Developer's Guide

CHAPTER 3: SESSION API VOICEXML PROGRAMMING GUIDE

FOR CISCO UNIFIED CUSTOMER VOICE PORTAL RELEASE 4.0(1)

Creating session data. This class does not allow the creation of element data because only

elements can do so (the start of call class cannot, for example).

Triggering custom logging events and warning events that are picked up by loggers.

Setting the maintainer, default audio path, application language and encoding, as well as the

call’s session timeout. At any point in the application, these settings can be changed.

Accessing the Global API to get and set application and global data (see the VoiceXML

Server User Guide for Cisco Unified Customer Voice Portal Chapter 2 in the section entitled

Variables for more on application and global data).

The following lists the API classes that are used for various components (also found in the

com.audium.server.session package). A detailed description of what each class provides is

given in the individual section for that component.

CallStartAPI. This class is sent as an argument to the start of call class.

CallEndAPI. This class is sent as an argument to the end of call class.

ElementAPI. This class is used by all standard and configurable element classes as well as

dynamic configuration classes. The following classes extend ElementAPI to provide

additional functionality required for different kinds of elements.

o ActionAPI / ActionElementData. The ActionAPI class is used by generic action element

classes and is extended by ActionElementData which is used by configurable action

element classes.

o DecisionElementData. This class is used by configurable decision element classes.

o VoiceElementData. This class is used by configurable voice element classes.

LoggerAPI. This class is sent as an argument to a logger’s execution method for handling a

logging event.

XML API

When a component uses the Java API, the Session API is accessed via an object passed to the

execution method. A similar setup exists when a component uses the XML API. The entire

contents of the Session API are made available via a set of XML documents passed to the

component in the HTTP request. Each component will receive this information whether they

need it or not, since the VoiceXML Server does not know in advance what information the

component could require. The component can choose to ignore these documents if the

information contained within is not required, or use a fast, event-based parser to extract only the

desired information from the documents.

Each component receives two POST arguments containing complete XML documents

representing the Session API. The first, named “inputs”, lists the session information

representing the state of the application up to the point when the component was reached. The

second argument, named “settings”, lists the current value for the application settings.