2010/1/12 Günther Schmidt <[email protected]>:
> Hi John,
>
> thanks for responding. As I said I've been using Parsec quite a lot, but
> wonder if there is a different approach possible/feasible to parsing. Parsec
> (2x) isn't an "online" parser, ie, it doesn't produce a result before the
> whole parse is completed.
>
This appears to be a common source of confusion.
You can produce a result by not trying to parse the entire input in
one go, instead parse as much as you want to consume, then return the
rest of the input with the tokens that you have collected. For
example, in Parsec 2 a regular approach I use is something like:
get1 = do token <- parse1token
remaining <- getInput
return (remaining,token)
Jeff
_______________________________________________
Haskell-Cafe mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/haskell-cafe
