Following my idea to lazy copy "tokens" into a buffer (instead of char  
by char) didn't improve the performance a lot (%1 improvement).
I'm trying to rewrite Ragel's code for parsing of "integer", "unsigned  
integer" and "float" numbers.

Your advices will be very welcome !

Le 6 sept. 09 à 22:38, Adrian Thurston a écrit :

> In the break actions I'm assuming that you're returning a token to  
> the caller?  I'm using email on my phone right now so it's hard for  
> me to read.
>
> If you want to squeeze more time out of it I would suggest calling  
> the parsing functions right from the ragel actions. I'm not sure how  
> much faster that will be though.
>
> Adrian
> -----Original Message-----
> From: Younès HAFRI <[email protected]>
>
> Date: Sat, 5 Sep 2009 01:05:17
> To: <[email protected]>
> Subject: [ragel-users] Ragel for JSON
>
>
> _______________________________________________
> ragel-users mailing list
> [email protected]
> http://www.complang.org/mailman/listinfo/ragel-users
>
> _______________________________________________
> ragel-users mailing list
> [email protected]
> http://www.complang.org/mailman/listinfo/ragel-users


_______________________________________________
ragel-users mailing list
[email protected]
http://www.complang.org/mailman/listinfo/ragel-users

Reply via email to