Debugging

Viewing generated SQL

For each supported database there is a method to directly generate SQL in that dialect for the purpose of inspection. These methods can be used without instantiating a DataSource. This makes them perfect for locally inspecting generated SQL in your IDE or in unit tests.

Dialect	Method	Import
MySQL	PerformableSql::generateMysqlSql	import io.koalaql.mysql.generateMysqlSql
Postgres	PerformableSql::generatePostgresSql	import io.koalaql.postgres.generatePostgresSql
H2	PerformableSql::generateH2Sql	import io.koalaql.h2.generateH2Sql

Example

Here is an example of using generatePostgresSql to print some generated SQL

Code

val generated: CompiledSql? = CustomerTable
    .where(CustomerTable.id eq 123)
    .generatePostgresSql()

println(generated?.parameterizedSql)

Output

SELECT T0."id" c0
, T0."shop" c1
, T0."name" c2
, T0."spent" c3
FROM "Customer" T0
WHERE T0."id" = ?

note

The return type is nullable because certain Koala statements may be no-ops. An example is attempting to insert with an empty list of values.

Using a DataSource

The methods above are handy if you want to quickly see the generated SQL for a dialect. If you want to see the exact SQL generated by a DataSource at runtime, you can use the generateSql method.

Code

val generated: CompiledSql? = CustomerTable
    .where(CustomerTable.id eq 123)
    .generateSql(db) /* pass our DataSource */

println(generated?.parameterizedSql)

Output

SELECT T0."id" c0
, T0."shop" c1
, T0."name" c2
, T0."spent" c3
FROM "Customer" T0
WHERE T0."id" = ?