Fix "whole" multi centroid computation (#1606) #1610

merged 2 commits into from Nov 30, 2012


None yet

2 participants


Closes #1606.
Still lacks a testcase

Sandro Santilli Fix "whole" multi centroid computation (#1606)
Still lacks a testcase

I'm having an hard time writing the testcase showing previosu implementation wrong.
The real-life case is a multipolygon composed by 22 elements, many of which having 4 points (triangles)
all within extent BOX(736908.312373676 4390316.66779138,2060770.68121407 5942345.52058283)

I think being far away from the 0,0 origin helps here. Trying harder..
May superceed #1613 and #1615


Alright, I resorted to add a new test

@strk strk referenced this pull request Nov 29, 2012

Centroid computer is bogus #1606

@springmeyer springmeyer merged commit 3ec3a7c into mapnik:2.1.x Nov 30, 2012

1 check failed

Details default The Travis build failed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment