Null parameter value handling improvement

[matthewadams]

Null handling in query method parameters doesn't meet our needs, because it seems to always produce an "= NULL" clause in our queries, which always returns false. We need null parameter values to be ignored in the query.

FYI, I've entered
https://jira.springsource.org/browse/DATAJPA-209
to track our enhancement request.

Basically, it would allow something like this (notice @NullMeans annotation):

public interface FooRepository extends JpaRepository<Foo> {
  List<Foo> findByBarAndGoo(@NullMeans(IGNORED) Bar bar, @NullMeans(IGNORED) Goo goo);
}

Allowed values of @NullMeans are EQUALS (current behavior), IS (proposed) and IGNORED (proposed).

WDYT?

-matthew

[matthewadams]

NB: this is a shameless troll for votes since more folks follow the forum than the issue tracker.

The "= NULL" v. "IS NULL" issue is indeed fixed in 1.1.0.RC1. However, the enhancement proposed in https://jira.springsource.org/browse/DATAJPA-209 (controlling null behavior) still holds, and the enum values are reduced to just IS and IGNORED.

The SD JPA team is waiting for community feedback about and votes for the issue. If you feel that being able to control null handling behavior in queries with @NullMeans(IS) to produce "IS NULL" conditions (the current behavior) or @NullMeans(IGNORED) to exclude the condition from the query, then please vote for https://jira.springsource.org/browse/DATAJPA-209.

-matthew