# coding: utf-8
from __future__ import unicode_literals
-import re
-import json
import datetime
+import itertools
+import json
+import re
-from .common import InfoExtractor
+from .common import InfoExtractor, SearchInfoExtractor
from ..postprocessor.ffmpeg import FFmpegPostProcessor
from ..compat import (
+ compat_str,
compat_parse_qs,
compat_urllib_parse_urlparse,
)
from ..utils import (
- dict_get,
ExtractorError,
- int_or_none,
+ dict_get,
float_or_none,
+ int_or_none,
OnDemandPagedList,
parse_duration,
parse_iso8601,
PostProcessingError,
remove_start,
+ str_or_none,
try_get,
unified_timestamp,
urlencode_postdata,
_TESTS = [{
'url': 'http://www.nicovideo.jp/watch/sm22312215',
- 'md5': 'd1a75c0823e2f629128c43e1212760f9',
+ 'md5': 'a5bad06f1347452102953f323c69da34s',
'info_dict': {
'id': 'sm22312215',
'ext': 'mp4',
_VALID_URL = r'https?://(?:www\.|secure\.|sp\.)?nicovideo\.jp/watch/(?P<id>(?:[a-z]{2})?[0-9]+)'
_NETRC_MACHINE = 'niconico'
+ _API_HEADERS = {
+ 'X-Frontend-ID': '6',
+ 'X-Frontend-Version': '0'
+ }
+
def _real_initialize(self):
self._login()
if compat_parse_qs(parts.query).get('message', [None])[0] == 'cant_login':
login_ok = False
if not login_ok:
- self._downloader.report_warning('unable to log in: bad username or password')
+ self.report_warning('unable to log in: bad username or password')
return login_ok
def _get_heartbeat_info(self, info_dict):
video_id, video_src_id, audio_src_id = info_dict['url'].split(':')[1].split('/')
- # Get video webpage for API data.
- webpage, handle = self._download_webpage_handle(
- 'http://www.nicovideo.jp/watch/' + video_id, video_id)
-
- api_data = self._parse_json(self._html_search_regex(
- 'data-api-data="([^"]+)"', webpage,
- 'API data', default='{}'), video_id)
+ api_data = (
+ info_dict.get('_api_data')
+ or self._parse_json(
+ self._html_search_regex(
+ 'data-api-data="([^"]+)"',
+ self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id),
+ 'API data', default='{}'),
+ video_id))
- session_api_data = try_get(api_data, lambda x: x['video']['dmcInfo']['session_api'])
+ session_api_data = try_get(api_data, lambda x: x['media']['delivery']['movie']['session'])
session_api_endpoint = try_get(session_api_data, lambda x: x['urls'][0])
- # ping
- self._download_json(
- 'https://nvapi.nicovideo.jp/v1/2ab0cbaa/watch', video_id,
- query={'t': try_get(api_data, lambda x: x['video']['dmcInfo']['tracking_id'])},
- headers={
- 'Origin': 'https://www.nicovideo.jp',
- 'Referer': 'https://www.nicovideo.jp/watch/' + video_id,
- 'X-Frontend-Id': '6',
- 'X-Frontend-Version': '0'
- })
+ def ping():
+ status = try_get(
+ self._download_json(
+ 'https://nvapi.nicovideo.jp/v1/2ab0cbaa/watch', video_id,
+ query={'t': try_get(api_data, lambda x: x['media']['delivery']['trackingId'])},
+ note='Acquiring permission for downloading video',
+ headers=self._API_HEADERS),
+ lambda x: x['meta']['status'])
+ if status != 200:
+ self.report_warning('Failed to acquire permission for playing video. The video may not download.')
