def get_rates(ds, **kwargs):
pg_hook = PostgresHook(postgres_conn_id='rates')
api_hook = HttpHook(http_conn_id='openexchangerates', method='GET')
# If either of these raises an exception then we'll be notified via
# Airflow
resp = api_hook.run('')
resp = json.loads(resp.content)
# These are the only valid pairs the DB supports at the moment. Anything
# else that turns up will be ignored.
valid_pairs = (
'AED', 'AFN', 'ALL', 'AMD', 'ANG', 'AOA', 'ARS',
'AUD', 'AWG', 'AZN', 'BAM', 'BBD', 'BDT', 'BGN',
'BHD', 'BIF', 'BMD', 'BND', 'BOB', 'BRL', 'BSD',
'BTC', 'BTN', 'BWP', 'BYN', 'BYR', 'BZD', 'CAD',
'CDF', 'CHF', 'CLF', 'CLP', 'CNY', 'COP', 'CRC',
'CUC', 'CUP', 'CVE', 'CZK', 'DJF', 'DKK', 'DOP',
'DZD', 'EEK', 'EGP', 'ERN', 'ETB', 'EUR', 'FJD',
'FKP', 'GBP', 'GEL', 'GGP', 'GHS', 'GIP', 'GMD',
'GNF', 'GTQ', 'GYD', 'HKD', 'HNL', 'HRK', 'HTG',
'HUF', 'IDR', 'ILS', 'IMP', 'INR', 'IQD', 'IRR',
'ISK', 'JEP', 'JMD', 'JOD', 'JPY', 'KES', 'KGS',
'KHR', 'KMF', 'KPW', 'KRW', 'KWD', 'KYD', 'KZT',
'LAK', 'LBP', 'LKR', 'LRD', 'LSL', 'LTL', 'LVL',
'LYD', 'MAD', 'MDL', 'MGA', 'MKD', 'MMK', 'MNT',
'MOP', 'MRO', 'MTL', 'MUR', 'MVR', 'MWK', 'MXN',
'MYR', 'MZN', 'NAD', 'NGN', 'NIO', 'NOK', 'NPR',
'NZD', 'OMR', 'PAB', 'PEN', 'PGK', 'PHP', 'PKR',
'PLN', 'PYG', 'QAR', 'RON', 'RSD', 'RUB', 'RWF',
'SAR', 'SBD', 'SCR', 'SDG', 'SEK', 'SGD', 'SHP',
'SLL', 'SOS', 'SRD', 'STD', 'SVC', 'SYP', 'SZL',
'THB', 'TJS', 'TMT', 'TND', 'TOP', 'TRY', 'TTD',
'TWD', 'TZS', 'UAH', 'UGX', 'USD', 'UYU', 'UZS',
'VEF', 'VND', 'VUV', 'WST', 'XAF', 'XAG', 'XAU',
'XCD', 'XDR', 'XOF', 'XPD', 'XPF', 'XPT', 'YER',
'ZAR', 'ZMK', 'ZMW', 'ZWL')
rates_insert = """INSERT INTO rates (pair, valid_until, rate)
VALUES (%s, %s, %s);"""
# If this raises an exception then we'll be notified via Airflow
valid_until = datetime.fromtimestamp(resp['timestamp'])
for (iso2, rate) in resp['rates'].iteritems():
# If converting the rate to a float fails for whatever reason then
# just move on.
try:
rate = float(rate)
except:
continue
iso2 = iso2.upper().strip()
if iso2 not in valid_pairs or rate < 0:
continue
pg_hook.run(rates_insert, parameters=(iso2,
valid_until,
rate))
def execute(self, context):
http = HttpHook(self.method, http_conn_id=self.http_conn_id)
logging.info("Calling HTTP method")
response = http.run(self.endpoint, self.data, self.headers,
self.extra_options)
if self.response_check:
if not self.response_check(response):
raise AirflowException("Response check returned False.")
def execute(self, context):
http = HttpHook(self.method, http_conn_id=self.http_conn_id)
logging.info("Calling HTTP method")
response = http.run(self.endpoint,
self.data,
self.headers,
self.extra_options)
if self.response_check:
if not self.response_check(response):
raise AirflowException("Response check returned False.")
def download_api_data(**kwargs):
tfl_hook = HttpHook(http_conn_id='tfl_parking_conn', method='GET')
resp = tfl_hook.run('', extra_options={'check_response': True})
filepostfix = str(time.time()) + '.json'
base_dir = Variable.get("tfl_park_base_dir")
filename = join(base_dir, filepostfix)
with open(filename, 'w') as f:
f.write(resp.text)
def get_people(**kwargs):
api_hook = HttpHook(http_conn_id='swapi_people', method='GET')
for i in range(9):
response = api_hook.run('', data={'page': i + 1})
response_json = json.loads(response.content)
print(response_json)
store_people(response_json['results'])
time.sleep(3)
return True
def __init__(
self,
spark_script,
session_kind="spark", # spark, pyspark, or sparkr
http_conn_id='http_default',
poll_interval=30,
*args,
**kwargs):
super(LivySparkOperator, self).__init__(*args, **kwargs)
self.spark_script = spark_script
self.session_kind = session_kind
self.http_conn_id = http_conn_id
self.poll_interval = poll_interval
self.http = HttpHook("GET", http_conn_id=self.http_conn_id)
def get_rates(ds, **kwargs):
# connection: a Airflow connection
pg_hook = PostgresHook(postgres_conn_id='crypto')
api_hook = HttpHook(http_conn_id='cryptocoincharts_eth', method='GET')
resp = api_hook.run('')
resp = json.loads(resp.content)
rates_insert = """INSERT INTO rates (id, price, last_price, volume)
VALUES (%s, %s, %s, %s);"""
markets_insert = """INSERT INTO markets (market, pair, price, volume_btc, volume)
VALUES (%s, %s, %s, %s, %s);"""
pg_hook.run(rates_insert, parameters=(resp['id'], resp['price'], resp['price_before_24h'], resp['volume_second']))
for market in resp['markets']:
pg_hook.run(markets_insert, parameters=(market['market'], PAIR, market['price'], market['volume_btc'], market['volume']))
class KairosDBOperator(BaseOperator):
"""
Operator to facilitate interacting with the kairosDB which executes Apache Spark code via a REST API.
:param query: Scala, the kairos query
:type spark_script: string
"""
@apply_defaults
def __init__(self, query, http_conn_id='http_kairosdb', *args, **kwargs):
super(KairosDBOperator, self).__init__(*args, **kwargs)
self.query = query
self.http_conn_id = http_conn_id
self.acceptable_response_codes = [200, 201]
self.http = HttpHook("GET", http_conn_id=self.http_conn_id)
def execute(self, context):
# Simple test
headers = {'Content-Type': 'application/json'}
response = self._http_rest_call("POST",
"/api/v1/datapoints/query",
data=json.dumps(self.query),
headers=headers)
logging.debug("Status code: %d" % response.status_code)
if not (response.status_code in self.acceptable_response_codes):
return None
r = response.json()
logging.debug("JSON response: %s" % r)
if r:
return r.get("queries")
else:
return None
def _http_rest_call(self,
method,
endpoint,
data=None,
headers=None,
extra_options=None):
if not extra_options:
extra_options = {}
logging.info("Performing HTTP REST call... (method: " + str(method) +
", endpoint: " + str(endpoint) + ", data: " + str(data) +
", headers: " + str(headers) + ")")
self.http.method = method
response = self.http.run(endpoint,
data,
headers,
extra_options=extra_options)
logging.debug("status_code: " + str(response.status_code))
logging.debug("response_as_json: " + str(response.json()))
return response
def get_stocks(ds, **context):
symbol = context['params']['symbol']
pg_hook = PostgresHook(postgres_conn_id='stocks')
api_hook = HttpHook(http_conn_id='alphavantage', method='GET')
# If either of these raises an exception then we'll be notified via
# Airflow
resp = api_hook.run(f'query?function=TIME_SERIES_DAILY_ADJUSTED&symbol={symbol}&apikey=537201H9R203WT4C&datatype=csv')
resp = json.loads(resp.content)
# These are the only valid stocks the DB supports at the moment. Anything
# else that turns up will be ignored.
stocks_insert = """INSERT INTO stocks (symbol, valid_until, price)
VALUES (%s, %s, %s);"""
# If this raises an exception then we'll be notified via Airflow
valid_until = datetime.fromtimestamp(resp['timestamp'])
for iso2, price in resp['stocks'].items():
# If converting the price to a float fails for whatever reason then
# just move on.
try:
price = float(price)
except:
continue
iso2 = iso2.upper().strip()
if iso2 not in stocks or price < 0:
continue
pg_hook.run(stocks_insert, parameters=(iso2,
valid_until,
price))
class LivySparkOperator(BaseOperator):
"""
Operator to facilitate interacting with the Livy Server which executes Apache Spark code via a REST API.
:param spark_script: Scala, Python or R code to submit to the Livy Server (templated)
:type spark_script: string
:param session_kind: Type of session to setup with Livy. This will determine which type of code will be accepted. Possible values include "spark" (executes Scala code), "pyspark" (executes Python code) or "sparkr" (executes R code).
:type session_kind: string
:param http_conn_id: The http connection to run the operator against
:type http_conn_id: string
:param poll_interval: The polling interval to use when checking if the code in spark_script has finished executing. In seconds. (default: 30 seconds)
:type poll_interval: integer
"""
template_fields = ['spark_script'] # todo : make sure this works
template_ext = ['.py', '.R', '.r']
ui_color = '#34a8dd' # Clouderas Main Color: Blue
acceptable_response_codes = [200, 201]
@apply_defaults
def __init__(
self,
spark_script,
session_kind="spark", # spark, pyspark, or sparkr
http_conn_id='http_default',
poll_interval=30,
*args,
**kwargs):
super(LivySparkOperator, self).__init__(*args, **kwargs)
self.spark_script = spark_script
self.session_kind = session_kind
self.http_conn_id = http_conn_id
self.poll_interval = poll_interval
self.http = HttpHook("GET", http_conn_id=self.http_conn_id)
def execute(self, context):
logging.info("Executing LivySparkOperator.execute(context)")
logging.info("Validating arguments...")
self._validate_arguments()
logging.info("Finished validating arguments")
logging.info("Creating a Livy Session...")
session_id = self._create_session()
logging.info("Finished creating a Livy Session. (session_id: " +
str(session_id) + ")")
logging.info("Submitting spark script...")
statement_id, overall_statements_state = self._submit_spark_script(
session_id=session_id)
logging.info("Finished submitting spark script. (statement_id: " +
str(statement_id) + ", overall_statements_state: " +
str(overall_statements_state) + ")")
poll_for_completion = (overall_statements_state == "running")
if poll_for_completion:
logging.info(
"Spark job did not complete immediately. Starting to Poll for completion..."
)
while overall_statements_state == "running": # todo: test execution_timeout
logging.info("Sleeping for " + str(self.poll_interval) +
" seconds...")
time.sleep(self.poll_interval)
logging.info(
"Finished sleeping. Checking if Spark job has completed...")
statements = self._get_session_statements(session_id=session_id)
is_all_complete = True
for statement in statements:
if statement["state"] == "running":
is_all_complete = False
if is_all_complete:
overall_statements_state = "available"
logging.info(
"Finished checking if Spark job has completed. (overall_statements_state: "
+ str(overall_statements_state) + ")")
if poll_for_completion:
logging.info("Finished Polling for completion.")
logging.info("Session Logs:\n" +
str(self._get_session_logs(session_id=session_id)))
for statement in self._get_session_statements(session_id):
logging.info("Statement '" + str(statement["id"]) + "' Output:\n" +
str(statement["output"]))
logging.info("Closing session...")
