diff --git a/README.md b/README.md index 70bcfaccf..4ba982907 100644 --- a/README.md +++ b/README.md @@ -667,7 +667,7 @@ The following numeric meta fields can be used with comparisons `<`, `<=`, `>`, ` - `asr`: Audio sampling rate in Hertz - `fps`: Frame rate -Also filtering work for comparisons `=` (equals), `!=` (not equals), `^=` (begins with), `$=` (ends with), `*=` (contains) and following string meta fields: +Also filtering work for comparisons `=` (equals), `^=` (starts with), `$=` (ends with), `*=` (contains) and following string meta fields: - `ext`: File extension - `acodec`: Name of the audio codec in use - `vcodec`: Name of the video codec in use @@ -675,6 +675,8 @@ Also filtering work for comparisons `=` (equals), `!=` (not equals), `^=` (begin - `protocol`: The protocol that will be used for the actual download, lower-case (`http`, `https`, `rtsp`, `rtmp`, `rtmpe`, `mms`, `f4m`, `ism`, `http_dash_segments`, `m3u8`, or `m3u8_native`) - `format_id`: A short description of the format +Any string comparison may be prefixed with negation `!` in order to produce an opposite comparison, e.g. `!*=` (does not contain). + Note that none of the aforementioned meta fields are guaranteed to be present since this solely depends on the metadata obtained by particular extractor, i.e. the metadata offered by the video hoster. Formats for which the value is not known are excluded unless you put a question mark (`?`) after the operator. You can combine format filters, so `-f "[height <=? 720][tbr>500]"` selects up to 720p videos (or videos where the height is not known) with a bitrate of at least 500 KBit/s. diff --git a/test/test_InfoExtractor.py b/test/test_InfoExtractor.py index 06be72616..75fa0bbb7 100644 --- a/test/test_InfoExtractor.py +++ b/test/test_InfoExtractor.py @@ -497,7 +497,64 @@ jwplayer("mediaplayer").setup({"abouttext":"Visit Indie DB","aboutlink":"http:\/ 'width': 1280, 'height': 720, }] - ) + ), + ( + # https://github.com/rg3/youtube-dl/issues/18923 + # https://www.ted.com/talks/boris_hesser_a_grassroots_healthcare_revolution_in_africa + 'ted_18923', + 'http://hls.ted.com/talks/31241.m3u8', + [{ + 'url': 'http://hls.ted.com/videos/BorisHesser_2018S/audio/600k.m3u8?nobumpers=true&uniqueId=76011e2b', + 'format_id': '600k-Audio', + 'vcodec': 'none', + }, { + 'url': 'http://hls.ted.com/videos/BorisHesser_2018S/audio/600k.m3u8?nobumpers=true&uniqueId=76011e2b', + 'format_id': '68', + 'vcodec': 'none', + }, { + 'url': 'http://hls.ted.com/videos/BorisHesser_2018S/video/64k.m3u8?nobumpers=true&uniqueId=76011e2b', + 'format_id': '163', + 'acodec': 'none', + 'width': 320, + 'height': 180, + }, { + 'url': 'http://hls.ted.com/videos/BorisHesser_2018S/video/180k.m3u8?nobumpers=true&uniqueId=76011e2b', + 'format_id': '481', + 'acodec': 'none', + 'width': 512, + 'height': 288, + }, { + 'url': 'http://hls.ted.com/videos/BorisHesser_2018S/video/320k.m3u8?nobumpers=true&uniqueId=76011e2b', + 'format_id': '769', + 'acodec': 'none', + 'width': 512, + 'height': 288, + }, { + 'url': 'http://hls.ted.com/videos/BorisHesser_2018S/video/450k.m3u8?nobumpers=true&uniqueId=76011e2b', + 'format_id': '984', + 'acodec': 'none', + 'width': 512, + 'height': 288, + }, { + 'url': 'http://hls.ted.com/videos/BorisHesser_2018S/video/600k.m3u8?nobumpers=true&uniqueId=76011e2b', + 'format_id': '1255', + 'acodec': 'none', + 'width': 640, + 'height': 360, + }, { + 'url': 'http://hls.ted.com/videos/BorisHesser_2018S/video/950k.m3u8?nobumpers=true&uniqueId=76011e2b', + 'format_id': '1693', + 'acodec': 'none', + 'width': 853, + 'height': 480, + }, { + 'url': 'http://hls.ted.com/videos/BorisHesser_2018S/video/1500k.m3u8?nobumpers=true&uniqueId=76011e2b', + 'format_id': '2462', + 'acodec': 'none', + 'width': 1280, + 'height': 720, + }] + ), ] for m3u8_file, m3u8_url, expected_formats in _TEST_CASES: diff --git a/test/test_YoutubeDL.py b/test/test_YoutubeDL.py index f0f5a8470..df8994b84 100644 --- a/test/test_YoutubeDL.py +++ b/test/test_YoutubeDL.py @@ -239,6 +239,52 @@ class TestFormatSelection(unittest.TestCase): downloaded = ydl.downloaded_info_dicts[0] self.assertEqual(downloaded['format_id'], 'vid-vcodec-dot') + def test_format_selection_string_ops(self): + formats = [ + {'format_id': 'abc-cba', 'ext': 'mp4', 'url': TEST_URL}, + ] + info_dict = _make_result(formats) + + # equals (=) + ydl = YDL({'format': '[format_id=abc-cba]'}) + ydl.process_ie_result(info_dict.copy()) + downloaded = ydl.downloaded_info_dicts[0] + self.assertEqual(downloaded['format_id'], 'abc-cba') + + # does not equal (!=) + ydl = YDL({'format': '[format_id!