yesno = lambda x: 'yes' if x else 'no'
# m3u8 (encryption)
- if 'encryption' in try_get(api_data, lambda x: x['video']['dmcInfo']) or {}:
+ if try_get(api_data, lambda x: x['media']['delivery']['encryption']) is not None:
protocol = 'm3u8'
+ encryption = self._parse_json(session_api_data['token'], video_id)['hls_encryption']
session_api_http_parameters = {
'parameters': {
'hls_parameters': {
'encryption': {
- 'hls_encryption_v1': {
- 'encrypted_key': try_get(api_data, lambda x: x['video']['dmcInfo']['encryption']['hls_encryption_v1']['encrypted_key']),
- 'key_uri': try_get(api_data, lambda x: x['video']['dmcInfo']['encryption']['hls_encryption_v1']['key_uri'])
+ encryption: {
+ 'encrypted_key': try_get(api_data, lambda x: x['media']['delivery']['encryption']['encryptedKey']),
+ 'key_uri': try_get(api_data, lambda x: x['media']['delivery']['encryption']['keyUri'])
}
},
'transfer_preset': '',
- 'use_ssl': yesno(session_api_endpoint['is_ssl']),
- 'use_well_known_port': yesno(session_api_endpoint['is_well_known_port']),
- 'segment_duration': 6000
+ 'use_ssl': yesno(session_api_endpoint['isSsl']),
+ 'use_well_known_port': yesno(session_api_endpoint['isWellKnownPort']),
+ 'segment_duration': 6000,
}
}
}
session_api_http_parameters = {
'parameters': {
'http_output_download_parameters': {
- 'use_ssl': yesno(session_api_endpoint['is_ssl']),
- 'use_well_known_port': yesno(session_api_endpoint['is_well_known_port']),
+ 'use_ssl': yesno(session_api_endpoint['isSsl']),
+ 'use_well_known_port': yesno(session_api_endpoint['isWellKnownPort']),
}
}
}
data=json.dumps({
'session': {
'client_info': {
- 'player_id': session_api_data.get('player_id'),
+ 'player_id': session_api_data.get('playerId'),
},
'content_auth': {
- 'auth_type': try_get(session_api_data, lambda x: x['auth_types'][session_api_data['protocols'][0]]),
- 'content_key_timeout': session_api_data.get('content_key_timeout'),
+ 'auth_type': try_get(session_api_data, lambda x: x['authTypes'][session_api_data['protocols'][0]]),
+ 'content_key_timeout': session_api_data.get('contentKeyTimeout'),
'service_id': 'nicovideo',
- 'service_user_id': session_api_data.get('service_user_id')
+ 'service_user_id': session_api_data.get('serviceUserId')
},
- 'content_id': session_api_data.get('content_id'),
+ 'content_id': session_api_data.get('contentId'),
'content_src_id_sets': [{
'content_src_ids': [{
'src_id_to_mux': {
'content_uri': '',
'keep_method': {
'heartbeat': {
- 'lifetime': session_api_data.get('heartbeat_lifetime')
+ 'lifetime': session_api_data.get('heartbeatLifetime')
}
},
'priority': session_api_data.get('priority'),
'http_parameters': session_api_http_parameters
}
},
- 'recipe_id': session_api_data.get('recipe_id'),
+ 'recipe_id': session_api_data.get('recipeId'),
'session_operation_auth': {
'session_operation_auth_by_signature': {
'signature': session_api_data.get('signature'),
'url': session_api_endpoint['url'] + '/' + session_response['data']['session']['id'] + '?_format=json&_method=PUT',
'data': json.dumps(session_response['data']),
# interval, convert milliseconds to seconds, then halve to make a buffer.
- 'interval': float_or_none(session_api_data.get('heartbeat_lifetime'), scale=2000),
+ 'interval': float_or_none(session_api_data.get('heartbeatLifetime'), scale=3000),
+ 'ping': ping
}
return info_dict, heartbeat_info_dict
format_id = '-'.join(map(lambda s: remove_start(s['id'], 'archive_'), [video_quality, audio_quality]))
vdict = parse_format_id(video_quality['id'])
adict = parse_format_id(audio_quality['id'])
- resolution = video_quality.get('resolution', {'height': vdict.get('res')})
+ resolution = try_get(video_quality, lambda x: x['metadata']['resolution'], dict) or {'height': vdict.get('res')}
+ vbr = try_get(video_quality, lambda x: x['metadata']['bitrate'], float)
return {
'url': '%s:%s/%s/%s' % (protocol, video_id, video_quality['id'], audio_quality['id']),
'format_id': format_id,
+ 'format_note': 'DMC %s' % try_get(video_quality, lambda x: x['metadata']['label'], compat_str),
'ext': 'mp4', # Session API are used in HTML5, which always serves mp4
'vcodec': vdict.get('codec'),
'acodec': adict.get('codec'),
