Imbruced commented on code in PR #2490:
URL: https://github.com/apache/sedona/pull/2490#discussion_r2518106081
##########
spark/common/src/main/scala/org/apache/sedona/stats/Weighting.scala:
##########
@@ -109,30 +109,42 @@ object Weighting {
val formattedDataFrame = dataframe.withColumn(ID_COLUMN,
sha2(to_json(struct("*")), 256))
- formattedDataFrame
+ val spatiallyJoined = formattedDataFrame
.alias("l")
.join(
formattedDataFrame.alias("r"),
- joinCondition && col(s"l.$ID_COLUMN") =!= col(
- s"r.$ID_COLUMN"
- ), // we will add self back later if self.includeSelf
+ joinCondition && col(s"l.$ID_COLUMN") =!= col(s"r.$ID_COLUMN"),
+ "inner")
+ .select(struct("l.*").alias("left"), struct("r.*").alias("right"))
+
+ val mapped = formattedDataFrame
+ .alias("f")
+ .join(
+ spatiallyJoined.alias("s"),
+ col(s"s.left.$ID_COLUMN") === col(s"f.$ID_COLUMN"),
"left")
.select(
- col(s"l.$ID_COLUMN"),
- struct("l.*").alias("left_contents"),
- struct(
- (
- savedAttributesWithGeom match {
- case null => struct(col("r.*")).dropFields(ID_COLUMN)
- case _ =>
- struct(savedAttributesWithGeom.map(c => col(s"r.$c")): _*)
- }
- ).alias("neighbor"),
- if (!binary)
- pow(distanceFunction(col(s"l.$geometryColumn"),
col(s"r.$geometryColumn")), alpha)
- .alias("value")
- else lit(1.0).alias("value")).alias("weight"))
- .groupBy(s"l.$ID_COLUMN")
+ col(ID_COLUMN),
+ struct("f.*").alias("left_contents"),
+ when(col(ID_COLUMN).isNull, lit(null))
Review Comment:
makes sense, this condition is not needed here
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
