youtube_dl/extractor/naver.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     int_or_none,
   8     update_url_query,
   9 )
  10
  11
  12 class NaverIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:m\.)?tv(?:cast)?\.naver\.com/v/(?P<id>\d+)'
  14
  15     _TESTS = [{
  16         'url': 'http://tv.naver.com/v/81652',
  17         'info_dict': {
  18             'id': '81652',
  19             'ext': 'mp4',
  20             'title': '[9월 모의고사 해설강의][수학_김상희] 수학 A형 16~20번',
  21             'description': '합격불변의 법칙 메가스터디 | 메가스터디 수학 김상희 선생님이 9월 모의고사 수학A형 16번에서 20번까지 해설강의를 공개합니다.',
  22             'upload_date': '20130903',
  23         },
  24     }, {
  25         'url': 'http://tv.naver.com/v/395837',
  26         'md5': '638ed4c12012c458fefcddfd01f173cd',
  27         'info_dict': {
  28             'id': '395837',
  29             'ext': 'mp4',
  30             'title': '9년이 지나도 아픈 기억, 전효성의 아버지',
  31             'description': 'md5:5bf200dcbf4b66eb1b350d1eb9c753f7',
  32             'upload_date': '20150519',
  33         },
  34         'skip': 'Georestricted',
  35     }, {
  36         'url': 'http://tvcast.naver.com/v/81652',
  37         'only_matching': True,
  38     }]
  39
  40     def _real_extract(self, url):
  41         video_id = self._match_id(url)
  42         webpage = self._download_webpage(url, video_id)
  43
  44         vid = self._search_regex(
  45             r'videoId["\']\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
  46             'video id', fatal=None, group='value')
  47         in_key = self._search_regex(
  48             r'inKey["\']\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
  49             'key', default=None, group='value')
  50
  51         if not vid or not in_key:
  52             error = self._html_search_regex(
  53                 r'(?s)<div class="(?:nation_error|nation_box|error_box)">\s*(?:<!--.*?-->)?\s*<p class="[^"]+">(?P<msg>.+?)</p>\s*</div>',
  54                 webpage, 'error', default=None)
  55             if error:
  56                 raise ExtractorError(error, expected=True)
  57             raise ExtractorError('couldn\'t extract vid and key')
  58         video_data = self._download_json(
  59             'http://play.rmcnmv.naver.com/vod/play/v2.0/' + vid,
  60             video_id, query={
  61                 'key': in_key,
  62             })
  63         meta = video_data['meta']
  64         title = meta['subject']
  65         formats = []
  66
  67         def extract_formats(streams, stream_type, query={}):
  68             for stream in streams:
  69                 stream_url = stream.get('source')
  70                 if not stream_url:
  71                     continue
  72                 stream_url = update_url_query(stream_url, query)
  73                 encoding_option = stream.get('encodingOption', {})
  74                 bitrate = stream.get('bitrate', {})
  75                 formats.append({
  76                     'format_id': '%s_%s' % (stream.get('type') or stream_type, encoding_option.get('id') or encoding_option.get('name')),
  77                     'url': stream_url,
  78                     'width': int_or_none(encoding_option.get('width')),
  79                     'height': int_or_none(encoding_option.get('height')),
  80                     'vbr': int_or_none(bitrate.get('video')),
  81                     'abr': int_or_none(bitrate.get('audio')),
  82                     'filesize': int_or_none(stream.get('size')),
  83                     'protocol': 'm3u8_native' if stream_type == 'HLS' else None,
  84                 })
  85
  86         extract_formats(video_data.get('videos', {}).get('list', []), 'H264')
  87         for stream_set in video_data.get('streams', []):
  88             query = {}
  89             for param in stream_set.get('keys', []):
  90                 query[param['name']] = param['value']
  91             stream_type = stream_set.get('type')
  92             videos = stream_set.get('videos')
  93             if videos:
  94                 extract_formats(videos, stream_type, query)
  95             elif stream_type == 'HLS':
  96                 stream_url = stream_set.get('source')
  97                 if not stream_url:
  98                     continue
  99                 formats.extend(self._extract_m3u8_formats(
 100                     update_url_query(stream_url, query), video_id,
 101                     'mp4', 'm3u8_native', m3u8_id=stream_type, fatal=False))
 102         self._sort_formats(formats)
 103
 104         subtitles = {}
 105         for caption in video_data.get('captions', {}).get('list', []):
 106             caption_url = caption.get('source')
 107             if not caption_url:
 108                 continue
 109             subtitles.setdefault(caption.get('language') or caption.get('locale'), []).append({
 110                 'url': caption_url,
 111             })
 112
 113         upload_date = self._search_regex(
 114             r'<span[^>]+class="date".*?(\d{4}\.\d{2}\.\d{2})',
 115             webpage, 'upload date', fatal=False)
 116         if upload_date:
 117             upload_date = upload_date.replace('.', '')
 118
 119         return {
 120             'id': video_id,
 121             'title': title,
 122             'formats': formats,
 123             'subtitles': subtitles,
 124             'description': self._og_search_description(webpage),
 125             'thumbnail': meta.get('cover', {}).get('source') or self._og_search_thumbnail(webpage),
 126             'view_count': int_or_none(meta.get('count')),
 127             'upload_date': upload_date,
 128         }