- 'vbr': float_or_none(video_quality.get('bitrate'), 1000) or float_or_none(vdict.get('br')),
+ 'vbr': float_or_none(vbr, 1000) or float_or_none(vdict.get('br')),
'abr': float_or_none(audio_quality.get('bitrate'), 1000) or float_or_none(adict.get('br')),
'height': int_or_none(resolution.get('height', vdict.get('res'))),
'width': int_or_none(resolution.get('width')),
formats = []
# Get HTML5 videos info
- try:
- dmc_info = api_data['video']['dmcInfo']
- except KeyError:
- raise ExtractorError('The video can\'t downloaded.',
- expected=True)
+ quality_info = try_get(api_data, lambda x: x['media']['delivery']['movie'])
+ if not quality_info:
+ raise ExtractorError('The video can\'t be downloaded', expected=True)
- quality_info = dmc_info.get('quality')
for audio_quality in quality_info.get('audios') or {}:
for video_quality in quality_info.get('videos') or {}:
- if not audio_quality.get('available') or not video_quality.get('available'):
+ if not audio_quality.get('isAvailable') or not video_quality.get('isAvailable'):
continue
formats.append(self._extract_format_for_quality(
api_data, video_id, audio_quality, video_quality))
# Get flv/swf info
+ timestamp = None
video_real_url = try_get(api_data, lambda x: x['video']['smileInfo']['url'])
- is_economy = video_real_url.endswith('low')
-
- if is_economy:
- self.report_warning('Site is currently in economy mode! You will only have access to lower quality streams')
-
- # Invoking ffprobe to determine resolution
- pp = FFmpegPostProcessor(self._downloader)
- cookies = self._get_cookies('https://nicovideo.jp').output(header='', sep='; path=/; domain=nicovideo.jp;\n')
-
- self.to_screen('%s: %s' % (video_id, 'Checking smile format with ffprobe'))
-
- try:
- metadata = pp.get_metadata_object(video_real_url, ['-cookies', cookies])
- except PostProcessingError as err:
- raise ExtractorError(err.msg, expected=True)
-
- v_stream = a_stream = {}
-
- # Some complex swf files doesn't have video stream (e.g. nm4809023)
- for stream in metadata['streams']:
- if stream['codec_type'] == 'video':
- v_stream = stream
- elif stream['codec_type'] == 'audio':
- a_stream = stream
-
- # Community restricted videos seem to have issues with the thumb API not returning anything at all
- filesize = int(
- (get_video_info_xml('size_high') if not is_economy else get_video_info_xml('size_low'))
- or metadata['format']['size']
- )
- extension = (
- get_video_info_xml('movie_type')
- or 'mp4' if 'mp4' in metadata['format']['format_name'] else metadata['format']['format_name']
- )
-
- # 'creation_time' tag on video stream of re-encoded SMILEVIDEO mp4 files are '1970-01-01T00:00:00.000000Z'.
- timestamp = (
- parse_iso8601(get_video_info_web('first_retrieve'))
- or unified_timestamp(get_video_info_web('postedDateTime'))
- )
- metadata_timestamp = (
- parse_iso8601(try_get(v_stream, lambda x: x['tags']['creation_time']))
- or timestamp if extension != 'mp4' else 0
- )
-
- # According to compconf, smile videos from pre-2017 are always better quality than their DMC counterparts
- smile_threshold_timestamp = parse_iso8601('2016-12-08T00:00:00+09:00')
-
- is_source = timestamp < smile_threshold_timestamp or metadata_timestamp > 0
-
- # If movie file size is unstable, old server movie is not source movie.
- if filesize > 1:
- formats.append({
- 'url': video_real_url,
- 'format_id': 'smile' if not is_economy else 'smile_low',
- 'format_note': 'SMILEVIDEO source' if not is_economy else 'SMILEVIDEO low quality',
- 'ext': extension,
- 'container': extension,
- 'vcodec': v_stream.get('codec_name'),
- 'acodec': a_stream.get('codec_name'),
- # Some complex swf files doesn't have total bit rate metadata (e.g. nm6049209)
- 'tbr': int_or_none(metadata['format'].get('bit_rate'), scale=1000),
- 'vbr': int_or_none(v_stream.get('bit_rate'), scale=1000),
- 'abr': int_or_none(a_stream.get('bit_rate'), scale=1000),
- 'height': int_or_none(v_stream.get('height')),
- 'width': int_or_none(v_stream.get('width')),
- 'source_preference': 5 if not is_economy else -2,
- 'quality': 5 if is_source and not is_economy else None,
- 'filesize': filesize
- })
-
- if len(formats) == 0:
- raise ExtractorError('Unable to find video info.')
