youtube-dl

Another place where youtube-dl lives on
git clone git://git.oshgnacknak.de/youtube-dl.git
Log | Files | Refs | README | LICENSE

playtvak.py (7271B)


      1 # coding: utf-8
      2 from __future__ import unicode_literals
      3 
      4 from .common import InfoExtractor
      5 from ..compat import (
      6     compat_urlparse,
      7     compat_urllib_parse_urlencode,
      8 )
      9 from ..utils import (
     10     ExtractorError,
     11     int_or_none,
     12     parse_iso8601,
     13     qualities,
     14 )
     15 
     16 
     17 class PlaytvakIE(InfoExtractor):
     18     IE_DESC = 'Playtvak.cz, iDNES.cz and Lidovky.cz'
     19     _VALID_URL = r'https?://(?:.+?\.)?(?:playtvak|idnes|lidovky|metro)\.cz/.*\?(?:c|idvideo)=(?P<id>[^&]+)'
     20     _TESTS = [{
     21         'url': 'http://www.playtvak.cz/vyzente-vosy-a-srsne-ze-zahrady-dn5-/hodinovy-manzel.aspx?c=A150730_150323_hodinovy-manzel_kuko',
     22         'md5': '4525ae312c324b4be2f4603cc78ceb4a',
     23         'info_dict': {
     24             'id': 'A150730_150323_hodinovy-manzel_kuko',
     25             'ext': 'mp4',
     26             'title': 'Vyžeňte vosy a sršně ze zahrady',
     27             'description': 'md5:4436e61b7df227a093778efb7e373571',
     28             'thumbnail': r're:(?i)^https?://.*\.(?:jpg|png)$',
     29             'duration': 279,
     30             'timestamp': 1438732860,
     31             'upload_date': '20150805',
     32             'is_live': False,
     33         }
     34     }, {  # live video test
     35         'url': 'http://slowtv.playtvak.cz/planespotting-0pr-/planespotting.aspx?c=A150624_164934_planespotting_cat',
     36         'info_dict': {
     37             'id': 'A150624_164934_planespotting_cat',
     38             'ext': 'flv',
     39             'title': 're:^Planespotting [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
     40             'description': 'Sledujte provoz na ranveji Letiště Václava Havla v Praze',
     41             'is_live': True,
     42         },
     43         'params': {
     44             'skip_download': True,  # requires rtmpdump
     45         },
     46     }, {  # another live stream, this one without Misc.videoFLV
     47         'url': 'https://slowtv.playtvak.cz/zive-sledujte-vlaky-v-primem-prenosu-dwi-/hlavni-nadrazi.aspx?c=A151218_145728_hlavni-nadrazi_plap',
     48         'info_dict': {
     49             'id': 'A151218_145728_hlavni-nadrazi_plap',
     50             'ext': 'flv',
     51             'title': 're:^Hlavní nádraží [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
     52             'is_live': True,
     53         },
     54         'params': {
     55             'skip_download': True,  # requires rtmpdump
     56         },
     57     }, {  # idnes.cz
     58         'url': 'http://zpravy.idnes.cz/pes-zavreny-v-aute-rozbijeni-okynek-v-aute-fj5-/domaci.aspx?c=A150809_104116_domaci_pku',
     59         'md5': '819832ba33cd7016e58a6658577fe289',
     60         'info_dict': {
     61             'id': 'A150809_104116_domaci_pku',
     62             'ext': 'mp4',
     63             'title': 'Zavřeli jsme mraženou pizzu do auta. Upekla se',
     64             'description': 'md5:01e73f02329e2e5760bd5eed4d42e3c2',
     65             'thumbnail': r're:(?i)^https?://.*\.(?:jpg|png)$',
     66             'duration': 39,
     67             'timestamp': 1438969140,
     68             'upload_date': '20150807',
     69             'is_live': False,
     70         }
     71     }, {  # lidovky.cz
     72         'url': 'http://www.lidovky.cz/dalsi-demonstrace-v-praze-o-migraci-duq-/video.aspx?c=A150808_214044_ln-video_ELE',
     73         'md5': 'c7209ac4ba9d234d4ad5bab7485bcee8',
     74         'info_dict': {
     75             'id': 'A150808_214044_ln-video_ELE',
     76             'ext': 'mp4',
     77             'title': 'Táhni! Demonstrace proti imigrantům budila emoce',
     78             'description': 'md5:97c81d589a9491fbfa323c9fa3cca72c',
     79             'thumbnail': r're:(?i)^https?://.*\.(?:jpg|png)$',
     80             'timestamp': 1439052180,
     81             'upload_date': '20150808',
     82             'is_live': False,
     83         }
     84     }, {  # metro.cz
     85         'url': 'http://www.metro.cz/video-pod-billboardem-se-na-vltavske-roztocil-kolotoc-deti-vozil-jen-par-hodin-1hx-/metro-extra.aspx?c=A141111_173251_metro-extra_row',
