yt_dlp/extractor/gedidigital.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     base_url,
   6     determine_ext,
   7     int_or_none,
   8     url_basename,
   9     urljoin,
  10 )
  11
  12
  13 class GediDigitalIE(InfoExtractor):
  14     _VALID_URL = r'''(?x:(?P<url>(?:https?:)//video\.
  15         (?:
  16             (?:
  17                 (?:espresso\.)?repubblica
  18                 |lastampa
  19                 |ilsecoloxix
  20                 |huffingtonpost
  21             )|
  22             (?:
  23                 iltirreno
  24                 |messaggeroveneto
  25                 |ilpiccolo
  26                 |gazzettadimantova
  27                 |mattinopadova
  28                 |laprovinciapavese
  29                 |tribunatreviso
  30                 |nuovavenezia
  31                 |gazzettadimodena
  32                 |lanuovaferrara
  33                 |corrierealpi
  34                 |lasentinella
  35             )\.gelocal
  36         )\.it(?:/[^/]+){2,4}/(?P<id>\d+))(?:$|[?&].*))'''
  37     _TESTS = [{
  38         'url': 'https://video.lastampa.it/politica/il-paradosso-delle-regionali-la-lega-vince-ma-sembra-aver-perso/121559/121683',
  39         'md5': '84658d7fb9e55a6e57ecc77b73137494',
  40         'info_dict': {
  41             'id': '121683',
  42             'ext': 'mp4',
  43             'title': 'Il paradosso delle Regionali: ecco perché la Lega vince ma sembra aver perso',
  44             'description': 'md5:de7f4d6eaaaf36c153b599b10f8ce7ca',
  45             'thumbnail': r're:^https://www\.repstatic\.it/video/photo/.+?-thumb-full-.+?\.jpg$',
  46             'duration': 125,
  47         },
  48     }, {
  49         'url': 'https://video.huffingtonpost.it/embed/politica/cotticelli-non-so-cosa-mi-sia-successo-sto-cercando-di-capire-se-ho-avuto-un-malore/29312/29276?responsive=true&el=video971040871621586700',
  50         'only_matching': True,
  51     }, {
  52         'url': 'https://video.espresso.repubblica.it/embed/tutti-i-video/01-ted-villa/14772/14870&width=640&height=360',
  53         'only_matching': True,
  54     }, {
  55         'url': 'https://video.repubblica.it/motori/record-della-pista-a-spa-francorchamps-la-pagani-huayra-roadster-bc-stupisce/367415/367963',
  56         'only_matching': True,
  57     }, {
  58         'url': 'https://video.ilsecoloxix.it/sport/cassani-e-i-brividi-azzurri-ai-mondiali-di-imola-qui-mi-sono-innamorato-del-ciclismo-da-ragazzino-incredibile-tornarci-da-ct/66184/66267',
  59         'only_matching': True,
  60     }, {
  61         'url': 'https://video.iltirreno.gelocal.it/sport/dentro-la-notizia-ferrari-cosa-succede-a-maranello/141059/142723',
  62         'only_matching': True,
  63     }, {
  64         'url': 'https://video.messaggeroveneto.gelocal.it/locale/maria-giovanna-elmi-covid-vaccino/138155/139268',
  65         'only_matching': True,
  66     }, {
  67         'url': 'https://video.ilpiccolo.gelocal.it/dossier/big-john/dinosauro-big-john-al-via-le-visite-guidate-a-trieste/135226/135751',
  68         'only_matching': True,
  69     }, {
  70         'url': 'https://video.gazzettadimantova.gelocal.it/locale/dal-ponte-visconteo-di-valeggio-l-and-8217sos-dei-ristoratori-aprire-anche-a-cena/137310/137818',
  71         'only_matching': True,
  72     }, {
  73         'url': 'https://video.mattinopadova.gelocal.it/dossier/coronavirus-in-veneto/covid-a-vo-un-anno-dopo-un-cuore-tricolore-per-non-dimenticare/138402/138964',
  74         'only_matching': True,
  75     }, {
  76         'url': 'https://video.laprovinciapavese.gelocal.it/locale/mede-zona-rossa-via-alle-vaccinazioni-per-gli-over-80/137545/138120',
  77         'only_matching': True,
  78     }, {
  79         'url': 'https://video.tribunatreviso.gelocal.it/dossier/coronavirus-in-veneto/ecco-le-prima-vaccinazioni-di-massa-nella-marca/134485/135024',
  80         'only_matching': True,
  81     }, {
  82         'url': 'https://video.nuovavenezia.gelocal.it/locale/camion-troppo-alto-per-il-ponte-ferroviario-perde-il-carico/135734/136266',
  83         'only_matching': True,
  84     }, {
  85         'url': 'https://video.gazzettadimodena.gelocal.it/locale/modena-scoperta-la-proteina-che-predice-il-livello-di-gravita-del-covid/139109/139796',
  86         'only_matching': True,
  87     }, {
  88         'url': 'https://video.lanuovaferrara.gelocal.it/locale/due-bombole-di-gpl-aperte-e-abbandonate-i-vigili-bruciano-il-gas/134391/134957',
  89         'only_matching': True,
  90     }, {
  91         'url': 'https://video.corrierealpi.gelocal.it/dossier/cortina-2021-i-mondiali-di-sci-alpino/mondiali-di-sci-il-timelapse-sulla-splendida-olympia/133760/134331',
  92         'only_matching': True,
  93     }, {
  94         'url': 'https://video.lasentinella.gelocal.it/locale/vestigne-centra-un-auto-e-si-ribalta/138931/139466',