+ if video_real_url:
+ is_economy = video_real_url.endswith('low')
+
+ if is_economy:
+ self.report_warning('Site is currently in economy mode! You will only have access to lower quality streams')
+
+ # Invoking ffprobe to determine resolution
+ pp = FFmpegPostProcessor(self._downloader)
+ cookies = self._get_cookies('https://nicovideo.jp').output(header='', sep='; path=/; domain=nicovideo.jp;\n')
+
+ self.to_screen('%s: %s' % (video_id, 'Checking smile format with ffprobe'))
+
+ try:
+ metadata = pp.get_metadata_object(video_real_url, ['-cookies', cookies])
+ except PostProcessingError as err:
+ raise ExtractorError(err.msg, expected=True)
+
+ v_stream = a_stream = {}
+
+ # Some complex swf files doesn't have video stream (e.g. nm4809023)
+ for stream in metadata['streams']:
+ if stream['codec_type'] == 'video':
+ v_stream = stream
+ elif stream['codec_type'] == 'audio':
+ a_stream = stream
+
+ # Community restricted videos seem to have issues with the thumb API not returning anything at all
+ filesize = int(
+ (get_video_info_xml('size_high') if not is_economy else get_video_info_xml('size_low'))
+ or metadata['format']['size']
+ )
+ extension = (
+ get_video_info_xml('movie_type')
+ or 'mp4' if 'mp4' in metadata['format']['format_name'] else metadata['format']['format_name']
+ )
+
+ # 'creation_time' tag on video stream of re-encoded SMILEVIDEO mp4 files are '1970-01-01T00:00:00.000000Z'.
+ timestamp = (
+ parse_iso8601(get_video_info_web('first_retrieve'))
+ or unified_timestamp(get_video_info_web('postedDateTime'))
+ )
+ metadata_timestamp = (
+ parse_iso8601(try_get(v_stream, lambda x: x['tags']['creation_time']))
+ or timestamp if extension != 'mp4' else 0
+ )
+
+ # According to compconf, smile videos from pre-2017 are always better quality than their DMC counterparts
+ smile_threshold_timestamp = parse_iso8601('2016-12-08T00:00:00+09:00')
+
+ is_source = timestamp < smile_threshold_timestamp or metadata_timestamp > 0
+
+ # If movie file size is unstable, old server movie is not source movie.
+ if filesize > 1:
+ formats.append({
+ 'url': video_real_url,
+ 'format_id': 'smile' if not is_economy else 'smile_low',
+ 'format_note': 'SMILEVIDEO source' if not is_economy else 'SMILEVIDEO low quality',
+ 'ext': extension,
+ 'container': extension,
+ 'vcodec': v_stream.get('codec_name'),
+ 'acodec': a_stream.get('codec_name'),
+ # Some complex swf files doesn't have total bit rate metadata (e.g. nm6049209)
+ 'tbr': int_or_none(metadata['format'].get('bit_rate'), scale=1000),
+ 'vbr': int_or_none(v_stream.get('bit_rate'), scale=1000),
+ 'abr': int_or_none(a_stream.get('bit_rate'), scale=1000),
+ 'height': int_or_none(v_stream.get('height')),
+ 'width': int_or_none(v_stream.get('width')),
+ 'source_preference': 5 if not is_economy else -2,
+ 'quality': 5 if is_source and not is_economy else None,
+ 'filesize': filesize
+ })
self._sort_formats(formats)
# Start extracting information
- title = get_video_info_web('originalTitle')
- if not title:
- title = self._og_search_title(webpage, default=None)
- if not title:
- title = self._html_search_regex(
+ title = (
+ get_video_info_xml('title') # prefer to get the untranslated original title
+ or get_video_info_web(['originalTitle', 'title'])
+ or self._og_search_title(webpage, default=None)
+ or self._html_search_regex(
r'<span[^>]+class="videoHeaderTitle"[^>]*>([^<]+)</span>',
- webpage, 'video title')
+ webpage, 'video title'))
watch_api_data_string = self._html_search_regex(
r'<div[^>]+id="watchAPIDataContainer"[^>]+>([^<]+)</div>',
thumbnail = (
self._html_search_regex(r'<meta property="og:image" content="([^"]+)">', webpage, 'thumbnail data', default=None)
- or get_video_info_web(['thumbnail_url', 'largeThumbnailURL', 'thumbnailURL'])
+ or dict_get( # choose highest from 720p to 240p
+ get_video_info_web('thumbnail'),
+ ['ogp', 'player', 'largeUrl', 'middleUrl', 'url'])
or self._html_search_meta('image', webpage, 'thumbnail', default=None)
or video_detail.get('thumbnail'))
timestamp = parse_iso8601(
video_detail['postedAt'].replace('/', '-'),
delimiter=' ', timezone=datetime.timedelta(hours=9))
+ timestamp = timestamp or try_get(api_data, lambda x: parse_iso8601(x['video']['registeredAt']))
view_count = int_or_none(get_video_info_web(['view_counter', 'viewCount']))
if not view_count:
webpage, 'view count', default=None)
if match:
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')
+ or try_get(api_data, lambda x: x['video']['count']['view']))
+
+ comment_count = (
+ int_or_none(get_video_info_web('comment_num'))
+ or video_detail.get('commentCount')
+ or try_get(api_data, lambda x: x['video']['count']['comment']))
- comment_count = (int_or_none(get_video_info_web('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>',
