I think as long as you can make it optional then it doesn't break any of the APIs and should be fine.
That is just my thoughts. Maybe see what others think first before doing it. On Tue, Jan 26, 2016 at 9:22 PM, Sandro Santilli <s...@keybit.net> wrote: > I've been watching memory corruptions happening during WKB parsing [1] > and found the internal QgsAbstractGeometryV2::fromWkb method not > taking a size parameter to allow preventing such memory errors from > occurring. This needs be changed before it spreads too much ! > > The current method is virtual: > > virtual bool fromWkb( const unsigned char * wkb ) = 0; > > It'll need to be changed to take a size parameter. > > Is that something we can do for this upcoming release or should > we just hope all our WKB providers will pass us good WKB ? > > Note this all started from OGR giving a WKB which may even be > considered valid but our parser did not expect (3D components > inside a marked-to-be-2D collection). > > [1] http://hub.qgis.org/issues/12416 > > --strk; > _______________________________________________ > Qgis-developer mailing list > Qgis-developer@lists.osgeo.org > List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer > Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer
_______________________________________________ Qgis-developer mailing list Qgis-developer@lists.osgeo.org List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer
