def create_timeseries_jhu(data, lookup_table, value_name):
try:
logger.info('timeseries: ' + value_name)
id_vars = "region"
var_name = "date"
timeseries = pd.melt(data,
id_vars=id_vars,
var_name=var_name,
value_name=value_name)
timeseries.loc[:, var_name] = pd.to_datetime(
timeseries.loc[:, var_name])
timeseries = pd.merge(
lookup_table[[
"iso2", "iso3", "code3", "Lat", "Lon", id_vars
]].groupby(id_vars).first(),
timeseries,
on=id_vars,
how="inner",
)
return timeseries
except BaseException:
logger.error('failed to create timeseries')
return None
def countries_geojson(self):
logger.info("geojson")
error_msg = "cannot load geojson data, no url provided"
try:
with urlopen(self.parser.get("urls",
"mapbox_countries_url")) as response:
countries = json.load(response)
return countries
except BaseException:
logger.error(error_msg)
return None
def read_prepare_data(url):
try:
# logger.info(url)
data_raw = pd.read_csv(self.parser.get("urls", url))
data_raw.rename(columns={"Country/Region": "region"},
inplace=True)
data = (data_raw.groupby("region").sum().drop(
columns=["Lat", "Long"]).reset_index())
logger.info('length {}: {}'.format(url.rstrip('_url'),
len(data)))
return data
except BaseException:
logger.error('prepare_data')
logger.error(error_msg)
return None
def post(self, cid):
logger.debug(f"{str(request).replace(request.url_root, '/')}")
post_data = request.form
name = post_data.get('name')
gender = post_data.get('gender')
birthday = post_data.get('birthday')
if gender == '男':
gender = 1
else:
gender = 0
student = Student()
student.cid = int(cid)
student.name = name
student.gender = gender
try:
birthday_date = DT.date_str2date(birthday)
student.birthday = birthday_date
db.session.add(student)
db.session.commit()
msg = f'添加成功！'
logger.success(msg)
status_code = 200
return make_response(
jsonify(
dict(
status_code=status_code,
msg=msg
)
), status_code
)
except Exception as e:
msg = f'error:{e} \n请按照 2001-01-01的格式输入生日'
logger.error(msg)
status_code = 400
return make_response(
jsonify(
dict(
status_code=status_code,
msg=msg
)
), status_code
)
def read_geonames_country_info(self):
logger.info("geonames")
error_msg = "cannot load geonames data, no url provided"
try:
res = requests.get(
self.parser.get("urls", "geonames_countries_url"))
soup = BeautifulSoup(res.content, "lxml")
table = soup.find_all("table", id="countries")
country_info = pd.read_html(str(table), keep_default_na=False)[0]
country_info.rename(
columns={
"ISO-3166alpha2": "iso_alpha2",
"ISO-3166alpha3": "iso_alpha",
"ISO-3166numeric": "iso_num",
"Country": "region",
"Population": "population",
"Continent": "continent",
"Area in km²": "area",
},
inplace=True,
)
country_info.loc[country_info["continent"] == "EU",
"continent"] = "Europe"
country_info.loc[country_info["continent"] == "NA",
"continent"] = "North-A."
country_info.loc[country_info["continent"] == "SA",
"continent"] = "South-A."
country_info.loc[country_info["continent"] == "AS",
"continent"] = "Asia"
country_info.loc[country_info["continent"] == "OC",
"continent"] = "Oceania"
country_info.loc[country_info["continent"] == "AF",
"continent"] = "Africa"
return country_info
except BaseException:
logger.error(error_msg)
return None
def load_jhu(self):
logger.info("load jhu")
error_msg = "cannot load JHU data, no url provided"
def read_prepare_data(url):
try:
# logger.info(url)
data_raw = pd.read_csv(self.parser.get("urls", url))
data_raw.rename(columns={"Country/Region": "region"},
inplace=True)
data = (data_raw.groupby("region").sum().drop(
columns=["Lat", "Long"]).reset_index())
logger.info('length {}: {}'.format(url.rstrip('_url'),
len(data)))
return data
except BaseException:
logger.error('prepare_data')
logger.error(error_msg)
return None
def create_timeseries_jhu(data, lookup_table, value_name):
try:
logger.info('timeseries: ' + value_name)
id_vars = "region"
var_name = "date"
timeseries = pd.melt(data,
id_vars=id_vars,
var_name=var_name,
value_name=value_name)
timeseries.loc[:, var_name] = pd.to_datetime(
timeseries.loc[:, var_name])
timeseries = pd.merge(
lookup_table[[
"iso2", "iso3", "code3", "Lat", "Lon", id_vars
]].groupby(id_vars).first(),
timeseries,
on=id_vars,
how="inner",
)
return timeseries
except BaseException:
logger.error('failed to create timeseries')
return None
try:
lookup_table = pd.read_csv(
self.parser.get("urls", "jhu_lookup_url"))
lookup_table.rename(columns={
"Country_Region": "region",
"Long_": "Lon"
},
inplace=True)
logger.info('length data ({}): {}'.format('lookup table',
len(lookup_table)))
except BaseException:
logger.error('lookup table')
logger.error(error_msg)
return None
if not lookup_table.empty:
confirmed_data = read_prepare_data("jhu_confirmed_url")
deaths_data = read_prepare_data("jhu_deaths_url")
confirmed = create_timeseries_jhu(confirmed_data, lookup_table,
"confirmed")
deaths = create_timeseries_jhu(deaths_data, lookup_table, "deaths")
data = pd.merge(
deaths[["date", "region", "iso3", "Lat", "Lon", "deaths"]],
confirmed[["date", "confirmed", "iso3"]],
on=["iso3", "date"],
how="inner",
)
# data = pd.merge(
# data,
# recovered[["date", "recovered", "iso3"]],
# on=["iso3", "date"],
# how="inner",
# )
data.rename(columns={"confirmed": "cases"}, inplace=True)
return data
else:
logger.error(error_msg)
return None