Unverified Commit 463e7216 authored by Yen Chi Hsuan's avatar Yen Chi Hsuan
Browse files

[niconico] Support HTML5-only videos (closes #13806)

parent 903a183b
version <unreleased>
Extractors
+ [niconico] Support HTML5-only videos (#13806)
version 2017.08.06 version 2017.08.06
Core Core
......
...@@ -11,10 +11,13 @@ from ..compat import ( ...@@ -11,10 +11,13 @@ from ..compat import (
) )
from ..utils import ( from ..utils import (
determine_ext, determine_ext,
dict_get,
ExtractorError, ExtractorError,
int_or_none, int_or_none,
parse_duration, parse_duration,
parse_iso8601, parse_iso8601,
try_get,
unified_timestamp,
urlencode_postdata, urlencode_postdata,
xpath_text, xpath_text,
) )
...@@ -31,12 +34,15 @@ class NiconicoIE(InfoExtractor): ...@@ -31,12 +34,15 @@ class NiconicoIE(InfoExtractor):
'id': 'sm22312215', 'id': 'sm22312215',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Big Buck Bunny', 'title': 'Big Buck Bunny',
'thumbnail': r're:https?://.*',
'uploader': 'takuya0301', 'uploader': 'takuya0301',
'uploader_id': '2698420', 'uploader_id': '2698420',
'upload_date': '20131123', 'upload_date': '20131123',
'timestamp': 1385182762, 'timestamp': 1385182762,
'description': '(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org', 'description': '(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org',
'duration': 33, 'duration': 33,
'view_count': int,
'comment_count': int,
}, },
'skip': 'Requires an account', 'skip': 'Requires an account',
}, { }, {
...@@ -48,6 +54,7 @@ class NiconicoIE(InfoExtractor): ...@@ -48,6 +54,7 @@ class NiconicoIE(InfoExtractor):
'ext': 'swf', 'ext': 'swf',
'title': '【鏡音リン】Dance on media【オリジナル】take2!', 'title': '【鏡音リン】Dance on media【オリジナル】take2!',
'description': 'md5:689f066d74610b3b22e0f1739add0f58', 'description': 'md5:689f066d74610b3b22e0f1739add0f58',
'thumbnail': r're:https?://.*',
'uploader': 'りょうた', 'uploader': 'りょうた',
'uploader_id': '18822557', 'uploader_id': '18822557',
'upload_date': '20110429', 'upload_date': '20110429',
...@@ -64,9 +71,11 @@ class NiconicoIE(InfoExtractor): ...@@ -64,9 +71,11 @@ class NiconicoIE(InfoExtractor):
'ext': 'unknown_video', 'ext': 'unknown_video',
'description': 'deleted', '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>', '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', 'upload_date': '20071224',
'timestamp': int, # timestamp field has different value if logged in 'timestamp': int, # timestamp field has different value if logged in
'duration': 304, 'duration': 304,
'view_count': int,
}, },
'skip': 'Requires an account', 'skip': 'Requires an account',
}, { }, {
...@@ -76,12 +85,31 @@ class NiconicoIE(InfoExtractor): ...@@ -76,12 +85,31 @@ class NiconicoIE(InfoExtractor):
'ext': 'mp4', '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~', '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', 'description': 'md5:b27d224bb0ff53d3c8269e9f8b561cf1',
'thumbnail': r're:https?://.*',
'timestamp': 1388851200, 'timestamp': 1388851200,
'upload_date': '20140104', 'upload_date': '20140104',
'uploader': 'アニメロチャンネル', 'uploader': 'アニメロチャンネル',
'uploader_id': '312', 'uploader_id': '312',
}, },
'skip': 'The viewing period of the video you were searching for has expired.', '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',
}, { }, {
'url': 'http://sp.nicovideo.jp/watch/sm28964488?ss_pos=1&cp_in=wt_tg', 'url': 'http://sp.nicovideo.jp/watch/sm28964488?ss_pos=1&cp_in=wt_tg',
'only_matching': True, 'only_matching': True,
...@@ -130,33 +158,51 @@ class NiconicoIE(InfoExtractor): ...@@ -130,33 +158,51 @@ class NiconicoIE(InfoExtractor):
if video_id.startswith('so'): if video_id.startswith('so'):
video_id = self._match_id(handle.geturl()) video_id = self._match_id(handle.geturl())
video_info = self._download_xml( api_data = self._parse_json(self._html_search_regex(
'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id, 'data-api-data="([^"]+)"', webpage,
note='Downloading video info page') 'API data', default='{}'), video_id)
video_real_url = try_get(
# Get flv info api_data, lambda x: x['video']['smileInfo']['url'])
flv_info_webpage = self._download_webpage(
'http://flapi.nicovideo.jp/api/getflv/' + video_id + '?as3=1', if video_real_url:
video_id, 'Downloading flv info') def get_video_info(items):
return dict_get(api_data['video'], items)
flv_info = compat_urlparse.parse_qs(flv_info_webpage) else:
if 'url' not in flv_info: # Get flv info
if 'deleted' in flv_info: flv_info_webpage = self._download_webpage(
raise ExtractorError('The video has been deleted.', 'http://flapi.nicovideo.jp/api/getflv/' + video_id + '?as3=1',
expected=True) video_id, 'Downloading flv info')
elif 'closed' in flv_info:
raise ExtractorError('Niconico videos now require logging in', flv_info = compat_urlparse.parse_qs(flv_info_webpage)
expected=True) if 'url' not in flv_info:
