def archive_files(yga, subdir=None):
logger = logging.getLogger(name="archive_files")
try:
if subdir:
file_json = yga.files(sfpath=subdir)
else:
file_json = yga.files()
except Exception:
logger.error("Couldn't access Files functionality for this group")
return
with open('fileinfo.json', 'wb') as f:
json.dump(file_json['dirEntries'],
codecs.getwriter('utf-8')(f),
ensure_ascii=False,
indent=4)
n = 0
sz = len(file_json['dirEntries'])
for path in file_json['dirEntries']:
n += 1
if path['type'] == 0:
# Regular file
name = html_unescape(path['fileName'])
logger.info("Fetching file '%s' (%d/%d)", name, n, sz)
with open(sanitise_file_name(name), 'wb') as f:
yga.download_file(path['downloadURL'], f)
elif path['type'] == 1:
# Directory
name = html_unescape(path['fileName'])
logger.info("Fetching directory '%s' (%d/%d)", name, n, sz)
with Mkchdir(name):
pathURI = unquote(path['pathURI'])
archive_files(yga, subdir=pathURI)
def archive_photos(yga):
logger = logging.getLogger(name="archive_photos")
try:
nb_albums = yga.albums(count=5)['total'] + 1
except Exception:
logger.error("Couldn't access Photos functionality for this group")
return
albums = yga.albums(count=nb_albums)
n = 0
with open('albums.json', 'wb') as f:
json.dump(albums['albums'],
codecs.getwriter('utf-8')(f),
ensure_ascii=False,
indent=4)
for a in albums['albums']:
n += 1
name = html_unescape(a['albumName'])
# Yahoo sometimes has an off-by-one error in the album count...
logger.info("Fetching album '%s' (%d/%d)", name, n, albums['total'])
folder = "%d-%s" % (a['albumId'], name)
with Mkchdir(folder):
photos = yga.albums(a['albumId'])
pages = int(photos['total'] / 100 + 1)
p = 0
for page in range(pages):
photos = yga.albums(a['albumId'], start=page * 100, count=100)
with open('photos-%d.json' % page, 'wb') as f:
json.dump(photos['photos'],
codecs.getwriter('utf-8')(f),
ensure_ascii=False,
indent=4)
for photo in photos['photos']:
p += 1
pname = html_unescape(photo['photoName'])
logger.info("Fetching photo '%s' (%d/%d)", pname, p,
photos['total'])
photoinfo = get_best_photoinfo(photo['photoInfo'])
fname = "%d-%s.jpg" % (photo['photoId'], pname)
with open(sanitise_file_name(fname), 'wb') as f:
for i in range(TRIES):
try:
yga.download_file(photoinfo['displayURL'], f)
break
except requests.exceptions.HTTPError as err:
logger.error(
"HTTP error (sleeping before retry, try %d: %s",
i, err)
time.sleep(HOLDOFF)
def archive_photos(yga):
logger = logging.getLogger(name="archive_photos")
try:
nb_albums = yga.albums(count=5)['total'] + 1
except Exception:
logger.error("Couldn't access Photos functionality for this group")
return
albums = yga.albums(count=nb_albums)
n = 0
with open('albums.json', 'wb') as f:
json.dump(albums['albums'],
codecs.getwriter('utf-8')(f),
ensure_ascii=False,
indent=4)
for a in albums['albums']:
n += 1
name = html_unescape(a['albumName'])
