I am trying to design an application using the Struts framework, however, I have come
across some contradicting information in the
documentation and was wondering what is the right way of implementing my design.
Suppose I have a Action object that calls a JavaBean to perform some business logic
returning a recordset to the action object. The
action object can then take this recordset save it in the appropriate scope for the
request and forward it onto the view to display
the results. (At least this is what I have understood of the Struts framework)
My problem is with the following statement, found in the User's Guide.
Section 1.3 ....an action object should pass the request to another object, usually a
JavaBean. To allow resuse on other platforms,
business-logic JavaBeans should not refer to any Web application objects. The action
object should translate needed details from the
HTTP request and pass those along to the business-logic beans as regular Java
variables....
Now in the Developers Guide under the section for Connection Pooling I found the
following.
DataSource dataSource = (DataSource)
servlet.getServletContext().getAttribute(Action.DATA_SOURCE_KEY);
conn = dataSource.getConnection();
... perform required functions as in the previous example ...
conn.close();
If this getConnection code was in my JavaBean then I have tied the JavaBean to a web
application object specifically the
servlet.getServletContext() which is a no-no.
Would I be better to allocate my connection in my Action object and then either set it
or pass it in to my JavaBean object. What
have people done to handle this situation.
Thanks
Scott Watson.
--
To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
