karl@haddock (08/18/86)
zircon!c8-rah (Max Schireson) writes: >Since one could declare an arbitrarily long array of void, sizeof(void *) >would have to be infinite, which does not seem possible. Since the "arbitarily long array" has total size zero, there's no problem. In fact, the compiler could store all void "objects" at the same address, and "void *" itself -- having only one value -- could have size zero. Part of the problem here, I think, is that (in X3J11) the keyword "void" now has three meanings: "void f()" or "(void)e", "void *", and "f(void)". The second meaning is based on the assumption that you can't have a pointer to a real void, and the third assumes that void expressions cannot be function arguments. If you want to make void a full-fledged datatype, you'd better use a different syntax for the other two meanings. Karl W. Z. Heuer (ihnp4!ima!haddock!karl), The Walking Lint