response = self._close_session(session_id=session_id)
logging.info("Finished closing session. (response: " + str(response) +
")")
logging.info("Finished executing LivySparkOperator.execute(context)")
def _validate_arguments(self):
if self.session_kind is None or self.session_kind == "":
raise AirflowException(
"session_kind argument is invalid. It is empty or None. (value: '"
+ str(self.session_kind) + "')")
elif self.session_kind not in ["spark", "pyspark", "sparkr"]:
raise AirflowException(
"session_kind argument is invalid. It should be set to 'spark', 'pyspark', or 'sparkr'. (value: '"
+ str(self.session_kind) + "')")
def _get_sessions(self):
method = "GET"
endpoint = "sessions"
response = self._http_rest_call(method=method, endpoint=endpoint)
if response.status_code in self.acceptable_response_codes:
return response.json()["sessions"]
else:
raise AirflowException("Call to get sessions didn't return " +
str(self.acceptable_response_codes) +
". Returned '" + str(response.status_code) +
"'.")
def _get_session(self, session_id):
sessions = self._get_sessions()
for session in sessions:
if session["id"] == session_id:
return session
def _get_session_logs(self, session_id):
method = "GET"
endpoint = "sessions/" + str(session_id) + "/log"
response = self._http_rest_call(method=method, endpoint=endpoint)
return response.json()
def _create_session(self):
method = "POST"
endpoint = "sessions"
data = {"kind": self.session_kind}
response = self._http_rest_call(method=method,
endpoint=endpoint,
data=data)
if response.status_code in self.acceptable_response_codes:
response_json = response.json()
session_id = response_json["id"]
session_state = response_json["state"]
if session_state == "starting":
logging.info(
"Session is starting. Polling to see if it is ready...")
session_state_polling_interval = 10
while session_state == "starting":
logging.info("Sleeping for " +
str(session_state_polling_interval) + " seconds")
time.sleep(session_state_polling_interval)
session_state_check_response = self._get_session(
session_id=session_id)
session_state = session_state_check_response["state"]
logging.info("Got latest session state as '" + session_state +
"'")
return session_id
else:
raise AirflowException(
"Call to create a new session didn't return " +
str(self.acceptable_response_codes) + ". Returned '" +
str(response.status_code) + "'.")
def _submit_spark_script(self, session_id):
method = "POST"
endpoint = "sessions/" + str(session_id) + "/statements"
logging.info("Executing Spark Script: \n" + str(self.spark_script))
data = {'code': textwrap.dedent(self.spark_script)}
response = self._http_rest_call(method=method,
endpoint=endpoint,
data=data)
if response.status_code in self.acceptable_response_codes:
response_json = response.json()
return response_json["id"], response_json["state"]
else:
raise AirflowException(
"Call to create a new statement didn't return " +
str(self.acceptable_response_codes) + ". Returned '" +
str(response.status_code) + "'.")
def _get_session_statements(self, session_id):
method = "GET"
endpoint = "sessions/" + str(session_id) + "/statements"
response = self._http_rest_call(method=method, endpoint=endpoint)
if response.status_code in self.acceptable_response_codes:
response_json = response.json()
statements = response_json["statements"]
return statements
else:
raise AirflowException(
"Call to get the session statement response didn't return " +
str(self.acceptable_response_codes) + ". Returned '" +
str(response.status_code) + "'.")
def _close_session(self, session_id):
method = "DELETE"
endpoint = "sessions/" + str(session_id)
return self._http_rest_call(method=method, endpoint=endpoint)
def _http_rest_call(self,
method,
endpoint,
data=None,
headers=None,
extra_options=None):
if not extra_options:
extra_options = {}
logging.debug("Performing HTTP REST call... (method: " + str(method) +
", endpoint: " + str(endpoint) + ", data: " + str(data) +
", headers: " + str(headers) + ")")
self.http.method = method
response = self.http.run(endpoint,
json.dumps(data),
headers,
extra_options=extra_options)
logging.debug("status_code: " + str(response.status_code))
logging.debug("response_as_json: " + str(response.json()))
return response
def __init__(self, query, http_conn_id='http_kairosdb', *args, **kwargs):
super(KairosDBOperator, self).__init__(*args, **kwargs)
self.query = query
self.http_conn_id = http_conn_id
self.acceptable_response_codes = [200, 201]
self.http = HttpHook("GET", http_conn_id=self.http_conn_id)
def httpcall(**kwargs):
api_hook = HttpHook(http_conn_id='http_default', method='GET')
resp = api_hook.run('')
resp = json.loads(resp.text)
print resp
