On Sun, 2011-01-30 at 13:13 -0200, Felipe Almeida Lessa wrote: > On Sun, Jan 30, 2011 at 1:03 PM, Maciej Piechotka <[email protected]> > wrote: > > It allows to achive in parsing a near-attoparsec levels of speed > > (benchmarks from attoparsec library shown a 0.450 ± 0.028 for > > attoparsec, 0.479 ± 0.043 for nanoparsec and 1.532 ± 0.084 for parsec 3) > > combining the flexibility of stream of parsec 3 with the iterative > > approach and speed of attoparsec. > > Nice! How does it compare to attoparsec-text performance-wise?
I haven't checked. So far there are specializations for ByteString but not Text. I would expect similar penalty for it anyway (I'll try to port tests and report values). > Although I'm its maintainer, duplicating libraries isn't funny =). > "Deduplication" was a goal of the nanoparsec. Most of the code and documentation is from attoparsec anyway. > Cheers, > Regards
signature.asc
Description: This is a digitally signed message part
_______________________________________________ Haskell-Cafe mailing list [email protected] http://www.haskell.org/mailman/listinfo/haskell-cafe
