[Featureserver] Why reverse sequence on kml decode?

Christopher Schmidt crschmidt at metacarta.com
Thu Apr 23 10:37:13 EDT 2009


On Thu, Apr 23, 2009 at 11:14:05AM +0100, Peter Robins wrote:
> line 119 on the current trunk of vectorformats/Formats/KML.py reverses the 
> sequence of the incoming placemark nodes. I've commented this line out for my 
> own purposes, so my linestrings are in the correct geographical order, but 
> presumably there is a reason for this reversing?


For the purposes of FeatureServer, we have a desire to be able to take a
KML doc from one server, download it, and post it to another: in order
for this to work effectively, the ordering has to be 'reversed' when
decoding, otherwise when we write them into the database, they'll get
all ofut of whack...

But there's no reason this should be teh default, it should instead be
done at the FeatureServer level or something.

Regards,
-- 
Christopher Schmidt
MetaCarta



More information about the Featureserver mailing list