asolimando commented on code in PR #4557:
URL: https://github.com/apache/calcite/pull/4557#discussion_r2381749784
##########
core/src/main/java/org/apache/calcite/rex/RexUtil.java:
##########
@@ -1718,7 +1731,71 @@ public static boolean isLosslessCast(RelDataType source,
RelDataType target) {
final int targetPrecision = target.getPrecision();
return targetPrecision == PRECISION_NOT_SPECIFIED || targetPrecision >=
sourceLength;
}
- // Return FALSE by default
+
+ // 4) DECIMAL -> DECIMAL
+ if (sourceSqlTypeName == SqlTypeName.DECIMAL
+ && targetSqlTypeName == SqlTypeName.DECIMAL) {
+ int sourcePrecision = source.getPrecision();
+ int sourceScale = Math.max(source.getScale(), 0);
+ int targetPrecision = target.getPrecision();
+ int targetScale = Math.max(target.getScale(), 0);
+ if (sourcePrecision <= 0 || targetPrecision <= 0) {
+ return false;
+ }
+ return targetScale >= sourceScale
+ && (targetPrecision - targetScale) >= (sourcePrecision -
sourceScale);
+ }
+
+ // 5) integer family (signed or unsigned) -> DECIMAL
+ if (SqlTypeUtil.isIntType(source)
+ && targetSqlTypeName == SqlTypeName.DECIMAL) {
+ int targetPrecision = target.getPrecision();
+ int targetScale = Math.max(target.getScale(), 0);
+ int sourcePrecision = source.getPrecision();
+ if (sourcePrecision <= 0) {
Review Comment:
For now I have turned this into an assert `assert sourcePrecision > 0 :
"integer types must have positive precision";`, I think it clarifies
intent/assumptions and gives a chance to spot the issue if run with `-ea`
--
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]
