[daum] improve info extraction

youtube_dl/extractor/__init__.py

@@ -119,7 +119,10 @@
     DailymotionUserIE,
     DailymotionCloudIE,
 )
-from .daum import DaumIE
+from .daum import (
+    DaumIE,
+    DaumClipIE,
+)
 from .dbtv import DBTVIE
 from .dcn import DCNIE
 from .dctp import DctpTvIE

youtube_dl/extractor/daum.py

@@ -2,16 +2,73 @@
 
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..compat import (
+from ..compat import compat_urllib_parse
-    compat_urllib_parse,
+from ..utils import int_or_none
-)
 
 
 class DaumIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/(?:v/|.*?clipid=)(?P<id>[^?#&]+)'
+    _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/v/(?P<id>[^?#&]+)'
+    IE_NAME = 'daum.net'
+
+    _TESTS = [{
+        'url': 'http://tvpot.daum.net/v/vab4dyeDBysyBssyukBUjBz',
+        'info_dict': {
+            'id': 'vab4dyeDBysyBssyukBUjBz',
+            'ext': 'mp4',
+            'title': '마크 헌트 vs 안토니오 실바',
+            'description': 'Mark Hunt vs Antonio Silva',
+            'upload_date': '20131217',
+            'duration': 2117,
+        },
+    }, {
+        'url': 'http://tvpot.daum.net/v/07dXWRka62Y%24',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        query = compat_urllib_parse.urlencode({'vid': video_id})
+        info = self._download_xml(
+            'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id,
+            'Downloading video info')
+        movie_data = self._download_json(
+            'http://videofarm.daum.net/controller/api/closed/v1_2/IntegratedMovieData.json?' + query,
+            video_id, 'Downloading video formats info')
+
+        formats = []
+        for format_el in movie_data['output_list']['output_list']:
+            profile = format_el['profile']
+            format_query = compat_urllib_parse.urlencode({
+                'vid': video_id,
+                'profile': profile,
+            })
+            url_doc = self._download_xml(
+                'http://videofarm.daum.net/controller/api/open/v1_2/MovieLocation.apixml?' + format_query,
+                video_id, note='Downloading video data for %s format' % profile)
+            format_url = url_doc.find('result/url').text
+            formats.append({
+                'url': format_url,
+                'format_id': profile,
+                'width': int_or_none(format_el.get('width')),
+                'height': int_or_none(format_el.get('height')),
+                'filesize': int_or_none(format_el.get('filesize')),
+            })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': info.find('TITLE').text,
+            'formats': formats,
+            'thumbnail': info.find('THUMB_URL').text,
+            'description': info.find('CONTENTS').text,
+            'duration': int_or_none(info.find('DURATION').text),
+            'upload_date': info.find('REGDTTM').text[:8],
+        }
+
+
+class DaumClipIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/.*?clipid=(?P<id>\d+)'
     IE_NAME = 'daum.net'
 
     _TESTS = [{
@@ -23,53 +80,23 @@ class DaumIE(InfoExtractor):
             'description': 'DOTA 2GETHER 시즌2 6회 - 2부',
             'upload_date': '20130831',
             'duration': 3868,
+            'view_count': int,
         },
-    }, {
-        'url': 'http://tvpot.daum.net/v/vab4dyeDBysyBssyukBUjBz',
-        'only_matching': True,
-    }, {
-        'url': 'http://tvpot.daum.net/v/07dXWRka62Y%24',
-        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
-        video_id = mobj.group('id')
+        clip_info = self._download_json('http://tvpot.daum.net/mypot/json/GetClipInfo.do?clipid=%s' % video_id, video_id)['clip_bean']
-        canonical_url = 'http://tvpot.daum.net/v/%s' % video_id
-        webpage = self._download_webpage(canonical_url, video_id)
-        full_id = self._search_regex(
-            r'src=["\']http://videofarm\.daum\.net/controller/video/viewer/Video\.html\?.*?vid=(.+?)[&"\']',
-            webpage, 'full id')
-        query = compat_urllib_parse.urlencode({'vid': full_id})
-        info = self._download_xml(
-            'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id,
-            'Downloading video info')
-        urls = self._download_xml(
-            'http://videofarm.daum.net/controller/api/open/v1_2/MovieData.apixml?' + query,
-            video_id, 'Downloading video formats info')
-
-        formats = []
-        for format_el in urls.findall('result/output_list/output_list'):
-            profile = format_el.attrib['profile']
-            format_query = compat_urllib_parse.urlencode({
-                'vid': full_id,
-                'profile': profile,
-            })
-            url_doc = self._download_xml(
-                'http://videofarm.daum.net/controller/api/open/v1_2/MovieLocation.apixml?' + format_query,
-                video_id, note='Downloading video data for %s format' % profile)
-            format_url = url_doc.find('result/url').text
-            formats.append({
-                'url': format_url,
-                'format_id': profile,
-            })
 
         return {
+            '_type': 'url_transparent',
             'id': video_id,
-            'title': info.find('TITLE').text,
+            'url': 'http://tvpot.daum.net/v/%s' % clip_info['vid'],
-            'formats': formats,
+            'title': clip_info['title'],
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'thumbnail': clip_info.get('thumb_url'),
-            'description': info.find('CONTENTS').text,
+            'description': clip_info.get('contents'),
-            'duration': int(info.find('DURATION').text),
+            'duration': int_or_none(clip_info.get('duration')),
-            'upload_date': info.find('REGDTTM').text[:8],
+            'upload_date': clip_info.get('up_date')[:8],
+            'view_count': int_or_none(clip_info.get('play_count')),
+            'ie_key': 'Daum',
         }