Re: [PATCH 07/10] d3dx9: Merge the d3dx_effect_GetInt() helper

[Matteo Bruni]

On Fri, Mar 8, 2019 at 12:15 AM Michael Stefaniuc  wrote:
> -
>        /* all components (3,4) are clamped (0,255) and put in the INT */
> -
>        *n = (INT)(min(max(0.0f, *((FLOAT *)param->data + 2)), 1.0f) * INT_FLOAT_MULTI);
> -
>        *n += ((INT)(min(max(0.0f, *((FLOAT *)param->data + 1)), 1.0f) * INT_FLOAT_MULTI)) << 8;
> -
>        *n += ((INT)(min(max(0.0f, *((FLOAT *)param->data + 0)), 1.0f) * INT_FLOAT_MULTI)) << 16;
> -
>        if (param->columns * param->rows > 3)
> -
>            *n += ((INT)(min(max(0.0f, *((FLOAT *)param->data + 3)), 1.0f) * INT_FLOAT_MULTI)) << 24;

One more of those "since you're touching this" cleanup: please use
plain (C standard, lowercase) int and float types in those casts. The
first (int) cast technically isn't required but it's fine to keep for
consistency.

Also, the comment above is a bit ugly. It could be reworded but I
don't see the point of it in the first place (the code is
self-explanatory), I'd just get rid of it.