X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/15b1d10671b48df598afd70e17ba21e9e64ac766..b0b67c374528aca7bc76a9476d554a1d3ab96a9e:/youtube_dl/extractor/firsttv.py diff --git a/youtube_dl/extractor/firsttv.py b/youtube_dl/extractor/firsttv.py index 08ceee4..4803a22 100644 --- a/youtube_dl/extractor/firsttv.py +++ b/youtube_dl/extractor/firsttv.py @@ -1,60 +1,155 @@ -# encoding: utf-8 +# coding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor -from ..utils import int_or_none +from ..compat import ( + compat_str, + compat_urlparse, +) +from ..utils import ( + int_or_none, + qualities, + unified_strdate, +) class FirstTVIE(InfoExtractor): - IE_NAME = 'firsttv' - IE_DESC = 'Видеоархив - Первый канал' - _VALID_URL = r'http://(?:www\.)?1tv\.ru/videoarchive/(?P\d+)' + IE_NAME = '1tv' + IE_DESC = 'Первый канал' + _VALID_URL = r'https?://(?:www\.)?1tv\.ru/(?:[^/]+/)+(?P[^/?#]+)' - _TEST = { - 'url': 'http://www.1tv.ru/videoarchive/73390', - 'md5': '3de6390cf0cca4a5eae1d1d83895e5ad', + _TESTS = [{ + # single format + 'url': 'http://www.1tv.ru/shows/naedine-so-vsemi/vypuski/gost-lyudmila-senchina-naedine-so-vsemi-vypusk-ot-12-02-2015', + 'md5': 'a1b6b60d530ebcf8daacf4565762bbaf', + 'info_dict': { + 'id': '40049', + 'ext': 'mp4', + 'title': 'Гость Людмила Сенчина. Наедине со всеми. Выпуск от 12.02.2015', + 'thumbnail': r're:^https?://.*\.(?:jpg|JPG)$', + 'upload_date': '20150212', + 'duration': 2694, + }, + }, { + # multiple formats + 'url': 'http://www.1tv.ru/shows/dobroe-utro/pro-zdorove/vesennyaya-allergiya-dobroe-utro-fragment-vypuska-ot-07042016', 'info_dict': { - 'id': '73390', + 'id': '364746', 'ext': 'mp4', - 'title': 'Олимпийские канатные дороги', - 'description': 'md5:cc730d2bf4215463e37fff6a1e277b13', - 'thumbnail': 'http://img1.1tv.ru/imgsize640x360/PR20140210114657.JPG', - 'duration': 149, + 'title': 'Весенняя аллергия. Доброе утро. Фрагмент выпуска от 07.04.2016', + 'thumbnail': r're:^https?://.*\.(?:jpg|JPG)$', + 'upload_date': '20160407', + 'duration': 179, + 'formats': 'mincount:3', + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://www.1tv.ru/news/issue/2016-12-01/14:00', + 'info_dict': { + 'id': '14:00', + 'title': 'Выпуск новостей в 14:00 1 декабря 2016 года. Новости. Первый канал', + 'description': 'md5:2e921b948f8c1ff93901da78ebdb1dfd', }, - 'skip': 'Only works from Russia', - } + 'playlist_count': 13, + }, { + 'url': 'http://www.1tv.ru/shows/tochvtoch-supersezon/vystupleniya/evgeniy-dyatlov-vladimir-vysockiy-koni-priveredlivye-toch-v-toch-supersezon-fragment-vypuska-ot-06-11-2016', + 'only_matching': True, + }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + display_id = self._match_id(url) - webpage = self._download_webpage(url, video_id, 'Downloading page') + webpage = self._download_webpage(url, display_id) + playlist_url = compat_urlparse.urljoin(url, self._search_regex( + r'data-playlist-url=(["\'])(?P(?:(?!\1).)+)\1', + webpage, 'playlist url', group='url')) - video_url = self._html_search_regex( - r'''(?s)jwplayer\('flashvideoportal_1'\)\.setup\({.*?'file': '([^']+)'.*?