TODO

Feature ideas:
- hover over character class should show an ascii map
- figure out a way to tell users that /foo$/ matches "foo\n"
- bookmark / share url


Things I wished I had time to implement, but will probably never have time for:
- Handle differences between various regexp implementations (e.g. php vs javascript)
  - python auto anchors
  - other languages can use % % as the regexp seperator
- More lint rules:
  - x*x => x+
  - x*x* => x*
  - xx* => x+
  - [0-9] => \d
    etc.
  - is it possible to detect rules which can cause a DOS?
    .*.*foo or (.*)*
  - detect www., .com or .net and
    suggest replacing "." with "\.".

A few things I learned while writing this code:

1.

  /\b/ matches word boundary
  /[\b]/ matches back space (\x08)
  so the identity escapes are useful in the following cases:
  - outside ranges: ^$\.*+?()[]{}|
  - inside ranges: \]-

2.
  the spec doesn't really talk about octal escapes. We'll ignore them for now.
    /[\92]/ matches '9' or '2'
    /[\12]/ doesn't match '2'. It matches \x0a.

  don't know how to handle \01, \010, etc. there is a potential
  confusion between octal & group ref?

  I decided to punt on this issue and just fail to parse such expressions.

3.
  Internet Explorer is weird when it comes to this:
  /[]x]/.test("x") is true in IE, but false in most other browsers.