youtube-dl

commit 9d303bf29b13632489ab563b0b8cbdeb53ea2899
parent d58cb3ec7e115cf07a6cf6e035cbba2b5c6861fd
Author: Yen Chi Hsuan <yan12125@gmail.com>
Date:   Fri,  4 Mar 2016 23:21:24 +0800

Merge branch 'mutantmonkey-kusi'

Diffstat:
M
youtube_dl/extractor/__init__.py
 | 
1
+
A
youtube_dl/extractor/kusi.py
 | 
98
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

2 files changed, 99 insertions(+), 0 deletions(-)
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
@@ -340,6 +340,7 @@ from .konserthusetplay import KonserthusetPlayIE
 from .kontrtube import KontrTubeIE
 from .krasview import KrasViewIE
 from .ku6 import Ku6IE
+from .kusi import KUSIIE
 from .kuwo import (
     KuwoIE,
     KuwoAlbumIE,
diff --git a/youtube_dl/extractor/kusi.py 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,
+        }