def _count(self, args):
key = args.get('key')
persons = Person.find()
total = len(persons)
df = pd.DataFrame(persons)
current_date = df['release_date'].max()
result = {
'status': 'success',
'current_date': current_date,
'total': total
}
if not key:
return result
sum = df[key].value_counts().to_dict()
rows = list(map(lambda k: {key: k, 'count': sum[k]}, sum))
rows = sorted(rows, key=lambda r: r[key])
result['rows'] = rows
return result
def _find(self, args={}):
filters = []
if args.get('age'):
filters.append({'key': 'age', 'value': request.args['age']})
if args.get('sex'):
filters.append({'key': 'sex', 'value': request.args['sex']})
if args.get('area'):
filters.append({'key': 'area', 'value': request.args['area']})
if args.get('reason'):
filters.append({'key': 'reason', 'value': request.args['reason']})
if args.get('status'):
filters.append({'key': 'status', 'value': request.args['status']})
if args.get('cluster_no'):
filters.append({
'key': 'cluster_no',
'value': int(request.args['cluster_no'])
})
if args.get('release_date'):
filters.append({
'key': 'release_date',
'value': request.args['release_date']
})
else:
if args.get('from_date'):
filters.append({
'key': 'release_date',
'symbol': '>=',
'value': request.args['from_date']
})
if args.get('to_date'):
filters.append({
'key': 'release_date',
'symbol': '<=',
'value': request.args['to_date']
})
offset = args.get('offset', '')
offset = int(offset) if str.isdecimal(offset) else 0
limit = args.get('limit', '')
limit = int(limit) if str.isdecimal(limit) else None
persons = Person.find(filters=filters, offset=offset, limit=limit)
total = Person.count(filters=filters)
current_date = Person.current_date()
result = {
'status': 'success',
'current_date': current_date,
'persons': persons,
'total': total
}
return result
def _cross(self, args):
row_key = args.get('row')
if not row_key:
return {
'status': 'failure',
'message': 'Parameter "row" not defined.',
}
col_key = args.get('col')
if not col_key:
return {
'status': 'failure',
'message': 'Parameter "col" not defined.',
}
persons = Person.find()
total = len(persons)
df = pd.DataFrame(persons)
current_date = df['release_date'].max()
row_total = df[row_key].value_counts().to_dict()
table = pd.crosstab(df[col_key], df[row_key])
data = table.to_dict()
rows = []
for key in data:
row = {
row_key: key,
'values': list(map(lambda name: {'name': name, 'count': data[key][name]}, data[key])),
'total': row_total.get(key) or 0
}
rows.append(row)
col_total = df[col_key].value_counts().to_dict()
col_total = list(map(lambda name: {'name': name, 'count': col_total[name]}, col_total))
result = {
'status': 'success',
'current_date': current_date,
'rows': rows,
'col_total': col_total,
'total': total
}
return result
# 'refer': {'government': 'nagoya', 'release_no': 305},
# 'release_date': '2020-07-15',
# 'remarks': [],
# 'route': {'area': '東京都', 'text': '東京都'},
# 'sex': 'female',
# })
return result
# Main
if __name__ == '__main__':
last_date = None
last_no = 0
current_persons = Person.find(order=['-no'], limit=1)
if len(current_persons) > 0:
last_date = current_persons[-1]['release_date']
last_no = current_persons[-1]['no']
releases = read_aichi_release(last_date=last_date)
for release in releases:
# filepath = os.path.join(
# config.DATA_DIR, 'aichi/releases',
# 'aichi_release_{}.pdf'.format(release['pdf']['current_date'].strftime(r'%Y%m%d')))
filepath = os.path.join(
config.DATA_DIR, 'aichi/releases', 'aichi_release_{}.pdf'.format(
re.sub(r'^.*\/(\d+)\.pdf$', r'\1', release['pdf']['url'])))
if not os.path.exists(filepath):
logger.info('Download PDF file %s', filepath)
util.download_file(release['pdf']['url'], filepath=filepath)
