def clear_cache(self):
"""
Clear the cache database.
"""
msg = 'Cached Data has been cleared'
requests_cache.get_cache().clear()
xbmcgui.Dialog().notification(_plugin, msg, _icon, 3000, False)
def cache_initiated():
"""Hackish function to test if there is an existing requests_cache"""
try:
requests_cache.get_cache()
return True
except AttributeError:
return False
def uninstall_cache():
"""Deactives caches"""
try:
requests_cache.get_cache()
except AttributeError:
pass
else:
requests_cache.uninstall_cache()
def get_one_page(self, url, usecache=True):
if not usecache:
if requests_cache.get_cache().has_url(url):
requests_cache.get_cache().delete_url(url)
print "# fetching: %s" % url
i = requests.get(url, auth=(self.username, self.password))
#print "# fetched: %s" % url
return i
def get_comments(self, usecache=True):
for k in self.datadict.keys():
if 'comments_url' in self.datadict[k]:
if not usecache:
if requests_cache.get_cache().has_url(self.datadict[k]['comments_url']):
requests_cache.get_cache().delete_url(self.datadict[k]['comments_url'])
i = self.get_one_page(self.datadict[k]['comments_url'])
idict = json.loads(i.content)
self.datadict[k]['comments'] = idict
def has_url(url):
try:
cache = requests_cache.get_cache()
except AttributeError:
return False
return cache.has_url(url)
def load_annotations(data_folder):
contents = os.listdir(path=data_folder)
try:
infile = [i for i in contents if '.tsv' in i][0]
except IndexError:
raise Exception(f"No .tsv found in {contents}")
with open(infile, 'r') as litcovid_tsv:
tsv_reader = csv.reader(litcovid_tsv, delimiter='\t')
for i in range(32):
next(tsv_reader)
pmids = [line[0] for line in tsv_reader]
doc_id_set = set()
requests_cache.install_cache('litcovid_cache')
requests_cache.clear()
s = requests_cache.CachedSession()
s.hooks = {'response': throttle}
logging.debug("requests_cache: %s", requests_cache.get_cache().responses.filename)
for i, pmid in enumerate(pmids,start=1):
# NCBI eutils API limits requests to 10/sec
if i % 100 == 0:
logging.info("litcovid.parser.load_annotations progress %s", i)
doc = getPubMedDataFor(pmid, session=s)
if doc['_id'] not in doc_id_set:
yield doc
doc_id_set.add(doc['_id'])
remove_expired(s)
def get_timestamp(url):
"""
get the timestamp of an HTTP get request
:param url: the URL of the request
:return the timestamp of the request, of None if the request is not in the cache
"""
def _to_bytes(s, encoding='utf-8'):
return bytes(s, encoding)
def create_key(request):
url, body = request.url, request.body
key = hashlib.sha256()
key.update(_to_bytes(request.method.upper()))
key.update(_to_bytes(url))
if request.body:
key.update(_to_bytes(body))
return key.hexdigest()
def url_to_key(url):
session = requests.Session()
return create_key(session.prepare_request(requests.Request('GET', url)))
# get the cache from request_cache
results = requests_cache.get_cache()
# create the key according to the url
key_url = url_to_key(url)
# results.responses is a dictionary and follows the following format:
# { 'key': (requests_cache.backends objects, timestamp), ..., }
# for example: '4c28e3e4a61e325e520d9c02e0caee99e30c00951a223e67':
# (<requests_cache.backends.base._Store object at 0x12697e630>,
# datetime.datetime(2018, 10, 16, 0, 19, 8, 130204)),
if key_url in results.responses:
back_obj, timestamp = results.responses[key_url]
return timestamp
return None
def __init__(self,
cache_name="cache",
expire_after=300,
http_get_timeout=5,
delay_between_requests=8,
headers=HEADERS):
self.http_get_timeout = http_get_timeout
self.delay_between_requests = delay_between_requests
self.headers = headers
self.expire_after = expire_after
self.cache_name = cache_name
# Install cache
requests_cache.install_cache(self.cache_name,
backend='sqlite',
expire_after=self.expire_after)
self._cache = requests_cache.get_cache()
# Establish a session to be used for the GET requests
# IMPORTANT: the session must be established after installing the cache
# if you want all your responses to be cached, i.e. monkey-patching
# requests. Ref.: https://bit.ly/2MCDCeD
self._req_session = requests.Session()
# Add headers to the request session
self._req_session.headers = self.headers
self._last_request_time = -sys.float_info.max
self.response = None
def load_annotations(data_folder):
infile = os.path.join(data_folder, "litcovid2BioCJSON.gz")
assert os.path.exists(infile)
with open_anyfile(infile, mode='r') as file:
