def validate_config(self):
"""
Validate provider config according to `cls.fields`
:param config: Ingest provider configuration
:type config: dict
:return:
"""
# validate required config fields
required_keys = [
field['id'] for field in self.fields
if field.get('required', False)
]
if not set(self.config.keys()).issuperset(required_keys):
raise SuperdeskIngestError.notConfiguredError(
Exception('{} are required.'.format(', '.join(required_keys))))
url = self.config.get('url').strip()
if not url:
try:
url_field = next({f for f in self.fields if f['id'] == u'url'})
except StopIteration:
url_required = False
else:
url_required = url_field.get('required', False)
if url_required:
raise SuperdeskIngestError.notConfiguredError(
Exception('URL is a required field.'))
else:
# validate url
if not url.startswith('http'):
raise SuperdeskIngestError.notConfiguredError(
Exception('URL must be a valid HTTP link.'))
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 _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 _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 _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 _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 validate_config(self):
"""
Validate provider config according to `cls.fields`
:param config: Ingest provider configuration
:type config: dict
:return:
"""
# validate required config fields
required_keys = [field['id'] for field in self.fields if field.get('required', False)]
if not set(self.config.keys()).issuperset(required_keys):
raise SuperdeskIngestError.notConfiguredError(
Exception('{} are required.'.format(', '.join(required_keys)))
)
# validate url
url = self.config.get('url')
if url and not url.strip().startswith('http'):
raise SuperdeskIngestError.notConfiguredError(
Exception('URL must be a valid HTTP link.')
)
def validate_config(self):
"""
Validate provider config according to `cls.fields`
:param config: Ingest provider configuration
:type config: dict
:return:
"""
# validate required config fields
required_keys = [
field["id"] for field in self.fields
if field.get("required", False)
]
if not set(self.config.keys()).issuperset(required_keys):
raise SuperdeskIngestError.notConfiguredError(
Exception("{} are required.".format(", ".join(required_keys))))
# validate url
url = self.config.get("url")
if url and not url.strip().startswith("http"):
raise SuperdeskIngestError.notConfiguredError(
Exception("URL must be a valid HTTP link."))
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 parse(self, data, provider=None):
if self.subjects_map is None:
self._set_metadata()
try:
stage_map = config.NIFS_STAGE_MAP
qcode_map = config.NIFS_QCODE_MAP
except KeyError:
raise SuperdeskIngestError.notConfiguredError(
Exception('NIFS maps are not found in settings'))
events = json.loads(data.decode('utf-8', 'ignore'))
items = []
try:
for event in events:
stage = stage_map.get(event['stageId'], '')
# we retrieve qcode from sportId, according to config, and sport name
try:
qcode = qcode_map[event['sportId']]
except KeyError:
logger.warning(
'no qcode registered for sportId {sport_id}'.format(
sport_id=event['sportId']))
qcode = ''
sport = ''
else:
sport = self.get_sport(qcode)
# name as requested by NTB
if stage or sport:
tpl_name = '{sport} {stage}, {rnd}. runde, {home} - {away}'
else:
tpl_name = '{rnd}. runde, {home} - {away}'
name = tpl_name.format(stage=stage,
sport=sport,
rnd=event['round'],
home=event['homeTeam']['name'],
away=event['awayTeam']['name']).strip()
event_start = dateutil.parser.parse(event['timestamp'])
# there is no end time specified in event
event_end = event_start + timedelta(hours=2)
# we have a common category and subject + a subject per sport
# cf. SDNTB-496
subject = [{'qcode': CAT, 'name': CAT, 'scheme': 'category'}]
subject.append({
'qcode':
MAIN_SUBJ_QCODE,
'name':
self.subjects_map.get(MAIN_SUBJ_QCODE, ''),
'scheme':
'subject_custom'
})
subject.append({
'qcode': qcode,
'name': sport,
'scheme': 'subject_custom'
})
service = {'qcode': SERVICE_QCODE, 'name': self.service_name}
item = {
'guid': event['uid'],
