From 469d4c89686afca46333d85442bb770e6010518c Mon Sep 17 00:00:00 2001 From: Will Sewell Date: Mon, 17 Nov 2014 17:52:00 -0500 Subject: [PATCH 01/11] [vk] Added a new information extractor for pages that are a list of a user\'s videos on vk.com. It works in a same way to playlist style pages for the YT information extractors. --- youtube_dl/extractor/__init__.py | 5 ++++- youtube_dl/extractor/vk.py | 37 +++++++++++++++++++++++++++++++- 2 files changed, 40 insertions(+), 2 deletions(-) diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index f45ce05ab..b687a56b4 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -452,7 +452,10 @@ VineUserIE, ) from .viki import VikiIE -from .vk import VKIE +from .vk import ( + VKIE, + VKUserVideosIE, +) from .vodlocker import VodlockerIE from .vporn import VpornIE from .vrt import VRTIE diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index 36cd7e52e..5223e5e2c 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -16,7 +16,7 @@ class VKIE(InfoExtractor): IE_NAME = 'vk.com' - _VALID_URL = r'https?://(?:m\.)?vk\.com/(?:video_ext\.php\?.*?\boid=(?P-?\d+).*?\bid=(?P\d+)|(?:.+?\?.*?z=)?video(?P.*?)(?:\?|%2F|$))' + _VALID_URL = r'https?://(?:m\.)?vk\.com/(?:video_ext\.php\?.*?\boid=(?P-?\d+).*?\bid=(?P\d+)|(?:.+?\?.*?z=)?video(?P[^s].*?)(?:\?|%2F|$))' _NETRC_MACHINE = 'vk' _TESTS = [ @@ -185,3 +185,38 @@ def _real_extract(self, url): 'uploader': data.get('md_author'), 'duration': data.get('duration') } + + +class VKUserVideosIE(InfoExtractor): + IE_NAME = 'vk.com:user-videos' + IE_DESC = 'All of a user\'s videos' + _VALID_URL = r'https?://(?:m\.)?vk\.com/videos([0-9]+)' + _TEMPLATE_URL = 'https://vk.com/videos' + _TEST = { + 'url': 'http://vk.com/videos205387401', + 'playlist_mincount': 4, + } + + def extract_videos_from_page(self, page): + ids_in_page = [] + for mobj in re.finditer(r'href="/video([0-9_]+)"', page): + if mobj.group(1) not in ids_in_page: + ids_in_page.append(mobj.group(1)) + return ids_in_page + + def _real_extract(self, url): + # Extract page id + mobj = re.match(self._VALID_URL, url) + if mobj is None: + raise ExtractorError('Invalid URL: %s' % url) + + # Download page and get video ids + page_id = mobj.group(1) + page = self._download_webpage(url, page_id) + video_ids = self.extract_videos_from_page(page) + + self._downloader.to_screen('[vk] User videos %s: Found %i videos' % (page_id, len(video_ids))) + + url_entries = [self.url_result('http://vk.com/video' + video_id, 'VK', video_id=video_id) + for video_id in video_ids] + return self.playlist_result(url_entries, page_id) \ No newline at end of file From 6fcd6e0e21839ae4b1753995dc44d2a93f72ac1f Mon Sep 17 00:00:00 2001 From: Will Sewell Date: Tue, 18 Nov 2014 19:34:12 +0000 Subject: [PATCH 02/11] [vk] Updated the regex for matching user video pages. It now matches optional URL parameters too. --- youtube_dl/extractor/vk.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index 5223e5e2c..3bcf50e28 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -190,7 +190,7 @@ def _real_extract(self, url): class VKUserVideosIE(InfoExtractor): IE_NAME = 'vk.com:user-videos' IE_DESC = 'All of a user\'s videos' - _VALID_URL = r'https?://(?:m\.)?vk\.com/videos([0-9]+)' + _VALID_URL = r'https?://(?:m\.)?vk\.com/videos([0-9]+)(?:m\?.*)?' _TEMPLATE_URL = 'https://vk.com/videos' _TEST = { 'url': 'http://vk.com/videos205387401', From 02a12f9fe69508525c9cad06782151f5cf950671 Mon Sep 17 00:00:00 2001 From: Will Sewell Date: Tue, 18 Nov 2014 20:19:56 +0000 Subject: [PATCH 03/11] [vk] date_added is now extracted from the video page. --- youtube_dl/extractor/vk.py | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index 3bcf50e28..deaad6c3e 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -11,7 +11,7 @@ compat_urllib_parse, compat_str, unescapeHTML, -) + unified_strdate) class VKIE(InfoExtractor): @@ -169,6 +169,12 @@ def _real_extract(self, url): data_json = self._search_regex(r'var vars = ({.*?});', info_page, 'vars') data = json.loads(data_json) + # Extract upload date + upload_date = None + mobj = re.search(r'id="mv_date_wrap".