a = file.read()
data_list = json.loads(a)
# First item is a comment by provider
data = data_list[1]
doc_id_set = set()
with requests_cache.enabled('litcovid_cache', expire_after=expire_after):
logging.debug("requests_cache: %s",
requests_cache.get_cache().responses.filename)
for i, rec in enumerate(data, start=1):
# NCBI eutils API limits requests to 10/sec
if i % 100 == 0:
logging.info("litcovid.parser.load_annotations progress %s", i)
doc = getPubMedDataFor(rec["pmid"])
if not doc['from_cache']:
time.sleep(.2)
doc.pop('from_cache')
if doc['_id'] not in doc_id_set:
yield doc
doc_id_set.add(doc['_id'])
requests_cache.core.remove_expired_responses()
def weather():
lat = request.form['lat']
lon = request.form['lon']
t_threshold = request.form['t_threshold']
w_threshold = request.form['w_threshold']
metric = request.form['metric']
r = requests.get(
'https://api.darksky.net/forecast/25526a9ace577c46efe91103749d39ed/' +
lat + ',' + lon + '?units=' + metric +
'&exclude=minutely,flags,hourly')
old_stdout = sys.stdout
log_file = open("cache.log", "w")
sys.stdout = log_file
print('Cache enabled for this request:', r.from_cache)
print('Request:', requests_cache.get_cache())
sys.stdout = old_stdout
log_file.close()
json = r.json()
epoch = json['currently']['time']
currenttime = time.strftime("%a, %d %b %Y %H:%M:%S %Z",
time.localtime(epoch))
return render_template('newweather.html',
json=json,
currenttime=currenttime,
t_threshold=t_threshold,
w_threshold=w_threshold)
def install(self):
"""install cache"""
requests # require `import requests` in case this is essential for monkey patching by requests_cache.
requests_cache.install_cache(self.path, include_get_headers=True)
self.cache = requests_cache.get_cache()
logging.info(
f"requests-cache starting with {len(self.cache.responses)} cached responses"
)
def _wait_for_limiting(self):
# https://api.github.com/users/whatever
url = 'https://api.github.com/users/' + self.username
sleeptime = True
while sleeptime:
if requests_cache.get_cache().has_url(url):
requests_cache.get_cache().delete_url(url)
i = requests.get(url, auth=(self.username, self.password))
sleeptime = i.headers.get('X-RateLimit-Reset', None)
if sleeptime:
sleeptime = calendar.timegm(time.gmtime()) - int(sleeptime)
if sleeptime < 0:
sleeptime = sleeptime * -1
n_time = time.time()
n_time = datetime.fromtimestamp(n_time).strftime('%Y-%m-%d $H:%M:%S')
print "# %s sleeping %s" % (n_time, sleeptime)
time.sleep(sleeptime)
def setup(self):
ds_args = dict(id=self.name)
defaults = dict(name='HelMet-kirjastot')
self.data_source, _ = DataSource.objects.get_or_create(
defaults=defaults, **ds_args)
self.tprek_data_source = DataSource.objects.get(id='tprek')
self.ahjo_data_source = DataSource.objects.get(id='ahjo')
system_data_source_defaults = {'user_editable': True}
self.system_data_source, _ = DataSource.objects.get_or_create(id=settings.SYSTEM_DATA_SOURCE_ID,
defaults=system_data_source_defaults)
org_args = dict(origin_id='u4804001010', data_source=self.ahjo_data_source)
defaults = dict(name='Helsingin kaupunginkirjasto')
self.organization, _ = Organization.objects.get_or_create(defaults=defaults, **org_args)
org_args = dict(origin_id='00001', data_source=self.ahjo_data_source)
defaults = dict(name='Helsingin kaupunki')
self.city, _ = Organization.objects.get_or_create(defaults=defaults, **org_args)
# Build a cached list of Places
loc_id_list = [l[1] for l in LOCATIONS.values()]
place_list = Place.objects.filter(
data_source=self.tprek_data_source
).filter(origin_id__in=loc_id_list)
self.tprek_by_id = {p.origin_id: p.id for p in place_list}
# Create "Tapahtuma vain internetissä" location if not present
defaults = dict(data_source=self.system_data_source,
publisher=self.city,
name='Internet',
description='Tapahtuma vain internetissä.',)
self.internet_location, _ = Place.objects.get_or_create(id=INTERNET_LOCATION_ID, defaults=defaults)
try:
yso_data_source = DataSource.objects.get(id='yso')
except DataSource.DoesNotExist:
yso_data_source = None
if yso_data_source:
# Build a cached list of YSO keywords
cat_id_set = set()
for yso_val in YSO_KEYWORD_MAPS.values():
if isinstance(yso_val, tuple):
for t_v in yso_val:
cat_id_set.add('yso:' + t_v)
else:
cat_id_set.add('yso:' + yso_val)
keyword_list = Keyword.objects.filter(data_source=yso_data_source).\
filter(id__in=cat_id_set)
self.yso_by_id = {p.id: p for p in keyword_list}
else:
