unabsorbedinsulin: variable size record

[bewest]

UnabsorbedInsulin

The head is two bytes, there is no date/time, making it a unique record.
Given a record header of 0x5c 0x08, the total record length is 8, and there are 6 bytes in the body:

# two byte header
#8 byte record size
bytearray([ 0x5c, 0x08,
                  # means body of 6 bytes
                  0x32, 0x4e, 0x04,
                  0x26, 0x58, 0x04, ])

Each unabsorbed record is a 3 byte record encoding the amount, age, and curve.
The amount needs to be divided by the pump model's strokes per unit in order to determine the number of units of insulin.

In the above example, since there are 6 bytes in the body, we know that there are 2 records:

  >>> print pformat(rec.parse( UnabsorbedInsulinBolus._test_1 ))
  [{'age': 78, 'amount': 1.25, 'curve': 4},
   {'age': 88, 'amount': 0.95, 'curve': 4}]

The body should always be a multiple of three bytes, thus the record size in general will always be the 2 byte header + some multiple of 3.

[bewest]

Closing this for now.