Comments for Java Blog - Java, J2EE, SOA, Spring and Hibernate http://javablog.info Wed, 10 Mar 2010 03:09:04 +0000 http://wordpress.org/?v=2.3.3 Comment on Search Pagination Design (Server side pagination using Hibernate) by Ravi Nallakukkala http://javablog.info/2007/03/30/search-pagination-design-server-side-pagination-using-hibernate/#comment-11 Ravi Nallakukkala Wed, 13 Jun 2007 07:36:37 +0000 http://javablog.info/2007/03/30/search-pagination-design-server-side-pagination-using-hibernate/#comment-11 Hi Gaurav, Answer for the first part of the question is dependent on your deployment structure. Lets take a simple example say you have 1 server instance(X) running with your server sider code (EAR containing EJB, Dao, utils) and M server instances with GUI code(War); Say EAR and War's are deployed in different machines and M GUI instance talk to the same server side deployed code. Now in the Criteria Caching based pagination on Server side, This server side code deployed instance (X) would hold all the (unique) criteria and its results on the server instance, where as in the Pagination on UI front approach, as the caching happens at UI front so each UI server instance will hold the search results. Towards your second question, slightly not clear with your question, you said you wanna to provide pagination and not to increase the memory on weblogic (no mention of any caching) sounds to me like "Real time pagination on the server side" approach; which is the first approach we discussed in the article. Say without caching if you wish to increase the page load time, may be you could consider using AJAX XML data for displaying the results and using real time pagination approach retrieve the page results and (re)populate the XML data. this way you could avoid the JSP page load time, which is quite significant in most the cases. Did it answer your second question? Remember the thumb rule "Sacrifice memory for time and vice-versa" :-) Thanks, Ravi Hi Gaurav,

Answer for the first part of the question is dependent on your deployment structure. Lets take a simple example say you have 1 server instance(X) running with your server sider code (EAR containing EJB, Dao, utils) and M server instances with GUI code(War); Say EAR and War’s are deployed in different machines and M GUI instance talk to the same server side deployed code.

Now in the Criteria Caching based pagination on Server side, This server side code deployed instance (X) would hold all the (unique) criteria and its results on the server instance, where as in the Pagination on UI front approach, as the caching happens at UI front so each UI server instance will hold the search results.

Towards your second question, slightly not clear with your question, you said you wanna to provide pagination and not to increase the memory on weblogic (no mention of any caching) sounds to me like “Real time pagination on the server side” approach; which is the first approach we discussed in the article.

Say without caching if you wish to increase the page load time, may be you could consider using AJAX XML data for displaying the results and using real time pagination approach retrieve the page results and (re)populate the XML data. this way you could avoid the JSP page load time, which is quite significant in most the cases.

Did it answer your second question?

Remember the thumb rule “Sacrifice memory for time and vice-versa” :-)

Thanks,
Ravi

]]> Comment on Search Pagination Design (Server side pagination using Hibernate) by Gaurav Bahl http://javablog.info/2007/03/30/search-pagination-design-server-side-pagination-using-hibernate/#comment-10 Gaurav Bahl Wed, 13 Jun 2007 06:35:41 +0000 http://javablog.info/2007/03/30/search-pagination-design-server-side-pagination-using-hibernate/#comment-10 Hi, I couldn't understand the diffrence between Criteria Caching based pagination on Server side and Pagination on the UI front. If we are using session variable to save the result then both are at server side only. Or are you talking about caching in database buffer. Currently i am in the same situation, but i don't want to increase the memory footprint at my weblogic server but want to provide pagination for my results. Can you tell me the way for that and please send me some details of that. Gaurav Hi,
I couldn’t understand the diffrence between Criteria Caching based pagination on Server side and Pagination on the UI front. If we are using session variable to save the result then both are at server side only. Or are you talking about caching in database buffer.

Currently i am in the same situation, but i don’t want to increase the memory footprint at my weblogic server but want to provide pagination for my results. Can you tell me the way for that and please send me some details of that.

Gaurav

]]> Comment on Equals and hashcode Approach/ Implementation for Business Objects (Bo) by Ravi Nallakukkala http://javablog.info/2007/04/15/equals-and-hashcode-implementation-for-business-objects-bo/#comment-8 Ravi Nallakukkala Mon, 07 May 2007 23:55:12 +0000 http://javablog.info/2007/04/15/equals-and-hashcode-implementation-for-business-objects-bo/#comment-8 Well, The recommendation was use a combination of Technical and business keys (and "not" Technical keys alone) "Include the Technical Keys and Business keys in equals and hashcode methods." Worst case scenario will be when your business keys are not unique. Thanks Well, The recommendation was use a combination of Technical and business keys (and “not” Technical keys alone)

“Include the Technical Keys and Business keys in equals and hashcode methods.”

Worst case scenario will be when your business keys are not unique.

