Documentation

testString1 and testString2 are convenient to use in GHCi when playing around with how parsing works.

See testString1.

Handle escaped characters correctly

>>> parseExprs $ "Foo \"hello \\\"world!\""
([Other "Foo ",StringLit "hello \\\"world!"],"")

Parses integers and reals, like 123 and 45.67.

To be more precise, any numbers matching the regex \d+(\.\d+)? should get parsed by this function.

>>> parseNumberLit '3' "456hello world []"
("3456","hello world []")
>>> parseNumberLit '0' ".12399880 foobar"
("0.12399880"," foobar")

This function consumes input, stopping only when it hits a special character or a digit. However, if the digit is in the middle of a Haskell-style identifier (e.g. foo123), then keep going anyway.

This is almost the same as the function

parseOtherSimple = span $ \c ->
  notElem c ("{[()]}\"," :: String) && not (isDigit c)

except parseOther ignores digits that appear in Haskell-like identifiers.

>>> parseOther "hello world []"
("hello world ","[]")
>>> parseOther "hello234 world"
("hello234 world","")
>>> parseOther "hello 234 world"
("hello ","234 world")
>>> parseOther "hello{[ 234 world"
("hello","{[ 234 world")
>>> parseOther "H3110 World"
("H3110 World","")