# Yahoo sometimes has an off-by-one error in the album count...
logger.info("Fetching album '%s' (%d/%d)", name, n, albums['total'])
folder = "%d-%s" % (a['albumId'], name)
with Mkchdir(folder):
photos = yga.albums(a['albumId'])
pages = int(photos['total'] / 100 + 1)
p = 0
for page in range(pages):
photos = yga.albums(a['albumId'], start=page * 100, count=100)
with open('photos-%d.json' % page, 'wb') as f:
json.dump(photos['photos'],
codecs.getwriter('utf-8')(f),
ensure_ascii=False,
indent=4)
for photo in photos['photos']:
p += 1
pname = html_unescape(photo['photoName'])
fname = sanitise_file_name("%d-%s.jpg" %
(photo['photoId'], pname))
if file_keep(fname, hacky_vars['file'], "photo: %s" %
(fname, )) is False:
logger.info("Fetching photo '%s' (%d/%d)", pname, p,
photos['total'])
with open(fname, 'wb') as f:
process_single_photo(yga, photo['photoInfo'], f)
set_mtime(fname, photo['creationDate'])
set_mtime(sanitise_folder_name(folder), a['modificationDate'])
async def work(worker_id: int):
i = 0
async with ClientSession(
cookie_jar=session.cookie_jar) as worker_session:
while True:
request_options = dict(
url=f'{origin}/owa/?ae=Dialog&t=AddressBook&ctx=1',
data={
**form_data, 'hidpg': worker_id + i * num_concurrent
},
headers={'Connection': 'Keep-Alive'})
async with worker_session.post(**request_options) as response:
data = (await response.content.read()).decode()
parsed_html_bs: BeautifulSoup = BeautifulSoup(
markup=html_unescape(data), features='html.parser')
page_names = {
td.text.rstrip()
for td in parsed_html_bs.select(
selector=
'table.lvw > tr:nth-child(n+4) > td:nth-child(3)')
if td.text.rstrip()
}
if len(page_names) == 0:
break
all_names.update(page_names)
i += 1
def cleanup_ywh_redirects_from_html(
ywh_domain: str,
html: str,
) -> str:
"""
Replace YesWeHack redirects with real URLs.
Args:
ywh_domain: a base domain of the YWH redirects
html: an html
Returns:
the cleaned html
"""
redirect_base_re = re.escape(f'{ywh_domain}/redirect?url=')
pattern = re.compile(f'"(https?://{redirect_base_re}[^ "]*)"')
redirect_urls = pattern.findall(html)
for redirect_url in redirect_urls:
real_url = _extract_real_url_from_redirect(
redirect_url=html_unescape(redirect_url), )
html = html.replace(
redirect_url,
html_escape(real_url or ''),
)
return html
def sanitize(text: str):
"""
Sanitize text removing html encoded parts, double spaces and so on
"""
if text is None:
return None
text = html_unescape(html_unescape(text))
for key, value in replaces.items():
text = text.replace(key, value)
text = re.sub(r'[ \t]+', ' ', text).strip()
return text
async def get_account_identity_from_ui(
session: ClientSession, origin: str,
**extra_request_keywords) -> Dict[str, str]:
"""
:param session: An authenticated `aiohttp.ClientSession`.
:param origin: The origin part of the URL of the OWA.
:return:
"""
# TODO: I would like something like Javascript Promises instead of this pattern.
json_results = None
def find_json_results(node, _) -> None:
if node.type == 'Property' and node.key.value == 'JsonResults':
nonlocal json_results
json_results = json_loads(node.value.value)
async with session.get(url=f'{origin}/ecp/PersonalSettings/HomePage.aspx',
**extra_request_keywords) as response:
data = (await response.content.read()).decode()
esprima_parse(code=pq(
html_unescape(data))('script[type="text/javascript"]')[-1].text,
delegate=find_json_results)
# TODO: Reconsider return value. Return `json_results` instead?
return json_results['Output'][0]
def get_last_n_filtered_tweets(twitter_oauth_session, screen_name, n):
base_URI = f'https://api.twitter.com/1.1/statuses/user_timeline.json?screen_name={screen_name}&count=200'
all_tweet_texts = []
last_id = 0
while (len(all_tweet_texts) < n):
current_tweets = json.loads(
twitter_oauth_session.request(
'GET',
f"{base_URI}&max_id={last_id}" if last_id else base_URI).text)
if current_tweets[-1]['id'] == last_id: break
for tweet in current_tweets:
#Skip replies
if current_tweets[0]['in_reply_to_status_id']: continue
#Skip retweets
if current_tweets[0]['retweeted']: continue
#Skip tweets with any users tagged in it
if bool(re_tagged_username_pattern.search(tweet['text'])): continue
#Redact all urls, twitter automatically converts urls to the t.co shortened format so we simply replace those
tweet['text'] = re_link_pattern.sub('[url redacted]',
tweet['text']).strip()
if tweet['text'] == '[url redacted]': continue
#If it passed all these checks, we add it to our all_tweet_texts list to eventually return
all_tweet_texts.append(html_unescape(tweet['text']))
last_id = current_tweets[-1]['id']
return all_tweet_texts[:n]
def get_item_group_from_feedparser(parser):
""" Retrieve all items from feedparser and return item group.
:type parser: 'feedparser.FeedParserDict'
:rtype: ItemGroup
"""
items = list()
logging.info('Loop for retrieving items.')
for item in parser.entries:
try:
text, img_links = format_description(item.description)
except AttributeError:
continue
if text:
new_item = Item(title=html_unescape(item.title),
date=item.published,
link=item.link,
text=text,
img_links=img_links)
items.append(new_item)
return ItemGroup(feed=parser.feed.title, items=items)
class Huya(Plugin):
_re_url = re.compile(r'https?://(?:www\.)?huya\.com/(?P<channel>[^/]+)')
_re_stream = re.compile(r'"stream"\s?:\s?"([^"]+)"')
_schema_data = validate.Schema(
{
# 'status': int,
# 'msg': validate.any(None, str),
'data': [{
'gameStreamInfoList': [{
'sCdnType': str,
'sStreamName': str,
'sFlvUrl': str,
'sFlvUrlSuffix': str,
'sFlvAntiCode': validate.all(str, validate.transform(lambda v: html_unescape(v))),
# 'sHlsUrl': str,
# 'sHlsUrlSuffix': str,
# 'sHlsAntiCode': validate.all(str, validate.transform(lambda v: html_unescape(v))),
validate.optional('iIsMultiStream'): int,
'iPCPriorityRate': int,
}]
}],
# 'vMultiStreamInfo': [{
# 'sDisplayName': str,
# 'iBitRate': int,
# }],
},
validate.get('data'),
validate.get(0),
validate.get('gameStreamInfoList'),
)
QUALITY_WEIGHTS = {}
@classmethod
def can_handle_url(cls, url):
return cls._re_url.match(url) is not None
@classmethod
def stream_weight(cls, key):
weight = cls.QUALITY_WEIGHTS.get(key)
if weight:
return weight, 'huya'
return Plugin.stream_weight(key)
def _get_streams(self):
res = self.session.http.get(self.url)
data = self._re_stream.search(res.text)
if not data:
return
data = parse_json(base64.b64decode(data.group(1)), schema=self._schema_data)
for info in data:
log.trace(f'{info!r}')
flv_url = f'{info["sFlvUrl"]}/{info["sStreamName"]}.{info["sFlvUrlSuffix"]}?{info["sFlvAntiCode"]}'
name = f'source_{info["sCdnType"].lower()}'
self.QUALITY_WEIGHTS[name] = info['iPCPriorityRate']
yield name, HTTPStream(self.session, flv_url)
log.debug(f'QUALITY_WEIGHTS: {self.QUALITY_WEIGHTS!r}')
def unescape(text):
"""
* Unescape HTML special chars\n
* @param string text\n
* @return string
"""
return html_unescape(text)
def _get_streams(self):
streamdata = None
if self.get_option("email"):
if self.login(self.get_option("email"), self.get_option("password")):
log.info("Logged in as {0}".format(self.get_option("email")))
self.save_cookies(lambda c: "steamMachineAuth" in c.name)
# Handle steam.tv URLs
if self.matches[1] is not None:
# extract the steam ID from the page
res = self.session.http.get(self.url)
for div in itertags(res.text, 'div'):
if div.attributes.get("id") == "webui_config":
broadcast_data = html_unescape(div.attributes.get("data-broadcast"))
steamid = parse_json(broadcast_data).get("steamid")
self.url = self._watch_broadcast_url + steamid
# extract the steam ID from the URL
steamid = self.match.group(1)
res = self.session.http.get(self.url) # get the page to set some cookies
sessionid = res.cookies.get('sessionid')
while streamdata is None or streamdata["success"] in ("waiting", "waiting_for_start"):
streamdata = self._get_broadcast_stream(steamid,
sessionid=sessionid)
if streamdata["success"] == "ready":
return DASHStream.parse_manifest(self.session, streamdata["url"])
elif streamdata["success"] == "unavailable":
log.error("This stream is currently unavailable")
return
else:
r = streamdata["retry"] / 1000.0
log.info("Waiting for stream, will retry again in {} seconds...".format(r))
time.sleep(r)
def get_home_title(url: str, home_title_list: List[str]) -> None:
"""Get homepage of the url and return it's title.
home_title_list will be used to return the thread result.
