From 79819f441beadb0dac3936c6b63464dca04c12da Mon Sep 17 00:00:00 2001 From: dirkf Date: Thu, 9 Jun 2022 23:42:06 +0100 Subject: [PATCH] [BitChute] Back-port from yt-dlp and upgrade BitChute * extract timestamp instead of upload date * improve title and description extraction BitChuteChannel * fix listing entries Throughout * relax regexes --- youtube_dl/extractor/bitchute.py | 107 +++++++++++++++++++------------ 1 file changed, 67 insertions(+), 40 deletions(-) diff --git a/youtube_dl/extractor/bitchute.py b/youtube_dl/extractor/bitchute.py index 0c773e66e..459602079 100644 --- a/youtube_dl/extractor/bitchute.py +++ b/youtube_dl/extractor/bitchute.py @@ -6,8 +6,15 @@ import re from .common import InfoExtractor from ..utils import ( + clean_html, + ExtractorError, + GeoRestrictedError, + get_element_by_class, + get_element_by_id, orderedSet, + strip_or_none, unified_strdate, + unified_timestamp, urlencode_postdata, ) @@ -15,16 +22,17 @@ from ..utils import ( class BitChuteIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?bitchute\.com/(?:video|embed|torrent/[^/]+)/(?P[^/?#&]+)' _TESTS = [{ - 'url': 'https://www.bitchute.com/video/szoMrox2JEI/', - 'md5': '66c4a70e6bfc40dcb6be3eb1d74939eb', + 'url': 'https://www.bitchute.com/video/UGlrF9o9b-Q/', + 'md5': '7e427d7ed7af5a75b5855705ec750e2b', 'info_dict': { - 'id': 'szoMrox2JEI', + 'id': 'UGlrF9o9b-Q', 'ext': 'mp4', - 'title': 'Fuck bitches get money', - 'description': 'md5:3f21f6fb5b1d17c3dee9cf6b5fe60b3a', + 'title': 'This is the first video on #BitChute !', + 'timestamp': 1483425420, + 'upload_date': '20170103', + 'description': 'md5:a0337e7b1fe39e32336974af8173a034', 'thumbnail': r're:^https?://.*\.jpg$', - 'uploader': 'Victoria X Rave', - 'upload_date': '20170813', + 'uploader': 'BitChute', }, }, { 'url': 'https://www.bitchute.com/embed/lbb5G1hjPhw/', @@ -34,6 +42,13 @@ class BitChuteIE(InfoExtractor): 'only_matching': True, }] + @staticmethod + def _extract_urls(webpage): + urls = re.finditer( + r'''<(?:script|iframe)\b[^>]+\bsrc\s*=\s*(["'])(?P%s)''' % (BitChuteIE._VALID_URL, ), + webpage) + return (mobj.group('url') for mobj in urls) + def _real_extract(self, url): video_id = self._match_id(url) @@ -42,43 +57,55 @@ class BitChuteIE(InfoExtractor): 'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.57 Safari/537.36', }) - title = self._html_search_regex( - (r'<[^>]+\bid=["\']video-title[^>]+>([^<]+)', r'([^<]+)'), - webpage, 'title', default=None) or self._html_search_meta( - 'description', webpage, 'title', - default=None) or self._og_search_description(webpage) + title = ( + self._og_search_title(webpage, default=None) + or strip_or_none(clean_html(get_element_by_id('video-title', webpage))) + or self._html_search_regex(r'(?s)<title\b[^>]*>.*?</title', webpage, 'title')) - format_urls = [] - for mobj in re.finditer( - r'addWebSeed\s*\(\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage): - format_urls.append(mobj.group('url')) - format_urls.extend(re.findall(r'as=(https?://[^&"\']+)', webpage)) + format_urls = [ + mobj.group('url') + for mobj in re.finditer( + r'''\baddWebSeed\s*\(\s*(["'])(?P<url>(?:(?!\1).)+)\1''', webpage)] + format_urls.extend(re.findall(r'''as=(https?://[^&"']+)''', webpage)) formats = [ {'url': format_url} for format_url in orderedSet(format_urls)] if not formats: - formats = self._parse_html5_media_entries( - url, webpage, video_id)[0]['formats'] + entries = self._parse_html5_media_entries( + url, webpage, video_id) + if not entries: + error = strip_or_none(clean_html(self.get_element_by_id('video-title'))) or 'Cannot find video' + if error == 'Video Unavailable': + raise GeoRestrictedError(error) + raise ExtractorError(error) + formats = entries[0]['formats'] self._check_formats(formats, video_id) self._sort_formats(formats) - description = self._html_search_regex( - r'(?s)<div\b[^>]+\bclass=["\']full hidden[^>]+>(.