ITEM_TYPE: CONTENT_TYPE.EVENT,
'dates': {
'start': event_start,
'end': event_end,
'tz': ''
},
'name': name,
'slugline': sport,
'subject': subject,
'anpa_category': [service],
'calendars': [self.calendar_item],
'firstcreated': utcnow(),
'versioncreated': utcnow()
}
items.append(item)
return items
except Exception as ex:
raise ParserError.parseMessageError(ex, provider)
def _update(self, provider, update):
try:
config = provider["config"]
id_list = config["idList"]
# before "products" was hardcoded as value for "idListType"
id_list_type = config.get("idListType", "products")
if not id_list.strip():
raise KeyError
except KeyError:
raise SuperdeskIngestError.notConfiguredError(Exception("idList is needed"))
# we check if the provider has been closed since the last update
try:
last_closed = provider["last_closed"]["closed_at"]
last_updated = provider["last_updated"]
except KeyError:
pass
else:
if last_closed > last_updated and "private" in provider:
# we reset the private data so only last page of items will be retrieved (cf. SDESK-4372)
logger.info("reseting private data for provider {source}".format(source=provider.get("source")))
del provider["private"]
# 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",
}
try:
min_date_time = provider["private"]["min_date_time"]
sequence_number = provider["private"]["sequence_number"]
except KeyError:
# the provider is new or re-opened, we want last items
# so we need reverse-chronological order
chronological = False
else:
params["minDateTime"] = min_date_time
params["sequenceNumber"] = sequence_number
params["sortOrder"] = "chronological"
chronological = True
r = self.get_url(params=params)
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)
if not chronological:
items.reverse()
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]
class NTBEventsApiFeedingService(HTTPFeedingServiceBase):
"""
Feeding Service class which can read events from NTB API using HTTP
"""
NAME = 'ntb_events_api'
ERRORS = [
SuperdeskIngestError.notConfiguredError().get_error_description(),
IngestApiError.apiTimeoutError().get_error_description(),
IngestApiError.apiConnectionError().get_error_description(),
]
REQUESTS_PER_UPDATE = 4
EVENTS_PER_REQUEST = 25
HTTP_TIMEOUT = 20
label = 'NTB Events API'
fields = [
{
'id': 'url', 'type': 'text', 'label': 'Feed URL',
'placeholder': 'Feed URL', 'required': True
}] + HTTPFeedingServiceBase.AUTH_FIELDS
service = 'events'
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.
"""
all_items = OrderedDict()
self._provider = provider
provider_private = self._provider.get('private', {})
offset = provider_private.get('search', {}).get('offset', 0)
for _ in range(self.REQUESTS_PER_UPDATE):
response = self._send_request(offset + len(all_items))
xml = etree.fromstring(response.content)
items = self._parse_events(xml=xml)
if items:
all_items.update(items)
else:
break
if all_items:
update['private'] = {
'search': {
'offset': offset + len(all_items)
}
}
all_items = self._filter_items(all_items)
else:
update['is_closed'] = True
update['last_closed'] = {
'closed_at': utcnow(),
'message': 'Ingesting was finished.'
}
return [all_items]
def _send_request(self, offset):
"""
Execute http request to external API
:param offset: offset provided in request payload
:type offset: int
:return: http response
:raises IngestApiError.apiTimeoutError
:raises IngestApiError.apiConnectionError
:raises IngestApiError.apiRequestError
:raises IngestApiError.apiGeneralError
:raises IngestApiError.apiAuthError
:raises IngestApiError.apiNotFoundError
"""
payload = {
'search.offset': offset,
'search.showNumResults': self.EVENTS_PER_REQUEST
}
url = self._provider['config']['url'].strip()
return self.get_url(url, params=payload)
def _parse_events(self, xml):
"""
Parse xml document and returns list of events
:param xml: xml document
:type xml: lxml.etree._Element
:return: a list of events
"""
parser = self.get_feed_parser(self._provider, article=xml)
return OrderedDict(
(item['ntb_id'], item) for item in parser.parse(xml)
)
def _filter_items(self, items):
"""
Remove events which are exist in the db.
:param items: dict with events, ntbId used as a key
:type items: dict
:return: a list of events
"""
req = ParsedRequest()
req.projection = json.dumps({'ntb_id': 1, 'guid': 1, ITEM_STATE: 1})
req.max_results = len(items)
existing_items = superdesk.get_resource_service('events').get_from_mongo(
req,
{
'ntb_id': {
'$in': [ntb_id for ntb_id in items.keys()]
}
}
)
for existing_item in existing_items:
if existing_item.get(ITEM_STATE) == WORKFLOW_STATE.INGESTED:
# update event
items[existing_item['ntb_id']][GUID_FIELD] = existing_item[GUID_FIELD]
else:
# remove event when it has a state different from 'ingested'
del items[existing_item['ntb_id']]
return [items[i] for i in items.keys()]
class APFeedingService(FeedingService):
"""
Feeding Service class which can retrieve articles from Associated Press web service
"""
NAME = 'ap'
ERRORS = [
IngestApiError.apiRequestError().get_error_description(),
SuperdeskIngestError.notConfiguredError().get_error_description()
]
label = 'AP feed API'
fields = [{
'id': 'username',
'type': 'text',
'label': 'Username',
'placeholder': 'Username',
'required': True
}, {
'id': 'password',
'type': 'password',
'label': 'Password',
'placeholder': 'Password',
'required': True
}, {
'id': 'idList',
'type': 'text',
'label': 'Id List',
'placeholder': 'use coma separated ids for multiple values',
'required': False
}]
def config_test(self, provider=None):
super().config_test(provider)
def _update(self, provider, update):
try:
config = provider['config']
user = config['username']
password = config['password']
id_list = config['idList']
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': '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:
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 get_url(self, url=None, **kwargs):
"""Do an HTTP Get on URL
:param string url: url to use (None to use self.HTTP_URL)
:param **kwargs: extra parameter for requests
:return requests.Response: response
"""
if not url:
url = self.HTTP_URL
config = self.config
user = config.get('username')
password = config.get('password')
if user:
user = user.strip()
if password:
password = password.strip()
auth_required = config.get('auth_required', self.HTTP_AUTH)
if auth_required is None:
# auth_required may not be user in the feeding service
# in this case with use authentification only if user
# and password are set.
auth_required = bool(user and password)
if auth_required:
if not user:
raise SuperdeskIngestError.notConfiguredError(
"user is not configured")
if not password:
raise SuperdeskIngestError.notConfiguredError(
"password is not configured")
kwargs.setdefault('auth', (user, password))
params = kwargs.pop("params", {})
if params or self.HTTP_DEFAULT_PARAMETERS:
# if we have default parameters, we want them to be overriden
# by conflicting params given in arguments
if self.HTTP_DEFAULT_PARAMETERS:
params.update(self.HTTP_DEFAULT_PARAMETERS)
kwargs["params"] = params
try:
response = requests.get(url, timeout=self.HTTP_TIMEOUT, **kwargs)
except requests.exceptions.Timeout as exception:
raise IngestApiError.apiTimeoutError(exception, self.provider)
except requests.exceptions.ConnectionError as exception:
raise IngestApiError.apiConnectionError(exception, self.provider)
except requests.exceptions.RequestException as exception:
raise IngestApiError.apiRequestError(exception, self.provider)
except Exception as exception:
traceback.print_exc()
raise IngestApiError.apiGeneralError(exception, self.provider)
if not response.ok:
