From 609ff8ca19f1c4c168a81121074b91cc0f0d4c47 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Wed, 5 Jul 2017 23:23:35 +0800 Subject: [PATCH 1/5] [utils] Support attributes with no values in get_elements_by_attribute() --- test/test_utils.py | 6 ++++++ youtube_dl/utils.py | 4 ++-- 2 files changed, 8 insertions(+), 2 deletions(-) diff --git a/test/test_utils.py b/test/test_utils.py index 2b93b3604..30738e7cf 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -1228,6 +1228,12 @@ part 3 self.assertEqual(get_element_by_attribute('class', 'foo', html), None) self.assertEqual(get_element_by_attribute('class', 'no-such-foo', html), None) + html = ''' + + ''' + + self.assertEqual(get_element_by_attribute('itemprop', 'author', html), 'foo') + def test_get_elements_by_class(self): html = ''' nicealso nice diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 39860e9d1..fdf5e29e7 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -365,9 +365,9 @@ def get_elements_by_attribute(attribute, value, html, escape_value=True): retlist = [] for m in re.finditer(r'''(?xs) <([a-zA-Z0-9:._-]+) - (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*? + (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*? \s+%s=['"]?%s['"]? - (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*? + (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*? \s*> (?P.*?) From babbc04d459a6e8b6a2e65c522107cce1ca2cbbb Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Tue, 13 Jun 2017 16:21:26 +0800 Subject: [PATCH 2/5] [xuite] Move to the new HTML5 API and reduce # of requests --- youtube_dl/extractor/xuite.py | 88 ++++++++++++++--------------------- 1 file changed, 35 insertions(+), 53 deletions(-) diff --git a/youtube_dl/extractor/xuite.py b/youtube_dl/extractor/xuite.py index e0818201a..0276c0dbb 100644 --- a/youtube_dl/extractor/xuite.py +++ b/youtube_dl/extractor/xuite.py @@ -1,14 +1,13 @@ # coding: utf-8 from __future__ import unicode_literals -import base64 - from .common import InfoExtractor -from ..compat import compat_urllib_parse_unquote from ..utils import ( ExtractorError, + float_or_none, + get_element_by_attribute, parse_iso8601, - parse_duration, + remove_end, ) @@ -24,6 +23,7 @@ class XuiteIE(InfoExtractor): 'id': '3860914', 'ext': 'mp3', 'title': '孤單南半球-歐德陽', + 'description': '孤單南半球-歐德陽', 'thumbnail': r're:^https?://.*\.jpg$', 'duration': 247.246, 'timestamp': 1314932940, @@ -44,7 +44,7 @@ class XuiteIE(InfoExtractor): 'duration': 596.458, 'timestamp': 1454242500, 'upload_date': '20160131', - 'uploader': 'yan12125', + 'uploader': '屁姥', 'uploader_id': '12158353', 'categories': ['個人短片'], 'description': 'http://download.blender.org/peach/bigbuckbunny_movies/BigBuckBunny_320x180.mp4', @@ -72,10 +72,10 @@ class XuiteIE(InfoExtractor): # from http://forgetfulbc.blogspot.com/2016/06/date.html 'url': 'http://vlog.xuite.net/embed/cE1xbENoLTI3NDQ3MzM2LmZsdg==?ar=0&as=0', 'info_dict': { - 'id': 'cE1xbENoLTI3NDQ3MzM2LmZsdg==', + 'id': '27447336', 'ext': 'mp4', 'title': '男女平權只是口號？專家解釋約會時男生是否該幫女生付錢 (中字)', - 'description': 'md5:f0abdcb69df300f522a5442ef3146f2a', + 'description': 'md5:1223810fa123b179083a3aed53574706', 'timestamp': 1466160960, 'upload_date': '20160617', 'uploader': 'B.C. & Lowy', @@ -86,29 +86,9 @@ class XuiteIE(InfoExtractor): 'only_matching': True, }] - @staticmethod - def base64_decode_utf8(data): - return base64.b64decode(data.encode('utf-8')).decode('utf-8') - - @staticmethod - def base64_encode_utf8(data): - return base64.b64encode(data.encode('utf-8')).decode('utf-8') - - def _extract_flv_config(self, encoded_media_id): - flv_config = self._download_xml( - 'http://vlog.xuite.net/flash/player?media=%s' % encoded_media_id, - 'flv config') - prop_dict = {} - for prop in flv_config.findall('./property'): - prop_id = self.base64_decode_utf8(prop.attrib['id']) - # CDATA may be empty in flv config - if not