elif 'error' in flv_info: if 'deleted' in flv_info:
raise ExtractorError('%s reports error: %s' % ( raise ExtractorError('The video has been deleted.',
self.IE_NAME, flv_info['error'][0]), expected=True) expected=True)
else: elif 'closed' in flv_info:
raise ExtractorError('Unable to find video URL') raise ExtractorError('Niconico videos now require logging in',
expected=True)
video_real_url = flv_info['url'][0] 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
# Start extracting information # Start extracting information
title = xpath_text(video_info, './/title') title = get_video_info('title')
if not title: if not title:
title = self._og_search_title(webpage, default=None) title = self._og_search_title(webpage, default=None)
if not title: if not title:
...@@ -170,18 +216,19 @@ class NiconicoIE(InfoExtractor): ...@@ -170,18 +216,19 @@ class NiconicoIE(InfoExtractor):
watch_api_data = self._parse_json(watch_api_data_string, video_id) if watch_api_data_string else {} 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', {}) video_detail = watch_api_data.get('videoDetail', {})
extension = xpath_text(video_info, './/movie_type') extension = get_video_info(['movie_type', 'movieType'])
if not extension: if not extension:
extension = determine_ext(video_real_url) extension = determine_ext(video_real_url)
thumbnail = ( thumbnail = (
xpath_text(video_info, './/thumbnail_url') or get_video_info(['thumbnail_url', 'thumbnailURL']) or
self._html_search_meta('image', webpage, 'thumbnail', default=None) or self._html_search_meta('image', webpage, 'thumbnail', default=None) or
video_detail.get('thumbnail')) video_detail.get('thumbnail'))
description = xpath_text(video_info, './/description') description = get_video_info('description')
timestamp = parse_iso8601(xpath_text(video_info, './/first_retrieve')) timestamp = (parse_iso8601(get_video_info('first_retrieve')) or
unified_timestamp(get_video_info('postedDateTime')))
if not timestamp: if not timestamp:
match = self._html_search_meta('datePublished', webpage, 'date published', default=None) match = self._html_search_meta('datePublished', webpage, 'date published', default=None)
if match: if match:
...@@ -191,7 +238,7 @@ class NiconicoIE(InfoExtractor): ...@@ -191,7 +238,7 @@ class NiconicoIE(InfoExtractor):
video_detail['postedAt'].replace('/', '-'), video_detail['postedAt'].replace('/', '-'),
delimiter=' ', timezone=datetime.timedelta(hours=9)) delimiter=' ', timezone=datetime.timedelta(hours=9))
view_count = int_or_none(xpath_text(video_info, './/view_counter')) view_count = int_or_none(get_video_info(['view_counter', 'viewCount']))
if not view_count: if not view_count:
match = self._html_search_regex( match = self._html_search_regex(
r'>Views: <strong[^>]*>([^<]+)</strong>', r'>Views: <strong[^>]*>([^<]+)</strong>',
...@@ -200,31 +247,28 @@ class NiconicoIE(InfoExtractor): ...@@ -200,31 +247,28 @@ class NiconicoIE(InfoExtractor):
view_count = int_or_none(match.replace(',', '')) view_count = int_or_none(match.replace(',', ''))
view_count = view_count or video_detail.get('viewCount') view_count = view_count or video_detail.get('viewCount')
comment_count = int_or_none(xpath_text(video_info, './/comment_num')) 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: if not comment_count:
match = self._html_search_regex( match = self._html_search_regex(
r'>Comments: <strong[^>]*>([^<]+)</strong>', r'>Comments: <strong[^>]*>([^<]+)</strong>',
webpage, 'comment count', default=None) webpage, 'comment count', default=None)
if match: if match:
comment_count = int_or_none(match.replace(',', '')) comment_count = int_or_none(match.replace(',', ''))
comment_count = comment_count or video_detail.get('commentCount')
duration = (parse_duration( duration = (parse_duration(
xpath_text(video_info, './/length') or get_video_info('length') or
self._html_search_meta( self._html_search_meta(
'video:duration', webpage, 'video duration', default=None)) or 'video:duration', webpage, 'video duration', default=None)) or
video_detail.get('length')) video_detail.get('length') or
get_video_info('duration'))
webpage_url = xpath_text(video_info, './/watch_url') or url webpage_url = get_video_info('watch_url') or url
if video_info.find('.//ch_id') is not None: owner = api_data.get('owner', {})
uploader_id = video_info.find('.//ch_id').text uploader_id = get_video_info(['ch_id', 'user_id']) or owner.get('id')
uploader = video_info.find('.//ch_name').text uploader = get_video_info(['ch_name', 'user_nickname']) or owner.get('nickname')
elif video_info.find('.//user_id') is not None:
uploader_id = video_info.find('.//user_id').text
uploader = video_info.find('.//user_nickname').text
else:
uploader_id = uploader = None
return { return {
'id': video_id, 'id': video_id,
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment