diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index c2e24def13..d302fe45fd 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -392,6 +392,10 @@ class InfoExtractor(object): url_or_request, video_id, note, errnote, fatal=fatal) if (not fatal) and json_string is False: return None + return self._parse_json( + json_string, video_id, transform_source=transform_source, fatal=fatal) + + def _parse_json(self, json_string, video_id, transform_source=None, fatal=True): if transform_source: json_string = transform_source(json_string) try: diff --git a/youtube_dl/extractor/howstuffworks.py b/youtube_dl/extractor/howstuffworks.py index fccc238840..9478e9e67e 100644 --- a/youtube_dl/extractor/howstuffworks.py +++ b/youtube_dl/extractor/howstuffworks.py @@ -1,10 +1,5 @@ from __future__ import unicode_literals -import re -import json -import random -import string - from .common import InfoExtractor from ..utils import find_xpath_attr @@ -16,97 +11,75 @@ class HowStuffWorksIE(InfoExtractor): '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$', + 'description': 'Cold sleds, freezing temps and warm dog breath... an Iditarod musher\'s dream. Kasey-Dee Gardner jumps on a sled to find out what the big deal is.', + 'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/5266.jpg', }, - 'params': { - # md5 is not consistent - 'skip_download': True - } }, { 'url': 'http://adventure.howstuffworks.com/7199-survival-zone-food-and-water-in-the-savanna-video.htm', 'info_dict': { 'id': '453464', - 'display_id': 'survival-zone-food-and-water-in-the-savanna', 'ext': 'mp4', 'title': 'Survival Zone: Food and Water In the Savanna', - 'description': 'md5:7e1c89f6411434970c15fa094170c371', - 'thumbnail': 're:^https?://.*\.jpg$', + 'description': 'Learn how to find both food and water while trekking in the African savannah. In this video from the Discovery Channel.', + 'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/7199.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$', + 'description': 'Video footage (1 of 3) used by permission of the owner Dan Meyer through Sword Swallowers Association International ', + 'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/118306353233.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') + display_id = self._match_id(url) webpage = self._download_webpage(url, display_id) + clip_info = self._search_regex('(?s)var clip = {(.*?)};', webpage, 'clip info') - 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)) + def extract_clip_info(key, clip_info, name=None, **kargs): + if name is None: + name = key + return self._html_search_regex( + r"\s*%s\s*: '?([^'\n]*[^'\n,])" % key, clip_info, name, **kargs) + video_id = extract_clip_info('content_id', clip_info, 'video id') formats = [] + m3u8_url = extract_clip_info('m3u8', clip_info, 'm3u8 url', default=None) + if m3u8_url is not None: + formats += self._extract_m3u8_formats(m3u8_url , video_id, 'mp4') + mp4 = self._parse_json( + extract_clip_info( + 'mp4', clip_info, 'formats').replace('},]','}]'), video_id) + for video in mp4: + formats.append({ + 'url': video['src'], + 'format_id': video['bitrate'], + 'vbr': int(video['bitrate'].rstrip('k')), + }) + + if not formats: + smil = self._download_xml( + 'http://services.media.howstuffworks.com/videos/%s/smil-service.smil' % video_id, + video_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') + + URL_SUFFIX = '?v=2.11.3&fp=LNX 11,2,202,356&r=A&g=A' - 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}')): + './{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), @@ -116,19 +89,12 @@ class HowStuffWorksIE(InfoExtractor): 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, + 'id': video_id, 'display_id': display_id, - 'title': title, - 'description': description, - 'thumbnail': thumbnail, + 'title': extract_clip_info('clip_title', clip_info, 'title'), + 'description': extract_clip_info('caption', clip_info, 'description', fatal=False), + 'thumbnail': extract_clip_info('video_still_url', clip_info, 'thumbnail'), + 'duration': extract_clip_info('duration', clip_info), 'formats': formats, }