GideonPotok commented on code in PR #46041:
URL: https://github.com/apache/spark/pull/46041#discussion_r1567344271
##########
sql/core/src/test/scala/org/apache/spark/sql/CollationStringExpressionsSuite.scala:
##########
@@ -212,6 +212,48 @@ class CollationStringExpressionsSuite
})
}
+ test("Support StringRPad string expressions with collation") {
+ // Supported collations
+ case class StringRPadTestCase[R](s: String, len: Int, pad: String, c:
String, result: R)
+ val testCases = Seq(
+ StringRPadTestCase("", 5, " ", "UTF8_BINARY", " "),
+ StringRPadTestCase("abc", 5, " ", "UNICODE", "abc "),
+ StringRPadTestCase("Hello", 7, "Wörld", "UTF8_BINARY_LCASE", "HelloWö"),
// scalastyle:ignore
+ StringRPadTestCase("1234567890", 5, "aaaAAa", "UNICODE_CI", "12345"),
+ StringRPadTestCase("aaAA", 2, " ", "UTF8_BINARY", "aa"),
+ StringRPadTestCase("ÀÃÂĀĂȦÄäåäáâãȻȻȻȻȻǢǼÆ℀℃", 2, "1", "UNICODE", "ÀÃ"),
// scalastyle:ignore
+ StringRPadTestCase("ĂȦÄäåäá", 20, "ÀÃÂĀĂȦÄäåäáâãȻȻȻȻȻǢǼÆ",
"UTF8_BINARY_LCASE", "ĂȦÄäåäáÀÃÂĀĂȦÄäåäáâã"), // scalastyle:ignore
+ StringRPadTestCase("aȦÄä", 8, "a1", "UNICODE_CI", "aȦÄäa1a1") //
scalastyle:ignore
+ )
+ testCases.foreach(t => {
+ val query = s"SELECT rpad(collate('${t.s}', '${t.c}'), ${t.len},
'${t.pad}')"
+ // Result & data type
+ checkAnswer(sql(query), Row(t.result))
+ assert(sql(query).schema.fields.head.dataType.sameType(StringType(t.c)))
+ })
+ }
+
+ test("Support StringLPad string expressions with collation") {
+ // Supported collations
+ case class StringLPadTestCase[R](s: String, len: Int, pad: String, c:
String, result: R)
+ val testCases = Seq(
+ StringLPadTestCase("", 5, " ", "UTF8_BINARY", " "),
+ StringLPadTestCase("abc", 5, " ", "UNICODE", " abc"),
+ StringLPadTestCase("Hello", 7, "Wörld", "UTF8_BINARY_LCASE",
"WöHello"), // scalastyle:ignore
+ StringLPadTestCase("1234567890", 5, "aaaAAa", "UNICODE_CI", "12345"),
+ StringLPadTestCase("aaAA", 2, " ", "UTF8_BINARY", "aa"),
+ StringLPadTestCase("ÀÃÂĀĂȦÄäåäáâãȻȻȻȻȻǢǼÆ℀℃", 2, "1", "UNICODE",
"ÀÃ"), // scalastyle:ignore
+ StringLPadTestCase("ĂȦÄäåäá", 20, "ÀÃÂĀĂȦÄäåäáâãȻȻȻȻȻǢǼÆ",
"UTF8_BINARY_LCASE", "ÀÃÂĀĂȦÄäåäáâãĂȦÄäåäá"), // scalastyle:ignore
+ StringLPadTestCase("aȦÄä", 8, "a1", "UNICODE_CI", "a1a1aȦÄä") //
scalastyle:ignore
+ )
+ testCases.foreach(t => {
+ val query = s"SELECT lpad(collate('${t.s}', '${t.c}'), ${t.len},
'${t.pad}')"
Review Comment:
That was super helpful. Thank you! I looked at that existing code and made
changes changes to the new string Expressions. contains functions are
predicates, so they don't deal with the exact same scenario. But I think this
scenario can be handled by relying on CollationTypeCasts.scala here/during
enforcement, in addition to during casting itself.
Tests are now passing locally. Please re-review.
--
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]
