Questions - EPSG conversion about sedona HOT 2 CLOSED

@anto , have you successfully got a SpatialRDD using SpatialRDD constructor?

If so, please use

public boolean CRSTransform(String sourceEpsgCRSCode, String targetEpsgCRSCode)

to do CRSTransformation on your spatialRDD.

Pseudo code:

PointRDD myPointRDD = new PointRDD(...)
myPointRDD.CRSTransform("epsg:32722","epsg:4326")

After that, you can persist it onto disk.
Pseudo code to store as TSV:

myPointRDD.rawSpatialRDD.saveAsTextFile(...)

Pseudo code to store as GeoJSON:

myPointRDD.saveAsGeoJSON(...)

from sedona.

Hello jiayuasu,

Thanks a lot for the hints. I have been away so I am sorry that I could not respond back earlier.

I can indeed manage to load CSV file, convert EPSG and save it back into TSV file. The example of my results is as below:

CSV Input Data with EPSG: 32772
780530,9839470,783767
780710,9840510,22004

Conversion result to EPSG: 4326
POINT (-1.4509475921881358 -48.47896284047234)	780530,9839470,783767
POINT (-1.4415458115407216 -48.47735667412911)	780710,9840510,22004

However, what I have been looking for is the processing without involving files, as the input data might be quite big. So I basically would like to load Spark RDD or Hive table into SpatialRDD, convert the EPSG and store the result back into different Spark RDD or Hive table.

I am not sure how to achieve that as it seems the only way to create SpatialRDD is load the data from a file, e.g. PointRDDInputLocation. Do you have any suggestion on this?

Thanks a lot in advance for your help.

Kind regards,

Anto

from sedona.