Re: [g-a-devel]test-simple patch ...

[Bill Haneman <bill haneman sun com>]

Padraig O'Briain wrote:
> 
> It looks like I am the culprit here.
> 
> I made a change to atktext.c on Februrary 20th which checked for and disallowed
> negative offsets. Are we saying that this change is wrong?
> 
> If so, can someone specify what the values for end_offset should be?

end_offset can be -1, to signify "last character".  This is
the indexing convention in GTK+ I believe for offsets of unknown
length.

-2, etc. would indeed be an error, and I think a start_offset of -1
would also be highly suspect.

-Bill

> Padraig