]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/facebook.py
937b28fcccf3bd58929adcca1bda9d05966460e5
[youtubedl] / youtube_dl / extractor / facebook.py
1 from __future__ import unicode_literals
2
3 import json
4 import re
5 import socket
6
7 from .common import InfoExtractor
8 from ..compat import (
9 compat_http_client,
10 compat_str,
11 compat_urllib_error,
12 compat_urllib_parse,
13 compat_urllib_request,
14 )
15 from ..utils import (
16 ExtractorError,
17 int_or_none,
18 limit_length,
19 urlencode_postdata,
20 )
21
22
23 class FacebookIE(InfoExtractor):
24 _VALID_URL = r'''(?x)
25 https?://(?:\w+\.)?facebook\.com/
26 (?:[^#]*?\#!/)?
27 (?:
28 (?:video/video\.php|photo\.php|video\.php|video/embed)\?(?:.*?)
29 (?:v|video_id)=|
30 [^/]+/videos/(?:[^/]+/)?
31 )
32 (?P<id>[0-9]+)
33 (?:.*)'''
34 _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
35 _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
36 _NETRC_MACHINE = 'facebook'
37 IE_NAME = 'facebook'
38 _TESTS = [{
39 'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf',
40 'md5': '6a40d33c0eccbb1af76cf0485a052659',
41 'info_dict': {
42 'id': '637842556329505',
43 'ext': 'mp4',
44 'title': 're:Did you know Kei Nishikori is the first Asian man to ever reach a Grand Slam',
45 }
46 }, {
47 'note': 'Video without discernible title',
48 'url': 'https://www.facebook.com/video.php?v=274175099429670',
49 'info_dict': {
50 'id': '274175099429670',
51 'ext': 'mp4',
52 'title': 'Facebook video #274175099429670',
53 }
54 }, {
55 'url': 'https://www.facebook.com/video.php?v=10204634152394104',
56 'only_matching': True,
57 }, {
58 'url': 'https://www.facebook.com/amogood/videos/1618742068337349/?fref=nf',
59 'only_matching': True,
60 }, {
61 'url': 'https://www.facebook.com/ChristyClarkForBC/videos/vb.22819070941/10153870694020942/?type=2&theater',
62 'only_matching': True,
63 }]
64
65 def _login(self):
66 (useremail, password) = self._get_login_info()
67 if useremail is None:
68 return
69
70 login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
71 login_page_req.add_header('Cookie', 'locale=en_US')
72 login_page = self._download_webpage(login_page_req, None,
73 note='Downloading login page',
74 errnote='Unable to download login page')
75 lsd = self._search_regex(
76 r'<input type="hidden" name="lsd" value="([^"]*)"',
77 login_page, 'lsd')
78 lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, 'lgnrnd')
79
80 login_form = {
81 'email': useremail,
82 'pass': password,
83 'lsd': lsd,
84 'lgnrnd': lgnrnd,
85 'next': 'http://facebook.com/home.php',
86 'default_persistent': '0',
87 'legacy_return': '1',
88 'timezone': '-60',
89 'trynum': '1',
90 }
91 request = compat_urllib_request.Request(self._LOGIN_URL, urlencode_postdata(login_form))
92 request.add_header('Content-Type', 'application/x-www-form-urlencoded')
93 try:
94 login_results = self._download_webpage(request, None,
95 note='Logging in', errnote='unable to fetch login page')
96 if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
97 self._downloader.report_warning('unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
98 return
99
100 check_form = {
101 'fb_dtsg': self._search_regex(r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg'),
102 'h': self._search_regex(
103 r'name="h"\s+(?:\w+="[^"]+"\s+)*?value="([^"]+)"', login_results, 'h'),
104 'name_action_selected': 'dont_save',
105 }
106 check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))
107 check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
108 check_response = self._download_webpage(check_req, None,
109 note='Confirming login')
110 if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
111 self._downloader.report_warning('Unable to confirm login, you have to login in your brower and authorize the login.')
112 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
113 self._downloader.report_warning('unable to log in: %s' % compat_str(err))
114 return
115
116 def _real_initialize(self):
117 self._login()
118
119 def _real_extract(self, url):
120 video_id = self._match_id(url)
121 url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
122 webpage = self._download_webpage(url, video_id)
123
124 BEFORE = '{swf.addParam(param[0], param[1]);});\n'
125 AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
126 m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
127 if not m:
128 m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
129 if m_msg is not None:
130 raise ExtractorError(
131 'The video is not available, Facebook said: "%s"' % m_msg.group(1),
132 expected=True)
133 else:
134 raise ExtractorError('Cannot parse data')
135 data = dict(json.loads(m.group(1)))
136 params_raw = compat_urllib_parse.unquote(data['params'])
137 params = json.loads(params_raw)
138 video_data = params['video_data'][0]
139
140 formats = []
141 for quality in ['sd', 'hd']:
142 src = video_data.get('%s_src' % quality)
143 if src is not None:
144 formats.append({
145 'format_id': quality,
146 'url': src,
147 })
148 if not formats:
149 raise ExtractorError('Cannot find video formats')
150
151 video_title = self._html_search_regex(
152 r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, 'title',
153 fatal=False)
154 if not video_title:
155 video_title = self._html_search_regex(
156 r'(?s)<span class="fbPhotosPhotoCaption".*?id="fbPhotoPageCaption"><span class="hasCaption">(.*?)</span>',
157 webpage, 'alternative title', default=None)
158 video_title = limit_length(video_title, 80)
159 if not video_title:
160 video_title = 'Facebook video #%s' % video_id
161
162 return {
163 'id': video_id,
164 'title': video_title,
165 'formats': formats,
166 'duration': int_or_none(video_data.get('video_duration')),
167 'thumbnail': video_data.get('thumbnail_src'),
168 }