youtube_dl/extractor/arcpublishing.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     extract_attributes,
   9     int_or_none,
  10     parse_iso8601,
  11     try_get,
  12 )
  13
  14
  15 class ArcPublishingIE(InfoExtractor):
  16     _UUID_REGEX = r'[\da-f]{8}-(?:[\da-f]{4}-){3}[\da-f]{12}'
  17     _VALID_URL = r'arcpublishing:(?P<org>[a-z]+):(?P<id>%s)' % _UUID_REGEX
  18     _TESTS = [{
  19         # https://www.adn.com/politics/2020/11/02/video-senate-candidates-campaign-in-anchorage-on-eve-of-election-day/
  20         'url': 'arcpublishing:adn:8c99cb6e-b29c-4bc9-9173-7bf9979225ab',
  21         'only_matching': True,
  22     }, {
  23         # https://www.bostonglobe.com/video/2020/12/30/metro/footage-released-showing-officer-talking-about-striking-protesters-with-car/
  24         'url': 'arcpublishing:bostonglobe:232b7ae6-7d73-432d-bc0a-85dbf0119ab1',
  25         'only_matching': True,
  26     }, {
  27         # https://www.actionnewsjax.com/video/live-stream/
  28         'url': 'arcpublishing:cmg:cfb1cf1b-3ab5-4d1b-86c5-a5515d311f2a',
  29         'only_matching': True,
  30     }, {
  31         # https://elcomercio.pe/videos/deportes/deporte-total-futbol-peruano-seleccion-peruana-la-valorizacion-de-los-peruanos-en-el-exterior-tras-un-2020-atipico-nnav-vr-video-noticia/
  32         'url': 'arcpublishing:elcomercio:27a7e1f8-2ec7-4177-874f-a4feed2885b3',
  33         'only_matching': True,
  34     }, {
  35         # https://www.clickondetroit.com/video/community/2020/05/15/events-surrounding-woodward-dream-cruise-being-canceled/
  36         'url': 'arcpublishing:gmg:c8793fb2-8d44-4242-881e-2db31da2d9fe',
  37         'only_matching': True,
  38     }, {
  39         # https://www.wabi.tv/video/2020/12/30/trenton-company-making-equipment-pfizer-covid-vaccine/
  40         'url': 'arcpublishing:gray:0b0ba30e-032a-4598-8810-901d70e6033e',
  41         'only_matching': True,
  42     }, {
  43         # https://www.lateja.cr/el-mundo/video-china-aprueba-con-condiciones-su-primera/dfcbfa57-527f-45ff-a69b-35fe71054143/video/
  44         'url': 'arcpublishing:gruponacion:dfcbfa57-527f-45ff-a69b-35fe71054143',
  45         'only_matching': True,
  46     }, {
  47         # https://www.fifthdomain.com/video/2018/03/09/is-america-vulnerable-to-a-cyber-attack/
  48         'url': 'arcpublishing:mco:aa0ca6fe-1127-46d4-b32c-be0d6fdb8055',
  49         'only_matching': True,
  50     }, {
  51         # https://www.vl.no/kultur/2020/12/09/en-melding-fra-en-lytter-endret-julelista-til-lewi-bergrud/
  52         'url': 'arcpublishing:mentormedier:47a12084-650b-4011-bfd0-3699b6947b2d',
  53         'only_matching': True,
  54     }, {
  55         # https://www.14news.com/2020/12/30/whiskey-theft-caught-camera-henderson-liquor-store/
  56         'url': 'arcpublishing:raycom:b89f61f8-79fa-4c09-8255-e64237119bf7',
  57         'only_matching': True,
  58     }, {
  59         # https://www.theglobeandmail.com/world/video-ethiopian-woman-who-became-symbol-of-integration-in-italy-killed-on/
  60         'url': 'arcpublishing:tgam:411b34c1-8701-4036-9831-26964711664b',
  61         'only_matching': True,
  62     }, {
  63         # https://www.pilotonline.com/460f2931-8130-4719-8ea1-ffcb2d7cb685-132.html
  64         'url': 'arcpublishing:tronc:460f2931-8130-4719-8ea1-ffcb2d7cb685',
  65         'only_matching': True,
  66     }]
  67     _POWA_DEFAULTS = [
  68         (['cmg', 'prisa'], '%s-config-prod.api.cdn.arcpublishing.com/video'),
  69         ([
  70             'adn', 'advancelocal', 'answers', 'bonnier', 'bostonglobe', 'demo',
  71             'gmg', 'gruponacion', 'infobae', 'mco', 'nzme', 'pmn', 'raycom',
  72             'spectator', 'tbt', 'tgam', 'tronc', 'wapo', 'wweek',
  73         ], 'video-api-cdn.%s.arcpublishing.com/api'),
  74     ]
  75
  76     @staticmethod
  77     def _extract_urls(webpage):
  78         entries = []
  79         # https://arcpublishing.atlassian.net/wiki/spaces/POWA/overview
  80         for powa_el in re.findall(r'(<div[^>]+class="[^"]*\bpowa\b[^"]*"[^>]+data-uuid="%s"[^>]*>)' % ArcPublishingIE._UUID_REGEX, webpage):
  81             powa = extract_attributes(powa_el) or {}
  82             org = powa.get('data-org')
  83             uuid = powa.get('data-uuid')
  84             if org and uuid:
  85                 entries.append('arcpublishing:%s:%s' % (org, uuid))
