Closed ircwaves closed 9 years ago
The bug is that it didn't pass gips/test/tile_geom.py
. 7bcbc94 makes it pass.
Looks like using a.intersects(b)
is just as fast as a.overlaps(b)
. a.intersection(b)
is ~20x slower.
In [17]: %timeit -n 10000 a.intersects(b)
10000 loops, best of 3: 12.4 µs per loop
In [18]: %timeit -n 10000 a.overlaps(b)
10000 loops, best of 3: 12.4 µs per loop
In [19]: %timeit -n 10000 a.intersection(b)
10000 loops, best of 3: 217 µs per loop
I didn't intend to push the changes to gipit/gips -- but it happened. So, I won't issue a pull request.
I will have to look at my git
settings. git push
pushed to upstream, instead of origin. Didn't think that was the default behaviour.
Actually, i had pulled those changes into master, already, so you only wrote the (now defunct) final0.8.0 branch. But no matter, I'll merge those with master
Request is as commented. Ignore phone call....have a look and let me know if there are issues with what I have done. Thanks.