while looking a the docs I noticed something that caught my attention
before:
<web-logic-docs>
By embedding a simple Java expression in the WLQL string, you can convert
the supplied maxBal value to another currency before querying the RDBMS. For
example, if maxBal is supplied in U.S. dollars and the conversion rate to
pounds is 1.6483, you can use a simple expression to multiply the value:
<finder>
<method-name>findSomeAccounts</method-name>
<method-params>
<method-param>double</method-param>
<method-param>string</method-param>
</method-params>
<finder-query>(& (< balance $0) (= owner $1))</finder-query>
<finder-expression>
<expression-number>0</expression-number>
<expression-text>@0 * 1.6483</expression-text>
<expression-type>long</expression-type>
</finder-expression>
</finder>
In the above example, $0 is replaced by the Java expression @0 * 1.6483,
which multiplies the value of maxBal by 1.6483. Because the EJB provider did
not override the value of $1, WLQL maps $1 to the second parameter in the
finder method signature, ownerID.
A more advanced version of this finder could use Java to determine the
conversion rate when converting maxBal:
<finder-expression>
<expression-number>0</expression-number>
<expression-text>@0 *
Double.parseDouble(System.getProperties().get("rate.pounds.dollars"))</expre
ssion-text>
<expression-type>long</expression-type>
</finder-expression>
</web-logic-docs>
I previously posted a message regarding improving and standardizing the
finder syntax and was hoping for some feedback but alas none came. So I will
*ignite* the issue again in this thread. Question: Is it just me but have
the web logic guys just totally lost it. I am all for impoving the
capabilities of the syntax like accessing a field within an object passed
across as a parameter (based on java beans design pattern) but I think we
are starting to push things just the the edge of silliness when we allow for
the syntax to include :
Double.parseDouble(System.getProperties().get("rate.pounds.dollars")) while
still keeping the ridiculus prefix notation : >(& (< balance $0) (= owner
$1)). Maybe I am old fashion but should not the finder syntax reasonable SQL
considering that most ejb installations will have a relational database
backend and most deployers would have knowledge of this over Java's api. I
think the web logic server team's time would have been better spent in
getting rid of the crappy prefix syntax - *we are not computers and do not
process instructions this way*.
To the web logic users who have used the features mentioned above. Could you
tell me why not include the * 1.6483 in the original query syntax is this
some limitation which was addressed in this convulated way or just an
example demonstrating a unwarranted feature.
William Louth
> -----Original Message-----
> From: Louth, William (Exchange)
> Sent: Tuesday, April 18, 2000 8:29 PM
> To: 'A mailing list for Enterprise JavaBeans development'
> Subject: RE: failover of stateful session beans in WebLogic
>
> as per the web logic 5.1 documentation
>
> http://www.weblogic.com/docs51/classdocs/API_ejb/EJB_environment.html
>
> ===========================================
> Stateful session EJBs
>
> Stateful session EJBs can utilize cluster-aware home stubs by setting
> home-is-clusterable to "true." This provides failover and load balancing
> for stateful EJB lookups. Stateful session EJBs cannot utilize
> replica-aware EJBObject stubs, and WebLogic Server does not provide
> failover services for method calls to stateful session EJBs.
>
> If you require cluster failover services for stateful objects, consider
> implementing the stateful session EJB as a servlet. Servlets can maintain
> state through failover in a cluster using either JDBC, an operating system
> file, or directly in memory. See Using session tracking from a servlet for
> more information.
>
> ===========================================
>
> I do not have any idea if there are plans. I suspose this will come with
> the next release possibly next year (ejb 2.0) since this release has still
> to come out of beta.
>
> William Louth
>
> -----Original Message-----
> From: John Hogan [SMTP:[EMAIL PROTECTED]]
> Sent: Tuesday, April 18, 2000 8:00 PM
> To: [EMAIL PROTECTED]
> Subject: failover of stateful session beans in WebLogic
>
> All,
>
> This feature seems to be missing from WebLogic and their documentation
> suggests achieving failover protection using servlet/session. Am I
> understanding this right, and does anyone have information on WebLogic's
> plans to support stateful session bean failover? Thanks.
>
> JohnH
>
> ==========================================================================
> =
> 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".
***********************************************************************
Bear Stearns is not responsible for any recommendation, solicitation,
offer or agreement or any information about any transaction, customer
account or account activity contained in this communication.
***********************************************************************
===========================================================================
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".
