Re: A problem of a name parser

[Aaron Jackson]

See below.

On Feb 17, 2008, at 8:22 AM, Tong King wrote:

> Dear All,
>   I write a .l file and a .y file to parse the struct:  
> somebody=age, and the
> result is to print somebody's age is .... But the parser can work  
> at the
> first time, at the second time it reports wrong although input is  
> right. I
> don't know why, and here is the code:
> 3.l
>
>
> %{
> #include "3.tab.h"
> #include<string.h>
> extern void yyerror( char*);
> %}
> %%
> [a-zA-Z]+       {yylval.str=strdup(yytext); return NAME;}
> [0-9]+                {yylval.num = atoi(yytext); return AGE;}
> [ \t\n]               {;}
> [=;]                  {return yytext[0];}
> .                     {ECHO; yyerror ("unexpected character");}
>
> %%
>
> int yywrap (void) {return 1;}
>
> ###############################
>
> 3.y
>
> %{
> #include<stdio.h>
> extern int yylex();
> void yyerror(const char*);
> %}
> %union{
> char* str;
> int   num;
> }
> %token <str> NAME
> //%token '='
> %token <num> AGE
>
> %%
exps: exp /* You need to include something to parse multiple  
statements */
    |    exps exp
    ;
> exp:
>  | NAME '=' AGE  {printf("the man %s's age is%d\n",$1,$3 );}
> ;
> %%
> void yyerror(const char*s)
> {
> fprintf(stderr, "%s\n", s);
> }
> int main()
> {
> return yyparse();
> }

Your parser works as written, but you wrote it to parse only *one*  
statement.  You need to include an additional parser rule to parse  
multiple statements see above.