Hello Chris
Le 07/06/13 22:40, Mattmann, Chris A (398J) a écrit :
+1 to make all 3 updates below.
Thanks, I went away. Javadoc is at [1]. I added a get/setFormat
accessor, where Format is an enum containing only BASE32 for now. But I
tried to design it in a way that make room for BASE36 is we want to
implement it later. I also added get/setPrecision accessor. On the
implementation side, I took the liberty to modify the code in order to
reduce the amount of array allocations. This would hopefully improve
performance, but I didn't benchmarked...
I noticed that the algorithm would be easy to generalize to the
n-dimensional case, thus encoding/decoding geohash for spatio-temporal
coordinates. However I don't know if there would be any use for that,
especially since it would be non-standard. Is anyone aware of
applications that would benefit from (x,y,z) or (x,y,z,t) geohashes,
despite being non-standard? If so, a JIRA task could be created for
collecting use cases.
Martin
[1]
https://builds.apache.org/job/sis-trunk/site/apidocs/org/apache/sis/index/GeoHashCoder.html
