JSON SerDe on SparkSQL throws ClassCastException

[randallwhitman]

java.lang.ClassCastException: java.lang.Long cannot be cast to org.apache.hadoop.io.Writable

Seen in SparkSQL shell; had not been seen when testing on Hive. Both with v1.1 and master.

[randallwhitman]

For integration-regression testing, I ran the following queries updated from #25:

create external table trips_by_origin_cell(leftlon double, botlat double, rightlon double, toplat double, totnum int, samedest int, pct double, destlhs double, destbot double, destrhs double, desttop double)
row format delimited fields terminated by '\t'
location '/user/rwhitman/out-trips-by-origin-cell';

create external table trip_origin_json  (totnum int, samedest int, pct double, destlhs double, destbot double, destrhs double, desttop double, shape binary)
row format serde 'com.esri.hadoop.hive.serde.EsriJsonSerDe'
stored as inputformat 'com.esri.json.hadoop.UnenclosedEsriJsonInputFormat'
outputformat 'org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat'
;
insert overwrite table trip_origin_json select totnum, samedest, pct, destlhs, destbot, destrhs, desttop,
  ST_Polygon(leftlon,botlat, rightlon,botlat, rightlon,toplat, leftlon,toplat) from trips_by_origin_cell;

create external table trip_origin_geojs  (totnum int, samedest int, pct double, destlhs double, destbot double, destrhs double, desttop double, shape binary)
row format serde 'com.esri.hadoop.hive.serde.GeoJsonSerDe'
stored as inputformat 'com.esri.json.hadoop.UnenclosedGeoJsonInputFormat'
outputformat 'org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat'
;
insert overwrite table trip_origin_geojs select totnum, samedest, pct, destlhs, destbot, destrhs, desttop,
  ST_Polygon(leftlon,botlat, rightlon,botlat, rightlon,toplat, leftlon,toplat) from trips_by_origin_cell;