yt_dlp/extractor/ant1newsgr.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import urllib.parse
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     HEADRequest,
  10     ExtractorError,
  11     determine_ext,
  12     scale_thumbnails_to_max_format_width,
  13     unescapeHTML,
  14 )
  15
  16
  17 class Ant1NewsGrBaseIE(InfoExtractor):
  18     def _download_and_extract_api_data(self, video_id, netloc, cid=None):
  19         url = f'{self.http_scheme()}//{netloc}{self._API_PATH}'
  20         info = self._download_json(url, video_id, query={'cid': cid or video_id})
  21         try:
  22             source = info['url']
  23         except KeyError:
  24             raise ExtractorError('no source found for %s' % video_id)
  25         formats, subs = (self._extract_m3u8_formats_and_subtitles(source, video_id, 'mp4')
  26                          if determine_ext(source) == 'm3u8' else ([{'url': source}], {}))
  27         self._sort_formats(formats)
  28         thumbnails = scale_thumbnails_to_max_format_width(
  29             formats, [{'url': info['thumb']}], r'(?<=/imgHandler/)\d+')
  30         return {
  31             'id': video_id,
  32             'title': info.get('title'),
  33             'thumbnails': thumbnails,
  34             'formats': formats,
  35             'subtitles': subs,
  36         }
  37
  38
  39 class Ant1NewsGrWatchIE(Ant1NewsGrBaseIE):
  40     IE_NAME = 'ant1newsgr:watch'
  41     IE_DESC = 'ant1news.gr videos'
  42     _VALID_URL = r'https?://(?P<netloc>(?:www\.)?ant1news\.gr)/watch/(?P<id>\d+)/'
  43     _API_PATH = '/templates/data/player'
  44
  45     _TESTS = [{
  46         'url': 'https://www.ant1news.gr/watch/1506168/ant1-news-09112021-stis-18-45',
  47         'md5': '95925e6b32106754235f2417e0d2dfab',
  48         'info_dict': {
  49             'id': '1506168',
  50             'ext': 'mp4',
  51             'title': 'md5:0ad00fa66ecf8aa233d26ab0dba7514a',
  52             'description': 'md5:18665af715a6dcfeac1d6153a44f16b0',
  53             'thumbnail': 'https://ant1media.azureedge.net/imgHandler/640/26d46bf6-8158-4f02-b197-7096c714b2de.jpg',
  54         },
  55     }]
  56
  57     def _real_extract(self, url):
  58         video_id, netloc = self._match_valid_url(url).group('id', 'netloc')
  59         webpage = self._download_webpage(url, video_id)
  60         info = self._download_and_extract_api_data(video_id, netloc)
  61         info['description'] = self._og_search_description(webpage)
  62         return info
  63
  64
  65 class Ant1NewsGrArticleIE(Ant1NewsGrBaseIE):
  66     IE_NAME = 'ant1newsgr:article'
  67     IE_DESC = 'ant1news.gr articles'
  68     _VALID_URL = r'https?://(?:www\.)?ant1news\.gr/[^/]+/article/(?P<id>\d+)/'
  69
  70     _TESTS = [{
  71         'url': 'https://www.ant1news.gr/afieromata/article/549468/o-tzeims-mpont-sta-meteora-oi-apeiles-kai-o-xesikomos-ton-kalogeron',
  72         'md5': '294f18331bb516539d72d85a82887dcc',
  73         'info_dict': {
  74             'id': '_xvg/m_cmbatw=',
  75             'ext': 'mp4',
  76             'title': 'md5:a93e8ecf2e4073bfdffcb38f59945411',
  77             'timestamp': 1603092840,
  78             'upload_date': '20201019',
  79             'thumbnail': 'https://ant1media.azureedge.net/imgHandler/640/756206d2-d640-40e2-b201-3555abdfc0db.jpg',
  80         },
  81     }, {
  82         'url': 'https://ant1news.gr/Society/article/620286/symmoria-anilikon-dikigoros-thymaton-ithelan-na-toys-apoteleiosoyn',
  83         'info_dict': {
  84             'id': '620286',
  85             'title': 'md5:91fe569e952e4d146485740ae927662b',
  86         },
  87         'playlist_mincount': 2,
  88         'params': {
  89             'skip_download': True,
  90         },
  91     }]
  92
  93     def _real_extract(self, url):
  94         video_id = self._match_id(url)
  95         webpage = self._download_webpage(url, video_id)
  96         info = self._search_json_ld(webpage, video_id, expected_type='NewsArticle')
  97         embed_urls = list(Ant1NewsGrEmbedIE._extract_urls(webpage))
  98         if not embed_urls:
  99             raise ExtractorError('no videos found for %s' % video_id, expected=True)
 100         return self.playlist_from_matches(
 101             embed_urls, video_id, info.get('title'), ie=Ant1NewsGrEmbedIE.ie_key(),
 102             video_kwargs={'url_transparent': True, 'timestamp': info.get('timestamp')})
 103
 104
 105 class Ant1NewsGrEmbedIE(Ant1NewsGrBaseIE):
 106     IE_NAME = 'ant1newsgr:embed'
 107     IE_DESC = 'ant1news.gr embedded videos'
 108     _BASE_PLAYER_URL_RE = r'(?:https?:)?//(?:[a-zA-Z0-9\-]+\.)?(?:antenna|ant1news)\.gr/templates/pages/player'
 109     _VALID_URL = rf'{_BASE_PLAYER_URL_RE}\?([^#]+&)?cid=(?P<id>[^#&]+)'
 110     _API_PATH = '/news/templates/data/jsonPlayer'
 111
 112     _TESTS = [{
 113         'url': 'https://www.antenna.gr/templates/pages/player?cid=3f_li_c_az_jw_y_u=&w=670&h=377',
 114         'md5': 'dfc58c3a11a5a9aad2ba316ed447def3',
 115         'info_dict': {
 116             'id': '3f_li_c_az_jw_y_u=',
 117             'ext': 'mp4',
 118             'title': 'md5:a30c93332455f53e1e84ae0724f0adf7',
 119             'thumbnail': 'https://ant1media.azureedge.net/imgHandler/640/bbe31201-3f09-4a4e-87f5-8ad2159fffe2.jpg',
 120         },
 121     }]
 122
 123     @classmethod
 124     def _extract_urls(cls, webpage):
 125         _EMBED_URL_RE = rf'{cls._BASE_PLAYER_URL_RE}\?(?:(?!(?P=_q1)).)+'
 126         _EMBED_RE = rf'<iframe[^>]+?src=(?P<_q1>["\'])(?P<url>{_EMBED_URL_RE})(?P=_q1)'
 127         for mobj in re.finditer(_EMBED_RE, webpage):
 128             url = unescapeHTML(mobj.group('url'))
 129             if not cls.suitable(url):
 130                 continue
 131             yield url
 132
 133     def _real_extract(self, url):
 134         video_id = self._match_id(url)
 135
 136         canonical_url = self._request_webpage(
 137             HEADRequest(url), video_id,
 138             note='Resolve canonical player URL',
 139             errnote='Could not resolve canonical player URL').geturl()
 140         _, netloc, _, _, query, _ = urllib.parse.urlparse(canonical_url)
 141         cid = urllib.parse.parse_qs(query)['cid'][0]
 142
 143         return self._download_and_extract_api_data(video_id, netloc, cid=cid)