exception = Exception(response.reason)
if response.status_code in (401, 403):
raise IngestApiError.apiAuthError(exception, self.provider)
elif response.status_code == 404:
raise IngestApiError.apiNotFoundError(exception, self.provider)
else:
raise IngestApiError.apiGeneralError(exception, self.provider)
return response
class RitzauFeedingService(HTTPFeedingServiceBase):
"""
Feeding Service class which can retrieve articles from Ritzau web service
"""
NAME = "ritzau"
ERRORS = [
IngestApiError.apiRequestError().get_error_description(),
SuperdeskIngestError.notConfiguredError().get_error_description(),
]
label = "Ritzau feed API"
fields = HTTPFeedingServiceBase.AUTH_FIELDS + [{
"id": "url",
"type": "text",
"label": "URL",
"placeholder": "fill this field only for advanced uses",
"required": False,
}]
HTTP_URL = "https://services.ritzau.dk/ritzaurest/Services.svc/xml/news/NewsQueue"
# auth is done with params
HTTP_AUTH = False
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]
class RitzauFeedingService(HTTPFeedingServiceBase):
"""
Feeding Service class which can retrieve articles from Ritzau web service
"""
NAME = 'ritzau'
ERRORS = [IngestApiError.apiRequestError().get_error_description(),
SuperdeskIngestError.notConfiguredError().get_error_description()]
label = 'Ritzau feed API'
fields = HTTPFeedingServiceBase.AUTH_FIELDS + [
{
'id': 'url', 'type': 'text', 'label': 'URL',
'placeholder': 'fill this field only for advanced uses', 'required': False
}
]
HTTP_URL = 'https://services.ritzau.dk/ritzaurest/Services.svc/xml/news/NewsQueue'
# auth is done with params
HTTP_AUTH = False
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]
class HTTPFeedingServiceBase(FeedingService):
"""
Base class for feeding services using HTTP.
This class contains helpers to make the creation of HTTP based feeding services
easier.
There are a couple of class attributes you can use:
======================= ===========
Attribute Explanation
======================= ===========
HTTP_URL Main URL of your service, will be used by default in get_url
HTTP_TIMEOUT Timeout of requests in seconds
HTTP_DEFAULT_PARAMETERS Parameters used in every ``get`` requests.
Will be updated with params set in arguments
HTTP_AUTH Indicate if HTTP authentication is needed for your service.
If None, the authentication will be determined by the existence
of user and password. Will be overriden by auth_required config
if it exists.
======================= ===========
In addition, you have some pre-filled fields:
=============== ===========
Field Explanation
=============== ===========
AUTH_FIELDS username and password fields
AUTH_REQ_FIELDS username and password fields + auth_required field to indicate
if they are needed
=============== ===========
When ingest is updated, the provider is automatically saved to ``self.provider``.
``config`` property allows to access easily the user configuration.
``auth_info`` property returns a dictionary with ``username`` and ``password``
``get_url`` method do a HTTP Get request. url can be ommited in which case HTTP_URL will be used.
Authentication parameters are set automatically, and errors are catched appropriately.
Extra arguments are used directly in *requests* call.
"""
ERRORS = [
IngestApiError.apiTimeoutError().get_error_description(),
IngestApiError.apiRequestError().get_error_description(),
IngestApiError.apiGeneralError().get_error_description(),
SuperdeskIngestError.notConfiguredError().get_error_description()
]