*?Added ([a-zA-Z]+ [0-9]+), ([0-9]+) at', info_page) + if mobj is not None: + upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2)) + formats = [{ 'format_id': k, 'url': v, @@ -183,7 +189,8 @@ def _real_extract(self, url): 'title': unescapeHTML(data['md_title']), 'thumbnail': data.get('jpg'), 'uploader': data.get('md_author'), - 'duration': data.get('duration') + 'duration': data.get('duration'), + 'upload_date': upload_date, } From 42e1ff8665ceb6eddb6b2067f57239be3a8ab209 Mon Sep 17 00:00:00 2001 From: Will Sewell Date: Fri, 21 Nov 2014 22:23:39 +0000 Subject: [PATCH 04/11] [vk.com] Added upload_date variable to the test cases that still work. --- youtube_dl/extractor/vk.py | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index deaad6c3e..de87bee1f 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -29,6 +29,7 @@ class VKIE(InfoExtractor): 'title': 'ProtivoGunz - Хуёвая песня', 'uploader': 're:Noize MC.*', 'duration': 195, + 'upload_date': '20120212', }, }, { @@ -52,6 +53,7 @@ class VKIE(InfoExtractor): 'uploader': 'Vladimir Gavrin', 'title': 'Lin Dan', 'duration': 101, + 'upload_date': '20120730', } }, { @@ -88,6 +90,7 @@ class VKIE(InfoExtractor): 'uploader': 'Киномания - лучшее из мира кино', 'title': ' ', 'duration': 7291, + 'upload_date': '20140328', }, 'skip': 'Requires vk account credentials', }, @@ -100,6 +103,7 @@ class VKIE(InfoExtractor): 'ext': 'mp4', 'title': 'Книга Илая', 'duration': 6771, + 'upload_date': '20140626', }, 'skip': 'Only works from Russia', }, From c52331f30c15ff715431cf1ca5fceec505efe599 Mon Sep 17 00:00:00 2001 From: Will Sewell Date: Fri, 21 Nov 2014 22:52:01 +0000 Subject: [PATCH 05/11] [vk.com] Updated a test video that has been removed, and added a comment for others to update two other test videos that are also now removed. --- youtube_dl/extractor/vk.py | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index de87bee1f..d6632cbb7 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -33,14 +33,15 @@ class VKIE(InfoExtractor): }, }, { - 'url': 'http://vk.com/video4643923_163339118', - 'md5': 'f79bccb5cd182b1f43502ca5685b2b36', + 'url': 'http://vk.com/video205387401_165548505', + 'md5': '6c0aeb2e90396ba97035b9cbde548700', 'info_dict': { - 'id': '163339118', + 'id': '165548505', 'ext': 'mp4', - 'uploader': 'Elya Iskhakova', - 'title': 'Dream Theater - Hollow Years Live at Budokan 720*', - 'duration': 558, + 'uploader': 'Tom Cruise', + 'title': 'No name', + 'duration': 9, + 'upload_date': '20130721' } }, { @@ -57,6 +58,8 @@ class VKIE(InfoExtractor): } }, { + # VIDEO NOW REMOVED + # please update if you find a video whose URL follows the same pattern 'url': 'http://vk.com/video-8871596_164049491', 'md5': 'a590bcaf3d543576c9bd162812387666', 'note': 'Only available for registered users', @@ -66,10 +69,13 @@ class VKIE(InfoExtractor): 'uploader': 'Триллеры', 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]', 'duration': 8352, + 'upload_date': '20121218' }, 'skip': 'Requires vk account credentials', }, { + # VIDEO NOW REMOVED + # please update if you find a video whose URL follows the same pattern 'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a', 'md5': 'd82c22e449f036282d1d3f7f4d276869', 'info_dict': { @@ -78,6 +84,7 @@ class VKIE(InfoExtractor): 'uploader': 'Киномания - лучшее из мира кино', 'title': 'Запах женщины (1992)', 'duration': 9392, + 'upload_date': '20130914' }, 'skip': 'Requires vk account credentials', }, @@ -177,6 +184,7 @@ def _real_extract(self, url): upload_date = None mobj = re.search(r'id="mv_date_wrap".*?Added ([a-zA-Z]+ [0-9]+), ([0-9]+) at', info_page) if mobj is not None: + x = mobj.group(1) + ' ' + mobj.group(2) upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2)) formats = [{ From cad985ab4d78ef00a77932e87cee5aba8a1e3320 Mon Sep 17 00:00:00 2001 From: Will Sewell Date: Fri, 21 Nov 2014 22:54:11 +0000 Subject: [PATCH 06/11] [vk.com] Updated the description to include vk.com. --- youtube_dl/extractor/vk.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index d6632cbb7..d96abd16e 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -208,7 +208,7 @@ def _real_extract(self, url): class VKUserVideosIE(InfoExtractor): IE_NAME = 'vk.com:user-videos' - IE_DESC = 'All of a user\'s videos' + IE_DESC = 'vk.com:All of a user\'s videos' _VALID_URL = r'https?://(?:m\.)?vk\.com/videos([0-9]+)(?:m\?.*)?' _TEMPLATE_URL = 'https://vk.com/videos' _TEST = { From 53d1cd1f779201af426548fc77e1724b6c70abd9 Mon Sep 17 00:00:00 2001 From: Will Sewell Date: Fri, 21 Nov 2014 23:03:31 +0000 Subject: [PATCH 07/11] [vk.com] Updated the _VALID_URL regex for the playlist IE. Removed optional m, and named the id group. --- youtube_dl/extractor/vk.