# Note: cannot use api_data.get('owner', {}) because owner may be set to "null"
# in the JSON, which will cause None to be returned instead of {}.
owner = try_get(api_data, lambda x: x.get('owner'), dict) or {}
- uploader_id = (
+ uploader_id = str_or_none(
get_video_info_web(['ch_id', 'user_id'])
or owner.get('id')
or channel_id
return {
'id': video_id,
+ '_api_data': api_data,
'title': title,
'formats': formats,
'thumbnail': thumbnail,
class NiconicoPlaylistIE(InfoExtractor):
- _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/(?:user/\d+/)?mylist/(?P<id>\d+)'
+ _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/(?:user/\d+/|my/)?mylist/(?P<id>\d+)'
_TESTS = [{
'url': 'http://www.nicovideo.jp/mylist/27411728',
'only_matching': True,
}]
+ _API_HEADERS = {
+ 'X-Frontend-ID': '6',
+ 'X-Frontend-Version': '0'
+ }
+
def _real_extract(self, url):
list_id = self._match_id(url)
- webpage = self._download_webpage(url, list_id)
-
- header = self._parse_json(self._html_search_regex(
- r'data-common-header="([^"]+)"', webpage,
- 'webpage header'), list_id)
- frontendId = header.get('initConfig').get('frontendId')
- frontendVersion = header.get('initConfig').get('frontendVersion')
def get_page_data(pagenum, pagesize):
return self._download_json(
'http://nvapi.nicovideo.jp/v2/mylists/' + list_id, list_id,
query={'page': 1 + pagenum, 'pageSize': pagesize},
- headers={
- 'X-Frontend-Id': frontendId,
- 'X-Frontend-Version': frontendVersion,
- }).get('data').get('mylist')
+ headers=self._API_HEADERS).get('data').get('mylist')
data = get_page_data(0, 1)
title = data.get('name')
'uploader_id': uploader_id,
'entries': OnDemandPagedList(pagefunc, 25),
}
+
+
+NicovideoSearchIE_NAME = 'nicovideo:search'
+
+
+class NicovideoSearchURLIE(InfoExtractor):
+ IE_NAME = f'{NicovideoSearchIE_NAME}_url'
+ IE_DESC = 'Nico video search URLs'
+ _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/search/(?P<id>[^?#&]+)?'
