Hey Bruno, > Le 19 mai 2019 à 14:01, Bruno Haible <br...@clisp.org> a écrit : > >> Or are you suggesting that users should define their own type. Or do you >> mean to provide a macro to instantiate such a type given a data type? Or >> are you suggesting to restrict to a single size data type? > > Yes, these are the possible options. > > As far as I understand, the type would consist of > - an int, an enum value, or even a pointer to a statically allocated object > defined by the programmer, > - a constant string: the option, > - (optional) a constant string: the documentation.
Yup. > Restricting the type of the first field does not look like a good idea to me. > If the users define their own type, they have manual work when you add the > documentation string or other new features to argmatch. Agreed. That was the motivation for the API change you referred to. > So I would prefer a macro with one argument, the type of the first field. That sounds good. Thanks!