Python module establishing database connection without __main__

Question

I am moving the code that interacts with my Neo4J database into a separate python module so I can stop repeating myself in other modules.

The problem I am having is that in each function call in the new module I am having to have a separate call to...

db = Graph('http://localhost/db/data')

...to establish a connection to the database.

This seems really silly and is not solving my goal of reducing the amount of unnecessary code.

Normally, I would establish the connection in the main function but because this module is being called from elsewhere I can't do this.

I am looking for a way of establishing a local variable that will persist between function calls, so I can forget about connecting to the db.

Answer 1

Apart from code bloat you are making it extremely inefficient by having every function connect to a database. If you want to stick with procedural programming, then make it a global variable. With classes you can either connect in the __init__ method or pass database connection to __init__ as an argument. Most of my classes support both:

class A:
   def __init__(self, ..., db_connection=None):
       self.db_connection = db_connection or DbConnection()
       ...

    def f(self):
       self.db_connection("SELECT A from B")