Re: Loops (was Re: do { quit; } else { })

On 22/04/2025 21:26, Kaz Kylheku wrote:

On 2025-04-22, bart <bc@freeuk.com> wrote:

BTW here is that macro definition from Lua:
>
   #define op_arith(L,iop,fop) {  \
     TValue *v1 = vRB(i);  \
     TValue *v2 = vRC(i);  \
     op_arith_aux(L, v1, v2, iop, fop); }

 I'm assuming that op_arith is called numerous times to define handling
for various VM opcodes.
 It *reliably* writes a bunch of code that would otherwise require a lot
of error-prone copy and paste.
 This is language extension; the LUA maintainers have written
themselves an op_arith language feature.
 

That's a good way to describe it.

The generated codee is large and complex, but at least you can see it.

Yes, but that is only needed by the people writing and debugging the macros involved.

 If that language feature were wired into the compiler, it would only
be harder to work with. All that morass of code would be hidden
inside, generated into some intermediate representation (that you
can perhaps intercept and get in some readable form, in a
language/notation you will have to learn).
 

It calls yet more macros, and those in turn call others; here are a few
for vRB:
>
    #define vRB(i) s2v(RB(i))
    #define s2v(o) (&(o)->val)
    #define RB(i) (base+GETARG_B(i))
    #define GETARG_A(i) getarg(i, POS_A, SIZE_A)
    #define getarg(i,pos,size) (cast_int(((i)>>(pos)) & MASK1(size,0)))
    #define cast_int(i) cast(int, (i))
    #define MASK1(n,p) ((~((~(Instruction)0)<<(n)))<<(p))
    #define cast(t, exp) ((t)(exp))

 Some of these look like low value.  For instance cast_int(x) saves no
characters over cast(int,x), and ony one character over cast(int, x),
and that is longer than just (int)(x), not to mention (int)x
situations where you can omit parentheses.

It is certainly split up more than I would have written, but there are perhaps good reasons for the style.  I haven't seen any more of the code than what Bart has quoted here - perhaps there are other situations in which "cast_int" is being passed as a parameter in another macro.  Or maybe there are historical reasons - "cast_int" came first, then there were other "cast_X" macros added later, then a generic "cast" macro and "cast_int" was redefined in terms of the generic one.  Maybe there are development or test versions of some of these macros where the plain casts are replaced by runtime checks before the conversion, and what we see here is just the release versions without the checks.
Or maybe they were just written by someone who likes to divide things up a bit more, or they were mixed from different sources or authors with minimal changes (that might explain the different capitalisation styles).