+ _TESTS = [{
+ 'url': 'http://www.nicovideo.jp/search/sm9',
+ 'info_dict': {
+ 'id': 'sm9',
+ 'title': 'sm9'
+ },
+ 'playlist_mincount': 40,
+ }, {
+ 'url': 'https://www.nicovideo.jp/search/sm9?sort=h&order=d&end=2020-12-31&start=2020-01-01',
+ 'info_dict': {
+ 'id': 'sm9',
+ 'title': 'sm9'
+ },
+ 'playlist_count': 31,
+ }]
+
+ def _entries(self, url, item_id, query=None, note='Downloading page %(page)s'):
+ query = query or {}
+ pages = [query['page']] if 'page' in query else itertools.count(1)
+ for page_num in pages:
+ query['page'] = str(page_num)
+ webpage = self._download_webpage(url, item_id, query=query, note=note % {'page': page_num})
+ results = re.findall(r'(?<=data-video-id=)["\']?(?P<videoid>.*?)(?=["\'])', webpage)
+ for item in results:
+ yield self.url_result(f'http://www.nicovideo.jp/watch/{item}', 'Niconico', item)
+ if not results:
+ break
+
+ def _real_extract(self, url):
+ query = self._match_id(url)
+ return self.playlist_result(self._entries(url, query), query, query)
+
+
+class NicovideoSearchIE(SearchInfoExtractor, NicovideoSearchURLIE):
+ IE_DESC = 'Nico video searches'
+ IE_NAME = NicovideoSearchIE_NAME
+ _SEARCH_KEY = 'nicosearch'
+ _TESTS = []
+
+ def _search_results(self, query):
+ return self._entries(
+ self._proto_relative_url(f'//www.nicovideo.jp/search/{query}'), query)
+
+
+class NicovideoSearchDateIE(NicovideoSearchIE):
+ IE_DESC = 'Nico video searches, newest first'
+ IE_NAME = f'{NicovideoSearchIE_NAME}:date'
+ _SEARCH_KEY = 'nicosearchdate'
+ _TESTS = [{
+ 'url': 'nicosearchdateall:a',
+ 'info_dict': {
+ 'id': 'a',
+ 'title': 'a'
+ },
+ 'playlist_mincount': 1610,
+ }]
+
+ _START_DATE = datetime.date(2007, 1, 1)
+ _RESULTS_PER_PAGE = 32
+ _MAX_PAGES = 50
+
+ def _entries(self, url, item_id, start_date=None, end_date=None):
+ start_date, end_date = start_date or self._START_DATE, end_date or datetime.datetime.now().date()
+
+ # If the last page has a full page of videos, we need to break down the query interval further
+ last_page_len = len(list(self._get_entries_for_date(
+ url, item_id, start_date, end_date, self._MAX_PAGES,
+ note=f'Checking number of videos from {start_date} to {end_date}')))
+ if (last_page_len == self._RESULTS_PER_PAGE and start_date != end_date):
+ midpoint = start_date + ((end_date - start_date) // 2)
+ yield from self._entries(url, item_id, midpoint, end_date)
+ yield from self._entries(url, item_id, start_date, midpoint)
+ else:
+ self.to_screen(f'{item_id}: Downloading results from {start_date} to {end_date}')
+ yield from self._get_entries_for_date(
+ url, item_id, start_date, end_date, note=' Downloading page %(page)s')
+
+ def _get_entries_for_date(self, url, item_id, start_date, end_date=None, page_num=None, note=None):
+ query = {
+ 'start': str(start_date),
+ 'end': str(end_date or start_date),
+ 'sort': 'f',
+ 'order': 'd',
+ }
+ if page_num:
+ query['page'] = str(page_num)
+
+ yield from NicovideoSearchURLIE._entries(self, url, item_id, query=query, note=note)
+
+
+class NiconicoUserIE(InfoExtractor):
+ _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/user/(?P<id>\d+)/?(?:$|[#?])'
+ _TEST = {
+ 'url': 'https://www.nicovideo.jp/user/419948',
+ 'info_dict': {
+ 'id': '419948',
+ },
+ 'playlist_mincount': 101,
+ }
+ _API_URL = "https://nvapi.nicovideo.jp/v1/users/%s/videos?sortKey=registeredAt&sortOrder=desc&pageSize=%s&page=%s"
+ _PAGE_SIZE = 100
+
+ _API_HEADERS = {
+ 'X-Frontend-ID': '6',
+ 'X-Frontend-Version': '0'
+ }
+
+ def _entries(self, list_id):
+ total_count = 1
+ count = page_num = 0
+ while count < total_count:
+ json_parsed = self._download_json(
+ self._API_URL % (list_id, self._PAGE_SIZE, page_num + 1), list_id,
+ headers=self._API_HEADERS,
+ note='Downloading JSON metadata%s' % (' page %d' % page_num if page_num else ''))
+ if not page_num:
+ total_count = int_or_none(json_parsed['data'].get('totalCount'))
+ for entry in json_parsed["data"]["items"]:
+ count += 1
+ yield self.url_result('https://www.nicovideo.jp/watch/%s' % entry['id'])
+ page_num += 1
+
+ def _real_extract(self, url):
+ list_id = self._match_id(url)
+ return self.playlist_result(self._entries(list_id), list_id, ie=NiconicoIE.ie_key())
projects / yt-dlp.git / blobdiff