dbatomic commented on code in PR #46180:
URL: https://github.com/apache/spark/pull/46180#discussion_r1592428205
##########
common/unsafe/src/test/scala/org/apache/spark/unsafe/types/CollationFactorySuite.scala:
##########
@@ -30,31 +33,101 @@ import
org.apache.spark.unsafe.types.UTF8String.{fromString => toUTF8}
class CollationFactorySuite extends AnyFunSuite with Matchers { //
scalastyle:ignore funsuite
test("collationId stability") {
- val utf8Binary = fetchCollation(0)
+ assert(UTF8_BINARY_COLLATION_ID == 0)
+ assert(INDETERMINATE_COLLATION_ID == 1 << 30)
+
+ val utf8Binary = fetchCollation(UTF8_BINARY_COLLATION_ID)
assert(utf8Binary.collationName == "UTF8_BINARY")
assert(utf8Binary.supportsBinaryEquality)
- val utf8BinaryLcase = fetchCollation(1)
+ val utf8BinaryLcase = fetchCollation(UTF8_BINARY_LCASE_COLLATION_ID)
assert(utf8BinaryLcase.collationName == "UTF8_BINARY_LCASE")
assert(!utf8BinaryLcase.supportsBinaryEquality)
- val unicode = fetchCollation(2)
+ val unicode = fetchCollation(UNICODE_COLLATION_ID)
assert(unicode.collationName == "UNICODE")
- assert(unicode.supportsBinaryEquality);
+ assert(unicode.supportsBinaryEquality)
- val unicodeCi = fetchCollation(3)
+ val unicodeCi = fetchCollation(UNICODE_CI_COLLATION_ID)
assert(unicodeCi.collationName == "UNICODE_CI")
assert(!unicodeCi.supportsBinaryEquality)
}
- test("fetch invalid collation name") {
- val error = intercept[SparkException] {
- fetchCollation("UTF8_BS")
+ test("UTF8_BINARY and ICU root locale collation names") {
+ // collation name already normalized
+ Seq(
+ "UTF8_BINARY",
+ "UTF8_BINARY_LCASE",
+ "UTF8_BINARY_UCASE",
+ "UNICODE",
+ "UNICODE_CI",
+ "UNICODE_AI",
+ "UNICODE_CI_AI",
+ "UNICODE_LCASE",
+ "UNICODE_UCASE"
+ ).foreach(collationName => {
+ val col = fetchCollation(collationName)
+ assert(col.collationName == collationName)
+ })
+ // collation name normalization
+ Seq(
+ // UTF8_BINARY
+ ("UTF8_BINARY_CS", "UTF8_BINARY"),
Review Comment:
I don't think that we should allow UTF8_BINARY_CS, UTF8_BINARY_CI and
UTF8_BINARY_AS and silently fallback to UTF8_BINARY. IMO, passing
UTF8_BINARY_CS should throw an error as invalid collation.
--
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]
