yt-dlc/youtube_dl/extractor/niconico.py

323 lines
12 KiB
Python
Raw Normal View History

2016-10-02 13:39:18 +02:00
# coding: utf-8
2014-03-27 21:01:09 +01:00
from __future__ import unicode_literals
2013-11-23 10:19:44 +01:00
import json
import datetime
2013-11-23 10:19:44 +01:00
from .common import InfoExtractor
from ..compat import (
2017-06-07 08:47:25 +02:00
compat_parse_qs,
2013-11-23 10:19:44 +01:00
compat_urlparse,
)
from ..utils import (
2017-06-07 08:47:25 +02:00
determine_ext,
dict_get,
ExtractorError,
int_or_none,
parse_duration,
parse_iso8601,
try_get,
unified_timestamp,
urlencode_postdata,
2017-06-07 08:47:25 +02:00
xpath_text,
2013-11-23 10:19:44 +01:00
)
2013-11-23 10:19:44 +01:00
class NiconicoIE(InfoExtractor):
2014-03-27 21:01:09 +01:00
IE_NAME = 'niconico'
IE_DESC = 'ニコニコ動画'
2013-11-23 10:19:44 +01:00
2015-03-20 15:53:14 +01:00
_TESTS = [{
2014-03-27 21:01:09 +01:00
'url': 'http://www.nicovideo.jp/watch/sm22312215',
'md5': 'd1a75c0823e2f629128c43e1212760f9',
'info_dict': {
'id': 'sm22312215',
'ext': 'mp4',
'title': 'Big Buck Bunny',
'thumbnail': r're:https?://.*',
2014-03-27 21:01:09 +01:00
'uploader': 'takuya0301',
'uploader_id': '2698420',
'upload_date': '20131123',
'timestamp': 1385182762,
2014-03-27 21:01:09 +01:00
'description': '(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org',
'duration': 33,
'view_count': int,
'comment_count': int,
2013-11-23 10:19:44 +01:00
},
'skip': 'Requires an account',
2015-03-20 15:53:14 +01:00
}, {
# File downloaded with and without credentials are different, so omit
# the md5 field
2015-03-20 15:53:14 +01:00
'url': 'http://www.nicovideo.jp/watch/nm14296458',
'info_dict': {
'id': 'nm14296458',
'ext': 'swf',
'title': '【鏡音リン】Dance on media【オリジナル】take2!',
'description': 'md5:689f066d74610b3b22e0f1739add0f58',
'thumbnail': r're:https?://.*',
2015-03-20 15:53:14 +01:00
'uploader': 'りょうた',
'uploader_id': '18822557',
'upload_date': '20110429',
'timestamp': 1304065916,
2015-03-20 15:53:14 +01:00
'duration': 209,
},
'skip': 'Requires an account',
}, {
# 'video exists but is marked as "deleted"
# md5 is unstable
'url': 'http://www.nicovideo.jp/watch/sm10000',
'info_dict': {
'id': 'sm10000',
'ext': 'unknown_video',
'description': 'deleted',
'title': 'ドラえもんエターナル第3話「決戦第3新東京市」前編',
'thumbnail': r're:https?://.*',
'upload_date': '20071224',
'timestamp': int, # timestamp field has different value if logged in
'duration': 304,
'view_count': int,
},
'skip': 'Requires an account',
}, {
'url': 'http://www.nicovideo.jp/watch/so22543406',
'info_dict': {
'id': '1388129933',
'ext': 'mp4',
'title': '【第1回】RADIOアニメロミックス ラブライブのぞえりRadio Garden',
'description': 'md5:b27d224bb0ff53d3c8269e9f8b561cf1',
'thumbnail': r're:https?://.*',
'timestamp': 1388851200,
'upload_date': '20140104',
'uploader': 'アニメロチャンネル',
'uploader_id': '312',
},
'skip': 'The viewing period of the video you were searching for has expired.',
}, {
# video not available via `getflv`
'url': 'http://www.nicovideo.jp/watch/sm1151009',
'info_dict': {
'id': 'sm1151009',
'ext': 'flv',
'title': 'マスターシステム本体内蔵のスペハリのメインテーマ(PSG版)',
'description': 'md5:6ee077e0581ff5019773e2e714cdd0b7',
'thumbnail': r're:https?://.*',
'duration': 184,
'timestamp': 1190868283,
'upload_date': '20070927',
'uploader': 'denden2',
'uploader_id': '1392194',
'view_count': int,
'comment_count': int,
},
'skip': 'Requires an account',
2017-06-25 21:30:05 +02:00
}, {
'url': 'http://sp.nicovideo.jp/watch/sm28964488?ss_pos=1&cp_in=wt_tg',
