youtube_dl/extractor/imdb.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_str
   7 from ..utils import (
   8     determine_ext,
   9     mimetype2ext,
  10     parse_duration,
  11     qualities,
  12 )
  13
  14
  15 class ImdbIE(InfoExtractor):
  16     IE_NAME = 'imdb'
  17     IE_DESC = 'Internet Movie Database trailers'
  18     _VALID_URL = r'https?://(?:www|m)\.imdb\.com/(?:video|title|list).+?[/-]vi(?P<id>\d+)'
  19
  20     _TESTS = [{
  21         'url': 'http://www.imdb.com/video/imdb/vi2524815897',
  22         'info_dict': {
  23             'id': '2524815897',
  24             'ext': 'mp4',
  25             'title': 'No. 2 from Ice Age: Continental Drift (2012)',
  26             'description': 'md5:87bd0bdc61e351f21f20d2d7441cb4e7',
  27         }
  28     }, {
  29         'url': 'http://www.imdb.com/video/_/vi2524815897',
  30         'only_matching': True,
  31     }, {
  32         'url': 'http://www.imdb.com/title/tt1667889/?ref_=ext_shr_eml_vi#lb-vi2524815897',
  33         'only_matching': True,
  34     }, {
  35         'url': 'http://www.imdb.com/title/tt1667889/#lb-vi2524815897',
  36         'only_matching': True,
  37     }, {
  38         'url': 'http://www.imdb.com/videoplayer/vi1562949145',
  39         'only_matching': True,
  40     }, {
  41         'url': 'http://www.imdb.com/title/tt4218696/videoplayer/vi2608641561',
  42         'only_matching': True,
  43     }, {
  44         'url': 'https://www.imdb.com/list/ls009921623/videoplayer/vi260482329',
  45         'only_matching': True,
  46     }]
  47
  48     def _real_extract(self, url):
  49         video_id = self._match_id(url)
  50         webpage = self._download_webpage(
  51             'https://www.imdb.com/videoplayer/vi' + video_id, video_id)
  52         video_metadata = self._parse_json(self._search_regex(
  53             r'window\.IMDbReactInitialState\.push\(({.+?})\);', webpage,
  54             'video metadata'), video_id)['videos']['videoMetadata']['vi' + video_id]
  55         title = self._html_search_meta(
  56             ['og:title', 'twitter:title'], webpage) or self._html_search_regex(
  57             r'<title>(.+?)</title>', webpage, 'title', fatal=False) or video_metadata['title']
  58
  59         quality = qualities(('SD', '480p', '720p', '1080p'))
  60         formats = []
  61         for encoding in video_metadata.get('encodings', []):
  62             if not encoding or not isinstance(encoding, dict):
  63                 continue
  64             video_url = encoding.get('videoUrl')
  65             if not video_url or not isinstance(video_url, compat_str):
  66                 continue
  67             ext = determine_ext(video_url, mimetype2ext(encoding.get('mimeType')))
  68             if ext == 'm3u8':
  69                 formats.extend(self._extract_m3u8_formats(
  70                     video_url, video_id, 'mp4', entry_protocol='m3u8_native',
  71                     m3u8_id='hls', fatal=False))
  72                 continue
  73             format_id = encoding.get('definition')
  74             formats.append({
  75                 'format_id': format_id,
  76                 'url': video_url,
  77                 'ext': ext,
  78                 'quality': quality(format_id),
  79             })
  80         self._sort_formats(formats)
  81
  82         return {
  83             'id': video_id,
  84             'title': title,
  85             'formats': formats,
  86             'description': video_metadata.get('description'),
  87             'thumbnail': video_metadata.get('slate', {}).get('url'),
  88             'duration': parse_duration(video_metadata.get('duration')),
  89         }
  90
  91
  92 class ImdbListIE(InfoExtractor):
  93     IE_NAME = 'imdb:list'
  94     IE_DESC = 'Internet Movie Database lists'
  95     _VALID_URL = r'https?://(?:www\.)?imdb\.com/list/ls(?P<id>\d{9})(?!/videoplayer/vi\d+)'
  96     _TEST = {
  97         'url': 'https://www.imdb.com/list/ls009921623/',
  98         'info_dict': {
  99             'id': '009921623',
 100             'title': 'The Bourne Legacy',
 101             'description': 'A list of trailers, clips, and more from The Bourne Legacy, starring Jeremy Renner and Rachel Weisz.',
 102         },
 103         'playlist_count': 8,
 104     }
 105
 106     def _real_extract(self, url):
 107         list_id = self._match_id(url)
 108         webpage = self._download_webpage(url, list_id)
 109         entries = [
 110             self.url_result('http://www.imdb.com' + m, 'Imdb')
 111             for m in re.findall(r'href="(/list/ls%s/videoplayer/vi[^"]+)"' % list_id, webpage)]
 112
 113         list_title = self._html_search_regex(
 114             r'<h1[^>]+class="[^"]*header[^"]*"[^>]*>(.*?)</h1>',
 115             webpage, 'list title')
 116         list_description = self._html_search_regex(
 117             r'<div[^>]+class="[^"]*list-description[^"]*"[^>]*><p>(.*?)</p>',
 118             webpage, 'list description')
 119
 120         return self.playlist_result(entries, list_id, list_title, list_description)