Insert Conflicts

Conflict handling decides whether an insert matching an existing row should fail, be ignored, or update that row. Typical scenarios include replayed jobs, data synchronization, repeated imports, and insert-or-update operations. It is an INSERT execution strategy, not a separate ordinary UPDATE.

Strategies

Strategy	Oracle statement	Result
Into	INSERT INTO ... VALUES (...)	The database reports a conflict
Ignore	MERGE ... WHEN NOT MATCHED THEN INSERT	Matched rows remain unchanged; unmatched rows are inserted
Update	MERGE ... WHEN MATCHED THEN UPDATE ... WHEN NOT MATCHED THEN INSERT	Matched rows are updated; unmatched rows are inserted

MERGE builds source data with SELECT ... FROM dual and uses mapped primary-key columns as its match condition.

Usage

int rows = lambda.insert(UserInfo.class)
        .onDuplicateStrategy(DuplicateKeyStrategy.Update)
        .applyEntity(user)
        .executeSumResult();

Use Ignore to skip matched rows. Omitting the strategy or selecting Into produces a regular INSERT. See Insert Operations for the common API.

Notes

• Ignore and Update require an entity primary-key mapping; dbVisitor does not select another business unique key.
• Mapping a primary key does not create a database constraint. Match columns should identify at most one row.
• Ignore skips matched rows only; other database errors still fail.
• The database executes one MERGE. Whether several entity writes commit together still depends on the transaction boundary.