Raphaël G. Git Repositories - youtubedl/blob - 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 ..compat import compat_urlparse
   8 from ..utils import (
   9     int_or_none,
  10 )
  11
  12
  13 class AppleTrailersIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/(?:trailers|ca)/(?P<company>[^/]+)/(?P<movie>[^/]+)'
  15     _TESTS = [{
  16         "url": "http://trailers.apple.com/trailers/wb/manofsteel/",
  17         'info_dict': {
  18             'id': 'manofsteel',
  19         },
  20         "playlist": [
  21             {
  22                 "md5": "d97a8e575432dbcb81b7c3acb741f8a8",
  23                 "info_dict": {
  24                     "id": "manofsteel-trailer4",
  25                     "ext": "mov",
  26                     "duration": 111,
  27                     "title": "Trailer 4",
  28                     "upload_date": "20130523",
  29                     "uploader_id": "wb",
  30                 },
  31             },
  32             {
  33                 "md5": "b8017b7131b721fb4e8d6f49e1df908c",
  34                 "info_dict": {
  35                     "id": "manofsteel-trailer3",
  36                     "ext": "mov",
  37                     "duration": 182,
  38                     "title": "Trailer 3",
  39                     "upload_date": "20130417",
  40                     "uploader_id": "wb",
  41                 },
  42             },
  43             {
  44                 "md5": "d0f1e1150989b9924679b441f3404d48",
  45                 "info_dict": {
  46                     "id": "manofsteel-trailer",
  47                     "ext": "mov",
  48                     "duration": 148,
  49                     "title": "Trailer",
  50                     "upload_date": "20121212",
  51                     "uploader_id": "wb",
  52                 },
  53             },
  54             {
  55                 "md5": "5fe08795b943eb2e757fa95cb6def1cb",
  56                 "info_dict": {
  57                     "id": "manofsteel-teaser",
  58                     "ext": "mov",
  59                     "duration": 93,
  60                     "title": "Teaser",
  61                     "upload_date": "20120721",
  62                     "uploader_id": "wb",
  63                 },
  64             },
  65         ]
  66     }, {
  67         'url': 'http://trailers.apple.com/ca/metropole/autrui/',
  68         'only_matching': True,
  69     }]
  70
  71     _JSON_RE = r'iTunes.playURL\((.*?)\);'
  72
  73     def _real_extract(self, url):
  74         mobj = re.match(self._VALID_URL, url)
  75         movie = mobj.group('movie')
  76         uploader_id = mobj.group('company')
  77
  78         playlist_url = compat_urlparse.urljoin(url, 'includes/playlists/itunes.inc')
  79
  80         def fix_html(s):
  81             s = re.sub(r'(?s)<script[^<]*?>.*?</script>', '', s)
  82             s = re.sub(r'<img ([^<]*?)>', r'<img \1/>', s)
  83             # The ' in the onClick attributes are not escaped, it couldn't be parsed
  84             # like: http://trailers.apple.com/trailers/wb/gravity/
  85
  86             def _clean_json(m):
  87                 return 'iTunes.playURL(%s);' % m.group(1).replace('\'', '&#39;')
  88             s = re.sub(self._JSON_RE, _clean_json, s)
  89             s = '<html>%s</html>' % s
  90             return s
  91         doc = self._download_xml(playlist_url, movie, transform_source=fix_html)
  92
  93         playlist = []
  94         for li in doc.findall('./div/ul/li'):
  95             on_click = li.find('.//a').attrib['onClick']
  96             trailer_info_json = self._search_regex(self._JSON_RE,
  97                                                    on_click, 'trailer info')
  98             trailer_info = json.loads(trailer_info_json)
  99             title = trailer_info['title']
 100             video_id = movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', title).lower()
 101             thumbnail = li.find('.//img').attrib['src']
 102             upload_date = trailer_info['posted'].replace('-', '')
 103
 104             runtime = trailer_info['runtime']
 105             m = re.search(r'(?P<minutes>[0-9]+):(?P<seconds>[0-9]{1,2})', runtime)
 106             duration = None
 107             if m:
 108                 duration = 60 * int(m.group('minutes')) + int(m.group('seconds'))
 109
 110             first_url = trailer_info['url']
 111             trailer_id = first_url.split('/')[-1].rpartition('_')[0].lower()
 112             settings_json_url = compat_urlparse.urljoin(url, 'includes/settings/%s.json' % trailer_id)
 113             settings = self._download_json(settings_json_url, trailer_id, 'Downloading settings json')
 114
 115             formats = []
 116             for format in settings['metadata']['sizes']:
 117                 # The src is a file pointing to the real video file
 118                 format_url = re.sub(r'_(\d*p.mov)', r'_h\1', format['src'])
 119                 formats.append({
 120                     'url': format_url,
 121                     'format': format['type'],
 122                     'width': int_or_none(format['width']),
 123                     'height': int_or_none(format['height']),
 124                 })
 125
 126             self._sort_formats(formats)
 127
 128             playlist.append({
 129                 '_type': 'video',
 130                 'id': video_id,
 131                 'formats': formats,
 132                 'title': title,
 133                 'duration': duration,
 134                 'thumbnail': thumbnail,
 135                 'upload_date': upload_date,
 136                 'uploader_id': uploader_id,
 137                 'http_headers': {
 138                     'User-Agent': 'QuickTime compatible (youtube-dl)',
 139                 },
 140             })
 141
 142         return {
 143             '_type': 'playlist',
 144             'id': movie,
 145             'entries': playlist,
 146         }