def _fetch_data(self, config, provider):
url = config['url']
api_key = config['api_key']
last_update = provider.get(
'last_updated', utcfromtimestamp(0)).strftime('%Y-%m-%dT%H:%M:%S')
# Results are pagified so we'll read this many at a time
offset_jump = 10
params = {'start': last_update, 'limit': offset_jump}
headers = {'apikey': api_key}
items = []
offset = 0
while True:
params['offset'] = offset
try:
response = requests.get(url,
params=params,
headers=headers,
timeout=30)
except requests.exceptions.ConnectionError as err:
raise IngestApiError.apiConnectionError(exception=err)
if response.ok:
# The total number of results are given to us in json, get them
# via a regex to read the field so we don't have to convert the
# whole thing to json pointlessly
item_ident = re.search('\"total\": *[0-9]*',
response.text).group()
results_str = re.search('[0-9]+', item_ident).group()
if results_str is None:
raise IngestApiError.apiGeneralError(
Exception(response.text), provider)
num_results = int(results_str)
if num_results > 0:
items.append(response.text)
if offset >= num_results:
return items
offset += offset_jump
else:
if re.match('Error: No API Key provided', response.text):
raise IngestApiError.apiAuthError(Exception(response.text),
provider)
elif response.status_code == 404:
raise IngestApiError.apiNotFoundError(
Exception(response.reason), provider)
else:
raise IngestApiError.apiGeneralError(
Exception(response.reason), provider)
return items
def _update(self, provider, update):
try:
config = provider['config']
user = config['username']
password = config['password']
id_list = config['idList']
except KeyError as e:
SuperdeskIngestError.notConfiguredError(
Exception('username, password and idList are needed'))
# we remove spaces and empty values from id_list to do a clean list
id_list = ','.join(
[id_.strip() for id_ in id_list.split(',') if id_.strip()])
params = {
'idList': id_list,
'idListType': 'products',
'format': '5',
'maxItems': '25',
'sortOrder': 'chronological'
}
try:
min_date_time = provider['private']['min_date_time']
sequence_number = provider['private']['sequence_number']
except KeyError:
pass
else:
params['minDateTime'] = min_date_time
params['sequenceNumber'] = sequence_number
try:
r = requests.get(URL, auth=(user, password), params=params)
except Exception as e:
raise IngestApiError.apiRequestError(
Exception('error while doing the request'))
try:
root_elt = etree.fromstring(r.content)
except Exception as e:
raise IngestApiError.apiRequestError(
Exception('error while doing the request'))
parser = self.get_feed_parser(provider)
items = parser.parse(root_elt, provider)
try:
min_date_time = root_elt.xpath(
'//iptc:timestamp[@role="minDateTime"]/text()',
namespaces=NS)[0].strip()
sequence_number = root_elt.xpath('//iptc:transmitId/text()',
namespaces=NS)[0].strip()
except IndexError:
raise IngestApiError.apiRequestError(
Exception('missing minDateTime or transmitId'))
else:
update.setdefault('private', {})
update['private']['min_date_time'] = min_date_time
update['private']['sequence_number'] = sequence_number
return [items]
def _fetch_data(self, config, provider):
"""Fetch the latest feed data.
:param dict config: RSS resource configuration
:param provider: data provider instance, needed as an argument when
raising ingest errors
:return: fetched RSS data
:rtype: str
:raises IngestApiError: if fetching data fails for any reason
(e.g. authentication error, resource not found, etc.)
"""
url = config["url"]
if config.get("auth_required", False):
auth = (config.get("username"), config.get("password"))
else:
auth = None
response = requests.get(url, auth=auth)
if response.ok:
return response.content
else:
if response.status_code in (401, 403):
raise IngestApiError.apiAuthError(Exception(response.reason), provider)
elif response.status_code == 404:
raise IngestApiError.apiNotFoundError(Exception(response.reason), provider)
else:
raise IngestApiError.apiGeneralError(Exception(response.reason), provider)
def _test(self, provider):
config = provider.get('config', {})
url = config['url']
api_key = config['api_key']
# limit the data to a single article and filter out all article fields
# to save bandwidth
params = {'limit': 1, 'fields': 'id'}
headers = {'apikey': api_key}
try:
response = requests.get(url,
params=params,
headers=headers,
timeout=30)
except requests.exceptions.ConnectionError as err:
raise IngestApiError.apiConnectionError(exception=err)
if not response.ok:
if response.status_code == 404:
raise IngestApiError.apiNotFoundError(
Exception(response.reason), provider)
else:
raise IngestApiError.apiGeneralError(
Exception(response.reason), provider)
def _update(self, provider, update):
config = self.config
try:
user, password = self.config["username"], self.config["password"]
except KeyError:
SuperdeskIngestError.notConfiguredError(
Exception("username and password are needed"))
url_override = config.get("url", "").strip()
if not url_override.startswith("http"):
SuperdeskIngestError.notConfiguredError(
Exception("if URL is set, it must be a valid http link"))
if url_override:
params = {"user": user, "password": password, "maksAntal": 50}
else:
params = {
"user": user,
"password": password,
"maksAntal": 50,
"waitAcknowledge": "true"
}
r = self.get_url(url_override, params=params)
try:
root_elt = etree.fromstring(r.text)
except Exception:
raise IngestApiError.apiRequestError(
Exception("error while parsing the request answer"))
try:
if root_elt.xpath("(//error/text())[1]")[0] != "0":
err_msg = root_elt.xpath("(//errormsg/text())[1]")[0]
raise IngestApiError.apiRequestError(
Exception("error code returned by API: {msg}".format(
msg=err_msg)))
except IndexError:
raise IngestApiError.apiRequestError(
Exception("Invalid XML, <error> element not found"))
parser = self.get_feed_parser(provider)
items = []
for elt in root_elt.xpath("//RBNews"):
item = parser.parse(elt, provider)
items.append(item)
if not url_override:
try:
queue_id = elt.xpath(".//ServiceQueueId/text()")[0]
except IndexError:
raise IngestApiError.apiRequestError(
Exception("missing ServiceQueueId element"))
ack_params = {
"user": user,
"password": password,
"servicequeueid": queue_id
}
self.get_url(URL_ACK, params=ack_params)
return [items]
def _generate_auth_token(self, provider):
"""
Generates Authentication Token as per the configuration in Ingest Provider.
:param provider: dict - Ingest provider details to which the current directory has been configured
:type provider: dict :py:class: `superdesk.io.ingest_provider_model.IngestProviderResource`
:return: token details if successfully authenticated
:rtype: str
:raises: IngestApiError.apiGeneralError() if auth_url is missing in the Ingest Provider configuration
"""
session = requests.Session()
session.mount('https://', SSLAdapter())
auth_url = provider.get('config', {}).get('auth_url', None)
if not auth_url:
raise IngestApiError.apiGeneralError(provider=provider,
exception=KeyError(
'''
Ingest Provider {} is missing Authentication URL.
Please check the configuration.
'''.format(provider['name']))
)
payload = {
'username': provider.get('config', {}).get('username', ''),
'password': provider.get('config', {}).get('password', ''),
}
response = session.get(auth_url, params=payload, verify=False, timeout=30)
if response.status_code < 200 or response.status_code >= 300:
raise IngestApiError.apiAuthError(provider=provider)
tree = etree.fromstring(response.content) # workaround for http mock lib
return tree.text
def _fetch_data(self, config, provider):
"""Fetch the latest feed data.
:param dict config: RSS resource configuration
:param provider: data provider instance, needed as an argument when
raising ingest errors
:return: fetched RSS data
:rtype: str
:raises IngestApiError: if fetching data fails for any reason
(e.g. authentication error, resource not found, etc.)
