def __init__(self):
self._API_SERVER = "https://data.nba.net"
self._cache_control_adapter = CacheControlAdapter(heuristic=LastModified())
self._requests_session = requests.Session()
self._requests_session.mount('http://', CacheControlAdapter())
self._requests_session.mount('https://', CacheControlAdapter())
self._TEAM_TRICODES = frozenset(('CHA', 'ATL', 'IND', 'MEM', 'DET',
'UTA', 'CHI', 'TOR', 'CLE', 'OKC',
'DAL', 'MIN', 'BOS', 'SAS', 'MIA',
'DEN', 'LAL', 'PHX', 'NOP', 'MIL',
'HOU', 'NYK', 'ORL', 'SAC', 'PHI',
'BKN', 'POR', 'GSW', 'LAC', 'WAS'))
self._STAT_CATEGORIES = frozenset(('ppg', 'trpg', 'apg', 'fgp', 'ftp',
'tpp', 'bpg', 'spg', 'tpg', 'pfpg'))
self._CONFERENCES = frozenset(('west', 'east'))
self._EASTERN_DIVISIONS = frozenset(('southeast', 'atlantic', 'central'))
self._WESTERN_DIVISIONS = frozenset(('southwest', 'pacific', 'northwest'))
self._DIVISIONS = {'west': self._WESTERN_DIVISIONS,
'east': self._EASTERN_DIVISIONS}
# Cached dictionaries. Saving these copies avoids having to
# re-parse JSONs when they are returned from the HTTP cache.
self._person_ids = None
self._team_ids_to_tricodes = None
self._team_tricodes_to_ids = None
def __init__(self, main_source_domain=None, start_page=None):
if main_source_domain is None and start_page is None:
raise ValueError('Need to specify main_source_domain or start_page.')
if main_source_domain:
self.MAIN_SOURCE_DOMAIN = main_source_domain.rstrip('/')
self.START_PAGE = self.MAIN_SOURCE_DOMAIN
if self.MAIN_SOURCE_DOMAIN is None:
parsedurl = urlparse(start_page)
self.MAIN_SOURCE_DOMAIN = parsedurl.scheme + '://' + parsedurl.netloc
if self.MAIN_SOURCE_DOMAIN not in self.SOURCE_DOMAINS:
self.SOURCE_DOMAINS.append(self.MAIN_SOURCE_DOMAIN)
if start_page:
self.START_PAGE = start_page
# make resolve any redirects
#verdict, head_response = self.is_html_file(self.START_PAGE)
is_new_url, content_type, content_length, return_url = self.get_url_type(self.START_PAGE)
if content_type == 'text/html':
self.START_PAGE = return_url
else:
raise ValueError('The Starting URL ' + self.START_PAGE + ' did not return any html.')
forever_adapter= CacheControlAdapter(heuristic=CacheForeverHeuristic(), cache=self.CACHE)
for source_domain in self.SOURCE_DOMAINS:
self.SESSION.mount(source_domain, forever_adapter) # TODO: change to less aggressive in final version
def __init__(self):
super(ProjectHolder, self).__init__()
app_name = __name__.split('.')[0]
self.cache_dir = user_cache_dir(app_name)
log.info("Using cache directory: {}.".format(self.cache_dir))
self.cache = FileCache(self.cache_dir)
cache_adapter = CacheControlAdapter(cache=self.cache)
self.mount("http://", cache_adapter)
self.mount("https://", cache_adapter)
self.headers.update(
{'User-Agent': '{}/{}'.format(app_name, __version__)})
log.info('Created instance of {}'.format(type(self).__name__))
self.branches = None
self.only = None
self.exclude = None
self.having_asset = None
self.hostname = None
# identifies project on a given hostname
self.repo = None
# short name for "repo", useful in URLs
self.name = None
# in some case we do not specify repo, but feed is discovered, no repo is given then
self.feed_url = None
def get_events_from_icalendars():
global now, midnight
now = localtz.localize(datetime.datetime.now())
midnight = localtz.localize(datetime.datetime.combine(now, datetime.time(0,0,0)))
cz = Calzone()
session = FuturesSession()
session.mount('https://', CacheControlAdapter(cache=FileCache('.webcache'), heuristic=ForceCacheHeuristic()))
