1
0
mirror of https://github.com/l1ving/youtube-dl synced 2025-03-20 05:38:46 +08:00

[naver] add support for other naver sites

This commit is contained in:
remitamine 2015-10-17 05:20:18 +01:00
parent 2ccb37beb9
commit aaf7499008

View File

@ -7,23 +7,28 @@ from .common import InfoExtractor
from ..compat import (
compat_urllib_parse,
compat_urlparse,
compat_urllib_request,
)
from ..utils import (
ExtractorError,
int_or_none,
float_or_none,
determine_ext,
)
class NaverIE(InfoExtractor):
_VALID_URL = r'https?://(?:m\.)?tvcast\.naver\.com/v/(?P<id>\d+)'
_VALID_URL = r'https?://(?:m\.)?(?:tvcast\.naver\.com/v/|news\.naver\.com/main/read\.nhn?.*aid=|sports\.news\.naver.com/(?:videoCenter|sports)/(?:index|video)\.nhn?.*id=|movie\.naver.com/movie/bi/mi/mediaView\.nhn?.*mid=|music\.naver\.com/artist/videoPlayer\.nhn?.*videoId=)(?P<id>\d+)'
_TESTS = [{
'url': 'http://tvcast.naver.com/v/81652',
'md5': '0fe25e226a0ec388cd75679981bd2a1a',
'info_dict': {
'id': '81652',
'id': '4B40C2B7F4BC7C7BBA5237C5E3CED1ADEAF5',
'ext': 'mp4',
'title': '[9월 모의고사 해설강의][수학_김상희] 수학 A형 16~20번',
'description': '합격불변의 법칙 메가스터디 | 메가스터디 수학 김상희 선생님이 9월 모의고사 수학A형 16번에서 20번까지 해설강의를 공개합니다.',
'upload_date': '20130903',
'uploader_id': 'megastudy',
'uploader': '합격불변의 법칙 메가스터디',
},
}, {
'url': 'http://tvcast.naver.com/v/395837',
@ -36,63 +41,120 @@ class NaverIE(InfoExtractor):
'upload_date': '20150519',
},
'skip': 'Georestricted',
}, {
'url': 'http://news.naver.com/main/read.nhn?mode=LSD&mid=tvh&oid=056&aid=0010235712&sid1=293',
'md5': 'ea02b7943173553618f49de08d6bd36e',
'info_dict': {
'id': '3C026B64F022C136ED9057AA820458275CD7',
'ext': 'mp4',
'title': '한미 정상 “북핵 문제 최우선”…공동성명 첫 채택',
'uploader_id': 'muploader_o',
'uploader': '',
},
}, {
'url': 'http://sports.news.naver.com/videoCenter/index.nhn?uCategory=esports&category=lol&id=158508',
'md5': '436254dbbb7dde42053c731039f6f14d',
'info_dict': {
'id': '3CD20A3B7B15044056189B09557FB349DE0B',
'ext': 'mp4',
'title': '\'전승 가도\' 절대강자 Faker의 인터뷰',
'uploader_id': 'muploader_n',
'uploader': '',
},
}, {
'url': 'http://movie.naver.com/movie/bi/mi/mediaView.nhn?code=115622&mid=27362',
'md5': 'ad9623ff5a23d6e9c0c7f451b063912f',
'info_dict': {
'id': '1132280440B9037B8E48DB0F569208440008',
'ext': 'mp4',
'title': '<인사이드 아웃> 메인 예고편',
'uploader_id': 'navermovie',
'uploader': '네이버 영화',
},
}]
def _extract_video_formats(self, formats_list):
formats = []
for format_el in formats_list:
url = format_el.get('source')
if url:
encoding_option = format_el.get('encodingOption')
bitrate = format_el.get('bitrate')
formats.append({
'format_id': encoding_option.get('id') or encoding_option.get('name'),
'url': format_el['source'],
'width': int_or_none(encoding_option.get('width')),
'height': int_or_none(encoding_option.get('height')),
'vbr': float_or_none(bitrate.get('video')),
'abr': float_or_none(bitrate.get('audio')),
'filesize': int_or_none(format_el.get('size')),
'vcodec': format_el.get('type'),
'ext': determine_ext(url, 'mp4'),
})
if formats:
self._sort_formats(formats)
return formats
def _extract_video_info(self, vid, key):
play_data = self._download_json(
'http://global.apis.naver.com/linetv/rmcnmv/vod_play_videoInfo.json?' + compat_urllib_parse.urlencode({'videoId': vid, 'key': key}),
vid, 'Downloading video info')
meta = play_data.get('meta')
user = meta.get('user')
thumbnails = []
for thumbnail in play_data['thumbnails']['list']:
thumbnails.append({'url': thumbnail['source']})
formats = self._extract_video_formats(play_data['videos']['list'])
if not formats:
video_info = self._download_json(
'http://serviceapi.rmcnmv.naver.com/mobile/getVideoInfo.nhn?' + compat_urllib_parse.urlencode({'videoId': vid, 'inKey': key, 'protocol': 'http'}),
