diff options
author | Yen Chi Hsuan <yan12125@gmail.com> | 2016-03-04 23:21:24 +0800 |
---|---|---|
committer | Yen Chi Hsuan <yan12125@gmail.com> | 2016-03-04 23:21:24 +0800 |
commit | 9d303bf29b13632489ab563b0b8cbdeb53ea2899 (patch) | |
tree | 3d8916804ba8edda4c39f7a0914ac01a99c23696 /youtube_dl/extractor/kusi.py | |
parent | d58cb3ec7e115cf07a6cf6e035cbba2b5c6861fd (diff) | |
parent | 5f1688f27106b6501a2a97d7e6fc1d9f079f1f9f (diff) |
Merge branch 'mutantmonkey-kusi'
Diffstat (limited to 'youtube_dl/extractor/kusi.py')
-rw-r--r-- | youtube_dl/extractor/kusi.py | 98 |
1 files changed, 98 insertions, 0 deletions
diff --git a/youtube_dl/extractor/kusi.py b/youtube_dl/extractor/kusi.py new file mode 100644 index 000000000..5d458e31e --- /dev/null +++ b/youtube_dl/extractor/kusi.py @@ -0,0 +1,98 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import random +import re + +from .common import InfoExtractor +from ..compat import compat_urllib_parse_unquote_plus +from ..utils import ( + int_or_none, + float_or_none, + timeconvert, + update_url_query, + xpath_text, +) + + +class KUSIIE(InfoExtractor): + _VALID_URL = r'http://(?:www\.)?kusi\.com/(?P<path>story/.+|video\?clipId=(?P<clipId>\d+))' + _TESTS = [{ + 'url': 'http://www.kusi.com/story/31183873/turko-files-case-closed-put-on-hold', + 'md5': 'f926e7684294cf8cb7bdf8858e1b3988', + 'info_dict': { + 'id': '12203019', + 'ext': 'mp4', + 'title': 'Turko Files: Case Closed! & Put On Hold!', + 'duration': 231, + }, + }, { + 'url': 'http://kusi.com/video?clipId=12203019', + 'info_dict': { + 'id': '12203019', + 'ext': 'mp4', + 'title': 'Turko Files: Case Closed! & Put On Hold!', + 'duration': 231.0, + 'upload_date': '20160210', + 'timestamp': 1455087571, + 'thumbnail': 're:^https?://.*\.jpg$' + }, + 'params': { + 'skip_download': True, # Same as previous one + }, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + clip_id = mobj.group('clipId') + video_id = clip_id or mobj.group('path') + + webpage = self._download_webpage(url, video_id) + + if clip_id is None: + video_id = clip_id = self._html_search_regex( + r'"clipId"\s*,\s*"(\d+)"', webpage, 'clip id') + + affiliate_id = self._search_regex( + r'affiliateId\s*:\s*\'([^\']+)\'', webpage, 'affiliate id') + + # See __Packages/worldnow/model/GalleryModel.as of WNGallery.swf + xml_url = update_url_query('http://www.kusi.com/build.asp', { + 'buildtype': 'buildfeaturexmlrequest', + 'featureType': 'Clip', + 'featureid': clip_id, + 'affiliateno': affiliate_id, + 'clientgroupid': '1', + 'rnd': int(round(random.random() * 1000000)), + }) + + doc = self._download_xml(xml_url, video_id) + + video_title = xpath_text(doc, 'HEADLINE') + duration = float_or_none( + xpath_text(doc, 'DURATION', fatal=False), scale=1000) + description = xpath_text(doc, 'ABSTRACT', fatal=False) + thumbnail = xpath_text(doc, './THUMBNAILIMAGE/FILENAME', fatal=False) + createtion_time = timeconvert( + xpath_text(doc, 'rfc822creationdate', fatal=False)) + + quality_options = doc.find('{http://search.yahoo.com/mrss/}group').findall('{http://search.yahoo.com/mrss/}content') + formats = [] + for quality in quality_options: + formats.append({ + 'url': compat_urllib_parse_unquote_plus(quality.attrib['url']), + 'height': int_or_none(quality.attrib.get('height')), + 'width': int_or_none(quality.attrib.get('width')), + 'vbr': float_or_none(quality.attrib.get('bitratebits'), scale=1024), + }) + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': video_title, + 'description': description, + 'duration': duration, + 'formats': formats, + 'thumbnail': thumbnail, + 'timestamp': createtion_time, + } |