cals = {k: session.get(u) for k,u in calendars.items()}
concurrent.futures.wait(cals.values())
for k,req in cals.items():
try:
cz.load(req.result().text)
except Exception as err:
print("Failed to load calendar '{}'".format(k))
print (err)
try:
events = cz.get_events(midnight, midnight + datetime.timedelta(days=90))
except Exception as e:
print (e)
events.sort(key=lambda e: e.start)
return events
def __init__(self, filename=""):
super(BabelNet, self).__init__()
if filename == "":
filename = "babelnet_cache"
self.mount('https://', CacheControlAdapter(cache=FileCache(filename)))
self.headers.update({'Accept-Encoding': 'gzip'})
self.params.update({'key': cfg.babelnet_key})
self.endpoint = "https://babelnet.io/v4/"
def request(self, method, url, headers=None, params=None, proxies=None, cache=True, verify=False, *args, **kwargs):
if headers is None: headers = {}
if params is None: params = {}
if proxies is None: proxies = {}
headers['Accept-Encoding'] = 'gzip, deflate'
headers["User-Agent"] = sickrage.srCore.USER_AGENT
# request session ssl verify
if sickrage.srCore.srConfig.SSL_VERIFY:
try:
verify = certifi.where()
except:
pass
# request session proxies
if 'Referer' not in headers and sickrage.srCore.srConfig.PROXY_SETTING:
sickrage.srCore.srLogger.debug("Using global proxy: " + sickrage.srCore.srConfig.PROXY_SETTING)
scheme, address = urllib2.splittype(sickrage.srCore.srConfig.PROXY_SETTING)
address = ('http://{}'.format(sickrage.srCore.srConfig.PROXY_SETTING),
sickrage.srCore.srConfig.PROXY_SETTING)[scheme]
proxies.update({"http": address, "https": address})
headers.update({'Referer': address})
# setup caching adapter
if cache:
adapter = CacheControlAdapter(DBCache(os.path.abspath(os.path.join(sickrage.DATA_DIR, 'sessions.db'))))
self.mount('http://', adapter)
self.mount('https://', adapter)
# get web response
response = super(srSession, self).request(
method,
url,
headers=headers,
params=params,
verify=verify,
proxies=proxies,
*args, **kwargs
)
try:
# check web response for errors
response.raise_for_status()
except requests.exceptions.SSLError as e:
if ssl.OPENSSL_VERSION_INFO < (1, 0, 1, 5):
sickrage.srCore.srLogger.info(
"SSL Error requesting url: '{}' You have {}, try upgrading OpenSSL to 1.0.1e+".format(
e.request.url, ssl.OPENSSL_VERSION))
if sickrage.srCore.srConfig.SSL_VERIFY:
sickrage.srCore.srLogger.info(
"SSL Error requesting url: '{}', try disabling cert verification in advanced settings".format(
e.request.url))
except Exception:
pass
return response
def set_cache_expiration_delay(seconds):
"""
Set a cache for requests with a given expiration time.
"""
adapter = CacheControlAdapter(
heuristic=ExpiresAfter(seconds=seconds)
)
session.mount('http://', adapter)
return session
def session(self):
if self._session is None:
self._session = real_requests.Session()
if CacheControlAdapter:
adapter = CacheControlAdapter(cache=FileCache(".webcache"))
self._session.mount("http://", adapter)
self._session.mount("https://", adapter)
print("Caching to .webcache")
return self._session
def get_session(*args, **kwargs):
session = OAuth2Session(*args, **kwargs)
cache_adapter = CacheControlAdapter(
cache=FileCache(CACHE_FILE),
pool_connections=config.http.connections,
pool_maxsize=config.http.connections,
max_retries=config.http.retries,
)
session.mount("http://", cache_adapter)
return session
def _get_requests_cache_adapter(heuristic):
"""
Given a heuristic, constructs and returns a
:class:`cachecontrol.CacheControlAdapter` attached to the instance's
:data:`requests_cache`.
