The Connection object allows us to establish connection sessions with data sources. In other words, the connection represents a physical connection to a data source. The connection we set up provides a mechanism for initializing and establishing the connection. Once the connection object is created we can use its methods and properties to use the data source (in this case the database) that we have connected to. We will also use the connection for executing queries and using transactions (we'll talk about these in detail later). The underlying OLE DB provider is used for connecting to the data source.
Let's briefly summarize the methods of the connection object:
Connection Object Methods | Description |
Open | Opens a new connection to a data source |
Close | Closes a connection as well as any dependant objects |
Execute | Executes a query, typically an SQL statement |
BeginTrans | Starts a new transaction |
CommitTrans | Commits changes made during the transaction |
RollBackTrans | Cancels any changes done during a transaction |
And here are the connection object's properties:
Connection Object Properties | Description |
ConnectionString | Contains the necessary information to establish a connection |
ConnectionTimeOut | Determines how long to wait to establish a connection. |
CommandTimeOut | Determines how long to wait while executing a command. |
State | Indicates if a connection is open. |
Provider | Indicates the name of the provider |
Version | Indicates the ADO version. |
CursorLocation | Sets/returns the location of the cursor engine. |
As we mentioned, the connection object allows us to establish sessions with data sources. But remember, behind the scenes the underlying OLE DB provider is actually used for connecting to the data source. That is how the magic is accomplished.
We program to the ADO interface. But remember when we selected the OLE DB provider when we selected "Build…" next to the Connection String option? It is the OLE DB provider that knows how to take commands from ADO and translate them into language that the specific data source will understand. Makes sense now, doesn't it? And as more and more suppliers create OLE DB providers for their data sources, ADO will take over the world.