1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..compat import compat_str
8 from ..utils import (
9 determine_ext,
10 extract_attributes,
11 int_or_none,
12 str_to_int,
13 url_or_none,
14 urlencode_postdata,
15 )
16
17
18 class ManyVidsIE(InfoExtractor):
19 _VALID_URL = r'(?i)https?://(?:www\.)?manyvids\.com/video/(?P<id>\d+)'
20 _TESTS = [{
21 # preview video
22 'url': 'https://www.manyvids.com/Video/133957/everthing-about-me/',
23 'md5': '03f11bb21c52dd12a05be21a5c7dcc97',
24 'info_dict': {
25 'id': '133957',
26 'ext': 'mp4',
27 'title': 'everthing about me (Preview)',
28 'uploader': 'ellyxxix',
29 'view_count': int,
30 'like_count': int,
31 },
32 }, {
33 # full video
34 'url': 'https://www.manyvids.com/Video/935718/MY-FACE-REVEAL/',
35 'md5': 'bb47bab0e0802c2a60c24ef079dfe60f',
36 'info_dict': {
37 'id': '935718',
38 'ext': 'mp4',
39 'title': 'MY FACE REVEAL',
40 'description': 'md5:ec5901d41808b3746fed90face161612',
41 'uploader': 'Sarah Calanthe',
42 'view_count': int,
43 'like_count': int,
44 },
45 }]
46
47 def _real_extract(self, url):
48 video_id = self._match_id(url)
49
50 real_url = 'https://www.manyvids.com/video/%s/gtm.js' % (video_id, )
51 try:
52 webpage = self._download_webpage(real_url, video_id)
53 except:
54 # probably useless fallback
55 webpage = self._download_webpage(url, video_id)
56
57 info = self._search_regex(
58 r'''(<div\b[^>]*\bid\s*=\s*(['"])pageMetaDetails\2[^>]*>)''',
59 webpage, 'meta details', default='')
60 info = extract_attributes(info)
61
62 player = self._search_regex(
63 r'''(<div\b[^>]*\bid\s*=\s*(['"])rmpPlayerStream\2[^>]*>)''',
64 webpage, 'player details', default='')
65 player = extract_attributes(player)
66
67 video_urls_and_ids = (
68 (info.get('data-meta-video'), 'video'),
69 (player.get('data-video-transcoded'), 'transcoded'),
70 (player.get('data-video-filepath'), 'filepath'),
71 (self._og_search_video_url(webpage, secure=False, default=None), 'og_video'),
72 )
73
74 def txt_or_none(s, default=None):
75 return (s.strip() or default) if isinstance(s, compat_str) else default
76
77 uploader = txt_or_none(info.get('data-meta-author'))
78
79 def mung_title(s):
80 if uploader:
81 s = re.sub(r'^\s*%s\s+[|-]' % (re.escape(uploader), ), '', s)
82 return txt_or_none(s)
83
84 title = (
85 mung_title(info.get('data-meta-title'))
86 or self._html_search_regex(
87 (r'<span[^>]+class=["\']item-title[^>]+>([^<]+)',
88 r'<h2[^>]+class=["\']h2 m-0["\'][^>]*>([^<]+)'),
89 webpage, 'title', default=None)
90 or self._html_search_meta(
91 'twitter:title', webpage, 'title', fatal=True))
92
93 title = re.sub(r'\s*[|-]\s+ManyVids\s*$', '', title) or title
94
95 if any(p in webpage for p in ('preview_videos', '_preview.mp4')):
96 title += ' (Preview)'
97
98 mv_token = self._search_regex(
99 r'data-mvtoken=(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
100 'mv token', default=None, group='value')
101
102 if mv_token:
103 # Sets some cookies
104 self._download_webpage(
105 'https://www.manyvids.com/includes/ajax_repository/you_had_me_at_hello.php',
106 video_id, note='Setting format cookies', fatal=False,
107 data=urlencode_postdata({
108 'mvtoken': mv_token,
109 'vid': video_id,
110 }), headers={
111 'Referer': url,
112 'X-Requested-With': 'XMLHttpRequest'
113 })
114
115 formats = []
116 for v_url, fmt in video_urls_and_ids:
117 v_url = url_or_none(v_url)
118 if not v_url:
119 continue
120 if determine_ext(v_url) == 'm3u8':
121 formats.extend(self._extract_m3u8_formats(
122 v_url, video_id, 'mp4', entry_protocol='m3u8_native',
123 m3u8_id='hls'))
124 else:
125 formats.append({
126 'url': v_url,
127 'format_id': fmt,
128 })
129
130 self._remove_duplicate_formats(formats)
131
132 for f in formats:
133 if f.get('height') is None:
134 f['height'] = int_or_none(
135 self._search_regex(r'_(\d{2,3}[02468])_', f['url'], 'video height', default=None))
136 if '/preview/' in f['url']:
137 f['format_id'] = '_'.join(filter(None, (f.get('format_id'), 'preview')))
138 f['preference'] = -10
139 if 'transcoded' in f['format_id']:
140 f['preference'] = f.get('preference', -1) - 1
141
142 self._sort_formats(formats)
143
144 def get_likes():
145 likes = self._search_regex(
146 r'''(<a\b[^>]*\bdata-id\s*=\s*(['"])%s\2[^>]*>)''' % (video_id, ),
147 webpage, 'likes', default='')
148 likes = extract_attributes(likes)
149 return int_or_none(likes.get('data-likes'))
150
151 def get_views():
152 return str_to_int(self._html_search_regex(
153 r'''(?s)<span\b[^>]*\bclass\s*=["']views-wrapper\b[^>]+>.+?<span\b[^>]+>\s*(\d[\d,.]*)\s*</span>''',
154 webpage, 'view count', default=None))
155
156 return {
157 'id': video_id,
158 'title': title,
159 'formats': formats,
160 'description': txt_or_none(info.get('data-meta-description')),
161 'uploader': txt_or_none(info.get('data-meta-author')),
162 'thumbnail': (
163 url_or_none(info.get('data-meta-image'))
164 or url_or_none(player.get('data-video-screenshot'))),
165 'view_count': get_views(),
166 'like_count': get_likes(),
167 }
|