self.yso_by_id = {}
if self.options['cached']:
requests_cache.install_cache('helmet')
self.cache = requests_cache.get_cache()
else:
self.cache = None
def get_one_page(self, url, usecache=True, ignoreerrors=True):
limited = True
while limited:
if not usecache:
if requests_cache.get_cache().has_url(url):
requests_cache.get_cache().delete_url(url)
if self.username and self.password:
print "# fetching (BASIC): %s" % url
i = requests.get(url, auth=(self.username, self.password))
elif self.token:
print "# fetching (TOKEN): %s" % url
i = requests.get(url, headers={'Authorization': "token %s" % self.token})
else:
print "# fetching (NONE): %s" % url
i = requests.get(url)
if not i.ok:
print "# ERROR: %s for %s " % (i.reason, url)
#import epdb; epdb.st()
if 'rate limit exceeded' in i.content:
self._wait_for_limiting()
elif i.reason == 'Not Found':
limited = False
else:
import epdb; epdb.st()
sys.exit(1)
else:
data = None
try:
data = json.loads(i.content)
except:
pass
if data:
if 'documentation_url' in data:
limited = True
print "# hit rate limit, sleeping 300s"
time.sleep(300)
else:
limited = False
else:
limited = False
return i
def run(self):
def _parse_date(str):
if str is None:
return datetime.new()
return datetime(*parsedate(str)[:6])
self.start_time = clock()
try:
requests_cache.install_cache('.cache')
if not self.enable_cache:
log.debug("removing '%s' from cache" % self.url)
requests_cache.get_cache().delete_url(self.url)
log.debug("fetching '%s'" % self.url)
if self.url.startswith('file://'):
path = self.url[7:]
if not os.path.exists(path):
raise IOError("file not found: %s" % path)
with open(path, 'r') as fd:
self.result = fd.read()
self.cached = False
self.date = datetime.now()
self.last_modified = datetime.fromtimestamp(
os.stat(path).st_mtime)
else:
self.resp = requests.get(self.url, timeout=60, verify=False)
self.last_modified = _parse_date(
self.resp.headers.get('last-modified',
self.resp.headers.get('date', None)))
self.date = _parse_date(self.resp.headers['date'])
self.cached = getattr(self.resp, 'from_cache', False)
self.status = self.resp.status_code
if self.resp.status_code != 200:
raise IOError(self.resp.reason)
self.result = self.resp.content
log.debug("got %d bytes from '%s'" % (len(self.result), self.url))
except Exception, ex:
traceback.print_exc()
log.warn("unable to fetch '%s': %s" % (self.url, ex))
self.ex = ex
self.result = None
def main():
# The real request will only be made once; afterward, the cached response is used
for i in range(5):
response = requests.get('http://httpbin.org/get')
# This is more obvious when calling a slow endpoint
for i in range(5):
response = requests.get('http://httpbin.org/delay/2')
# Caching can be disabled if we want to get a fresh page and not cache it
with requests_cache.disabled():
print(requests.get('http://httpbin.org/ip').text)
# Get some debugging info about the cache
print(requests_cache.get_cache())
print('Cached URLS:', requests_cache.get_cache().urls)
# Uninstall to remove caching from all requests functions
requests_cache.uninstall_cache()
def run(self):
def _parse_date(str):
if str is None:
return datetime.new()
return datetime(*parsedate(str)[:6])
self.start_time = clock()
try:
requests_cache.install_cache('.cache')
if not self.enable_cache:
log.debug("removing '%s' from cache" % self.url)
requests_cache.get_cache().delete_url(self.url)
log.debug("fetching '%s'" % self.url)
if self.url.startswith('file://'):
path = self.url[7:]
if not os.path.exists(path):
raise IOError("file not found: %s" % path)
with open(path, 'r') as fd:
self.result = fd.read()
self.cached = False
self.date = datetime.now()
self.last_modified = datetime.fromtimestamp(os.stat(path).st_mtime)
else:
self.resp = requests.get(self.url, timeout=60, verify=False)
self.last_modified = _parse_date(self.resp.headers.get('last-modified', self.resp.headers.get('date', None)))
self.date = _parse_date(self.resp.headers['date'])
self.cached = getattr(self.resp, 'from_cache', False)
self.status = self.resp.status_code
if self.resp.status_code != 200:
raise IOError(self.resp.reason)
self.result = self.resp.content
log.debug("got %d bytes from '%s'" % (len(self.result), self.url))
except Exception, ex:
traceback.print_exc()
log.warn("unable to fetch '%s': %s" % (self.url, ex))
self.ex = ex
self.result = None
def _url_in_cache(url):
"""
If requests_cache is in use, return whether or not the URL is in the cache.
