1 from __future__ import unicode_literals
6 from .common import InfoExtractor
8 compat_etree_fromstring,
11 compat_urllib_parse_unquote,
12 compat_urllib_parse_unquote_plus,
28 class FacebookIE(InfoExtractor):
32 (?:[\w-]+\.)?(?:facebook\.com|facebookcorewwwi\.onion)/
41 )\?(?:.*?)(?:v|video_id|story_fbid)=|
42 [^/]+/videos/(?:[^/]+/)?|
44 groups/[^/]+/permalink/
50 _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
51 _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
52 _NETRC_MACHINE = 'facebook'
55 _CHROME_USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/48.0.2564.97 Safari/537.36'
57 _VIDEO_PAGE_TEMPLATE = 'https://www.facebook.com/video/video.php?v=%s'
60 'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf',
61 'md5': '6a40d33c0eccbb1af76cf0485a052659',
63 'id': '637842556329505',
65 'title': 're:Did you know Kei Nishikori is the first Asian man to ever reach a Grand Slam',
66 'uploader': 'Tennis on Facebook',
67 'upload_date': '20140908',
68 'timestamp': 1410199200,
71 'note': 'Video without discernible title',
72 'url': 'https://www.facebook.com/video.php?v=274175099429670',
74 'id': '274175099429670',
76 'title': 'Asif Nawab Butt posted a video to his Timeline.',
77 'uploader': 'Asif Nawab Butt',
78 'upload_date': '20140506',
79 'timestamp': 1399398998,
81 'expected_warnings': [
85 'note': 'Video with DASH manifest',
86 'url': 'https://www.facebook.com/video.php?v=957955867617029',
87 'md5': 'b2c28d528273b323abe5c6ab59f0f030',
89 'id': '957955867617029',
91 'title': 'When you post epic content on instagram.com/433 8 million followers, this is ...',
92 'uploader': 'Demy de Zeeuw',
93 'upload_date': '20160110',
94 'timestamp': 1452431627,
97 'url': 'https://www.facebook.com/maxlayn/posts/10153807558977570',
98 'md5': '037b1fa7f3c2d02b7a0d7bc16031ecc6',
100 'id': '544765982287235',
102 'title': '"What are you doing running in the snow?"',
103 'uploader': 'FailArmy',
105 'skip': 'Video gone',
107 'url': 'https://m.facebook.com/story.php?story_fbid=1035862816472149&id=116132035111903',
108 'md5': '1deb90b6ac27f7efcf6d747c8a27f5e3',
110 'id': '1035862816472149',
112 'title': 'What the Flock Is Going On In New Zealand Credit: ViralHog',
113 'uploader': 'S. Saint',
115 'skip': 'Video gone',
117 'note': 'swf params escaped',
118 'url': 'https://www.facebook.com/barackobama/posts/10153664894881749',
119 'md5': '97ba073838964d12c70566e0085c2b91',
121 'id': '10153664894881749',
123 'title': 'Facebook video #10153664894881749',
126 # have 1080P, but only up to 720p in swf params
127 'url': 'https://www.facebook.com/cnn/videos/10155529876156509/',
128 'md5': '0d9813160b146b3bc8744e006027fcc6',
130 'id': '10155529876156509',
132 'title': 'Holocaust survivor becomes US citizen',
133 'timestamp': 1477818095,
134 'upload_date': '20161030',
138 # bigPipe.onPageletArrive ... onPageletArrive pagelet_group_mall
139 'url': 'https://www.facebook.com/yaroslav.korpan/videos/1417995061575415/',
141 'id': '1417995061575415',
143 'title': 'md5:a7b86ca673f51800cd54687b7f4012fe',
144 'timestamp': 1486648217,
145 'upload_date': '20170209',
146 'uploader': 'Yaroslav Korpan',
149 'skip_download': True,
152 'url': 'https://www.facebook.com/LaGuiaDelVaron/posts/1072691702860471',
154 'id': '1072691702860471',
156 'title': 'md5:ae2d22a93fbb12dad20dc393a869739d',
157 'timestamp': 1477305000,
158 'upload_date': '20161024',
159 'uploader': 'La Guía Del Varón',
162 'skip_download': True,
165 'url': 'https://www.facebook.com/groups/1024490957622648/permalink/1396382447100162/',
