Parsing apple health data

health.py

from datetime import datetime
from collections import namedtuple
from lxml import etree

EXPORT_FILE = 'apple_health_export/export.xml'

BodyMass = namedtuple('BodyMass', ('date_time', 'value'))
StepCount = namedtuple('StepCount', ('date_time', 'value'))
Distance = namedtuple('Distance', ('date_time', 'value'))
Stairs = namedtuple('Stairs', ('date_time', 'value'))

TYPE_MAP = {
    'HKQuantityTypeIdentifierBodyMass': BodyMass,
    'HKQuantityTypeIdentifierStepCount': StepCount,
    'HKQuantityTypeIdentifierDistanceWalkingRunning': Distance,
    'HKQuantityTypeIdentifierFlightsClimbed': Stairs,
}

UNIT_MAP = {
    'km': 0.621371,  # kilometer to miles
    'kg': 2.20462,  # kilograms to pounds
}


def parse_record(record):
    item_class = TYPE_MAP.get(record.get('type'))

    if not item_class:
        return

    value = record.get('value')
    multiplier = UNIT_MAP.get(record.get('unit'))
    if multiplier:
        value = value * multiplier

    parsed_date = datetime.strptime(record.get('creationDate'), '%Y-%m-%d %H:%M:%S %z')
    yield item_class(date_time=parsed_date, value=value)


def parse(file_like_object):
    doc = etree.parse(file_like_object)

    for record in doc.iter('Record'):
        yield from parse_record(record)


if __name__ == '__main__':
    with open(EXPORT_FILE, 'r') as handle:
        for item in parse(handle):
            print(item)