[weibo] fix retweets (#2825, #3874, #5263)

- handle 快转 retweets
- disable 'retweets' by default
- skip all retweet media when 'retweets' are disabled
- extract all retweet media when 'retweets' is set to "original"

docs/configuration.rst

@@ -4039,7 +4039,7 @@ extractor.weibo.retweets
 Type
     ``bool``
 Default
-    ``true``
+    ``false``
 Description
     Fetch media from retweeted posts.
 

gallery_dl/extractor/weibo.py

@@ -30,9 +30,9 @@ class WeiboExtractor(Extractor):
         self._prefix, self.user = match.groups()
 
     def _init(self):
-        self.retweets = self.config("retweets", True)
-        self.videos = self.config("videos", True)
         self.livephoto = self.config("livephoto", True)
+        self.retweets = self.config("retweets", False)
+        self.videos = self.config("videos", True)
         self.gifs = self.config("gifs", True)
         self.gifs_video = (self.gifs == "video")
 
@@ -59,15 +59,25 @@ class WeiboExtractor(Extractor):
 
         for status in self.statuses():
 
-            files = []
-            if self.retweets and "retweeted_status" in status:
+            if "ori_mid" in status and not self.retweets:
+                self.log.debug("Skipping %s (快转 retweet)", status["id"])
+                continue
+
+            if "retweeted_status" in status:
+                if not self.retweets:
+                    self.log.debug("Skipping %s (retweet)", status["id"])
+                    continue
+
+                # videos of the original post are in status
+                # images of the original post are in status["retweeted_status"]
+                files = []
+                self._extract_status(status, files)
+                self._extract_status(status["retweeted_status"], files)
+
                 if original_retweets:
                     status = status["retweeted_status"]
-                    self._extract_status(status, files)
-                else:
-                    self._extract_status(status, files)
-                    self._extract_status(status["retweeted_status"], files)
             else:
+                files = []
                 self._extract_status(status, files)
 
             status["date"] = text.parse_datetime(

test/results/weibo.py

@@ -80,11 +80,11 @@ __tests__ = (
     "#category": ("", "weibo", "home"),
     "#class"   : weibo.WeiboHomeExtractor,
     "#range"   : "1-30",
-    "#count"   : 30,
+    "#count"   : 0,
 },
 
 {
-    "#url"     : "https://weibo.com/1758989602?tabtype=feed",
+    "#url"     : "https://weibo.com/2553930725?tabtype=feed",
     "#category": ("", "weibo", "feed"),
     "#class"   : weibo.WeiboFeedExtractor,
     "#range"   : "1-30",
@@ -194,6 +194,28 @@ __tests__ = (
     "#class"   : weibo.WeiboStatusExtractor,
 },
 
+{
+    "#url"     : "https://weibo.cn/detail/4600272267522211",
+    "#comment" : "retweet",
+    "#category": ("", "weibo", "status"),
+    "#class"   : weibo.WeiboStatusExtractor,
+    "#count"   : 0,
+},
+
+{
+    "#url"     : "https://weibo.cn/detail/4600272267522211",
+    "#comment" : "retweet",
+    "#category": ("", "weibo", "status"),
+    "#class"   : weibo.WeiboStatusExtractor,
+    "#options" : {"retweets": True},
+    "#count"   : 2,
+
+    "status": {
+        "id"                     : 4600272267522211,
+        "retweeted_status": {"id": 4600167083287033},
+    },
+},
+
 {
     "#url"     : "https://m.weibo.cn/detail/4600272267522211",
     "#comment" : "original retweets (#1542)",