"""
url = config['url']
if config.get('auth_required', False):
auth = (config.get('username'), config.get('password'))
else:
auth = None
response = requests.get(url, auth=auth)
if response.ok:
return response.content
else:
if response.status_code in (401, 403):
raise IngestApiError.apiAuthError(
Exception(response.reason), provider)
elif response.status_code == 404:
raise IngestApiError.apiNotFoundError(
Exception(response.reason), provider)
else:
raise IngestApiError.apiGeneralError(
Exception(response.reason), provider)
class WufooFeedingService(FeedingService):
"""
Feeding Service class which can read article(s) using Wufoo API
"""
NAME = "wufoo"
ERRORS = [
IngestApiError.apiTimeoutError().get_error_description(),
IngestApiError.apiRedirectError().get_error_description(),
IngestApiError.apiRequestError().get_error_description(),
IngestApiError.apiGeneralError().get_error_description(),
]
label = "Wufoo feed API"
fields = [
{
"id": "wufoo_username",
"type": "text",
"label": "Login",
"placeholder": "Wufoo login",
"required": True
},
{
"id": "wufoo_api_key",
"type": "password",
"label": "API key",
"placeholder": "Wufoo API Key",
"required": True,
},
]
def __init__(self):
super().__init__()
self.fields_cache = {}
def _update(self, provider, update):
user = provider["config"]["wufoo_username"]
wufoo_data = {
"url": WUFOO_URL.format(subdomain=user),
"user": user,
"api_key": provider["config"]["wufoo_api_key"],
"form_query_entries_tpl": WUFOO_QUERY_FORM + WUFOO_QUERY_ENTRIES,
"update": update,
}
try:
parser = self.get_feed_parser(provider, None)
except requests.exceptions.Timeout as ex:
raise IngestApiError.apiTimeoutError(ex, provider)
except requests.exceptions.TooManyRedirects as ex:
raise IngestApiError.apiRedirectError(ex, provider)
except requests.exceptions.RequestException as ex:
raise IngestApiError.apiRequestError(ex, provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
items = parser.parse(wufoo_data, provider)
return [items]
def _update(self, provider, update):
updated = utcnow()
last_updated = provider.get('last_updated')
ttl_minutes = app.config['INGEST_EXPIRY_MINUTES']
if not last_updated or last_updated < updated - datetime.timedelta(
minutes=ttl_minutes):
last_updated = updated - datetime.timedelta(minutes=ttl_minutes)
self.provider = provider
provider_config = provider.get('config')
if not provider_config:
provider_config = {}
provider['config'] = provider_config
self.URL = provider_config.get('url')
payload = {}
parser = self.get_feed_parser(provider)
try:
response = requests.get(self.URL, params=payload, timeout=15)
# TODO: check if file has been updated since provider last_updated
# although some ptovider do not include 'Last-Modified' in headers
# so unsure how to do this
logger.info('Http Headers: %s', response.headers)
except requests.exceptions.Timeout as ex:
# Maybe set up for a retry, or continue in a retry loop
raise IngestApiError.apiTimeoutError(ex, self.provider)
except requests.exceptions.TooManyRedirects as ex:
# Tell the user their URL was bad and try a different one
raise IngestApiError.apiRedirectError(ex, self.provider)
except requests.exceptions.RequestException as ex:
# catastrophic error. bail.
raise IngestApiError.apiRequestError(ex, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
if response.status_code == 404:
raise LookupError('Not found %s' % payload)
logger.info('Ingesting: %s', str(response.content))
if isinstance(parser, NTBEventXMLFeedParser):
xml = ET.fromstring(response.content)
items = parser.parse(xml, provider)
elif isinstance(parser, IcsTwoFeedParser):
cal = Calendar.from_ical(response.content)
items = parser.parse(cal, provider)
else:
items = parser.parse(response.content)
if isinstance(items, list):
yield items
else:
yield [items]
class WufooFeedingService(FeedingService):
"""
Feeding Service class which can read article(s) using Wufoo API
"""
NAME = 'wufoo'
ERRORS = [
IngestApiError.apiTimeoutError().get_error_description(),
IngestApiError.apiRedirectError().get_error_description(),
IngestApiError.apiRequestError().get_error_description(),
IngestApiError.apiGeneralError().get_error_description()
]
label = 'Wufoo feed API'
fields = [{
'id': 'wufoo_username',
'type': 'text',
'label': 'Login',
'placeholder': 'Wufoo login',
'required': True
}, {
'id': 'wufoo_api_key',
'type': 'password',
'label': 'API key',
'placeholder': 'Wufoo API Key',
'required': True
}]
parser_restricted_values = ['wufoo']
def __init__(self):
self.fields_cache = {}
def _update(self, provider, update):
user = provider['config']['wufoo_username']
wufoo_data = {
"url": WUFOO_URL.format(subdomain=user),
"user": user,
"api_key": provider['config']['wufoo_api_key'],
"form_query_entries_tpl": WUFOO_QUERY_FORM + WUFOO_QUERY_ENTRIES,
"update": update
}
try:
parser = self.get_feed_parser(provider, None)
except requests.exceptions.Timeout as ex:
raise IngestApiError.apiTimeoutError(ex, provider)
except requests.exceptions.TooManyRedirects as ex:
raise IngestApiError.apiRedirectError(ex, provider)
except requests.exceptions.RequestException as ex:
raise IngestApiError.apiRequestError(ex, provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
items = parser.parse(wufoo_data, provider)
return [items]
def _fetch_data(self, config, provider):
url = config['url']
api_key = config['api_key']
last_update = provider.get('last_updated', utcfromtimestamp(0)).strftime('%Y-%m-%dT%H:%M:%S')
# Results are pagified so we'll read this many at a time
offset_jump = 10
params = {'start': last_update, 'limit': offset_jump}
headers = {'apikey': api_key}
items = []
offset = 0
while True:
params['offset'] = offset
try:
response = requests.get(url, params=params, headers=headers, timeout=30)
except requests.exceptions.ConnectionError as err:
raise IngestApiError.apiConnectionError(exception=err)
if response.ok:
# The total number of results are given to us in json, get them
# via a regex to read the field so we don't have to convert the
# whole thing to json pointlessly
item_ident = re.search('\"total\": *[0-9]*', response.text).group()
results_str = re.search('[0-9]+', item_ident).group()
if results_str is None:
raise IngestApiError.apiGeneralError(
Exception(response.text), provider)
num_results = int(results_str)
if num_results > 0:
items.append(response.text)
if offset >= num_results:
return items
offset += offset_jump
else:
if re.match('Error: No API Key provided', response.text):
raise IngestApiError.apiAuthError(
Exception(response.text), provider)
elif response.status_code == 404:
raise IngestApiError.apiNotFoundError(
Exception(response.reason), provider)
else:
raise IngestApiError.apiGeneralError(
Exception(response.reason), provider)
return items
def _update(self, provider, update):
try:
config = provider['config']
user = config['username']
password = config['password']
id_list = config['idList']
# before "products" was hardcoded as value for "idListType"
id_list_type = config.get('idListType', 'products')
if not user.strip() or not password.strip() or not id_list.strip():
raise KeyError
except KeyError:
raise SuperdeskIngestError.notConfiguredError(Exception('username, password and idList are needed'))
# we remove spaces and empty values from id_list to do a clean list
id_list = ','.join([id_.strip() for id_ in id_list.split(',') if id_.strip()])
params = {'idList': id_list,
'idListType': id_list_type,
'format': '5',
'maxItems': '25',
'sortOrder': 'chronological'}
try:
min_date_time = provider['private']['min_date_time']
sequence_number = provider['private']['sequence_number']
except KeyError:
pass
else:
params['minDateTime'] = min_date_time
params['sequenceNumber'] = sequence_number
try:
r = requests.get(URL, auth=(user, password), params=params)
except Exception:
raise IngestApiError.apiRequestError(Exception('error while doing the request'))
try:
root_elt = etree.fromstring(r.content)
except Exception:
raise IngestApiError.apiRequestError(Exception('error while doing the request'))
parser = self.get_feed_parser(provider)
items = parser.parse(root_elt, provider)
try:
min_date_time = root_elt.xpath('//iptc:timestamp[@role="minDateTime"]/text()', namespaces=NS)[0].strip()
sequence_number = root_elt.xpath('//iptc:transmitId/text()', namespaces=NS)[0].strip()
except IndexError:
raise IngestApiError.apiRequestError(Exception('missing minDateTime or transmitId'))
else:
update.setdefault('private', {})
update['private']['min_date_time'] = min_date_time
update['private']['sequence_number'] = sequence_number
return [items]
def _update(self, provider, update):
# Each update run will retrieve the data for a single "market"
market_index = provider.get('private', {}).get('market_index', 0)
markets = json.loads(
provider.get('config', {}).get('market_definitions',
[]).replace('\'', '"'))
market = markets[market_index]
logger.info('Retrieving fuel data for the {} market'.format(
market.get('market')))
try:
self.session_token = self._get_token(provider).get('id')
prices = self._get_prices(provider, market)
self._save(prices, market)
except Exception as ex:
raise IngestApiError.apiGeneralError(ex, self.provider)
finally:
# Save the next market to process
market_index = (market_index + 1) % len(markets)
get_resource_service('ingest_providers').system_update(
provider.get('_id'),
{'private': {
'market_index': market_index
}}, provider)
return None
def _generate_auth_token(self, provider):
"""
Generates Authentication Token as per the configuration in Ingest Provider.
:param provider: dict - Ingest provider details to which the current directory has been configured
:type provider: dict :py:class: `superdesk.io.ingest_provider_model.IngestProviderResource`
:return: token details if successfully authenticated
:rtype: str
:raises: IngestApiError.apiGeneralError() if auth_url is missing in the Ingest Provider configuration
"""
session = requests.Session()
session.mount('https://', SSLAdapter())
auth_url = provider.get('config', {}).get('auth_url', None)
if not auth_url:
raise IngestApiError.apiGeneralError(provider=provider,
exception=KeyError('''
Ingest Provider {} is missing Authentication URL.
Please check the configuration.
'''.format(
provider['name'])))
payload = {
'username': provider.get('config', {}).get('username', ''),
'password': provider.get('config', {}).get('password', ''),
}
response = session.get(auth_url,
params=payload,
verify=False,
timeout=30)
tree = etree.fromstring(
response.content) # workaround for http mock lib
return tree.text
def _get_worksheet(self, provider):
"""Get worksheet from google spreadsheet
:return: worksheet
:rtype: object
"""
scope = [
'https://spreadsheets.google.com/feeds',
'https://www.googleapis.com/auth/drive',
]
config = provider.get('config', {})
url = config.get('url', '')
service_account = config.get('service_account', '')
title = config.get('worksheet_title', '')
try:
service_account = json.loads(service_account)
credentials = ServiceAccountCredentials.from_json_keyfile_dict(
service_account, scope)
gc = gspread.authorize(credentials)
spreadsheet = gc.open_by_url(url)
permission = spreadsheet.list_permissions()[0]
if permission['role'] != 'writer':
raise IngestSpreadsheetError.SpreadsheetPermissionError()
worksheet = spreadsheet.worksheet(title)
return worksheet
except (json.decoder.JSONDecodeError, AttributeError, ValueError) as e:
# both permission and credential raise Value error
if e.args[0] == 15100:
raise IngestSpreadsheetError.SpreadsheetPermissionError()
raise IngestSpreadsheetError.SpreadsheetCredentialsError()
except gspread.exceptions.NoValidUrlKeyFound:
raise IngestApiError.apiNotFoundError()
except gspread.exceptions.WorksheetNotFound:
raise IngestSpreadsheetError.WorksheetNotFoundError()
except gspread.exceptions.APIError as e:
error = e.response.json()['error']
response_code = error['code']
logger.error('Provider %s: %s', provider.get('name'),
error['message'])
if response_code == 403:
raise IngestSpreadsheetError.SpreadsheetPermissionError()
elif response_code == 429:
raise IngestSpreadsheetError.SpreadsheetQuotaLimitError()
else:
raise IngestApiError.apiNotFoundError()
def _get_tree(self, endpoint, payload=None):
"""Get xml response for given API endpoint and payload.
:param: endpoint
:type endpoint: str
:param: payload
:type payload: str
"""
if payload is None:
payload = {}
payload['token'] = self._get_auth_token(self.provider, update=True)
url = self._get_absolute_url(endpoint)
if not self.session:
self.session = requests.Session()
retries = 0
while True:
try:
response = self.session.get(url,
params=payload,
timeout=(30, 15))
except requests.exceptions.Timeout as ex:
if retries < 3:
logger.warn(
'Reuters API timeout retrying, retries {}'.format(
retries))
retries += 1
continue
raise IngestApiError.apiTimeoutError(ex, self.provider)
except requests.exceptions.TooManyRedirects as ex:
# Tell the user their URL was bad and try a different one
raise IngestApiError.apiRedirectError(ex, self.provider)
except requests.exceptions.RequestException as ex:
# catastrophic error. bail.
raise IngestApiError.apiRequestError(ex, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
if response.status_code == 404:
raise LookupError(
_('Not found {payload}').format(payload=payload))
break
try:
return etree.fromstring(
response.content) # workaround for http mock lib
except UnicodeEncodeError as error:
traceback.print_exc()
raise IngestApiError.apiUnicodeError(error, self.provider)
except ParseError as error:
traceback.print_exc()
raise IngestApiError.apiParseError(error, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
def _update(self, provider, update):
config = self.config
try:
user, password = self.config['username'], self.config['password']
except KeyError:
SuperdeskIngestError.notConfiguredError(Exception('username and password are needed'))
url_override = config.get('url', '').strip()
if not url_override.startswith('http'):
SuperdeskIngestError.notConfiguredError(Exception('if URL is set, it must be a valid http link'))
if url_override:
params = {'user': user, 'password': password, 'maksAntal': 50}
else:
params = {'user': user, 'password': password, 'maksAntal': 50, 'waitAcknowledge': 'true'}
r = self.get_url(url_override, params=params)
try:
root_elt = etree.fromstring(r.text)
except Exception:
raise IngestApiError.apiRequestError(Exception('error while parsing the request answer'))
try:
if root_elt.xpath('(//error/text())[1]')[0] != '0':
err_msg = root_elt.xpath('(//errormsg/text())[1]')[0]
raise IngestApiError.apiRequestError(Exception('error code returned by API: {msg}'.format(msg=err_msg)))
except IndexError:
raise IngestApiError.apiRequestError(Exception('Invalid XML, <error> element not found'))
parser = self.get_feed_parser(provider)
items = []
for elt in root_elt.xpath('//RBNews'):
item = parser.parse(elt, provider)
items.append(item)
if not url_override:
try:
queue_id = elt.xpath('.//ServiceQueueId/text()')[0]
except IndexError:
raise IngestApiError.apiRequestError(Exception('missing ServiceQueueId element'))
ack_params = {'user': user, 'password': password, 'servicequeueid': queue_id}
self.get_url(URL_ACK, params=ack_params)
return [items]
def _generate_auth_token(self, provider):
"""
Generates Authentication Token as per the configuration in Ingest Provider.
:param provider: dict - Ingest provider details to which the current directory has been configured
:type provider: dict :py:class: `superdesk.io.ingest_provider_model.IngestProviderResource`
:return: token details if successfully authenticated
:rtype: str
:raises: IngestApiError.apiGeneralError() if auth_url is missing in the Ingest Provider configuration
"""
session = requests.Session()
session.mount("https://", SSLAdapter())
auth_url = provider.get("config", {}).get("auth_url", None)
if not auth_url:
raise IngestApiError.apiGeneralError(
provider=provider,
exception=KeyError(
"""
Ingest Provider {} is missing Authentication URL.
