On 16/04/13 01:09, Martin Owens wrote:
On Tue, 2013-04-16 at 00:51 +0200, Alexander Brock wrote:
The third and the fourth argument should be switched
Nice work Alexander! is there any way to turn this into a test so we can detect such a problem in the future?
You could test if the created points lie on the original curve, but I'd rather test if deleting previously inserted, unneccessary nodes leads to a set of control points within a sufficiently small neighbourhood of the original points.
I've been watching your adventure with the deletion of nodes, I oft been bitten by massive inaccuracies which gnash at my workflow. I'm keen to test the patch, although I don't have a local build infrastructure atm and inkscape is non-trivial to build.
Which platform are you running? I'm running debian testing and building worked without any problems.
Alexander