[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / appleconnect.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from .common import InfoExtractor
5 from ..utils import (
6     str_to_int,
7     ExtractorError
8 )
9
10
11 class AppleConnectIE(InfoExtractor):
12     _VALID_URL = r'https?://itunes\.apple\.com/\w{0,2}/?post/idsa\.(?P<id>[\w-]+)'
13     _TEST = {
14         'url': 'https://itunes.apple.com/us/post/idsa.4ab17a39-2720-11e5-96c5-a5b38f6c42d3',
15         'md5': 'e7c38568a01ea45402570e6029206723',
16         'info_dict': {
17             'id': '4ab17a39-2720-11e5-96c5-a5b38f6c42d3',
18             'ext': 'm4v',
19             'title': 'Energy',
20             'uploader': 'Drake',
21             'thumbnail': r're:^https?://.*\.jpg$',
22             'upload_date': '20150710',
23             'timestamp': 1436545535,
24         },
25     }
26
27     def _real_extract(self, url):
28         video_id = self._match_id(url)
29         webpage = self._download_webpage(url, video_id)
30
31         try:
32             video_json = self._html_search_regex(
33                 r'class="auc-video-data">(\{.*?\})', webpage, 'json')
34         except ExtractorError:
35             raise ExtractorError('This post doesn\'t contain a video', expected=True)
36
37         video_data = self._parse_json(video_json, video_id)
38         timestamp = str_to_int(self._html_search_regex(r'data-timestamp="(\d+)"', webpage, 'timestamp'))
39         like_count = str_to_int(self._html_search_regex(r'(\d+) Loves', webpage, 'like count'))
40
41         return {
42             'id': video_id,
43             'url': video_data['sslSrc'],
44             'title': video_data['title'],
45             'description': video_data['description'],
46             'uploader': video_data['artistName'],
47             'thumbnail': video_data['artworkUrl'],
48             'timestamp': timestamp,
49             'like_count': like_count,
50         }