=abc-cba]'}) + self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy()) + + # starts with (^=) + ydl = YDL({'format': '[format_id^=abc]'}) + ydl.process_ie_result(info_dict.copy()) + downloaded = ydl.downloaded_info_dicts[0] + self.assertEqual(downloaded['format_id'], 'abc-cba') + + # does not start with (!^=) + ydl = YDL({'format': '[format_id!^=abc-cba]'}) + self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy()) + + # ends with ($=) + ydl = YDL({'format': '[format_id$=cba]'}) + ydl.process_ie_result(info_dict.copy()) + downloaded = ydl.downloaded_info_dicts[0] + self.assertEqual(downloaded['format_id'], 'abc-cba') + + # does not end with (!$=) + ydl = YDL({'format': '[format_id!$=abc-cba]'}) + self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy()) + + # contains (*=) + ydl = YDL({'format': '[format_id*=-]'}) + ydl.process_ie_result(info_dict.copy()) + downloaded = ydl.downloaded_info_dicts[0] + self.assertEqual(downloaded['format_id'], 'abc-cba') + + # does not contain (!*=) + ydl = YDL({'format': '[format_id!*=-]'}) + self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy()) + def test_youtube_format_selection(self): order = [ '38', '37', '46', '22', '45', '35', '44', '18', '34', '43', '6', '5', '17', '36', '13', diff --git a/test/test_utils.py b/test/test_utils.py index 9e28e008f..409482c3b 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -507,6 +507,8 @@ class TestUtil(unittest.TestCase): self.assertEqual(urljoin('http://foo.de/', ''), None) self.assertEqual(urljoin('http://foo.de/', ['foobar']), None) self.assertEqual(urljoin('http://foo.de/a/b/c.txt', '.././../d.txt'), 'http://foo.de/d.txt') + self.assertEqual(urljoin('http://foo.de/a/b/c.txt', 'rtmp://foo.de'), 'rtmp://foo.de') + self.assertEqual(urljoin(None, 'rtmp://foo.de'), 'rtmp://foo.de') def test_url_or_none(self): self.assertEqual(url_or_none(None), None) diff --git a/test/testdata/m3u8/ted_18923.m3u8 b/test/testdata/m3u8/ted_18923.m3u8 new file mode 100644 index 000000000..52a27118b --- /dev/null +++ b/test/testdata/m3u8/ted_18923.m3u8 @@ -0,0 +1,28 @@ +#EXTM3U +#EXT-X-VERSION:4 +#EXT-X-STREAM-INF:AUDIO="600k",BANDWIDTH=1255659,PROGRAM-ID=1,CODECS="avc1.42c01e,mp4a.40.2",RESOLUTION=640x360 +/videos/BorisHesser_2018S/video/600k.m3u8?nobumpers=true&uniqueId=76011e2b +#EXT-X-STREAM-INF:AUDIO="600k",BANDWIDTH=163154,PROGRAM-ID=1,CODECS="avc1.42c00c,mp4a.40.2",RESOLUTION=320x180 +/videos/BorisHesser_2018S/video/64k.m3u8?nobumpers=true&uniqueId=76011e2b +#EXT-X-STREAM-INF:AUDIO="600k",BANDWIDTH=481701,PROGRAM-ID=1,CODECS="avc1.42c015,mp4a.40.2",RESOLUTION=512x288 +/videos/BorisHesser_2018S/video/180k.m3u8?nobumpers=true&uniqueId=76011e2b +#EXT-X-STREAM-INF:AUDIO="600k",BANDWIDTH=769968,PROGRAM-ID=1,CODECS="avc1.42c015,mp4a.40.2",RESOLUTION=512x288 +/videos/BorisHesser_2018S/video/320k.m3u8?nobumpers=true&uniqueId=76011e2b +#EXT-X-STREAM-INF:AUDIO="600k",BANDWIDTH=984037,PROGRAM-ID=1,CODECS="avc1.42c015,mp4a.40.2",RESOLUTION=512x288 +/videos/BorisHesser_2018S/video/450k.m3u8?nobumpers=true&uniqueId=76011e2b +#EXT-X-STREAM-INF:AUDIO="600k",BANDWIDTH=1693925,PROGRAM-ID=1,CODECS="avc1.4d401f,mp4a.40.2",RESOLUTION=853x480 +/videos/BorisHesser_2018S/video/950k.m3u8?nobumpers=true&uniqueId=76011e2b +#EXT-X-STREAM-INF:AUDIO="600k",BANDWIDTH=2462469,PROGRAM-ID=1,CODECS="avc1.640028,mp4a.40.2",RESOLUTION=1280x720 +/videos/BorisHesser_2018S/video/1500k.m3u8?nobumpers=true&uniqueId=76011e2b +#EXT-X-STREAM-INF:AUDIO="600k",BANDWIDTH=68101,PROGRAM-ID=1,CODECS="mp4a.40.2",DEFAULT=YES +/videos/BorisHesser_2018S/audio/600k.m3u8?nobumpers=true&uniqueId=76011e2b + +#EXT-X-I-FRAME-STREAM-INF:BANDWIDTH=74298,PROGRAM-ID=1,CODECS="avc1.42c00c",RESOLUTION=320x180,URI="/videos/BorisHesser_2018S/video/64k_iframe.m3u8?nobumpers=true&uniqueId=76011e2b" +#EXT-X-I-FRAME-STREAM-INF:BANDWIDTH=216200,PROGRAM-ID=1,CODECS="avc1.42c015",RESOLUTION=512x288,URI="/videos/BorisHesser_2018S/video/180k_iframe.m3u8?nobumpers=true&uniqueId=76011e2b" +#EXT-X-I-FRAME-STREAM-INF:BANDWIDTH=304717,PROGRAM-ID=1,CODECS="avc1.42c015",RESOLUTION=512x288,URI="/videos/BorisHesser_2018S/video/320k_iframe.m3u8?nobumpers=true&uniqueId=76011e2b" +#EXT-X-I-FRAME-STREAM-INF:BANDWIDTH=350933,PROGRAM-ID=1,CODECS="avc1.42c015",RESOLUTION=512x288,URI="/videos/BorisHesser_2018S/video/450k_iframe.m3u8?nobumpers=true&uniqueId=76011e2b" +#EXT-X-I-FRAME-STREAM-INF:BANDWIDTH=495850,PROGRAM-ID=1,CODECS="avc1.42c01e",RESOLUTION=640x360,URI="/videos/BorisHesser_2018S/video/600k_iframe.m3u8?nobumpers=true&uniqueId=76011e2b" +#EXT-X-I-FRAME-STREAM-INF:BANDWIDTH=810750,PROGRAM-ID=1,CODECS="avc1.4d401f",RESOLUTION=853x480,URI="/videos/BorisHesser_2018S/video/950k_iframe.m3u8?nobumpers=true&uniqueId=76011e2b" +#EXT-X-I-FRAME-STREAM-INF:BANDWIDTH=1273700,PROGRAM-ID=1,CODECS="avc1.640028",RESOLUTION=1280x720,URI="/videos/BorisHesser_2018S/video/1500k_iframe.m3u8?nobumpers=true&uniqueId=76011e2b" + +#EXT-X-MEDIA:TYPE=AUDIO,GROUP-ID="600k",LANGUAGE="en",NAME="Audio",AUTOSELECT=YES,DEFAULT=YES,URI="/videos/BorisHesser_2018S/audio/600k.m3u8?nobumpers=true&uniqueId=76011e2b",BANDWIDTH=614400 diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index 4493fd0e1..a827414dc 100755 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -1063,21 +1063,24 @@ class YoutubeDL(object): if not m: STR_OPERATORS = { '=': operator.eq, - '!