prop.text: - continue - encoded_content = self.base64_decode_utf8(prop.text) - prop_dict[prop_id] = compat_urllib_parse_unquote(encoded_content) - return prop_dict - def _real_extract(self, url): + # /play/ URLs provide embedded video URL and more metadata + url = url.replace('/embed/', '/play/') video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) @@ -121,51 +101,53 @@ class XuiteIE(InfoExtractor): '%s returned error: %s' % (self.IE_NAME, error_msg), expected=True) - encoded_media_id = self._search_regex( - r'attributes\.name\s*=\s*"([^"]+)"', webpage, - 'encoded media id', default=None) - if encoded_media_id is None: - video_id = self._html_search_regex( - r'data-mediaid="(\d+)"', webpage, 'media id') - encoded_media_id = self.base64_encode_utf8(video_id) - flv_config = self._extract_flv_config(encoded_media_id) + media_info = self._parse_json(self._search_regex( + r'var\s+mediaInfo\s*=\s*({.*});', webpage, 'media info'), video_id) - FORMATS = { - 'audio': 'mp3', - 'video': 'mp4', - } + video_id = media_info['MEDIA_ID'] formats = [] - for format_tag in ('src', 'hq_src'): - video_url = flv_config.get(format_tag) + for key in ('html5Url', 'html5HQUrl'): + video_url = media_info.get(key) if not video_url: continue format_id = self._search_regex( - r'\bq=(.+?)\b', video_url, 'format id', default=format_tag) + r'\bq=(.+?)\b', video_url, 'format id', default=None) formats.append({ 'url': video_url, - 'ext': FORMATS.get(flv_config['type'], 'mp4'), + 'ext': 'mp4' if format_id.isnumeric() else format_id, 'format_id': format_id, 'height': int(format_id) if format_id.isnumeric() else None, }) self._sort_formats(formats) - timestamp = flv_config.get('publish_datetime') + timestamp = media_info.get('PUBLISH_DATETIME') if timestamp: timestamp = parse_iso8601(timestamp + ' +0800', ' ') - category = flv_config.get('category') + category = media_info.get('catName') categories = [category] if category else [] + uploader = media_info.get('NICKNAME') + uploader_url = None + + author_div = get_element_by_attribute('itemprop', 'author', webpage) + if author_div: + uploader = uploader or self._html_search_meta('name', author_div) + uploader_url = self._html_search_regex( + r']+itemprop="url"[^>]+href="([^"]+)"', author_div, + 'uploader URL', fatal=False) + return { 'id': video_id, - 'title': flv_config['title'], - 'description': flv_config.get('description'), - 'thumbnail': flv_config.get('thumb'), + 'title': media_info['TITLE'], + 'description': remove_end(media_info.get('metaDesc'), ' (Xuite 影音)'), + 'thumbnail': media_info.get('ogImageUrl'), 'timestamp': timestamp, - 'uploader': flv_config.get('author_name'), - 'uploader_id': flv_config.get('author_id'), - 'duration': parse_duration(flv_config.get('duration')), + 'uploader': uploader, + 'uploader_id': media_info.get('MEMBER_ID'), + 'uploader_url': uploader_url, + 'duration': float_or_none(media_info.get('MEDIA_DURATION'), 1000000), 'categories': categories, 'formats': formats, } From 0a2e1b2e30045de7834aca880d35253c5e8a3812 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Wed, 5 Jul 2017 22:13:47 +0700 Subject: [PATCH 3/5] [vier] Adapt extraction to redesign (#13575) --- youtube_dl/extractor/vier.py | 45 +++++++++++++++++++++++++++++------- 1 file changed, 37 insertions(+), 8 deletions(-) diff --git a/youtube_dl/extractor/vier.py b/youtube_dl/extractor/vier.py index 3e67eb8c2..dbd5ba9ba 100644 --- a/youtube_dl/extractor/vier.py +++ b/youtube_dl/extractor/vier.py @@ -15,7 +15,21 @@ from ..utils import ( class VierIE(InfoExtractor): IE_NAME = 'vier' IE_DESC = 'vier.be and vijf.be' - _VALID_URL = r'https?://(?:www\.)?(?Pvier|vijf)\.be/(?:[^/]+/videos/(?P[^/]+)(?:/(?P\d+))?|video/v3/embed/(?P\d+))' + _VALID_URL = r'''(?x) + https?:// + (?:www\.)?(?Pvier|vijf)\.be/ + (?: + (?: + [^/]+/videos| + video(?:/[^/]+)* + )/ + (?P[^/]+)(?:/(?P\d+))?