Getting 2 Words From a Double Word

[PulsarSL]

Hi guys,

I'm using the Windows MCI to write a program. One of the status commands I'm using returns information I need about the levels in both the left and right audio channels in a dword. The low-order word is the right channel and the high-order word is the left channel.

How do I seperate the dword into the component words?

Thanks,

Pulsar

 

[xwb]

You could use the macros LOWORD and HIWORD. You could use masks and shifts but that depends on whether you are working in a 16, 32 or 64 bit environment and whether the data that is coming back comes from a 16, 32 or 64 bit system.

On 16 bits, everything was sane, words were 16 bits, DWORDS were 32 bits. On 32 bits, both WORDs and DWORDs are both 32 bits. No idea what they are on 64 bits.

If we assume wsize is the word size and wmask is the word mask, then

WORD wsize = 16; // assuming 16 bit env
WORD wmask = 0xFFFF;
WORD loword = (WORD)(dw & wmask);
WORD hiword = (WORD)((dw >> wsize) & wmask);

 

[PulsarSL]

Thanks for the prompt response... I'll try the macros.

Pulsar

 

[dEVooXiAm]

On 32 bits, both WORDs and DWORDs are both 32 bits.

WORD itself is a 16bit "structure". Like BYTE will never become 16bit or 32bit. I really doubt Microsoft would start making such drastic macro redefinitions. Can you please paste in some proof of where WORD is considered a 32bit value?

------------------
When you do it, do it right.

 

[B00gyeMan]

Another approach would be to use a union defined like this:

typedef struct
{
WORD w1;
WORD w2;
}MyDWORD;

typedef union {
MyDWORD aDWORD;
DWORD dwValue;
}NameMe;

The trick is that the union occupies as much as its largest member (which in this case is the aDWORD variable) and that all the data occupies the same memory location.
So, you can put a DWORD value in the dwValue member and then read the two WORDs from aDWORD.w1 and aDWORD.w2
Should work both ways.