WebFund 2014W Lecture 20

Basics of Regular Expressions

• start and end: /
• . represents any single character
• * is 0 or more repeats, + is one or more repeats
• Thus .* matches any number of characters (including none)
• () denote groups, normally for extraction or later substitution
• Each group is numbered, so first () is $1 (or something like that)
• can include letter ranges in [], e.g. [a-z]
• An all lowercase word with at least one character is: /[a-z]+/
• | means or (as usual), and is implicit

Apparently there are regular expression decoders online somewhere

Escaped characters

• \ is used to treat special characters as literals
• % followed by hex numbers denotes character codes

Chomsky hierarchy

Regular expressions are used for input sanitization

• make sure externally provided input is "safe"
• two strategies: whitelists and blacklists
  • whilelist: explicit list of "good" things
  • blacklist: explicit list of "bad" things
• use whitelists when possible
• whitelists are more work because if you leave anything out things break