youtube_dl/extractor/imdb.py

   1 from __future__ import unicode_literals
   2
   3 import base64
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     determine_ext,
  10     mimetype2ext,
  11     parse_duration,
  12     qualities,
  13     try_get,
  14     url_or_none,
  15 )
  16
  17
  18 class ImdbIE(InfoExtractor):
  19     IE_NAME = 'imdb'
  20     IE_DESC = 'Internet Movie Database trailers'
  21     _VALID_URL = r'https?://(?:www|m)\.imdb\.com/(?:video|title|list).*?[/-]vi(?P<id>\d+)'
  22
  23     _TESTS = [{
  24         'url': 'http://www.imdb.com/video/imdb/vi2524815897',
  25         'info_dict': {
  26             'id': '2524815897',
  27             'ext': 'mp4',
  28             'title': 'No. 2',
  29             'description': 'md5:87bd0bdc61e351f21f20d2d7441cb4e7',
  30             'duration': 152,
  31         }
  32     }, {
  33         'url': 'http://www.imdb.com/video/_/vi2524815897',
  34         'only_matching': True,
  35     }, {
  36         'url': 'http://www.imdb.com/title/tt1667889/?ref_=ext_shr_eml_vi#lb-vi2524815897',
  37         'only_matching': True,
  38     }, {
  39         'url': 'http://www.imdb.com/title/tt1667889/#lb-vi2524815897',
  40         'only_matching': True,
  41     }, {
  42         'url': 'http://www.imdb.com/videoplayer/vi1562949145',
  43         'only_matching': True,
  44     }, {
  45         'url': 'http://www.imdb.com/title/tt4218696/videoplayer/vi2608641561',
  46         'only_matching': True,
  47     }, {
  48         'url': 'https://www.imdb.com/list/ls009921623/videoplayer/vi260482329',
  49         'only_matching': True,
  50     }]
  51
  52     def _real_extract(self, url):
  53         video_id = self._match_id(url)
  54
  55         data = self._download_json(
  56             'https://www.imdb.com/ve/data/VIDEO_PLAYBACK_DATA', video_id,
  57             query={
  58                 'key': base64.b64encode(json.dumps({
  59                     'type': 'VIDEO_PLAYER',
  60                     'subType': 'FORCE_LEGACY',
  61                     'id': 'vi%s' % video_id,
  62                 }).encode()).decode(),
  63             })[0]
  64
  65         quality = qualities(('SD', '480p', '720p', '1080p'))
  66         formats = []
  67         for encoding in data['videoLegacyEncodings']:
  68             if not encoding or not isinstance(encoding, dict):
  69                 continue
  70             video_url = url_or_none(encoding.get('url'))
  71             if not video_url:
  72                 continue
  73             ext = mimetype2ext(encoding.get(
  74                 'mimeType')) or determine_ext(video_url)
  75             if ext == 'm3u8':
  76                 formats.extend(self._extract_m3u8_formats(
  77                     video_url, video_id, 'mp4', entry_protocol='m3u8_native',
  78                     preference=1, m3u8_id='hls', fatal=False))
  79                 continue
  80             format_id = encoding.get('definition')
  81             formats.append({
  82                 'format_id': format_id,
  83                 'url': video_url,
  84                 'ext': ext,
  85                 'quality': quality(format_id),
  86             })
  87         self._sort_formats(formats)
  88
  89         webpage = self._download_webpage(
  90             'https://www.imdb.com/video/vi' + video_id, video_id)
  91         video_metadata = self._parse_json(self._search_regex(
  92             r'args\.push\(\s*({.+?})\s*\)\s*;', webpage,
  93             'video metadata'), video_id)
  94
  95         video_info = video_metadata.get('VIDEO_INFO')
  96         if video_info and isinstance(video_info, dict):
  97             info = try_get(
  98                 video_info, lambda x: x[list(video_info.keys())[0]][0], dict)
  99         else:
 100             info = {}
 101
 102         title = self._html_search_meta(
 103             ['og:title', 'twitter:title'], webpage) or self._html_search_regex(
 104             r'<title>(.+?)</title>', webpage, 'title',
 105             default=None) or info['videoTitle']
 106
 107         return {
 108             'id': video_id,
 109             'title': title,
 110             'alt_title': info.get('videoSubTitle'),
 111             'formats': formats,
 112             'description': info.get('videoDescription'),
 113             'thumbnail': url_or_none(try_get(
 114                 video_metadata, lambda x: x['videoSlate']['source'])),
 115             'duration': parse_duration(info.get('videoRuntime')),
 116         }
 117
 118
 119 class ImdbListIE(InfoExtractor):
 120     IE_NAME = 'imdb:list'
 121     IE_DESC = 'Internet Movie Database lists'
 122     _VALID_URL = r'https?://(?:www\.)?imdb\.com/list/ls(?P<id>\d{9})(?!/videoplayer/vi\d+)'
 123     _TEST = {
 124         'url': 'https://www.imdb.com/list/ls009921623/',
 125         'info_dict': {
 126             'id': '009921623',
 127             'title': 'The Bourne Legacy',
 128             'description': 'A list of trailers, clips, and more from The Bourne Legacy, starring Jeremy Renner and Rachel Weisz.',
 129         },
 130         'playlist_count': 8,
 131     }
 132
 133     def _real_extract(self, url):
 134         list_id = self._match_id(url)
 135         webpage = self._download_webpage(url, list_id)
 136         entries = [
 137             self.url_result('http://www.imdb.com' + m, 'Imdb')
 138             for m in re.findall(r'href="(/list/ls%s/videoplayer/vi[^"]+)"' % list_id, webpage)]
 139
 140         list_title = self._html_search_regex(
 141             r'<h1[^>]+class="[^"]*header[^"]*"[^>]*>(.*?)</h1>',
 142             webpage, 'list title')
 143         list_description = self._html_search_regex(
 144             r'<div[^>]+class="[^"]*list-description[^"]*"[^>]*><p>(.*?)</p>',
 145             webpage, 'list description')
 146
 147         return self.playlist_result(entries, list_id, list_title, list_description)