Query on use of UBXMessage bitfield constructor

[semuadmin]

(originally raised by Bob in BuyMeACoffee)

The output from this doesn't obey the principle of least astonishment for my brain, so maybe I'm missing something about how mask and individual parameters interact?

resetMsg = UBXMessage(6, 4, SET, navBbrMask=b'\xffff', resetMode=1)
print(resetMsg)
resetMsg = UBXMessage(6, 4, SET, eph=1, alm=1, health=1, klob=1, pos=1, clkd=1, osc=1, utc=1, rtc=1, aop=1, resetMode=1)
print(resetMsg)

<UBX(CFG-RST, eph=0, alm=0, health=0, klob=0, pos=0, clkd=0, osc=0, utc=0, rtc=0, aop=0, resetMode=1, reserved0=0)>
<UBX(CFG-RST, eph=1, alm=1, health=1, klob=1, pos=1, clkd=1, osc=1, utc=1, rtc=1, aop=1, resetMode=1, reserved0=0)>

[semuadmin]

Hi Bob,

pyubx2 parses bitfields ('X' type attributes) according to the boolean parsebitfield argument. By default, this is set to True, meaning that individual bit fields are set or parsed explicitly (any you don't set explicitly will be set to their type defaults, which is what has happened in the first of your two commands above. Reserved bits default to 0). In the case of the CFG-RST message, the navBbrMask bitfield is defined thus:

    "CFG-RST": {
        "navBbrMask": (
            X2,
            {
                "eph": U1,
                "alm": U1,
                "health": U1,
                "klob": U1,
                "pos": U1,
                "clkd": U1,
                "osc": U1,
                "utc": U1,
                "rtc": U1,
                "reserved2": U6,
                "aop": U1,
            },
        ),
        "resetMode": U1,
        "reserved0": U1,
    },

If you want to address the bitfield group as a single attribute, you can set parsebitfield=0, e.g.

resetMsg = UBXMessage(6, 4, SET, navBbrMask=b'\xff\xff', resetMode=1, parsebitfield=0)
print(resetMsg)
print(resetMsg.serialize())

<UBX(CFG-RST, navBbrMask=b'\xff\xff', resetMode=1, reserved0=0)>
b'\xb5\x62\x06\x04\x04\x00\xff\xff\x01\x00\x0d\x5f'

But I would generally recommend setting the individual bit fields, as this is less ambiguous. Bear in mind that b'\xffff' (len 3) is not the same as b'\xff\xff' (len 2).

[bobvan]

Thanks! Much less astonished now :-)