An Inprise employee told me that some of the database guys at Inprise were
asked by Sun to contribute to the JDBC 2.0 spec. The CachedRowSet certainly
seems like an idea that could of come out of those conversations. It is
almost identical to the implementation that JBuilder/Delphi support from the
DataExpress architecture.
In that article, Sun states that they will provide three implementations of
the CachedRowSet interface. One that they did not mention would be an
implementation that feed a large JTable or JList or JTree on demand in a
distributed environment. I believe the server side component of this
implementation would be well suited as a SessionBean. There is another
thread a couple of days ago that touched upon this concept:
"Large volume data exchange", by Gabby Tal <[EMAIL PROTECTED]> 8/23/99
jim
----- Original Message -----
From: Mike Walsh <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, August 26, 1999 4:09 PM
Subject: CachedRowSet: is anyone using it?
> Sun has made available an early access release of a CachedRowSet class
> which is an implementation of the Rowset interface which is part of
> the JDBC 2.0 Standard Extension API. I've seen the CachedRowSet
> mentioned once on this list, but, I haven't seen much discussion.
>
> The web page at http://developer.java.sun.com/developer/earlyAccess/crs
> describes the CachedRowSet as a "disconnected, serializable,
> scrollable container for tabular data" that can be "passed between
> different components of a distributed application, such as Enterprise
> JavaBeans(TM) components."
>
> The CachedRowSet is a lot like a disconnected ResultSet. A client
> invokes a method on an EJB Session Bean which then creates and
> populates a CachedRowSet from a ResultSet. The serializable
> CachedRowSet is passed back to the client where any number of rows of
> data are inserted, deleted or modified. The client then sends the
> CachedRowSet back to the server where an optimistic concurrency
> algorigthm resolves the modifications against some persistent storage.
>
> Do you consider this a valid part of an application architected
> around EJB?
>
> Perhaps it is a valid approach for an application domain that
> benefits by using Session Beans only?
>
> This use of the CachedRowSet sounds similar to "View Objects" in
> Oracle's JDeveloper 3.0 (see white paper at
> http://technet.oracle.com/product/tools/appjava/info/techwp20/wp.html,
> note: you'll need to get a free Oracle Technology Network
> Username/Password... sorry about that.) The white paper describes
> View Objects as fully-scrollable, fully-updateable, serializable
> rowsets that can be used just as I have described the CachedRowSet
> usage above. Warning: the white paper takes some serious jabs at
> Entity Beans.
>
> I haven't seen any commercial implementations of the CachedRowSet
> available and I haven't heard of any plans to produce one from several
> JDBC vendors that I have contacted.
>
> I hope to hear what everyone thinks of the CachedRowSet.
>
> -mike-
> Principal Software Engineer
> Riverton Software
>
>
===========================================================================
> To unsubscribe, send email to [EMAIL PROTECTED] and include in the
body
> of the message "signoff EJB-INTEREST". For general help, send email to
> [EMAIL PROTECTED] and include in the body of the message "help".
>
>
===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff EJB-INTEREST". For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".
