[youtube] Fix extraction.

[youtube-dl] / youtube_dl / extractor / lecture2go.py

diff --git a/youtube_dl/extractor/lecture2go.py b/youtube_dl/extractor/lecture2go.py
index 0075b8a2e4b02de44a78c00b436ef17480423e0c..81b5d41be4a676e55c795fe233591913e7a691c8 100644 (file)
--- a/youtube_dl/extractor/lecture2go.py
+++ b/youtube_dl/extractor/lecture2go.py
@@ -4,19 +4,29 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import determine_ext
+from ..utils import (
+    determine_ext,
+    determine_protocol,
+    parse_duration,
+    int_or_none,
+)
 
 
 class Lecture2GoIE(InfoExtractor):
     _VALID_URL = r'https?://lecture2go\.uni-hamburg\.de/veranstaltungen/-/v/(?P<id>\d+)'
     _TEST = {
         'url': 'https://lecture2go.uni-hamburg.de/veranstaltungen/-/v/17473',
-        'md5': 'a9e76f83b3ef58019c4b7dbc35f406c1',
+        'md5': 'ac02b570883020d208d405d5a3fd2f7f',
         'info_dict': {
             'id': '17473',
             'ext': 'mp4',
-            'url': 'https://fms1.rrz.uni-hamburg.de/abo/64.050_FrankHeitmann_2015-04-13_14-35.mp4',
-            'title': '2 - Endliche Automaten und reguläre Sprachen'
+            'title': '2 - Endliche Automaten und reguläre Sprachen',
+            'creator': 'Frank Heitmann',
+            'duration': 5220,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
         }
     }
 
@@ -27,24 +37,35 @@ class Lecture2GoIE(InfoExtractor):
         title = self._html_search_regex(r'<em[^>]+class="title">(.+)</em>', webpage, 'title')
 
         formats = []
-        for url in set(re.findall(r'"src","([^"]+)"', webpage)):
+        for url in set(re.findall(r'var\s+playerUri\d+\s*=\s*"([^"]+)"', webpage)):
             ext = determine_ext(url)
+            protocol = determine_protocol({'url': url})
             if ext == 'f4m':
-                formats.extend(self._extract_f4m_formats(url, video_id))
+                formats.extend(self._extract_f4m_formats(url, video_id, f4m_id='hds'))
             elif ext == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(url, video_id))
+                formats.extend(self._extract_m3u8_formats(url, video_id, ext='mp4', m3u8_id='hls'))
             else:
+                if protocol == 'rtmp':
+                    continue  # XXX: currently broken
                 formats.append({
+                    'format_id': protocol,
                     'url': url,
                 })
 
         self._sort_formats(formats)
 
-        creator = self._html_search_regex(r'<div[^>]+id="description">([^<]+)</div>', webpage, 'creator')
+        creator = self._html_search_regex(
+            r'<div[^>]+id="description">([^<]+)</div>', webpage, 'creator', fatal=False)
+        duration = parse_duration(self._html_search_regex(
+            r'Duration:\s*</em>\s*<em[^>]*>([^<]+)</em>', webpage, 'duration', fatal=False))
+        view_count = int_or_none(self._html_search_regex(
+            r'Views:\s*</em>\s*<em[^>]+>(\d+)</em>', webpage, 'view count', fatal=False))
 
         return {
             'id': video_id,
             'title': title,
             'formats': formats,
-            'creator': creator
+            'creator': creator,
+            'duration': duration,
+            'view_count': view_count,
         }