uros-db commented on code in PR #48386:
URL: https://github.com/apache/spark/pull/48386#discussion_r1793050474
##########
common/unsafe/src/main/java/org/apache/spark/sql/catalyst/util/CollationFactory.java:
##########
@@ -538,27 +542,63 @@ private static boolean isValidCollationId(int
collationId) {
@Override
protected Collation buildCollation() {
if (caseSensitivity == CaseSensitivity.UNSPECIFIED) {
+ Comparator<UTF8String> comparator;
+ ToLongFunction<UTF8String> hashFunction;
+ BiFunction<UTF8String, UTF8String, Boolean> equalsFunction;
+ boolean supportsSpaceTrimming = spaceTrimming != SpaceTrimming.NONE;
+
+ if(spaceTrimming == SpaceTrimming.NONE) {
+ comparator = UTF8String::binaryCompare;
+ hashFunction = s -> (long) s.hashCode();
+ equalsFunction = UTF8String::equals;
+ }else {
+ comparator = (s1, s2) -> applyTrimmingPolicy(s1,
spaceTrimming).binaryCompare(
+ applyTrimmingPolicy(s2, spaceTrimming));
+ hashFunction = s -> (long) applyTrimmingPolicy(s,
spaceTrimming).hashCode();
+ equalsFunction = (s1, s2) -> applyTrimmingPolicy(s1,
spaceTrimming).equals(
+ applyTrimmingPolicy(s2, spaceTrimming));
+ }
+
return new Collation(
normalizedCollationName(),
PROVIDER_SPARK,
null,
- UTF8String::binaryCompare,
+ comparator,
"1.0",
- s -> (long) s.hashCode(),
+ hashFunction,
+ equalsFunction,
/* supportsBinaryEquality = */ true,
/* supportsBinaryOrdering = */ true,
- /* supportsLowercaseEquality = */ false);
+ /* supportsLowercaseEquality = */ false,
+ spaceTrimming != SpaceTrimming.NONE);
} else {
+ Comparator<UTF8String> comparator;
+ ToLongFunction<UTF8String> hashFunction;
+
+ if(spaceTrimming == SpaceTrimming.NONE ) {
+ comparator = CollationAwareUTF8String::compareLowerCase;
+ hashFunction = s ->
+ (long)
CollationAwareUTF8String.lowerCaseCodePoints(s).hashCode();
+ }else{
+ comparator = (s1, s2) -> CollationAwareUTF8String.compareLowerCase(
+ applyTrimmingPolicy(s1, spaceTrimming),
+ applyTrimmingPolicy(s2, spaceTrimming));
Review Comment:
```suggestion
comparator = (s1, s2) ->
CollationAwareUTF8String.compareLowerCase(
applyTrimmingPolicy(s1, spaceTrimming),
applyTrimmingPolicy(s2, spaceTrimming));
```
##########
common/unsafe/src/main/java/org/apache/spark/sql/catalyst/util/CollationFactory.java:
##########
@@ -538,27 +542,63 @@ private static boolean isValidCollationId(int
collationId) {
@Override
protected Collation buildCollation() {
if (caseSensitivity == CaseSensitivity.UNSPECIFIED) {
+ Comparator<UTF8String> comparator;
+ ToLongFunction<UTF8String> hashFunction;
+ BiFunction<UTF8String, UTF8String, Boolean> equalsFunction;
+ boolean supportsSpaceTrimming = spaceTrimming != SpaceTrimming.NONE;
+
+ if(spaceTrimming == SpaceTrimming.NONE) {
+ comparator = UTF8String::binaryCompare;
+ hashFunction = s -> (long) s.hashCode();
+ equalsFunction = UTF8String::equals;
+ }else {
+ comparator = (s1, s2) -> applyTrimmingPolicy(s1,
spaceTrimming).binaryCompare(
+ applyTrimmingPolicy(s2, spaceTrimming));
+ hashFunction = s -> (long) applyTrimmingPolicy(s,
spaceTrimming).hashCode();
+ equalsFunction = (s1, s2) -> applyTrimmingPolicy(s1,
spaceTrimming).equals(
+ applyTrimmingPolicy(s2, spaceTrimming));
+ }
+
return new Collation(
normalizedCollationName(),
PROVIDER_SPARK,
null,
- UTF8String::binaryCompare,
+ comparator,
"1.0",
- s -> (long) s.hashCode(),
+ hashFunction,
+ equalsFunction,
/* supportsBinaryEquality = */ true,
/* supportsBinaryOrdering = */ true,
- /* supportsLowercaseEquality = */ false);
+ /* supportsLowercaseEquality = */ false,
+ spaceTrimming != SpaceTrimming.NONE);
} else {
+ Comparator<UTF8String> comparator;
+ ToLongFunction<UTF8String> hashFunction;
+
+ if(spaceTrimming == SpaceTrimming.NONE ) {
+ comparator = CollationAwareUTF8String::compareLowerCase;
+ hashFunction = s ->
+ (long)
CollationAwareUTF8String.lowerCaseCodePoints(s).hashCode();
+ }else{
+ comparator = (s1, s2) -> CollationAwareUTF8String.compareLowerCase(
+ applyTrimmingPolicy(s1, spaceTrimming),
+ applyTrimmingPolicy(s2, spaceTrimming));
+ hashFunction = s -> (long) CollationAwareUTF8String.
+ lowerCaseCodePoints(applyTrimmingPolicy(s,
spaceTrimming)).hashCode();
Review Comment:
```suggestion
hashFunction = s -> (long) CollationAwareUTF8String.
lowerCaseCodePoints(applyTrimmingPolicy(s,
spaceTrimming)).hashCode();
```
--
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]