If not, return False.
"""
try:
return requests_cache.get_cache().has_url(url)
except AttributeError as e: # requests_cache not enabled
if e.message == "'Session' object has no attribute 'cache'":
return False
raise
def generate_csl_items(args, citekeys_df):
"""
General CSL (citeproc) items for standard_citekeys in citekeys_df.
Writes references.json to disk and logs warnings for potential problems.
"""
# Read manual references (overrides) in JSON CSL
manual_refs = load_manual_references(args.manual_references_paths)
requests_cache.install_cache(args.requests_cache_path,
include_get_headers=True)
cache = requests_cache.get_cache()
if args.clear_requests_cache:
logging.info('Clearing requests-cache')
requests_cache.clear()
logging.info(
f'requests-cache starting with {len(cache.responses)} cached responses'
)
csl_items = list()
failures = list()
for standard_citekey in citekeys_df.standard_citekey.unique():
if standard_citekey in manual_refs:
csl_items.append(manual_refs[standard_citekey])
continue
elif standard_citekey.startswith('raw:'):
logging.error(
f'CSL JSON Data with a standard_citekey of {standard_citekey!r} not found in manual-references.json. '
'Metadata must be provided for raw citekeys.')
failures.append(standard_citekey)
try:
csl_item = citekey_to_csl_item(standard_citekey)
csl_items.append(csl_item)
except Exception:
logging.exception(
f'Citeproc retrieval failure for {standard_citekey!r}')
failures.append(standard_citekey)
logging.info(
f'requests-cache finished with {len(cache.responses)} cached responses'
)
requests_cache.uninstall_cache()
if failures:
message = 'CSL JSON Data retrieval failed for the following standardized citation keys:\n{}'.format(
'\n'.join(failures))
logging.error(message)
# Write JSON CSL bibliography for Pandoc.
with args.references_path.open('w', encoding='utf-8') as write_file:
json.dump(csl_items, write_file, indent=2, ensure_ascii=False)
write_file.write('\n')
return csl_items
def main():
# Once cached, delayed page will be taken from cache
# redirects also handled
for i in range(5):
requests.get('http://httpbin.org/delay/2')
r = requests.get('http://httpbin.org/redirect/5')
print(r.text)
# And if we need to get fresh page or don't want to cache it?
with requests_cache.disabled():
print(requests.get('http://httpbin.org/ip').text)
# Debugging info about cache
print(requests_cache.get_cache())
def main():
# Once cached, delayed page will be taken from cache
# redirects also handled
for i in range(5):
requests.get('http://httpbin.org/delay/2')
r = requests.get('http://httpbin.org/redirect/5')
print(r.text)
# And if we need to get fresh page or don't want to cache it?
with requests_cache.disabled():
print(requests.get('http://httpbin.org/ip').text)
# Debugging info about cache
print(requests_cache.get_cache())
def setup(self):
ds_args = dict(id=self.name)
defaults = dict(name='HelMet-kirjastot')
self.data_source, _ = DataSource.objects.get_or_create(
defaults=defaults, **ds_args)
self.tprek_data_source = DataSource.objects.get(id='tprek')
ahjo_ds, _ = DataSource.objects.get_or_create(defaults=defaults,
**ds_args)
org_args = dict(id='ahjo:45400')
defaults = dict(name='Helsingin kaupunginkirjasto',
data_source=ahjo_ds)
self.organization, _ = Organization.objects.get_or_create(
defaults=defaults, **org_args)
# Build a cached list of Places
loc_id_list = [l[1] for l in LOCATIONS.values()]
place_list = Place.objects.filter(
data_source=self.tprek_data_source).filter(
origin_id__in=loc_id_list)
self.tprek_by_id = {p.origin_id: p.id for p in place_list}
try:
yso_data_source = DataSource.objects.get(id='yso')
except DataSource.DoesNotExist:
yso_data_source = None
if yso_data_source:
# Build a cached list of YSO keywords
cat_id_set = set()
for yso_val in YSO_KEYWORD_MAPS.values():
if isinstance(yso_val, tuple):
for t_v in yso_val:
cat_id_set.add('yso:' + t_v)
else:
cat_id_set.add('yso:' + yso_val)
keyword_list = Keyword.objects.filter(data_source=yso_data_source).\
filter(id__in=cat_id_set)
self.yso_by_id = {p.id: p for p in keyword_list}
else:
self.yso_by_id = {}
if self.options['cached']:
requests_cache.install_cache('helmet')
self.cache = requests_cache.get_cache()
else:
self.cache = None
def _is_response_cached(method, full_url):
"""Returns True if response to GET request is in requests_cache.
