X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fpandoratv.py;h=538738c090a515c296969dab9762ee1d643d1382;hb=HEAD;hp=a0a0c114a35c9d504cd6711bc7453b7f70e751f8;hpb=9accfed4e724d3048e894c2323b3a478951d670d;p=youtube-dl

diff --git a/youtube_dl/extractor/pandoratv.py b/youtube_dl/extractor/pandoratv.py
index a0a0c114a..538738c09 100644
--- a/youtube_dl/extractor/pandoratv.py
+++ b/youtube_dl/extractor/pandoratv.py
@@ -1,56 +1,134 @@
-# encoding: utf-8
+# coding: utf-8
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
+import re
 
+from .common import InfoExtractor
 from ..compat import (
+    compat_str,
     compat_urlparse,
 )
 from ..utils import (
     ExtractorError,
+    float_or_none,
+    parse_duration,
+    str_to_int,
+    urlencode_postdata,
 )
 
 
 class PandoraTVIE(InfoExtractor):
-    _VALID_URL = r'http://(?:.+?\.)?channel.pandora.tv/channel/video.ptv\?'
+    IE_NAME = 'pandora.tv'
+    IE_DESC = '판도라TV'
+    _VALID_URL = r'''(?x)
+                        https?://
+                            (?:
+                                (?:www\.)?pandora\.tv/view/(?P<user_id>[^/]+)/(?P<id>\d+)|  # new format
+                                (?:.+?\.)?channel\.pandora\.tv/channel/video\.ptv\?|        # old format
+                                m\.pandora\.tv/?\?                                          # mobile
+                            )
+                    '''
     _TESTS = [{
         'url': 'http://jp.channel.pandora.tv/channel/video.ptv?c1=&prgid=53294230&ch_userid=mikakim&ref=main&lot=cate_01_2',
         'info_dict': {
-            'description': '\u982d\u3092\u64ab\u3067\u3066\u304f\u308c\u308b\uff1f',
-            'ext': 'mp4',
             'id': '53294230',
-            'title': '\u982d\u3092\u64ab\u3067\u3066\u304f\u308c\u308b\uff1f',
+            'ext': 'flv',
+            'title': '頭を撫でてくれる？',
+            'description': '頭を撫でてくれる？',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 39,
             'upload_date': '20151218',
+            'uploader': 'カワイイ動物まとめ',
+            'uploader_id': 'mikakim',
+            'view_count': int,
+            'like_count': int,
         }
+    }, {
+        'url': 'http://channel.pandora.tv/channel/video.ptv?ch_userid=gogoucc&prgid=54721744',
+        'info_dict': {
+            'id': '54721744',
+            'ext': 'flv',
+            'title': '[HD] JAPAN COUNTDOWN 170423',
+            'description': '[HD] JAPAN COUNTDOWN 170423',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 1704.9,
+            'upload_date': '20170423',
+            'uploader': 'GOGO_UCC',
+            'uploader_id': 'gogoucc',
+            'view_count': int,
+            'like_count': int,
+        },
+        'params': {
+            # Test metadata only
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://www.pandora.tv/view/mikakim/53294230#36797454_new',
+        'only_matching': True,
+    }, {
+        'url': 'http://m.pandora.tv/?c=view&ch_userid=mikakim&prgid=54600346',
+        'only_matching': True,
     }]
 
-
     def _real_extract(self, url):
-        qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
-        video_id = qs.get('prgid', [None])[0]
-        user_id = qs.get('ch_userid', [None])[0]
-        if any(not f for f in (video_id, user_id,)):
-            raise ExtractorError('Invalid URL', expected=True)
-
-        data_url ='http://m.pandora.tv/?c=view&m=viewJsonApi&ch_userid={userid}&prgid={prgid}'.format(userid=user_id,prgid=video_id)
-        data = self._download_json(data_url, video_id)
+        mobj = re.match(self._VALID_URL, url)
+        user_id = mobj.group('user_id')
+        video_id = mobj.group('id')
+
+        if not user_id or not video_id:
+            qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
+            video_id = qs.get('prgid', [None])[0]
+            user_id = qs.get('ch_userid', [None])[0]
+            if any(not f for f in (video_id, user_id,)):
+                raise ExtractorError('Invalid URL', expected=True)
+
+        data = self._download_json(
+            'http://m.pandora.tv/?c=view&m=viewJsonApi&ch_userid=%s&prgid=%s'
+            % (user_id, video_id), video_id)
+
         info = data['data']['rows']['vod_play_info']['result']
 
         formats = []
-        for format_id in sorted([k for k in info if k.startswith('v') and k.endswith('Url') and info[k]]):
+        for format_id, format_url in info.items():
+            if not format_url:
+                continue
+            height = self._search_regex(
+                r'^v(\d+)[Uu]rl$', format_id, 'height', default=None)
+            if not height:
+                continue
+
+            play_url = self._download_json(
+                'http://m.pandora.tv/?c=api&m=play_url', video_id,
+                data=urlencode_postdata({
+                    'prgid': video_id,
+                    'runtime': info.get('runtime'),
+                    'vod_url': format_url,
+                }),
+                headers={
+                    'Origin': url,
+                    'Content-Type': 'application/x-www-form-urlencoded',
+                })
+            format_url = play_url.get('url')
+            if not format_url:
+                continue
+
             formats.append({
-                'format_id': format_id,
-                'url': info[format_id],
-                'ext': 'mp4',
-                'height': int(format_id[1:-3]),
+                'format_id': '%sp' % height,
+                'url': format_url,
+                'height': int(height),
             })
+        self._sort_formats(formats)
 
         return {
-            'description': info['body'],
-            'thumbnail': info['thumbnail'],
-            'formats': formats,
             'id': video_id,
             'title': info['subject'],
-            'upload_date': info['fid'][:8],
-            'view_count': info['hit'],
+            'description': info.get('body'),
+            'thumbnail': info.get('thumbnail') or info.get('poster'),
+            'duration': float_or_none(info.get('runtime'), 1000) or parse_duration(info.get('time')),
+            'upload_date': info['fid'].split('/')[-1][:8] if isinstance(info.get('fid'), compat_str) else None,
+            'uploader': info.get('nickname'),
+            'uploader_id': info.get('upload_userid'),
+            'view_count': str_to_int(info.get('hit')),
+            'like_count': str_to_int(info.get('likecnt')),
+            'formats': formats,
         }