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     parse_duration,
  11     unified_strdate,
  12 )
  13
  14
  15 class AppleTrailersIE(InfoExtractor):
  16     IE_NAME = 'appletrailers'
  17     _VALID_URL = r'https?://(?:www\.|movie)?trailers\.apple\.com/(?:trailers|ca)/(?P<company>[^/]+)/(?P<movie>[^/]+)'
  18     _TESTS = [{
  19         'url': 'http://trailers.apple.com/trailers/wb/manofsteel/',
  20         'info_dict': {
  21             'id': '5111',
  22             'title': 'Man of Steel',
  23         },
  24         'playlist': [
  25             {
  26                 'md5': 'd97a8e575432dbcb81b7c3acb741f8a8',
  27                 'info_dict': {
  28                     'id': 'manofsteel-trailer4',
  29                     'ext': 'mov',
  30                     'duration': 111,
  31                     'title': 'Trailer 4',
  32                     'upload_date': '20130523',
  33                     'uploader_id': 'wb',
  34                 },
  35             },
  36             {
  37                 'md5': 'b8017b7131b721fb4e8d6f49e1df908c',
  38                 'info_dict': {
  39                     'id': 'manofsteel-trailer3',
  40                     'ext': 'mov',
  41                     'duration': 182,
  42                     'title': 'Trailer 3',
  43                     'upload_date': '20130417',
  44                     'uploader_id': 'wb',
  45                 },
  46             },
  47             {
  48                 'md5': 'd0f1e1150989b9924679b441f3404d48',
  49                 'info_dict': {
  50                     'id': 'manofsteel-trailer',
  51                     'ext': 'mov',
  52                     'duration': 148,
  53                     'title': 'Trailer',
  54                     'upload_date': '20121212',
  55                     'uploader_id': 'wb',
  56                 },
  57             },
  58             {
  59                 'md5': '5fe08795b943eb2e757fa95cb6def1cb',
  60                 'info_dict': {
  61                     'id': 'manofsteel-teaser',
  62                     'ext': 'mov',
  63                     'duration': 93,
  64                     'title': 'Teaser',
  65                     'upload_date': '20120721',
  66                     'uploader_id': 'wb',
  67                 },
  68             },
  69         ]
  70     }, {
  71         'url': 'http://trailers.apple.com/trailers/magnolia/blackthorn/',
  72         'info_dict': {
  73             'id': '4489',
  74             'title': 'Blackthorn',
  75         },
  76         'playlist_mincount': 2,
  77         'expected_warnings': ['Unable to download JSON metadata'],
  78     }, {
  79         # json data only available from http://trailers.apple.com/trailers/feeds/data/15881.json
  80         'url': 'http://trailers.apple.com/trailers/fox/kungfupanda3/',
  81         'info_dict': {
  82             'id': '15881',
  83             'title': 'Kung Fu Panda 3',
  84         },
  85         'playlist_mincount': 4,
  86     }, {
  87         'url': 'http://trailers.apple.com/ca/metropole/autrui/',
  88         'only_matching': True,
  89     }, {
  90         'url': 'http://movietrailers.apple.com/trailers/focus_features/kuboandthetwostrings/',
  91         'only_matching': True,
  92     }]
  93
  94     _JSON_RE = r'iTunes.playURL\((.*?)\);'
  95
  96     def _real_extract(self, url):
  97         mobj = re.match(self._VALID_URL, url)
  98         movie = mobj.group('movie')
  99         uploader_id = mobj.group('company')
 100
 101         webpage = self._download_webpage(url, movie)
 102         film_id = self._search_regex(r"FilmId\s*=\s*'(\d+)'", webpage, 'film id')
 103         film_data = self._download_json(
 104             'http://trailers.apple.com/trailers/feeds/data/%s.json' % film_id,
 105             film_id, fatal=False)
 106
 107         if film_data:
 108             entries = []
 109             for clip in film_data.get('clips', []):
 110                 clip_title = clip['title']
 111
 112                 formats = []
 113                 for version, version_data in clip.get('versions', {}).items():
 114                     for size, size_data in version_data.get('sizes', {}).items():
 115                         src = size_data.get('src')
 116                         if not src:
 117                             continue
 118                         formats.append({
 119                             'format_id': '%s-%s' % (version, size),
 120                             'url': re.sub(r'_(\d+p.mov)', r'_h\1', src),
 121                             'width': int_or_none(size_data.get('width')),
 122                             'height': int_or_none(size_data.get('height')),
 123                             'language': version[:2],
 124                         })
 125                 self._sort_formats(formats)
 126
 127                 entries.append({
 128                     'id': movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', clip_title).lower(),
 129                     'formats': formats,
 130                     'title': clip_title,
 131                     'thumbnail': clip.get('screen') or clip.get('thumb'),
 132                     'duration': parse_duration(clip.get('runtime') or clip.get('faded')),
 133                     'upload_date': unified_strdate(clip.get('posted')),
 134                     'uploader_id': uploader_id,
 135                 })
 136
 137             page_data = film_data.get('page', {})
 138             return self.playlist_result(entries, film_id, page_data.get('movie_title'))
 139
 140         playlist_url = compat_urlparse.urljoin(url, 'includes/playlists/itunes.inc')
 141
 142         def fix_html(s):
 143             s = re.sub(r'(?s)<script[^<]*?>.*?</script>', '', s)