Please check the configuration.
""".format(
provider["name"]
)
),
)
payload = {
"username": provider.get("config", {}).get("username", ""),
"password": provider.get("config", {}).get("password", ""),
}
response = session.get(auth_url, params=payload, verify=False, timeout=30)
if response.status_code < 200 or response.status_code >= 300:
try:
response.raise_for_status()
except Exception:
err = IngestApiError.apiAuthError(provider=provider)
self.close_provider(provider, err, force=True)
raise err
tree = etree.fromstring(response.content) # workaround for http mock lib
return tree.text
def _fetch_data(self, config, provider):
"""Fetch the latest feed data.
:param dict config: RSS resource configuration
:param provider: data provider instance, needed as an argument when
raising ingest errors
:return: fetched RSS data
:rtype: str
:raises IngestApiError: if fetching data fails for any reason
(e.g. authentication error, resource not found, etc.)
"""
url = config['url']
if config.get('auth_required', False):
auth = (config.get('username'), config.get('password'))
self.auth_info = {
'username': config.get('username', ''),
'password': config.get('password', '')
}
else:
auth = None
try:
response = requests.get(url, auth=auth, timeout=30)
except requests.exceptions.ConnectionError as err:
raise IngestApiError.apiConnectionError(exception=err, provider=provider)
except requests.exceptions.RequestException as err:
raise IngestApiError.apiURLError(exception=err, provider=provider)
if response.ok:
return response.content
else:
if response.status_code in (401, 403):
raise IngestApiError.apiAuthError(
Exception(response.reason), provider)
elif response.status_code == 404:
raise IngestApiError.apiNotFoundError(
Exception(response.reason), provider)
else:
raise IngestApiError.apiGeneralError(
Exception(response.reason), provider)
def _request(self, url):
try:
response = requests.get(url, params={}, timeout=120)
except requests.exceptions.Timeout as ex:
# Maybe set up for a retry, or continue in a retry loop
raise IngestApiError.apiTimeoutError(ex, self.provider)
except requests.exceptions.TooManyRedirects as ex:
# Tell the user their URL was bad and try a different one
raise IngestApiError.apiRedirectError(ex, self.provider)
except requests.exceptions.RequestException as ex:
# catastrophic error. bail.
raise IngestApiError.apiRequestError(ex, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
if response.status_code == 404:
raise LookupError('Not found')
return response.content
def _request(self, url):
try:
response = requests.get(url, params={}, timeout=120)
except requests.exceptions.Timeout as ex:
# Maybe set up for a retry, or continue in a retry loop
raise IngestApiError.apiTimeoutError(ex, self.provider)
except requests.exceptions.TooManyRedirects as ex:
# Tell the user their URL was bad and try a different one
raise IngestApiError.apiRedirectError(ex, self.provider)
except requests.exceptions.RequestException as ex:
# catastrophic error. bail.
raise IngestApiError.apiRequestError(ex, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
if response.status_code == 404:
raise LookupError('Not found')
return response.content
def _update(self, provider, update):
user = provider['config']['wufoo_username']
wufoo_data = {
"url": WUFOO_URL.format(subdomain=user),
"user": user,
"api_key": provider['config']['wufoo_api_key'],
"form_query_entries_tpl": WUFOO_QUERY_FORM + WUFOO_QUERY_ENTRIES,
"update": update}
try:
parser = self.get_feed_parser(provider, None)
except requests.exceptions.Timeout as ex:
raise IngestApiError.apiTimeoutError(ex, provider)
except requests.exceptions.TooManyRedirects as ex:
raise IngestApiError.apiRedirectError(ex, provider)
except requests.exceptions.RequestException as ex:
raise IngestApiError.apiRequestError(ex, provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
items = parser.parse(wufoo_data, provider)
return [items]
def _update(self, provider, update):
user = provider['config']['wufoo_username']
wufoo_data = {
"url": WUFOO_URL.format(subdomain=user),
"user": user,
"api_key": provider['config']['wufoo_api_key'],
"form_query_entries_tpl": WUFOO_QUERY_FORM + WUFOO_QUERY_ENTRIES,
"update": update}
try:
parser = self.get_feed_parser(provider, None)
except requests.exceptions.Timeout as ex:
raise IngestApiError.apiTimeoutError(ex, provider)
except requests.exceptions.TooManyRedirects as ex:
raise IngestApiError.apiRedirectError(ex, provider)
except requests.exceptions.RequestException as ex:
raise IngestApiError.apiRequestError(ex, provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
items = parser.parse(wufoo_data, provider)
return [items]
def test_raise_apiNotFoundError(self):
with assert_raises(IngestApiError) as error_context:
ex = Exception("Testing apiNotFoundError")
raise IngestApiError.apiNotFoundError(ex, self.provider)
exception = error_context.exception
self.assertTrue(exception.code == 4006)
self.assertTrue(exception.message == "API service not found(404) error")
self.assertIsNotNone(exception.system_exception)
self.assertEqual(exception.system_exception.args[0], "Testing apiNotFoundError")
self.assertEqual(len(self.mock_logger_handler.messages['error']), 1)
self.assertEqual(self.mock_logger_handler.messages['error'][0],
"IngestApiError Error 4006 - API service not found(404) error: "
"Testing apiNotFoundError on channel TestProvider")
def _test(self, provider):
config = provider.get('config', {})
url = config['url']
api_key = config['api_key']
# limit the data to a single article and filter out all article fields
# to save bandwidth
params = {'limit': 1, 'fields': 'id'}
headers = {'apikey': api_key}
try:
response = requests.get(url, params=params, headers=headers, timeout=30)
except requests.exceptions.ConnectionError as err:
raise IngestApiError.apiConnectionError(exception=err)
if not response.ok:
if response.status_code == 404:
raise IngestApiError.apiNotFoundError(
Exception(response.reason), provider)
else:
raise IngestApiError.apiGeneralError(
Exception(response.reason), provider)
def test_raise_apiRedirectError(self):
with assert_raises(IngestApiError) as error_context:
ex = Exception("Testing apiRedirectError")
raise IngestApiError.apiRedirectError(ex, self.provider)
exception = error_context.exception
self.assertTrue(exception.code == 4002)
self.assertTrue(exception.message == "API ingest has too many redirects")
self.assertIsNotNone(exception.system_exception)
self.assertEqual(exception.system_exception.args[0], "Testing apiRedirectError")
self.assertEqual(len(self.mock_logger_handler.messages['error']), 1)
self.assertEqual(self.mock_logger_handler.messages['error'][0],
"IngestApiError Error 4002 - API ingest has too many redirects: "
"Testing apiRedirectError on channel TestProvider")
def _get_tree(self, endpoint, payload=None):
"""Get xml response for given API endpoint and payload.
:param: endpoint
:type endpoint: str
:param: payload
:type payload: str
"""
if payload is None:
payload = {}
payload['token'] = self._get_auth_token(self.provider, update=True)
url = self._get_absolute_url(endpoint)
if not self.session:
self.session = requests.Session()
retries = 0
while True:
try:
response = self.session.get(url, params=payload, timeout=(30, 15))
except requests.exceptions.Timeout as ex:
if retries < 3:
logger.warn('Reuters API timeout retrying, retries {}'.format(retries))
retries += 1
continue
raise IngestApiError.apiTimeoutError(ex, self.provider)
except requests.exceptions.TooManyRedirects as ex:
# Tell the user their URL was bad and try a different one
raise IngestApiError.apiRedirectError(ex, self.provider)
except requests.exceptions.RequestException as ex:
# catastrophic error. bail.
raise IngestApiError.apiRequestError(ex, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
if response.status_code == 404:
raise LookupError('Not found %s' % payload)
break
try:
return etree.fromstring(response.content) # workaround for http mock lib
except UnicodeEncodeError as error:
traceback.print_exc()
raise IngestApiError.apiUnicodeError(error, self.provider)
except ParseError as error:
traceback.print_exc()
raise IngestApiError.apiParseError(error, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
class NewsworthyFeedingService(FeedingService):
"""
Feeding Service class which can retrieve articles from Newsworthy web service
"""
NAME = 'newsworthy'
ERRORS = [IngestApiError.apiRequestError().get_error_description(),
SuperdeskIngestError.notConfiguredError().get_error_description()]
label = 'Newsworthy'
fields = [
{
'id': 'url', 'type': 'text', 'label': 'Use this URL for webhook',
'default_value': '',
'readonly': True,
},
{
'id': 'username', 'type': 'text', 'label': 'Username',
'required': True
},
{
'id': 'password', 'type': 'password', 'label': 'Password',
'required': True
},
{
'id': 'secret', 'type': 'password', 'label': 'Shared Secret',
'placeholder': 'Shared Secret', 'required': False
},
]
def _update(self, provider, update):
try:
data = provider['newsworthy_data']
except IndexError:
return [[]]
if data['hook']['event'] == EVENT_UNPUBLISHED:
logger.info("ignoring unpublish event on following data:\n{data}".format(data=data))
return [[]]
# we have to write to a temporary file because feed parser expect a file path
# FIXME: it would be better to use the data directly
with NamedTemporaryFile('w') as f:
json.dump(data['data'], f)
f.seek(0)
parser = self.get_feed_parser(provider, f.name)
items = parser.parse(f.name, provider)
return [items]
def test_raise_apiParseError(self):
with assert_raises(IngestApiError) as error_context:
ex = Exception("Testing apiParseError")
raise IngestApiError.apiParseError(ex, self.provider)
exception = error_context.exception
self.assertTrue(exception.code == 4005)
self.assertTrue(exception.message == "API ingest xml parse error")
self.assertIsNotNone(exception.system_exception)
self.assertEquals(exception.system_exception.args[0], "Testing apiParseError")
self.assertEqual(len(self.mock_logger_handler.messages["error"]), 1)
self.assertEqual(
self.mock_logger_handler.messages["error"][0],
"IngestApiError Error 4005 - API ingest xml parse error: " "Testing apiParseError on channel TestProvider",
)
def test_raise_apiRequestError(self):
with assert_raises(IngestApiError) as error_context:
ex = Exception("Testing apiRequestError")
raise IngestApiError.apiRequestError(ex, self.provider)
exception = error_context.exception
self.assertTrue(exception.code == 4003)
self.assertTrue(exception.message == "API ingest has request error")
self.assertIsNotNone(exception.system_exception)
self.assertEquals(exception.system_exception.args[0],
"Testing apiRequestError")
self.assertEqual(len(self.mock_logger_handler.messages['error']), 1)
self.assertEqual(
self.mock_logger_handler.messages['error'][0],
"IngestApiError Error 4003 - API ingest has request error: "
"Testing apiRequestError on channel TestProvider")
def test_raise_apiNotFoundError(self):
with assert_raises(IngestApiError) as error_context:
try:
ex = Exception("Testing apiNotFoundError")
raise ex
except Exception:
raise IngestApiError.apiNotFoundError(ex, self.provider)
exception = error_context.exception
self.assertTrue(exception.code == 4006)
self.assertTrue(exception.message == "API service not found(404) error")
self.assertIsNotNone(exception.system_exception)
self.assertEqual(exception.system_exception.args[0], "Testing apiNotFoundError")
self.assertEqual(len(self.mock_logger_handler.messages['error']), 1)
self.assertEqual(self.mock_logger_handler.messages['error'][0],
"IngestApiError Error 4006 - API service not found(404) error: "
"Testing apiNotFoundError on channel TestProvider")
def test_raise_apiTimeoutError(self):
with assert_raises(IngestApiError) as error_context:
try:
ex = Exception("Testing apiTimeoutError")
raise ex
except Exception:
raise IngestApiError.apiTimeoutError(ex, self.provider)
exception = error_context.exception
self.assertTrue(exception.code == 4001)
self.assertTrue(exception.message == "API ingest connection has timed out.")
self.assertIsNotNone(exception.system_exception)
self.assertEqual(exception.system_exception.args[0], "Testing apiTimeoutError")
self.assertEqual(len(self.mock_logger_handler.messages['error']), 1)
self.assertEqual(self.mock_logger_handler.messages['error'][0],
"IngestApiError Error 4001 - API ingest connection has timed out.: "
"Testing apiTimeoutError on channel TestProvider")
def test_raise_apiUnicodeError(self):
with assert_raises(IngestApiError) as error_context:
try:
ex = Exception("Testing apiUnicodeError")
raise ex
except Exception:
raise IngestApiError.apiUnicodeError(ex, self.provider)
exception = error_context.exception
self.assertTrue(exception.code == 4004)
self.assertTrue(exception.message == "API ingest Unicode Encode Error")
self.assertIsNotNone(exception.system_exception)
self.assertEqual(exception.system_exception.args[0], "Testing apiUnicodeError")
self.assertEqual(len(self.mock_logger_handler.messages['error']), 1)
self.assertEqual(self.mock_logger_handler.messages['error'][0],
"IngestApiError Error 4004 - API ingest Unicode Encode Error: "
"Testing apiUnicodeError on channel TestProvider")
def _update(self, provider, update):
try:
config = provider['config']
user = config['username']
password = config['password']
except KeyError:
SuperdeskIngestError.notConfiguredError(Exception('username and password are needed'))
url_override = config.get('url', '').strip()
if not url_override.startswith('http'):
SuperdeskIngestError.notConfiguredError(Exception('if URL is set, it must be a valid http link'))
if url_override:
params = {'user': user, 'password': password, 'maksAntal': 50}
else:
params = {'user': user, 'password': password, 'maksAntal': 50, 'waitAcknowledge': 'true'}
try:
r = requests.get(url_override or URL, params=params)
except Exception:
raise IngestApiError.apiRequestError(Exception('error while doing the request'))
try:
root_elt = etree.fromstring(r.text)
except Exception:
raise IngestApiError.apiRequestError(Exception('error while parsing the request answer'))
try:
if root_elt.xpath('(//error/text())[1]')[0] != '0':
err_msg = root_elt.xpath('(//errormsg/text())[1]')[0]
raise IngestApiError.apiRequestError(Exception('error code returned by API: {msg}'.format(msg=err_msg)))
except IndexError:
raise IngestApiError.apiRequestError(Exception('Invalid XML, <error> element not found'))
parser = self.get_feed_parser(provider)
items = []
for elt in root_elt.xpath('//RBNews'):
item = parser.parse(elt, provider)
items.append(item)
if not url_override:
try:
queue_id = elt.xpath('.//ServiceQueueId/text()')[0]
except IndexError:
raise IngestApiError.apiRequestError(Exception('missing ServiceQueueId element'))
ack_params = {'user': user, 'password': password, 'servicequeueid': queue_id}
try:
requests.get(URL_ACK, params=ack_params)
except Exception:
raise IngestApiError.apiRequestError(Exception('error while doing the request'))
return [items]
def test_raise_apiGeneralError(self):
with assert_raises(IngestApiError) as error_context:
ex = Exception("Testing general API error")
raise IngestApiError.apiGeneralError(ex, self.provider)
exception = error_context.exception
self.assertEqual(exception.code, 4000)
self.assertEqual(exception.message, "Unknown API ingest error")
self.assertEqual(exception.provider_name, "TestProvider")
self.assertIsNotNone(exception.system_exception)
self.assertEqual(exception.system_exception.args[0], "Testing general API error")
self.assertEqual(len(self.mock_logger_handler.messages["error"]), 1)
self.assertEqual(
self.mock_logger_handler.messages["error"][0],
"IngestApiError Error 4000 - Unknown API ingest error: "
"Testing general API error on channel TestProvider",
)
def test_raise_apiGeneralError(self):
with assert_raises(IngestApiError) as error_context:
ex = Exception("Testing general API error")
raise IngestApiError.apiGeneralError(ex, self.provider)
exception = error_context.exception
self.assertEqual(exception.code, 4000)
self.assertEqual(exception.message, "Unknown API ingest error")
self.assertEqual(exception.provider_name, "TestProvider")
self.assertIsNotNone(exception.system_exception)
self.assertEqual(exception.system_exception.args[0],
"Testing general API error")
self.assertEqual(len(self.mock_logger_handler.messages['error']), 1)
self.assertEqual(
self.mock_logger_handler.messages['error'][0],
"IngestApiError Error 4000 - Unknown API ingest error: "
"Testing general API error on channel TestProvider")
def test_raise_apiParseError(self):
with assert_raises(IngestApiError) as error_context:
try:
ex = Exception("Testing apiParseError")
raise ex
except Exception:
raise IngestApiError.apiParseError(ex, self.provider)
exception = error_context.exception
self.assertTrue(exception.code == 4005)
self.assertTrue(exception.message == "API ingest xml parse error")
self.assertIsNotNone(exception.system_exception)
self.assertEqual(exception.system_exception.args[0],
"Testing apiParseError")
self.assertEqual(len(self.mock_logger_handler.messages["error"]), 1)
self.assertEqual(
self.mock_logger_handler.messages["error"][0],
"IngestApiError Error 4005 - API ingest xml parse error: "
"Testing apiParseError on channel TestProvider",
)
def _get_tree(self, endpoint, payload=None):
"""
Get xml response for given API endpoint and payload.
:param: endpoint
:type endpoint: str
:param: payload
:type payload: str
"""
if payload is None:
payload = {}
payload['token'] = self._get_auth_token(self.provider, update=True)
url = self._get_absolute_url(endpoint)
try:
response = requests.get(url, params=payload, timeout=15)
except requests.exceptions.Timeout as ex:
# Maybe set up for a retry, or continue in a retry loop
raise IngestApiError.apiTimeoutError(ex, self.provider)
except requests.exceptions.TooManyRedirects as ex:
# Tell the user their URL was bad and try a different one
raise IngestApiError.apiRedirectError(ex, self.provider)
except requests.exceptions.RequestException as ex:
# catastrophic error. bail.
raise IngestApiError.apiRequestError(ex, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
if response.status_code == 404:
raise LookupError('Not found %s' % payload)
try:
return etree.fromstring(
response.content) # workaround for http mock lib
except UnicodeEncodeError as error:
traceback.print_exc()
raise IngestApiError.apiUnicodeError(error, self.provider)
except ParseError as error:
traceback.print_exc()
raise IngestApiError.apiParseError(error, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
def get_tree(self, endpoint, payload=None):
"""Get xml response for given API endpoint and payload."""
if payload is None:
payload = {}
payload['token'] = self.get_token()
url = self.get_url(endpoint)
try:
response = requests.get(url, params=payload, timeout=21.0)
except requests.exceptions.Timeout as ex:
# Maybe set up for a retry, or continue in a retry loop
raise IngestApiError.apiTimeoutError(ex, self.provider)
except requests.exceptions.TooManyRedirects as ex:
# Tell the user their URL was bad and try a different one
raise IngestApiError.apiRedirectError(ex, self.provider)
except requests.exceptions.RequestException as ex:
# catastrophic error. bail.
raise IngestApiError.apiRequestError(ex, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError(error, self.provider)
if response.status_code == 404:
raise IngestApiError.apiNotFoundError(
LookupError('Not found %s' % payload), self.provider)
try:
# workaround for httmock lib
# return etree.fromstring(response.text.encode('utf-8'))
return etree.fromstring(response.content)
except UnicodeEncodeError as error:
traceback.print_exc()
raise IngestApiError.apiUnicodeError(error, self.provider)
except ParseError as error:
traceback.print_exc()
raise IngestApiError.apiParseError(error, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError(error, self.provider)
def _get_tree(self, endpoint, payload=None):
"""
Get xml response for given API endpoint and payload.
:param: endpoint
:type endpoint: str
:param: payload
:type payload: str
"""
if payload is None:
payload = {}
payload['token'] = self._get_auth_token(self.provider, update=True)
url = self._get_absolute_url(endpoint)
try:
response = requests.get(url, params=payload, timeout=15)
except requests.exceptions.Timeout as ex:
# Maybe set up for a retry, or continue in a retry loop
raise IngestApiError.apiTimeoutError(ex, self.provider)
except requests.exceptions.TooManyRedirects as ex:
# Tell the user their URL was bad and try a different one
raise IngestApiError.apiRedirectError(ex, self.provider)
except requests.exceptions.RequestException as ex:
# catastrophic error. bail.
raise IngestApiError.apiRequestError(ex, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
if response.status_code == 404:
raise LookupError('Not found %s' % payload)
try:
return etree.fromstring(response.content) # workaround for http mock lib
except UnicodeEncodeError as error:
traceback.print_exc()
raise IngestApiError.apiUnicodeError(error, self.provider)
except ParseError as error:
traceback.print_exc()
raise IngestApiError.apiParseError(error, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
def get_url(self, url=None, **kwargs):
"""Do an HTTP Get on URL and validate response.
:param string url: url to use (None to use self.HTTP_URL)
:param **kwargs: extra parameter for requests
:return dict: response content data
"""
response = super().get_url(url=url, **kwargs)
content = response.json()
if content['hasError']:
msg = "Error in GET: '{}'. ErrorCode: '{}'. Description: '{}'".format(
url,
content['data']['errorCode'],
content['data']['description']
)
logger.error(msg)
raise IngestApiError.apiGeneralError(Exception(msg), self.provider)
return content['data']
def _fetch_data(self):
url = self.config['url']
api_key = self.config['api_key']
last_update = self.provider.get(
'last_updated', utcfromtimestamp(0)).strftime('%Y-%m-%dT%H:%M:%S')
# Results are pagified so we'll read this many at a time
offset_jump = 10
params = {'start': last_update, 'limit': offset_jump}
headers = {'apikey': api_key}
items = []
offset = 0
while True:
params['offset'] = offset
response = self.get_url(url, params=params, headers=headers)
# The total number of results are given to us in json, get them
# via a regex to read the field so we don't have to convert the
# whole thing to json pointlessly
item_ident = re.search('\"total\": *[0-9]*', response.text).group()
results_str = re.search('[0-9]+', item_ident).group()
if results_str is None:
raise IngestApiError.apiGeneralError(Exception(response.text),
self.provider)
num_results = int(results_str)
if num_results > 0:
items.append(response.text)
if offset >= num_results:
return items
offset += offset_jump
return items
def _update(self, provider, update):
# Each update run will retrieve the data for a single "market"
market_index = provider.get('private', {}).get('market_index', 0)
markets = json.loads(provider.get('config', {}).get('market_definitions', []).replace('\'', '"'))
market = markets[market_index]
logger.info('Retrieving fuel data for the {} market'.format(market.get('market')))
try:
self.session_token = self._get_token(provider).get('id')
prices = self._get_prices(provider, market)
self._save(prices, market)
except Exception as ex:
raise IngestApiError.apiGeneralError(ex, self.provider)
finally:
# Save the next market to process
market_index = (market_index + 1) % len(markets)
get_resource_service('ingest_providers').system_update(provider.get('_id'),
{'private': {'market_index': market_index}},
provider)
return None
class EventHTTPFeedingService(HTTPFeedingServiceBase):
"""
Feeding Service class which can read events using HTTP
"""
NAME = 'event_http'
label = 'Event HTTP feed'
service = 'events'
fields = [
{
'id': 'url', 'type': 'text', 'label': 'Feed URL',
'placeholder': 'Feed URL', 'required': True
}
]
ERRORS = [IngestApiError.apiTimeoutError().get_error_description(),
IngestApiError.apiRedirectError().get_error_description(),
IngestApiError.apiRequestError().get_error_description(),
IngestApiError.apiUnicodeError().get_error_description(),
IngestApiError.apiParseError().get_error_description(),
IngestApiError.apiGeneralError().get_error_description()]
HTTP_AUTH = False
def _update(self, provider, update):
"""
Fetch events from external API.
