wenhuitang commented on a change in pull request #1990:
URL: https://github.com/apache/calcite/pull/1990#discussion_r431575566
##########
File path:
cassandra/src/main/java/org/apache/calcite/adapter/cassandra/CassandraFilter.java
##########
@@ -174,14 +174,13 @@ private String translateMatch(RexNode condition) {
}
}
- /** Convert the value of a literal to a string.
+ /** Returns the value of the literal.
*
* @param literal Literal to translate
- * @return String representation of the literal
+ * @return The value of the literal in the form of the actual type.
*/
- private static String literalValue(RexLiteral literal) {
- Object value = literal.getValue2();
- return String.valueOf(value);
+ private static Object literalValue(RexLiteral literal) {
+ return literal.getValue2();
}
Review comment:
> For `Integral` and `string` it is correct, but i'm not sure if it is
for other data types literal, such as `TIMESTAMP(3)`, can you make sure that ?
Thanks a lot for reviewing. This PR has been updated.
I have tested for type like Date, Timestamp and Time.
1.When the type of primary key 'f_time' is Time, it will throw exception
when the filter condition is "f_time = Time '13:30:54.234'", because the type
of 'f_time' is Bigint.
It seems another issue.
2.As for Timestamp(3), CassandraSchema create a Sql type Timestamp with
precision 0. So It is possible to lose precision when get timestamp data from
Cassandra.
As for above two issues, Whether we should solve them at this PR or creat
new jira issues for them?
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
