diff options
Diffstat (limited to 'hypervideo_dl/extractor/naver.py')
-rw-r--r-- | hypervideo_dl/extractor/naver.py | 172 |
1 files changed, 158 insertions, 14 deletions
diff --git a/hypervideo_dl/extractor/naver.py b/hypervideo_dl/extractor/naver.py index a6821ba..e2e6e97 100644 --- a/hypervideo_dl/extractor/naver.py +++ b/hypervideo_dl/extractor/naver.py @@ -1,16 +1,19 @@ -# coding: utf-8 -from __future__ import unicode_literals - +import itertools import re +from urllib.parse import urlparse, parse_qs from .common import InfoExtractor from ..utils import ( + ExtractorError, clean_html, dict_get, - ExtractorError, int_or_none, + join_nonempty, + merge_dicts, parse_duration, + traverse_obj, try_get, + unified_timestamp, update_url_query, ) @@ -65,19 +68,16 @@ class NaverBaseIE(InfoExtractor): formats.extend(self._extract_m3u8_formats( update_url_query(stream_url, query), video_id, 'mp4', 'm3u8_native', m3u8_id=stream_type, fatal=False)) - self._sort_formats(formats) replace_ext = lambda x, y: re.sub(self._CAPTION_EXT_RE, '.' + y, x) def get_subs(caption_url): if re.search(self._CAPTION_EXT_RE, caption_url): - return [{ - 'url': replace_ext(caption_url, 'ttml'), - }, { - 'url': replace_ext(caption_url, 'vtt'), - }] - else: - return [{'url': caption_url}] + return [ + replace_ext(caption_url, 'ttml'), + replace_ext(caption_url, 'vtt'), + ] + return [caption_url] automatic_captions = {} subtitles = {} @@ -86,7 +86,13 @@ class NaverBaseIE(InfoExtractor): if not caption_url: continue sub_dict = automatic_captions if caption.get('type') == 'auto' else subtitles - sub_dict.setdefault(dict_get(caption, ('locale', 'language')), []).extend(get_subs(caption_url)) + lang = caption.get('locale') or join_nonempty('language', 'country', from_dict=caption) or 'und' + if caption.get('type') == 'fan': + lang += '_fan%d' % next(i for i in itertools.count(1) if f'{lang}_fan{i}' not in sub_dict) + sub_dict.setdefault(lang, []).extend({ + 'url': sub_url, + 'name': join_nonempty('label', 'fanName', from_dict=caption, delim=' - '), + } for sub_url in get_subs(caption_url)) user = meta.get('user', {}) @@ -237,7 +243,6 @@ class NaverLiveIE(InfoExtractor): quality.get('url'), video_id, 'mp4', m3u8_id=quality.get('qualityId'), live=True )) - self._sort_formats(formats) return { 'id': video_id, @@ -250,3 +255,142 @@ class NaverLiveIE(InfoExtractor): 'categories': [meta.get('categoryId')], 'is_live': True } + + +class NaverNowIE(NaverBaseIE): + IE_NAME = 'navernow' + _VALID_URL = r'https?://now\.naver\.com/s/now\.(?P<id>\w+)' + _API_URL = 'https://apis.naver.com/now_web/oldnow_web/v4' + _TESTS = [{ + 'url': 'https://now.naver.com/s/now.4759?shareReplayId=26331132#replay=', + 'md5': 'e05854162c21c221481de16b2944a0bc', + 'info_dict': { + 'id': '4759-26331132', + 'title': '아이키X노제\r\n💖꽁냥꽁냥💖(1)', + 'ext': 'mp4', + 'thumbnail': r're:^https?://.*\.jpg', + 'timestamp': 1650369600, + 'upload_date': '20220419', + 'uploader_id': 'now', + 'view_count': int, + 'uploader_url': 'https://now.naver.com/show/4759', + 'uploader': '아이키의 떰즈업', + }, + 'params': { + 'noplaylist': True, + } + }, { + 'url': 'https://now.naver.com/s/now.4759?shareHightlight=26601461#highlight=', + 'md5': '9f6118e398aa0f22b2152f554ea7851b', + 'info_dict': { + 'id': '4759-26601461', + 'title': '아이키: 나 리정한테 흔들렸어,,, 질투 폭발하는 노제 여보😾 [아이키의 떰즈업]ㅣ네이버 NOW.', + 'ext': 'mp4', + 'thumbnail': r're:^https?://.