From 664718ff63bd8c61380486ce8f1dc96651bbd2e1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Thu, 14 Aug 2014 20:17:31 +0700 Subject: [PATCH 1/2] [livestream] Improve extraction (Closes #3513) --- youtube_dl/extractor/livestream.py | 23 +++++++++++++++++------ 1 file changed, 17 insertions(+), 6 deletions(-) diff --git a/youtube_dl/extractor/livestream.py b/youtube_dl/extractor/livestream.py index c0c2d9b09d..281a0ce405 100644 --- a/youtube_dl/extractor/livestream.py +++ b/youtube_dl/extractor/livestream.py @@ -111,17 +111,28 @@ def _real_extract(self, url): event_name = mobj.group('event_name') webpage = self._download_webpage(url, video_id or event_name) - if video_id is None: - # This is an event page: + og_video = self._og_search_video_url(webpage, 'player url', fatal=False, default=None) + if og_video is None: config_json = self._search_regex( r'window.config = ({.*?});', webpage, 'window config') info = json.loads(config_json)['event'] + + def is_relevant(vdata, vid): + result = vdata['type'] == 'video' + if video_id is not None: + result = result and compat_str(vdata['data']['id']) == vid + return result + videos = [self._extract_video_info(video_data['data']) - for video_data in info['feed']['data'] - if video_data['type'] == 'video'] - return self.playlist_result(videos, info['id'], info['full_name']) + for video_data in info['feed']['data'] + if is_relevant(video_data, video_id)] + if video_id is None: + # This is an event page: + return self.playlist_result(videos, info['id'], info['full_name']) + else: + if videos: + return videos[0] else: - og_video = self._og_search_video_url(webpage, 'player url') query_str = compat_urllib_parse_urlparse(og_video).query query = compat_urlparse.parse_qs(query_str) api_url = query['play_url'][0].replace('.smil', '') From c990bb36338c422d4da9f092384da1dfbb528d87 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Fri, 15 Aug 2014 21:38:41 +0700 Subject: [PATCH 2/2] [howstuffworks] Add extractor (#3500) Content-length is invalid for final download links. --- youtube_dl/extractor/__init__.py | 1 + youtube_dl/extractor/howstuffworks.py | 134 ++++++++++++++++++++++++++ 2 files changed, 135 insertions(+) create mode 100644 youtube_dl/extractor/howstuffworks.py diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 2ea2c73d24..27602e0c03 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -126,6 +126,7 @@ from .hentaistigma import HentaiStigmaIE from .hotnewhiphop import HotNewHipHopIE from .howcast import HowcastIE +from .howstuffworks import HowStuffWorksIE from .huffpost import HuffPostIE from .hypem import HypemIE from .iconosquare import IconosquareIE diff --git a/youtube_dl/extractor/howstuffworks.py b/youtube_dl/extractor/howstuffworks.py new file mode 100644 index 0000000000..68684b9974 --- /dev/null +++ b/youtube_dl/extractor/howstuffworks.py @@ -0,0 +1,134 @@ +from __future__ import unicode_literals + +import re +import json +import random +import string + +from .common import InfoExtractor +from ..utils import find_xpath_attr + + +class HowStuffWorksIE(InfoExtractor): + _VALID_URL = r'https?://[\da-z-]+\.howstuffworks\.com/(?:[^/]+/)*\d+-(?P.+?)-video\.htm' + _TESTS = [ + { + 'url': 'http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm', + 'info_dict': { + 'id': '450221', + 'display_id': 'cool-jobs-iditarod-musher', + 'ext': 'flv', + 'title': 'Cool Jobs - Iditarod Musher', + 'description': 'md5:82bb58438a88027b8186a1fccb365f90', + 'thumbnail': 're:^https?://.*\.jpg$', + }, + 'params': { + # md5 is not consistent + 'skip_download': True + } + }, + { + 'url': 'http://adventure.howstuffworks.com/39516-deadliest-catch-jakes-farewell-pots-video.htm', + 'info_dict': { + 'id': '553470', + 'display_id': 'deadliest-catch-jakes-farewell-pots', + 'ext': 'mp4', + 'title': 'Deadliest Catch: Jake\'s Farewell Pots', + 'description': 'md5:9632c346d5e43ee238028c9cefd8dbbc', + 'thumbnail': 're:^https?://.*\.jpg$', + }, + 'params': { + # md5 is not consistent + 'skip_download': True + } + }, + { + 'url': 'http://entertainment.howstuffworks.com/arts/2706-sword-swallowing-1-by-dan-meyer-video.htm', + 'info_dict': { + 'id': '440011', + 'display_id': 'sword-swallowing-1-by-dan-meyer', + 'ext': 'flv', + 'title': 'Sword Swallowing #1 by Dan Meyer', + 'description': 'md5:b2409e88172913e2e7d3d1159b0ef735', + 'thumbnail': 're:^https?://.*\.jpg$', + }, + 'params': { + # md5 is not consistent + 'skip_download': True + } + }, + ] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + display_id = mobj.group('id') + webpage = self._download_webpage(url, display_id) + + content_id = self._search_regex(r'var siteSectionId="(\d+)";', webpage, 'content id') + + mp4 = self._search_regex( + r'''(?xs)var\s+clip\s*=\s*{\s* + .+?\s* + content_id\s*:\s*%s\s*,\s* + .+?\s* + mp4\s*:\s*\[(.*?),?\]\s* + };\s* + videoData\.push\(clip\);''' % content_id, + webpage, 'mp4', fatal=False, default=None) + + smil = self._download_xml( + 'http://services.media.howstuffworks.com/videos/%s/smil-service.smil' % content_id, + content_id, 'Downloading video SMIL') + + http_base = find_xpath_attr( + smil, + './{0}head/{0}meta'.format('{http://www.w3.org/2001/SMIL20/Language}'), + 'name', + 'httpBase').get('content') + + def random_string(str_len=0): + return ''.join([random.choice(string.ascii_uppercase) for _ in range(str_len)]) + + URL_SUFFIX = '?v=2.11.3&fp=LNX 11,2,202,356&r=%s&g=%s' % (random_string(5), random_string(12)) + + formats = [] + + if mp4: + for video in json.loads('[%s]' % mp4): + bitrate = video['bitrate'] + fmt = { + 'url': video['src'].replace('http://pmd.video.howstuffworks.com', http_base) + URL_SUFFIX, + 'format_id': bitrate, + } + m = re.search(r'(?P\d+)[Kk]', bitrate) + if m: + fmt['vbr'] = int(m.group('vbr')) + formats.append(fmt) + else: + for video in smil.findall( + './/{0}body/{0}switch/{0}video'.format('{http://www.w3.org/2001/SMIL20/Language}')): + vbr = int(video.attrib['system-bitrate']) / 1000 + formats.append({ + 'url': '%s/%s%s' % (http_base, video.attrib['src'], URL_SUFFIX), + 'format_id': '%dk' % vbr, + 'vbr': vbr, + }) + + self._sort_formats(formats) + + title = self._og_search_title(webpage) + TITLE_SUFFIX = ' : HowStuffWorks' + if title.endswith(TITLE_SUFFIX): + title = title[:-len(TITLE_SUFFIX)] + + description = self._og_search_description(webpage) + thumbnail = self._og_search_thumbnail(webpage) + + return { + 'id': content_id, + 'display_id': display_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'formats': formats, + }