 144             s = re.sub(r'<img ([^<]*?)/?>', r'<img \1/>', s)
 145             # The ' in the onClick attributes are not escaped, it couldn't be parsed
 146             # like: http://trailers.apple.com/trailers/wb/gravity/
 147
 148             def _clean_json(m):
 149                 return 'iTunes.playURL(%s);' % m.group(1).replace('\'', '&#39;')
 150             s = re.sub(self._JSON_RE, _clean_json, s)
 151             s = '<html>%s</html>' % s
 152             return s
 153         doc = self._download_xml(playlist_url, movie, transform_source=fix_html)
 154
 155         playlist = []
 156         for li in doc.findall('./div/ul/li'):
 157             on_click = li.find('.//a').attrib['onClick']
 158             trailer_info_json = self._search_regex(self._JSON_RE,
 159                                                    on_click, 'trailer info')
 160             trailer_info = json.loads(trailer_info_json)
 161             first_url = trailer_info.get('url')
 162             if not first_url:
 163                 continue
 164             title = trailer_info['title']
 165             video_id = movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', title).lower()
 166             thumbnail = li.find('.//img').attrib['src']
 167             upload_date = trailer_info['posted'].replace('-', '')
 168
 169             runtime = trailer_info['runtime']
 170             m = re.search(r'(?P<minutes>[0-9]+):(?P<seconds>[0-9]{1,2})', runtime)
 171             duration = None
 172             if m:
 173                 duration = 60 * int(m.group('minutes')) + int(m.group('seconds'))
 174
 175             trailer_id = first_url.split('/')[-1].rpartition('_')[0].lower()
 176             settings_json_url = compat_urlparse.urljoin(url, 'includes/settings/%s.json' % trailer_id)
 177             settings = self._download_json(settings_json_url, trailer_id, 'Downloading settings json')
 178
 179             formats = []
 180             for format in settings['metadata']['sizes']:
 181                 # The src is a file pointing to the real video file
 182                 format_url = re.sub(r'_(\d*p.mov)', r'_h\1', format['src'])
 183                 formats.append({
 184                     'url': format_url,
 185                     'format': format['type'],
 186                     'width': int_or_none(format['width']),
 187                     'height': int_or_none(format['height']),
 188                 })
 189
 190             self._sort_formats(formats)
 191
 192             playlist.append({
 193                 '_type': 'video',
 194                 'id': video_id,
 195                 'formats': formats,
 196                 'title': title,
 197                 'duration': duration,
 198                 'thumbnail': thumbnail,
 199                 'upload_date': upload_date,
 200                 'uploader_id': uploader_id,
 201                 'http_headers': {
 202                     'User-Agent': 'QuickTime compatible (youtube-dl)',
 203                 },
 204             })
 205
 206         return {
 207             '_type': 'playlist',
 208             'id': movie,
 209             'entries': playlist,
 210         }
 211
 212
 213 class AppleTrailersSectionIE(InfoExtractor):
 214     IE_NAME = 'appletrailers:section'
 215     _SECTIONS = {
 216         'justadded': {
 217             'feed_path': 'just_added',
 218             'title': 'Just Added',
 219         },
 220         'exclusive': {
 221             'feed_path': 'exclusive',
 222             'title': 'Exclusive',
 223         },
 224         'justhd': {
 225             'feed_path': 'just_hd',
 226             'title': 'Just HD',
 227         },
 228         'mostpopular': {
 229             'feed_path': 'most_pop',
 230             'title': 'Most Popular',
 231         },
 232         'moviestudios': {
 233             'feed_path': 'studios',
 234             'title': 'Movie Studios',
 235         },
 236     }
 237     _VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/#section=(?P<id>%s)' % '|'.join(_SECTIONS)
 238     _TESTS = [{
 239         'url': 'http://trailers.apple.com/#section=justadded',
 240         'info_dict': {
 241             'title': 'Just Added',
 242             'id': 'justadded',
 243         },
 244         'playlist_mincount': 80,
 245     }, {
 246         'url': 'http://trailers.apple.com/#section=exclusive',
 247         'info_dict': {
 248             'title': 'Exclusive',
 249             'id': 'exclusive',
 250         },
 251         'playlist_mincount': 80,
 252     }, {
 253         'url': 'http://trailers.apple.com/#section=justhd',
 254         'info_dict': {
 255             'title': 'Just HD',
 256             'id': 'justhd',
 257         },
 258         'playlist_mincount': 80,
 259     }, {
 260         'url': 'http://trailers.apple.com/#section=mostpopular',
 261         'info_dict': {
 262             'title': 'Most Popular',
 263             'id': 'mostpopular',
 264         },
 265         'playlist_mincount': 30,
 266     }, {
 267         'url': 'http://trailers.apple.com/#section=moviestudios',
 268         'info_dict': {
 269             'title': 'Movie Studios',
 270             'id': 'moviestudios',
 271         },
 272         'playlist_mincount': 80,
 273     }]
 274
 275     def _real_extract(self, url):
 276         section = self._match_id(url)
 277         section_data = self._download_json(
 278             'http://trailers.apple.com/trailers/home/feeds/%s.json' % self._SECTIONS[section]['feed_path'],
 279             section)
 280         entries = [
 281             self.url_result('http://trailers.apple.com' + e['location'])
 282             for e in section_data]
 283         return self.playlist_result(entries, section, self._SECTIONS[section]['title'])