# override this parameter with the main URL to use
HTTP_URL = None
# timeout in seconds
HTTP_TIMEOUT = 30
# if some parameters are used in every request, put them here
HTTP_DEFAULT_PARAMETERS = None
# Set to True if authentication is mandatory, False if there is no authentication
# and None to add authentication if user and password are defined.
# If auth_required is defined in config fields, it will override this value.
HTTP_AUTH = True
# use this when auth is always required
AUTH_FIELDS = [{
'id': 'username',
'type': 'text',
'label': 'Username',
'placeholder': 'Username',
'required': True
}, {
'id': 'password',
'type': 'password',
'label': 'Password',
'placeholder': 'Password',
'required': True
}]
# use this when auth depends of a "auth_required" flag (set by user)
AUTH_REQ_FIELDS = [{
'id': 'auth_required',
'type': 'boolean',
'label': 'Requires Authentication',
'placeholder': 'Requires Authentication',
'required': False
}, {
'id': 'username',
'type': 'text',
'label': 'Username',
'placeholder': 'Username',
'required_expression': '{auth_required}',
'show_expression': '{auth_required}'
}, {
'id': 'password',
'type': 'password',
'label': 'Password',
'placeholder': 'Password',
'required_expression': '{auth_required}',
'show_expression': '{auth_required}'
}]
def __init__(self):
self.token = None
@property
def auth_info(self):
"""Helper method to retrieve a dict with username and password when set"""
username = self.config.get('username', '')
password = self.config.get('password', '')
if not username or not password:
return None
return {'username': username, 'password': password}
@property
def config(self):
return self.provider.setdefault('config', {})
def validate_config(self):
"""
Validate provider config according to `cls.fields`
:param config: Ingest provider configuration
:type config: dict
:return:
"""
# validate required config fields
required_keys = [
field['id'] for field in self.fields
if field.get('required', False)
]
if not set(self.config.keys()).issuperset(required_keys):
raise SuperdeskIngestError.notConfiguredError(
Exception('{} are required.'.format(', '.join(required_keys))))
url = self.config.get('url').strip()
if not url:
try:
url_field = next({f for f in self.fields if f['id'] == u'url'})
except StopIteration:
url_required = False
else:
url_required = url_field.get('required', False)
if url_required:
raise SuperdeskIngestError.notConfiguredError(
Exception('URL is a required field.'))
else:
# validate url
if not url.startswith('http'):
raise SuperdeskIngestError.notConfiguredError(
Exception('URL must be a valid HTTP link.'))
def get_url(self, url=None, **kwargs):
"""Do an HTTP Get on URL
:param string url: url to use (None to use self.HTTP_URL)
:param **kwargs: extra parameter for requests
:return requests.Response: response
"""
if not url:
url = self.HTTP_URL
config = self.config
user = config.get('username')
password = config.get('password')
if user:
user = user.strip()
if password:
password = password.strip()
auth_required = config.get('auth_required', self.HTTP_AUTH)
if auth_required is None:
# auth_required may not be user in the feeding service
# in this case with use authentification only if user
# and password are set.
auth_required = bool(user and password)
if auth_required:
if not user:
raise SuperdeskIngestError.notConfiguredError(
"user is not configured")
if not password:
raise SuperdeskIngestError.notConfiguredError(
"password is not configured")
kwargs.setdefault('auth', (user, password))
params = kwargs.pop("params", {})
if params or self.HTTP_DEFAULT_PARAMETERS:
# if we have default parameters, we want them to be overriden
# by conflicting params given in arguments
if self.HTTP_DEFAULT_PARAMETERS:
params.update(self.HTTP_DEFAULT_PARAMETERS)
kwargs["params"] = params
try:
response = requests.get(url, timeout=self.HTTP_TIMEOUT, **kwargs)
except requests.exceptions.Timeout as exception:
raise IngestApiError.apiTimeoutError(exception, self.provider)
except requests.exceptions.ConnectionError as exception:
raise IngestApiError.apiConnectionError(exception, self.provider)
except requests.exceptions.RequestException as exception:
raise IngestApiError.apiRequestError(exception, self.provider)
except Exception as exception:
traceback.print_exc()
raise IngestApiError.apiGeneralError(exception, self.provider)
if not response.ok:
exception = Exception(response.reason)
if response.status_code in (401, 403):
raise IngestApiError.apiAuthError(exception, self.provider)
elif response.status_code == 404:
raise IngestApiError.apiNotFoundError(exception, self.provider)
else:
raise IngestApiError.apiGeneralError(exception, self.provider)
return response
def update(self, provider, update):
self.provider = provider
self.validate_config()
return super().update(provider, update)