From 86f63633c8e7c62ce245d1352d4d381efb614466 Mon Sep 17 00:00:00 2001 From: Remita Amine Date: Thu, 17 Oct 2019 13:20:16 +0100 Subject: [PATCH 1/3] [audioboom] improve metadata extraction --- youtube_dl/extractor/audioboom.py | 34 +++++++++++++++++-------------- 1 file changed, 19 insertions(+), 15 deletions(-) diff --git a/youtube_dl/extractor/audioboom.py b/youtube_dl/extractor/audioboom.py index 393f381c6..c51837b40 100644 --- a/youtube_dl/extractor/audioboom.py +++ b/youtube_dl/extractor/audioboom.py @@ -2,22 +2,25 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import float_or_none +from ..utils import ( + clean_html, + float_or_none, +) class AudioBoomIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?audioboom\.com/(?:boos|posts)/(?P[0-9]+)' _TESTS = [{ - 'url': 'https://audioboom.com/boos/4279833-3-09-2016-czaban-hour-3?t=0', - 'md5': '63a8d73a055c6ed0f1e51921a10a5a76', + 'url': 'https://audioboom.com/posts/7398103-asim-chaudhry', + 'md5': '7b00192e593ff227e6a315486979a42d', 'info_dict': { - 'id': '4279833', + 'id': '7398103', 'ext': 'mp3', - 'title': '3/09/2016 Czaban Hour 3', - 'description': 'Guest: Nate Davis - NFL free agency, Guest: Stan Gans', - 'duration': 2245.72, - 'uploader': 'SB Nation A.M.', - 'uploader_url': r're:https?://(?:www\.)?audioboom\.com/channel/steveczabanyahoosportsradio', + 'title': 'Asim Chaudhry', + 'description': 'md5:2f3fef17dacc2595b5362e1d7d3602fc', + 'duration': 4000.99, + 'uploader': 'Sue Perkins: An hour or so with...', + 'uploader_url': r're:https?://(?:www\.)?audioboom\.com/channel/perkins', } }, { 'url': 'https://audioboom.com/posts/4279833-3-09-2016-czaban-hour-3?t=0', @@ -32,8 +35,8 @@ class AudioBoomIE(InfoExtractor): clip = None clip_store = self._parse_json( - self._search_regex( - r'data-new-clip-store=(["\'])(?P{.*?"clipId"\s*:\s*%s.*?})\1' % video_id, + self._html_search_regex( + r'data-new-clip-store=(["\'])(?P{.+?})\1', webpage, 'clip store', default='{}', group='json'), video_id, fatal=False) if clip_store: @@ -47,14 +50,15 @@ class AudioBoomIE(InfoExtractor): audio_url = from_clip('clipURLPriorToLoading') or self._og_search_property( 'audio', webpage, 'audio url') - title = from_clip('title') or self._og_search_title(webpage) - description = from_clip('description') or self._og_search_description(webpage) + title = from_clip('title') or self._html_search_meta( + ['og:title', 'og:audio:title', 'audio_title'], webpage) + description = from_clip('description') or clean_html(from_clip('formattedDescription')) or self._og_search_description(webpage) duration = float_or_none(from_clip('duration') or self._html_search_meta( 'weibo:audio:duration', webpage)) - uploader = from_clip('author') or self._og_search_property( - 'audio:artist', webpage, 'uploader', fatal=False) + uploader = from_clip('author') or self._html_search_meta( + ['og:audio:artist', 'twitter:audio:artist_name', 'audio_artist'], webpage, 'uploader') uploader_url = from_clip('author_url') or self._html_search_meta( 'audioboo:channel', webpage, 'uploader url') From 755541a4c8ac3dd4e8b9abd0c7df95182a1f3fd4 Mon Sep 17 00:00:00 2001 From: Remita Amine Date: Thu, 17 Oct 2019 13:21:44 +0100 Subject: [PATCH 2/3] [mangomolo] fix video format extraction and add support for player URLs --- youtube_dl/extractor/generic.py | 8 ++++++-- youtube_dl/extractor/mangomolo.py | 17 +++++++++++------ 2 files changed, 17 insertions(+), 8 deletions(-) diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index ec43c5ae4..5ed952b29 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -2962,10 +2962,14 @@ class GenericIE(InfoExtractor): # Look for Mangomolo embeds mobj = re.search( - r'''(?x)]+src=(["\'])(?P(?:https?:)?