Regular expressions on the web

On the web, static pages fade into the past; web sites use server and client-side scripting techniques to improve the user experience. For client-side scripting, JavaScript is by far the most commonly used dynamic language. In these scripts, regular expressions are widely used for several purposes, e.g., for text filtering and form validation. In this paper, we take a closer look at regular expressions on the web. We investigate historical data and determine the trends of the use of regular expressions over time on various web sites. Then, we also take a snapshot of the current status. We find that although regular expressions are being used more and more often, their use is highly repetitive: on the most popular web sites only 4% of the regular expressions are unique. Based on this result, we show that web browsers can cut down regular expression compilation time to the third by applying caching techniques.