jabolol / pandoc

Document converter aiming to be a robust pandoc subset

mypandoc

Document converter implemented from scratch in Haskell. Aims to be a robust subset of pandoc.

usage

Warning

This tool only supports markdown, json and xml

./mypandoc -i ifile -f ofmt [-o ofile] [-e ifmt]

    -i ifile  : input file
    -f ofmt   : output format
    -o ofile  : output file (default: stdout)
    -e ifmt   : input format (default: auto)

setup

Clone the repository

git clone [email protected]:Jabolol/pandoc.git .

Build the project

gmake

Run the converter

./mypandoc -i examples/syntaxe.md -f json -o output/mine.json

overview

This project uses in-house parsers and printers to convert between formats. Full support between markdown, json and xml is provided. This means that you can convert to and from any of these formats.

parser

The main parser type is defined as follows:

newtype Parser i o = Parser {parse :: i -> Maybe (i, o)}

This type is a wrapper around a function that takes an input and returns a Maybe tuple of the remaining input and the output. This allows for the composition of parsers.

functor

The Functor typeclass is implemented to allow for the mapping of a function over the output of a parser. This is done by applying the function to the output of the parser.

instance Functor (Parser i) where
  fmap :: (a -> b) -> Parser i a -> Parser i b
  fmap f parser = Parser $ fmap (fmap f) . parse parser

applicative

The Applicative typeclass is implemented to allow for the chaining of parsers. This is done by applying the output of one parser to the input of another.

instance Applicative (Parser i) where
  pure :: a -> Parser i a
  pure x = Parser $ pure . (,x)

  (<*>) :: Parser i (a -> b) -> Parser i a -> Parser i b
  x <*> y = Parser $ \i -> case parse x i of
    Just (i', f) -> fmap f <$> parse y i'
    Nothing -> Nothing

alternative

The Alternative typeclass is implemented to allow for the chaining of parsers with a fallback. This is done by trying the first parser and if it fails, trying the second parser and so on.

instance Alternative (Parser i) where
  empty :: Parser i a
  empty = Parser $ const empty

  (<|>) :: Parser i a -> Parser i a -> Parser i a
  x <|> y = Parser $ \i -> parse x i <|> parse y i

monad

The Monad typeclass is implemented to allow for the chaining of parsers with the ability to change the parser based on the output of the previous parser. This is done by applying a function to the output of a parser and then parsing the result.

instance Monad (Parser i) where
  (>>=) :: Parser i a -> (a -> Parser i b) -> Parser i b
  x >>= f = Parser $ \i -> case parse x i of
    Just (i', a) -> parse (f a) i'
    Nothing -> Nothing

result

Take for example the xml parser:

xTag :: S.Parser String XValue
xTag = do
  _ <- S.char '<'
  n <- A.some (S.matches C.isAlphaNum)
  a <- A.many $ do
    k <- S.spaces *> A.some (S.matches C.isAlphaNum)
    _ <- S.spaces *> S.char '='
    _ <- S.spaces *> S.char '"'
    v <- A.many (S.matches (/= '"'))
    _ <- S.char '"' <* S.spaces
    pure (k, v)
  _ <- S.char '>'
  c <- S.tabs *> A.many (xTag A.<|> xText) <* S.tabs
  _ <- S.string "</" *> S.string n <* S.char '>' <* S.tabs
  pure $ XTag n a c

This parser is constructed from smaller parsers that parse individual parts of the xml format. It uses the Applicative and Monad typeclasses to chain parsers together to form a larger parser. It also uses the Alternative typeclass to provide fallbacks in case a parser fails, effectively providing error handling.