//(?:www\.)?admin\.mangomolo\.com/analytics/index\.php/customers/embed/ + r'''(?x)]+src=(["\'])(?P(?:https?:)?// + (?: + admin\.mangomolo\.com/analytics/index\.php/customers/embed| + player\.mangomolo\.com/v1 + )/ (?: video\?.*?\bid=(?P\d+)| - index\?.*?\bchannelid=(?P(?:[A-Za-z0-9+/=]|%2B|%2F|%3D)+) + (?:index|live)\?.*?\bchannelid=(?P(?:[A-Za-z0-9+/=]|%2B|%2F|%3D)+) ).+?)\1''', webpage) if mobj is not None: info = { diff --git a/youtube_dl/extractor/mangomolo.py b/youtube_dl/extractor/mangomolo.py index 482175a34..acee370e9 100644 --- a/youtube_dl/extractor/mangomolo.py +++ b/youtube_dl/extractor/mangomolo.py @@ -10,18 +10,21 @@ from ..utils import int_or_none class MangomoloBaseIE(InfoExtractor): + _BASE_REGEX = r'https?://(?:admin\.mangomolo\.com/analytics/index\.php/customers/embed/|player\.mangomolo\.com/v1/)' + def _get_real_id(self, page_id): return page_id def _real_extract(self, url): page_id = self._get_real_id(self._match_id(url)) - webpage = self._download_webpage(url, page_id) + webpage = self._download_webpage( + 'https://player.mangomolo.com/v1/%s?%s' % (self._TYPE, url.split('?')[1]), page_id) hidden_inputs = self._hidden_inputs(webpage) m3u8_entry_protocol = 'm3u8' if self._IS_LIVE else 'm3u8_native' format_url = self._html_search_regex( [ - r'file\s*:\s*"(https?://[^"]+?/playlist\.m3u8)', + r'(?:file|src)\s*:\s*"(https?://[^"]+?/playlist\.m3u8)', r']+href="(rtsp://[^"]+)"' ], webpage, 'format url') formats = self._extract_wowza_formats( @@ -39,14 +42,16 @@ class MangomoloBaseIE(InfoExtractor): class MangomoloVideoIE(MangomoloBaseIE): - IE_NAME = 'mangomolo:video' - _VALID_URL = r'https?://admin\.mangomolo\.com/analytics/index\.php/customers/embed/video\?.*?\bid=(?P\d+)' + _TYPE = 'video' + IE_NAME = 'mangomolo:' + _TYPE + _VALID_URL = MangomoloBaseIE._BASE_REGEX + r'video\?.*?\bid=(?P\d+)' _IS_LIVE = False class MangomoloLiveIE(MangomoloBaseIE): - IE_NAME = 'mangomolo:live' - _VALID_URL = r'https?://admin\.mangomolo\.com/analytics/index\.php/customers/embed/index\?.*?\bchannelid=(?P(?:[A-Za-z0-9+/=]|%2B|%2F|%3D)+)' + _TYPE = 'live' + IE_NAME = 'mangomolo:' + _TYPE + _VALID_URL = MangomoloBaseIE._BASE_REGEX + r'(live|index)\?.*?\bchannelid=(?P(?:[A-Za-z0-9+/=]|%2B|%2F|%3D)+)' _IS_LIVE = True def _get_real_id(self, page_id): From 59296bae7ec6d15b0df37dce34bdd96381c0e743 Mon Sep 17 00:00:00 2001 From: Remita Amine Date: Thu, 17 Oct 2019 13:26:45 +0100 Subject: [PATCH 3/3] [xfileshare] clean extractor - update the list of domains - add support for aa-encoded video data - improve jwplayer format extraction - add support for Clappr sources closes #17032 closes #17906 closes #18237 closes #18239 --- youtube_dl/extractor/xfileshare.py | 192 +++++++++++++---------------- 1 file changed, 86 insertions(+), 106 deletions(-) diff --git a/youtube_dl/extractor/xfileshare.py b/youtube_dl/extractor/xfileshare.py index b38c7a7b3..48ef07ed1 100644 --- a/youtube_dl/extractor/xfileshare.py +++ b/youtube_dl/extractor/xfileshare.py @@ -4,37 +4,64 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_chr from ..utils import ( decode_packed_codes, determine_ext, ExtractorError, int_or_none, - NO_DEFAULT, + js_to_json, urlencode_postdata, ) +# based on openload_decode from 2bfeee69b976fe049761dd3012e30b637ee05a58 +def aa_decode(aa_code): + symbol_table = [ + ('7', '((ﾟｰﾟ) + (o^_^o))'), + ('6', '((o^_^o) +(o^_^o))'), + ('5', '((ﾟｰﾟ) + (ﾟΘﾟ))'), + ('2', '((o^_^o) - (ﾟΘﾟ))'), + ('4', '(ﾟｰﾟ)'), + ('3', '(o^_^o)'), + ('1', '(ﾟΘﾟ)'), + ('0', '(c^_^o)'), + ] + delim = '(ﾟДﾟ)[ﾟεﾟ]+' + ret = '' + for aa_char in aa_code.split(delim): + for val, pat in symbol_table: + aa_char = aa_char.replace(pat, val) + aa_char = aa_char.replace('+ ', '') + m = re.match(r'^\d+', aa_char) + if m: + ret += compat_chr(int(m.group(0), 8)) + else: + m = re.match(r'^u([\da-f]+)', aa_char) + if m: + ret += compat_chr(int(m.group(1), 16)) + return ret + + class XFileShareIE(InfoExtractor): _SITES = ( - (r'daclips\.(?:in|com)', 'DaClips'), - (r'filehoot\.com', 'FileHoot'), - (r'gorillavid\.(?:in|com)', 'GorillaVid'), - (r'movpod\.in', 'MovPod'), - (r'powerwatch\.pw', 'PowerWatch'), - (r'rapidvideo\.ws', 'Rapidvideo.ws'), + (r'clipwatching\.com', 'ClipWatching'), + (r'gounlimited\.to', 'GoUnlimited'), + (r'govid\.me', 'GoVid'), + (r'holavid\.com', 'HolaVid'), + (r'streamty\.com', 'Streamty'), (r'thevideobee\.to', 'TheVideoBee'), - (r'vidto\.(?:me|se)', 'Vidto'), - (r'streamin\.to', 'Streamin.To'), - (r'xvidstage\.com', 'XVIDSTAGE'), - (r'vidabc\.com', 'Vid ABC'), + (r'uqload\.com', 'Uqload'), (r'vidbom\.com', 'VidBom'), (r'vidlo\.us', 'vidlo'), - (r'rapidvideo\.(?:cool|org)', 'RapidVideo.TV'), - (r'fastvideo\.me', 'FastVideo.me'), + (r'vidlocker\.xyz', 'VidLocker'), + (r'vidshare\.tv', 'VidShare'), + (r'vup\.to', 'VUp'), + (r'xvideosharing\.com', 'XVideoSharing'), ) IE_DESC = 'XFileShare based sites: %s' % ', '.join(list(zip(*_SITES))[1]) - _VALID_URL = (r'https?://(?P(?:www\.)?(?:%s))/(?:embed-)?(?P[0-9a-zA-Z]+)' + _VALID_URL = (r'https?://(?:www\.)?(?P%s)/(?:embed-)?(?P[0-9a-zA-Z]+)' % '|'.join(site for site in list(zip(*_SITES))[0])) _FILE_NOT_FOUND_REGEXES = ( @@ -43,82 +70,14 @@ class XFileShareIE(InfoExtractor): ) _TESTS = [{ - 'url': 'http://gorillavid.in/06y9juieqpmi', - 'md5': '5ae4a3580620380619678ee4875893ba', + 'url': 'http://xvideosharing.com/fq65f94nd2ve', + 'md5': '4181f63957e8fe90ac836fa58dc3c8a6', 'info_dict': { - 'id': '06y9juieqpmi', + 'id': 'fq65f94nd2ve', 'ext': 'mp4', - 'title': 'Rebecca Black My Moment Official Music Video Reaction-6GK87Rc8bzQ', + 'title': 'sample', 'thumbnail': r're:http://.*\.jpg', }, - }, { - 'url': 'http://gorillavid.in/embed-z08zf8le23c6-960x480.html', - 'only_matching': True, - }, { - 'url': 'http://daclips.in/3rso4kdn6f9m', - 'md5': '1ad8fd39bb976eeb66004d3a4895f106', - 'info_dict': { - 'id': '3rso4kdn6f9m', - 'ext': 'mp4', - 'title': 'Micro Pig piglets ready on 16th July 2009-bG0PdrCdxUc', - 'thumbnail': r're:http://.*\.jpg', - } - }, { - 'url': 'http://movpod.in/0wguyyxi1yca', - 'only_matching': True, - }, { - 'url': 'http://filehoot.com/3ivfabn7573c.html', - 'info_dict': { - 'id': '3ivfabn7573c', - 'ext': 'mp4', - 'title': 'youtube-dl test video \'äBaW_jenozKc.mp4.mp4', - 'thumbnail': r're:http://.*\.jpg', - }, - 'skip': 'Video removed', - }, { - 'url': 'http://vidto.me/ku5glz52nqe1.html', - 'info_dict': { - 'id': 'ku5glz52nqe1', - 'ext': 'mp4', - 'title': 'test' - } - }, { - 'url': 'http://powerwatch.pw/duecjibvicbu', - 'info_dict': { - 'id': 'duecjibvicbu', - 'ext': 'mp4', - 'title': 'Big Buck Bunny trailer', - }, - }, { - 'url': 'http://xvidstage.com/e0qcnl03co6z', - 'info_dict': { - 'id': 'e0qcnl03co6z', - 'ext': 'mp4', - 'title': 'Chucky Prank 2015.mp4', - }, - }, { - # removed by administrator - 'url': 'http://xvidstage.com/amfy7atlkx25', - 'only_matching': True, - }, { - 'url': 'http://vidabc.com/i8ybqscrphfv', - 'info_dict': { - 'id': 'i8ybqscrphfv', - 'ext': 'mp4', - 'title': 're:Beauty and