167 'id': '1396382447100162',
169 'title': 'md5:e2d2700afdf84e121f5d0f999bad13a3',
170 'timestamp': 1486035494,
171 'upload_date': '20170202',
172 'uploader': 'Elisabeth Ahtn',
175 'skip_download': True,
178 'url': 'https://www.facebook.com/video.php?v=10204634152394104',
179 'only_matching': True,
181 'url': 'https://www.facebook.com/amogood/videos/1618742068337349/?fref=nf',
182 'only_matching': True,
184 'url': 'https://www.facebook.com/ChristyClarkForBC/videos/vb.22819070941/10153870694020942/?type=2&theater',
185 'only_matching': True,
187 'url': 'facebook:544765982287235',
188 'only_matching': True,
190 'url': 'https://www.facebook.com/groups/164828000315060/permalink/764967300301124/',
191 'only_matching': True,
193 'url': 'https://zh-hk.facebook.com/peoplespower/videos/1135894589806027/',
194 'only_matching': True,
196 'url': 'https://www.facebookcorewwwi.onion/video.php?v=274175099429670',
197 'only_matching': True,
201 def _extract_url(webpage):
203 r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
205 return mobj.group('url')
208 # see https://developers.facebook.com/docs/plugins/embedded-video-player
209 mobj = re.search(r'''(?x)<div[^>]+
210 class=(?P<q1>[\'"])[^\'"]*\bfb-(?:video|post)\b[^\'"]*(?P=q1)[^>]+
211 data-href=(?P<q2>[\'"])(?P<url>(?:https?:)?//(?:www\.)?facebook.com/.+?)(?P=q2)''', webpage)
213 return mobj.group('url')
216 (useremail, password) = self._get_login_info()
217 if useremail is None:
220 login_page_req = sanitized_Request(self._LOGIN_URL)
221 self._set_cookie('facebook.com', 'locale', 'en_US')
222 login_page = self._download_webpage(login_page_req, None,
223 note='Downloading login page',
224 errnote='Unable to download login page')
225 lsd = self._search_regex(
226 r'<input type="hidden" name="lsd" value="([^"]*)"',
228 lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, 'lgnrnd')
235 'next': 'http://facebook.com/home.php',
236 'default_persistent': '0',
237 'legacy_return': '1',
241 request = sanitized_Request(self._LOGIN_URL, urlencode_postdata(login_form))
242 request.add_header('Content-Type', 'application/x-www-form-urlencoded')
244 login_results = self._download_webpage(request, None,
245 note='Logging in', errnote='unable to fetch login page')
246 if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
247 error = self._html_search_regex(
248 r'(?s)<div[^>]+class=(["\']).*?login_error_box.*?\1[^>]*><div[^>]*>.*?</div><div[^>]*>(?P<error>.+?)</div>',
249 login_results, 'login error', default=None, group='error')
251 raise ExtractorError('Unable to login: %s' % error, expected=True)
252 self._downloader.report_warning('unable to log in: bad username/password, or exceeded login rate limit (~3/min). Check credentials or wait.')
255 fb_dtsg = self._search_regex(
256 r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg', default=None)
257 h = self._search_regex(
258 r'name="h"\s+(?:\w+="[^"]+"\s+)*?value="([^"]+)"', login_results, 'h', default=None)
260 if not fb_dtsg or not h:
266 'name_action_selected': 'dont_save',
268 check_req = sanitized_Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))
269 check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
270 check_response = self._download_webpage(check_req, None,
271 note='Confirming login')
272 if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
273 self._downloader.report_warning('Unable to confirm login, you have to login in your browser and authorize the login.')