Args:
method (str): http verb ('GET', 'POST', etc.)
full_url (str): url, including the protocol
Returns:
is_cached (bool):
"""
if method != 'GET':
return False # pragma: no cover
try:
cache = requests_cache.get_cache()
except AttributeError: # pragma: no cover
cache = None
return cache.has_url(full_url) if cache else False
def _is_response_cached(method, full_url): # pragma: no cover
"""Returns True if response to GET request is in requests_cache.
Args:
method (str): http verb ('GET', 'POST', etc.)
full_url (str): url, including the protocol
Returns:
is_cached (bool):
"""
if method != 'GET':
return False
try:
cache = requests_cache.get_cache()
except AttributeError:
cache = None
return cache.has_url(full_url) if cache else False
def generate_csl_items(args, citation_df):
"""
General CSL (citeproc) items for standard_citations in citation_df.
Writes references.json to disk and logs warnings for potential problems.
"""
# Read manual references (overrides) in JSON CSL
manual_refs = read_manual_references(args.manual_references_path)
requests_cache.install_cache(args.requests_cache_path,
include_get_headers=True)
cache = requests_cache.get_cache()
if args.clear_requests_cache:
logging.info('Clearing requests-cache')
requests_cache.clear()
logging.info(
f'requests-cache starting with {len(cache.responses)} cached responses'
)
csl_items = list()
failures = list()
for citation in citation_df.standard_citation.unique():
if citation in manual_refs:
csl_items.append(manual_refs[citation])
continue
try:
citeproc = citation_to_citeproc(citation)
csl_items.append(citeproc)
except Exception as error:
logging.exception(f'Citeproc retrieval failure for {citation}')
failures.append(citation)
logging.info(
f'requests-cache finished with {len(cache.responses)} cached responses'
)
requests_cache.uninstall_cache()
if failures:
message = 'Citeproc retrieval failed for:\n{}'.format(
'\n'.join(failures))
logging.error(message)
# Write JSON CSL bibliography for Pandoc.
with args.references_path.open('w') as write_file:
json.dump(csl_items, write_file, indent=2, ensure_ascii=False)
write_file.write('\n')
return csl_items
def call_api(url, doi):
req = requests.Request('GET', url + doi)
cache = requests_cache.get_cache()
prepped = requests.Session().prepare_request(req)
cache_key = cache.create_key(prepped)
try:
response = cache.get_response(cache_key)
except (ImportError, TypeError):
response = None
if response:
return response.json()
return call_api_server(url, doi)
def setup(self):
ds_args = dict(id=self.name)
defaults = dict(name='HelMet-kirjastot')
self.data_source, _ = DataSource.objects.get_or_create(
defaults=defaults, **ds_args)
self.tprek_data_source = DataSource.objects.get(id='tprek')
ahjo_ds, _ = DataSource.objects.get_or_create(defaults=defaults, **ds_args)
org_args = dict(id='ahjo:45400')
defaults = dict(name='Helsingin kaupunginkirjasto', data_source=ahjo_ds)
self.organization, _ = Organization.objects.get_or_create(defaults=defaults, **org_args)
# Build a cached list of Places
loc_id_list = [l[1] for l in LOCATIONS.values()]
place_list = Place.objects.filter(
data_source=self.tprek_data_source
).filter(origin_id__in=loc_id_list)
self.tprek_by_id = {p.origin_id: p.id for p in place_list}
try:
yso_data_source = DataSource.objects.get(id='yso')
except DataSource.DoesNotExist:
yso_data_source = None
if yso_data_source:
# Build a cached list of YSO keywords
cat_id_set = set()
for yso_val in YSO_KEYWORD_MAPS.values():
if isinstance(yso_val, tuple):
for t_v in yso_val:
cat_id_set.add('yso:' + t_v)
else:
cat_id_set.add('yso:' + yso_val)
keyword_list = Keyword.objects.filter(data_source=yso_data_source).\
filter(id__in=cat_id_set)
self.yso_by_id = {p.id: p for p in keyword_list}
else:
self.yso_by_id = {}
if self.options['cached']:
requests_cache.install_cache('helmet')
self.cache = requests_cache.get_cache()
else:
self.cache = None
def get_cached_ids(self):
with requests_cache.enabled(cache_name=GENAPI_CACHE,
backend=CACHE_BACKEND):
cached_object = requests_cache.get_cache()
responses = [
cached_object.get_response(response)
for response in cached_object.responses
]
gen_ids = []
for url in [response.url for response in responses]:
gen_id = re.search(r'{}(.*?)/'.format(self._data_endpoint),
url)
if gen_id is not None:
gen_ids.append(gen_id.group(1))
return gen_ids
def setup(self):
self.tprek_data_source = DataSource.objects.get(id='tprek')
ds_args = dict(id=self.name)
ds_defaults = dict(name='City of Espoo')
self.data_source, _ = DataSource.objects.get_or_create(defaults=ds_defaults, **ds_args)
org_args = dict(origin_id='kaupunki', data_source=self.data_source)
org_defaults = dict(name='Espoon kaupunki')
self.organization, _ = Organization.objects.get_or_create(defaults=org_defaults, **org_args)
self._build_cache_places()
self._cache_yso_keywords()
if self.options['cached']:
requests_cache.install_cache('espoo')
self.cache = requests_cache.get_cache()
else:
self.cache = None
def setup(self):
self.tprek_data_source = DataSource.objects.get(id='tprek')
ds_args = dict(id=self.name)
ds_defaults = dict(name='City of Espoo')
self.data_source, _ = DataSource.objects.get_or_create(defaults=ds_defaults, **ds_args)
org_args = dict(id='espoo:kaupunki')
org_defaults = dict(name='Espoon kaupunki', data_source=self.data_source)
self.organization, _ = Organization.objects.get_or_create(defaults=org_defaults, **org_args)
self._build_cache_places()
self._cache_yso_keywords()
if self.options['cached']:
requests_cache.install_cache('espoo')
self.cache = requests_cache.get_cache()
else:
self.cache = None
def _is_url_in_cache(*args, **kwargs):
""" Return True if request has been cached or False otherwise. """
