Module query.database in plugin tabular v0.5.3

Execute a sql query against a (sqlite) database.

Author(s)	Markus Binsteiner (markus@frkl.io)
Tags	tabular
Python class	kiara_plugin.tabular.modules.db.QueryDatabaseModule

Configuration

Module configuration options

Configuration class: kiara_plugin.tabular.modules.db.QueryDatabaseConfig

Name	Description	Type	Required?	Default
constants	Value constants for this module.	object	false	null
defaults	Value defaults for this module.	object	false	null
query	The query.	anyOf: [{'type': 'string'}, {'type': 'null'}]	false	null

Source code

Module source code

class QueryDatabaseModule(KiaraModule):
    """Execute a sql query against a (sqlite) database."""

    _config_cls = QueryDatabaseConfig
    _module_type_name = "query.database"

    def create_inputs_schema(
        self,
    ) -> ValueMapSchema:

        result: Dict[str, Dict[str, Any]] = {
            "database": {"type": "database", "doc": "The database to query."}
        }

        if not self.get_config_value("query"):
            result["query"] = {"type": "string", "doc": "The query to execute."}

        return result

    def create_outputs_schema(
        self,
    ) -> ValueMapSchema:

        return {"query_result": {"type": "table", "doc": "The query result."}}

    def process(self, inputs: ValueMap, outputs: ValueMap):

        import pyarrow as pa

        database: KiaraDatabase = inputs.get_value_data("database")
        query = self.get_config_value("query")
        if query is None:
            query = inputs.get_value_data("query")

        # TODO: make this memory efficent

        result_columns: Dict[str, List[Any]] = {}
        with database.get_sqlalchemy_engine().connect() as con:
            result = con.execute(text(query))
            for r in result:
                for k, v in dict(r).items():
                    result_columns.setdefault(k, []).append(v)

        table = pa.Table.from_pydict(result_columns)
        outputs.set_value("query_result", table)