=': operator.ne, '^=': lambda attr, value: attr.startswith(value), '$=': lambda attr, value: attr.endswith(value), '*=': lambda attr, value: value in attr, } str_operator_rex = re.compile(r'''(?x) \s*(?Pext|acodec|vcodec|container|protocol|format_id) - \s*(?P%s)(?P\s*\?)? + \s*(?P!\s*)?(?P%s)(?P\s*\?)? \s*(?P[a-zA-Z0-9._-]+) \s*$ ''' % '|'.join(map(re.escape, STR_OPERATORS.keys()))) m = str_operator_rex.search(filter_spec) if m: comparison_value = m.group('value') - op = STR_OPERATORS[m.group('op')] + str_op = STR_OPERATORS[m.group('op')] + if m.group('negation'): + op = lambda attr, value: not str_op + else: + op = str_op if not m: raise ValueError('Invalid filter specification %r' % filter_spec) diff --git a/youtube_dl/extractor/cartoonnetwork.py b/youtube_dl/extractor/cartoonnetwork.py index 6aeebd7b3..48b33617f 100644 --- a/youtube_dl/extractor/cartoonnetwork.py +++ b/youtube_dl/extractor/cartoonnetwork.py @@ -1,20 +1,19 @@ # coding: utf-8 from __future__ import unicode_literals -import re - from .turner import TurnerBaseIE +from ..utils import int_or_none class CartoonNetworkIE(TurnerBaseIE): _VALID_URL = r'https?://(?:www\.)?cartoonnetwork\.com/video/(?:[^/]+/)+(?P[^/?#]+)-(?:clip|episode)\.html' _TEST = { - 'url': 'http://www.cartoonnetwork.com/video/teen-titans-go/starfire-the-cat-lady-clip.html', + 'url': 'https://www.cartoonnetwork.com/video/ben-10/how-to-draw-upgrade-episode.html', 'info_dict': { - 'id': '8a250ab04ed07e6c014ef3f1e2f9016c', + 'id': '6e3375097f63874ebccec7ef677c1c3845fa850e', 'ext': 'mp4', - 'title': 'Starfire the Cat Lady', - 'description': 'Robin decides to become a cat so that Starfire will finally love him.', + 'title': 'How to Draw Upgrade', + 'description': 'md5:2061d83776db7e8be4879684eefe8c0f', }, 'params': { # m3u8 download @@ -25,18 +24,39 @@ class CartoonNetworkIE(TurnerBaseIE): def _real_extract(self, url): display_id = self._match_id(url) webpage = self._download_webpage(url, display_id) - id_type, video_id = re.search(r"_cnglobal\.cvp(Video|Title)Id\s*=\s*'([^']+)';", webpage).groups() - query = ('id' if id_type == 'Video' else 'titleId') + '=' + video_id - return self._extract_cvp_info( - 'http://www.cartoonnetwork.com/video-seo-svc/episodeservices/getCvpPlaylist?networkName=CN2&' + query, video_id, { - 'secure': { - 'media_src': 'http://androidhls-secure.cdn.turner.com/toon/big', - 'tokenizer_src': 'https://token.vgtf.net/token/token_mobile', - }, - }, { + + def find_field(global_re, name, content_re=None, value_re='[^"]+', fatal=False): + metadata_re = '' + if content_re: + metadata_re = r'|video_metadata\.content_' + content_re + return self._search_regex( + r'(?:_cnglobal\.currentVideo\.%s%s)\s*=\s*"(%s)";' % (global_re, metadata_re, value_re), + webpage, name, fatal=fatal) + + media_id = find_field('mediaId', 'media id', 'id', '[0-9a-f]{40}', True) + title = find_field('episodeTitle', 'title', '(?:episodeName|name)', fatal=True) + + info = self._extract_ngtv_info( + media_id, {'networkId': 'cartoonnetwork'}, { 'url': url, 'site_name': 'CartoonNetwork', - 'auth_required': self._search_regex( - r'_cnglobal\.cvpFullOrPreviewAuth\s*=\s*(true|false);', - webpage, 'auth required', default='false') == 'true', + 'auth_required': find_field('authType', 'auth type') != 'unauth', }) + + series = find_field( + 'propertyName', 'series', 'showName') or self._html_search_meta('partOfSeries', webpage) + info.update({ + 'id': media_id, + 'display_id': display_id, + 'title': title, + 'description': self._html_search_meta('description', webpage), + 'series': series, + 'episode': title, + }) + + for field in ('season', 'episode'): + field_name = field + 'Number' + info[field + '_number'] = int_or_none(find_field( + field_name, field + ' number', value_re=r'\d+') or self._html_search_meta(field_name, webpage)) + + return info diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 9e7febcad..95456b291 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -1596,6 +1596,7 @@ class InfoExtractor(object): # References: # 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-21 # 