Updates, generated keys, and batches

JdbcClient supports write operations, generated keys, and efficient batch updates without leaving the fluent API. The examples below cover the common patterns used when modifying data in an application.

Updating data

Use update() for SQL statements such as INSERT, UPDATE, and DELETE. The method returns the number of rows affected.

int rows = client
        .sql("UPDATE engineers SET dev_name = :name WHERE id = :id")
        .param("name", "Ada")
        .param("id", 1L)
        .update();

Handling generated keys

When inserting a row and generated keys are required, pass a GeneratedKeyHolder to the update call.

GeneratedKeyHolder keys = new GeneratedKeyHolder();

int rows = client
        .sql("INSERT INTO engineers (dev_name) VALUES (:name)")
        .param("name", "Ada")
        .update(keys);

Object generatedKey = keys.getKey();
Map<String, Object> firstRow = keys.getKeys();
List<Map<String, Object>> allRows = keys.getKeyList();

The JDBC driver must support generated keys. getKey() returns the first column from the first generated-key row, while getKeys() and getKeyList() preserve column labels and multiple rows.

Executing batch updates

Use a list of named-argument maps for batch updates:

int[] results = client
        .sql("UPDATE engineers SET active = :active WHERE id = :id")
        .batchUpdate(List.of(
                Map.of("active", true, "id", 1L),
                Map.of("active", false, "id", 2L)
        ));

For positional SQL, pass an array for each batch entry:

int[] results = client
        .sql("UPDATE engineers SET active = ? WHERE id = ?")
        .batchUpdate(new Object[][]{
                {true, 1L},
                {false, 2L}
        });

These patterns keep write operations compact and consistent while exposing the database-specific capabilities needed for generated keys and bulk updates.

For query execution, parameter binding, and result mapping, see querying and result mapping.