Codegen CLI 🔗

The squery-cli_2.13 artifact runs ba.sake.squery.cli.SqueryMain. JDBC drivers must be supplied on the launch classpath. The generated code targets Scala 3.

cs launch org.xerial:sqlite-jdbc:3.46.1.0 \
  ba.sake:squery-cli_2.13:0.13.0 \
  -M ba.sake.squery.cli.SqueryMain -- --help

Options 🔗

Names below match the current CLI's --help, including the mixed spelling of explicitly named options. Camel-case aliases such as --jdbcUrl are not the help-listed options. --ironNewtypes is not supported; use --newtypes iron.

Option	Default	Value
--jdbc-url	Required	JDBC connection URL
--schema-mappings	Empty	Repeatable schema:package mapping; supply one mapping per occurrence
--base-folder	src/main/scala	Output source root
--col-name-identifier-mapper	camelcase	camelcase or noop
--type-name-mapper	camelcase	camelcase or noop
--row-type-suffix	Row	Row class suffix
--dao-type-suffix	Dao	DAO class suffix
--typeMappingRule	Empty	Repeatable column-name-regex|declared-type-regex|Scala-type rule
--includeTables	All tables	Repeatable regex matching schema.table
--excludeTables	Empty	Repeatable regex matching schema.table; exclusions win
--newtypes	none	none or iron; iron requires the application dependency squery-iron
--help	—	Display usage

An empty schema mapping list generates no schemas. SQLite's schema name is main. Output packages contain models and daos directories.

Matching rules 🔗

Type mapping rules match the complete column name and the JDBC driver's complete declared type. Matching is case-sensitive unless the regex enables an option such as (?i:...). The first matching rule wins, before built-in mappings. Use fully qualified Scala type names for types outside the generated imports:

--typeMappingRule '.*_id|(?i:TEXT)|java.util.UUID'

Table filters also match complete names. A table must match an include pattern and no exclude pattern. Omitting includes selects all tables. For example:

--includeTables 'main\..*' \
--excludeTables 'main\.flyway_schema_history' \
--excludeTables 'main\.sqlite_.*'

With Iron keys, referenced tables must be included so their PK types can be resolved. Excluding a referenced table causes generation to fail before writing files. Use --newtypes none when primitive keys are required instead.

See Generating Code for a complete SQLite example.