yt_dlp/extractor/tvnet.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     int_or_none,
   6     unescapeHTML,
   7     url_or_none,
   8 )
   9
  10
  11 class TVNetIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:[^/]+)\.tvnet\.gov\.vn/[^/]+/(?:\d+/)?(?P<id>\d+)(?:/|$)'
  13     _TESTS = [{
  14         # video
  15         'url': 'http://de.tvnet.gov.vn/video/109788/vtv1---bac-tuyet-tai-lao-cai-va-ha-giang/tin-nong-24h',
  16         'md5': 'b4d7abe0252c9b47774760b7519c7558',
  17         'info_dict': {
  18             'id': '109788',
  19             'ext': 'mp4',
  20             'title': 'VTV1 - Bắc tuyết tại Lào Cai và Hà Giang',
  21             'thumbnail': r're:(?i)https?://.*\.(?:jpg|png)',
  22             'is_live': False,
  23             'view_count': int,
  24         },
  25     }, {
  26         # audio
  27         'url': 'http://vn.tvnet.gov.vn/radio/27017/vov1---ban-tin-chieu-10062018/doi-song-va-xa-hoi',
  28         'md5': 'b5875ce9b0a2eecde029216d0e6db2ae',
  29         'info_dict': {
  30             'id': '27017',
  31             'ext': 'm4a',
  32             'title': 'VOV1 - Bản tin chiều (10/06/2018)',
  33             'thumbnail': r're:(?i)https?://.*\.(?:jpg|png)',
  34             'is_live': False,
  35         },
  36     }, {
  37         'url': 'http://us.tvnet.gov.vn/video/118023/129999/ngay-0705',
  38         'info_dict': {
  39             'id': '129999',
  40             'ext': 'mp4',
  41             'title': 'VTV1 - Quốc hội với cử tri (11/06/2018)',
  42             'thumbnail': r're:(?i)https?://.*\.(?:jpg|png)',
  43             'is_live': False,
  44         },
  45         'params': {
  46             'skip_download': True,
  47         },
  48     }, {
  49         # live stream
  50         'url': 'http://us.tvnet.gov.vn/kenh-truyen-hinh/1011/vtv1',
  51         'info_dict': {
  52             'id': '1011',
  53             'ext': 'mp4',
  54             'title': r're:^VTV1 \| LiveTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
  55             'thumbnail': r're:(?i)https?://.*\.(?:jpg|png)',
  56             'is_live': True,
  57         },
  58         'params': {
  59             'skip_download': True,
  60         },
  61     }, {
  62         # radio live stream
  63         'url': 'http://vn.tvnet.gov.vn/kenh-truyen-hinh/1014',
  64         'info_dict': {
  65             'id': '1014',
  66             'ext': 'm4a',
  67             'title': r're:VOV1 \| LiveTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
  68             'thumbnail': r're:(?i)https?://.*\.(?:jpg|png)',
  69             'is_live': True,
  70         },
  71         'params': {
  72             'skip_download': True,
  73         },
  74     }, {
  75         'url': 'http://us.tvnet.gov.vn/phim/6136/25510/vtv3---ca-mot-doi-an-oan-tap-1-50/phim-truyen-hinh',
  76         'only_matching': True,
  77     }]
  78
  79     def _real_extract(self, url):
  80         video_id = self._match_id(url)
  81
  82         webpage = self._download_webpage(url, video_id)
  83
  84         title = self._og_search_title(
  85             webpage, default=None) or self._html_search_meta(
  86             'title', webpage, default=None) or self._search_regex(
  87             r'<title>([^<]+)<', webpage, 'title')
  88         title = re.sub(r'\s*-\s*TV Net\s*$', '', title)
  89
  90         if '/video/' in url or '/radio/' in url:
  91             is_live = False
  92         elif '/kenh-truyen-hinh/' in url:
  93             is_live = True
  94         else:
  95             is_live = None
  96
  97         data_file = unescapeHTML(self._search_regex(
  98             r'data-file=(["\'])(?P<url>(?:https?:)?//.+?)\1', webpage,
  99             'data file', group='url'))
 100
 101         stream_urls = set()
 102         formats = []
 103         for stream in self._download_json(data_file, video_id):
 104             if not isinstance(stream, dict):
 105                 continue
 106             stream_url = url_or_none(stream.get('url'))
 107             if stream_url in stream_urls or not stream_url:
 108                 continue
 109             stream_urls.add(stream_url)
 110             formats.extend(self._extract_m3u8_formats(
 111                 stream_url, video_id, 'mp4', live=is_live, m3u8_id='hls', fatal=False))
 112
 113         # better support for radio streams
 114         if title.startswith('VOV'):
 115             for f in formats:
 116                 f.update({
 117                     'ext': 'm4a',
 118                     'vcodec': 'none',
 119                 })
 120
 121         thumbnail = self._og_search_thumbnail(
 122             webpage, default=None) or unescapeHTML(
 123             self._search_regex(
 124                 r'data-image=(["\'])(?P<url>(?:https?:)?//.+?)\1', webpage,
 125                 'thumbnail', default=None, group='url'))
 126
 127         view_count = int_or_none(self._search_regex(
 128             r'(?s)<div[^>]+\bclass=["\'].*?view-count[^>]+>.*?(\d+).*?</div>',
 129             webpage, 'view count', default=None))
 130
 131         return {
 132             'id': video_id,
 133             'title': title,
 134             'thumbnail': thumbnail,
 135             'is_live': is_live,
 136             'view_count': view_count,
 137             'formats': formats,
 138         }