📘 Overview

The SQLExecutionFilter is an operator designed to filter data entries by executing their associated SQL queries against a specified database. It retains only the entries for which the SQL query executes successfully, effectively cleaning the dataset by removing invalid or non-executable queries.

__init__ function

def __init__(self, database_manager: DatabaseManager):

init parameter description

Parameter Name	Type	Default Value	Description
database_manager	DatabaseManager	Required	An instance for managing database connections and executing queries.

Prompt Template Descriptions

Prompt Template Name	Primary Use	Applicable Scenarios	Feature Description

run function

def run(self, storage: DataFlowStorage, input_sql_key: str = "sql", input_db_id_key: str = "db_id")

Parameters

Name	Type	Default Value	Description
storage	DataFlowStorage	Required	DataFlow storage instance for reading and writing data.
input_sql_key	str	"sql"	The column name in the input data that contains the SQL queries.
input_db_id_key	str	"db_id"	The column name in the input data that contains the database identifiers.

🧠 Example Usage

🧾 Default output format (Output Format)

The operator filters the input DataFrame, so the output format consists of the same columns as the input, but only contains the rows where the SQL query was successfully executed.

Field	Type	Description
...		Original columns from the input are preserved.
sql	str	The SQL query that was successfully executed.
db_id	str	The database ID against which the query was executed.