  95         'only_matching': True,
  96     }, {
  97         'url': 'https://video.espresso.repubblica.it/tutti-i-video/01-ted-villa/14772',
  98         'only_matching': True,
  99     }]
 100
 101     @staticmethod
 102     def _sanitize_urls(urls):
 103         # add protocol if missing
 104         for i, e in enumerate(urls):
 105             if e.startswith('//'):
 106                 urls[i] = 'https:%s' % e
 107         # clean iframes urls
 108         for i, e in enumerate(urls):
 109             urls[i] = urljoin(base_url(e), url_basename(e))
 110         return urls
 111
 112     @staticmethod
 113     def _extract_urls(webpage):
 114         entries = [
 115             mobj.group('eurl')
 116             for mobj in re.finditer(r'''(?x)
 117             (?:
 118                 data-frame-src=|
 119                 <iframe[^\n]+src=
 120             )
 121             (["'])(?P<eurl>%s)\1''' % GediDigitalIE._VALID_URL, webpage)]
 122         return GediDigitalIE._sanitize_urls(entries)
 123
 124     @staticmethod
 125     def _extract_url(webpage):
 126         urls = GediDigitalIE._extract_urls(webpage)
 127         return urls[0] if urls else None
 128
 129     @staticmethod
 130     def _clean_formats(formats):
 131         format_urls = set()
 132         clean_formats = []
 133         for f in formats:
 134             if f['url'] not in format_urls:
 135                 if f.get('audio_ext') != 'none' and not f.get('acodec'):
 136                     continue
 137                 format_urls.add(f['url'])
 138                 clean_formats.append(f)
 139         formats[:] = clean_formats
 140
 141     def _real_extract(self, url):
 142         video_id = self._match_id(url)
 143         url = self._match_valid_url(url).group('url')
 144         webpage = self._download_webpage(url, video_id)
 145         title = self._html_search_meta(
 146             ['twitter:title', 'og:title'], webpage, fatal=True)
 147         player_data = re.findall(
 148             r"PlayerFactory\.setParam\('(?P<type>format|param)',\s*'(?P<name>[^']+)',\s*'(?P<val>[^']+)'\);",
 149             webpage)
 150
 151         formats = []
 152         duration = thumb = None
 153         for t, n, v in player_data:
 154             if t == 'format':
 155                 if n in ('video-hds-vod-ec', 'video-hls-vod-ec', 'video-viralize', 'video-youtube-pfp'):
 156                     continue
 157                 elif n.endswith('-vod-ak'):
 158                     formats.extend(self._extract_akamai_formats(
 159                         v, video_id, {'http': 'media.gedidigital.it'}))
 160                 else:
 161                     ext = determine_ext(v)
 162                     if ext == 'm3u8':
 163                         formats.extend(self._extract_m3u8_formats(
 164                             v, video_id, 'mp4', 'm3u8_native', m3u8_id=n, fatal=False))
 165                         continue
 166                     f = {
 167                         'format_id': n,
 168                         'url': v,
 169                     }
 170                     if ext == 'mp3':
 171                         abr = int_or_none(self._search_regex(
 172                             r'-mp3-audio-(\d+)', v, 'abr', default=None))
 173                         f.update({
 174                             'abr': abr,
 175                             'tbr': abr,
 176                             'acodec': ext,
 177                             'vcodec': 'none'
 178                         })
 179                     else:
 180                         mobj = re.match(r'^video-rrtv-(\d+)(?:-(\d+))?$', n)
 181                         if mobj:
 182                             f.update({
 183                                 'height': int(mobj.group(1)),
 184                                 'vbr': int_or_none(mobj.group(2)),
 185                             })
 186                         if not f.get('vbr'):
 187                             f['vbr'] = int_or_none(self._search_regex(
 188                                 r'-video-rrtv-(\d+)', v, 'abr', default=None))
 189                     formats.append(f)
 190             elif t == 'param':
 191                 if n in ['image_full', 'image']:
 192                     thumb = v
 193                 elif n == 'videoDuration':
 194                     duration = int_or_none(v)
 195
 196         self._clean_formats(formats)
 197         self._sort_formats(formats)
 198
 199         return {
 200             'id': video_id,
 201             'title': title,
 202             'description': self._html_search_meta(
 203                 ['twitter:description', 'og:description', 'description'], webpage),
 204             'thumbnail': thumb or self._og_search_thumbnail(webpage),
 205             'formats': formats,
 206             'duration': duration,
 207         }