'only_matching': True,
2015-03-20 15:53:14 +01:00
}]
2013-11-23 10:19:44 +01:00
2017-06-25 21:30:05 +02:00
_VALID_URL = r'https?://(?:www\.|secure\.|sp\.)?nicovideo\.jp/watch/(?P<id>(?:[a-z]{2})?[0-9]+)'
2013-11-23 10:19:44 +01:00
_NETRC_MACHINE = 'niconico'
def _real_initialize(self):
self._login()
2013-11-23 10:19:44 +01:00
def _login(self):
(username, password) = self._get_login_info()
# No authentication to be performed
if not username:
return True
2013-11-23 10:19:44 +01:00
# Log in
2017-06-07 08:47:25 +02:00
login_ok = True
2013-11-23 10:19:44 +01:00
login_form_strs = {
2017-06-07 08:47:25 +02:00
'mail_tel': username,
2014-03-27 21:01:09 +01:00
'password': password,
2013-11-23 10:19:44 +01:00
}
2017-06-07 08:47:25 +02:00
urlh = self._request_webpage(
'https://account.nicovideo.jp/api/v1/login', None,
note='Logging in', errnote='Unable to log in',
data=urlencode_postdata(login_form_strs))
if urlh is False:
login_ok = False
else:
parts = compat_urlparse.urlparse(urlh.geturl())
if compat_parse_qs(parts.query).get('message', [None])[0] == 'cant_login':
login_ok = False
if not login_ok:
2014-03-27 21:01:09 +01:00
self._downloader.report_warning('unable to log in: bad username or password')
2017-06-07 08:47:25 +02:00
return login_ok
2013-11-23 10:19:44 +01:00
def _real_extract(self, url):
2015-03-09 18:12:41 +01:00
video_id = self._match_id(url)
2013-11-23 10:19:44 +01:00
# Get video webpage. We are not actually interested in it for normal
# cases, but need the cookies in order to be able to download the
# info webpage
webpage, handle = self._download_webpage_handle(
'http://www.nicovideo.jp/watch/' + video_id, video_id)
if video_id.startswith('so'):
video_id = self._match_id(handle.geturl())
2013-11-23 10:19:44 +01:00
api_data = self._parse_json(self._html_search_regex(
'data-api-data="([^"]+)"', webpage,
'API data', default='{}'), video_id)
video_real_url = try_get(
api_data, lambda x: x['video']['smileInfo']['url'])
if video_real_url:
def get_video_info(items):
return dict_get(api_data['video'], items)
else:
# Get flv info
flv_info_webpage = self._download_webpage(
'http://flapi.nicovideo.jp/api/getflv/' + video_id + '?as3=1',
video_id, 'Downloading flv info')
flv_info = compat_urlparse.parse_qs(flv_info_webpage)
if 'url' not in flv_info:
if 'deleted' in flv_info:
raise ExtractorError('The video has been deleted.',
expected=True)
elif 'closed' in flv_info:
raise ExtractorError('Niconico videos now require logging in',
expected=True)
elif 'error' in flv_info:
raise ExtractorError('%s reports error: %s' % (
self.IE_NAME, flv_info['error'][0]), expected=True)
else:
raise ExtractorError('Unable to find video URL')
video_real_url = flv_info['url'][0]
video_info_xml = self._download_xml(
'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id,
video_id, note='Downloading video info page')
def get_video_info(items):
if not isinstance(items, list):
items = [items]
for item in items:
ret = xpath_text(video_info_xml, './/' + item)
if ret:
return ret
2013-11-23 10:19:44 +01:00
# Start extracting information
title = get_video_info('title')
if not title:
title = self._og_search_title(webpage, default=None)
if not title:
title = self._html_search_regex(
r'<span[^>]+class="videoHeaderTitle"[^>]*>([^<]+)</span>',
webpage, 'video title')
watch_api_data_string = self._html_search_regex(
r'<div[^>]+id="watchAPIDataContainer"[^>]+>([^<]+)</div>',
webpage, 'watch api data', default=None)