2. https://github.com/rg3/youtube-dl/issues/12211 + # 3. https://github.com/rg3/youtube-dl/issues/18923 # We should try extracting formats only from master playlists [1, 4.3.4], # i.e. playlists that describe available qualities. On the other hand @@ -1667,11 +1668,16 @@ class InfoExtractor(object): rendition = stream_group[0] return rendition.get('NAME') or stream_group_id + # parse EXT-X-MEDIA tags before EXT-X-STREAM-INF in order to have the + # chance to detect video only formats when EXT-X-STREAM-INF tags + # precede EXT-X-MEDIA tags in HLS manifest such as [3]. + for line in m3u8_doc.splitlines(): + if line.startswith('#EXT-X-MEDIA:'): + extract_media(line) + for line in m3u8_doc.splitlines(): if line.startswith('#EXT-X-STREAM-INF:'): last_stream_inf = parse_m3u8_attributes(line) - elif line.startswith('#EXT-X-MEDIA:'): - extract_media(line) elif line.startswith('#') or not line.strip(): continue else: @@ -2624,7 +2630,7 @@ class InfoExtractor(object): 'id': this_video_id, 'title': unescapeHTML(video_data['title'] if require_title else video_data.get('title')), 'description': video_data.get('description'), - 'thumbnail': self._proto_relative_url(video_data.get('image')), + 'thumbnail': urljoin(base_url, self._proto_relative_url(video_data.get('image'))), 'timestamp': int_or_none(video_data.get('pubdate')), 'duration': float_or_none(jwplayer_data.get('duration') or video_data.get('duration')), 'subtitles': subtitles, @@ -2651,12 +2657,9 @@ class InfoExtractor(object): for source in jwplayer_sources_data: if not isinstance(source, dict): continue - source_url = self._proto_relative_url(source.get('file')) - if not source_url: - continue - if base_url: - source_url = compat_urlparse.urljoin(base_url, source_url) - if source_url in urls: + source_url = urljoin( + base_url, self._proto_relative_url(source.get('file'))) + if not source_url or source_url in urls: continue urls.append(source_url) source_type = source.get('type') or '' diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index de38c6641..574a47e6d 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -452,6 +452,7 @@ from .hellporno import HellPornoIE from .helsinki import HelsinkiIE from .hentaistigma import HentaiStigmaIE from .hgtv import HGTVComShowIE +from .hketv import HKETVIE from .hidive import HiDiveIE from .historicfilms import HistoricFilmsIE from .hitbox import HitboxIE, HitboxLiveIE @@ -494,7 +495,11 @@ from .ina import InaIE from .inc import IncIE from .indavideo import IndavideoEmbedIE from .infoq import InfoQIE -from .instagram import InstagramIE, InstagramUserIE +from .instagram import ( + InstagramIE, + InstagramUserIE, + InstagramTagIE, +) from .internazionale import InternazionaleIE from .internetvideoarchive import InternetVideoArchiveIE from .iprima import IPrimaIE diff --git a/youtube_dl/extractor/hketv.py b/youtube_dl/extractor/hketv.py new file mode 100644 index 000000000..b57927fc1 --- /dev/null +++ b/youtube_dl/extractor/hketv.py @@ -0,0 +1,191 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import compat_str +from ..utils import ( + clean_html, + ExtractorError, + int_or_none, + merge_dicts, + parse_count, + str_or_none, + try_get, + unified_strdate, + urlencode_postdata, + urljoin, +) + + +class HKETVIE(InfoExtractor): + IE_NAME = 'hketv' + IE_DESC = '香港教育局教育電視 (HKETV) Educational Television, Hong Kong Educational Bureau' + _GEO_BYPASS = False + _GEO_COUNTRIES = ['HK'] + _VALID_URL = r'https?://(?:www\.)?hkedcity\.net/etv/resource/(?P[0-9]+)' + _TESTS = [{ + 'url': 'https://www.hkedcity.net/etv/resource/2932360618', + 'md5': 'f193712f5f7abb208ddef3c5ea6ed0b7', + 'info_dict': { + 'id': '2932360618', + 'ext': 'mp4', + 'title': '喜閱一生(共享閱讀樂) (中、英文字幕可供選擇)', + 'description': 'md5:d5286d05219ef50e0613311cbe96e560', + 'upload_date': '20181024', + 'duration': 900, + 'subtitles': 'count:2', + }, + 'skip': 'Geo restricted to HK', + }, { + 'url': 'https://www.hkedcity.net/etv/resource/972641418', + 'md5': '1ed494c1c6cf7866a8290edad9b07dc9', + 'info_dict': { + 'id': '972641418', + 'ext': 'mp4', + 'title': '衣冠楚楚 (天使系列之一)', + 'description': 'md5:10bb3d659421e74f58e5db5691627b0f', + 'upload_date': '20070109', + 'duration': 907, + 'subtitles': {}, + }, + 'params': { + 'geo_verification_proxy': '', + }, + 'skip': 'Geo restricted to HK', + }] + + _CC_LANGS = { + '中文（繁體中文）': 'zh-Hant', + '中文（简体中文）': 'zh-Hans', + 'English': 'en', + 'Bahasa Indonesia': 'id', + '\u0939\u093f\u0928\u094d\u0926\u0940': 'hi', + '\u0928\u0947\u092a\u093e\u0932\u0940': 'ne', + 'Tagalog': 'tl', + '\u0e44\u0e17\u0e22': 'th', + '\u0627\u0631\u062f\u0648': 'ur', + } + _FORMAT_HEIGHTS = { + 'SD': 360, + 'HD': 720, + } + _APPS_BASE_URL = 'https://apps.hkedcity.net' + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + title = ( + self._html_search_meta( + ('ed_title', 'search.ed_title'), webpage, default=None) or + self._search_regex( + r'data-favorite_title_(?:eng|chi)=(["\'])(?P(?:(?!\1).)+)\1', + webpage, 'title', default=None, group='url') or + self._html_search_regex( + r'

([^<]+)

', webpage, 'title', default=None) or + self._og_search_title(webpage) + ) + + file_id = self._search_regex( + r'post_var\[["\']file_id["\']\s*\]\s*=\s*(.+?);', + webpage, 'file ID') + curr_url = self._search_regex( + r'post_var\[["\']curr_url["\']\s*\]\s*=\s*"(.+?)";', + webpage, 'curr URL') + data = { + 'action': 'get_info', + 'curr_url': curr_url, + 'file_id': file_id, + 'video_url': file_id, + } + + response = self._download_json( + self._APPS_BASE_URL + '/media/play/handler.php', video_id, + data=urlencode_postdata(data), + headers=merge_dicts({ + 'Content-Type': 'application/x-www-form-urlencoded'}, + self.geo_verification_headers())) + + result = response['result'] + + if not response.get('success') or not response.get('access'): + error = clean_html(response.get('access_err_msg')) + if 'Video streaming is not available in your country' in error: + self.raise_geo_restricted( + msg=error, countries=self._GEO_COUNTRIES) + else: + raise ExtractorError(error, expected=True) + + formats = [] + + width = int_or_none(result.get('width')) + height = int_or_none(result.get('height')) + + playlist0 = result['playlist'][0] + for fmt in playlist0['sources']: + file_url = urljoin(self._APPS_BASE_URL, fmt.get('file')) + if not file_url: + continue + # If we ever wanted to provide the final resolved URL that + # does not require cookies, albeit with a shorter lifespan: + # urlh = self._downloader.urlopen(file_url) + # resolved_url = urlh.geturl() + label = fmt.get('label') + h = self._FORMAT_HEIGHTS.get(label) + w = h * width // height if h and width and height else None + formats.append({ + 'format_id': label, + 'ext': fmt.get('type'), + 'url': file_url, + 'width': w, + 'height': h, + }) + self._sort_formats(formats) + + subtitles = {} + tracks = try_get(playlist0, lambda x: x['tracks'], list) or [] + for track in tracks: + if not isinstance(track, dict): + continue + track_kind = str_or_none(track.get('kind')) + if not track_kind or not isinstance(track_kind, compat_str): + continue + if track_kind.lower() not in ('captions', 'subtitles'): + continue + track_url = urljoin(self._APPS_BASE_URL, track.get('file')) + if not track_url: + continue + track_label = track.get('label') + subtitles.setdefault(self._CC_LANGS.get( + track_label, track_label), []).append({ + 'url': self._proto_relative_url(track_url), + 'ext': 'srt', + }) + + # Likes + emotion = self._download_json( + 'https://emocounter.hkedcity.net/handler.php', video_id, + data=urlencode_postdata({ + 'action': 'get_emotion', + 'data[bucket_id]': 'etv', + 'data[identifier]': video_id, + }), + headers={'Content-Type': 'application/x-www-form-urlencoded'}, + fatal=False) or {} + like_count = int_or_none(try_get( + emotion, lambda x: x['data']['emotion_data'][0]['count'])) + + return { + 'id': video_id, + 'title': title, + 'description': self._html_search_meta( + 'description', webpage, fatal=False), + 'upload_date': unified_strdate(self._html_search_meta( + 'ed_date', webpage, fatal=False), day_first=False), + 'duration': int_or_none(result.get('length')), + 'formats': formats, + 'subtitles': subtitles, + 'thumbnail': urljoin(self._APPS_BASE_URL, result.get('image')), + 'view_count': parse_count(result.get('view_count')), + 'like_count': like_count, + } diff --git a/youtube_dl/extractor/instagram.py b/youtube_dl/extractor/instagram.py index 7e0e838f0..ffd87b55f 100644 --- a/youtube_dl/extractor/instagram.py +++ b/youtube_dl/extractor/instagram.py @@ -227,44 +227,37 @@ class InstagramIE(InfoExtractor): } -class InstagramUserIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?instagram\.com/(?P[^/]{2,})/?(?:$|[?#])' - IE_DESC = 'Instagram user profile' - IE_NAME = 'instagram:user' - _TEST = { - 'url': 'https://instagram.com/porsche', - 'info_dict': { - 'id': 'porsche', - 'title': 'porsche', - }, - 'playlist_count': 5, - 'params': { - 'extract_flat': True, - 'skip_download': True, - 'playlistend': 5, - } - } +class InstagramPlaylistIE(InfoExtractor): + # A superclass for handling any kind of query based on GraphQL which + # results in a playlist. - _gis_tmpl = None + _gis_tmpl = None # used to cache GIS request type - def _entries(self, data): + def _parse_graphql(self, webpage, item_id): + # Reads a webpage and returns its GraphQL data. + return self._parse_json( + self._search_regex( + r'sharedData\s*=\s*({.