1 from __future__
import unicode_literals
7 from .common
import InfoExtractor
9 compat_etree_fromstring
,
12 compat_urllib_parse_unquote
,
13 compat_urllib_parse_unquote_plus
,
26 class FacebookIE(InfoExtractor
):
30 (?:\w+\.)?facebook\.com/
39 )\?(?:.*?)(?:v|video_id|story_fbid)=|
40 [^/]+/videos/(?:[^/]+/)?|
42 groups/[^/]+/permalink/
48 _LOGIN_URL
= 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
49 _CHECKPOINT_URL
= 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
50 _NETRC_MACHINE
= 'facebook'
53 _CHROME_USER_AGENT
= 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/48.0.2564.97 Safari/537.36'
55 _VIDEO_PAGE_TEMPLATE
= 'https://www.facebook.com/video/video.php?v=%s'
58 'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf',
59 'md5': '6a40d33c0eccbb1af76cf0485a052659',
61 'id': '637842556329505',
63 'title': 're:Did you know Kei Nishikori is the first Asian man to ever reach a Grand Slam',
64 'uploader': 'Tennis on Facebook',
67 'note': 'Video without discernible title',
68 'url': 'https://www.facebook.com/video.php?v=274175099429670',
70 'id': '274175099429670',
72 'title': 'Facebook video #274175099429670',
73 'uploader': 'Asif Nawab Butt',
75 'expected_warnings': [
79 'note': 'Video with DASH manifest',
80 'url': 'https://www.facebook.com/video.php?v=957955867617029',
81 'md5': '54706e4db4f5ad58fbad82dde1f1213f',
83 'id': '957955867617029',
85 'title': 'When you post epic content on instagram.com/433 8 million followers, this is ...',
86 'uploader': 'Demy de Zeeuw',
89 'url': 'https://www.facebook.com/maxlayn/posts/10153807558977570',
90 'md5': '037b1fa7f3c2d02b7a0d7bc16031ecc6',
92 'id': '544765982287235',
94 'title': '"What are you doing running in the snow?"',
95 'uploader': 'FailArmy',
98 'url': 'https://m.facebook.com/story.php?story_fbid=1035862816472149&id=116132035111903',
99 'md5': '1deb90b6ac27f7efcf6d747c8a27f5e3',
101 'id': '1035862816472149',
103 'title': 'What the Flock Is Going On In New Zealand Credit: ViralHog',
104 'uploader': 'S. Saint',
107 'note': 'swf params escaped',
108 'url': 'https://www.facebook.com/barackobama/posts/10153664894881749',
109 'md5': '97ba073838964d12c70566e0085c2b91',
111 'id': '10153664894881749',
113 'title': 'Facebook video #10153664894881749',
116 'url': 'https://www.facebook.com/video.php?v=10204634152394104',
117 'only_matching': True,
119 'url': 'https://www.facebook.com/amogood/videos/1618742068337349/?fref=nf',
120 'only_matching': True,
122 'url': 'https://www.facebook.com/ChristyClarkForBC/videos/vb.22819070941/10153870694020942/?type=2&theater',
123 'only_matching': True,
125 'url': 'facebook:544765982287235',
126 'only_matching': True,
128 'url': 'https://www.facebook.com/groups/164828000315060/permalink/764967300301124/',
129 'only_matching': True,
133 def _extract_url(webpage
):
135 r
'<iframe[^>]+?src=(["\'])(?P
<url
>https
://www\
.facebook\
.com
/video
/embed
.+?
)\
1', webpage)
137 return mobj.group('url
')
140 # see https://developers.facebook.com/docs/plugins/embedded-video-player
141 mobj = re.search(r'''(?x)<div[^>]+
142 class=(?P<q1>[\'"])[^\'"]*\bfb-(?:video|post)\b[^\'"]*(?P=q1)[^>]+
143 data-href=(?P<q2>[\'"])(?P<url>(?:https?:)?//(?:www\.)?facebook.com/.+?)(?P=q2)''', webpage)
145 return mobj.group('url
')
148 (useremail, password) = self._get_login_info()
149 if useremail is None:
152 login_page_req = sanitized_Request(self._LOGIN_URL)
153 self._set_cookie('facebook
.com
', 'locale
', 'en_US
')
154 login_page = self._download_webpage(login_page_req, None,
155 note='Downloading login page
',
156 errnote='Unable to download login page
')
157 lsd = self._search_regex(
158 r'<input type="hidden" name
="lsd" value
="([^"]*)"',
160 lgnrnd = self._search_regex(r'name="lgnrnd
" value="([^
"]*?)"', login_page, 'lgnrnd
')
167 'next
': 'http
://facebook
.com
/home
.php
',
168 'default_persistent
': '0',
169 'legacy_return
': '1',
173 request = sanitized_Request(self._LOGIN_URL, urlencode_postdata(login_form))
174 request.add_header('Content
-Type
', 'application
/x
-www
-form
-urlencoded
')
176 login_results = self._download_webpage(request, None,
177 note='Logging
in', errnote='unable to fetch login page
')
178 if re.search(r'<form(.*)name
="login"(.*)</form
>', login_results) is not None:
179 error = self._html_search_regex(
180 r'(?s
)<div
[^
>]+class=(["\']).*?login_error_box.*?\1[^>]*><div[^>]*>.*?</div><div[^>]*>(?P<error>.+?)</div>',
181 login_results, 'login error', default=None, group='error')
183 raise ExtractorError('Unable to login: %s' % error, expected=True)
184 self._downloader.report_warning('unable to log in: bad username/password, or exceeded login rate limit (~3/min). Check credentials or wait.')
