def __init__(self, enable_debug):
self.log = logging.getLogger(__name__)
today = datetime.date.today()
previousMonth = today - datetime.timedelta(days=31)
prevoiusMonthSubDir = previousMonth.strftime("%Y-%m")
cache_dir_to_remove = os.path.join(os.path.normpath(os.path.expanduser("~/.epguide/cache")), prevoiusMonthSubDir)
if os.path.exists(cache_dir_to_remove):
self.log.info("Removing previous month cache dir: " + cache_dir_to_remove)
try:
shutil.rmtree(cache_dir_to_remove)
self.log.info("Removed.")
except Exception:
self.log.exception("Exception while removing dir " + cache_dir_to_remove)
else:
self.log.info("Previous month cache dir not exists: " + cache_dir_to_remove)
thisMonthSubDir = today.strftime("%Y-%m")
cache_dir = os.path.join(os.path.normpath(os.path.expanduser("~/.epguide/cache")), thisMonthSubDir)
self.log.info("Cache dir: " + cache_dir)
self.cache = FileCache(cache_dir)
self.http = Http(cache = self.cache)
self.user_agent = "Mozilla/5.0 (Windows NT 6.1; WOW64; rv:18.0) Gecko/20100101 Firefox/18.0"
if(enable_debug):
httplib2.debuglevel = 255
def test_epgguide(self):
cache_dir = os.getcwd() + "/.epguide"
# os.makedirs(cache_dir)
print cache_dir
cache = FileCache(cache_dir)
h = Http(cache=cache)
user_agent = "Mozilla/5.0 (Windows NT 6.1; WOW64; rv:18.0) Gecko/20100101 Firefox/18.0"
httplib2.debuglevel = 255
resp, content = h.request(
"http://www.teleman.pl/program-tv/stacje/TVP-2",
headers={'user-agent': user_agent})
print "content:"
print content
print "response:"
print resp
print "fromcache:" + str(resp.fromcache)
print "status:" + str(resp.status)
#url = "http://www.teleman.pl/program-tv/stacje/TVP-2"
url = "http://www.teleman.pl/tv/Dr-House-7-152-885990"
resp, content = h.request(url, headers={'user-agent': user_agent})
print "fromcache:" + str(resp.fromcache)
print "status:" + str(resp.status)
safe = httplib2.safename(url)
print "safe:" + safe
cached_value = cache.get(url)
info, cached = cached_value.split('\r\n\r\n', 1)
print "===="
print content
print "===="
print cached
print "===="
self.assertEqual(content, cached)
def test_epgguide(self):
cache_dir = os.getcwd() + "/.epguide"
# os.makedirs(cache_dir)
print cache_dir
cache = FileCache(cache_dir)
h = Http(cache = cache)
user_agent = "Mozilla/5.0 (Windows NT 6.1; WOW64; rv:18.0) Gecko/20100101 Firefox/18.0"
httplib2.debuglevel = 255
resp, content = h.request("http://www.teleman.pl/program-tv/stacje/TVP-2", headers={'user-agent':user_agent})
print "content:"
print content
print "response:"
print resp
print "fromcache:" + str(resp.fromcache)
print "status:" + str(resp.status)
#url = "http://www.teleman.pl/program-tv/stacje/TVP-2"
url = "http://www.teleman.pl/tv/Dr-House-7-152-885990"
resp, content = h.request(url, headers={'user-agent':user_agent})
print "fromcache:" + str(resp.fromcache)
print "status:" + str(resp.status)
safe = httplib2.safename(url)
print "safe:" + safe
cached_value = cache.get(url)
info, cached = cached_value.split('\r\n\r\n', 1)
print "===="
print content
print "===="
print cached
print "===="
self.assertEqual(content, cached)
def do_GET(self):
from httplib2 import FileCache
logging.basicConfig(level=logging.DEBUG)
store = FileCache("cache4")
opener = urllib2.build_opener(CacheHandle(store))
urllib2.install_opener(opener)
response = opener.open(self.path)
(scm, netloc, path, params, query,
fragment) = urlparse.urlparse(self.path, 'http')
if scm not in ('http', 'ftp') or fragment or not netloc:
self.send_error(400, "bad url %s" % self.path)
return
soc = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
try:
if scm == 'http':
if self._connect_to(netloc, soc):
self.log_request()
soc.send("%s %s %s\r\n" %
(self.command,
urlparse.urlunparse(('', '', path, params, query,
'')), self.request_version))
self.headers['Connection'] = 'close'
del self.headers['Proxy-Connection']
for key_val in self.headers.items():
soc.send("%s: %s\r\n" % key_val)
soc.send("\r\n")
self._read_write(soc)
finally:
soc.close()
self.connection.close()
class HttpHelper(object):
'''
classdocs
'''
def __init__(self, enable_debug):
self.log = logging.getLogger(__name__)
today = datetime.date.today()
previousMonth = today - datetime.timedelta(days=31)
prevoiusMonthSubDir = previousMonth.strftime("%Y-%m")
cache_dir_to_remove = os.path.join(os.path.normpath(os.path.expanduser("~/.epguide/cache")), prevoiusMonthSubDir)
if os.path.exists(cache_dir_to_remove):
self.log.info("Removing previous month cache dir: " + cache_dir_to_remove)
try:
shutil.rmtree(cache_dir_to_remove)
self.log.info("Removed.")
except Exception:
self.log.exception("Exception while removing dir " + cache_dir_to_remove)
else:
self.log.info("Previous month cache dir not exists: " + cache_dir_to_remove)
thisMonthSubDir = today.strftime("%Y-%m")
cache_dir = os.path.join(os.path.normpath(os.path.expanduser("~/.epguide/cache")), thisMonthSubDir)
self.log.info("Cache dir: " + cache_dir)
self.cache = FileCache(cache_dir)
self.http = Http(cache = self.cache)
self.user_agent = "Mozilla/5.0 (Windows NT 6.1; WOW64; rv:18.0) Gecko/20100101 Firefox/18.0"
if(enable_debug):
httplib2.debuglevel = 255
def get(self, url, force_cache = False, charset = "UTF-8"):
if(force_cache):
cached_value = self.cache.get(url)
if cached_value:
try:
info, content = cached_value.split('\r\n\r\n', 1)
if(content):
return content.decode(charset)
else:
return self.get(url, False, charset)
except (IndexError, ValueError):
return self.get(url, False, charset)
else:
return self.get(url, False, charset)
else:
resp, content = self.http.request(url, headers={'user-agent':self.user_agent})
self.log.debug("result content: %s response: %s fromcache: %s status: %s)" % (content, resp, str(resp.fromcache), str(resp.status)))
c = content.decode(charset)
return c
def __init__(self, credentials):
super().__init__()
# do the authentication outside
assert credentials is not None and credentials.invalid is False, "Invalid or misssing credentials"
cache = FileCache(osJoin(gettempdir(), ".httpcache"),
safe=_SafeCacheName)
http = Http(cache, timeout=60)
http = credentials.authorize(http)
self.drive = build("drive", "v3", http=http)
_meta = self._meta = {
"case_insensitive":
False, # it will even let you have 2 identical filenames in the same directory!
"invalid_path_chars": ":", # not sure what else
"max_path_length": None, # don't know what the limit is
"max_sys_path_length": None, # there's no syspath
"network": True,
"read_only": False,
"supports_rename": False # since we don't have a syspath...
}