274 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
275 self._downloader.report_warning('unable to log in: %s' % error_to_compat_str(err))
278 def _real_initialize(self):
281 def _extract_from_url(self, url, video_id, fatal_if_no_video=True):
282 req = sanitized_Request(url)
283 req.add_header('User-Agent', self._CHROME_USER_AGENT)
284 webpage = self._download_webpage(req, video_id)
288 def extract_video_data(instances):
289 for item in instances:
290 if item[1][0] == 'VideoConfig':
291 video_item = item[2][0]
292 if video_item.get('video_id'):
293 return video_item['videoData']
295 server_js_data = self._parse_json(self._search_regex(
296 r'handleServerJS\(({.+})(?:\);|,")', webpage,
297 'server js data', default='{}'), video_id, fatal=False)
300 video_data = extract_video_data(server_js_data.get('instances', []))
303 server_js_data = self._parse_json(
305 r'bigPipe\.onPageletArrive\(({.+?})\)\s*;\s*}\s*\)\s*,\s*["\']onPageletArrive\s+(?:stream_pagelet|pagelet_group_mall)',
306 webpage, 'js data', default='{}'),
307 video_id, transform_source=js_to_json, fatal=False)
309 video_data = extract_video_data(try_get(
310 server_js_data, lambda x: x['jsmods']['instances'],
314 if not fatal_if_no_video:
315 return webpage, False
316 m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
317 if m_msg is not None:
318 raise ExtractorError(
319 'The video is not available, Facebook said: "%s"' % m_msg.group(1),
321 elif '>You must log in to continue' in webpage:
322 self.raise_login_required()
324 raise ExtractorError('Cannot parse data')
328 format_id = f['stream_type']
329 if f and isinstance(f, dict):
331 if not f or not isinstance(f, list):
333 for quality in ('sd', 'hd'):
334 for src_type in ('src', 'src_no_ratelimit'):
335 src = f[0].get('%s_%s' % (quality, src_type))
337 preference = -10 if format_id == 'progressive' else 0
341 'format_id': '%s_%s_%s' % (format_id, quality, src_type),
343 'preference': preference,
345 dash_manifest = f[0].get('dash_manifest')
347 formats.extend(self._parse_mpd_formats(
348 compat_etree_fromstring(compat_urllib_parse_unquote_plus(dash_manifest))))
350 raise ExtractorError('Cannot find video formats')
352 self._sort_formats(formats)
354 video_title = self._html_search_regex(
355 r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>([^<]*)</h2>', webpage, 'title',
358 video_title = self._html_search_regex(
359 r'(?s)<span class="fbPhotosPhotoCaption".*?id="fbPhotoPageCaption"><span class="hasCaption">(.*?)</span>',
360 webpage, 'alternative title', default=None)
362 video_title = self._html_search_meta(
363 'description', webpage, 'title')
365 video_title = limit_length(video_title, 80)
367 video_title = 'Facebook video #%s' % video_id
368 uploader = clean_html(get_element_by_id(
369 'fbPhotoPageAuthorName', webpage)) or self._search_regex(
370 r'ownerName\s*:\s*"([^"]+)"', webpage, 'uploader', fatal=False)
371 timestamp = int_or_none(self._search_regex(
372 r'<abbr[^>]+data-utime=["\'](\d+)', webpage,
373 'timestamp', default=None))
377 'title': video_title,
379 'uploader': uploader,
380 'timestamp': timestamp,
383 return webpage, info_dict
385 def _real_extract(self, url):
386 video_id = self._match_id(url)
388 real_url = self._VIDEO_PAGE_TEMPLATE % video_id if url.startswith('facebook:') else url
389 webpage, info_dict = self._extract_from_url(real_url, video_id, fatal_if_no_video=False)
396 self.url_result('facebook:%s' % vid, FacebookIE.ie_key())
397 for vid in self._parse_json(
399 r'(["\'])video_ids\1\s*:\s*(?P<ids>\[.+?\])',
400 webpage, 'video ids', group='ids'),
403 return self.playlist_result(entries, video_id)
405 _, info_dict = self._extract_from_url(
406 self._VIDEO_PAGE_TEMPLATE % video_id,
407 video_id, fatal_if_no_video=True)
411 class FacebookPluginsVideoIE(InfoExtractor):
412 _VALID_URL = r'https?://(?:[\w-]+\.)?facebook\.com/plugins/video\.php\?.*?\bhref=(?P<id>https.+)'
415 'url': 'https://www.facebook.com/plugins/video.php?href=https%3A%2F%2Fwww.facebook.com%2Fgov.sg%2Fvideos%2F10154383743583686%2F&show_text=0&width=560',
416 'md5': '5954e92cdfe51fe5782ae9bda7058a07',
418 'id': '10154383743583686',
420 'title': 'What to do during the haze?',
421 'uploader': 'Gov.sg',
422 'upload_date': '20160826',
423 'timestamp': 1472184808,
425 'add_ie': [FacebookIE.ie_key()],
427 'url': 'https://www.facebook.com/plugins/video.php?href=https%3A%2F%2Fwww.facebook.com%2Fvideo.php%3Fv%3D10204634152394104',
428 'only_matching': True,
430 'url': 'https://www.facebook.com/plugins/video.php?href=https://www.facebook.com/gov.sg/videos/10154383743583686/&show_text=0&width=560',
431 'only_matching': True,
434 def _real_extract(self, url):
435 return self.url_result(
436 compat_urllib_parse_unquote(self._match_id(url)),