import re
import time
+from urllib.parse import urlparse
+
from .common import InfoExtractor, SearchInfoExtractor
-from ..compat import (
- compat_HTTPError,
-)
+from ..networking import Request
+from ..networking.exceptions import HTTPError
from ..utils import (
ExtractorError,
OnDemandPagedList,
_VALID_URL = r'https?://(?:(?:www\.|secure\.|sp\.)?nicovideo\.jp/watch|nico\.ms)/(?P<id>(?:[a-z]{2})?[0-9]+)'
_NETRC_MACHINE = 'niconico'
- _COMMENT_API_ENDPOINTS = (
- 'https://nvcomment.nicovideo.jp/legacy/api.json',
- 'https://nmsg.nicovideo.jp/api.json',)
_API_HEADERS = {
'X-Frontend-ID': '6',
'X-Frontend-Version': '0',
webpage, handle = self._download_webpage_handle(
'https://www.nicovideo.jp/watch/' + video_id, video_id)
if video_id.startswith('so'):
- video_id = self._match_id(handle.geturl())
+ video_id = self._match_id(handle.url)
api_data = self._parse_json(self._html_search_regex(
'data-api-data="([^"]+)"', webpage,
'https://www.nicovideo.jp/api/watch/v3/%s?_frontendId=6&_frontendVersion=0&actionTrackId=AAAAAAAAAA_%d' % (video_id, round(time.time() * 1000)), video_id,
note='Downloading API JSON', errnote='Unable to fetch data')['data']
except ExtractorError:
- if not isinstance(e.cause, compat_HTTPError):
+ if not isinstance(e.cause, HTTPError):
raise
- webpage = e.cause.read().decode('utf-8', 'replace')
+ webpage = e.cause.response.read().decode('utf-8', 'replace')
error_msg = self._html_search_regex(
r'(?s)<section\s+class="(?:(?:ErrorMessage|WatchExceptionPage-message)\s*)+">(.+?)</section>',
webpage, 'error reason', default=None)
parse_duration(self._html_search_meta('video:duration', webpage, 'video duration', default=None))
or get_video_info('duration')),
'webpage_url': url_or_none(url) or f'https://www.nicovideo.jp/watch/{video_id}',
- 'subtitles': self.extract_subtitles(video_id, api_data, session_api_data),
+ 'subtitles': self.extract_subtitles(video_id, api_data),
}
- def _get_subtitles(self, video_id, api_data, session_api_data):
- comment_user_key = traverse_obj(api_data, ('comment', 'keys', 'userKey'))
- user_id_str = session_api_data.get('serviceUserId')
-
- thread_ids = traverse_obj(api_data, ('comment', 'threads', lambda _, v: v['isActive']))
- raw_danmaku = self._extract_all_comments(video_id, thread_ids, user_id_str, comment_user_key)
- if not raw_danmaku:
+ def _get_subtitles(self, video_id, api_data):
+ comments_info = traverse_obj(api_data, ('comment', 'nvComment', {dict})) or {}
+ danmaku = traverse_obj(self._download_json(
+ f'{comments_info.get("server")}/v1/threads', video_id, data=json.dumps({
+ 'additionals': {},
+ 'params': comments_info.get('params'),
+ 'threadKey': comments_info.get('threadKey'),
+ }).encode(), fatal=False,
+ headers={
+ 'Referer': 'https://www.nicovideo.jp/',
+ 'Origin': 'https://www.nicovideo.jp',
+ 'Content-Type': 'text/plain;charset=UTF-8',
+ 'x-client-os-type': 'others',
+ 'x-frontend-id': '6',
+ 'x-frontend-version': '0',
+ },
+ note='Downloading comments', errnote='Failed to download comments'),
+ ('data', 'threads', ..., 'comments', ...))