187 fb_dtsg = self._search_regex(
188 r'name="fb_dtsg
" value="(.+?
)"', login_results, 'fb_dtsg', default=None)
189 h = self._search_regex(
190 r'name="h
"\s+(?:\w+="[^
"]+"\s
+)*?value
="([^"]+)"', login_results, 'h', default=None)
192 if not fb_dtsg or not h:
198 'name_action_selected': 'dont_save',
200 check_req = sanitized_Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))
201 check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
202 check_response = self._download_webpage(check_req, None,
203 note='Confirming login')
204 if re.search(r'id="checkpointSubmitButton
"', check_response) is not None:
205 self._downloader.report_warning('Unable to confirm login, you have to login in your browser and authorize the login.')
206 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
207 self._downloader.report_warning('unable to log in: %s' % error_to_compat_str(err))
210 def _real_initialize(self):
213 def _extract_from_url(self, url, video_id, fatal_if_no_video=True):
214 req = sanitized_Request(url)
215 req.add_header('User-Agent', self._CHROME_USER_AGENT)
216 webpage = self._download_webpage(req, video_id)
220 BEFORE = '{swf.addParam(param[0], param[1]);});'
221 AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
222 PATTERN = re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER)
224 for m in re.findall(PATTERN, webpage):
225 swf_params = m.replace('\\\\', '\\').replace('\\"', '"')
226 data = dict(json.loads(swf_params))
227 params_raw = compat_urllib_parse_unquote(data['params'])
228 video_data_candidate = json.loads(params_raw)['video_data']
229 for _, f in video_data_candidate.items():
232 if isinstance(f, dict):
234 if not isinstance(f, list):
236 if f[0].get('video_id') == video_id:
237 video_data = video_data_candidate
242 def video_data_list2dict(video_data):
244 for item in video_data:
245 format_id = item['stream_type']
246 ret.setdefault(format_id, []).append(item)
250 server_js_data = self._parse_json(self._search_regex(
251 r'handleServerJS\(({.+})\);', webpage, 'server js data', default='{}'), video_id)
252 for item in server_js_data.get('instances', []):
253 if item[1][0] == 'VideoConfig':
254 video_data = video_data_list2dict(item[2][0]['videoData'])
258 if not fatal_if_no_video:
259 return webpage, False
260 m_msg = re.search(r'class="[^
"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
261 if m_msg is not None:
262 raise ExtractorError(
263 'The video is not available, Facebook said: "%s"' % m_msg.group(1),
266 raise ExtractorError('Cannot parse data')
269 for format_id, f in video_data.items():
270 if f and isinstance(f, dict):
272 if not f or not isinstance(f, list):
274 for quality in ('sd', 'hd'):
275 for src_type in ('src', 'src_no_ratelimit'):
276 src = f[0].get('%s_%s' % (quality, src_type))
278 preference = -10 if format_id == 'progressive' else 0
282 'format_id': '%s_%s_%s' % (format_id, quality, src_type),
284 'preference': preference,
286 dash_manifest = f[0].get('dash_manifest')
288 formats.extend(self._parse_mpd_formats(
289 compat_etree_fromstring(compat_urllib_parse_unquote_plus(dash_manifest))))
291 raise ExtractorError('Cannot find video formats')
293 self._sort_formats(formats)
295 video_title = self._html_search_regex(
296 r'<h2\s+[^>]*class="uiHeaderTitle
"[^>]*>([^<]*)</h2>', webpage, 'title',
299 video_title = self._html_search_regex(
300 r'(?s)<span class="fbPhotosPhotoCaption
".*?id="fbPhotoPageCaption
"><span class="hasCaption
">(.*?)</span>',
301 webpage, 'alternative title', default=None)
302 video_title = limit_length(video_title, 80)
304 video_title = 'Facebook video #%s' % video_id
305 uploader = clean_html(get_element_by_id('fbPhotoPageAuthorName', webpage))
309 'title': video_title,
311 'uploader': uploader,
314 return webpage, info_dict
316 def _real_extract(self, url):
317 video_id = self._match_id(url)
319 real_url = self._VIDEO_PAGE_TEMPLATE % video_id if url.startswith('facebook:') else url
320 webpage, info_dict = self._extract_from_url(real_url, video_id, fatal_if_no_video=False)
327 self.url_result('facebook:%s' % vid, FacebookIE.ie_key())
328 for vid in self._parse_json(
330 r'(["\'])video_ids\
1\s
*:\s
*(?P
<ids
>\
[.+?\
])',
331 webpage, 'video ids
', group='ids
'),
334 return self.playlist_result(entries, video_id)
336 _, info_dict = self._extract_from_url(
337 self._VIDEO_PAGE_TEMPLATE % video_id,
338 video_id, fatal_if_no_video=True)