Why does c allow main() even when it is not int ma

Question

While reading the K&R 2nd edition I noticed that the programs always began with "main(){". I had always thought that main() had to have int or void before it. So that it would look like "int main()" or "void main()". What is just "main()" and what is the difference?

Answer 1

Because this is supported in by an old version of c.

main()

is equivalent to

int main()

Answer 2

main() is the old K&R style where the int was omitted as the return type defaults to int if not specified (you should specify it). Additionally, empty parentheses is in K&R style to show it takes no arguments.. in C99 this should now be void to indicate such. Empty parentheses means that the function will accept any number of arguments of any type, which is clearly not what you want. So the final result is:

int main(void) { ... }

main() should return int.. convention says a return 0; statement at the end will help indicate to the caller that the program executed successfully - non-0 return values indicate abnormal termination.

A more direct answer to your question would be that main() { ... } works because it's not wrong. The compiler sees that no return type was declared for the main function so it defaults to int. The empty parentheses indicates to it that main takes any number of arguments of any type, which is not wrong either. However, to conform to C99 style/standard, use

int main(void) { ... }

Answer 3

Syntax most of times depends on the compiler. For example, when you use visual c++ you write "void main" but when you use GCC, you should write "int main()" and then return 0 or 1 if the program finished good or bad.