Link Details

This was a nice article, unfortunately it gave some bad information. For example, it suggested /^[a-zA-Z0-9_]{3,16}$/ to match a user name. That is way too complex when you can just use /^\w{3,16}$/ to match the exact same thing. Also the suggestion to use regular expressions for xml should be heavily qualified to point out there are a lot of times regular expressions just won't work for parsing xml. For example, he suggests {]*>(.*?)} to find the contents of a tag. This will work fine if you never have the same tag embedded within itself, but if you are doing webscraping like he suggests, you will often have things like " <div>This is a div <div>This is an embedded div</div><div>This is another embeded div</div></div>". In this case his regular expressions will match " This is a div This is an embedded div" and not what was intended.