Python Programming On Win32

Lazy Fetches

The examples in this chapter used two tables, one of clients and one of invoices, with a master-detail relationship. In terms of objects, you’d say that a client has a list of related invoices. However, it might be expensive to fetch these every time when a user might want to see them only some of the time.

Let’s implement a lazy fetch to get the data on demand. This Customer class inherits from our Record class. Given a database connection, it fetches its main attributes when explicitly asked, but the invoice list is retrieved on demand:

class Customer(Record):
    def __getattr__(self, key):
        #trap attempts to fetch the list of invoices
        if key == 'Invoices':
            self.fetchInvoices()
            return self.Invoices
        else:
            #call the inherited method
            return Record.__getattr__(self, key)

    def fetch(self, conn, key):
        self.conn = conn
        cursor = self.conn.cursor()
        cursor.execute("SELECT * FROM Clients \
                        WHERE ClientID = '%s'" % key)
        dicts = DataSetFromCursor(cursor).asDicts()
        assert len(dicts) == 1, 'Error fetching data!'
        self.loadFromDict(dicts[0])
        
    def fetchInvoices(self):
        #presumes an attribute pointing to the database
        cursor = self.conn.cursor()
        cursor.execute("SELECT * FROM Invoices \ 
                      WHERE ClientID = '%s'" % self.ClientID)
        ds = DataSetFromCursor(cursor)
        self.Invoices = ds.asObjects()

Using this class is straightforward:

>>> c = laundry.Customer() >>> c.fetch(conn, 'MEGAWAD') # assume an open connection >>> c.CompanyName # see if we got data... 'MegaWad Investments' >>> c.pp() # let's see all ...