cv — convert binary data files between representation formats
Convert a binary data file from one representation (format) to another. The particular representation desired is expressed as a set of choices from the pattern: [h|n][s|u] c|s|i|l|d|8|16|32|64. For example, "hui" is host unsigned int, "nl" is network (signed) long. The default is NETWORK format.
Host specific format is the native internal number representation used inside the computer that cv is running on. There is great variation between vendors in the host-specific formats that they use. Especially for double-precision floating point, DEC, Cray, and IBM have peculiar formats and byte orderings. The network-independent byte orderings used within BRL-CAD is big-endian (most significant bit first, on the left side of the word), twos complement. The network-independent double-precision floating point format used within BRL-CAD is 64-bit IEEE 754, transmitted in big-endian order.
Signed or unsigned integer formats can be selected. This has no effect on the double-precision format.
"char", "short", "int", "long", and "double" provide their C language equivalent on this processor when used in "h" host-specific mode. "8", "16", "32", and "64" allow more precise specification of datum sizes for "n" network-independent mode.
For example, to change a file of host-specific (machine-specific) signed shorts to a file of network-independent 32-bit integers, the input format would be "hss", and the output format would be "n32".
pix-bw(1), dpix-pix(1), bw-pix(1); M. J. Muuss, ``Workstations, Networking, Distributed Graphics, and Parallel Processing`` in Computer Graphics Techniques: Theory and Practice, edited by D. F. Rogers, R. A. Earnshaw, Springer-Verlag, 1990