Is there any objection if I had a "throws ParseException" to the
GeoHashCoder.decode(String) method signature? This is a checked exception.
Additionally, would it be okay to change the return type - currently
double[], to DirectPosition?
Likewise, what about changing the argument type of encode(double,
double) to encode(DirectPosition)?
One more issue: the encode(double, double) method currently expect
arguments in (latitude, longitude), which is the correct order from the
traditional geographer perspective. However the programmatic order (in
PostGIS, GDAL, some legacy OGC specifications, etc.) tends to be
(longitude, latitude). This is a kind of "programmatic structure versus
textual representation" issue. We may be better to use a consistent axis
order for API where no CoordinateReferenceSystem is specified (this will
be an other story when we will have a CoordinateReferenceSystem at
hand). What about (longitude, latitude), keeping in mind that this is
only for the API, not for the coordinates to be show to the end users?
Martin
- Shoudn't GeoHashCoder.decode(String) throws Parse... Martin Desruisseaux
-