Thanks

]]> Comment on Equals and hashcode Approach/ Implementation for Business Objects (Bo) by Daniel http://javablog.info/2007/04/15/equals-and-hashcode-implementation-for-business-objects-bo/#comment-7 Daniel Mon, 07 May 2007 23:39:57 +0000 http://javablog.info/2007/04/15/equals-and-hashcode-implementation-for-business-objects-bo/#comment-7 Are you sure about the undesired results. Even Hibernate recommends using only the "business key" for equals and hashcode [1]. [1] <a href="http://www.hibernate.org/109.html" rel="nofollow">http://www.hibernate.org/109.html</a> Are you sure about the undesired results. Even Hibernate recommends using only the “business key” for equals and hashcode [1].

[1] http://www.hibernate.org/109.html

]]> Comment on Equals and hashcode Approach/ Implementation for Business Objects (Bo) by Ravi Nallakukkala http://javablog.info/2007/04/15/equals-and-hashcode-implementation-for-business-objects-bo/#comment-6 Ravi Nallakukkala Mon, 07 May 2007 23:10:29 +0000 http://javablog.info/2007/04/15/equals-and-hashcode-implementation-for-business-objects-bo/#comment-6 Hey Daniel, Greeting to you from hot summer land; Phoenix :) Say, we have two object with identical values; one object stored in Database (with technical key) and another which is not stored in database (without a technical key). In this case, if you don't include technical keys in hash code and equals methods, Hibernate execution will lead to undesired results. I totally agree with if removing technical keys in equals/ hash code if I'm working on Dto objects than a Business Objects. If you are concern about handling of null cases, commons-lang package takes care of it. Thanks Hey Daniel,

Greeting to you from hot summer land; Phoenix :)

Say, we have two object with identical values; one object stored in Database (with technical key) and another which is not stored in database (without a technical key). In this case, if you don’t include technical keys in hash code and equals methods, Hibernate execution will lead to undesired results.

I totally agree with if removing technical keys in equals/ hash code if I’m working on Dto objects than a Business Objects.

If you are concern about handling of null cases, commons-lang package takes care of it.

Thanks

]]> Comment on Equals and hashcode Approach/ Implementation for Business Objects (Bo) by Daniel http://javablog.info/2007/04/15/equals-and-hashcode-implementation-for-business-objects-bo/#comment-5 Daniel Sun, 06 May 2007 01:07:15 +0000 http://javablog.info/2007/04/15/equals-and-hashcode-implementation-for-business-objects-bo/#comment-5 I wouldn't include the technical key in hashcode calculation or equals comparison. Sometimes the object to check is not stored in the database so technical key is not set (null) but the business key can be equal so it won't work. Greetings from Atlanta :-) I wouldn’t include the technical key in hashcode calculation or equals comparison. Sometimes the object to check is not stored in the database so technical key is not set (null) but the business key can be equal so it won’t work.

Greetings from Atlanta :-)

]]> Comment on Things I hate in IntelliJ compared to Eclipse by Ravi Nallakukkala http://javablog.info/2007/04/11/things-i-hate-in-intellij-compared-to-eclipse-ide/#comment-3 Ravi Nallakukkala Thu, 12 Apr 2007 23:12:24 +0000 http://javablog.info/2007/04/11/things-i-hate-in-intellij-compared-to-eclipse-ide/#comment-3 Thanks Dmitry, was a handy tip. Could play around writing my own templates now. Support!!! That's one really good thing about intelliJ, I missed out (After being in Eclispe world for long :-) ). Cheers Thanks Dmitry, was a handy tip. Could play around writing my own templates now.

Support!!!
That’s one really good thing about intelliJ, I missed out (After being in Eclispe world for long :-) ).

Cheers

]]> Comment on Things I hate in IntelliJ compared to Eclipse by Dmitry Jemerov http://javablog.info/2007/04/11/things-i-hate-in-intellij-compared-to-eclipse-ide/#comment-2 Dmitry Jemerov Thu, 12 Apr 2007 09:22:39 +0000 http://javablog.info/2007/04/11/things-i-hate-in-intellij-compared-to-eclipse-ide/#comment-2 Hello Ravi, Actually IntelliJ IDEA does not expect that you would configure the classpath twice. It uses the same classpath for compile and run/debug, just as you would expect. To configure the code templates for new files created in the IDE, you can use Settings | File Templates. To configure code snippets which can be inserted into existing files, you can use Settings | Live Templates (which has a number of predefined templates which you can use as examples for creating your own templates). If you have any questions regarding IntelliJ IDEA, don't hesitate to contact our support at http://www.jetbrains.com/support/idea/index.html Hello Ravi,

Actually IntelliJ IDEA does not expect that you would configure the classpath twice. It uses the same classpath for compile and run/debug, just as you would expect.

To configure the code templates for new files created in the IDE, you can use Settings | File Templates. To configure code snippets which can be inserted into existing files, you can use Settings | Live Templates (which has a number of predefined templates which you can use as examples for creating your own templates).

If you have any questions regarding IntelliJ IDEA, don’t hesitate to contact our support at http://www.jetbrains.com/support/idea/index.html

]]>