Re: [PATCH] Fix compilation erorr when --enable-gcc-warnings passed

[Paul Eggert]

On 01/22/2016 02:54 AM, Alexander Kuleshov wrote:
> @@ -4615,16 +4622,15 @@ colors_in_color_table (int *n)
>   static unsigned long
>   lookup_rgb_color (struct frame *f, int r, int g, int b)
>   {
> -  unsigned long pixel;
> -
>   #ifdef HAVE_NTGUI
> -  pixel = PALETTERGB (r >> 8, g >> 8, b >> 8);
> +  return PALETTERGB (r >> 8, g >> 8, b >> 8);
>   #endif /* HAVE_NTGUI */
>   
>   #ifdef HAVE_NS
> -  pixel = RGB_TO_ULONG (r >> 8, g >> 8, b >> 8);
> +  return RGB_TO_ULONG (r >> 8, g >> 8, b >> 8);
>   #endif /* HAVE_NS */
> -  return pixel;
> +
> +  return 0;
>   }

Does this change cause lookup_rgb_color to always return 0 on platforms 
other than MS-Windows and NextStep? Is this the right thing to do? This 
part of the change isn't mentioned in the draft ChangeLog entry, so I'm 
a bit lost as to the motivation.