mode setter appears wrong

[peterhinch]

I'm writing a MicroPython driver for your board. In my testing it is only possible to change the device mode if you first change the mode to _CONFIG_MODE. Unfortunately I don't have any hardware capable of running CircuitPython but I think the following test will fail:

import adafruit_bno055
from busio import I2C
from board import SDA, SCL

i2c = I2C(SCL, SDA)
imu = adafruit_bno055.BNO055(i2c)
print(imu.euler)  # Check it's working
imu.mode = 5
print(imu.mode)  # I think this will be unchanged

The need to set _CONFIG_MODE prior to setting any other mode is contradicted in the chip datasheet (section 3.3.1). Consequently before I submit a PR I would be grateful if someone could perform the above test to confirm my findings.

For reference this is my mode setting code. This includes delays mandated by the device datasheet.

    def mode(self, new_mode=None):
        old_mode = self._read(_MODE_REGISTER)
        if new_mode is not None:
            self._write(_MODE_REGISTER, _CONFIG_MODE)  # This is empirically necessary if the mode is to be changed
            time.sleep_ms(20)  # Datasheet table 3.6
            if new_mode != _CONFIG_MODE:
                self._write(_MODE_REGISTER, new_mode)
                time.sleep_ms(10)  # Table 3.6
        return old_mode

[caternuson]

Hmmm...I think you are correct.

Adafruit CircuitPython 3.1.2 on 2019-01-07; Adafruit ItsyBitsy M4 Express with samd51g19
>>> import adafruit_bno055
>>> from busio import I2C
>>> from board import SDA, SCL
>>> i2c = I2C(SCL, SDA)
>>> imu = adafruit_bno055.BNO055(i2c)
>>> imu.mode
12
>>> imu.mode = 5
>>> imu.mode
12
>>> imu.mode = 0
>>> imu.mode
0
>>> imu.mode = 5
>>> imu.mode
5
>>>

[peterhinch]

Thank you for confirming. I'll submit a PR in a day or two.

[caternuson]

Fixed by #27

Thanks @peterhinch!

[tannewt]

@peterhinch email me (scott@adafruit.com) if you'd like some hardware to test CircuitPython with. We really appreciate your fixes to the drivers.