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index d96abd16e..7136a0cb2 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -209,7 +209,7 @@ def _real_extract(self, url): class VKUserVideosIE(InfoExtractor): IE_NAME = 'vk.com:user-videos' IE_DESC = 'vk.com:All of a user\'s videos' - _VALID_URL = r'https?://(?:m\.)?vk\.com/videos([0-9]+)(?:m\?.*)?' + _VALID_URL = r'https?://vk\.com/videos(?P[0-9]+)(?:m\?.*)?' _TEMPLATE_URL = 'https://vk.com/videos' _TEST = { 'url': 'http://vk.com/videos205387401', From e1e8b6897b2e2610c45eb53fe44e1e07c3c39e82 Mon Sep 17 00:00:00 2001 From: Will Sewell Date: Fri, 21 Nov 2014 23:16:12 +0000 Subject: [PATCH 08/11] [vk.com] Updated the extract_videos_from_page function with a much simpler 1-liner. --- youtube_dl/extractor/vk.py | 12 +++--------- 1 file changed, 3 insertions(+), 9 deletions(-) diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index 7136a0cb2..e8d7cdeae 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -11,7 +11,8 @@ compat_urllib_parse, compat_str, unescapeHTML, - unified_strdate) + unified_strdate, + orderedSet) class VKIE(InfoExtractor): @@ -216,13 +217,6 @@ class VKUserVideosIE(InfoExtractor): 'playlist_mincount': 4, } - def extract_videos_from_page(self, page): - ids_in_page = [] - for mobj in re.finditer(r'href="/video([0-9_]+)"', page): - if mobj.group(1) not in ids_in_page: - ids_in_page.append(mobj.group(1)) - return ids_in_page - def _real_extract(self, url): # Extract page id mobj = re.match(self._VALID_URL, url) @@ -232,7 +226,7 @@ def _real_extract(self, url): # Download page and get video ids page_id = mobj.group(1) page = self._download_webpage(url, page_id) - video_ids = self.extract_videos_from_page(page) + video_ids = orderedSet(m.group(1) for m in re.finditer(r'href="/video([0-9_]+)"', page)) self._downloader.to_screen('[vk] User videos %s: Found %i videos' % (page_id, len(video_ids))) From 021a0db8f787521757919fde5a2303205312abff Mon Sep 17 00:00:00 2001 From: Will Sewell Date: Fri, 21 Nov 2014 23:21:03 +0000 Subject: [PATCH 09/11] [vk.com] Simplified the page_id acquisition by using the id matched in the URL earlier on. --- youtube_dl/extractor/vk.py | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index e8d7cdeae..b4e261fad 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -218,13 +218,7 @@ class VKUserVideosIE(InfoExtractor): } def _real_extract(self, url): - # Extract page id - mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError('Invalid URL: %s' % url) - - # Download page and get video ids - page_id = mobj.group(1) + page_id = self._match_id(url) page = self._download_webpage(url, page_id) video_ids = orderedSet(m.group(1) for m in re.finditer(r'href="/video([0-9_]+)"', page)) From b9272e8f8f785fee4d77258bdb1cf1d85f947ecf Mon Sep 17 00:00:00 2001 From: Will Sewell Date: Fri, 21 Nov 2014 23:22:52 +0000 Subject: [PATCH 10/11] [vk.com] Removed redundant log message -- this information is already being logged. --- youtube_dl/extractor/vk.py | 3 --- 1 file changed, 3 deletions(-) diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index b4e261fad..248b7d54d 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -221,9 +221,6 @@ def _real_extract(self, url): page_id = self._match_id(url) page = self._download_webpage(url, page_id) video_ids = orderedSet(m.group(1) for m in re.finditer(r'href="/video([0-9_]+)"', page)) - - self._downloader.to_screen('[vk] User videos %s: Found %i videos' % (page_id, len(video_ids))) - url_entries = [self.url_result('http://vk.com/video' + video_id, 'VK', video_id=video_id) for video_id in video_ids] return self.playlist_result(url_entries, page_id) \ No newline at end of file From 9262867e86cb415ed1994f9f956536eba39fc457 Mon Sep 17 00:00:00 2001 From: Will Sewell Date: Fri, 21 Nov 2014 23:25:05 +0000 Subject: [PATCH 11/11] [vk.com] Added newline at the end of the file. --- youtube_dl/extractor/vk.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index 248b7d54d..daf615af9 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -223,4 +223,4 @@ def _real_extract(self, url): video_ids = orderedSet(m.group(1) for m in re.finditer(r'href="/video([0-9_]+)"', page)) url_entries = [self.url_result('http://vk.com/video' + video_id, 'VK', video_id=video_id) for video_id in video_ids] - return self.playlist_result(url_entries, page_id) \ No newline at end of file + return self.playlist_result(url_entries, page_id)