+?})\s*;\s*[<\n]', webpage, 'data'), + item_id) + + def _extract_graphql(self, data, url): + # Parses GraphQL queries containing videos and generates a playlist. def get_count(suffix): return int_or_none(try_get( node, lambda x: x['edge_media_' + suffix]['count'])) - uploader_id = data['entry_data']['ProfilePage'][0]['graphql']['user']['id'] + uploader_id = self._match_id(url) csrf_token = data['config']['csrf_token'] rhx_gis = data.get('rhx_gis') or '3c7ca9dcefcf966d11dacf1f151335e8' - self._set_cookie('instagram.com', 'ig_pr', '1') - cursor = '' for page_num in itertools.count(1): - variables = json.dumps({ - 'id': uploader_id, + variables = { 'first': 12, 'after': cursor, - }) + } + variables.update(self._query_vars_for(data)) + variables = json.dumps(variables) if self._gis_tmpl: gis_tmpls = [self._gis_tmpl] @@ -276,21 +269,26 @@ class InstagramUserIE(InfoExtractor): '%s:%s:%s' % (rhx_gis, csrf_token, std_headers['User-Agent']), ] + # try all of the ways to generate a GIS query, and not only use the + # first one that works, but cache it for future requests for gis_tmpl in gis_tmpls: try: - media = self._download_json( + json_data = self._download_json( 'https://www.instagram.com/graphql/query/', uploader_id, 'Downloading JSON page %d' % page_num, headers={ 'X-Requested-With': 'XMLHttpRequest', 'X-Instagram-GIS': hashlib.md5( ('%s:%s' % (gis_tmpl, variables)).encode('utf-8')).hexdigest(), }, query={ - 'query_hash': '42323d64886122307be10013ad2dcc44', + 'query_hash': self._QUERY_HASH, 'variables': variables, - })['data']['user']['edge_owner_to_timeline_media'] + }) + media = self._parse_timeline_from(json_data) self._gis_tmpl = gis_tmpl break except ExtractorError as e: + # if it's an error caused by a bad query, and there are + # more GIS templates to try, ignore it and keep trying if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403: if gis_tmpl != gis_tmpls[-1]: continue @@ -348,14 +346,80 @@ class InstagramUserIE(InfoExtractor): break def _real_extract(self, url): - username = self._match_id(url) + user_or_tag = self._match_id(url) + webpage = self._download_webpage(url, user_or_tag) + data = self._parse_graphql(webpage, user_or_tag) - webpage = self._download_webpage(url, username) - - data = self._parse_json( - self._search_regex( - r'sharedData\s*=\s*({.+?})\s*;\s*[<\n]', webpage, 'data'), - username) + self._set_cookie('instagram.com', 'ig_pr', '1') return self.playlist_result( - self._entries(data), username, username) + self._extract_graphql(data, url), user_or_tag, user_or_tag) + + +class InstagramUserIE(InstagramPlaylistIE): + _VALID_URL = r'https?://(?:www\.)?instagram\.com/(?P[^/]{2,})/?(?:$|[?#])' + IE_DESC = 'Instagram user profile' + IE_NAME = 'instagram:user' + _TEST = { + 'url': 'https://instagram.com/porsche', + 'info_dict': { + 'id': 'porsche', + 'title': 'porsche', + }, + 'playlist_count': 5, + 'params': { + 'extract_flat': True, + 'skip_download': True, + 'playlistend': 5, + } + } + + _QUERY_HASH = '42323d64886122307be10013ad2dcc44', + + @staticmethod + def _parse_timeline_from(data): + # extracts the media timeline data from a GraphQL result + return data['data']['user']['edge_owner_to_timeline_media'] + + @staticmethod + def _query_vars_for(data): + # returns a dictionary of variables to add to the timeline query based + # on the GraphQL of the original page + return { + 'id': data['entry_data']['ProfilePage'][0]['graphql']['user']['id'] + } + + +class InstagramTagIE(InstagramPlaylistIE): + _VALID_URL = r'https?://(?:www\.)?instagram\.com/explore/tags/(?P[^/]+)' + IE_DESC = 'Instagram hashtag search' + IE_NAME = 'instagram:tag' + _TEST = { + 'url': 'https://instagram.com/explore/tags/lolcats', + 'info_dict': { + 'id': 'lolcats', + 'title': 'lolcats', + }, + 'playlist_count': 50, + 'params': { + 'extract_flat': True, + 'skip_download': True, + 'playlistend': 50, + } + } + + _QUERY_HASH = 'f92f56d47dc7a55b606908374b43a314', + + @staticmethod + def _parse_timeline_from(data): + # extracts the media timeline data from a GraphQL result + return data['data']['hashtag']['edge_hashtag_to_media'] + + @staticmethod + def _query_vars_for(data): + # returns a dictionary of variables to add to the timeline query based + # on the GraphQL of the original page + return { + 'tag_name': + data['entry_data']['TagPage'][0]['graphql']['hashtag']['name'] + } diff --git a/youtube_dl/extractor/radiocanada.py b/youtube_dl/extractor/radiocanada.py index b952e59b4..302f67d96 100644 --- a/youtube_dl/extractor/radiocanada.py +++ b/youtube_dl/extractor/radiocanada.py @@ -49,6 +49,16 @@ class RadioCanadaIE(InfoExtractor): # m3u8 download 'skip_download': True, }, + }, + { + # with protectionType but not actually DRM protected + 'url': 'radiocanada:toutv:140872', + 'info_dict': { + 'id': '140872', + 'title': 'Épisode 1', + 'series': 'District 31', + }, + 'only_matching': True, } ] @@ -67,8 +77,10 @@ class RadioCanadaIE(InfoExtractor): el = find_xpath_attr(metadata, './/Meta', 'name', name) return el.text if el is not None else None + # protectionType does not necessarily mean the video is DRM protected (see + # https://github.com/rg3/youtube-dl/pull/18609). if get_meta('protectionType'): - raise ExtractorError('This video is DRM protected.', expected=True) + self.report_warning('This video is probably DRM protected.') device_types = ['ipad'] if not smuggled_data: diff --git a/youtube_dl/extractor/streamango.py b/youtube_dl/extractor/streamango.py index fcaa5ac0b..efb259f96 100644 --- a/youtube_dl/extractor/streamango.py +++ b/youtube_dl/extractor/streamango.py @@ -14,7 +14,7 @@ from ..utils import ( class StreamangoIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?streamango\.com/(?:f|embed)/(?P[^/?#&]+)' + _VALID_URL = r'https?://(?:www\.)?(?:streamango\.com|fruithosts\.net)/(?:f|embed)/(?P[^/?#&]+)' _TESTS = [{ 'url': 'https://streamango.com/f/clapasobsptpkdfe/20170315_150006_mp4', 'md5': 'e992787515a182f55e38fc97588d802a', @@ -38,6 +38,9 @@ class StreamangoIE(InfoExtractor): }, { 'url': 'https://streamango.com/embed/clapasobsptpkdfe/20170315_150006_mp4', 'only_matching': True, + }, { + 'url': 'https://fruithosts.net/f/mreodparcdcmspsm/w1f1_r4lph_2018_brrs_720p_latino_mp4', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/ted.py b/youtube_dl/extractor/ted.py index d3e4205f5..645942dfd 100644 --- a/youtube_dl/extractor/ted.py +++ b/youtube_dl/extractor/ted.py @@ -265,6 +265,8 @@ class TEDIE(InfoExtractor): 'format_id': m3u8_format['format_id'].replace('hls', 'http'), 'protocol': 'http', }) + if f.get('acodec') == 'none': + del f['acodec'] formats.append(f) audio_download = talk_info.get('audioDownload') diff --git a/youtube_dl/extractor/tnaflix.py b/youtube_dl/extractor/tnaflix.py index 6798ef4c3..b3573c6e0 100644 --- a/youtube_dl/extractor/tnaflix.py +++ b/youtube_dl/extractor/tnaflix.py @@ -96,7 +96,7 @@ class TNAFlixNetworkBaseIE(InfoExtractor): cfg_xml = self._download_xml( cfg_url, display_id, 'Downloading metadata', - transform_source=fix_xml_ampersands) + transform_source=fix_xml_ampersands, headers={'Referer': url}) formats = [] diff --git a/youtube_dl/extractor/videomore.py b/youtube_dl/extractor/videomore.py index 9b56630de..e3eda3327 100644 --- a/youtube_dl/extractor/videomore.py +++ b/youtube_dl/extractor/videomore.py @@ -4,8 +4,14 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_str from ..utils import ( int_or_none, + orderedSet, + parse_duration, + str_or_none, + unified_strdate, + url_or_none, xpath_element, xpath_text, ) @@ -13,7 +19,19 @@ from ..utils import ( class VideomoreIE(InfoExtractor): IE_NAME = 'videomore' - _VALID_URL = r'videomore:(?P\d+)$|https?://videomore\.ru/(?:(?:embed|[^/]+/[^/]+)/|[^/]+\?.*\btrack_id=)(?P\d+)(?:[/?#&]|\.(?:xml|json)|$)' + _VALID_URL = r'''(?x) + videomore:(?P\d+)$| + https?://(?:player\.)?videomore\.ru/ + (?: + (?: + embed| + [^/]+/[^/]+ + )/| + [^/]*\?.*?\btrack_id= + ) + (?P\d+) + (?:[/?#&]|\.(?:xml|json)|$) + ''' _TESTS = [{ 'url': 'http://videomore.ru/kino_v_detalayah/5_sezon/367617', 'md5': '44455a346edc0d509ac5b5a5b531dc35', @@ -79,6 +97,9 @@ class VideomoreIE(InfoExtractor): }, { 'url': 'videomore:367617', 'only_matching': True, + }, { + 'url': 'https://player.videomore.ru/?partner_id=97&track_id=736234&autoplay=0&userToken=', + 'only_matching': True, }] @staticmethod @@ -136,7 +157,7 @@ class VideomoreIE(InfoExtractor): class VideomoreVideoIE(InfoExtractor): IE_NAME = 'videomore:video' - _VALID_URL = r'https?://videomore\.ru/(?:(?:[^/]+/){2})?(?P[^/?#&]+)[/?#&]*$' + _VALID_URL = r'https?://videomore\.ru/(?:(?:[^/]+/){2})?(?P[^/?#&]+)(?:/*|[?#&].*?)$' _TESTS = [{ # single video with og:video:iframe 'url': 'http://videomore.ru/elki_3', @@ -176,6 +197,9 @@ class VideomoreVideoIE(InfoExtractor): 'params': { 'skip_download': True, }, + }, { + 'url': 'https://videomore.ru/molodezhka/6_sezon/29_seriya?utm_so', + 'only_matching': True, }] @classmethod @@ -196,13 +220,16 @@ class VideomoreVideoIE(InfoExtractor): r'track-id=["\'](\d+)', r'xcnt_product_id\s*=\s*(\d+)'), webpage, 'video id') video_url = 'videomore:%s' % video_id + else: + video_id = None - return self.url_result(video_url, VideomoreIE.ie_key()) + return self.url_result( + video_url, ie=VideomoreIE.ie_key(), video_id=video_id) class VideomoreSeasonIE(InfoExtractor): IE_NAME = 'videomore:season' - _VALID_URL = r'https?://videomore\.ru/(?!embed)(?P[^/]+/[^/?