This function is invoked through a thread.
"""
# Todo: cache the result.
home_url = '://'.join(urlparse(url)[:2])
with requests_get(home_url,
stream=True,
headers=USER_AGENT_HEADER,
timeout=15) as r:
try:
check_response_headers(r)
except (
RequestException,
StatusCodeError,
ContentTypeError,
ContentLengthError,
):
return
content = next(r.iter_content(MAX_RESPONSE_LENGTH))
m = CHARSET(content)
html = content.decode(m[1].decode() if m else r.encoding)
m = TITLE_TAG(html)
title = html_unescape(m['result']) if m else None
home_title_list.append(title)
def format_tweet_as_text(tweet):
tweet = tweet.AsDict()
assert not any(key[0] == '_' and key[-1] == '_' for key in tweet.keys())
retweet = tweet.get('retweeted_status')
if retweet:
assert not any(key[0] == '_' and key[-1] == '_' for key in retweet.keys())
retweet['_first_username_'] = tweet['user']['name']
tweet = retweet
username = tweet['user']['name']
tweet['_username_'] = click.style(' ' + username + ' ',
fg='white', bg='black')
created_at = parse_time(tweet['created_at'])
tweet['_time_ago_'] = click.style(time_ago(created_at), fg='red')
# Decorate text
text = html_unescape(tweet['text'])
urls = tweet.get('urls', []) + tweet.get('media', [])
url_pairs = [(url['url'], url['expanded_url']) for url in urls]
text = expand_urls(text, url_pairs)
mentions = [mention['screen_name'] for mention in tweet.get('user_mentions', [])]
text = decorate_user_mentions(text, mentions, underline=True)
hashtags = [hashtag['text'] for hashtag in tweet.get('hashtags', [])]
text = decorate_hashtags(text, hashtags, underline=True)
tweet['_aligned_text_'] = align_text(text, margin='\t', skip_first_line=True)
return _text_formatter.format(FORMAT_RETWEET if retweet else FORMAT_TWEET,
created_at,
**tweet)
def find_iframe(self, url):
self.session.http.headers.update({"User-Agent": useragents.CHROME})
res = self.session.http.get(self.url)
for iframe_url in self.iframe_re.findall(res.text):
if "googletagmanager" not in iframe_url:
iframe_url = html_unescape(iframe_url)
return update_scheme(self.url, iframe_url)
def archive_files(yga, subdir=None):
logger = logging.getLogger(name="archive_files")
try:
if subdir:
file_json = yga.files(sfpath=subdir)
else:
file_json = yga.files()
except Exception:
logger.error("Couldn't access Files functionality for this group")
return
with open('fileinfo.json', 'wb') as f:
json.dump(file_json['dirEntries'],
codecs.getwriter('utf-8')(f),
ensure_ascii=False,
indent=4)
n = 0
sz = len(file_json['dirEntries'])
for path in file_json['dirEntries']:
n += 1
if path['type'] == 0:
# Regular file
name = html_unescape(path['fileName'])
new_name = sanitise_file_name("%d_%s" % (n, name))
if file_keep(new_name, hacky_vars['file'], ": %s" %
(new_name, )) is False:
logger.info("Fetching file '%s' as '%s' (%d/%d)", name,
new_name, n, sz)
with open(new_name, 'wb') as f:
try:
yga.download_file(path['downloadURL'], f)
except:
pass # Bad size exceptions can sometimes cause issues going from -f to -i.