"""
return CacheControlAdapter(cache=requests_cache,
heuristic=heuristic,
cache_etags=False)
def __init__(self, headers=None, cookies=None, cache_name=None, delay=1, expire_hours=12, as_string=False):
'''
Base class for common scraping tasks
Args:
headers: dict of headers
cookies: cookiejar object
cache_name: should be full path
delay: int (be polite!!!)
expire_hours: int - default 4
as_string: get string rather than parsed json
'''
logging.getLogger(__name__).addHandler(logging.NullHandler())
if not cookies:
try:
import cookielib
cookies = cookielib.MozillaCookieJar()
except (NameError, ImportError) as e:
try:
import http.cookiejar
cookies = http.cookiejar.MozillaCookieJar()
except Exception as e:
pass
_s = requests.Session()
_s.cookies = cookies
if headers:
_s.headers.update(headers)
else:
_s.headers.update({'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/55.0.2883.87 Safari/537.36'})
if cache_name:
if not '/' in cache_name:
cache_name = os.path.join('/tmp', cache_name)
try:
from cachecontrol import CacheControlAdapter
from cachecontrol.heuristics import ExpiresAfter
from cachecontrol.caches import FileCache
_s.mount('http://', CacheControlAdapter(cache=FileCache(cache_name), cache_etags = False, heuristic=ExpiresAfter(hours=expire_hours)))
except ImportError as e:
try:
import requests_cache
requests_cache.install_cache(cache_name)
except:
pass
self.s = _s
self.urls = []
self.as_string = as_string
if delay > 0:
self.delay = delay
else:
self.delay = None
def add_cache_control(session, cache_control_config):
"""Add cache_control adapter to session object."""
adapter = CacheControlAdapter(
DictCache(),
cache_etags=cache_control_config.get('cache_etags', True),
serializer=cache_control_config.get('serializer', None),
heuristic=cache_control_config.get('heuristic', None),
)
session.mount('http://', adapter)
session.mount('https://', adapter)
session.cache_controller = adapter.controller
def _create_session(self, max_retries, proxies, backoff_factor, cache):
sess = Session()
# Retry only on idempotent methods and only when too many requests
retries = Retry(total=max_retries, backoff_factor=backoff_factor, status_forcelist=[429], method_whitelist=['GET', 'UPDATE', 'DELETE'])
retries_adapter = HTTPAdapter(max_retries=retries)
if cache:
cache_adapter = CacheControlAdapter(cache_etags=True)
sess.mount('http://', retries_adapter)
sess.mount('http://', cache_adapter)
sess.proxies.update(proxies)
return sess
def getsession(timeframe):
if timeframe in sessions:
return sessions[timeframe]
else:
sess = requests.Session()
sess.cookies = __cookie
if timeframe == 0:
sess.mount("http://", CacheControlAdapter(cache=__cache))
sess.mount("https://", CacheControlAdapter(cache=__cache))
else:
sess.mount(
"http://",
CacheControlAdapter(cache=__cache,
heuristic=timecache(timeframe)))
sess.mount(
"https://",
CacheControlAdapter(cache=__cache,
heuristic=timecache(timeframe)))
sessions[timeframe] = sess
return sess
def __init__(self, proxies=None, cache=True):
super(WebSession, self).__init__()
# setup caching adapter
if cache:
adapter = CacheControlAdapter()
self.mount('http://', adapter)
self.mount('https://', adapter)
# add proxies
self.proxies = proxies or _add_proxies()
# add hooks
self.hooks['response'] += [WebHooks.log_url, WebHooks.cloudflare]
def __init__(self, url, max_retries, expires_after_sec):
self.hasError = False
self.fromCache = False
self.url = url
self.cacheEnabled = expires_after_sec > 0
self.session = requests.Session()
retryPolicy = urllib3.util.Retry(max_retries, status_forcelist=[400])
if self.cacheEnabled:
self.session.mount(
url,
CacheControlAdapter(
max_retries=retryPolicy,
heuristic=ExpiresAfter(seconds=expires_after_sec)))
else:
self.session.mount(
url, requests.adapters.HTTPAdapter(max_retries=retryPolicy))
def _build_session():
"""Builds a requests session that caches responses where possible, making redirects faster.