#&]+)[/?#&]*$' + _VALID_URL = r'https?://videomore\.ru/(?!embed)(?P[^/]+/[^/?#&]+)(?:/*|[?#&].*?)$' _TESTS = [{ 'url': 'http://videomore.ru/molodezhka/sezon_promo', 'info_dict': { @@ -210,8 +237,16 @@ class VideomoreSeasonIE(InfoExtractor): 'title': 'Молодежка Промо', }, 'playlist_mincount': 12, + }, { + 'url': 'http://videomore.ru/molodezhka/sezon_promo?utm_so', + 'only_matching': True, }] + @classmethod + def suitable(cls, url): + return (False if (VideomoreIE.suitable(url) or VideomoreVideoIE.suitable(url)) + else super(VideomoreSeasonIE, cls).suitable(url)) + def _real_extract(self, url): display_id = self._match_id(url) @@ -219,9 +254,54 @@ class VideomoreSeasonIE(InfoExtractor): title = self._og_search_title(webpage) - entries = [ - self.url_result(item) for item in re.findall( - r']+href="((?:https?:)?//videomore\.ru/%s/[^/]+)"[^>]+class="widget-item-desc"' - % display_id, webpage)] + data = self._parse_json( + self._html_search_regex( + r'\bclass=["\']seasons-tracks["\'][^>]+\bdata-custom-data=(["\'])(?P{.+?})\1', + webpage, 'data', default='{}', group='value'), + display_id, fatal=False) + + entries = [] + + if data: + episodes = data.get('episodes') + if isinstance(episodes, list): + for ep in episodes: + if not isinstance(ep, dict): + continue + ep_id = int_or_none(ep.get('id')) + ep_url = url_or_none(ep.get('url')) + if ep_id: + e = { + 'url': 'videomore:%s' % ep_id, + 'id': compat_str(ep_id), + } + elif ep_url: + e = {'url': ep_url} + else: + continue + e.update({ + '_type': 'url', + 'ie_key': VideomoreIE.ie_key(), + 'title': str_or_none(ep.get('title')), + 'thumbnail': url_or_none(ep.get('image')), + 'duration': parse_duration(ep.get('duration')), + 'episode_number': int_or_none(ep.get('number')), + 'upload_date': unified_strdate(ep.get('date')), + }) + entries.append(e) + + if not entries: + entries = [ + self.url_result( + 'videomore:%s' % video_id, ie=VideomoreIE.ie_key(), + video_id=video_id) + for video_id in orderedSet(re.findall( + r':(?:id|key)=["\'](\d+)["\']', webpage))] + + if not entries: + entries = [ + self.url_result(item) for item in re.findall( + r']+href="((?:https?:)?//videomore\.ru/%s/[^/]+)"[^>]+class="widget-item-desc"' + % display_id, webpage)] return self.playlist_result(entries, display_id, title) diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index 3415ca597..bbbdd706c 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -436,6 +436,8 @@ class VimeoIE(VimeoBaseInfoExtractor): 'url': 'https://vimeo.com/160743502/abd0e13fb4', 'only_matching': True, } + # https://gettingthingsdone.com/workflowmap/ + # vimeo embed with check-password page protected by Referer header ] @staticmethod @@ -466,20 +468,22 @@ class VimeoIE(VimeoBaseInfoExtractor): urls = VimeoIE._extract_urls(url, webpage) return urls[0] if urls else None - def _verify_player_video_password(self, url, video_id): + def _verify_player_video_password(self, url, video_id, headers): password = self._downloader.params.get('videopassword') if password is None: raise ExtractorError('This video is protected by a password, use the --video-password option') data = urlencode_postdata({ 'password': base64.b64encode(password.encode()), }) - pass_url = url + '/check-password' - password_request = sanitized_Request(pass_url, data) - password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - password_request.add_header('Referer', url) - return self._download_json( - password_request, video_id, - 'Verifying the password', 'Wrong password') + headers = merge_dicts(headers, { + 'Content-Type': 'application/x-www-form-urlencoded', + }) + checked = self._download_json( + url + '/check-password', video_id, + 'Verifying the password', data=data, headers=headers) + if checked is False: + raise ExtractorError('Wrong video password', expected=True) + return checked def _real_initialize(self): self._login() @@ -592,7 +596,7 @@ class VimeoIE(VimeoBaseInfoExtractor): cause=e) else: if config.get('view') == 4: - config = self._verify_player_video_password(redirect_url, video_id) + config = self._verify_player_video_password(redirect_url, video_id, headers) vod = config.get('video', {}).get('vod', {}) diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index d2d3c1a9f..d0cb65814 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -1868,7 +1868,7 @@ def urljoin(base, path): path = path.decode('utf-8') if not isinstance(path, compat_str) or not path: return None - if re.match(r'^(?:https?:)?//', path): + if re.match(r'^(?:[a-zA-Z][a-zA-Z0-9+-.]*:)?//', path): return path if isinstance(base, bytes): base = base.decode('utf-8')