yt_dlp/extractor/la7.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     float_or_none,
   6     HEADRequest,
   7     int_or_none,
   8     parse_duration,
   9     unified_strdate,
  10 )
  11
  12
  13 class LA7IE(InfoExtractor):
  14     IE_NAME = 'la7.it'
  15     _VALID_URL = r'''(?x)https?://(?:
  16         (?:www\.)?la7\.it/([^/]+)/(?:rivedila7|video|news)/|
  17         tg\.la7\.it/repliche-tgla7\?id=
  18     )(?P<id>.+)'''
  19
  20     _TESTS = [{
  21         # single quality video
  22         'url': 'http://www.la7.it/crozza/video/inccool8-02-10-2015-163722',
  23         'md5': '8b613ffc0c4bf9b9e377169fc19c214c',
  24         'info_dict': {
  25             'id': 'inccool8-02-10-2015-163722',
  26             'ext': 'mp4',
  27             'title': 'Inc.Cool8',
  28             'description': 'Benvenuti nell\'incredibile mondo della INC. COOL. 8. dove “INC.” sta per “Incorporated” “COOL” sta per “fashion” ed Eight sta per il gesto atletico',
  29             'thumbnail': 're:^https?://.*',
  30             'upload_date': '20151002',
  31             'formats': 'count:4',
  32         },
  33     }, {
  34         # multiple quality video
  35         'url': 'https://www.la7.it/calcio-femminile/news/il-gol-di-lindsey-thomas-fiorentina-vs-milan-serie-a-calcio-femminile-26-11-2022-461736',
  36         'md5': 'd2370e78f75e8d1238cb3a0db9a2eda3',
  37         'info_dict': {
  38             'id': 'il-gol-di-lindsey-thomas-fiorentina-vs-milan-serie-a-calcio-femminile-26-11-2022-461736',
  39             'ext': 'mp4',
  40             'title': 'Il gol di Lindsey Thomas | Fiorentina vs Milan | Serie A Calcio Femminile',
  41             'description': 'Il gol di Lindsey Thomas | Fiorentina vs Milan | Serie A Calcio Femminile',
  42             'thumbnail': 're:^https?://.*',
  43             'upload_date': '20221126',
  44             'formats': 'count:8',
  45         },
  46     }, {
  47         'url': 'http://www.la7.it/omnibus/rivedila7/omnibus-news-02-07-2016-189077',
  48         'only_matching': True,
  49     }]
  50     _HOST = 'https://awsvodpkg.iltrovatore.it'
  51
  52     def _generate_mp4_url(self, quality, m3u8_formats):
  53         for f in m3u8_formats:
  54             if f['vcodec'] != 'none' and quality in f['url']:
  55                 http_url = f'{self._HOST}{quality}.mp4'
  56
  57                 urlh = self._request_webpage(
  58                     HEADRequest(http_url), quality,
  59                     note='Check filesize', fatal=False)
  60                 if urlh:
  61                     http_f = f.copy()
  62                     del http_f['manifest_url']
  63                     http_f.update({
  64                         'format_id': http_f['format_id'].replace('hls-', 'https-'),
  65                         'url': http_url,
  66                         'protocol': 'https',
  67                         'filesize_approx': int_or_none(urlh.headers.get('Content-Length', None)),
  68                     })
  69                     return http_f
  70                 return None
  71
  72     def _real_extract(self, url):
  73         video_id = self._match_id(url)
  74         webpage = self._download_webpage(url, video_id)
  75
  76         if re.search(r'(?i)(drmsupport\s*:\s*true)\s*', webpage):
  77             self.report_drm(video_id)
  78
  79         video_path = self._search_regex(
  80             r'(/content/[\w/,]+?)\.mp4(?:\.csmil)?/master\.m3u8', webpage, 'video_path')
  81
  82         formats = self._extract_mpd_formats(
  83             f'{self._HOST}/local/dash/,{video_path}.mp4.urlset/manifest.mpd',
  84             video_id, mpd_id='dash', fatal=False)
  85         m3u8_formats = self._extract_m3u8_formats(
  86             f'{self._HOST}/local/hls/,{video_path}.mp4.urlset/master.m3u8',
  87             video_id, 'mp4', m3u8_id='hls', fatal=False)
  88         formats.extend(m3u8_formats)
  89
  90         for q in filter(None, video_path.split(',')):
  91             http_f = self._generate_mp4_url(q, m3u8_formats)
  92             if http_f:
  93                 formats.append(http_f)
  94
  95         return {
  96             'id': video_id,
  97             'title': self._og_search_title(webpage, default=None),
  98             'description': self._og_search_description(webpage, default=None),
  99             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 100             'formats': formats,
 101             'upload_date': unified_strdate(self._search_regex(r'datetime="(.+?)"', webpage, 'upload_date', fatal=False))
 102         }
 103
 104
 105 class LA7PodcastEpisodeIE(InfoExtractor):
 106     IE_NAME = 'la7.it:pod:episode'
 107     _VALID_URL = r'https?://(?:www\.)?la7\.it/[^/]+/podcast/([^/]+-)?(?P<id>\d+)'
 108
 109     _TESTS = [{
 110         'url': 'https://www.la7.it/voicetown/podcast/la-carezza-delle-memoria-di-carlo-verdone-23-03-2021-371497',
 111         'md5': '7737d4d79b3c1a34b3de3e16297119ed',
 112         'info_dict': {
 113             'id': '371497',
 114             'ext': 'mp3',
 115             'title': '"La carezza delle memoria" di Carlo Verdone',
 116             'description': 'md5:5abf07c3c551a687db80af3f9ceb7d52',
 117             'thumbnail': 'https://www.la7.it/sites/default/files/podcast/371497.jpg',
 118             'upload_date': '20210323',
 119         },
 120     }, {
 121         # embed url
 122         'url': 'https://www.la7.it/embed/podcast/371497',
 123         'only_matching': True,
 124     }, {
 125         # date already in the title
 126         'url': 'https://www.la7.it/propagandalive/podcast/lintervista-di-diego-bianchi-ad-annalisa-cuzzocrea-puntata-del-1932021-20-03-2021-371130',
 127         'only_matching': True,
 128     }, {
 129         # title same as show_title
 130         'url': 'https://www.la7.it/otto-e-mezzo/podcast/otto-e-mezzo-26-03-2021-372340',
 131         'only_matching': True,
 132     }]
 133
 134     def _extract_info(self, webpage, video_id=None, ppn=None):
 135         if not video_id:
 136             video_id = self._search_regex(
 137                 r'data-nid=([\'"])(?P<vid>\d+)\1',
 138                 webpage, 'video_id', group='vid')
 139
 140         media_url = self._search_regex(
 141             (r'src\s*:\s*([\'"])(?P<url>\S+?mp3.+?)\1',
 142              r'data-podcast\s*=\s*([\'"])(?P<url>\S+?mp3.+?)\1'),
 143             webpage, 'media_url', group='url')
 144         formats = [{
 145             'url': media_url,
 146             'format_id': 'http-mp3',
 147             'ext': 'mp3',
 148             'acodec': 'mp3',
 149             'vcodec': 'none',
 150         }]
 151
 152         title = self._html_search_regex(
 153             (r'<div class="title">(?P<title>.+?)</',
 154              r'<title>(?P<title>[^<]+)</title>',
 155              r'title:\s*([\'"])(?P<title>.+?)\1'),
 156             webpage, 'title', group='title')
 157
 158         description = (
 159             self._html_search_regex(
 160                 (r'<div class="description">(.+?)</div>',
 161                  r'<div class="description-mobile">(.+?)</div>',
 162                  r'<div class="box-txt">([^<]+?)</div>',
 163                  r'<div class="field-content"><p>(.+?)</p></div>'),
 164                 webpage, 'description', default=None)
 165             or self._html_search_meta('description', webpage))
 166
 167         thumb = self._html_search_regex(
 168             (r'<div class="podcast-image"><img src="(.+?)"></div>',
 169              r'<div class="container-embed"[^<]+url\((.+?)\);">',
 170              r'<div class="field-content"><img src="(.+?)"'),
 171             webpage, 'thumbnail', fatal=False, default=None)
 172
 173         duration = parse_duration(self._html_search_regex(
 174             r'<span class="(?:durata|duration)">([\d:]+)</span>',
 175             webpage, 'duration', fatal=False, default=None))
 176
 177         date = self._html_search_regex(
 178             r'class="data">\s*(?:<span>)?([\d\.]+)\s*</',
 179             webpage, 'date', default=None)
 180
 181         date_alt = self._search_regex(
 182             r'(\d+[\./]\d+[\./]\d+)', title, 'date_alt', default=None)
 183         ppn = ppn or self._search_regex(
 184             r'ppN:\s*([\'"])(?P<ppn>.+?)\1',
 185             webpage, 'ppn', group='ppn', default=None)
 186         # if the date is not in the title
 187         # and title is the same as the show_title
 188         # add the date to the title
 189         if date and not date_alt and ppn and ppn.lower() == title.lower():
 190             title = f'{title} del {date}'
 191         return {
 192             'id': video_id,
 193             'title': title,
 194             'description': description,
 195             'duration': float_or_none(duration),
 196             'formats': formats,
 197             'thumbnail': thumb,
 198             'upload_date': unified_strdate(date),
 199         }
 200
 201     def _real_extract(self, url):
 202         video_id = self._match_id(url)
 203         webpage = self._download_webpage(url, video_id)
 204
 205         return self._extract_info(webpage, video_id)
 206
 207
 208 class LA7PodcastIE(LA7PodcastEpisodeIE):  # XXX: Do not subclass from concrete IE
 209     IE_NAME = 'la7.it:podcast'
 210     _VALID_URL = r'https?://(?:www\.)?la7\.it/(?P<id>[^/]+)/podcast/?(?:$|[#?])'
 211
 212     _TESTS = [{
 213         'url': 'https://www.la7.it/propagandalive/podcast',
 214         'info_dict': {
 215             'id': 'propagandalive',
 216             'title': "Propaganda Live",
 217         },
 218         'playlist_count_min': 10,
 219     }]
 220
 221     def _real_extract(self, url):
 222         playlist_id = self._match_id(url)
 223         webpage = self._download_webpage(url, playlist_id)
 224
 225         title = (
 226             self._html_search_regex(
 227                 r'<h1.*?>(.+?)</h1>', webpage, 'title', fatal=False, default=None)
 228             or self._og_search_title(webpage))
 229         ppn = self._search_regex(
 230             r'window\.ppN\s*=\s*([\'"])(?P<ppn>.+?)\1',
 231             webpage, 'ppn', group='ppn', default=None)
 232
 233         entries = []
 234         for episode in re.finditer(
 235                 r'<div class="container-podcast-property">([\s\S]+?)(?:</div>\s*){3}',
 236                 webpage):
 237             entries.append(self._extract_info(episode.group(1), ppn=ppn))
 238
 239         return self.playlist_result(entries, playlist_id, title)