def prepare(self):
try:
self._session = self._mendeley.start_client_credentials_flow(
).authenticate()
self._initialized = True
except Exception as e:
log.critical(e)
def get_profile_by_id(self, profile_id: str) -> Profile:
if not self._initialized:
log.critical("get_profile_by_id has been fired but the SDKCrawler was not initialized")
return []
profile = self._session.profiles.get(profile_id)
return get_profile_from_json(profile.json)
def get_profile_by_id(self, profile_id: str) -> Profile:
if not self._initialized:
log.critical(
"get_profile_by_id has been fired but the SDKCrawler was not initialized"
)
return []
profile = self._session.profiles.get(profile_id)
return get_profile_from_json(profile.json)
def get_documents_by_group_id(self, group_id: str) -> [Document]:
if not self._initialized:
log.critical("get_documents_by_group_id has been fired but the SDKCrawler was not initialized")
return []
results = []
documents = self._session.group_documents(group_id).iter(view='all')
for document in documents:
d = get_document_from_json(document.json)
results.append(d)
return results
def get_group_members(self, group_id: str) -> [Member]:
if not self._initialized:
log.critical("get_group_members has been fired but the SDKCrawler was not initialized")
return []
results = []
members = self._session.group_members(group_id).iter()
for member in members:
m = get_member_from_json(member.member_json)
if m.role != 'follower':
results.append(m)
return results
def get_documents_by_group_id(self, group_id: str) -> [Document]:
if not self._initialized:
log.critical(
"get_documents_by_group_id has been fired but the SDKCrawler was not initialized"
)
return []
results = []
documents = self._session.group_documents(group_id).iter(view='all')
for document in documents:
d = get_document_from_json(document.json)
results.append(d)
return results
def get_group_members(self, group_id: str) -> [Member]:
if not self._initialized:
log.critical(
"get_group_members has been fired but the SDKCrawler was not initialized"
)
return []
results = []
members = self._session.group_members(group_id).iter()
for member in members:
m = get_member_from_json(member.member_json)
if m.role != 'follower':
results.append(m)
return results
def get_documents_by_profile_id(self, profile_id: str) -> [Document]:
if not self._initialized:
log.critical("get_documents_by_profile_id has been fired but the SDKCrawler was not initialized")
return []
results = []
"""
Unfortunately the official Mendeley SDK has no support for document queries by non-logged-in profile-ids
Therefore i'll hack around that and reuse the session object to authenticate my own call.
Critical SDK class:
https://github.com/Mendeley/mendeley-python-sdk/blob/master/mendeley/resources/documents.py
"""
documents = ExtendedDocuments(self._session).iter(view='all', profile_id=profile_id, authored='true')
for document in documents:
d = get_document_from_json(document.json)
results.append(d)
return results
def get_documents_by_profile_id(self, profile_id: str) -> [Document]:
if not self._initialized:
log.critical(
"get_documents_by_profile_id has been fired but the SDKCrawler was not initialized"
)
return []
results = []
"""
Unfortunately the official Mendeley SDK has no support for document queries by non-logged-in profile-ids
Therefore i'll hack around that and reuse the session object to authenticate my own call.
Critical SDK class:
https://github.com/Mendeley/mendeley-python-sdk/blob/master/mendeley/resources/documents.py
"""
documents = ExtendedDocuments(self._session).iter(
view='all', profile_id=profile_id, authored='true')
for document in documents:
d = get_document_from_json(document.json)
results.append(d)
return results
configuration.load()
log.info("Configuration has been loaded")
# Create data controller and assert schema
# That will remove the race conditions of the gunicorn worker if it's done on every startup
data_controller = DataController(configuration.database)
data_controller.assert_schema()
# Pipeline runner
elif command == "pipeline":
config = ServiceConfiguration()
config.load()
data_controller = DataController(config.database)
if not data_controller.is_initialized():
log.critical("Database is not initialized")
exit()
crawler = None
if not config.uses_mendeley:
log.info("Pipeline uses FileCrawler")
crawler = FileCrawler()
else:
from mendeleycache.crawler.sdk_crawler import SDKCrawler
log.info("Pipeline uses SDKCrawler".format(
app_id=config.crawler.app_id,
app_secret=config.crawler.app_secret))
crawler = SDKCrawler(app_id=config.crawler.app_id,
app_secret=config.crawler.app_secret)
crawl_controller = CrawlController(crawler,
configuration.load()
log.info("Configuration has been loaded")
# Create data controller and assert schema
# That will remove the race conditions of the gunicorn worker if it's done on every startup
data_controller = DataController(configuration.database)
data_controller.assert_schema()
# Pipeline runner
elif command == "pipeline":
config = ServiceConfiguration()
config.load()
data_controller = DataController(config.database)
if not data_controller.is_initialized():
log.critical("Database is not initialized")
exit()
crawler = None
if not config.uses_mendeley:
log.info("Pipeline uses FileCrawler")
crawler = FileCrawler()
else:
from mendeleycache.crawler.sdk_crawler import SDKCrawler
log.info("Pipeline uses SDKCrawler".format(
app_id=config.crawler.app_id,
app_secret=config.crawler.app_secret
))
crawler = SDKCrawler(
app_id=config.crawler.app_id,
app_secret=config.crawler.app_secret
def prepare(self):
try:
self._session = self._mendeley.start_client_credentials_flow().authenticate()
self._initialized = True
except Exception as e:
log.critical(e)
