invalidated bounds interfaces

[Udo Giacomozzi]

Hello strk,

Thursday, November 2, 2006, 10:14:46 PM, you wrote:
s> Ok, since we now do have a testcase I'd like to go on
s> with the change. Beside, valgrind reports m_invalidated
s> to be used while still uninitialized, which isn't nice,
s> maybe this is a chance to fix that as well.

Ok, we should fix this compiler warning. In reality it will always be
initialized because it gets set once it is added to a DisplayList, but
agreed, this should be improved.


s> Before proceeding, do invalidated bounds work fine with 
s> the embedded-gravity movie ?

Yes, works fine.


s> And, also better, how could the automated testcase check
s> for it ? Do we have non-random values for the rect
s> that would be returned by a call to get_invalidated_bounds
s> on the root movie ?

I could create a simple animation with fixed bounds. However, I wonder
if a automated test case helps much because there are so many
situations where the mechanism must work correctly and we can't write
a test movie for all.

Udo



Udo