Gabriel Buheler

I think as long as it is UML 2.0 compliant, is perfectly fine.

Of course, if you know how to use it (that's a PLUS).

I'm one who used JMS 'cause that way is simple
to have response timeout - valuable feature in
network communications. Beside, container is able to manage
clients' threads while waiting for response, which makes
that solution scalable.

I do not like the idea for using JMS for this:
1. "Simple to have response time." Remember one of the SLA with TransMaster of no more than 3 sec during peak time.
2. Scalability is perfectly achieved by SLSB.
3. Thread-safe too.

MDB's, when dealing with asynchronous, clearly refer to requests that might take some long time to response. Do you really think that 3 sec, is a required?

Of course, you can still use JMS components. As long as you justify the reason, specially over other J2EE components.

Yes it is part of it. I would totally recommend you to take a look at it.

Definitely the answer is 4 and 2.

Choice 1: is not talking about scalability, that is fail over, and that depends on the implementation of the application server. Something not defined in specification.

Choice 3: does not exist with Session beans.
Choice 5: NA

Then the benefits of activate/passivate is that the EJB container would know when it performs these operations to be more efficient, when holding several instances.

And remember one of the benefits of EJB's are that they are thread-safe.

Gabriel

First congratulations for your final result.

How much detail did you add to the class diagram? what about others?
Can you give us a brief description of the kind of objects/patterns you used on each diagram?

Thanks

Synchronization is one of the services offered by the EJB container. That is, among others, the beauties that it can offer us.

Instead of worry about that, focus on the business requirement.

It seems fine to use JMS architecture.

I am assuming your system is going to be implemented within an intranet.

Regarding scability, it will offered by the EJB container when implement your MDB's. There are more options after a good design/architecture to do it more scalable.

Adding resources to your system, or clustering.

I will recommend publish-subscribe, with durable configuration.

Reliability is the ability to ensure consistency and integrity of the application and all its transaction.

Then, any means to improve this can be:

- J2EE services (yes like transaction manager)
- and also we can improve through horizontal scalability (adding more servers).

Then if we increase reliability, availability goes up too!

Regards,
Gabriel

I have a basic XSLT example below. First I will show you the xml, then the xsl sheet, and then the ouput.
All these works fine.
But when I try to use a variable name, is not working.

XML>
<page>
<group>ColumnB</group>
<rows>
<row>
<ColumnA>any value</ColumnA>
<ColumnB>id1</ColumnB>
<ColumnC>any vlaue</ColumnC>
</row>
<row>
<ColumnA>any value</ColumnA>
<ColumnB>id2</ColumnB>
<ColumnC>any value</ColumnC>
</row>
<row>
<ColumnA>any value</ColumnA>
<ColumnB>id1</ColumnB>
<ColumnC>any value</ColumnC>
</row>
</rows>
</page>

XSL>
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
<xsl:template match="/page">
<!-- Any stuff -->

<!-- BODY -->
<xsl:call-template name="template2">
<xsl:with-param name="data" select="rows/row" />
</xsl:call-template>

<!-- Any stuff -->
</xsl:template>

<xsl:template name="template2">
<xsl aram name="data" />

<xsl:variable name="filter-column" select="group"/>
<xsl:variable name="filter-identifier" select="$data[1]/*[name() = $filter-column]"/>
<xsl:variable name="filtered-data" select="$data[ColumnB = $filter-identifier]"/>

<xsl:text> </xsl:text>
<xsl:value-of select="$filter-column"/>
<xsl:text> </xsl:text>
<xsl:value-of select="$filter-identifier"/>
<xsl:text> </xsl:text>
<xsl:value-of select="count($filtered-data)"/>
</xsl:template>
</xsl:stylesheet>

OUTPUT:
ColumnB
id1
2

This is all fine, but where I try to use a variable. Instead of:
<xsl:variable name="filtered-data" select="$data[ColumnB = $filter-identifier]"/>

Use this(changing just ColumnB literal to its variable, that is the same content):
<xsl:variable name="filtered-data" select="$data[$filter-column = $filter-identifier]"/>

Does not work, because it comes up with this:
ColumnB
id1
0

Where the last number, should be 2.

Can any body help me with this?
Thanks!

You can also take a look at:

http://www.castor.org/

I think it is easier.

Rgrd,
Gabriel

Also you can look at this API:

XSOM

There it is few documentation, but I think that support included schemas.


Now,
Could you please help me with a schema parser?
Basically I need:

I want to parse a xsd file.

For example, �test.xsd� file:

<xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema">
<xsd:element name="note" type="NoteType"/>
<xsd:complexType name="NoteType">
<xsd:sequence>
<xsd:element name="to" type="xsd:string"/>
<xsd:element name="from" type="xsd:string"/>
<xsd:element name="heading" type="xsd:string"/>
<xsd:element name="body" type="xsd:string"/>
</xsd:sequence>
</xsd:complexType>
</xsd:schema>

And I want to print important elements:

Like:

Note
To
From
Heading
Body

How can I do that? Could your provide me with an example?

Thanks in advance,

Gabriel

Also you can evaluate this option:



FormattedDataSet

Gabriel

I want to parse a xsd file.

For example, �test.xsd� file:

<xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema">
<xsd:element name="note" type="NoteType"/>

<xsd:complexType name="NoteType">
<xsd:sequence>
<xsd:element name="to" type="xsd:string"/>
<xsd:element name="from" type="xsd:string"/>
<xsd:element name="heading" type="xsd:string"/>
<xsd:element name="body" type="xsd:string"/>
</xsd:sequence>
</xsd:complexType>
</xsd:schema>

And I want to print out basic elements:

Like:
Note
To
From
Heading
Body

How can I do that? Could your provide me with an example?

Thanks in advance,

Gabriel

Thank you for you help.

It really makes me get it in a clearer way.

Good luck,
Gabriel Enriquez

What's true about Container suppor for messagge-driven beans?

The Container must ensure that the bean instances are non-reentrant.

What does reentrant mean?
Is there other ejb type that use reentrant option?

thanks