Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Replace teh-chin with ramer-douglas-peucker
|
|
I think my implementation of this algorithm is broken. It does
something, but the output looks bad.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
We initially used Cairo for its bezier flattening algorithm. That
algorithm turned out to be a bit too imprecise at the scales we're
working at here (#17), so I ended up porting over some code from
Antigrain Graphics. The only other thing we used Cairo for was debug
output and coordinate transforms, so I just wrote the relevant vector
math in a small header file, deleted all debug output code and thus
eliminated the cairo dependency. This is a step towards Windows builds.
|
|
This also fixes an issue where non-closed curves were not dilated
properly.
|
|
|
|
|
|
|