Skip to content

shouldn't intersection detection work in latlong space, instead of in pixel space? #205

Open
wiswaud opened this Issue Oct 18, 2013 · 0 comments

1 participant

@wiswaud
wiswaud commented Oct 18, 2013

It would seem that intersection detection works in pixel space, which is fine when you're checking what someone is drawing now. However, when you're loaded polygons which either were drawn previously at a deeper zoom level, or which were defined elsewhere in lat/longs, this leads to points looking like they're all in the same pixel at the current, coarser zoom level. This leads to non-intersecting polygons being unsavable - the algo will think they're intersection, because when translated to pixel-space at the current zoom level, they do intersect.
Is this something others have a problem with?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Something went wrong with that request. Please try again.