regex-weburl.js

//
// Regular Expression for URL validation
//
// Author: Diego Perini
// Updated: 2010/12/05
//
// the regular expression composed & commented
// could be easily tweaked for RFC compliance,
// it was expressly modified to fit & satisfy
// these test for an URL shortener:
//
//   http://mathiasbynens.be/demo/url-regex
//
// Notes on possible differences from a standard/generic validation:
//
// - utf-8 char class take in consideration the full Unicode range
// - TLDs have been made mandatory so single names like "localhost" fails
// - protocols have been restricted to ftp, http and https only as requested
//
// Compressed one-line version (228 characters):
//
// var re_weburl = /^(?:(?:https?|ftp):\/\/)(?:\S+(?::\S*)?@|\d{1,3}(?:\.\d{1,3}){3}|(?:(?:[a-z\d\u00a1-\uffff]+-?)*[a-z\d\u00a1-\uffff]+)(?:\.(?:[a-z\d\u00a1-\uffff]+-?)*[a-z\d\u00a1-\uffff]+)*(?:\.[a-z\u00a1-\uffff]{2,6}))(?::\d+)?(?:[^\s]*)?$/i;
//
var re_weburl = new RegExp(
  "^" +
    // protocol identifier
    "(?:(?:https?|ftp)://)" +
    "(?:" +
      // user:pass@ authentication
      "\\S+(?::\\S*)?@|" +
      // IP dotted octets notation
      "\\d{1,3}(?:\\.\\d{1,3}){3}|" +
      // host name part
      "(?:(?:[a-z\\d\\u00a1-\\uffff]+-?)*[a-z\\d\\u00a1-\\uffff]+)" +
      // domain name part
      "(?:\\.(?:[a-z\\d\\u00a1-\\uffff]+-?)*[a-z\\d\\u00a1-\\uffff]+)*" +
      // TLD identifier part
      "(?:\\.[a-z\\u00a1-\\uffff]{2,6})" +
    ")" +
    // port number
    "(?::\\d+)?" +
    // resource path
    "(?:[^\\s]*)?" +
  "$", "i"
);