Re: GHashTable and const

[Mark Mielke <mark mark mielke cc>]

Morten Welinder wrote:
> "const" in C does not propagate as usefully as you would like.  Therefore,
> the following sniplet is not violating C rules:
>
> struct Foo { int *x; };
> int foo (const struct Foo *p) { *(p->x) = 1; }
>   
I don't think most languages propagate a "const"-like type. However, at 
least in theory, I think const would allow:

   struct Foo p;
   p->x = malloc(sizeof(int));
   foo(p);
   printf("%d\n", *(p->x));

To optimize the second p->x to be equal to a value from a register used 
during the initial assignment. If called in a tight inner loop with many 
iterations, the effect could be more significant.

> While I cannot change p->x, I can change what
> That said, it still wouldn't hurt to add const for cases like g_hash_table_size

I prefer prototypes to document intent as best as possible, even if the 
language does not allow for a direct match between intent and actual 
implementation. :-)

But, this is probably a waste of a thread as glib has done what glib has 
done.

Cheers,
mark

--
Mark Mielke <mark mielke cc>