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 webpage = self._download_webpage(url, video_id)
51
52 info = self._search_regex(
53 r'''(<div\b[^>]*\bid\s*=\s*(['"])pageMetaDetails\2[^>]*>)''',
54 webpage, 'meta details', default='')
55 info = extract_attributes(info)
56
57 player = self._search_regex(
58 r'''(<div\b[^>]*\bid\s*=\s*(['"])rmpPlayerStream\2[^>]*>)''',
59 webpage, 'player details', default='')
60 player = extract_attributes(player)
61
62 video_urls_and_ids = (
63 (info.get('data-meta-video'), 'video'),
64 (player.get('data-video-transcoded'), 'transcoded'),
65 (player.get('data-video-filepath'), 'filepath'),
66 (self._og_search_video_url(webpage, secure=False, default=None), 'og_video'),
67 )
68
69 def txt_or_none(s, default=None):
70 return (s.strip() or default) if isinstance(s, compat_str) else default
71
72 uploader = txt_or_none(info.get('data-meta-author'))
73
74 def mung_title(s):
75 if uploader:
76 s = re.sub(r'^\s*%s\s+[|-]' % (re.escape(uploader), ), '', s)
77 return txt_or_none(s)
78
79 title = (
80 mung_title(info.get('data-meta-title'))
81 or self._html_search_regex(
82 (r'<span[^>]+class=["\']item-title[^>]+>([^<]+)',
83 r'<h2[^>]+class=["\']h2 m-0["\'][^>]*>([^<]+)'),
84 webpage, 'title', default=None)
85 or self._html_search_meta(
86 'twitter:title', webpage, 'title', fatal=True))
87
88 title = re.sub(r'\s*[|-]\s+ManyVids\s*$', '', title) or title
89
90 if any(p in webpage for p in ('preview_videos', '_preview.mp4')):
91 title += ' (Preview)'
92
93 mv_token = self._search_regex(
94 r'data-mvtoken=(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
95 'mv token', default=None, group='value')
96
97 if mv_token:
98 # Sets some cookies
99 self._download_webpage(
100 'https://www.manyvids.com/includes/ajax_repository/you_had_me_at_hello.php',
101 video_id, fatal=False, data=urlencode_postdata({
102 'mvtoken': mv_token,
103 'vid': video_id,
104 }), headers={
105 'Referer': url,
106 'X-Requested-With': 'XMLHttpRequest'
107 })
108
109 formats = []
110 for v_url, fmt in video_urls_and_ids:
111 v_url = url_or_none(v_url)
112 if not v_url:
113 continue
114 if determine_ext(v_url) == 'm3u8':
115 formats.extend(self._extract_m3u8_formats(
116 v_url, video_id, 'mp4', entry_protocol='m3u8_native',
117 m3u8_id='hls'))
118 else:
119 formats.append({
120 'url': v_url,
121 'format_id': fmt,
122 })
123
124 self._remove_duplicate_formats(formats)
125
126 for f in formats:
127 if f.get('height') is None:
128 f['height'] = int_or_none(
129 self._search_regex(r'_(\d{2,3}[02468])_', f['url'], 'video height', default=None))
130 if '/preview/' in f['url']:
131 f['format_id'] = '_'.join(filter(None, (f.get('format_id'), 'preview')))
132 f['preference'] = -10
133 if 'transcoded' in f['format_id']:
134 f['preference'] = f.get('preference', -1) - 1
135
136 self._sort_formats(formats)
137
138 def get_likes():
139 likes = self._search_regex(
140 r'''(<a\b[^>]*\bdata-id\s*=\s*(['"])%s\2[^>]*>)''' % (video_id, ),
141 webpage, 'likes', default='')
142 likes = extract_attributes(likes)
143 return int_or_none(likes.get('data-likes'))
144
145 def get_views():
146 return str_to_int(self._html_search_regex(
147 r'''(?s)<span\b[^>]*\bclass\s*=["']views-wrapper\b[^>]+>.+?<span\b[^>]+>\s*(\d[\d,.]*)\s*</span>''',
148 webpage, 'view count', default=None))
149
150 return {
151 'id': video_id,
152 'title': title,
153 'formats': formats,
154 'description': txt_or_none(info.get('data-meta-description')),
155 'uploader': txt_or_none(info.get('data-meta-author')),
156 'thumbnail': (
157 url_or_none(info.get('data-meta-image'))
158 or url_or_none(player.get('data-video-screenshot'))),
159 'view_count': get_views(),
160 'like_count': get_likes(),
161 }
|