watch_api_data = self._parse_json(watch_api_data_string, video_id) if watch_api_data_string else {}
video_detail = watch_api_data.get('videoDetail', {})
extension = get_video_info(['movie_type', 'movieType'])
if not extension:
extension = determine_ext(video_real_url)
thumbnail = (
get_video_info(['thumbnail_url', 'thumbnailURL']) or
self._html_search_meta('image', webpage, 'thumbnail', default=None) or
video_detail.get('thumbnail'))
description = get_video_info('description')
timestamp = (parse_iso8601(get_video_info('first_retrieve')) or
unified_timestamp(get_video_info('postedDateTime')))
if not timestamp:
match = self._html_search_meta('datePublished', webpage, 'date published', default=None)
if match:
timestamp = parse_iso8601(match.replace('+', ':00+'))
if not timestamp and video_detail.get('postedAt'):
timestamp = parse_iso8601(
video_detail['postedAt'].replace('/', '-'),
delimiter=' ', timezone=datetime.timedelta(hours=9))
view_count = int_or_none(get_video_info(['view_counter', 'viewCount']))
if not view_count:
match = self._html_search_regex(
r'>Views: <strong[^>]*>([^<]+)</strong>',
webpage, 'view count', default=None)
if match:
view_count = int_or_none(match.replace(',', ''))
view_count = view_count or video_detail.get('viewCount')
comment_count = (int_or_none(get_video_info('comment_num')) or
video_detail.get('commentCount') or
try_get(api_data, lambda x: x['thread']['commentCount']))
if not comment_count:
match = self._html_search_regex(
r'>Comments: <strong[^>]*>([^<]+)</strong>',
webpage, 'comment count', default=None)
if match:
comment_count = int_or_none(match.replace(',', ''))
duration = (parse_duration(
get_video_info('length') or
self._html_search_meta(
'video:duration', webpage, 'video duration', default=None)) or
video_detail.get('length') or
get_video_info('duration'))
webpage_url = get_video_info('watch_url') or url
owner = api_data.get('owner', {})
uploader_id = get_video_info(['ch_id', 'user_id']) or owner.get('id')
uploader = get_video_info(['ch_name', 'user_nickname']) or owner.get('nickname')
2013-11-23 10:19:44 +01:00
return {
2014-03-27 21:01:09 +01:00
'id': video_id,
'url': video_real_url,
'title': title,
'ext': extension,
2015-06-14 23:43:33 +02:00
'format_id': 'economy' if video_real_url.endswith('low') else 'normal',
'thumbnail': thumbnail,
'description': description,
'uploader': uploader,
'timestamp': timestamp,
'uploader_id': uploader_id,
'view_count': view_count,
'comment_count': comment_count,
'duration': duration,
'webpage_url': webpage_url,
2013-11-23 10:19:44 +01:00
}
class NiconicoPlaylistIE(InfoExtractor):
2016-09-08 13:29:05 +02:00
_VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/mylist/(?P<id>\d+)'
_TEST = {
'url': 'http://www.nicovideo.jp/mylist/27411728',
'info_dict': {
'id': '27411728',
'title': 'AKB48のオールナイトニッポン',
},
'playlist_mincount': 225,
}
def _real_extract(self, url):
list_id = self._match_id(url)
webpage = self._download_webpage(url, list_id)
entries_json = self._search_regex(r'Mylist\.preload\(\d+, (\[.*\])\);',
2014-11-23 21:39:15 +01:00
webpage, 'entries')
entries = json.loads(entries_json)
entries = [{
'_type': 'url',
'ie_key': NiconicoIE.ie_key(),
'url': ('http://www.nicovideo.jp/watch/%s' %
2014-11-23 21:39:15 +01:00
entry['item_data']['video_id']),
} for entry in entries]
return {
'_type': 'playlist',
'title': self._search_regex(r'\s+name: "(.*?)"', webpage, 'title'),
'id': list_id,
'entries': entries,
}