import csv
import arrow
import os
import collections
import json
import re
import matplotlib

ref_time = arrow.get()
data_dir = '/Users/fluffy/Library/Mobile Documents/iCloud~com~ifunography~HealthExport/Documents/'

start_time, _ = ref_time.shift(days=-30).span('day')
_, end_time = ref_time.span('day', bounds='[]')

metrics = collections.defaultdict(list)

def filter(date):
    if date >= start_time and date < end_time:
        return date
    return None

def scan_dir(path):
    for file in os.scandir(path):
        if file.is_dir():
            scan_dir(file.path)
        elif file.is_file():
            scan_file(file)

def scan_file(file):
    match = re.match(r'HealthAutoExport-([0-9\-]+)\.json', file.name)
    if not match:
        return
    file_date = arrow.get(match[1])
    if filter(file_date):
        file_data = json.load(open(file.path,'r'))

        scan_metrics(file_data['data']['metrics'])

def get_value(item):
    if 'qty' in item:
        return item['qty']
    elif 'Avg' in item:
        return item['Avg']
    elif 'asleep' in item:
        print(item)
        return None
    else:
        raise KeyError(f"unknown format; keys: {item.keys()}")

def scan_metrics(data):
    for metric in data:
        if len(data):
            sink = metrics[metric['name']]
            for item in metric['data']:
                if when := filter(arrow.get(item['date'], 'YYYY-MM-DD HH:mm:ss Z')):
                    sink.append((when, item))

scan_dir(data_dir)
metrics = {k:sorted(v,key=lambda i:i[0]) for k,v in metrics.items() if len(v)}

for key,vals in metrics.items():
    print(key,len(vals))