Returns:
requests.Session -- A shared session to use for the notebook
"""
result = requests.session()
# Set up caching. Particularly obey and cache 307 redirects to avoid duplicate expensive calls when we already
# have a result
cache_adapter = CacheControlAdapter()
cache_adapter.controller = CacheController(cache=cache_adapter.cache, status_codes=(200, 203, 300, 301, 307))
result.mount('http://', cache_adapter)
result.mount('https://', cache_adapter)
return result
def __init__(self, proxies=None, cache=True):
super(WebSession, self).__init__()
# setup caching adapter
if cache:
adapter = CacheControlAdapter()
self.mount('http://', adapter)
self.mount('https://', adapter)
# add proxies
self.proxies = proxies or _add_proxies()
# add hooks
self.hooks['response'] += [WebHooks.log_url, WebHooks.cloudflare]
# add headers
self.headers.update({'Accept-Encoding': 'gzip, deflate', 'User-Agent': sickrage.app.user_agent})
def __init__(
self,
user=None,
prefix_url=settings.SITE_URL,
save_auth=None,
verify=settings.CERN_SSL_CHAIN,
):
super(Session, self).__init__()
self.user = user if user else User(save_auth=save_auth)
self.auth = self._authorize
self.prefix_url = prefix_url
self.verify = verify
# store last call
self._response = None
# add caching
super(Session, self).mount(
self.prefix_url, CacheControlAdapter(cache=FileCache('.webcache'))
)
def __init__(self, *args, **kwargs):
retries = kwargs.pop("retries", 0)
cache = kwargs.pop("cache", None)
super(PypiSession, self).__init__(*args, **kwargs)
# Attach our User Agent to the request
self.headers["User-Agent"] = user_agent()
# Attach our Authentication handler to the session
self.auth = MultiDomainBasicAuth()
# Create our urllib3.Retry instance which will allow us to customize
# how we handle retries.
retries = urllib3.Retry(
# Set the total number of retries that a particular request can
# have.
total=retries,
# A 503 error from PyPI typically means that the Fastly -> Origin
# connection got interupted in some way. A 503 error in general
# is typically considered a transient error so we'll go ahead and
# retry it.
status_forcelist=[503],
# Add a small amount of back off between failed requests in
# order to prevent hammering the service.
backoff_factor=0.25,
)
# We want to _only_ cache responses on securely fetched origins. We do
# this because we can't validate the response of an insecurely fetched
# origin, and we don't want someone to be able to poison the cache and
# require manual eviction from the cache to fix it.
if cache:
secure_adapter = CacheControlAdapter(
cache=SafeFileCache(cache, use_dir_lock=True),
max_retries=retries,
)
else:
secure_adapter = HTTPAdapter(max_retries=retries)
self.mount("https://", secure_adapter)
self.mount("file://", LocalFSAdapter())
def get_feed(feed_url):
"""
Return feed parsed feed
"""
requests_timeout = getattr(settings, 'FEED_TIMOUT', 1)
cache_adapter = CacheControlAdapter(
cache=FileCache('.web_cache'),
heuristic=ExpiresAfter(hours=1),
)
session = requests.Session()
session.mount('http://', cache_adapter)
session.mount('https://', cache_adapter)
show_exceptions = getattr(settings, 'DEBUG', True)
feed_request = session.get(feed_url, timeout=requests_timeout)
return feedparser.parse(feed_request.text)
def __init__(self, main_source_domain=None, start_page=None):
if main_source_domain is None and start_page is None:
raise ValueError(
'Need to specify main_source_domain or start_page.')