set_mtime(new_name, path['createdTime'])
elif path['type'] == 1:
# Directory
name = html_unescape(path['fileName'])
new_name = "%d_%s" % (n, name)
logger.info("Fetching directory '%s' as '%s' (%d/%d)", name,
sanitise_folder_name(new_name), n, sz)
with Mkchdir(new_name): # (new_name sanitised again by Mkchdir)
pathURI = unquote(path['pathURI'])
archive_files(yga, subdir=pathURI)
set_mtime(sanitise_folder_name(new_name), path['createdTime'])
async def quote(message, *_, author: str = ""):
"""Affiche une citation au hazard ou pour un auteur donné"""
global dayly_timer, twenty_timer
now = datetime.now()
# Update dayly quotes everyday
if now - dayly_timer >= timedelta(days=1):
dayly_timer = now
await get_quotes()
# One quote allow each 20 minutes
if now - twenty_timer < timedelta(seconds=5):
await thenumberone.send_message(message.channel,
"Je manque d'inspiration...")
return
twenty_timer = now
# If an author is specified, select one randomly from that author
if author:
if author.casefold() in "chuck norris":
async with thenumberone.http.session.get(
"http://www.chucknorrisfacts.fr/api/get?data=tri:alea;nb:3"
) as resp:
assert resp.status == 200
quote = Quote(
"Chuck Norris",
html_unescape(
max(await resp.json(),
key=itemgetter("points"))["fact"]))
else:
author_quotes = list(
filter(
lambda quote: author.casefold() in quote.author.casefold(),
quotes))
if not author_quotes:
await thenumberone.send_message(
message.channel, "Je ne connais pas cet auteur.")
return
quote = choice(author_quotes)
# If there are dayly_quotes left use one
elif dayly_quotes:
quote = dayly_quotes.pop()
# Otherwise use our quotes
else:
quote = quotes.pop()
quotes.appendleft(quote)
await thenumberone.send_message(message.channel,
f"{quote.author}: *{quote.text}*")
def val(self):
'''Unescape HTML entities into corresponding Unicode characters.
Named (&), decimal (&), and hex (& and &) formats
are supported. Unknown entities are left intact.
As of Python 2.7 and Python 3.6 the following 252 named entities are
recognized and unescaped:
https://github.com/python/cpython/blob/2.7/Lib/htmlentitydefs.py
https://github.com/python/cpython/blob/3.6/Lib/html/entities.py
'''
return html_unescape(self.raw_val)
def extract_next_page_url(archive_page_raw_html, settings):
"""
extracts and returns next page URL from an HTML code if there is one...
"""
next_page_url_format_re = settings['NEXT_PAGE_URL_FORMAT_RE']
code_line = next_page_url_format_re.search(archive_page_raw_html)
if code_line is not None:
code_line = code_line.group(0)
code_line = settings['BEFORE_NEXT_PAGE_URL_RE'].sub(
settings['before_next_page_url_repl'], code_line)
code_line = settings['AFTER_NEXT_PAGE_URL_RE'].sub(
settings['after_next_page_url_repl'], code_line)
code_line = html_unescape(code_line)
return code_line
def extract_article_urls_from_page(archive_page_raw_html, settings):
"""
extracts and returns as a list the URLs belonging to articles from an HTML code
"""
urls = set()
article_url_format_re = settings['ARTICLE_URL_FORMAT_RE']
for code_line in article_url_format_re.findall(archive_page_raw_html):
code_line = settings['BEFORE_ARTICLE_URL_RE'].sub(
settings['before_article_url_repl'], code_line)
code_line = settings['AFTER_ARTICLE_URL_RE'].sub(
settings['after_article_url_repl'], code_line)
code_line = html_unescape(code_line)
urls.add(code_line)
return urls
async def get_trivias():
if TRIVIA_REQUEST_LOCK.locked():
await TRIVIA_REQUEST_LOCK
return
async with TRIVIA_REQUEST_LOCK:
async with SLASH_CLIENT.http.get(TRIVIA_URL,
params={
'amount': 100,
'category': 31
}) as response:
json = await response.json()
for trivia_data in json['results']:
trivia = (
html_unescape(trivia_data['question']),
html_unescape(trivia_data['correct_answer']),
[
html_unescape(element)
for element in trivia_data['incorrect_answers']
],
)
TRIVIA_QUEUE.append(trivia)
def get_title(self):
if self.title is None:
if not self.html_text:
self.html_text = self._res_text(self.url)
_og_title_re = re.compile(r'<meta\s*property="og:title"\s*content="(?P<title>[^<>]+)"\s*/?>')