+
+ if not danmaku:
self.report_warning(f'Failed to get comments. {bug_reports_message()}')
return
+
return {
'comments': [{
'ext': 'json',
- 'data': json.dumps(raw_danmaku),
+ 'data': json.dumps(danmaku),
}],
}
- def _extract_all_comments(self, video_id, threads, user_id, user_key):
- auth_data = {
- 'user_id': user_id,
- 'userkey': user_key,
- } if user_id and user_key else {'user_id': ''}
-
- # Request Start
- post_data = [{'ping': {'content': 'rs:0'}}]
- for i, thread in enumerate(threads):
- thread_id = thread['id']
- thread_fork = thread['fork']
- # Post Start (2N)
- post_data.append({'ping': {'content': f'ps:{i * 2}'}})
- post_data.append({'thread': {
- 'fork': thread_fork,
- 'language': 0,
- 'nicoru': 3,
- 'scores': 1,
- 'thread': thread_id,
- 'version': '20090904',
- 'with_global': 1,
- **auth_data,
- }})
- # Post Final (2N)
- post_data.append({'ping': {'content': f'pf:{i * 2}'}})
-
- # Post Start (2N+1)
- post_data.append({'ping': {'content': f'ps:{i * 2 + 1}'}})
- post_data.append({'thread_leaves': {
- # format is '<bottom of minute range>-<top of minute range>:<comments per minute>,<total last comments'
- # unfortunately NND limits (deletes?) comment returns this way, so you're only able to grab the last 1000 per language
- 'content': '0-999999:999999,999999,nicoru:999999',
- 'fork': thread_fork,
- 'language': 0,
- 'nicoru': 3,
- 'scores': 1,
- 'thread': thread_id,
- **auth_data,
- }})
- # Post Final (2N+1)
- post_data.append({'ping': {'content': f'pf:{i * 2 + 1}'}})
- # Request Final
- post_data.append({'ping': {'content': 'rf:0'}})
-
- for api_url in self._COMMENT_API_ENDPOINTS:
- comments = self._download_json(
- api_url, video_id, data=json.dumps(post_data).encode(), fatal=False,
- headers={
- 'Referer': 'https://www.nicovideo.jp/watch/%s' % video_id,
- 'Origin': 'https://www.nicovideo.jp',
- 'Content-Type': 'text/plain;charset=UTF-8',
- },
- note='Downloading comments', errnote=f'Failed to access endpoint {api_url}')
- if comments:
- return comments
-
class NiconicoPlaylistBaseIE(InfoExtractor):
_PAGE_SIZE = 100
class NiconicoSeriesIE(InfoExtractor):
IE_NAME = 'niconico:series'
- _VALID_URL = r'https?://(?:(?:www\.|sp\.)?nicovideo\.jp|nico\.ms)/series/(?P<id>\d+)'
+ _VALID_URL = r'https?://(?:(?:www\.|sp\.)?nicovideo\.jp(?:/user/\d+)?|nico\.ms)/series/(?P<id>\d+)'
_TESTS = [{
- 'url': 'https://www.nicovideo.jp/series/110226',
+ 'url': 'https://www.nicovideo.jp/user/44113208/series/110226',
'info_dict': {
'id': '110226',
'title': 'ご立派ァ！のシリーズ',
def _real_extract(self, url):
list_id = self._match_id(url)
- webpage = self._download_webpage(f'https://www.nicovideo.jp/series/{list_id}', list_id)
+ webpage = self._download_webpage(url, list_id)
title = self._search_regex(
(r'<title>「(.+)（全',
webpage, 'title', fatal=False)
if title:
title = unescapeHTML(title)
- playlist = [
- self.url_result(f'https://www.nicovideo.jp/watch/{v_id}', video_id=v_id)
- for v_id in re.findall(r'data-href=[\'"](?:https://www\.nicovideo\.jp)?/watch/([a-z0-9]+)', webpage)]
- return self.playlist_result(playlist, list_id, title)
+ json_data = next(self._yield_json_ld(webpage, None, fatal=False))
+ return self.playlist_from_matches(
+ traverse_obj(json_data, ('itemListElement', ..., 'url')), list_id, title, ie=NiconicoIE)
class NiconicoHistoryIE(NiconicoPlaylistBaseIE):
IE_NAME = 'niconico:history'
- IE_DESC = 'NicoNico user history. Requires cookies.'
