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