the Beast 2017', - }, - 'params': { - 'skip_download': True, - }, - }, { - 'url': 'http://www.rapidvideo.cool/b667kprndr8w', - 'only_matching': True, - }, { - 'url': 'http://www.fastvideo.me/k8604r8nk8sn/FAST_FURIOUS_8_-_Trailer_italiano_ufficiale.mp4.html', - 'only_matching': True, - }, { - 'url': 'http://vidto.se/1tx1pf6t12cg.html', - 'only_matching': True, }] @staticmethod @@ -131,10 +90,9 @@ class XFileShareIE(InfoExtractor): webpage)] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + host, video_id = re.match(self._VALID_URL, url).groups() - url = 'http://%s/%s' % (mobj.group('host'), video_id) + url = 'https://%s/' % host + ('embed-%s.html' % video_id if host in ('govid.me', 'vidlo.us') else video_id) webpage = self._download_webpage(url, video_id) if any(re.search(p, webpage) for p in self._FILE_NOT_FOUND_REGEXES): @@ -142,7 +100,7 @@ class XFileShareIE(InfoExtractor): fields = self._hidden_inputs(webpage) - if fields['op'] == 'download1': + if fields.get('op') == 'download1': countdown = int_or_none(self._search_regex( r'(?:[Ww]ait)?\s*(\d+)\s*(?:seconds?)?', webpage, 'countdown', default=None)) @@ -160,13 +118,37 @@ class XFileShareIE(InfoExtractor): (r'style="z-index: [0-9]+;">([^<]+)', r'([^<]+)', r'h4-fine[^>]*>([^<]+)<', - r'>Watch (.+) ', + r'>Watch (.+)[ <]', r'

([^<]+)

', - r'

]*>([^<]+)<'), # streamin.to + r'

]*>([^<]+)<', # streamin.to + r'title\s*:\s*"([^"]+)"'), # govid.me webpage, 'title', default=None) or self._og_search_title( webpage, default=None) or video_id).strip() - def extract_formats(default=NO_DEFAULT): + for regex, func in ( + (r'(eval\(function\(p,a,c,k,e,d\){.+)', decode_packed_codes), + (r'(ﾟ.+)', aa_decode)): + obf_code = self._search_regex(regex, webpage, 'obfuscated code', default=None) + if obf_code: + webpage = webpage.replace(obf_code, func(obf_code)) + + formats = [] + + jwplayer_data = self._search_regex( + [ + r'jwplayer\("[^"]+"\)\.load\(\[({.+?})\]\);', + r'jwplayer\("[^"]+"\)\.setup\(({.+?})\);', + ], webpage, + 'jwplayer data', default=None) + if jwplayer_data: + jwplayer_data = self._parse_json( + jwplayer_data.replace(r"\'", "'"), video_id, js_to_json) + if jwplayer_data: + formats = self._parse_jwplayer_data( + jwplayer_data, video_id, False, + m3u8_id='hls', mpd_id='dash')['formats'] + + if not formats: urls = [] for regex in ( r'(?:file|src)\s*:\s*(["\'])(?Phttp(?:(?!\1).)+\.(?:m3u8|mp4|flv)(?:(?!\1).)*)\1', @@ -177,6 +159,12 @@ class XFileShareIE(InfoExtractor): video_url = mobj.group('url') if video_url not in urls: urls.append(video_url) + + sources = self._search_regex( + r'sources\s*:\s*(\[(?!{)[^\]]+\])', webpage, 'sources', default=None) + if sources: + urls.extend(self._parse_json(sources, video_id)) + formats = [] for video_url in urls: if determine_ext(video_url) == 'm3u8': @@ -189,21 +177,13 @@ class XFileShareIE(InfoExtractor): 'url': video_url, 'format_id': 'sd', }) - if not formats and default is not NO_DEFAULT: - return default - self._sort_formats(formats) - return formats - - formats = extract_formats(default=None) - - if not formats: - webpage = decode_packed_codes(self._search_regex( - r"(}\('(.+)',(\d+),(\d+),'[^']*\b(?:file|embed)\b[^']*'\.split\('\|'\))", - webpage, 'packed code')) - formats = extract_formats() + self._sort_formats(formats) thumbnail = self._search_regex( - r'image\s*:\s*["\'](http[^"\']+)["\'],', webpage, 'thumbnail', default=None) + [ + r']+poster="([^"]+)"', + r'(?:image|poster)\s*:\s*["\'](http[^"\']+)["\'],', + ], webpage, 'thumbnail', default=None) return { 'id': video_id,