[imdb] Improve _VALID_URL (Closes #9788)
[youtube-dl] / youtube_dl / extractor / imdb.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..utils import (
7     mimetype2ext,
8     qualities,
9 )
10
11
12 class ImdbIE(InfoExtractor):
13     IE_NAME = 'imdb'
14     IE_DESC = 'Internet Movie Database trailers'
15     _VALID_URL = r'https?://(?:www|m)\.imdb\.com/(?:video/[^/]+/|title/tt\d+.*?#lb-)vi(?P<id>\d+)'
16
17     _TESTS = [{
18         'url': 'http://www.imdb.com/video/imdb/vi2524815897',
19         'info_dict': {
20             'id': '2524815897',
21             'ext': 'mp4',
22             'title': 'Ice Age: Continental Drift Trailer (No. 2) - IMDb',
23             'description': 'md5:9061c2219254e5d14e03c25c98e96a81',
24         }
25     }, {
26         'url': 'http://www.imdb.com/video/_/vi2524815897',
27         'only_matching': True,
28     }, {
29         'url': 'http://www.imdb.com/title/tt1667889/?ref_=ext_shr_eml_vi#lb-vi2524815897',
30         'only_matching': True,
31     }, {
32         'url': 'http://www.imdb.com/title/tt1667889/#lb-vi2524815897',
33         'only_matching': True,
34     }]
35
36     def _real_extract(self, url):
37         video_id = self._match_id(url)
38         webpage = self._download_webpage('http://www.imdb.com/video/imdb/vi%s' % video_id, video_id)
39         descr = self._html_search_regex(
40             r'(?s)<span itemprop="description">(.*?)</span>',
41             webpage, 'description', fatal=False)
42         player_url = 'http://www.imdb.com/video/imdb/vi%s/imdb/single' % video_id
43         player_page = self._download_webpage(
44             player_url, video_id, 'Downloading player page')
45         # the player page contains the info for the default format, we have to
46         # fetch other pages for the rest of the formats
47         extra_formats = re.findall(r'href="(?P<url>%s.*?)".*?>(?P<name>.*?)<' % re.escape(player_url), player_page)
48         format_pages = [
49             self._download_webpage(
50                 f_url, video_id, 'Downloading info for %s format' % f_name)
51             for f_url, f_name in extra_formats]
52         format_pages.append(player_page)
53
54         quality = qualities(('SD', '480p', '720p', '1080p'))
55         formats = []
56         for format_page in format_pages:
57             json_data = self._search_regex(
58                 r'<script[^>]+class="imdb-player-data"[^>]*?>(.*?)</script>',
59                 format_page, 'json data', flags=re.DOTALL)
60             info = self._parse_json(json_data, video_id, fatal=False)
61             if not info:
62                 continue
63             format_info = info.get('videoPlayerObject', {}).get('video', {})
64             if not format_info:
65                 continue
66             video_info_list = format_info.get('videoInfoList')
67             if not video_info_list or not isinstance(video_info_list, list):
68                 continue
69             video_info = video_info_list[0]
70             if not video_info or not isinstance(video_info, dict):
71                 continue
72             video_url = video_info.get('videoUrl')
73             if not video_url:
74                 continue
75             format_id = format_info.get('ffname')
76             formats.append({
77                 'format_id': format_id,
78                 'url': video_url,
79                 'ext': mimetype2ext(video_info.get('videoMimeType')),
80                 'quality': quality(format_id),
81             })
82         self._sort_formats(formats)
83
84         return {
85             'id': video_id,
86             'title': self._og_search_title(webpage),
87             'formats': formats,
88             'description': descr,
89             'thumbnail': format_info['slate'],
90         }
91
92
93 class ImdbListIE(InfoExtractor):
94     IE_NAME = 'imdb:list'
95     IE_DESC = 'Internet Movie Database lists'
96     _VALID_URL = r'https?://www\.imdb\.com/list/(?P<id>[\da-zA-Z_-]{11})'
97     _TEST = {
98         'url': 'http://www.imdb.com/list/JFs9NWw6XI0',
99         'info_dict': {
100             'id': 'JFs9NWw6XI0',
101             'title': 'March 23, 2012 Releases',
102         },
103         'playlist_count': 7,
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="(/video/imdb/vi[^"]+)"\s+data-type="playlist"', webpage)]
112
113         list_title = self._html_search_regex(
114             r'<h1 class="header">(.*?)</h1>', webpage, 'list title')
115
116         return self.playlist_result(entries, list_id, list_title)