[puls4] Improve and extract more metadata

youtube_dl/extractor/puls4.py

@@ -1,61 +1,88 @@
 # -*- coding: utf-8 -*-
-
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-
+from ..utils import (
-import re
+    ExtractorError,
+    unified_strdate,
+    int_or_none,
+)
 
 
 class Puls4IE(InfoExtractor):
-
+    _VALID_URL = r'https?://(?:www\.)?puls4\.com/video/[^/]+/play/(?P<id>[0-9]+)'
-    _VALID_URL = r'https?://www.puls4.com/video/.+?/play/(?P<id>[0-9]+)'
     _TESTS = [{
         'url': 'http://www.puls4.com/video/pro-und-contra/play/2716816',
         'md5': '49f6a6629747eeec43cef6a46b5df81d',
         'info_dict': {
             'id': '2716816',
             'ext': 'mp4',
-            'title': 'Pro und Contra vom 23.02.2015'}},
+            'title': 'Pro und Contra vom 23.02.2015',
-        {
+            'description': 'md5:293e44634d9477a67122489994675db6',
+            'duration': 2989,
+            'upload_date': '20150224',
+            'uploader': 'PULS_4',
+        },
+        'skip': 'Only works from Germany',
+    }, {
         'url': 'http://www.puls4.com/video/kult-spielfilme/play/1298106',
         'md5': '6a48316c8903ece8dab9b9a7bf7a59ec',
         'info_dict': {
             'id': '1298106',
             'ext': 'mp4',
-            'title': 'Lucky Fritz'}}
+            'title': 'Lucky Fritz',
-    ]
+        },
+        'skip': 'Only works from Germany',
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        # if fsk-button
+        error_message = self._html_search_regex(
-        real_url = self._html_search_regex(r'\"fsk-button\".+?href=\"([^"]+)',
+            r'<div class="message-error">(.+?)</div>',
+            webpage, 'error message', default=None)
+        if error_message:
+            raise ExtractorError(
+                '%s returned error: %s' % (self.IE_NAME, error_message), expected=True)
+
+        real_url = self._html_search_regex(
+            r'\"fsk-button\".+?href=\"([^"]+)',
             webpage, 'fsk_button', default=None)
         if real_url:
             webpage = self._download_webpage(real_url, video_id)
 
-        title = self._html_search_regex(
+        player = self._search_regex(
-            r'<div id="bg_brandableContent">.+?<h1>(.+?)</h1>',
+            r'p4_video_player(?:_iframe)?\("video_\d+_container"\s*,(.+?)\);\s*\}',
-            webpage, 'title', flags=re.DOTALL)
+            webpage, 'player')
 
-        sd_url = self._html_search_regex(
+        player_json = self._parse_json(
-            r'{\"url\":\"([^"]+?)\",\"hd\":false',
+            '[%s]' % player, video_id,
-            webpage, 'sd_url').replace('\\', '')
+            transform_source=lambda s: s.replace('undefined,', ''))
 
-        formats = [{'format_id': 'sd', 'url': sd_url, 'quality': -2}]
+        formats = None
+        result = None
 
-        hd_url = self._html_search_regex(
+        for v in player_json:
-            r'{\"url\":\"([^"]+?)\",\"hd\":true',
+            if isinstance(v, list) and not formats:
-            webpage, 'hd_url', default=None)
+                formats = [{
-        if hd_url:
+                    'url': f['url'],
-            hd_url = hd_url.replace('\\', '')
+                    'format': 'hd' if f.get('hd') else 'sd',
-            formats.append({'format_id': 'hd', 'url': hd_url, 'quality': -1})
+                    'width': int_or_none(f.get('size_x')),
-
+                    'height': int_or_none(f.get('size_y')),
-        return {
+                    'tbr': int_or_none(f.get('bitrate')),
+                } for f in v]
+                self._sort_formats(formats)
+            elif isinstance(v, dict) and not result:
+                result = {
                     'id': video_id,
-            'title': title,
+                    'title': v['videopartname'].strip(),
-            'formats': formats,
+                    'description': v.get('videotitle'),
-            'ext': 'mp4'
+                    'duration': int_or_none(v.get('videoduration') or v.get('episodeduration')),
+                    'upload_date': unified_strdate(v.get('clipreleasetime')),
+                    'uploader': v.get('channel'),
                 }
+
+        result['formats'] = formats
+
+        return result