# Only include allowed arguments for a PreparedRequest.
allowed_args = inspect.getargspec(
requests.models.PreparedRequest.prepare).args
# self is in there as .prepare() is a method.
allowed_args.remove('self')
kwargs_cleaned = {}
for key, value in dict(kwargs).items():
if key in allowed_args:
kwargs_cleaned[key] = value
prepared_request = _prepare(*args, **kwargs_cleaned)
request_hash = _get_hash(prepared_request)
try:
return requests_cache.get_cache().has_key(request_hash)
except AttributeError as e: # requests_cache not enabled
if str(e) == "'Session' object has no attribute 'cache'":
return False
raise
def main(argv):
parser = build_cli_parser()
opts, args = parser.parse_args(argv)
if not opts.url or not opts.token:
parser.print_help()
sys.exit(-1)
global cache_file_name
cache_file_name = opts.cache_name
requests_cache.install_cache(cache_file_name,
allowable_methods=('GET', 'POST'))
global cbserverurl
cbserverurl = opts.url
#
# build a cbapi object
#
global cb
cb = cbapi.CbApi(opts.url, token=opts.token, ssl_verify=opts.ssl_verify)
#
# Run Tests to get cached responses
#
large_process_search()
large_binary_search()
test_sensors()
test_watchlist()
test_base_endpoints()
test_feeds()
cache = requests_cache.get_cache()
def main(argv):
parser = build_cli_parser()
opts, args = parser.parse_args(argv)
if not opts.url or not opts.token:
parser.print_help()
sys.exit(-1)
global cache_file_name
cache_file_name = opts.cache_name
requests_cache.install_cache(cache_file_name, allowable_methods=('GET', 'POST'))
global cbserverurl
cbserverurl = opts.url
#
# build a cbapi object
#
global cb
cb = cbapi.CbApi(opts.url, token=opts.token, ssl_verify=opts.ssl_verify)
#
# Run Tests to get cached responses
#
large_process_search()
large_binary_search()
test_sensors()
test_watchlist()
test_base_endpoints()
test_feeds()
cache = requests_cache.get_cache()
def extract_datareader(tickers, data_source="av-daily-adjusted", pause=None):
"""
Retrieve daily data with web.DataReader.
Parameters
----------
tickers : str or list of strs
Ticker or tickers to extract
data_source : str, optional
The data source ("quandl", "av-daily-adjusted", "iex", "fred", "ff", etc.)
pause : float, optional
Time, in seconds, to pause between consecutive queries of chunks
Warns
-----
On remote data error or invalid ticker. If using AlphaVantage API, extra
pause is made in those cases (see notes).
Yields
------
ticker : str
Short name for the extract data
data : dataframe
A dataframe with all extract data
metadata : dict
Extra information about the extract data
Notes
-----
Index is forced to datetime.
Attempt is made not to go over AlphaVantage API limits (maximum of five
calls per minute for non-premium accounts). This means that some pauses
are made, but they are not on the conservative side so be aware of remote
data errors and try again latter.
