This example's XHR DataSource instance points to a custom PHP script that
returns flat-file data. The DataSource schema is used to parse the tab- and
linebreak-delimited data. For maximum cache performance, the
queryMatchSubset property has been enabled, and the cache
itself has been turned up to 60 entries for fewer round trips to the server.
On this page there are three separate AutoComplete instances that all point to a single DataSource, which further maximizes cache efficiency. Queries can be delimited using the semi-colon (;) character. For demonstration purposes, each instance decreases the query delay slightly, and each instance presents the return data in the container using slightly different markup.