Custom log listeners
When the log method of a logging bean is called, the feedback facility
generates a log event. A log event is also generated when a rule is executed.
Log listeners process these log events and either store the event data or
perform custom processing with these events. The Feedback component provides
two default log listeners, the LMListener that collects data for use by the
LikeMinds recommendation engine and the FeedbackListener that collects data
for use in the feedback reports. Custom log listeners can be used to modify
the default behavior of the FeedbackListener or to implement a listener that
provides user-specific behavior.
There are a number of reasons that you might want to provide a customized
feedback listener. Some of these are:
- To collect request parameters, referral parameters, or cookies. By default,
the feedback listener does not collect this data. You must implement a customized
feedback listener that enables the collection of this data.
- To prevent private information from being stored. For instance, you could
mask the userid or IP address to ensure the privacy of your users.
- To augment the event data. Suppose your event data contains a product
id number and you would like to report on products by product name and id.
You could perform a custom lookup during event processing and add the product
name to the event.
There are a number of reasons that you might want to provide a new custom
listener class. Some of these are:
- To store event data in a separate database. Suppose that you want to capture
rating data in a preference database. You can do this through your own Web
application or you could do this as by adding a custom listener and capturing
rating events. The custom listener could facilitate real-time rating results.
- To generate e-mail when an event occurs. Perhaps you want to send e-mail
to customers after they purchase a large order. By processing an action event
with the purchase amount included as action data, you could do this with a
custom log listener.
- To generate notifications when an event occurs. You can detect the frequency
of shopping cart abandons and generate a notification to the site administrator
to check site availability and performance.
Parent topic: Listeners and persistence
|
|
|
Portal, Express Beta Version 6.1