From: Philipp Hagemeister Date: Sat, 22 Mar 2014 14:15:01 +0000 (+0100) Subject: [toypics] Separate user and video extraction (#2601) X-Git-Url: http://git.oshgnacknak.de/?a=commitdiff_plain;h=231f76b530d214e64762c8129861c0eff7d9494f;p=youtube-dl [toypics] Separate user and video extraction (#2601) --- diff --git a/test/test_playlists.py b/test/test_playlists.py index fbeed1c8c..4c9c34057 100644 --- a/test/test_playlists.py +++ b/test/test_playlists.py @@ -37,6 +37,7 @@ from youtube_dl.extractor import ( GoogleSearchIE, GenericIE, TEDIE, + ToypicsUserIE, ) @@ -269,5 +270,13 @@ class TestPlaylists(unittest.TestCase): self.assertEqual(result['title'], 'Who are the hackers?') self.assertTrue(len(result['entries']) >= 6) + def test_toypics_user(self): + dl = FakeYDL() + ie = ToypicsUserIE(dl) + result = ie.extract('http://videos.toypics.net/Mikey') + self.assertIsPlaylist(result) + self.assertEqual(result['id'], 'Mikey') + self.assertTrue(len(result['entries']) >= 17) + if __name__ == '__main__': unittest.main() diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 5ca6eb16c..b8c843515 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -239,7 +239,7 @@ from .theplatform import ThePlatformIE from .thisav import ThisAVIE from .tinypic import TinyPicIE from .toutv import TouTvIE -from .toypics import ToypicsIE +from .toypics import ToypicsUserIE, ToypicsIE from .traileraddict import TrailerAddictIE from .trilulilu import TriluliluIE from .trutube import TruTubeIE diff --git a/youtube_dl/extractor/toypics.py b/youtube_dl/extractor/toypics.py index 3cbfe2e7e..33a6988ca 100644 --- a/youtube_dl/extractor/toypics.py +++ b/youtube_dl/extractor/toypics.py @@ -2,43 +2,26 @@ from .common import InfoExtractor from math import ceil import re + class ToypicsIE(InfoExtractor): - _VALID_URL = r'(?:http://)?videos\.toypics\.net/.*' + IE_DESC = 'Toypics user profile' + _VALID_URL = r'http://videos\.toypics\.net/view/(?P[0-9]+)/.*' _TEST = { 'url': 'http://videos.toypics.net/view/514/chancebulged,-2-1/', - #'md5': '8a8b546956bbd0e769dbe28f6e80abb3', == $head -c10K 12929646011616163504.mp4 |md5sum //no idea why it fails + 'md5': '16e806ad6d6f58079d210fe30985e08b', 'info_dict': { 'id': '514', 'ext': 'mp4', 'title': 'Chance-Bulge\'d, 2', - 'age_limit': 18 + 'age_limit': 18, + 'uploader': 'kidsune', } } - PAGINATED=8 def _real_extract(self, url): - mobj = re.match(r'(http://)?videos\.toypics\.net/(?P[^/?]+)$', url) - if not mobj: - return self.extract_one(url) - return [self.extract_one(u) for u in self.process_paginated(url, - r'public/">Public Videos \((?P[0-9]+)\)', - r'

\n\s*' - )] - - def process_paginated(self, profile_url, re_total, re_video_page): - profile_page = self._download_webpage(profile_url, 'profile' , 'getting profile page: '+profile_url) - videos_count = self._html_search_regex(re_total, profile_page, 'videos count') - lst = [] - for n in xrange(1,int(ceil(float(videos_count)/self.PAGINATED)) +1): - lpage_url = profile_url +'/public/%d'%n - lpage = self._download_webpage(lpage_url, 'page %d'%n) - lst.extend(re.findall(re_video_page, lpage)) - return lst - - def extract_one(self,url): - mobj = re.match(r'(http://)?videos\.toypics\.net/view/(?P[0-9]+)/.*', url) - video_id = mobj.group('videoid') - page = self._download_webpage(url, video_id, 'getting page: '+url) + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + page = self._download_webpage(url, video_id) video_url = self._html_search_regex( r'src:\s+"(http://static[0-9]+\.toypics\.net/flvideo/[^"]+)"', page, 'video URL') title = self._html_search_regex( @@ -48,8 +31,46 @@ class ToypicsIE(InfoExtractor): return { 'id': video_id, 'url': video_url, - 'ext': video_url[-3:], 'title': title, 'uploader': username, - 'age_limit': 18 + 'age_limit': 18, + } + + +class ToypicsUserIE(InfoExtractor): + IE_DESC = 'Toypics user profile' + _VALID_URL = r'http://videos\.toypics\.net/(?P[^/?]+)(?:$|[?#])' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + username = mobj.group('username') + + profile_page = self._download_webpage( + url, username, note='Retrieving profile page') + + video_count = int(self._search_regex( + r'public/">Public Videos \(([0-9]+)\)', profile_page, + 'video count')) + + PAGE_SIZE = 8 + urls = [] + page_count = (video_count + PAGE_SIZE + 1) // PAGE_SIZE + for n in range(1, page_count + 1): + lpage_url = url + '/public/%d' % n + lpage = self._download_webpage( + lpage_url, username, + note='Downloading page %d/%d' % (n, page_count)) + urls.extend( + re.findall( + r'

\n\s*', + lpage)) + + return { + '_type': 'playlist', + 'id': username, + 'entries': [{ + '_type': 'url', + 'url': url, + 'ie_key': 'Toypics', + } for url in urls] }