_title_re = re.compile(r'<title[^<>]*>(?P<title>[^<>]+)</title>')
m = _og_title_re.search(self.html_text) or _title_re.search(self.html_text)
if m:
self.title = re.sub(r'[\s]+', ' ', m.group('title'))
self.title = re.sub(r'^\s*|\s*$', '', self.title)
self.title = html_unescape(self.title)
if self.title is None:
# fallback if there is no <title>
self.title = self.url
return self.title
def _get_streams(self):
"""
Find the streams for vk.com
:return:
"""
self.session.http.headers.update({'User-Agent': useragents.IPHONE_6})
# If this is a 'videos' catalog URL
# with an video ID in the GET request, get that instead
url = self.follow_vk_redirect(self.url)
m = self._url_re.match(url)
if not m:
log.error('URL is not compatible: {0}'.format(url))
return
video_id = m.group('video_id')
log.debug('video ID: {0}'.format(video_id))
params = {
'act': 'show_inline',
'al': '1',
'video': video_id,
}
res = self.session.http.post(self.API_URL, params=params)
for _i in itertags(res.text, 'iframe'):
if _i.attributes.get('src'):
iframe_url = update_scheme(self.url, _i.attributes['src'])
log.debug('Found iframe: {0}'.format(iframe_url))
yield from self.session.streams(iframe_url).items()
for _i in itertags(res.text.replace('\\', ''), 'source'):
if _i.attributes.get('type') == 'application/vnd.apple.mpegurl':
video_url = html_unescape(_i.attributes['src'])
streams = HLSStream.parse_variant_playlist(
self.session, video_url)
if not streams:
yield 'live', HLSStream(self.session, video_url)
else:
yield from streams.items()
elif _i.attributes.get('type') == 'video/mp4':
q = 'vod'
video_url = _i.attributes['src']
m = self._vod_quality_re.search(video_url)
if m:
q = '{0}p'.format(m.group(1))
yield q, HTTPStream(self.session, video_url)
def url2dict(url: str) -> Dict[str, Any]:
"""Get url and return the result as a dictionary."""
d = defaultdict(lambda: None)
# Creating a thread to request homepage title in background
home_title_list = [] # A mutable variable used to get the thread result
home_title_thread = Thread(target=get_home_title,
args=(url, home_title_list))
home_title_thread.start()
html = get_html(url)
d['url'] = find_url(html, url)
m = TITLE_TAG(html)
html_title = html_unescape(m['result']) if m else None
if html_title:
d['html_title'] = html_title
# d['html_title'] is used in waybackmechine.py.
authors = find_authors(html)
if authors:
d['authors'] = authors
d['issn'] = find_issn(html)
d['pmid'] = find_pmid(html)
d['doi'] = find_doi(html)
d['volume'] = find_volume(html)
d['issue'] = find_issue(html)
d['page'] = find_pages(html)
d['journal'] = find_journal(html)
if d['journal']:
d['cite_type'] = 'journal'
else:
d['cite_type'] = 'web'
d['website'] = find_site_name(html, html_title, url, authors,
home_title_list, home_title_thread)
d['title'] = find_title(html, html_title, url, authors, home_title_list,
home_title_thread)
date = find_date(html, url)
if date:
d['date'] = date
d['year'] = str(date.year)
lang_match = LANG_SEARCH(html)
if lang_match is not None:
d['language'] = lang_match[1]
else:
d['language'] = classify(html)[0]
return d
def test_string_markdown_link(self):
# markdown2 and markdown escape the email address
try:
from html import unescape as html_unescape
except ImportError:
from HTMLParser import HTMLParser
html_unescape = HTMLParser().unescape
p = StringHTMLProperty(self.client, 'test', '1', None, 'test',
u2s(u'A link <*****@*****.**>'))
m = html_unescape(p.markdown().strip())
m = self.mangleMarkdown2(m)
self.assertEqual(
m,
u2s(u'<p>A link <a href="mailto:[email protected]">[email protected]</a></p>'
))
def _parse_streams(self, res):
_found_stream_url = False
for meta in itertags(res.text, "meta"):
if meta.attributes.get("property") == "og:video:url":
stream_url = html_unescape(meta.attributes.get("content"))
if ".mpd" in stream_url:
for s in DASHStream.parse_manifest(self.session,
stream_url).items():
yield s
_found_stream_url = True
elif ".mp4" in stream_url:
yield "vod", HTTPStream(self.session, stream_url)
_found_stream_url = True
break
else:
log.debug("No meta og:video:url")
