[aliexpress:live] Fix issues (closes #13698, closes #13707)

youtube_dl/extractor/aliexpress.py

@@ -1,40 +1,53 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-
-import re
-
 from .common import InfoExtractor
-from ..utils import try_get, float_or_none
 from ..compat import compat_str
+from ..utils import (
+    float_or_none,
+    try_get,
+)
 
 
 class AliExpressLiveIE(InfoExtractor):
-
+    _VALID_URL = r'https?://live\.aliexpress\.com/live/(?P<id>\d+)'
-    _VALID_URL = r'https?://live\.aliexpress\.com/live/(?P<id>[0-9]{16})'
     _TEST = {
         'url': 'https://live.aliexpress.com/live/2800002704436634',
-        'md5': '7ac2bc46afdd18f0b45a0a340fc47ffe',
+        'md5': 'e729e25d47c5e557f2630eaf99b740a5',
         'info_dict': {
             'id': '2800002704436634',
-            'ext': 'm3u8',
+            'ext': 'mp4',
             'title': 'CASIMA7.22',
+            'thumbnail': r're:http://.*\.jpg',
             'uploader': 'CASIMA Official Store',
-            'upload_date': '20170714',
+            'timestamp': 1500717600,
-            'timestamp': 1500027138,
+            'upload_date': '20170722',
         },
     }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        page = self._download_webpage(url, video_id)
+
-        run_params_json = self._search_regex(r'runParams = (.+)[\s+]var myCtl', page, 'runParams', flags=re.DOTALL)
+        webpage = self._download_webpage(url, video_id)
-        run_params = self._parse_json(run_params_json, video_id)
+
+        data = self._parse_json(
+            self._search_regex(
+                r'(?s)runParams\s*=\s*({.+?})\s*;?\s*var',
+                webpage, 'runParams'),
+            video_id)
+
+        title = data['title']
+
+        formats = self._extract_m3u8_formats(
+            data['replyStreamUrl'], video_id, 'mp4',
+            entry_protocol='m3u8_native', m3u8_id='hls')
 
         return {
             'id': video_id,
-            'title': run_params['title'],
+            'title': title,
-            'url': run_params['replyStreamUrl'],
+            'thumbnail': data.get('coverUrl'),
-            'uploader': try_get(run_params, lambda x: x['followBar']['name'], compat_str),
+            'uploader': try_get(
-            'timestamp': float_or_none(try_get(run_params, lambda x: x['followBar']['createTime']) / 1000),
+                data, lambda x: x['followBar']['name'], compat_str),
+            'timestamp': float_or_none(data.get('startTimeLong'), scale=1000),
+            'formats': formats,
         }