Re: [Haskell-cafe] Impact of "try" on Parsec performance

On Fri, Mar 2, 2012 at 10:35 PM, Erik de Castro Lopo wrote:

Omari Norman wrote:

...

So my question is: what is the practical impact of using try?

My experience is based mostly on trying to improve the error location reporting in Ben Lippmeier's DDC compiler. What I found was that removing Parsec.try's almost always resulted in the parser generating better location information.

Roman Cheplyaka made some tweaks to 'try' error reporting in parsec 3.1.2, so this might be different now. Antoine