if main_source_domain:
self.MAIN_SOURCE_DOMAIN = main_source_domain.rstrip('/')
self.START_PAGE = self.MAIN_SOURCE_DOMAIN
if self.MAIN_SOURCE_DOMAIN is None:
parsedurl = urlparse(start_page)
self.MAIN_SOURCE_DOMAIN = parsedurl.scheme + '://' + parsedurl.netloc
if self.MAIN_SOURCE_DOMAIN not in self.SOURCE_DOMAINS:
self.SOURCE_DOMAINS.append(self.MAIN_SOURCE_DOMAIN)
if start_page:
self.START_PAGE = start_page
# keep track of broken links
self.broken_links = []
forever_adapter = CacheControlAdapter(
heuristic=CacheForeverHeuristic(), cache=self.CACHE)
for source_domain in self.SOURCE_DOMAINS:
self.SESSION.mount(
source_domain, forever_adapter
) # TODO: change to less aggressive in final version
def _build_session(self, max_retries):
from requests.adapters import HTTPAdapter
if not isinstance(max_retries, int):
raise ValueError(f'int expected, found {type(max_retries)}.')
elif max_retries < 1:
raise ValueError('max_retries should be greater or equal to 1.')
session = requests.Session()
# mount retries adapter
session.mount(
'http://',
HTTPAdapter(max_retries=Retry(total=max_retries,
method_whitelist=frozenset(
['GET', 'POST']))))
# mount cache adapter
session.mount('http://',
CacheControlAdapter(heuristic=ExpiresAfter(hours=1)))
session.headers['User-Agent'] = USER_AGENT
self.session = session
import zipfile
from typing import Any, Dict, Optional
import aiohttp
import requests
from cachecontrol import CacheControl, CacheControlAdapter
from cachecontrol.caches.file_cache import FileCache
from cachecontrol.heuristics import ExpiresAfter
from shared import configuration, perf
from shared.pd_exception import OperationalException
SESSION = CacheControl(requests.Session(),
cache=FileCache(configuration.get('web_cache')))
SESSION.mount('http://whatsinstandard.com',
CacheControlAdapter(heuristic=ExpiresAfter(days=14)))
AIOSESSION = aiohttp.ClientSession()
def unzip(url: str, path: str) -> str:
location = '{scratch_dir}/zip'.format(
scratch_dir=configuration.get('scratch_dir'))
def remove_readonly(func, path, _):
os.chmod(path, stat.S_IWRITE)
func(path)
shutil.rmtree(location, True, remove_readonly)
os.mkdir(location)
store(url, '{location}/zip.zip'.format(location=location))
def __init__(self, **kwargs):
'''
Base class for common scraping tasks
Args:
'''
logging.getLogger(__name__).addHandler(logging.NullHandler())
self.urls = []
# use requests HTML to aid parsing
# has all same methods as requests.Session
_s = HTMLSession()
# delay/expire
if kwargs.get('delay'):
self.delay = kwargs['delay']
else:
self.delay = 2
if kwargs.get('expire_hours'):
self.expire_hours = kwargs['expire_hours']
else:
self.expire_hours = 168
# add cookies
if kwargs.get('cookies'):
_s.cookies = kwargs['cookies']
else:
try:
import cookielib
_s.cookies = cookielib.MozillaCookieJar()
except (NameError, ImportError):
import http.cookiejar
_s.cookies = http.cookiejar.MozillaCookieJar()
# add headers
if kwargs.get('headers'):
_s.headers = kwargs['headers']
else:
ua = ('Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 '
'(KHTML, like Gecko) Chrome/55.0.2883.87 Safari/537.36')
_s.headers = {'User-Agent': ua}
# add proxies
if kwargs.get('proxies'):
_s.proxies = kwargs['proxies']
# add cache
if not '/' in kwargs.get('cache_name', ''):
self.cache_name = os.path.join('/tmp', kwargs['cache_name'])
try:
from cachecontrol import CacheControlAdapter
from cachecontrol.heuristics import ExpiresAfter
from cachecontrol.caches import FileCache
_s.mount('http://', CacheControlAdapter(cache=FileCache(self.cache_name),
cache_etags = False,
heuristic=ExpiresAfter(hours=self.expire_hours)))
except ImportError as e:
try:
import requests_cache
requests_cache.install_cache(self.cache_name)
except:
logging.exception('could not install cache')
self.s = _s
def cli(url, repositories, search, table, rows, minstar, report, description,
