Hi Roman You would need different behaviour for the /lexeme/ parser in Parsec.Token at least - this is the combinator that promotes a parser to also consume trailing whitespace.
I suspect you would have to recode most of Parsec.Token module - the TokenParser is a parameterized module (in the sense of Sheard and Pasalic [1]), unfortunately /lexeme/ is not one of the parameters and other combintators are defined using it within the makeTokenParser function which instantiates the parameterized module. Alternative you could remake a set of token parsers - this is covered in the Parsec manual [2] - section 2.11 "Advanced: Separate scanners". [1] http://web.cecs.pdx.edu/~sheard/papers/JfpPearl.ps [2] http://legacy.cs.uu.nl/daan/download/parsec/parsec.pdf Best wishes Stephen _______________________________________________ Haskell-Cafe mailing list [email protected] http://www.haskell.org/mailman/listinfo/haskell-cafe
