[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / bostonglobe.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7
8 from ..utils import (
9     extract_attributes,
10 )
11
12
13 class BostonGlobeIE(InfoExtractor):
14     _VALID_URL = r'(?i)https?://(?:www\.)?bostonglobe\.com/.*/(?P<id>[^/]+)/\w+(?:\.html)?'
15     _TESTS = [
16         {
17             'url': 'http://www.bostonglobe.com/metro/2017/02/11/tree-finally-succumbs-disease-leaving-hole-neighborhood/h1b4lviqzMTIn9sVy8F3gP/story.html',
18             'md5': '0a62181079c85c2d2b618c9a738aedaf',
19             'info_dict': {
20                 'title': 'A tree finally succumbs to disease, leaving a hole in a neighborhood',
21                 'id': '5320421710001',
22                 'ext': 'mp4',
23                 'description': 'It arrived as a sapling when the Back Bay was in its infancy, a spindly American elm tamped down into a square of dirt cut into the brick sidewalk of 1880s Marlborough Street, no higher than the first bay window of the new brownstone behind it.',
24                 'timestamp': 1486877593,
25                 'upload_date': '20170212',
26                 'uploader_id': '245991542',
27             },
28         },
29         {
30             # Embedded youtube video; we hand it off to the Generic extractor.
31             'url': 'https://www.bostonglobe.com/lifestyle/names/2017/02/17/does-ben-affleck-play-matt-damon-favorite-version-batman/ruqkc9VxKBYmh5txn1XhSI/story.html',
32             'md5': '582b40327089d5c0c949b3c54b13c24b',
33             'info_dict': {
34                 'title': "Who Is Matt Damon's Favorite Batman?",
35                 'id': 'ZW1QCnlA6Qc',
36                 'ext': 'mp4',
37                 'upload_date': '20170217',
38                 'description': 'md5:3b3dccb9375867e0b4d527ed87d307cb',
39                 'uploader': 'The Late Late Show with James Corden',
40                 'uploader_id': 'TheLateLateShow',
41             },
42             'expected_warnings': ['404'],
43         },
44     ]
45
46     def _real_extract(self, url):
47         page_id = self._match_id(url)
48         webpage = self._download_webpage(url, page_id)
49
50         page_title = self._og_search_title(webpage, default=None)
51
52         # <video data-brightcove-video-id="5320421710001" data-account="245991542" data-player="SJWAiyYWg" data-embed="default" class="video-js" controls itemscope itemtype="http://schema.org/VideoObject">
53         entries = []
54         for video in re.findall(r'(?i)(<video[^>]+>)', webpage):
55             attrs = extract_attributes(video)
56
57             video_id = attrs.get('data-brightcove-video-id')
58             account_id = attrs.get('data-account')
59             player_id = attrs.get('data-player')
60             embed = attrs.get('data-embed')
61
62             if video_id and account_id and player_id and embed:
63                 entries.append(
64                     'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
65                     % (account_id, player_id, embed, video_id))
66
67         if len(entries) == 0:
68             return self.url_result(url, 'Generic')
69         elif len(entries) == 1:
70             return self.url_result(entries[0], 'BrightcoveNew')
71         else:
72             return self.playlist_from_matches(entries, page_id, page_title, ie='BrightcoveNew')