youtube_dl/extractor/tvnet.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     int_or_none,
   9     unescapeHTML,
  10     url_or_none,
  11 )
  12
  13
  14 class TVNetIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:[^/]+)\.tvnet\.gov\.vn/[^/]+/(?:\d+/)?(?P<id>\d+)(?:/|$)'
  16     _TESTS = [{
  17         # video
  18         'url': 'http://de.tvnet.gov.vn/video/109788/vtv1---bac-tuyet-tai-lao-cai-va-ha-giang/tin-nong-24h',
  19         'md5': 'b4d7abe0252c9b47774760b7519c7558',
  20         'info_dict': {
  21             'id': '109788',
  22             'ext': 'mp4',
  23             'title': 'VTV1 - Bắc tuyết tại Lào Cai và Hà Giang',
  24             'thumbnail': r're:(?i)https?://.*\.(?:jpg|png)',
  25             'is_live': False,
  26             'view_count': int,
  27         },
  28     }, {
  29         # audio
  30         'url': 'http://vn.tvnet.gov.vn/radio/27017/vov1---ban-tin-chieu-10062018/doi-song-va-xa-hoi',
  31         'md5': 'b5875ce9b0a2eecde029216d0e6db2ae',
  32         'info_dict': {
  33             'id': '27017',
  34             'ext': 'm4a',
  35             'title': 'VOV1 - Bản tin chiều (10/06/2018)',
  36             'thumbnail': r're:(?i)https?://.*\.(?:jpg|png)',
  37             'is_live': False,
  38         },
  39     }, {
  40         'url': 'http://us.tvnet.gov.vn/video/118023/129999/ngay-0705',
  41         'info_dict': {
  42             'id': '129999',
  43             'ext': 'mp4',
  44             'title': 'VTV1 - Quốc hội với cử tri (11/06/2018)',
  45             'thumbnail': r're:(?i)https?://.*\.(?:jpg|png)',
  46             'is_live': False,
  47         },
  48         'params': {
  49             'skip_download': True,
  50         },
  51     }, {
  52         # live stream
  53         'url': 'http://us.tvnet.gov.vn/kenh-truyen-hinh/1011/vtv1',
  54         'info_dict': {
  55             'id': '1011',
  56             'ext': 'mp4',
  57             'title': r're:^VTV1 \| LiveTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
  58             'thumbnail': r're:(?i)https?://.*\.(?:jpg|png)',
  59             'is_live': True,
  60         },
  61         'params': {
  62             'skip_download': True,
  63         },
  64     }, {
  65         # radio live stream
  66         'url': 'http://vn.tvnet.gov.vn/kenh-truyen-hinh/1014',
  67         'info_dict': {
  68             'id': '1014',
  69             'ext': 'm4a',
  70             'title': r're:VOV1 \| LiveTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
  71             'thumbnail': r're:(?i)https?://.*\.(?:jpg|png)',
  72             'is_live': True,
  73         },
  74         'params': {
  75             'skip_download': True,
  76         },
  77     }, {
  78         'url': 'http://us.tvnet.gov.vn/phim/6136/25510/vtv3---ca-mot-doi-an-oan-tap-1-50/phim-truyen-hinh',
  79         'only_matching': True,
  80     }]
  81
  82     def _real_extract(self, url):
  83         video_id = self._match_id(url)
  84
  85         webpage = self._download_webpage(url, video_id)
  86
  87         title = self._og_search_title(
  88             webpage, default=None) or self._html_search_meta(
  89             'title', webpage, default=None) or self._search_regex(
  90             r'<title>([^<]+)<', webpage, 'title')
  91         title = re.sub(r'\s*-\s*TV Net\s*$', '', title)
  92
  93         if '/video/' in url or '/radio/' in url:
  94             is_live = False
  95         elif '/kenh-truyen-hinh/' in url:
  96             is_live = True
  97         else:
  98             is_live = None
  99
 100         data_file = unescapeHTML(self._search_regex(
 101             r'data-file=(["\'])(?P<url>(?:https?:)?//.+?)\1', webpage,
 102             'data file', group='url'))
 103
 104         stream_urls = set()
 105         formats = []
 106         for stream in self._download_json(data_file, video_id):
 107             if not isinstance(stream, dict):
 108                 continue
 109             stream_url = url_or_none(stream.get('url'))
 110             if stream_url in stream_urls or not stream_url:
 111                 continue
 112             stream_urls.add(stream_url)
 113             formats.extend(self._extract_m3u8_formats(
 114                 stream_url, video_id, 'mp4',
 115                 entry_protocol='m3u8' if is_live else 'm3u8_native',
 116                 m3u8_id='hls', fatal=False))
 117         self._sort_formats(formats)
 118
 119         # better support for radio streams
 120         if title.startswith('VOV'):
 121             for f in formats:
 122                 f.update({
 123                     'ext': 'm4a',
 124                     'vcodec': 'none',
 125                 })
 126
 127         thumbnail = self._og_search_thumbnail(
 128             webpage, default=None) or unescapeHTML(
 129             self._search_regex(
 130                 r'data-image=(["\'])(?P<url>(?:https?:)?//.+?)\1', webpage,
 131                 'thumbnail', default=None, group='url'))
 132
 133         if is_live:
 134             title = self._live_title(title)
 135
 136         view_count = int_or_none(self._search_regex(
 137             r'(?s)<div[^>]+\bclass=["\'].*?view-count[^>]+>.*?(\d+).*?</div>',
 138             webpage, 'view count', default=None))
 139
 140         return {
 141             'id': video_id,
 142             'title': title,
 143             'thumbnail': thumbnail,
 144             'is_live': is_live,
 145             'view_count': view_count,
 146             'formats': formats,
 147         }