This example demonstrates AutoComplete's queryMatchSubset property. The first instance of AutoComplete has queryMatchSubset enabled for maximum cache performance such that as you type, the query is searched within previously cached results. For best results, the DataSource should return a complete set of results when a single letter is queried such that subset matching will also return a complete set of results.
The second AutoComplete instance does not enable queryMatchSubset
so each typed letter results in a new request to the server.
Note the custom CSS that is needed for stacking AutoComplete instances.
Implementers who are working with data from third-party sources, user input data, or otherwise untrustworthy sources should be sure to read the Security Considerations section of the AutoComplete user guide.
Note: The flat-file database accessed here has a limited number of terms; for best results, type one-letter at at time and let the AutoComplete instance return — if you type a full, highly-specifc phrase (such as your name) you'll probably get no results from the small dataset.