*\.jpg', + 'upload_date': '20220504', + 'timestamp': 1651648311, + 'uploader_id': 'now', + 'view_count': int, + 'uploader_url': 'https://now.naver.com/show/4759', + 'uploader': '아이키의 떰즈업', + }, + 'params': { + 'noplaylist': True, + }, + }, { + 'url': 'https://now.naver.com/s/now.4759', + 'info_dict': { + 'id': '4759', + 'title': '아이키의 떰즈업', + }, + 'playlist_mincount': 101 + }, { + 'url': 'https://now.naver.com/s/now.4759?shareReplayId=26331132#replay', + 'info_dict': { + 'id': '4759', + 'title': '아이키의 떰즈업', + }, + 'playlist_mincount': 101, + }, { + 'url': 'https://now.naver.com/s/now.4759?shareHightlight=26601461#highlight=', + 'info_dict': { + 'id': '4759', + 'title': '아이키의 떰즈업', + }, + 'playlist_mincount': 101, + }, { + 'url': 'https://now.naver.com/s/now.kihyunplay?shareReplayId=30573291#replay', + 'only_matching': True, + }] + + def _extract_replay(self, show_id, replay_id): + vod_info = self._download_json(f'{self._API_URL}/shows/now.{show_id}/vod/{replay_id}', replay_id) + in_key = self._download_json(f'{self._API_URL}/shows/now.{show_id}/vod/{replay_id}/inkey', replay_id)['inKey'] + return merge_dicts({ + 'id': f'{show_id}-{replay_id}', + 'title': traverse_obj(vod_info, ('episode', 'title')), + 'timestamp': unified_timestamp(traverse_obj(vod_info, ('episode', 'start_time'))), + 'thumbnail': vod_info.get('thumbnail_image_url'), + }, self._extract_video_info(replay_id, vod_info['video_id'], in_key)) + + def _extract_show_replays(self, show_id): + page_size = 15 + page = 1 + while True: + show_vod_info = self._download_json( + f'{self._API_URL}/vod-shows/now.{show_id}', show_id, + query={'page': page, 'page_size': page_size}, + note=f'Downloading JSON vod list for show {show_id} - page {page}' + )['response']['result'] + for v in show_vod_info.get('vod_list') or []: + yield self._extract_replay(show_id, v['id']) + + if len(show_vod_info.get('vod_list') or []) < page_size: + break + page += 1 + + def _extract_show_highlights(self, show_id, highlight_id=None): + page_size = 10 + page = 1 + while True: + highlights_videos = self._download_json( + f'{self._API_URL}/shows/now.{show_id}/highlights/videos/', show_id, + query={'page': page, 'page_size': page_size}, + note=f'Downloading JSON highlights for show {show_id} - page {page}') + + for highlight in highlights_videos.get('results') or []: + if highlight_id and highlight.get('clip_no') != int(highlight_id): + continue + yield merge_dicts({ + 'id': f'{show_id}-{highlight["clip_no"]}', + 'title': highlight.get('title'), + 'timestamp': unified_timestamp(highlight.get('regdate')), + 'thumbnail': highlight.get('thumbnail_url'), + }, self._extract_video_info(highlight['clip_no'], highlight['video_id'], highlight['video_inkey'])) + + if len(highlights_videos.get('results') or []) < page_size: + break + page += 1 + + def _extract_highlight(self, show_id, highlight_id): + try: + return next(self._extract_show_highlights(show_id, highlight_id)) + except StopIteration: + raise ExtractorError(f'Unable to find highlight {highlight_id} for show {show_id}') + + def _real_extract(self, url): + show_id = self._match_id(url) + qs = parse_qs(urlparse(url).query) + + if not self._yes_playlist(show_id, qs.get('shareHightlight')): + return self._extract_highlight(show_id, qs['shareHightlight'][0]) + elif not self._yes_playlist(show_id, qs.get('shareReplayId')): + return self._extract_replay(show_id, qs['shareReplayId'][0]) + + show_info = self._download_json( + f'{self._API_URL}/shows/now.{show_id}/', show_id, + note=f'Downloading JSON vod list for show {show_id}') + + return self.playlist_result( + itertools.chain(self._extract_show_replays(show_id), self._extract_show_highlights(show_id)), + show_id, show_info.get('title')) |