> Hm, shouldn't transactions last only as short as possible? We are using a > system where the transaction is opened and closed in a wrapper for the > execute() method, but I always thought this is a design flaw ... ?
I would have thought that in most applications, the time spent processing a user request is dominated by the time spent doing data access (not the time spent rendering a JSP), in which case holding the transaction open while forwarding to the JSP would not be a huge impact. > Writing a filter would extend the time the transaction is open again. Isn't > there a better solution? I have no experience with EJB, but don't you always > have a transaction automatically opened? How do they handle that? This kind of approach would be relevant to use of Hibernate in a 2-tiered architecture (ie. the controller accesses the business domain model directly - and data access is done in the servlet engine process). With a 3-tiered architecture (eg. using session beans) you certainly *can't* do data access while rendering the view (that breaks the application layering). ------------------------------------------------------- This sf.net email is sponsored by: See the NEW Palm Tungsten T handheld. Power & Color in a compact size! http://ads.sourceforge.net/cgi-bin/redirect.pl?palm0001en _______________________________________________ Opensymphony-webwork mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/opensymphony-webwork
