Mailing-List: contact jetspeed-dev-help@jakarta.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Jetspeed Developers List" <jetspeed-dev@jakarta.apache.org>
Message-ID: <20594216.1087217300286.JavaMail.apache@nagoya>
Date: Mon, 14 Jun 2004 05:48:20 -0700 (PDT)
From: jira@apache.org
To: jetspeed-dev@jakarta.apache.org
Subject: [jira] Commented: (JS2-74) Refactor PAM and Descriptor Utilities
In-Reply-To: <1390748211.1086791170010.JavaMail.apache@nagoya>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

The following comment has been added to this issue:

     Author: Ate Douma
    Created: Mon, 14 Jun 2004 5:47 AM
       Body:
I finally found out the problem with the web.xml infusion.
The insertion point was based on the element index, *not* the content index. These two can, and usually are (comments, white space etc.) are !=.

Furthermore, added (again!) an EntityResolver to be able to build/run offline. 
Not just with maven -o mind you, really offline without a network connection!
I had the same problem with TestPortletDescriptor so I added an EntityResolver, looking up the web.xml dtd (web-app_2_3.dtd) locally using the PortletApplicationWar class.

I kindly request future new xml parsing code also to be tested without a network connection (or blocked/disabled whatever) before it is committed because at my client I'm completely blind from the internet. 
---------------------------------------------------------------------
View this comment:
  http://issues.apache.org/jira/browse/JS2-74?page=comments#action_36141

---------------------------------------------------------------------
View the issue:
  http://issues.apache.org/jira/browse/JS2-74

Here is an overview of the issue:
---------------------------------------------------------------------
        Key: JS2-74
    Summary: Refactor PAM and Descriptor Utilities
       Type: Improvement

     Status: Closed
   Priority: Major
 Resolution: FIXED

    Project: Jetspeed 2
 Components: 
             Deployment
   Fix Fors:
             2.0-dev/cvs
             2.0-a1
   Versions:
             2.0-dev/cvs
             2.0-a1

   Assignee: Scott T Weaver
   Reporter: Scott T Weaver

    Created: Wed, 9 Jun 2004 7:25 AM
    Updated: Mon, 14 Jun 2004 5:47 AM
Environment: Mandrake Linux 10, Tomcat 4.1.30, HSQL

Description:
I am refactoring all of the Jetspeed desccriptor utility classes from static, utility classes into objects.

PortletDescriptorUtilities has been renamed to PortletApplicationDescriptor.

WebDescriptorUtilities has been refactored into WebApplicationDescriptor.

JetspeedDescriptorUtilities has been refactored into ExtendedPortletMetadata.

Besides the rename, the static methods have been converted into instance methods.

I have created a composite object, PortletApplicationWar, that uses these three metadata classes to build up registry objects from a war file, which is either an actual war file itself OR a war-like structure on the file system.  I am using commons-vfs to manipulate the WAR as this allows me to work on the war as a FileObject, regardless if it is a file system directory or a WAR file.

All these descriptor classes have been moved to the org.apache.jetspeed.util.descriptor package.

FileSystemPAM uses the PortletApplicationWar class exclusively instead of directly using the metadata classes themselves.  This makes the FSPAM code much more readable and easier to debug.

Looking over the logic for processWebXML, it does not appear to be adding the JetspeedContainer servlet nor its mapping.  Also logic used to decide where to put the new elements could possibly cause the elements to be placed in the wrong spot relative to the DTD definition.

I rewrote this logic using JDom and XPath expressions.  I am using an XPath query to check for a specific servelet and servlet-mapping with the servlet-name "JetspeedContainer".  If these elements do not exist, I use JDom to walk the top-level of elements until it finds the correct location, per the DTD, to insert both the servlet and servlet-mapping elements. 

I have added an additional test, testInfuseWebXML, to TestPortletDescriptor to verify that the infusing works correctly. 


---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.

If you think it was sent incorrectly contact one of the administrators:
   http://issues.apache.org/jira/secure/Administrators.jspa

If you want more information on JIRA, or have a bug to report see:
   http://www.atlassian.com/software/jira


---------------------------------------------------------------------
To unsubscribe, e-mail: jetspeed-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jetspeed-dev-help@jakarta.apache.org