vid, 'Downloading video info')
formats = self._extract_video_formats(video_info['videos']['list'])
return {
'id': vid,
'title': meta['subject'],
'formats': formats,
'thumbnail': meta.get('cover', {}).get('source'),
'thumbnails': thumbnails,
'view_count': int_or_none(meta.get('count')),
'uploader_id': user.get('id'),
'uploader': user.get('name'),
}
def _extract_id_and_key(self, webpage):
m_id = re.search(r'(?s)new\s+nhn.rmcnmv.RMCVideoPlayer\(\s*["\']([^"\']+)["\']\s*,\s*(?:{[^}]*?value[^:]*?:\s*?)?["\']([^"\']+)["\']', webpage)
if not m_id:
m_id = re.search(r'(?s)_sVid\s*=\s*["\']([^"\']+)["\'];\s*var\s+_sInkey\s*=\s*["\']([^"\']+)["\'];', webpage)
return m_id
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
m_id = re.search(r'var rmcPlayer = new nhn.rmcnmv.RMCVideoPlayer\("(.+?)", "(.+?)"',
webpage)
if m_id is None:
error = self._html_search_regex(
r'(?s)<div class="(?:nation_error|nation_box|error_box)">\s*(?:<!--.*?-->)?\s*<p class="[^"]+">(?P<msg>.+?)</p>\s*</div>',
webpage, 'error', default=None)
if error:
raise ExtractorError(error, expected=True)
raise ExtractorError('couldn\'t extract vid and key')
vid = m_id.group(1)
key = m_id.group(2)
query = compat_urllib_parse.urlencode({'vid': vid, 'inKey': key, })
query_urls = compat_urllib_parse.urlencode({
'masterVid': vid,
'protocol': 'p2p',
'inKey': key,
})
info = self._download_xml(
'http://serviceapi.rmcnmv.naver.com/flash/videoInfo.nhn?' + query,
video_id, 'Downloading video info')
urls = self._download_xml(
'http://serviceapi.rmcnmv.naver.com/flash/playableEncodingOption.nhn?' + query_urls,
video_id, 'Downloading video formats info')
formats = []
for format_el in urls.findall('EncodingOptions/EncodingOption'):
domain = format_el.find('Domain').text
uri = format_el.find('uri').text
f = {
'url': compat_urlparse.urljoin(domain, uri),
'ext': 'mp4',
'width': int(format_el.find('width').text),
'height': int(format_el.find('height').text),
}
if domain.startswith('rtmp'):
# urlparse does not support custom schemes
# https://bugs.python.org/issue18828
f.update({
'url': domain + uri,
'ext': 'flv',
'rtmp_protocol': '1', # rtmpt
})
formats.append(f)
self._sort_formats(formats)
return {
'id': video_id,
'title': info.find('Subject').text,
'formats': formats,
'description': self._og_search_description(webpage),
'thumbnail': self._og_search_thumbnail(webpage),
'upload_date': info.find('WriteDate').text.replace('.', ''),
'view_count': int(info.find('PlayCount').text),
}
m_id = self._extract_id_and_key(webpage)
if not m_id:
iframe_urls = re.findall(r'<(?:iframe|IFRAME)[^>]+src="((?:/main/readVod|/movie/bi/mi/videoPlayer|http://serviceapi\.rmcnmv\.naver\.com/flash/outKeyPlayer)\.nhn[^"]+)"', webpage)
if iframe_urls:
entries = []
for iframe_url in iframe_urls:
if iframe_url.startswith('/'):
iframe_url = compat_urlparse.urljoin(url, iframe_url)
request = compat_urllib_request.Request(iframe_url, headers={'Referer': url})
iframe_webpage = self._download_webpage(request, video_id, 'Downloading iframe webpage')
m_id = self._extract_id_and_key(iframe_webpage)
if m_id:
vid, key = m_id.groups()
entries.append(self._extract_video_info(vid, key))
return entries[0] if len(entries) == 1 else self.playlist_result(entries)
else:
error = self._html_search_regex(
r'(?s)<div class="(?:nation_error|nation_box|error_box)">\s*(?:<!--.*?-->)?\s*<p class="[^"]+">(?P<msg>.+?)</p>\s*</div>',
webpage, 'error', default=None)
if error:
raise ExtractorError(error, expected=True)
raise ExtractorError('couldn\'t extract vid and key')
vid, key = m_id.groups()
return self._extract_video_info(vid, key)