-- Hoogle documentation, generated by Haddock
-- See Hoogle, http://www.haskell.org/hoogle/


-- | Simple tokenizer for English text.
--   
--   Simple tokenizer for English text.
@package tokenize
@version 0.1.0

module NLP.Tokenize

-- | The EitherList is a newtype-wrapped list of Eithers.
newtype EitherList a b
E :: [Either a b] -> EitherList a b
unE :: EitherList a b -> [Either a b]

-- | A Tokenizer is function which takes a list and returns a list of
--   Eithers (wrapped in a newtype). Right Strings will be passed on for
--   processing to tokenizers down the pipeline. Left Strings will be
--   passed through the pipeline unchanged. Use a Left String in a
--   tokenizer to protect certain tokens from further processing (e.g. see
--   the <a>uris</a> tokenizer).
type Tokenizer = String -> EitherList String String

-- | Split string into words using the default tokenizer pipeline
tokenize :: String -> [String]

-- | Run a tokenizer
run :: Tokenizer -> (String -> [String])

-- | Split string on whitespace. This is just a wrapper for <a>words</a>
whitespace :: Tokenizer

-- | Detect common uris and freeze them
uris :: Tokenizer

-- | Split off initial and final punctuation
punctuation :: Tokenizer

-- | Split off word-final punctuation
finalPunctuation :: Tokenizer

-- | Split off word-initial punctuation
initialPunctuation :: Tokenizer

-- | Split words ending in <a>n't</a>, and freeze <a>n't</a>
negatives :: Tokenizer
instance Monad (EitherList a)