--
-- hzsff, a program to do things to ZSFF files.
-- See http://www.zedshaw.com/blog/2008-01-13.html.
--
-- Copyright Michael Orlitzky
--
-- http://michael.orlitzky.com/
--
-- This program is free software: you can redistribute it and/or modify
-- it under the terms of the GNU General Public License as published by
-- the Free Software Foundation, either version 3 of the License, or
-- (at your option) any later version.
--
-- This program is distributed in the hope that it will be useful,
-- but WITHOUT ANY WARRANTY; without even the implied warranty of
-- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-- GNU General Public License for more details.
--
-- http://www.fsf.org/licensing/licenses/gpl.html

import System.Environment (getArgs)
import System.Exit
import Text.ParserCombinators.Parsec


data Version = Version (String)
             deriving (Show)
                      
data Header = Header (String, String)
            deriving (Show)

data Url = Url (String)
         deriving (Show)

data ZsffFeed = ZsffFeed (Version, [Header], [Url])
                deriving (Show)


version_characters :: GenParser Char st Char
version_characters = digit <|>
                     char '.'
                      
version_parser :: Parser Version
version_parser = do
  string "ZSFF "
  val <- many1 version_characters
  char '\n'
  return $ Version (val)


header_characters :: GenParser Char st Char
header_characters = alphaNum <|>
                    char '-'
                         

header_parser :: Parser Header
header_parser = do
  val1 <- many1 header_characters
  char ':'
  skipMany space
  val2 <- manyTill anyChar (char '\n')
  return $ Header (val1, val2)


headers_parser :: Parser [Header]
headers_parser = do
  headers <- many1 header_parser
  return headers


-- Ad-hoc compilation of characters that I think
-- are valid in URLs
url_characters :: GenParser Char st Char
url_characters = alphaNum <|>
                 char '.' <|>
                 char ':' <|>
                 char '_' <|>
                 char '-' <|>
                 char '%' <|>
                 char '~' <|>
                 char '?' <|>
                 char '/' <|>
                 char '#' <|>
                 char '&' <|>
                 char '+'


-- Will parse one or more url_characters
url_parser :: Parser Url
url_parser = do
  val <- manyTill url_characters (char '\n')
  return $ Url (val)


urls_parser :: Parser [Url]
urls_parser = do
  urls <- many1 url_parser
  return urls


main_parser :: Parser ZsffFeed
main_parser = do
  version <- version_parser
  headers <- headers_parser
  char '\n'
  urls <- urls_parser
  return $ ZsffFeed (version, headers, urls)


main = do
  args <- getArgs
  feed <- readFile (args !! 0)
  case (parse main_parser "" feed) of
    Left err -> do
         putStrLn ("Error " ++ (show err))
         exitFailure
    Right ast -> do
         putStrLn (show ast)
         exitWith ExitSuccess