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