Examples
--------
>>> for ticker, data, metadata in extract_datareader("^BVSP"):
... print(ticker, metadata)
^BVSP {'price_column': 'adjusted close'}
>>> for ticker, data, metadata in extract_datareader(["PETR4.SAO", "ITUB4.SAO"], pause=1.0):
... print(ticker, metadata)
PETR4.SAO {'price_column': 'adjusted close'}
ITUB4.SAO {'price_column': 'adjusted close'}
>>> for ticker, data, metadata in extract_datareader("BCB/11", data_source="quandl"):
... print(ticker, metadata)
BCB/11 {'price_column': 'Value'}
"""
global _last_api_call
if isinstance(tickers, str):
tickers = [tickers]
if pause is None:
if data_source == "av-daily-adjusted":
pause = 12.0 # max. 5 calls per minute for non-premium accounts
else:
pause = 1.0
extra_pause = 0.25 * pause
if data_source == "quandl":
metadata = {"price_column": "Value"}
else:
metadata = {"price_column": "adjusted close"}
for ticker in tickers:
end = timer()
if end - _last_api_call > pause:
_last_api_call = end
else:
time_interval = end - _last_api_call + np.random.uniform(
0.0, extra_pause)
logging.info(f"Waiting {time_interval:.3f} seconds.")
time.sleep(time_interval)
_last_api_call = timer()
logging.info(f"Attempting to retrieve {ticker}.")
try:
# TODO: support start_date and end_date
data = web.DataReader(ticker, data_source)
except ValueError as e:
logging.warning(e)
time_interval = np.random.uniform(0.0, 2.0 * pause)
logging.info(f"Waiting {time_interval:.3f} extra seconds.")
time.sleep(time_interval)
continue
except pandas_datareader._utils.RemoteDataError as e:
logging.warning(f"Remote data error{e} for {ticker}.")
time_interval = np.random.uniform(0.0, 2.0 * pause)
logging.info(f"Waiting {time_interval:.3f} extra seconds.")
time.sleep(time_interval)
requests_cache.get_cache().remove_old_entries(
datetime.datetime.now())
continue
data.index = pd.to_datetime(data.index)
yield ticker, data, metadata
from pandas import read_csv
import time
from datetime import datetime
import json
from biothings.utils.common import open_anyfile
from biothings import config
logger = config.logger
import requests_cache
expire_after = datetime.timedelta(days=7)
requests_cache.install_cache('litcovidtopics_cache', expire_after=expire_after)
logger.debug("requests_cache: %s",
requests_cache.get_cache().responses.filename)
def get_pmids(res):
data = []
litcovid_data = res.text.split('\n')[34:]
for line in litcovid_data:
if line.startswith('#') or line.startswith('p'):
continue
if len(line.strip()) < 5:
continue
data.append('pmid' + line.split('\t')[0])
return (data)
def get_topics():
def schema_validate(instance, options):
"""Perform STIX JSON Schema validation against the input JSON.
Find the correct schema by looking at the 'type' property of the
`instance` JSON object.
Args:
instance: A STIX JSON string.
options: ValidationOptions instance with validation options for this
validation run.
Returns:
A dictionary of validation results
"""
if 'type' not in instance:
raise ValidationError(
"Input must be an object with a 'type' property.")
# Find and load the schema
try:
schema_path = find_schema(options.schema_dir, instance['type'])
schema = load_schema(schema_path)
except (KeyError, TypeError):
# Assume a custom object with no schema
try:
schema_path = find_schema(options.schema_dir, 'core')
schema = load_schema(schema_path)
except (KeyError, TypeError):
raise SchemaInvalidError("Cannot locate a schema for the object's "
"type, nor the base schema (core.json).")
# Validate against schemas for specific object types later
if instance['type'] == 'bundle':
schema['properties']['objects'] = {
"objects": {
"type": "array",
"minItems": 1
}
}
elif instance['type'] == 'observed-data':
schema['allOf'][1]['properties']['objects'] = {
"objects": {
"type": "object",
"minProperties": 1
}
}
# Validate the schema first
try:
CustomDraft4Validator.check_schema(schema)
except schema_exceptions.SchemaError as e:
raise SchemaInvalidError('Invalid JSON schema: ' + str(e))
# Cache data from external sources; used in some checks
if not options.no_cache:
requests_cache.install_cache(expire_after=datetime.timedelta(weeks=1))
if options.refresh_cache:
now = datetime.datetime.utcnow()
requests_cache.get_cache().remove_old_entries(now)
validator = load_validator(schema_path, schema, options)
output.info("Running the following additional checks: %s." %
", ".join(x.__name__ for x in validator.get_list()))
# Actual validation of JSON document
try:
some_errors = validator.iter_errors(instance)
more_errors = validator.iter_errors_more(instance)
warnings = validator.iter_errors_more(instance, False)
if options.strict:
chained_errors = chain(some_errors, more_errors, warnings)
warnings = []
else:
chained_errors = chain(some_errors, more_errors)
warnings = [pretty_error(x, options.verbose) for x in warnings]
except schema_exceptions.RefResolutionError:
raise SchemaInvalidError('Invalid JSON schema: a JSON reference '
'failed to resolve')
# List of error generators and message prefixes (to denote which object the
# error comes from)
error_gens = [(chained_errors, '')]
# Validate each object in a bundle separately
if instance['type'] == 'bundle' and 'objects' in instance:
for sdo in instance['objects']:
object_validate(sdo, options, error_gens)
else:
object_validate(instance, options, error_gens)
# Clear requests cache if commandline flag was set
if options.clear_cache:
now = datetime.datetime.utcnow()
requests_cache.get_cache().remove_old_entries(now)
# Prepare the list of errors
error_list = []
for gen, prefix in error_gens:
for error in gen:
msg = prefix + pretty_error(error, options.verbose)
error_list.append(SchemaError(msg))
if error_list:
valid = False
else:
valid = True
return ValidationResults(is_valid=valid,
errors=error_list,
warnings=warnings)
def install_cache(path='cache'):
"""Activates cache located at path"""
try:
requests_cache.get_cache()
except AttributeError:
requests_cache.install_cache(path)
def generate_csl_items(
citekeys: list,
manual_refs: dict = {},
requests_cache_path: Optional[str] = None,
clear_requests_cache: Optional[bool] = False,
) -> list:
"""
General CSL (citeproc) items for standard_citekeys in citekeys_df.
