Minor tidy up of String.prototype.htmlEncode and String.prototype.htmlDecode

[Martin Budden]

Both these functions would benefit in terms of readability, size and
performance by changing their RegExp strings into regular expression
literals. (Unlike in the matchers there is no strong case for retaining
the RegExp string form.)

(Actually htmlDecode is not used, but it probably should be retained on
symmetry grounds).

// Convert & to "&amp;", < to "&lt;", > to "&gt;" and " to "&quot;"
String.prototype.htmlEncode = function()
{
return(this.replace(/&/mg,"&amp;").replace(/</mg,"&lt;").replace(/>/mg,"&gt;").replace(/\"/mg,"&quot;"));
}

// Convert "&amp;" to &, "&lt;" to <, "&gt;" to > and "&quot;" to "
String.prototype.htmlDecode = function()
{
return(this.replace(/&amp;/mg,"&").replace(/&lt;/mg,"<").replace(/&gt;/mg,">").replace(/&quot;/mg,"\""));
}