[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / tele13.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from .common import InfoExtractor
5 from .youtube import YoutubeIE
6 from ..utils import (
7     js_to_json,
8     qualities,
9     determine_ext,
10 )
11
12
13 class Tele13IE(InfoExtractor):
14     _VALID_URL = r'^https?://(?:www\.)?t13\.cl/videos(?:/[^/]+)+/(?P<id>[\w-]+)'
15     _TESTS = [
16         {
17             'url': 'http://www.t13.cl/videos/actualidad/el-circulo-de-hierro-de-michelle-bachelet-en-su-regreso-a-la-moneda',
18             'md5': '4cb1fa38adcad8fea88487a078831755',
19             'info_dict': {
20                 'id': 'el-circulo-de-hierro-de-michelle-bachelet-en-su-regreso-a-la-moneda',
21                 'ext': 'mp4',
22                 'title': 'El círculo de hierro de Michelle Bachelet en su regreso a La Moneda',
23             },
24             'params': {
25                 # HTTP Error 404: Not Found
26                 'skip_download': True,
27             },
28         },
29         {
30             'url': 'http://www.t13.cl/videos/mundo/tendencias/video-captan-misteriosa-bola-fuego-cielos-bangkok',
31             'md5': '867adf6a3b3fef932c68a71d70b70946',
32             'info_dict': {
33                 'id': 'rOoKv2OMpOw',
34                 'ext': 'mp4',
35                 'title': 'Shooting star seen on 7-Sep-2015',
36                 'description': 'md5:7292ff2a34b2f673da77da222ae77e1e',
37                 'uploader': 'Porjai Jaturongkhakun',
38                 'upload_date': '20150906',
39                 'uploader_id': 'UCnLY_3ezwNcDSC_Wc6suZxw',
40             },
41             'add_ie': ['Youtube'],
42         }
43     ]
44
45     def _real_extract(self, url):
46         display_id = self._match_id(url)
47         webpage = self._download_webpage(url, display_id)
48
49         setup_js = self._search_regex(
50             r"(?s)jwplayer\('player-vivo'\).setup\((\{.*?\})\)",
51             webpage, 'setup code')
52         sources = self._parse_json(self._search_regex(
53             r'sources\s*:\s*(\[[^\]]+\])', setup_js, 'sources'),
54             display_id, js_to_json)
55
56         preference = qualities(['Móvil', 'SD', 'HD'])
57         formats = []
58         urls = []
59         for f in sources:
60             format_url = f['file']
61             if format_url and format_url not in urls:
62                 ext = determine_ext(format_url)
63                 if ext == 'm3u8':
64                     formats.extend(self._extract_m3u8_formats(
65                         format_url, display_id, 'mp4', 'm3u8_native',
66                         m3u8_id='hls', fatal=False))
67                 elif YoutubeIE.suitable(format_url):
68                     return self.url_result(format_url, 'Youtube')
69                 else:
70                     formats.append({
71                         'url': format_url,
72                         'format_id': f.get('label'),
73                         'preference': preference(f.get('label')),
74                         'ext': ext,
75                     })
76                 urls.append(format_url)
77         self._sort_formats(formats)
78
79         return {
80             'id': display_id,
81             'title': self._search_regex(
82                 r'title\s*:\s*"([^"]+)"', setup_js, 'title'),
83             'description': self._html_search_meta(
84                 'description', webpage, 'description'),
85             'thumbnail': self._search_regex(
86                 r'image\s*:\s*"([^"]+)"', setup_js, 'thumbnail', default=None),
87             'formats': formats,
88         }