lex-无法识别的规则-使用正则表达式

Question

I am new to working with lex and just wanted to try and see whether I could try making a file that would locate comments in a C file using regular expressions.

%%

(/\*([^*]|[^]|(\*+([^*/]|[^])))*\*+/)|(//.*){   
    return 5;
}
.   ;

%%

int yywrap(void)
{
    return1;
}

So the idea here would be that everytime a comment is located, it will "return 5", otherwise, nothing happens.

The problem is, the regular expression is not being recognized and i'm met with the error:

line 3: unrecognized rule line 3: unrecognized rule line 3: unrecognized rule line 3: unrecognized rule ...etc

Any help would be appreciated, thanks in advance.

Answer 1

In (f)lex regular expressions, as in standard Posix regexes, you can include ] in a character class as the first character in the class. (That's true whether or not you use a positive ( [...] ) or negative ( [^...] ) character class.)

So the first [^] starts a character class, which is then terminated by the next ] , and the next one is unterminated. It's not clear to me what you expected [^] to mean, so I can't offer an alternative suggestion.

Also, you need a space before the action, in order to define the end of the regular expression.