Business::Associates::Stylesheet 1.00

Business::Associates::Stylesheet is a Perl information on the Associates XSLT stylesheets and how to modify the data processing.

THE PROCESSING MODEL

Before we get going on the details, let us first remind ourselves on what we might call the processing model of the Associates package.

The first step is to retrieve the raw XML data from Amazon.com (or, technically, from the local cache if we already have it). This is structured data that describes the (currently) fifteen best-selling titles at Amazon.com for the selection criteria we have chosen. Those criteria can be keyword searches or a look-up of a specific category of product, and are typically chosen by the HTML editor or by the code that generates the HTML. The perl code provided by this package retrieves the XML.

The second step is to transform this raw XML into a format that the device responsible for the visual display can handle. In our case, that device is ultimately the web browser, so we need to transform the XML data into HTML or xhtml data. However, the same process model could be used to render the data in another format, say for inclusion to a postscript document. This data transformation is done used XSLT, a standard and a very powerful tool for transforming XML documents. This is the subject of this document.

The third step is the visual formatting of the data. At this stage we have the data in a format that the visual display device can understand, and we need to tell it of the details of the formatting. In out model and for HTML (or xhtml) output, that is done using traditional Cascading Style Sheets (CSS). That is the topic of a separate document and concerns us only so far as it reminds us that we need to be able to identify each element type in the output such that the CSS designer can say, for example, "lets have all book titles in red text". In HTML, this identification is done by defining class attributes.