- _VALID_URL = r'https?://(?:www\.|sp\.)?nicovideo\.jp/my/history'
+ IE_DESC = 'NicoNico user history or likes. Requires cookies.'
+ _VALID_URL = r'https?://(?:www\.|sp\.)?nicovideo\.jp/my/(?P<id>history(?:/like)?)'
_TESTS = [{
'note': 'PC page, with /video',
'note': 'mobile page, without /video',
'url': 'https://sp.nicovideo.jp/my/history',
'only_matching': True,
+ }, {
+ 'note': 'PC page',
+ 'url': 'https://www.nicovideo.jp/my/history/like',
+ 'only_matching': True,
+ }, {
+ 'note': 'Mobile page',
+ 'url': 'https://sp.nicovideo.jp/my/history/like',
+ 'only_matching': True,
}]
def _call_api(self, list_id, resource, query):
+ path = 'likes' if list_id == 'history/like' else 'watch/history'
return self._download_json(
- 'https://nvapi.nicovideo.jp/v1/users/me/watch/history', 'history',
- f'Downloading {resource}', query=query,
- headers=self._API_HEADERS)['data']
+ f'https://nvapi.nicovideo.jp/v1/users/me/{path}', list_id,
+ f'Downloading {resource}', query=query, headers=self._API_HEADERS)['data']
def _real_extract(self, url):
- list_id = 'history'
+ list_id = self._match_id(url)
try:
- mylist = self._call_api(list_id, 'list', {
- 'pageSize': 1,
- })
+ mylist = self._call_api(list_id, 'list', {'pageSize': 1})
except ExtractorError as e:
- if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
- self.raise_login_required('You have to be logged in to get your watch history')
+ if isinstance(e.cause, HTTPError) and e.cause.status == 401:
+ self.raise_login_required('You have to be logged in to get your history')
raise
return self.playlist_result(self._entries(list_id), list_id, **self._parse_owner(mylist))
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())
+
+
+class NiconicoLiveIE(InfoExtractor):
+ IE_NAME = 'niconico:live'
+ IE_DESC = 'ニコニコ生放送'
+ _VALID_URL = r'https?://(?:sp\.)?live2?\.nicovideo\.jp/(?:watch|gate)/(?P<id>lv\d+)'
+ _TESTS = [{
+ 'note': 'this test case includes invisible characters for title, pasting them as-is',
+ 'url': 'https://live.nicovideo.jp/watch/lv339533123',
+ 'info_dict': {
+ 'id': 'lv339533123',
+ 'title': '激辛ペヤング食べます( ;ᯅ; )（歌枠オーディション参加中）',
+ 'view_count': 1526,
+ 'comment_count': 1772,
+ 'description': '初めましてもかって言います❕\nのんびり自由に適当に暮らしてます',
+ 'uploader': 'もか',
+ 'channel': 'ゲストさんのコミュニティ',
+ 'channel_id': 'co5776900',
+ 'channel_url': 'https://com.nicovideo.jp/community/co5776900',
+ 'timestamp': 1670677328,
+ 'is_live': True,
+ },
+ 'skip': 'livestream',
+ }, {
+ 'url': 'https://live2.nicovideo.jp/watch/lv339533123',
+ 'only_matching': True,
+ }, {
+ 'url': 'https://sp.live.nicovideo.jp/watch/lv339533123',
+ 'only_matching': True,
+ }, {
+ 'url': 'https://sp.live2.nicovideo.jp/watch/lv339533123',
+ 'only_matching': True,
+ }]
+
+ _KNOWN_LATENCY = ('high', 'low')
+
+ def _real_extract(self, url):
+ video_id = self._match_id(url)
+ webpage, urlh = self._download_webpage_handle(f'https://live.nicovideo.jp/watch/{video_id}', video_id)
+
+ embedded_data = self._parse_json(unescapeHTML(self._search_regex(
+ r'<script\s+id="embedded-data"\s*data-props="(.+?)"', webpage, 'embedded data')), video_id)
+
+ ws_url = traverse_obj(embedded_data, ('site', 'relive', 'webSocketUrl'))
+ if not ws_url:
+ raise ExtractorError('The live hasn\'t started yet or already ended.', expected=True)
+ ws_url = update_url_query(ws_url, {
+ 'frontend_id': traverse_obj(embedded_data, ('site', 'frontendId')) or '9',
+ })
+
+ hostname = remove_start(urlparse(urlh.url).hostname, 'sp.')
