Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/dvtv.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     ExtractorError,
  10     int_or_none,
  11     js_to_json,
  12     mimetype2ext,
  13     unescapeHTML,
  14 )
  15
  16
  17 class DVTVIE(InfoExtractor):
  18     IE_NAME = 'dvtv'
  19     IE_DESC = 'http://video.aktualne.cz/'
  20
  21     _VALID_URL = r'https?://video\.aktualne\.cz/(?:[^/]+/)+r~(?P<id>[0-9a-f]{32})'
  22
  23     _TESTS = [{
  24         'url': 'http://video.aktualne.cz/dvtv/vondra-o-ceskem-stoleti-pri-pohledu-na-havla-mi-bylo-trapne/r~e5efe9ca855511e4833a0025900fea04/',
  25         'md5': '67cb83e4a955d36e1b5d31993134a0c2',
  26         'info_dict': {
  27             'id': 'dc0768de855511e49e4b0025900fea04',
  28             'ext': 'mp4',
  29             'title': 'Vondra o Českém století: Při pohledu na Havla mi bylo trapně',
  30             'duration': 1484,
  31         }
  32     }, {
  33         'url': 'http://video.aktualne.cz/dvtv/dvtv-16-12-2014-utok-talibanu-boj-o-kliniku-uprchlici/r~973eb3bc854e11e498be002590604f2e/',
  34         'info_dict': {
  35             'title': r're:^DVTV 16\. 12\. 2014: útok Talibanu, boj o kliniku, uprchlíci',
  36             'id': '973eb3bc854e11e498be002590604f2e',
  37         },
  38         'playlist': [{
  39             'md5': 'da7ca6be4935532241fa9520b3ad91e4',
  40             'info_dict': {
  41                 'id': 'b0b40906854d11e4bdad0025900fea04',
  42                 'ext': 'mp4',
  43                 'title': 'Drtinová Veselovský TV 16. 12. 2014: Témata dne',
  44                 'description': 'md5:0916925dea8e30fe84222582280b47a0',
  45                 'timestamp': 1418760010,
  46                 'upload_date': '20141216',
  47             }
  48         }, {
  49             'md5': '5f7652a08b05009c1292317b449ffea2',
  50             'info_dict': {
  51                 'id': '420ad9ec854a11e4bdad0025900fea04',
  52                 'ext': 'mp4',
  53                 'title': 'Školní masakr možná změní boj s Talibanem, říká novinářka',
  54                 'description': 'md5:ff2f9f6de73c73d7cef4f756c1c1af42',
  55                 'timestamp': 1418760010,
  56                 'upload_date': '20141216',
  57             }
  58         }, {
  59             'md5': '498eb9dfa97169f409126c617e2a3d64',
  60             'info_dict': {
  61                 'id': '95d35580846a11e4b6d20025900fea04',
  62                 'ext': 'mp4',
  63                 'title': 'Boj o kliniku: Veřejný zájem, nebo právo na majetek?',
  64                 'description': 'md5:889fe610a70fee5511dc3326a089188e',
  65                 'timestamp': 1418760010,
  66                 'upload_date': '20141216',
  67             }
  68         }, {
  69             'md5': 'b8dc6b744844032dab6ba3781a7274b9',
  70             'info_dict': {
  71                 'id': '6fe14d66853511e4833a0025900fea04',
  72                 'ext': 'mp4',
  73                 'title': 'Pánek: Odmítání syrských uprchlíků je ostudou české vlády',
  74                 'description': 'md5:544f86de6d20c4815bea11bf2ac3004f',
  75                 'timestamp': 1418760010,
  76                 'upload_date': '20141216',
  77             }
  78         }],
  79     }, {
  80         'url': 'https://video.aktualne.cz/dvtv/zeman-si-jen-leci-mindraky-sobotku-nenavidi-a-babis-se-mu-te/r~960cdb3a365a11e7a83b0025900fea04/',
  81         'md5': 'f8efe9656017da948369aa099788c8ea',
  82         'info_dict': {
  83             'id': '3c496fec365911e7a6500025900fea04',
  84             'ext': 'mp4',
  85             'title': 'Zeman si jen léčí mindráky, Sobotku nenávidí a Babiš se mu teď hodí, tvrdí Kmenta',
  86             'duration': 1103,
  87         },
  88         'params': {
