1. From a directory that is on the CLASSPATH. The CLASSPATH of the JavaWebServer includes service_root/classes/ which is where the system classes reside.
2. From the /servlets/ directory. This is *not* in the server's classpath. A classloader is used to create servlets from this directory. New servlets can be added - existing servlets can be recompiled and the server will notice these changes.
3. From a remote location. For this a codebase like http://nine.eng/classes/foo/ is required in addition to the servlet's class name. Refer to the admin GUI docs on servlet section to see how to set this up.

For information on how to load servlets using the Administration Tool, see Servlet Loading.

Loading Remote Servlets

Remote servlets can be loaded by:

1. configuring the Admin Tool to setup automatic loading of remote servlets
2. setting up server side include tags in .shtml files
3. defining a filter chain configuration

• a virtual name that is assigned to the servlet via the Servlet Loading section of the Admin Tool.
• by its own class name if the servlet is dropped into the /servlets/ directory and invoked by its class name.

Servlets can be invoked by a client in the following ways:

• The client can ask for a document that is served by the servlet.

  The server gets a request for a document, looks up the configuration parameters, finds out that the document is not a static document residing on the disk, but a live document generated by a servlet object, and forwards the request to the servlet which services the request by generating the output. This method is the same as used by traditional web servers for invoking CGI scripts.

• The client (browser) can invoke the servlet directly using a URL, once it has been mapped using the Servlet Aliases section of the admin GUI.

  The client invokes the servlet with a URL of the form:

  http://server_host_name/servlet/<servlet URL>

  where the <servlet URL> is a regular URL that points to the location of the servlet. The host where servlet resides might be different from the host where the server is running. In this case the servlet class will be dynamically downloaded to the server, instantiated and then run. The reason this invocation is useful is that it does not need any special understanding of servlets on the part of the browser.

• The servlet can be invoked through server side include tags.

  Any file ending with .shtml is a server parsed file. In .shtml, if a servlet tag or server-insert tag is present, then the server will run the servlet and insert the output generated by the servlet in the place of the insert tag.

• The servlet can be invoked by placing it in the servlets/ directory.

  Servlets can be dropped in to the servlets/ directory relative to the root of the server. Servlets placed in this directory can be invoked using their class names. (For example, if HelloWorld.class was placed in the server_root/servlets directory):

  http://server_host_name:8080/servlet/HelloWorld

  A class_file.initArgs file can be placed in the same directory for passing init args to the servlet in question. For example to pass two init args A=B and C=D to the HelloWorld servlet. A HelloWorld.initArgs file must be created in the extensions dir whose syntax would be:

  
  	A=B
  	C=B
  

  When Servlets are recompiled in this directory - the new version will be automatically loaded by the server.

• The servlet can be invoked by using it in a filter chain

  This happens when a servlet is configured to be invoked when a particular mime type is set as the response. See the Filters and Servlet Chaining section for more on this topic.

Top