| + (?: + video/v3/embed| + embed/video/public + )/(?P\d+) + ) + ''' _NETRC_MACHINE = 'vier' _TESTS = [{ 'url': 'http://www.vier.be/planb/videos/het-wordt-warm-de-moestuin/16129', @@ -83,6 +97,15 @@ class VierIE(InfoExtractor): }, { 'url': 'http://www.vier.be/video/v3/embed/16129', 'only_matching': True, + }, { + 'url': 'https://www.vijf.be/embed/video/public/4093', + 'only_matching': True, + }, { + 'url': 'https://www.vier.be/video/blockbusters/in-juli-en-augustus-summer-classics', + 'only_matching': True, + }, { + 'url': 'https://www.vier.be/video/achter-de-rug/2017/achter-de-rug-seizoen-1-aflevering-6', + 'only_matching': True, }] def _real_initialize(self): @@ -133,14 +156,20 @@ class VierIE(InfoExtractor): video_id = self._search_regex( [r'data-nid="(\d+)"', r'"nid"\s*:\s*"(\d+)"'], webpage, 'video id', default=video_id or display_id) - application = self._search_regex( - [r'data-application="([^"]+)"', r'"application"\s*:\s*"([^"]+)"'], - webpage, 'application', default=site + '_vod') - filename = self._search_regex( - [r'data-filename="([^"]+)"', r'"filename"\s*:\s*"([^"]+)"'], - webpage, 'filename') - playlist_url = 'http://vod.streamcloud.be/%s/_definst_/mp4:%s.mp4/playlist.m3u8' % (application, filename) + playlist_url = self._search_regex( + r'data-file=(["\'])(?P(?:https?:)?//[^/]+/.+?\.m3u8.*?)\1', + webpage, 'm3u8 url', default=None, group='url') + + if not playlist_url: + application = self._search_regex( + [r'data-application="([^"]+)"', r'"application"\s*:\s*"([^"]+)"'], + webpage, 'application', default=site + '_vod') + filename = self._search_regex( + [r'data-filename="([^"]+)"', r'"filename"\s*:\s*"([^"]+)"'], + webpage, 'filename') + playlist_url = 'http://vod.streamcloud.be/%s/_definst_/mp4:%s.mp4/playlist.m3u8' % (application, filename) + formats = self._extract_wowza_formats( playlist_url, display_id, skip_protocols=['dash']) self._sort_formats(formats) From 655470825231eaa03b4b82cbc1314d551e72a01e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Wed, 5 Jul 2017 23:20:50 +0700 Subject: [PATCH 4/5] [kaltura] Fix typo in subtitles extraction (closes #13569) --- youtube_dl/extractor/kaltura.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/youtube_dl/extractor/kaltura.py b/youtube_dl/extractor/kaltura.py index 41c1f3d96..138d4844d 100644 --- a/youtube_dl/extractor/kaltura.py +++ b/youtube_dl/extractor/kaltura.py @@ -324,7 +324,7 @@ class KalturaIE(InfoExtractor): if captions: for caption in captions.get('objects', []): # Continue if caption is not ready - if f.get('status') != 2: + if caption.get('status') != 2: continue if not caption.get('id'): continue From dee2ff1d818bebd74990b7cebbc698f22163a43a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Thu, 6 Jul 2017 00:25:37 +0700 Subject: [PATCH 5/5] [test_utils] Fix tests under Windows --- test/test_utils.py | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/test/test_utils.py b/test/test_utils.py index 30738e7cf..7803e5bc7 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -98,6 +98,7 @@ from youtube_dl.compat import ( compat_chr, compat_etree_fromstring, compat_getenv, + compat_os_name, compat_setenv, compat_urlparse, compat_parse_qs, @@ -448,7 +449,9 @@ class TestUtil(unittest.TestCase): def test_shell_quote(self): args = ['ffmpeg', '-i', encodeFilename('ñ€ß\'.mp4')] - self.assertEqual(shell_quote(args), """ffmpeg -i 'ñ€ß'"'"'.mp4'""") + self.assertEqual( + shell_quote(args), + """ffmpeg -i 'ñ€ß'"'"'.mp4'""" if compat_os_name != 'nt' else '''ffmpeg -i "ñ€ß'.mp4"''') def test_str_to_int(self): self.assertEqual(str_to_int('123,456'), 123456) @@ -932,7 +935,7 @@ class TestUtil(unittest.TestCase): def test_args_to_str(self): self.assertEqual( args_to_str(['foo', 'ba/r', '-baz', '2 be', '']), - 'foo ba/r -baz \'2 be\' \'\'' + 'foo ba/r -baz \'2 be\' \'\'' if compat_os_name != 'nt' else 'foo ba/r -baz "2 be" ""' ) def test_parse_filesize(self):