Revert "Workaround for regex engine limitation"
[youtube-dl] / youtube_dl / extractor / appletrailers.py
1 from __future__ import unicode_literals
2
3 import re
4 import json
5
6 from .common import InfoExtractor
7 from ..utils import (
8     compat_urlparse,
9     determine_ext,
10 )
11
12
13 class AppleTrailersIE(InfoExtractor):
14     _VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/trailers/(?P<company>[^/]+)/(?P<movie>[^/]+)'
15     _TEST = {
16         "url": "http://trailers.apple.com/trailers/wb/manofsteel/",
17         "playlist": [
18             {
19                 "file": "manofsteel-trailer4.mov",
20                 "md5": "d97a8e575432dbcb81b7c3acb741f8a8",
21                 "info_dict": {
22                     "duration": 111,
23                     "title": "Trailer 4",
24                     "upload_date": "20130523",
25                     "uploader_id": "wb",
26                 },
27             },
28             {
29                 "file": "manofsteel-trailer3.mov",
30                 "md5": "b8017b7131b721fb4e8d6f49e1df908c",
31                 "info_dict": {
32                     "duration": 182,
33                     "title": "Trailer 3",
34                     "upload_date": "20130417",
35                     "uploader_id": "wb",
36                 },
37             },
38             {
39                 "file": "manofsteel-trailer.mov",
40                 "md5": "d0f1e1150989b9924679b441f3404d48",
41                 "info_dict": {
42                     "duration": 148,
43                     "title": "Trailer",
44                     "upload_date": "20121212",
45                     "uploader_id": "wb",
46                 },
47             },
48             {
49                 "file": "manofsteel-teaser.mov",
50                 "md5": "5fe08795b943eb2e757fa95cb6def1cb",
51                 "info_dict": {
52                     "duration": 93,
53                     "title": "Teaser",
54                     "upload_date": "20120721",
55                     "uploader_id": "wb",
56                 },
57             }
58         ]
59     }
60
61     _JSON_RE = r'iTunes.playURL\((.*?)\);'
62
63     def _real_extract(self, url):
64         mobj = re.match(self._VALID_URL, url)
65         movie = mobj.group('movie')
66         uploader_id = mobj.group('company')
67
68         playlist_url = compat_urlparse.urljoin(url, u'includes/playlists/itunes.inc')
69         def fix_html(s):
70             s = re.sub(r'(?s)<script[^<]*?>.*?</script>', u'', s)
71             s = re.sub(r'<img ([^<]*?)>', r'<img \1/>', s)
72             # The ' in the onClick attributes are not escaped, it couldn't be parsed
73             # like: http://trailers.apple.com/trailers/wb/gravity/
74             def _clean_json(m):
75                 return u'iTunes.playURL(%s);' % m.group(1).replace('\'', '&#39;')
76             s = re.sub(self._JSON_RE, _clean_json, s)
77             s = u'<html>' + s + u'</html>'
78             return s
79         doc = self._download_xml(playlist_url, movie, transform_source=fix_html)
80
81         playlist = []
82         for li in doc.findall('./div/ul/li'):
83             on_click = li.find('.//a').attrib['onClick']
84             trailer_info_json = self._search_regex(self._JSON_RE,
85                 on_click, u'trailer info')
86             trailer_info = json.loads(trailer_info_json)
87             title = trailer_info['title']
88             video_id = movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', title).lower()
89             thumbnail = li.find('.//img').attrib['src']
90             upload_date = trailer_info['posted'].replace('-', '')
91
92             runtime = trailer_info['runtime']
93             m = re.search(r'(?P<minutes>[0-9]+):(?P<seconds>[0-9]{1,2})', runtime)
94             duration = None
95             if m:
96                 duration = 60 * int(m.group('minutes')) + int(m.group('seconds'))
97
98             first_url = trailer_info['url']
99             trailer_id = first_url.split('/')[-1].rpartition('_')[0].lower()
100             settings_json_url = compat_urlparse.urljoin(url, 'includes/settings/%s.json' % trailer_id)
101             settings_json = self._download_webpage(settings_json_url, trailer_id, u'Downloading settings json')
102             settings = json.loads(settings_json)
103
104             formats = []
105             for format in settings['metadata']['sizes']:
106                 # The src is a file pointing to the real video file
107                 format_url = re.sub(r'_(\d*p.mov)', r'_h\1', format['src'])
108                 formats.append({
109                     'url': format_url,
110                     'ext': determine_ext(format_url),
111                     'format': format['type'],
112                     'width': format['width'],
113                     'height': int(format['height']),
114                 })
115
116             self._sort_formats(formats)
117
118             playlist.append({
119                 '_type': 'video',
120                 'id': video_id,
121                 'title': title,
122                 'formats': formats,
123                 'title': title,
124                 'duration': duration,
125                 'thumbnail': thumbnail,
126                 'upload_date': upload_date,
127                 'uploader_id': uploader_id,
128                 'user_agent': 'QuickTime compatible (youtube-dl)',
129             })
130
131         return {
132             '_type': 'playlist',
133             'id': movie,
134             'entries': playlist,
135         }