[Tutor] creating a connection class

[Alan Gauld]

On 30/03/18 15:55, Glenn Schultz wrote:

> I can create a connection as follows and it works 

Well done, so you get a connection object back as a result.

> but I think it is best to have a connection class that opens and closes.  

Why? You already have a connection object.
Classes are created so that you can create instances.
What would an instance of your Connection class offer
that the API connection object does not?

Creating classes introduces significant effort and
overhead, if you don't need them don't build them.
If the instances of your class don't contribute
some new functionality what is their point? The
more code you create the more likely that there
is a bug.

> model_prod_conn = pyodbc.connect(
> driver = '{ODBC Driver 13 for SQL Server}',
> server = 'foo',
> database = 'foo',
> username = 'foo',
> password = 'foo',
> trusted_connection = 'yes')
> 
> 
> class modelingdb(object):
>   def __init__(self):
>    self._db_connection = pyodbc.connect(
>    driver = '{ODBC Driver 13 for SQL Server}',
>    server = 'foo',
>    database = 'foo',
>    username = 'foo',
>    password = 'foo',
>    trusted_connection = 'yes')
>    self._db_cur = self._db_connection.cursor()
> 
> def __del__(self):
> self._db_connection.close()

There is a great talk somewhere on YouTube that
provides guidance on when not to create a class.
One of the cases is when you only have constructor
and destructor.

You may want to continue as an academic exercise in
wrapping an API object. That's fair enough. But your
current class offers nothing that the API does not.
(The db_cur attribute doesn't really count since
it's just a call to the connection object anyway.


-- 
Alan G
Author of the Learn to Program web site
http://www.alan-g.me.uk/
http://www.amazon.com/author/alan_gauld
Follow my photo-blog on Flickr at:
http://www.flickr.com/photos/alangauldphotos