Re: [Flex-help] Python Lexical Analysis

[Philip H. <red...@gc...>]

Hey guys

I've actually got a lot of stuff going at the moment now, but one
problem is, say i am using a rule like:

"\n"                    { return NEWLINE; }
"    "+                 {
                          int top_len = VEC_index( gpy_int, gpy_indent_stack,
						   VEC_length( gpy_int, gpy_indent_stack )-1 );
			  if( top_len <= yyleng )
			    {
			      if( top_len != yyleng )
				VEC_safe_push( gpy_int, gc, gpy_indent_stack, yyleng );

			      return INDENT;
			    }
			  else if( top_len > yyleng )
			    {
			      VEC_pop( gpy_int, gpy_indent_stack );
			      return DEDENT;
			    }
                         }

<<EOF>>                {
                         int top_len = VEC_index( gpy_int, gpy_indent_stack,
						  VEC_length( gpy_int, gpy_indent_stack )-1 );
			 if( top_len != 0 )
			   {
			     VEC_pop( gpy_int, gpy_indent_stack );
			     return DEDENT;
			   }
			 else
			   return 0;
                       }

So far thats what i do but its not right since, i need some mechanism
for counting the number of "    " rules are matched since yytext will
return the text of the single rule. There is probably some simple way
of doing that this works fine for simple things like:

x = 2
y = x

def foo( .. ):
  ...
  ...
  ..

But fails on more indentation, any suggestions will be great when the
parsing code is more concrete i will put up a separate a flex and
bison parser code to simply parse python code so you can do what ever
you want with it :)

--Phil