def url(self):
if self.parent_handle is not None:
return urljoin(BASE_FACEBOOK_URL,
'/%s/posts/%s' % (self.parent_handle, self.id))
if self.parent_id is not None:
return urljoin(
BASE_FACEBOOK_URL, '/permalink.php?story_fbid=%s&id=%s' %
(self.id, self.parent_id))
if self.group_id is not None:
return urljoin(
BASE_FACEBOOK_URL,
'/groups/%s/permalink/%s' % (self.group_id, self.id))
def infer_redirection(url):
"""
Function returning the url that the given url will redirect to. This is done
by finding obvious hints in the GET parameters that the given url is in
fact a redirection.
Args:
url (string): Target url.
Returns:
string: Redirected url or the original url if nothing was found.
"""
redirection_split = REDIRECTION_DOMAINS_RE.split(url, 1)
if len(redirection_split) > 1:
return infer_redirection('https://' + redirection_split[1])
obvious_redirect_match = re.search(OBVIOUS_REDIRECTS_RE, url)
if obvious_redirect_match is not None:
target = unquote(obvious_redirect_match.group(1))
if target.startswith('http://') or target.startswith('https://'):
return target
if target.startswith('/'):
return urljoin(url, target)
return url
def url(self):
if self.group_id:
return urljoin(
BASE_FACEBOOK_URL,
'/photo.php?fbid=%s&set=g.%s' % (self.id, self.group_id))
if self.parent_id:
return urljoin(
BASE_FACEBOOK_URL,
'/%s/a.%s/%s' % (self.parent_id, self.album_id, self.id))
if self.parent_handle:
return urljoin(
BASE_FACEBOOK_URL,
'/%s/a.%s/%s' % (self.parent_handle, self.album_id, self.id))
return urljoin(BASE_FACEBOOK_URL, '/photo.php?fbid=%s' % self.id)
def parse_facebook_url(url, allow_relative_urls=False):
# Allowing relative urls scraped from facebook?
if (allow_relative_urls and not url.startswith('http://')
and not url.startswith('https://') and 'facebook.' not in url):
url = urljoin(BASE_FACEBOOK_URL, url)
else:
if not is_facebook_url(url):
return None
splitted = safe_urlsplit(url)
if not splitted.path or splitted.path == '/':
return None
# Obvious post path
if '/posts/' in splitted.path:
parts = urlpathsplit(splitted.path)
parent_id_or_handle = parts[0]
if NUMERIC_ID_RE.match(parent_id_or_handle):
return FacebookPost(parts[2], parent_id=parent_id_or_handle)
return FacebookPost(parts[2], parent_handle=parent_id_or_handle)
# Ye olded permalink path
if splitted.query and '/permalink.php' in splitted.path:
query = parse_qs(splitted.query)
return FacebookPost(query['story_fbid'][0], parent_id=query['id'][0])
# Group permalink path
if '/groups/' in splitted.path and '/permalink/' in splitted.path:
parts = urlpathsplit(splitted.path)
return FacebookPost(parts[3], group_id=parts[1])
# Profile path
if splitted.path == '/profile.php':
query = parse_qs(splitted.query)
user_id = query['id'][0]
return FacebookUser(user_id)
# People path
if splitted.path.startswith('/people'):
parts = urlpathsplit(splitted.path)
user_id = parts[2]
return FacebookUser(user_id)
# Handle path
if splitted.path:
parts = urlpathsplit(splitted.path)
if not parts[0].endswith('.php'):
return FacebookHandle(parts[0])
return None
def infer_redirection(url, recursive=True):
"""
Function returning the url that the given url will redirect to. This is done
by finding obvious hints in the GET parameters that the given url is in
fact a redirection.
Args:
url (string): Target url.
recursive (bool): Whether to apply the function recursively until
no redirection can be inferred. Defaults to `True`.
Returns:
string: Redirected url or the original url if nothing was found.
