[DOC] Document how to load GeoJSON using Spark SQL's built-in JSON da…

[umartin]

…ta source.

Did you read the Contributor Guide?

• Yes, I have read Contributor Rules and Contributor Development Guide

Is this PR related to a JIRA ticket?

• No, this is a documentation update. The PR name follows the format [DOCS] my subject.

What changes were proposed in this PR?

Document how to use the built-in JSON data source in Spark SQL to read GeoJSON

How was this patch tested?

Did this PR include necessary documentation updates?

[umartin]

I also added documentation on how to work with JDBC data sources.

Reading geometries should be quite general across different databases. Writing I think is more specific. I only added documentation on how to write to Postgis.

[umartin]

I accidentally pushed some other changes. Now this PR only contains documentation updates. I really shouldn't multitask.

[jiayuasu]

Thank you, @umartin ! I don't even know we could do this using Spark's built-in JSON data source. Lesson learned!

Please fix two minor changes.

BTW, is there a way that we could leverage Spark built-in JSON data source and ST_AsGeoJSON to save data back to GeoJSON file?

[umartin]

BTW, is there a way that we could leverage Spark built-in JSON data source and ST_AsGeoJSON to save data back to GeoJSON file?

That's a good question. I haven't tried it but in theory you could build the coordinate part with ST_AsGeoJSON and then build the rest of the json with the to_json function or the json data source. For this to work there needs to be some option in Spark to not quote the output from ST_AsGeoJSON.

Processing GeoJSON in Spark SQL is a bit of a hack. GeoJSON is really the worst imaginable big data format. Since all rows (features) are wrapped up in an envelope object there is no way it can be written or read in parallel. Reading and writing that kind of format in Spark SQL means all data is pushed to a singe task.