token):
MODE = os.environ.get("GHTOPDEP_ENV")
BASE_URL = 'https://437w61gcj1.execute-api.us-west-2.amazonaws.com/api'
if MODE == "development":
BASE_URL = 'http://127.0.0.1:8080'
if report:
try:
result = requests.get('{}/repos?url={}'.format(BASE_URL, url))
if result.status_code != 404:
sorted_repos = sort_repos(result.json()['deps'], rows)
repos = readable_stars(sorted_repos)
click.echo(tabulate(repos, headers="keys", tablefmt="github"))
sys.exit()
except requests.exceptions.ConnectionError as e:
click.echo(e)
if (description or search) and token:
gh = github3.login(token=token)
CacheControl(gh.session,
cache=FileCache(CACHE_DIR),
heuristic=OneDayHeuristic())
elif (description or search) and not token:
click.echo("Please provide token")
sys.exit()
destination = "repository"
destinations = "repositories"
if not repositories:
destination = "package"
destinations = "packages"
repos = []
more_than_zero_count = 0
total_repos_count = 0
spinner = Halo(text="Fetching information about {0}".format(destinations),
spinner="dots")
spinner.start()
sess = requests.session()
retries = Retry(total=15, backoff_factor=15, status_forcelist=[429])
adapter = CacheControlAdapter(max_retries=retries,
cache=FileCache(CACHE_DIR),
heuristic=OneDayHeuristic())
sess.mount("http://", adapter)
sess.mount("https://", adapter)
page_url = get_page_url(sess, url, destination)
while True:
response = sess.get(page_url)
parsed_node = HTMLParser(response.text)
dependents = parsed_node.css(ITEM_SELECTOR)
total_repos_count += len(dependents)
for dep in dependents:
repo_stars_list = dep.css(STARS_SELECTOR)
# only for ghost or private? packages
if repo_stars_list:
repo_stars = repo_stars_list[0].text().strip()
repo_stars_num = int(repo_stars.replace(",", ""))
else:
continue
if repo_stars_num != 0:
more_than_zero_count += 1
if repo_stars_num >= minstar:
relative_repo_url = dep.css(
REPO_SELECTOR)[0].attributes["href"]
repo_url = "{0}{1}".format(GITHUB_URL, relative_repo_url)
# can be listed same package
is_already_added = already_added(repo_url, repos)
if not is_already_added and repo_url != url:
if description:
repo_description = fetch_description(
gh, relative_repo_url)
repos.append({
"url": repo_url,
"stars": repo_stars_num,
"description": repo_description
})
else:
repos.append({
"url": repo_url,
"stars": repo_stars_num
})
node = parsed_node.css(NEXT_BUTTON_SELECTOR)
if len(node) == 2:
page_url = node[1].attributes["href"]
elif len(node) == 0 or node[0].text() == "Previous":
spinner.stop()
break
elif node[0].text() == "Next":
page_url = node[0].attributes["href"]
if report:
try:
requests.post('{}/repos'.format(BASE_URL),
json={
"url": url,
"deps": repos
})
except requests.exceptions.ConnectionError as e:
click.echo(e)
sorted_repos = sort_repos(repos, rows)
if search:
for repo in repos:
repo_path = urlparse(repo["url"]).path[1:]
for s in gh.search_code("{0} repo:{1}".format(search, repo_path)):
click.echo("{0} with {1} stars".format(s.html_url,
repo["stars"]))
else:
show_result(sorted_repos, total_repos_count, more_than_zero_count,
destinations, table)
@oauth_authorized.connect_via(github_bp)
def github_logged_in(blueprint, token):
if not token:
flash("Failed to log in with Github")
if "error_reason" in token:
msg = "Access denied. Reason={reason} error={error}".format(
reason=request.args["error_reason"],
error=request.args["error_description"],
)
flash(msg)
else:
flash("Successfully signed in with Github")
# install CacheControl for github session, so we don't eat up API usage unnecessarily
github_bp.session.mount(github_bp.session.base_url, CacheControlAdapter())
## UTILITY FUNCTIONS ##
def jira_get(*args, **kwargs):
"""
JIRA sometimes returns an empty response to a perfectly valid GET request,
so this will retry it a few times if that happens.