  89             'skip_download': True,
  90         },
  91     }, {
  92         'url': 'http://video.aktualne.cz/v-cechach-poprve-zazni-zelenkova-zrestaurovana-mse/r~45b4b00483ec11e4883b002590604f2e/',
  93         'only_matching': True,
  94     }]
  95
  96     def _parse_video_metadata(self, js, video_id, live_js=None):
  97         data = self._parse_json(js, video_id, transform_source=js_to_json)
  98         if live_js:
  99             data.update(self._parse_json(
 100                 live_js, video_id, transform_source=js_to_json))
 101
 102         title = unescapeHTML(data['title'])
 103
 104         formats = []
 105         for video in data['sources']:
 106             video_url = video.get('file')
 107             if not video_url:
 108                 continue
 109             video_type = video.get('type')
 110             ext = determine_ext(video_url, mimetype2ext(video_type))
 111             if video_type == 'application/vnd.apple.mpegurl' or ext == 'm3u8':
 112                 formats.extend(self._extract_m3u8_formats(
 113                     video_url, video_id, 'mp4', entry_protocol='m3u8_native',
 114                     m3u8_id='hls', fatal=False))
 115             elif video_type == 'application/dash+xml' or ext == 'mpd':
 116                 formats.extend(self._extract_mpd_formats(
 117                     video_url, video_id, mpd_id='dash', fatal=False))
 118             else:
 119                 label = video.get('label')
 120                 height = self._search_regex(
 121                     r'^(\d+)[pP]', label or '', 'height', default=None)
 122                 format_id = ['http']
 123                 for f in (ext, label):
 124                     if f:
 125                         format_id.append(f)
 126                 formats.append({
 127                     'url': video_url,
 128                     'format_id': '-'.join(format_id),
 129                     'height': int_or_none(height),
 130                 })
 131         self._sort_formats(formats)
 132
 133         return {
 134             'id': data.get('mediaid') or video_id,
 135             'title': title,
 136             'description': data.get('description'),
 137             'thumbnail': data.get('image'),
 138             'duration': int_or_none(data.get('duration')),
 139             'timestamp': int_or_none(data.get('pubtime')),
 140             'formats': formats
 141         }
 142
 143     def _real_extract(self, url):
 144         video_id = self._match_id(url)
 145
 146         webpage = self._download_webpage(url, video_id)
 147
 148         # live content
 149         live_item = self._search_regex(
 150             r'(?s)embedData[0-9a-f]{32}\.asset\.liveStarter\s*=\s*(\{.+?\});',
 151             webpage, 'video', default=None)
 152
 153         # single video
 154         item = self._search_regex(
 155             r'(?s)embedData[0-9a-f]{32}\[["\']asset["\']\]\s*=\s*(\{.+?\});',
 156             webpage, 'video', default=None)
 157
 158         if item:
 159             return self._parse_video_metadata(item, video_id, live_item)
 160
 161         # playlist
 162         items = re.findall(
 163             r"(?s)BBX\.context\.assets\['[0-9a-f]{32}'\]\.push\(({.+?})\);",
 164             webpage)
 165         if not items:
 166             items = re.findall(r'(?s)var\s+asset\s*=\s*({.+?});\n', webpage)
 167
 168         if items:
 169             return {
 170                 '_type': 'playlist',
 171                 'id': video_id,
 172                 'title': self._og_search_title(webpage),
 173                 'entries': [self._parse_video_metadata(i, video_id) for i in items]
 174             }
 175
 176         raise ExtractorError('Could not find neither video nor playlist')