Raj Murthy

Greenhorn
+ Follow
since Oct 25, 2005
Merit badge: grant badges
For More
Cows and Likes
Cows
Total received
In last 30 days
0
Forums and Threads

Recent posts by Raj Murthy

One more note... we can ignore the c# client for now and assume that there will be 5 different servlet/jsp applications connecting to the business/persistence component.

Also, I have not used spring-framework, is there an answer to my question in spring-framework?

Does it allow you to bundle POJOs as services and deploy inside a container?

Thanks,
Raj
Zen,

Thanks for the reply. Here are the responses for your questions.


How maintainable it has to be?
>>>wanted to answer with a question.. .. depends on what maintainable means. If we do end up deploying the persistence component inside a container (such as web container) it can monitored & configured fairly easily.


How many web applications/clients will be accessing the persistence component (the torque thing deployed on a separate tomcat)?
>>>around 4-5

is it necessary that the separate persistence component be deployed on a separate machine?
>>>not 'necessary', but we will have 5 web apps & the persistence on same machine. we are looking to distribute the components

are they geographically separated?
>>>No

What are the security concerns?
>>>the persistence will be visible only to intended clients. So do not have big security clients.

How much transaction throughput must the persistence component be able to handle?
>>>the current web app has average response time of around 2-4 secs per page on average.
we are looking to keep this response time as close as possible.


is all client written in Java? Any technology constraint?
>>>4-5 clients will be servlet/jsp web applications; one will be C# based windows application.


TIA,
Raj

[ February 13, 2008: Message edited by: Raj Murthy ]
[ February 13, 2008: Message edited by: Raj Murthy ]
We are looking to deploy apache-torque based business/persistence layer as a separate component
that can be accessed by multiple clients. Currently we have it inside a war file.

We need to do this since we will have multiple web & non-web clients accessing the business/persistence layer.

One approach is to wrap the torque based business/persistence component as a REST service
and deploy it in a separate tomcat instance or as separate web-app.

We are using Tomcat for app server.

Any other ideas?

Thanks,
Raj
[ February 13, 2008: Message edited by: Rajiv Rangarajan ]
I moved my jboss server from c-drive to f-drive.
Now i get the error when i try to start the server from within eclipse:

org.eclipse.core.runtime.CoreException: Working directory does not exist: C:/jboss-4.0.2/bin

TIA,
Rajiv
Paul,
Yes, I did find the solution for the xml I posted
I have listed the same below for anyone interested.



Cheers,
Rajiv
Need to remove duplicates in a xml file based on a element's value.
Just use the first node with the element in the resulting list;

Here is a sample xml:



The result needs to be:

Title1 Tom
Title3 Jim

In this case use the first occurance of the 'Author' element.
How can i accomplish this in xslt?
Any hints on which xsl functions i can use for this?

Thanks in Advance,
Rajiv
[ December 28, 2005: Message edited by: Rajiv Rangarajan ]
Can you pls. clarify what you mean by java-application (stand-alone, web etc.) and also how the word document is used once it is fetched?

Depending upon on your security needs, you can expose the word doc in a web folder (through a web-server on windows) and try to fetch the document through a URL connection.

In case your 'java-app' is a web application, it would be much simpler, you will
just need a link to the windows web folder containing the word doc.
Thanks. Will give that a try.
[ October 28, 2005: Message edited by: Rajiv Rangarajan ]
So far we have been including the server and shared classes in both server and client
machines.
The client dynamically getting the stubs and shared-classes would be preferable.
What would be the performance implications, if any?
Please let me know if you post some command/code snippets that might help.

TIA,
Rajiv
19 years ago
Connection pooling (in the context of 'connection' referring to a connection to a service type such as database etc.) is specialization of object-pool pattern, a creational design pattern

An Object Pool manages the reuse of objects for a type of object that is expensive to create or only a limited number of a kind of object can be created. These instances of a class can be used interchangeably.

This can be used inconjunction with Factory-Method, Singleton and cache-management patterns.
Can think of two options...
*DOM
*JAXB - bind xml to java objects

HTH,
Rajiv
[ October 28, 2005: Message edited by: Rajiv Rangarajan ]
Thanks for your suggestions.

Here is the modified code(for anyone interested):



Not to prolong this thread, what if i use a JDOMSource instead of streamsource will it help the performance?
Since the calling program has the liberty of passing either xml-string of jdom source.

>>JDOMSource source = new JDOMSource(sourceDoc);

instead-of

>>StreamSource s1 = new StreamSource(new ByteArrayInputStream(strXml.getBytes()));


Cheers
Ulf,
Thanks for your response.

All,

I am able to read xml, apply xsl and convert it to html.
I have included the code below.

------------------------

-------------------------------------

Couple of notes on the same:

1. I am creating a new StreamReader from string to pass to transform method, not sure
if this is the most efficient way?

2. Rather than write to file, need to return the transform output as a String.

3. I can possibly optimize the xslSource by creating it only once, since this method will be called multiple times, sometimes possibly in a concurrent manner?


Any hints?

Thanks
Hello All,

We are working on a system that will get xml feed (through webservice or some other means.. to be decided) and display the results on a browser.

The platform is weblogic 8.1.3, we need to decide on the xml libraries to be used for parsing and displaying of a data, though weblogic offers certain jars for xml based processing we are thinking something more open source would be helpful incase we need to move the code to a different platform.

We are considering apache xerces.jar (for parsing the xml data), with xslt and xalan.jar for the transformations to convert it into html?

What is the latest/best versions of these two?

What are the alternatives to xerces and xalan, if any?

Any inputs on this will be greatly appreciated.

Thanks,
Rajiv