if _found_stream_url:
return
for match in self._src_re.finditer(res.text):
stream_url = match.group("url")
if "\\/" in stream_url:
# if the URL is json encoded, decode it
stream_url = parse_json("\"{}\"".format(stream_url))
if ".mpd" in stream_url:
yield from DASHStream.parse_manifest(self.session,
stream_url).items()
elif ".mp4" in stream_url:
yield match.group(1), HTTPStream(self.session, stream_url)
else:
log.debug("Non-dash/mp4 stream: {0}".format(stream_url))
match = self._dash_manifest_re.search(res.text)
if match:
# facebook replaces "<" characters with the substring "\\x3C"
manifest = match.group("manifest").replace("\\/", "/")
manifest = bytes(unquote_plus(manifest),
"utf-8").decode("unicode_escape")
# Ignore unsupported manifests until DASH SegmentBase support is implemented
if "SegmentBase" in manifest:
log.error("Skipped DASH manifest with SegmentBase streams")
else:
yield from DASHStream.parse_manifest(self.session,
manifest).items()
def parse(self, tree):
xpath = '//h1[contains(@class, "ArticleTitle")]/text()'
self.update_item('title', tree.xpath(xpath))
xpath = '//*[@class="authors__list"]//*[@class="authors__name"]/text()'
authors = tree.xpath(xpath)
authors = [html_unescape(author) for author in authors]
self.update_item('authors', authors)
xpath = '//*[contains(@class, "KeywordGroup")]//*[contains(@class, "Keyword")]/text()'
keywords = tree.xpath(xpath)
self.update_item('keywords', keywords)
xpath = '//*[@id="article-actions"]//*[contains(@class, "download-article")]/a[1]/@href'
download_link = tree.xpath(xpath)
download_link = self.urljoin(download_link[0]) if download_link else ''
self.update_item('download_link', download_link)
xpath = '//*[@id="doi-url"]/text()'
doi_link = tree.xpath(xpath)
doi_link = doi_link[0] if doi_link else ''
self.update_item('doi_link', doi_link)
def find_title(
html: str,
html_title: str,
url: str,
authors: List[Tuple[str, str]],
home_title: List[str],
thread: Thread,
) -> Optional[str]:
"""Return (title_string, where_info)."""
m = TITLE_SEARCH(html)
if m:
return parse_title(
html_unescape(m['result']), url, authors, home_title, thread,
)[1]
elif html_title:
return parse_title(html_title, url, authors, home_title, thread)[1]
else:
return None
def stackexchange(auth_token):
user = requests.get(
'https://api.stackexchange.com/2.2/me',
params={
'site': 'codegolf',
'filter':
'!)RwcIFN1JaCrhVpgyYeR_oO*', # Constant obtained from SE API explorer
'access_token': auth_token,
'key': auth.get('stackexchange.com').get('key')
}).json().get('items')[0]
display_name = html_unescape(user.get('display_name'))
return user.get('user_id'), {
'name': display_name,
'avatar': user.get('profile_image'),
'email': None,
'identifier': display_name
}
def _get_streams(self):
p = urlparse(self.url)
if "ott.streann.com" != p.netloc:
self._domain = p.netloc
res = self.session.http.get(self.url)
for iframe in itertags(res.text, "iframe"):
iframe_url = html_unescape(iframe.attributes.get("src"))
if "ott.streann.com" in iframe_url:
self.url = iframe_url
break
else:
log.error("Could not find 'ott.streann.com' iframe")
return
if not self._domain and self.get_option("url"):
self._domain = urlparse(self.get_option("url")).netloc
if self._domain is None:
log.error("Missing source URL use --streann-url")
return
self.session.http.headers.update({"Referer": self.url})
# Get the query string
encrypted_data = urlparse(self.url).query
data = base64.b64decode(encrypted_data)
# and decrypt it
passphrase = self.passphrase()
if passphrase:
log.debug("Found passphrase")
params = decrypt_openssl(data, passphrase)
config = parse_qsd(params.decode("utf8"))
log.trace(f"config: {config!r}")
token = self.get_token(**config)
if not token:
return
hls_url = self.stream_url.format(time=self.time,
deviceId=self.device_id,
token=token,
**config)
log.debug("URL={0}".format(hls_url))
return HLSStream.parse_variant_playlist(
self.session, hls_url, acceptable_status=(200, 403, 404, 500))
def url2dict(url: str) -> Dict[str, Any]:
"""Get url and return the result as a dictionary."""
