python dump sqlalchemy json array result into json file

user_6751456 0
Viewed 319

A postgres query returns json array and I am trying to dump this into json file.

query = f"""
           SELECT array_to_json(array_agg(results))
           FROM 
           (SELECT * from lookup_brands limit 1000) as results
        """ 
        json_results = exec_query(self.conn_string, query)
        print(json_results) # outputs <sqlalchemy.engine.cursor.LegacyCursorResult object at 0x7f30d64bd8b0>
        print(type(json_results) # outputs <class 'sqlalchemy.engine.cursor.LegacyCursorResult'>

I tried writing json_results to json like:

with open(output_path, 'w') as outfile:
     json.dump(json_results, outfile)

And I get this error

error Object of type LegacyCursorResult is not JSON serializable

which is expected.

How can I dump this json array result to the file?

sqlalchemy postgresql python json
1 Answers

SQLAlchemy's LegacyCursorResult object is an iterator over the resultset, and values in the results are converted to their Python equivalents. You need to create an object that can be serialised yourself from the results, for example:

with engine.connect() as conn:    
    result = conn.execute(sql)    
    print(json.dumps([dict(row) for row in result.mappings()]))

This seems wasteful, as Postgresql has already serialised the data. We can get around this by casting the query result to text on the database side, so that we receive it as a valid JSON str:

sql = """\
SELECT array_to_json(array_agg(results))::text AS data
FROM
(SELECT * FROM t70692551) AS results
"""

with engine.connect() as conn:
    result = conn.execute(sql)
    print(result.scalar_one())
user_53209060
Related