OK I see what your getting at now. Your right, the second example is really no different than the first. The data was simply declared differently in the data buffer. The data that is in the buffer is what viWrite() will send. The vi functions do not perform any conversion on the data. While the data buffer in the viWrite() and viRead() functions is prototyped to something that originates with unsigned char, it could just as well be a void type. The data that they send and receive over the serial port is ultimately binary bytes, there is no data type at the hardware level.
You can pass any type of buffer to the vi functions by using a cast. Conversions between different types of binary formats are not needed as long as you are not transfering data between systems with different "endian" formats for data representation.
int myData[5];
myData={0x0001,0x0002,0x0003}
viWrite (handle, (ViBuf*)Data, sizeof(myData), &count);
viRead(handle, (ViBuf*)Data, sizeof(myData), &count);
Any conversion of data between string/char and binary formats must be handled before the buffer is sent to the vi functions. This is where Scan and Fmt can be useful.
As far as declaring a bit field as a string type, that does not exist in C outside of bit fields used in structures. It would seem that most C programmers are comfortable enough with hex representations that there has never been a need to add binary formats to standard c. I have seen representations like 0b0101 used in embedded compilers, since bit level manipulation is more common in microcontrollers, but the formats vary from vendor to vendor.
It probably would not be too difficult to modify the bin2dec or dec2bin routines commonly found in basic code to c. You can find examples of them through Google. I also came across this very simple example in c
and this which is in c++ but the nuts and bolts part is in c
Message Edited by mvr on 09-06-2006 08:48 AM
