But here are  all the precedence rules I have:

/***** Precedence Rules  *****/
%left GT LT
%left PLUS MINUS
%left STAR DIVIDE
%left ATOB
%nonassoc prec_unary_minus


-Angela


On Oct 31, 2007, at 9:14 AM, Thomas Gazagnaire wrote:

> The following code works perfectly for me. Maybe you introduce some  
> undesired precedence rules in your grammar rules.
>
>
> ----
>
> %token <int> INT
> %token PLUS MINUS DIVIDE STAR ATOB END
>
> %start main
> %type <int> main
>
> %left PLUS MINUS
> %left STAR DIVIDE
> %left ATOB
>
> %%
>
>
> main:
> | expr END { $1 }
>
> expr:
> | INT { $1 }
> | expr PLUS expr { $1 + $3 }
> | expr MINUS expr { $1 - $3 }
> | expr DIVIDE expr { $1 / $3 }
> | expr STAR expr { $1 * $3 }
> | expr ATOB expr { int_of_float ( (float_of_int $1) **  
> (float_of_int $3) ) }
> ;
>
> ---
>
> And then "2^2+7\n" gives me "11"
>
> Cheers,
> Thomas
>
> Angela Zhu a écrit :
>> On Oct 31, 2007, at 6:52 AM, Peter Ilberg wrote:
>>>
>>> I have no experience with ocamlyacc, but looking at your grammar  
>>> below, it seems that you don't need the 'value PLUS exp' etc  
>>> rules. All these cases should be covered already by the 'exp PLUS  
>>> exp' rules at the beginning and the 'value' rule at the end.
>>>
>>> Try removing the 'value PLUS exp' rules. Maybe ocamlyacc gets  
>>> confused if it has two sets of productions that it has to  
>>> disambiguate with precedence rules.
>> I removed 'value PLUS exp' rules.
>> The precedence is still not correct, what is more, 1+ t (with t  
>> declared) gives a syntax error.
>> Thanks,
>> Angela
>>>
>>> --- Peter
>> _______________________________________________
>> Caml-list mailing list. Subscription management:
>> http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list
>> Archives: http://caml.inria.fr
>> Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
>> Bug reports: http://caml.inria.fr/bin/caml-bugs
>
> _______________________________________________
> Caml-list mailing list. Subscription management:
> http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list
> Archives: http://caml.inria.fr
> Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
> Bug reports: http://caml.inria.fr/bin/caml-bugs
>