Initialize Processing

Once the user has submitted a search request, Watson™ Explorer Engine extracts the CGI parameters from the URL that specifies the user's search and uses them to initialize the rest of its processing. For instance if the user has filled out and submitted an HTML search form with the fields q (for query) and title with the string values, big OR large and travel, respectively, the resulting URL might look something like:

  http://example.com/search?v:project=sample&input-form=sample-form&q=big&title=travel

Watson Explorer Engine parses the CGI parameters in this URL (the part after the question mark) and represents them as follows:

  <param name="v:project" value="sample"/>
  <param name="input-form" value="sample-form"/>
  <param name="q" value="big"/>
  <param name="title" value="travel"/>

In this example, v:project is a key parameter for Watson Explorer Engine. It specifies where to get all the configuration information that will be used to drive the rest of the processing. Watson Explorer Engine creates a new data explorer object and loads into it (from the repository) all the configuration information that is associated with the project called sample. It adds the other CGI parameters to this state and proceeds to process it. This section describes the processing in Watson Explorer Engine as a series of consecutive steps. In reality, it is a single process of interpretation of this initialized state of the Watson Explorer object, which in practice results in these steps. The state of the Watson Explorer object can be represented at any time by a large XML structure. This overview description of Watson Explorer Engine processing refers liberally to fragments of this XML structure that are relevant to particular aspects of the processing.

The next step in processing is normalizing the query input. See Normalize Query Input.