Raw binary values

VILLASnode supports sending samples formatted as raw binary data.

Signal values are encoded according to their configured data-type in either integer or IEEE-754 floating point values.

Implementation

The source code of the format-type is available here: https://github.com/VILLASframework/node/blob/master/lib/formats/raw.cpp

Configuration

bits	integer Default: 32 Number of bits per signal. Must be one of 8, 16, 32, 64 or 128.
endianess	string Default: "little" Enum: "big" "little" The endianess of the data.
fake	boolean Default: false Send and interpret the first three signals of each sample as the following header fields: sequence number timestamp seconds timestamp nano-seconds
real_precision	integer Default: 17 Output all real numbers with at most n digits of precision. The valid range for this setting is between 0 and 31 (inclusive), and other values result in an undefined behavior. By default, the precision is 17, to correctly and losslessly encode all IEEE 754 double precision floating point numbers.
ts_origin	boolean Default: true If set, include the origin timestamp in the output.
ts_received	boolean Default: true If set, include the received timestamp in the output.
sequence	boolean Default: true If set, include the sequence number in the output.
data	boolean Default: true If set, include the data in the output.
offset	boolean Default: true If set, include the offset between origin and received timestamp in the output.

{

• "bits": 32,
• "endianess": "big",
• "fake": false,
• "real_precision": 17,
• "ts_origin": true,
• "ts_received": true,
• "sequence": true,
• "data": true,
• "offset": true

}

Example Configuration

node/etc/examples/formats/raw.conf

nodes = {
    node = {
        type = "file"
        uri = "/dev/null"

        format = {
            type = "raw"

            bits = 32
            endianess = "little"
            fake = false
        }
    }
}

Notes

• Floating-point encoding is only supported for width larger or equal to 32 bit.
• The raw format-type does not support vectors. Only a single sample can be encoded per payload as there is no way to identify sample boundaries in the raw format.
• Complex sample values are supported and are encoded by real and imaginary parts with half of the precision as noted in the "Floating-point Precision" column above. E.g. in 64 bit mode real and imaginary components are encoded as two single-precision values encoded right after each other.
• Support for 128-bit wide values depends on you compiler support.