d = defaultdict(lambda: None)
# Creating a thread to request homepage title in background
home_title_list = [] # A mutable variable used to get the thread result
home_title_thread = Thread(
target=get_home_title, args=(url, home_title_list))
home_title_thread.start()
html = get_html(url)
d['url'] = find_url(html, url)
m = TITLE_TAG(html)
html_title = html_unescape(m['result']) if m else None
if html_title:
d['html_title'] = html_title
# d['html_title'] is used in waybackmechine.py.
authors = find_authors(html)
if authors:
d['authors'] = authors
d['issn'] = find_issn(html)
d['pmid'] = find_pmid(html)
d['doi'] = find_doi(html)
d['volume'] = find_volume(html)
d['issue'] = find_issue(html)
d['page'] = find_pages(html)
d['journal'] = find_journal(html)
if d['journal']:
d['cite_type'] = 'journal'
else:
d['cite_type'] = 'web'
d['website'] = find_site_name(
html, html_title, url, authors, home_title_list, home_title_thread)
d['title'] = find_title(
html, html_title, url, authors, home_title_list, home_title_thread)
date = find_date(html, url)
if date:
d['date'] = date
d['year'] = str(date.year)
d['language'] = classify(html)[0]
return d
def _get_streams(self):
if "cdn.bg" in urlparse(self.url).netloc:
iframe_url = self.url
h = self.session.get_option("http-headers")
if h and h.get("Referer"):
_referer = h.get("Referer")
else:
log.error(
"Missing Referer for iframe URL, use --http-header \"Referer=URL\" "
)
return
else:
_referer = self.url
res = self.session.http.get(self.url)
m = self._re_frame.search(res.text)
if m:
iframe_url = m.group(1)
else:
for iframe in itertags(res.text, "iframe"):
iframe_url = iframe.attributes.get("src")
if iframe_url and "cdn.bg" in iframe_url:
iframe_url = update_scheme("https://",
html_unescape(iframe_url),
force=False)
break
else:
return
log.debug(f"Found iframe: {iframe_url}")
res = self.session.http.get(iframe_url, headers={"Referer": _referer})
stream_url = self.stream_schema.validate(res.text)
if "geoblock" in stream_url:
log.error("Geo-restricted content")
return
return HLSStream.parse_variant_playlist(
self.session,
update_scheme(iframe_url, stream_url),
headers={"Referer": "https://i.cdn.bg/"},
)
def get_home_title(url: str, home_title_list: List[str]) -> None:
"""Get homepage of the url and return it's title.
home_title_list will be used to return the thread result.
This function is invoked through a thread.
"""
# Todo: cache the result.
home_url = '://'.join(urlparse(url)[:2])
with request(
home_url, spoof=True, stream=True
) as r:
try:
check_response_headers(r)
except (
RequestException, StatusCodeError,
ContentTypeError, ContentLengthError,
):
return
content = next(r.iter_content(MAX_RESPONSE_LENGTH))
m = CHARSET(content)
html = content.decode(m[1].decode() if m else r.encoding)
m = TITLE_TAG(html)
title = html_unescape(m['result']) if m else None
home_title_list.append(title)
def render_template(template, **kw):
return html_unescape(jin_env.get_template(template).render(**kw))
def unescape(string):
if string is None:
return None
return html_unescape(string)