Parameters:
- citekeys: list of standard_citekeys
- manual_refs: mapping from standard_citekey to csl_item for manual references
- requests_cache_path: path for the requests cache database.
Passed as cache_name to `requests_cache.install_cache`.
requests_cache may append an extension to this path, so it is not always the exact
path to the cache. If None, do not use requests_cache.
- clear_requests_cache: If True, clear the requests cache before generating citekey metadata.
"""
# Deduplicate citations
citekeys = list(dict.fromkeys(citekeys))
# Install cache
if requests_cache_path is not None:
requests # require `import requests` in case this is essential for monkey patching by requests_cache.
requests_cache.install_cache(requests_cache_path,
include_get_headers=True)
cache = requests_cache.get_cache()
if clear_requests_cache:
logging.info("Clearing requests-cache")
requests_cache.clear()
logging.info(
f"requests-cache starting with {len(cache.responses)} cached responses"
)
csl_items = list()
failures = list()
for standard_citekey in citekeys:
if standard_citekey in manual_refs:
csl_items.append(manual_refs[standard_citekey])
continue
elif standard_citekey.startswith("raw:"):
logging.error(
f"CSL JSON Data with a standard_citekey of {standard_citekey!r} not found in manual-references.json. "
"Metadata must be provided for raw citekeys.")
failures.append(standard_citekey)
try:
csl_item = citekey_to_csl_item(standard_citekey)
csl_items.append(csl_item)
except Exception:
logging.exception(
f"Citeproc retrieval failure for {standard_citekey!r}")
failures.append(standard_citekey)
# Uninstall cache
if requests_cache_path is not None:
logging.info(
f"requests-cache finished with {len(cache.responses)} cached responses"
)
requests_cache.uninstall_cache()
if failures:
message = "CSL JSON Data retrieval failed for the following standardized citation keys:\n{}".format(
"\n".join(failures))
logging.error(message)
return csl_items
from datetime import datetime, timedelta
import requests
import requests_cache
requests_cache.install_cache('demo_cache')
CACHE = requests_cache.get_cache()
from django.core.management.base import BaseCommand
from django.utils.text import slugify
from organisations.models import (Organisation, OrganisationDivision,
OrganisationDivisionSet)
from organisations.constants import PARENT_TO_CHILD_AREAS
class Command(BaseCommand):
skip_gss = [
"E12000007",
"W08000001" # See import_welsh_areas
]
BASE = "http://mapit.mysociety.org"
def add_arguments(self, parser):
parser.add_argument('--always_pick_option',
action='store',
type=int,
default=0)
def handle(self, **options):
self.always_pick_option = int(options['always_pick_option'])
def delete_cached_url(url):
"""Deletes the given URL from the cache"""
try:
requests_cache.get_cache().delete_url(url)
except AttributeError:
pass
def clear_requests_cache():
"""
Clears all cached responses.
"""
now = datetime.datetime.utcnow()
requests_cache.get_cache().remove_old_entries(now)
import requests_cache
requests_cache.configure("example_cache")
def main():
# Once cached, delayed page will be taken from cache
# redirects also handled
for i in range(5):
requests.get("http://httpbin.org/delay/2")
r = requests.get("http://httpbin.org/redirect/5")
print(r.text)
# What about async? It's also supported!
rs = [async.get("http://httpbin.org/delay/%s" % i) for i in range(5)]
for r in async.map(rs):
print(r.text)
# And if we need to get fresh page or don't want to cache it?
with requests_cache.disabled():
print(requests.get("http://httpbin.org/ip").text)
# Debugging info about cache
print(requests_cache.get_cache())
if __name__ == "__main__":
t = time.time()
main()
print("Elapsed: %.3f seconds" % (time.time() - t))