:param provider: Ingest Provider Details.
:type provider: dict
:param update: Any update that is required on provider.
:type update: dict
:return: a list of events which can be saved.
"""
response = self.get_url(self.config['url'])
parser = self.get_feed_parser(provider)
logger.info('Ingesting events with {} parser'.format(parser.__class__.__name__))
logger.info('Ingesting content: {} ...'.format(str(response.content)[:4000]))
if hasattr(parser, 'parse_http'):
items = parser.parse_http(response.content, provider)
else:
items = parser.parse(response.content)
if isinstance(items, list):
yield items
else:
yield [items]
def test_raise_apiAuthError(self):
with assert_raises(IngestApiError) as error_context:
ex = Exception("Testing API authorization error")
raise IngestApiError.apiAuthError(ex, self.provider)
exception = error_context.exception
self.assertEqual(exception.code, 4007)
self.assertEqual(
exception.message, "API authorization error")
self.assertEqual(exception.provider_name, "TestProvider")
self.assertIsNotNone(exception.system_exception)
self.assertEqual(
exception.system_exception.args[0],
"Testing API authorization error")
self.assertEqual(len(self.mock_logger_handler.messages['error']), 1)
self.assertEqual(
self.mock_logger_handler.messages['error'][0],
"IngestApiError Error 4007 - API authorization error: "
"Testing API authorization error on channel TestProvider")
def _update(self, provider, update):
def convert_date(epoch):
dt = local_to_utc(config.DEFAULT_TIMEZONE, datetime.fromtimestamp(int(str(epoch)[:10])))
return dt
username = provider.get('config', {}).get('username')
password = provider.get('config', {}).get('password')
url = provider.get('config', {}).get('api_url')
try:
response = requests.get(url, auth=(username, password))
response.raise_for_status()
except Exception as ex:
raise IngestApiError.apiGeneralError(ex, self.provider)
data = json.loads(response.content.decode('UTF-8'))
service = get_resource_service('traffic_incidents')
incidents = []
for feature in data.get('features', []):
props = feature.get('properties', {})
incident = {
'guid': int(props.get('id')),
'start_date': convert_date(props.get('startDate')),
'end_date': convert_date(props.get('endDate')),
'incident_type': props.get('type'),
'incident_description': props.get('description'),
'city': props.get('city'),
'state': props.get('state'),
'from_street_name': props.get('fromStreetName'),
'from_cross_street_name': props.get('fromCrossStreetName'),
'to_street_name': props.get('toStreetName'),
'to_cross_street_name': props.get('toCrossStreetName'),
'geometry': feature.get('geometry')
}
incident.get('geometry').pop('crs')
incidents.append(incident)
service.delete(lookup={})
service.post(incidents)
def test_raise_apiAuthError(self):
with assert_raises(IngestApiError) as error_context:
try:
ex = Exception("Testing API authorization error")
raise ex
except:
raise IngestApiError.apiAuthError(ex, self.provider)
exception = error_context.exception
self.assertEqual(exception.code, 4007)
self.assertEqual(exception.message, "API authorization error")
self.assertEqual(exception.provider_name, "TestProvider")
self.assertIsNotNone(exception.system_exception)
self.assertEqual(exception.system_exception.args[0],
"Testing API authorization error")
self.assertEqual(len(self.mock_logger_handler.messages['error']), 1)
self.assertEqual(
self.mock_logger_handler.messages['error'][0],
"IngestApiError Error 4007 - API authorization error: "
"Testing API authorization error on channel TestProvider")
def get_tree(self, endpoint, payload=None):
"""Get xml response for given API endpoint and payload."""
if payload is None:
payload = {}
payload['token'] = self.get_token()
url = self.get_url(endpoint)
try:
response = requests.get(url, params=payload, timeout=21.0)
except requests.exceptions.Timeout as ex:
# Maybe set up for a retry, or continue in a retry loop
raise IngestApiError.apiTimeoutError(ex, self.provider)
except requests.exceptions.TooManyRedirects as ex:
# Tell the user their URL was bad and try a different one
raise IngestApiError.apiRedirectError(ex, self.provider)
except requests.exceptions.RequestException as ex:
# catastrophic error. bail.
raise IngestApiError.apiRequestError(ex, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError(error, self.provider)
if response.status_code == 404:
raise IngestApiError.apiNotFoundError(LookupError('Not found %s' % payload), self.provider)
try:
# workaround for httmock lib
# return etree.fromstring(response.text.encode('utf-8'))
return etree.fromstring(response.content)
except UnicodeEncodeError as error:
traceback.print_exc()
raise IngestApiError.apiUnicodeError(error, self.provider)
except ParseError as error:
traceback.print_exc()
raise IngestApiError.apiParseError(error, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError(error, self.provider)
class ReutersHTTPFeedingService(HTTPFeedingService):
"""
Feeding Service class which can read article(s) using HTTP provided by Reuters.
"""
NAME = "reuters_http"
ERRORS = [
IngestApiError.apiTimeoutError().get_error_description(),
IngestApiError.apiRedirectError().get_error_description(),
IngestApiError.apiRequestError().get_error_description(),
IngestApiError.apiUnicodeError().get_error_description(),
IngestApiError.apiParseError().get_error_description(),
IngestApiError.apiGeneralError().get_error_description(),
]
DATE_FORMAT = "%Y.%m.%d.%H.%M"
label = "Reuters feed API"
fields = [
{
"id": "url",
"type": "text",
"label": "Feed URL",
"placeholder": "Feed URL",
"required": True,
"default": "http://rmb.reuters.com/rmd/rest/xml",
},
{
"id": "auth_url",
"type": "text",
"label": "URL for Authentication",
"placeholder": "authentication url",
"required": True,
"default": "https://commerce.reuters.com/rmd/rest/xml/login",
},
{"id": "username", "type": "text", "label": "Username", "placeholder": "Username", "required": True},
{"id": "password", "type": "password", "label": "Password", "placeholder": "Password", "required": True},
]
session = None
def _update(self, provider, update):
updated = utcnow()
last_updated = provider.get("last_updated")
ttl_minutes = app.config["INGEST_EXPIRY_MINUTES"]
if not last_updated or last_updated < updated - datetime.timedelta(minutes=ttl_minutes):
last_updated = updated - datetime.timedelta(minutes=ttl_minutes)
self.provider = provider
provider_config = provider.get("config")
if not provider_config:
provider_config = {}
provider["config"] = provider_config
provider_config.setdefault("url", "http://rmb.reuters.com/rmd/rest/xml")
provider_config.setdefault("auth_url", "https://commerce.reuters.com/rmd/rest/xml/login")
self.URL = provider_config.get("url")
for channel in self._get_channels():
ids = self._get_article_ids(channel, last_updated, updated)
for id in ids:
try:
items = self.fetch_ingest(id)
if items:
yield items
# if there was an exception processing the one of the bunch log it and continue
except Exception as ex:
logger.warn("Reuters item {} has not been retrieved".format(id))
logger.exception(ex)
def _get_channels(self):
"""Get subscribed channels."""
channels = []
tree = self._get_tree("channels")
for channel in tree.findall("channelInformation"):
channels.append(channel.find("alias").text)
return channels
def _get_tree(self, endpoint, payload=None):
"""Get xml response for given API endpoint and payload.