}\);''', webpage, 'video URL') + parsed_url = compat_urlparse.urlparse(playlist_url) + qs = compat_urlparse.parse_qs(parsed_url.query) + item_ids = qs.get('videos_ids[]') or qs.get('news_ids[]') + + items = self._download_json(playlist_url, display_id) + + if item_ids: + items = [ + item for item in items + if item.get('uid') and compat_str(item['uid']) in item_ids] + else: + items = [items[0]] + + entries = [] + QUALITIES = ('ld', 'sd', 'hd', ) + + for item in items: + title = item['title'] + quality = qualities(QUALITIES) + formats = [] + path = None + for f in item.get('mbr', []): + src = f.get('src') + if not src or not isinstance(src, compat_str): + continue + tbr = int_or_none(self._search_regex( + r'_(\d{3,})\.mp4', src, 'tbr', default=None)) + if not path: + path = self._search_regex( + r'//[^/]+/(.+?)_\d+\.mp4', src, + 'm3u8 path', default=None) + formats.append({ + 'url': src, + 'format_id': f.get('name'), + 'tbr': tbr, + 'source_preference': quality(f.get('name')), + # quality metadata of http formats may be incorrect + 'preference': -1, + }) + # m3u8 URL format is reverse engineered from [1] (search for + # master.m3u8). dashEdges (that is currently balancer-vod.1tv.ru) + # is taken from [2]. + # 1. http://static.1tv.ru/player/eump1tv-current/eump-1tv.all.min.js?rnd=9097422834:formatted + # 2. http://static.1tv.ru/player/eump1tv-config/config-main.js?rnd=9097422834 + if not path and len(formats) == 1: + path = self._search_regex( + r'//[^/]+/(.+?$)', formats[0]['url'], + 'm3u8 path', default=None) + if path: + if len(formats) == 1: + m3u8_path = ',' + else: + tbrs = [compat_str(t) for t in sorted(f['tbr'] for f in formats)] + m3u8_path = '_,%s,%s' % (','.join(tbrs), '.mp4') + formats.extend(self._extract_m3u8_formats( + 'http://balancer-vod.1tv.ru/%s%s.urlset/master.m3u8' + % (path, m3u8_path), + display_id, 'mp4', + entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)) + self._sort_formats(formats) + + thumbnail = item.get('poster') or self._og_search_thumbnail(webpage) + duration = int_or_none(item.get('duration') or self._html_search_meta( + 'video:duration', webpage, 'video duration', fatal=False)) + upload_date = unified_strdate(self._html_search_meta( + 'ya:ovs:upload_date', webpage, 'upload date', default=None)) + + entries.append({ + 'id': compat_str(item.get('id') or item['uid']), + 'thumbnail': thumbnail, + 'title': title, + 'upload_date': upload_date, + 'duration': int_or_none(duration), + 'formats': formats + }) title = self._html_search_regex( - r'
\s*

([^<]*)', webpage, 'title') + (r'
\s*

([^<]*)', + r"'title'\s*:\s*'([^']+)'"), + webpage, 'title', default=None) or self._og_search_title( + webpage, default=None) description = self._html_search_regex( - r'
\s*
 
\s*

([^<]*)

', webpage, 'description', fatal=False) - - thumbnail = self._og_search_thumbnail(webpage) - duration = self._og_search_property('video:duration', webpage, 'video duration', fatal=False) - - like_count = self._html_search_regex(r'title="Понравилось".*?/> \[(\d+)\]', - webpage, 'like count', fatal=False) - dislike_count = self._html_search_regex(r'title="Не понравилось".*?/> \[(\d+)\]', - webpage, 'dislike count', fatal=False) - - return { - 'id': video_id, - 'url': video_url, - 'thumbnail': thumbnail, - 'title': title, - 'description': description, - 'duration': int_or_none(duration), - 'like_count': int_or_none(like_count), - 'dislike_count': int_or_none(dislike_count), - } + r'
\s*
 
\s*

([^<]*)

', + webpage, 'description', default=None) or self._html_search_meta( + 'description', webpage, 'description', default=None) + + return self.playlist_result(entries, display_id, title, description)