1
0
mirror of https://github.com/l1ving/youtube-dl synced 2024-11-23 10:12:51 +08:00
youtube-dl/youtube_dl/extractor/facebook.py

133 lines
5.5 KiB
Python
Raw Normal View History

2013-06-24 02:59:45 +08:00
import json
import re
import socket
from .common import InfoExtractor
from ..utils import (
compat_http_client,
compat_str,
compat_urllib_error,
compat_urllib_parse,
compat_urllib_request,
ExtractorError,
)
class FacebookIE(InfoExtractor):
"""Information Extractor for Facebook"""
_VALID_URL = r'^(?:https?://)?(?:\w+\.)?facebook\.com/(?:video/video|photo)\.php\?(?:.*?)v=(?P<ID>\d+)(?:.*)'
_LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
_CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
2013-06-24 02:59:45 +08:00
_NETRC_MACHINE = 'facebook'
IE_NAME = u'facebook'
2013-06-28 02:46:46 +08:00
_TEST = {
u'url': u'https://www.facebook.com/photo.php?v=120708114770723',
u'file': u'120708114770723.mp4',
u'md5': u'48975a41ccc4b7a581abd68651c1a5a8',
u'info_dict': {
u"duration": 279,
u"title": u"PEOPLE ARE AWESOME 2013"
}
}
2013-06-24 02:59:45 +08:00
def report_login(self):
"""Report attempt to log in."""
self.to_screen(u'Logging in')
def _login(self):
(useremail, password) = self._get_login_info()
2013-06-24 02:59:45 +08:00
if useremail is None:
return
login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
login_page_req.add_header('Cookie', 'locale=en_US')
self.report_login()
login_page = self._download_webpage(login_page_req, None, note=False,
errnote=u'Unable to download login page')
lsd = self._search_regex(r'"lsd":"(\w*?)"', login_page, u'lsd')
lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, u'lgnrnd')
2013-06-24 02:59:45 +08:00
login_form = {
'email': useremail,
'pass': password,
'lsd': lsd,
'lgnrnd': lgnrnd,
'next': 'http://facebook.com/home.php',
'default_persistent': '0',
'legacy_return': '1',
'timezone': '-60',
'trynum': '1',
2013-06-24 02:59:45 +08:00
}
request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
request.add_header('Content-Type', 'application/x-www-form-urlencoded')
2013-06-24 02:59:45 +08:00
try:
login_results = compat_urllib_request.urlopen(request).read()
if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
self._downloader.report_warning(u'unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
return
check_form = {
'fb_dtsg': self._search_regex(r'"fb_dtsg":"(.*?)"', login_results, u'fb_dtsg'),
'nh': self._search_regex(r'name="nh" value="(\w*?)"', login_results, u'nh'),
'name_action_selected': 'dont_save',
'submit[Continue]': self._search_regex(r'<input value="(.*?)" name="submit\[Continue\]"', login_results, u'continue'),
}
check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, compat_urllib_parse.urlencode(check_form))
check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
check_response = compat_urllib_request.urlopen(check_req).read()
if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
self._downloader.report_warning(u'Unable to confirm login, you have to login in your brower and authorize the login.')
2013-06-24 02:59:45 +08:00
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
return
def _real_initialize(self):
self._login()
2013-06-24 02:59:45 +08:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
video_id = mobj.group('ID')
url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
webpage = self._download_webpage(url, video_id)
BEFORE = '{swf.addParam(param[0], param[1]);});\n'
AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
if not m:
m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
if m_msg is not None:
raise ExtractorError(
u'The video is not available, Facebook said: "%s"' % m_msg.group(1),
expected=True)
else:
raise ExtractorError(u'Cannot parse data')
2013-06-24 02:59:45 +08:00
data = dict(json.loads(m.group(1)))
params_raw = compat_urllib_parse.unquote(data['params'])
params = json.loads(params_raw)
video_data = params['video_data'][0]
video_url = video_data.get('hd_src')
if not video_url:
video_url = video_data['sd_src']
if not video_url:
raise ExtractorError(u'Cannot find video URL')
video_duration = int(video_data['video_duration'])
thumbnail = video_data['thumbnail_src']
video_title = self._html_search_regex(
r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, u'title')
2013-06-24 02:59:45 +08:00
info = {
'id': video_id,
'title': video_title,
'url': video_url,
'ext': 'mp4',
'duration': video_duration,
'thumbnail': thumbnail,
}
return [info]