Re:Zig's dot star syntax (value.*)
by toasterlovin on Saturday March 08, 2025 @07:14PM
(#43304856)
> Yes, but unlike C's terribly confusing "declaration follows usage" style, which no tells you about btw
I happen to be reading the K&R C book right now and they do in fact tell you this.
Yeah, not sure where that came from. The ANSI C standard was extensively documented in books and articles and specs at all levels of rigor starting from the mid-80's. No one has ever lacked for a reference for how function pointer declaration looks.
That said, C's function pointer declaration syntax is indeed awful. But really that's because ANSI took a very hardline "no incompatible changes" tact when adding prototypes to the language, which limited the ways they could express them. That decision is one of the reasons we're still writing C today. Any yahoo can come up with a new language, kids do it all the time. ANSI's job was to add features to the language in which Unix was already written.
In Zig, you can pretty much read the types aloud and it makes, your brain does not need to peek for parsing. For me it's too late, I got used to the C way but I still want the better thing to get adopted. No more int ((foo)(int))[5]; nonsense—just T, [N]T, or *T, making intent crystal clear.