:param: endpoint
:type endpoint: str
:param: payload
:type payload: str
"""
if payload is None:
payload = {}
payload["token"] = self._get_auth_token(self.provider, update=True)
url = self._get_absolute_url(endpoint)
if not self.session:
self.session = requests.Session()
retries = 0
while True:
try:
response = self.session.get(url, params=payload, timeout=(30, 15))
except requests.exceptions.Timeout as ex:
if retries < 3:
logger.warn("Reuters API timeout retrying, retries {}".format(retries))
retries += 1
continue
raise IngestApiError.apiTimeoutError(ex, self.provider)
except requests.exceptions.TooManyRedirects as ex:
# Tell the user their URL was bad and try a different one
raise IngestApiError.apiRedirectError(ex, self.provider)
except requests.exceptions.RequestException as ex:
# catastrophic error. bail.
raise IngestApiError.apiRequestError(ex, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
if response.status_code == 404:
raise LookupError(_("Not found {payload}").format(payload=payload))
break
try:
return etree.fromstring(response.content) # workaround for http mock lib
except UnicodeEncodeError as error:
traceback.print_exc()
raise IngestApiError.apiUnicodeError(error, self.provider)
except ParseError as error:
traceback.print_exc()
raise IngestApiError.apiParseError(error, self.provider)
except Exception as error:
traceback.print_exc()
raise IngestApiError.apiGeneralError(error, self.provider)
def _get_absolute_url(self, endpoint):
"""
Get absolute URL for given endpoint.
:param: endpoint
:type endpoint: str
"""
return "/".join([self.URL, endpoint])
def _get_article_ids(self, channel, last_updated, updated):
"""
Get article ids which should be upserted also save the poll token that is returned.
"""
ids = set()
payload = {"channel": channel, "fieldsRef": "id"}
# check if the channel has a pollToken if not fall back to dateRange
last_poll_token = self._get_poll_token(channel)
if last_poll_token is not None:
logger.info("Reuters requesting channel {} with poll token {}".format(channel, last_poll_token))
payload["pollToken"] = last_poll_token
else:
payload["dateRange"] = "%s-%s" % (self._format_date(last_updated), self._format_date(updated))
logger.info("Reuters requesting channel {} with dateRange {}".format(channel, payload["dateRange"]))
tree = self._get_tree("items", payload)
status_code = tree.find("status").get("code") if tree.tag == "results" else tree.get("code")
# check the returned status
if status_code != "10":
logger.warn("Reuters channel request returned status code {}".format(status_code))
# status code 30 indicates failure
if status_code == "30":
# invalid token
logger.warn(
"Reuters error on channel {} code {} {}".format(
channel, tree.find("error").get("code"), tree.find("error").text
)
)
if tree.find("error").get("code") == "2100":
self._save_poll_token(channel, None)
logger.warn("Reuters channel invalid token reseting {}".format(status_code))
return ids
# extract the returned poll token if there is one
poll_token = tree.find("pollToken")
if poll_token is not None:
# a new token indicated new content
if poll_token.text != last_poll_token:
logger.info("Reuters channel {} new token {}".format(channel, poll_token.text))
self._save_poll_token(channel, poll_token.text)
else:
# the token has not changed, so nothing new
logger.info("Reuters channel {} nothing new".format(channel))
return ids
else:
logger.info("Reuters channel {} retrieved no token".format(channel))
return ids
for result in tree.findall("result"):
id = result.find("id").text
ids.add(id)
logger.info("Reuters id : {}".format(id))
return ids
def _save_poll_token(self, channel, poll_token):
"""Saves the poll token for the passed channel in the config section of the
:param channel:
:param poll_token:
:return:
"""
# get the provider in case it has been updated by another channel
ingest_provider_service = superdesk.get_resource_service("ingest_providers")
provider = ingest_provider_service.find_one(req=None, _id=self.provider[superdesk.config.ID_FIELD])
provider_token = provider.get("tokens")
if "poll_tokens" not in provider_token:
provider_token["poll_tokens"] = {channel: poll_token}
else:
provider_token["poll_tokens"][channel] = poll_token
upd_provider = {"tokens": provider_token}
ingest_provider_service.system_update(self.provider[superdesk.config.ID_FIELD], upd_provider, self.provider)
def _get_poll_token(self, channel):
"""Get the poll token from provider config if it is available.
:param channel:
:return: token
"""
if "tokens" in self.provider and "poll_tokens" in self.provider["tokens"]:
return self.provider.get("tokens").get("poll_tokens").get(channel, None)
def _format_date(self, date):
return date.strftime(self.DATE_FORMAT)
def fetch_ingest(self, id):
items = self._parse_items(id)
result_items = []
while items:
item = items.pop()
self.localize_timestamps(item)
try:
items.extend(self._fetch_items_in_package(item))
result_items.append(item)
except LookupError as err:
self.log_item_error(err, item, self.provider)
return []
return result_items
def _parse_items(self, id):
"""
Parse item message and return given items.
"""
payload = {"id": id}
tree = self._get_tree("item", payload)
parser = self.get_feed_parser(self.provider, tree)
items = parser.parse(tree, self.provider)
return items
def _fetch_items_in_package(self, item):
"""
Fetch remote assets for given item.
"""
items = []
for group in item.get("groups", []):
for ref in group.get("refs", []):
if "residRef" in ref:
items.extend(self._parse_items(ref.get("residRef")))
return items
def prepare_href(self, href, mimetype=None):
(scheme, netloc, path, params, query, fragment) = urlparse(href)
new_href = urlunparse((scheme, netloc, path, "", "", ""))
return "%s?auth_token=%s" % (new_href, self._get_auth_token(self.provider, update=True))
import requests
from superdesk.io.ingest_service import IngestService
from superdesk.utc import utcnow
from superdesk.etree import etree, ParseError
from superdesk.io import register_provider
from .newsml_2_0 import NewsMLTwoParser
from .reuters_token import get_token
from superdesk.errors import IngestApiError
from flask import current_app as app
PROVIDER = 'reuters'
errors = [IngestApiError.apiTimeoutError().get_error_description(),
IngestApiError.apiRedirectError().get_error_description(),
IngestApiError.apiRequestError().get_error_description(),
IngestApiError.apiUnicodeError().get_error_description(),
IngestApiError.apiParseError().get_error_description(),
IngestApiError.apiGeneralError().get_error_description()]
class ReutersIngestService(IngestService):
"""Reuters ingest service."""
DATE_FORMAT = '%Y.%m.%d.%H.%M'
URL = 'http://rmb.reuters.com/rmd/rest/xml'
token = None
def __init__(self):
from datetime import datetime
from superdesk.errors import IngestApiError, ParserError
from superdesk.io import register_provider
from superdesk.io.ingest_service import IngestService
from superdesk.utils import merge_dicts
from urllib.parse import quote as urlquote, urlsplit, urlunsplit
PROVIDER = "rss"
utcfromtimestamp = datetime.utcfromtimestamp
errors = [
IngestApiError.apiAuthError().get_error_description(),
IngestApiError.apiNotFoundError().get_error_description(),
IngestApiError.apiGeneralError().get_error_description(),
ParserError.parseMessageError().get_error_description(),
]
class RssIngestService(IngestService):
"""Ingest service for providing feeds received in RSS 2.0 format.
(NOTE: it should also work with other syndicated feeds formats, too, since
the underlying parser supports them, but for our needs RSS 2.0 is assumed)
"""
ItemField = namedtuple("ItemField", ["name", "name_in_data", "type"])