+?)</div>', - webpage, 'description', fatal=False) - thumbnail = self._og_search_thumbnail( - webpage, default=None) or self._html_search_meta( - 'twitter:image:src', webpage, 'thumbnail') + description = ( + self._og_search_description(webpage) + or clean_html(get_element_by_id('video-description', webpage)) + or self._html_search_regex( + r'(?s)<div\b[^>]+\bclass=["\']full hidden[^>]+>(.+?)</div>', + webpage, 'description', fatal=False)) + thumbnail = self._html_search_meta( + ('og:image', 'twitter:image:src'), webpage, 'thumbnail', fatal=False) uploader = self._html_search_regex( - (r'(?s)<div class=["\']channel-banner.*?<p\b[^>]+\bclass=["\']name[^>]+>(.+?)</p>', - r'(?s)<p\b[^>]+\bclass=["\']video-author[^>]+>(.+?)</p>'), + (r'''(?s)<div\b[^>]+?\bclass\s*=\s*["']channel-banner.*?<p\b[^>]+\bclass\s*=\s*["']name\b[^>]+>(.+?)</p>''', + r'''(?s)<p\b[^>]+\bclass\s*=\s*["']video-author\b[^>]+>(.+?)</p>'''), webpage, 'uploader', fatal=False) - upload_date = unified_strdate(self._search_regex( - r'class=["\']video-publish-date[^>]+>[^<]+ at \d+:\d+ UTC on (.+?)\.', - webpage, 'upload date', fatal=False)) + def more_unified_timestamp(x): + # ... at hh:mm TZ on month nth. + y = re.split(r'\s+at\s+', x or '')[-1] + y = re.sub(r'(?:^\s+|\s+$|\.+$|(?<=\d)(?:st|nd|rd|th))', '', y) + y = ' '.join(reversed(re.split(r'\s+on\s+', y, 1))) + return unified_timestamp(y) or unified_timestamp(x) + + timestamp = more_unified_timestamp(get_element_by_class('video-publish-date', webpage)) return { 'id': video_id, @@ -86,7 +113,7 @@ class BitChuteIE(InfoExtractor): 'description': description, 'thumbnail': thumbnail, 'uploader': uploader, - 'upload_date': upload_date, + 'timestamp': timestamp, 'formats': formats, } @@ -94,22 +121,22 @@ class BitChuteIE(InfoExtractor): class BitChuteChannelIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?bitchute\.com/channel/(?P<id>[^/?#&]+)' _TEST = { - 'url': 'https://www.bitchute.com/channel/victoriaxrave/', - 'playlist_mincount': 185, + 'url': 'https://www.bitchute.com/channel/livesonnet/', + 'playlist_mincount': 135, 'info_dict': { - 'id': 'victoriaxrave', + 'id': 'livesonnet', }, } _TOKEN = 'zyG6tQcGPE5swyAEFLqKUwMuMMuF6IO2DZ6ZDQjGfsL0e4dcTLwqkTTul05Jdve7' def _entries(self, channel_id): - channel_url = 'https://www.bitchute.com/channel/%s/' % channel_id + channel_url = 'https://www.bitchute.com/channel/%s' % (channel_id, ) offset = 0 for page_num in itertools.count(1): data = self._download_json( - '%sextend/' % channel_url, channel_id, - 'Downloading channel page %d' % page_num, + channel_url + '/extend/', channel_id, + 'Downloading channel page %d' % (page_num, ), data=urlencode_postdata({ 'csrfmiddlewaretoken': self._TOKEN, 'name': '', @@ -118,7 +145,7 @@ class BitChuteChannelIE(InfoExtractor): 'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8', 'Referer': channel_url, 'X-Requested-With': 'XMLHttpRequest', - 'Cookie': 'csrftoken=%s' % self._TOKEN, + 'Cookie': 'csrftoken=' + self._TOKEN, }) if data.get('success') is False: break @@ -126,14 +153,14 @@ class BitChuteChannelIE(InfoExtractor): if not html: break video_ids = re.findall( - r'class=["\']channel-videos-image-container[^>]+>\s*<a\b[^>]+\bhref=["\']/video/([^"\'/]+)', + r'''class\s*=\s*["']channel-videos-image-container[^>]+>\s*<a\b[^>]+\bhref\s*=\s*["']/video/([^"'/]+)''', html) if not video_ids: break offset += len(video_ids) for video_id in video_ids: yield self.url_result( - 'https://www.bitchute.com/video/%s' % video_id, + 'https://www.bitchute.com/video/' + video_id, ie=BitChuteIE.ie_key(), video_id=video_id) def _real_extract(self, url):