  86         return entries
  87
  88     def _real_extract(self, url):
  89         org, uuid = re.match(self._VALID_URL, url).groups()
  90         for orgs, tmpl in self._POWA_DEFAULTS:
  91             if org in orgs:
  92                 base_api_tmpl = tmpl
  93                 break
  94         else:
  95             base_api_tmpl = '%s-prod-cdn.video-api.arcpublishing.com/api'
  96         if org == 'wapo':
  97             org = 'washpost'
  98         video = self._download_json(
  99             'https://%s/v1/ansvideos/findByUuid' % (base_api_tmpl % org),
 100             uuid, query={'uuid': uuid})[0]
 101         title = video['headlines']['basic']
 102         is_live = video.get('status') == 'live'
 103
 104         urls = []
 105         formats = []
 106         for s in video.get('streams', []):
 107             s_url = s.get('url')
 108             if not s_url or s_url in urls:
 109                 continue
 110             urls.append(s_url)
 111             stream_type = s.get('stream_type')
 112             if stream_type == 'smil':
 113                 smil_formats = self._extract_smil_formats(
 114                     s_url, uuid, fatal=False)
 115                 for f in smil_formats:
 116                     if f['url'].endswith('/cfx/st'):
 117                         f['app'] = 'cfx/st'
 118                         if not f['play_path'].startswith('mp4:'):
 119                             f['play_path'] = 'mp4:' + f['play_path']
 120                         if isinstance(f['tbr'], float):
 121                             f['vbr'] = f['tbr'] * 1000
 122                             del f['tbr']
 123                             f['format_id'] = 'rtmp-%d' % f['vbr']
 124                 formats.extend(smil_formats)
 125             elif stream_type in ('ts', 'hls'):
 126                 m3u8_formats = self._extract_m3u8_formats(
 127                     s_url, uuid, 'mp4', 'm3u8' if is_live else 'm3u8_native',
 128                     m3u8_id='hls', fatal=False)
 129                 if all([f.get('acodec') == 'none' for f in m3u8_formats]):
 130                     continue
 131                 for f in m3u8_formats:
 132                     if f.get('acodec') == 'none':
 133                         f['preference'] = -40
 134                     elif f.get('vcodec') == 'none':
 135                         f['preference'] = -50
 136                     height = f.get('height')
 137                     if not height:
 138                         continue
 139                     vbr = self._search_regex(
 140                         r'[_x]%d[_-](\d+)' % height, f['url'], 'vbr', default=None)
 141                     if vbr:
 142                         f['vbr'] = int(vbr)
 143                 formats.extend(m3u8_formats)
 144             else:
 145                 vbr = int_or_none(s.get('bitrate'))
 146                 formats.append({
 147                     'format_id': '%s-%d' % (stream_type, vbr) if vbr else stream_type,
 148                     'vbr': vbr,
 149                     'width': int_or_none(s.get('width')),
 150                     'height': int_or_none(s.get('height')),
 151                     'filesize': int_or_none(s.get('filesize')),
 152                     'url': s_url,
 153                     'preference': -1,
 154                 })
 155         self._sort_formats(
 156             formats, ('preference', 'width', 'height', 'vbr', 'filesize', 'tbr', 'ext', 'format_id'))
 157
 158         subtitles = {}
 159         for subtitle in (try_get(video, lambda x: x['subtitles']['urls'], list) or []):
 160             subtitle_url = subtitle.get('url')
 161             if subtitle_url:
 162                 subtitles.setdefault('en', []).append({'url': subtitle_url})
 163
 164         return {
 165             'id': uuid,
 166             'title': self._live_title(title) if is_live else title,
 167             'thumbnail': try_get(video, lambda x: x['promo_image']['url']),
 168             'description': try_get(video, lambda x: x['subheadlines']['basic']),
 169             'formats': formats,
 170             'duration': int_or_none(video.get('duration'), 100),
 171             'timestamp': parse_iso8601(video.get('created_date')),
 172             'subtitles': subtitles,
 173             'is_live': is_live,
 174         }