Raphaël G. Git Repositories - youtubedl/blob - 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\.)?tv(?:cast)?\.naver\.com/v/(?P<id>\d+)'
  16
  17     _TESTS = [{
  18         'url': 'http://tv.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://tv.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         'url': 'http://tvcast.naver.com/v/81652',
  39         'only_matching': True,
  40     }]
  41
  42     def _real_extract(self, url):
  43         video_id = self._match_id(url)
  44         webpage = self._download_webpage(url, video_id)
  45
  46         m_id = re.search(r'var rmcPlayer = new nhn.rmcnmv.RMCVideoPlayer\("(.+?)", "(.+?)"',
  47                          webpage)
  48         if m_id is None:
  49             error = self._html_search_regex(
  50                 r'(?s)<div class="(?:nation_error|nation_box|error_box)">\s*(?:<!--.*?-->)?\s*<p class="[^"]+">(?P<msg>.+?)</p>\s*</div>',
  51                 webpage, 'error', default=None)
  52             if error:
  53                 raise ExtractorError(error, expected=True)
  54             raise ExtractorError('couldn\'t extract vid and key')
  55         video_data = self._download_json(
  56             'http://play.rmcnmv.naver.com/vod/play/v2.0/' + m_id.group(1),
  57             video_id, query={
  58                 'key': m_id.group(2),
  59             })
  60         meta = video_data['meta']
  61         title = meta['subject']
  62         formats = []
  63
  64         def extract_formats(streams, stream_type, query={}):
  65             for stream in streams:
  66                 stream_url = stream.get('source')
  67                 if not stream_url:
  68                     continue
  69                 stream_url = update_url_query(stream_url, query)
  70                 encoding_option = stream.get('encodingOption', {})
  71                 bitrate = stream.get('bitrate', {})
  72                 formats.append({
  73                     'format_id': '%s_%s' % (stream.get('type') or stream_type, encoding_option.get('id') or encoding_option.get('name')),
  74                     'url': stream_url,
  75                     'width': int_or_none(encoding_option.get('width')),
  76                     'height': int_or_none(encoding_option.get('height')),
  77                     'vbr': int_or_none(bitrate.get('video')),
  78                     'abr': int_or_none(bitrate.get('audio')),
  79                     'filesize': int_or_none(stream.get('size')),
  80                     'protocol': 'm3u8_native' if stream_type == 'HLS' else None,
  81                 })
  82
  83         extract_formats(video_data.get('videos', {}).get('list', []), 'H264')
  84         for stream_set in video_data.get('streams', []):
  85             query = {}
  86             for param in stream_set.get('keys', []):
  87                 query[param['name']] = param['value']
  88             stream_type = stream_set.get('type')
  89             videos = stream_set.get('videos')
  90             if videos:
  91                 extract_formats(videos, stream_type, query)
  92             elif stream_type == 'HLS':
  93                 stream_url = stream_set.get('source')
  94                 if not stream_url:
  95                     continue
  96                 formats.extend(self._extract_m3u8_formats(
  97                     update_url_query(stream_url, query), video_id,
  98                     'mp4', 'm3u8_native', m3u8_id=stream_type, fatal=False))
  99         self._sort_formats(formats)
 100
 101         subtitles = {}
 102         for caption in video_data.get('captions', {}).get('list', []):
 103             caption_url = caption.get('source')
 104             if not caption_url:
 105                 continue
 106             subtitles.setdefault(caption.get('language') or caption.get('locale'), []).append({
 107                 'url': caption_url,
 108             })
 109
 110         upload_date = self._search_regex(
 111             r'<span[^>]+class="date".*?(\d{4}\.\d{2}\.\d{2})',
 112             webpage, 'upload date', fatal=False)
 113         if upload_date:
 114             upload_date = upload_date.replace('.', '')
 115
 116         return {
 117             'id': video_id,
 118             'title': title,
 119             'formats': formats,
 120             'subtitles': subtitles,
 121             'description': self._og_search_description(webpage),
 122             'thumbnail': meta.get('cover', {}).get('source') or self._og_search_thumbnail(webpage),
 123             'view_count': int_or_none(meta.get('count')),
 124             'upload_date': upload_date,
 125         }