dbatomic commented on code in PR #46247:
URL: https://github.com/apache/spark/pull/46247#discussion_r1600126986
##########
sql/core/src/test/scala/org/apache/spark/sql/streaming/StreamingDeduplicationSuite.scala:
##########
@@ -484,6 +486,52 @@ class StreamingDeduplicationSuite extends
StateStoreMetricsTest {
CheckLastBatch(("c", 9, "c"))
)
}
+
+ test("collation aware deduplication") {
+ val inputData = MemoryStream[(String, Int)]
+ val result = inputData.toDF()
+ .select(col("_1")
+ .try_cast(StringType("UNICODE")).as("str"),
+ col("_2").as("int"))
+ .dropDuplicates("str")
+
+ testStream(result, Append)(
+ AddData(inputData, "a" -> 1),
+ CheckLastBatch("a" -> 1),
+ assertNumStateRows(total = 1, updated = 1, droppedByWatermark = 0),
+ AddData(inputData, "a" -> 2), // Dropped
+ CheckLastBatch(),
+ assertNumStateRows(total = 1, updated = 0, droppedByWatermark = 0),
+ // scalastyle:off
+ AddData(inputData, "ä" -> 1),
+ CheckLastBatch("ä" -> 1),
+ // scalastyle:on
+ assertNumStateRows(total = 2, updated = 1, droppedByWatermark = 0)
+ )
+ }
+
+ test("non-binary collation aware deduplication not supported") {
+ val inputData = MemoryStream[(String)]
+ val result = inputData.toDF()
+ .select(col("value")
+ .try_cast(StringType("UTF8_BINARY_LCASE")).as("str"))
Review Comment:
That's right, UTF8_BINARY_LCASE is a non-binary collation.
In other words, "AAA" and "aaa" are considered equal, even though binary
representations are clearly different.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