"""
redirection_split = REDIRECTION_DOMAINS_RE.split(url, 1)
target = None
if len(redirection_split) > 1:
target = 'https://' + redirection_split[1]
else:
obvious_redirect_match = re.search(OBVIOUS_REDIRECTS_RE, url)
if obvious_redirect_match is not None:
potential_target = unquote(obvious_redirect_match.group(1))
if potential_target.startswith(
'http://') or potential_target.startswith('https://'):
target = potential_target
if potential_target.startswith('/'):
target = urljoin(url, potential_target)
if target is None:
return url
if recursive:
return infer_redirection(target, recursive=True)
return target
def url(self):
return urljoin(BASE_FACEBOOK_URL, '/%s' % self.handle)
def url(self):
if self.handle is None:
return urljoin(BASE_FACEBOOK_URL, '/profile.php?id=%s' % self.id)
return urljoin(BASE_FACEBOOK_URL, '/%s' % self.handle)
def parse_facebook_url(url, allow_relative_urls=False):
# Allowing relative urls scraped from facebook?
if (allow_relative_urls and not url.startswith('http://')
and not url.startswith('https://') and 'facebook.' not in url):
url = urljoin(BASE_FACEBOOK_URL, url)
else:
if not is_facebook_url(url):
return None
splitted = safe_urlsplit(url)
if not splitted.path or splitted.path == '/':
return None
# Videos
if '/watch' in splitted.path:
query = parse_qs(splitted.query)
if 'v' not in query:
return None
video_id = query['v'][0]
return FacebookVideo(video_id)
if '/videos/' in splitted.path:
parts = urlpathsplit(splitted.path)
return FacebookVideo(parts[2], parent_id=parts[0])
# Photos
if splitted.query and (splitted.path.endswith('/photo.php')
or splitted.path.endswith('/photo')):
query = parse_qs(splitted.query)
if 'fbid' not in query:
return None
group_id = None
album_id = None
if 'set' in query:
sets = query['set']
group_id = next((s for s in sets if s.startswith('g.')), None)
if group_id:
group_id = group_id.split('g.', 1)[1]
album_id = next((s for s in sets if s.startswith('a.')), None)
if album_id:
album_id = album_id.split('a.', 1)[1]
return FacebookPhoto(query['fbid'][0],
group_id=group_id,
album_id=album_id)
if '/photos/' in splitted.path:
parts = urlpathsplit(splitted.path)
parent_id_or_handle = parts[0]
album_id = parts[2].replace('a.', '')
photo_id = parts[3]
if is_facebook_id(parent_id_or_handle):
return FacebookPhoto(photo_id,
album_id=album_id,
parent_id=parent_id_or_handle)
return FacebookPhoto(photo_id,
album_id=album_id,
parent_handle=parent_id_or_handle)
# Obvious post path
if '/posts/' in splitted.path:
parts = urlpathsplit(splitted.path)
parent_id_or_handle = parts[0]
if NUMERIC_ID_RE.match(parent_id_or_handle):
return FacebookPost(parts[2], parent_id=parent_id_or_handle)
return FacebookPost(parts[2], parent_handle=parent_id_or_handle)
# Ye olded permalink path
if splitted.query and ('/permalink.php' in splitted.path
or '/story.php' in splitted.path):
query = parse_qs(splitted.query)
return FacebookPost(query['story_fbid'][0], parent_id=query['id'][0])
# Group permalink path
if '/groups/' in splitted.path:
parts = urlpathsplit(splitted.path)
if '/permalink/' in splitted.path:
if is_facebook_id(parts[1]):
return FacebookPost(parts[3], group_id=parts[1])
return FacebookPost(parts[3], group_handle=parts[1])
if is_facebook_id(parts[1]):
return FacebookGroup(id=parts[1])
return FacebookGroup(handle=parts[1])
# Profile path
if splitted.path == '/profile.php':
query = parse_qs(splitted.query)
user_id = query['id'][0]
return FacebookUser(user_id)
# People path
if splitted.path.startswith('/people'):
parts = urlpathsplit(splitted.path)
user_id = parts[2]
return FacebookUser(user_id)
# Handle path
if splitted.path:
parts = urlpathsplit(splitted.path)
if not parts[0].endswith('.php'):
return FacebookHandle(parts[0])
return None
def url(self):
if self.parent_id is None:
return urljoin(BASE_FACEBOOK_URL, '/watch/?v=%s' % self.id)
return urljoin(BASE_FACEBOOK_URL,
'/%s/videos/%s' % (self.parent_id, self.id))
def url(self):
if self.handle is not None:
return urljoin(BASE_FACEBOOK_URL, 'groups/%s' % self.handle)
return urljoin(BASE_FACEBOOK_URL, 'groups/%s' % self.id)