"""
for _ in range(3):
resp = jira_bp.session.get(*args, **kwargs)
if resp.content:
return resp
return jira_bp.session.get(*args, **kwargs)
def __init__(self, **kwargs):
"""
"""
logging.getLogger(__name__).addHandler(logging.NullHandler())
self.urls = []
# use requests HTML to aid parsing
# has all same methods as requests.Session
_s = HTMLSession()
self.delay = kwargs.get("delay", 2)
self.expire_hours = kwargs.get("expire_hours", 168)
# add cookies
if kwargs.get("cookies"):
_s.cookies = kwargs["cookies"]
else:
import http.cookiejar
_s.cookies = http.cookiejar.MozillaCookieJar()
# add headers
default_headers = {
"User-Agent": random.choice(USER_AGENTS),
"accept-encoding": "gzip, deflate, br",
"accept-language": "en-US,en;q=0.9",
"accept": "application/json, text/plain, */*",
}
_s.headers.update(default_headers)
if kwargs.get("headers"):
_s.headers.update(kwargs["headers"])
# add proxies
if kwargs.get("proxies"):
_s.proxies = kwargs["proxies"]
# add cache
if not kwargs.get("cache_name"):
self.cache_name = os.path.join("/tmp", random_string(32))
elif "/" not in kwargs.get("cache_name", ""):
self.cache_name = os.path.join("/tmp", kwargs["cache_name"])
else:
self.cache_name = kwargs.get("cache_name")
try:
from cachecontrol import CacheControlAdapter
from cachecontrol.heuristics import ExpiresAfter
from cachecontrol.caches import FileCache
_s.mount(
"http://",
CacheControlAdapter(
cache=FileCache(self.cache_name),
cache_etags=False,
heuristic=ExpiresAfter(hours=self.expire_hours),
),
)
except ImportError:
try:
import requests_cache
requests_cache.install_cache(self.cache_name)
except BaseException:
logging.exception("could not install cache")
self.session = _s
import os, datetime
import requests
from .data.rfeed import Item, Feed
from flask import Flask, jsonify, request as flask_request
from cachecontrol import CacheControlAdapter
from cachecontrol.heuristics import LastModified
app = Flask(__name__)
adapter = CacheControlAdapter(heuristic=LastModified())
sess = requests.Session()
sess.mount('http://', adapter)
sess.mount('https://', adapter)
SERVICE_NAME = os.path.splitext(os.path.basename(__file__))[0]
@app.route("/rss/summary", methods=['GET'])
def latest_articles():
if flask_request.method == 'GET':
response = sess.get('http://localhost/article/collect/10')
article_collection = []
if response.status_code == requests.codes.ok:
articles = response.json()['success']
for article in articles:
article_collection.append(
Item(
title=article['title'],
author=article['author'],
pubDate=datetime.datetime.strptime(
TIMEOUT_SECONDS = 10 # Seconds before URL query timeout is raised
PROVIDERS_URLS = [
"https://providers.optimade.org/v1/links",
"https://raw.githubusercontent.com/Materials-Consortia/providers/master/src"
"/links/v1/providers.json",
]
CACHE_DIR = Path(appdirs.user_cache_dir("optimade-client", "CasperWA"))
CACHE_DIR.mkdir(parents=True, exist_ok=True)
CACHED_PROVIDERS = CACHE_DIR / "cached_providers.json"
SESSION = requests.Session()
SESSION_ADAPTER = CacheControlAdapter(
cache=FileCache(CACHE_DIR / ".requests_cache"), heuristic=ExpiresAfter(days=1)
)
SESSION_ADAPTER_DEBUG = CacheControlAdapter()
SESSION.mount("http://", SESSION_ADAPTER)
SESSION.mount("https://", SESSION_ADAPTER)
SESSION.mount("http://localhost", SESSION_ADAPTER_DEBUG)
SESSION.mount("http://127.0.0.1", SESSION_ADAPTER_DEBUG)
# Currently known providers' development OPTIMADE base URLs
DEVELOPMENT_PROVIDERS = {"mcloud": "https://dev-www.materialscloud.org/optimade"}
try:
DEVELOPMENT_MODE = bool(int(os.getenv("OPTIMADE_CLIENT_DEVELOPMENT_MODE", "0")))
except ValueError:
LOGGER.debug(
(