     86         'md5': '84fc1deedcac37b7d4a6ccae7c716668',
     87         'info_dict': {
     88             'id': 'A141111_173251_metro-extra_row',
     89             'ext': 'mp4',
     90             'title': 'Recesisté udělali z billboardu kolotoč',
     91             'description': 'md5:7369926049588c3989a66c9c1a043c4c',
     92             'thumbnail': r're:(?i)^https?://.*\.(?:jpg|png)$',
     93             'timestamp': 1415725500,
     94             'upload_date': '20141111',
     95             'is_live': False,
     96         }
     97     }, {
     98         'url': 'http://www.playtvak.cz/embed.aspx?idvideo=V150729_141549_play-porad_kuko',
     99         'only_matching': True,
    100     }]
    101 
    102     def _real_extract(self, url):
    103         video_id = self._match_id(url)
    104 
    105         webpage = self._download_webpage(url, video_id)
    106 
    107         info_url = self._html_search_regex(
    108             r'Misc\.video(?:FLV)?\(\s*{\s*data\s*:\s*"([^"]+)"', webpage, 'info url')
    109 
    110         parsed_url = compat_urlparse.urlparse(info_url)
    111 
    112         qs = compat_urlparse.parse_qs(parsed_url.query)
    113         qs.update({
    114             'reklama': ['0'],
    115             'type': ['js'],
    116         })
    117 
    118         info_url = compat_urlparse.urlunparse(
    119             parsed_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
    120 
    121         json_info = self._download_json(
    122             info_url, video_id,
    123             transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1])
    124 
    125         item = None
    126         for i in json_info['items']:
    127             if i.get('type') == 'video' or i.get('type') == 'stream':
    128                 item = i
    129                 break
    130         if not item:
    131             raise ExtractorError('No suitable stream found')
    132 
    133         quality = qualities(('low', 'middle', 'high'))
    134 
    135         formats = []
    136         for fmt in item['video']:
    137             video_url = fmt.get('file')
    138             if not video_url:
    139                 continue
    140 
    141             format_ = fmt['format']
    142             format_id = '%s_%s' % (format_, fmt['quality'])
    143             preference = None
    144 
    145             if format_ in ('mp4', 'webm'):
    146                 ext = format_
    147             elif format_ == 'rtmp':
    148                 ext = 'flv'
    149             elif format_ == 'apple':
    150                 ext = 'mp4'
    151                 # Some streams have mp3 audio which does not play
    152                 # well with ffmpeg filter aac_adtstoasc
    153                 preference = -1
    154             elif format_ == 'adobe':  # f4m manifest fails with 404 in 80% of requests
    155                 continue
    156             else:  # Other formats not supported yet
    157                 continue
    158 
    159             formats.append({
    160                 'url': video_url,
    161                 'ext': ext,
    162                 'format_id': format_id,
    163                 'quality': quality(fmt.get('quality')),
    164                 'preference': preference,
    165             })
    166         self._sort_formats(formats)
    167 
    168         title = item['title']
    169         is_live = item['type'] == 'stream'
    170         if is_live:
    171             title = self._live_title(title)
    172         description = self._og_search_description(webpage, default=None) or self._html_search_meta(
    173             'description', webpage, 'description', default=None)
    174         timestamp = None
    175         duration = None
    176         if not is_live:
    177             duration = int_or_none(item.get('length'))
    178             timestamp = item.get('published')
    179             if timestamp:
    180                 timestamp = parse_iso8601(timestamp[:-5])
    181 
    182         return {
    183             'id': video_id,
    184             'title': title,
    185             'description': description,
    186             'thumbnail': item.get('image'),
    187             'duration': duration,
    188             'timestamp': timestamp,
    189             'is_live': is_live,
    190             'formats': formats,
    191         }