+ latency = try_get(self._configuration_arg('latency'), lambda x: x[0])
+ if latency not in self._KNOWN_LATENCY:
+ latency = 'high'
+
+ ws = self._request_webpage(
+ Request(ws_url, headers={'Origin': f'https://{hostname}'}),
+ video_id=video_id, note='Connecting to WebSocket server')
+
+ self.write_debug('[debug] Sending HLS server request')
+ ws.send(json.dumps({
+ 'type': 'startWatching',
+ 'data': {
+ 'stream': {
+ 'quality': 'abr',
+ 'protocol': 'hls+fmp4',
+ 'latency': latency,
+ 'chasePlay': False
+ },
+ 'room': {
+ 'protocol': 'webSocket',
+ 'commentable': True
+ },
+ 'reconnect': False,
+ }
+ }))
+
+ while True:
+ recv = ws.recv()
+ if not recv:
+ continue
+ data = json.loads(recv)
+ if not isinstance(data, dict):
+ continue
+ if data.get('type') == 'stream':
+ m3u8_url = data['data']['uri']
+ qualities = data['data']['availableQualities']
+ break
+ elif data.get('type') == 'disconnect':
+ self.write_debug(recv)
+ raise ExtractorError('Disconnected at middle of extraction')
+ elif data.get('type') == 'error':
+ self.write_debug(recv)
+ message = traverse_obj(data, ('body', 'code')) or recv
+ raise ExtractorError(message)
+ elif self.get_param('verbose', False):
+ if len(recv) > 100:
+ recv = recv[:100] + '...'
+ self.write_debug('Server said: %s' % recv)
+
+ title = traverse_obj(embedded_data, ('program', 'title')) or self._html_search_meta(
+ ('og:title', 'twitter:title'), webpage, 'live title', fatal=False)
+
+ raw_thumbs = traverse_obj(embedded_data, ('program', 'thumbnail')) or {}
+ thumbnails = []
+ for name, value in raw_thumbs.items():
+ if not isinstance(value, dict):
+ thumbnails.append({
+ 'id': name,
+ 'url': value,
+ **parse_resolution(value, lenient=True),
+ })
+ continue
+
+ for k, img_url in value.items():
+ res = parse_resolution(k, lenient=True) or parse_resolution(img_url, lenient=True)
+ width, height = res.get('width'), res.get('height')
+
+ thumbnails.append({
+ 'id': f'{name}_{width}x{height}',
+ 'url': img_url,
+ **res,
+ })
+
+ formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4', live=True)
+ for fmt, q in zip(formats, reversed(qualities[1:])):
+ fmt.update({
+ 'format_id': q,
+ 'protocol': 'niconico_live',
+ 'ws': ws,
+ 'video_id': video_id,
+ 'live_latency': latency,
+ 'origin': hostname,
+ })
+
+ return {
+ 'id': video_id,
+ 'title': title,
+ **traverse_obj(embedded_data, {
+ 'view_count': ('program', 'statistics', 'watchCount'),
+ 'comment_count': ('program', 'statistics', 'commentCount'),
+ 'uploader': ('program', 'supplier', 'name'),
+ 'channel': ('socialGroup', 'name'),
+ 'channel_id': ('socialGroup', 'id'),
+ 'channel_url': ('socialGroup', 'socialGroupPageUrl'),
+ }),
+ 'description': clean_html(traverse_obj(embedded_data, ('program', 'description'))),
+ 'timestamp': int_or_none(traverse_obj(embedded_data, ('program', 'openTime'))),
+ 'is_live': True,
+ 'thumbnails': thumbnails,
+ 'formats': formats,
+ }
projects / yt-dlp.git / blobdiff