[twitter] extract 'source_id' and 'source_user' metadata (#7470 #7640)

gallery_dl/extractor/twitter.py

@@ -123,6 +123,11 @@ class TwitterExtractor(Extractor):
             tdata.update(metadata)
             tdata["count"] = len(files)
             yield Message.Directory, tdata
+
+            del tdata["source_id"]
+            if "source_user" in tdata:
+                del tdata["source_user"]
+
             for tdata["num"], file in enumerate(files, 1):
                 file.update(tdata)
                 url = file.pop("url")
@@ -212,8 +217,22 @@ class TwitterExtractor(Extractor):
             file["width"] = media["original_info"].get("width", 0)
             file["height"] = media["original_info"].get("height", 0)
             file["description"] = media.get("ext_alt_text")
+            self._extract_media_source(file, media)
             files.append(file)
 
+    def _extract_media_source(self, dest, media):
+        dest["source_id"] = 0
+
+        if "source_status_id_str" in media:
+            try:
+                dest["source_id"] = text.parse_int(
+                    media["source_status_id_str"])
+                dest["source_user"] = self._transform_user(
+                    media["additional_media_info"]["source_user"]
+                    ["user_results"]["result"])
+            except Exception:
+                pass
+
     def _image_fallback(self, base):
         for fmt in self._size_fallback:
             yield base + fmt
@@ -338,6 +357,7 @@ class TwitterExtractor(Extractor):
                 tget("in_reply_to_status_id_str")),
             "conversation_id": text.parse_int(
                 tget("conversation_id_str")),
+            "source_id"     : 0,
             "date"          : date,
             "author"        : author,
             "user"          : self._user or author,
@@ -401,6 +421,9 @@ class TwitterExtractor(Extractor):
             tdata["reply_to"] = legacy["in_reply_to_screen_name"]
         if "quoted_by" in legacy:
             tdata["quote_by"] = legacy["quoted_by"]
+        if "extended_entities" in legacy:
+            self._extract_media_source(
+                tdata, legacy["extended_entities"]["media"][0])
         if tdata["retweet_id"]:
             tdata["content"] = "RT @{}: {}".format(
                 author["name"], tdata["content"])

test/results/twitter.py

@@ -334,7 +334,9 @@ __tests__ = (
     "#class"   : twitter.TwitterTweetExtractor,
     "#sha1_url": "3a2a43dc5fb79dd5432c701d8e55e87c4e551f47",
 
-    "type": "photo",
+    "type"        : "photo",
+    "source_id"   : 0,
+    "!source_user": dict,
 },
 
 {
@@ -657,6 +659,22 @@ The Washington Post writes, "Three weeks after the toxic train derailment in Ohi
     "#urls"    : "https://pbs.twimg.com/grok-img-share/1866736156786008064.jpg",
 },
 
+{
+    "#url"     : "https://x.com/gdldev/status/1932109706354733077",
+    "#comment" : "'source_id' and 'source_user' metadata (#7470, #7640)",
+    "#category": ("", "twitter", "tweet"),
+    "#class"   : twitter.TwitterTweetExtractor,
+    "#urls"    : (
+        "https://video.twimg.com/amplify_video/1932079443264376832/vid/avc1/640x336/7xo7NCPkMLRWb8NZ.mp4?tag=14",
+        "https://video.twimg.com/ext_tw_video/1930425322333229056/pu/vid/avc1/1024x576/6f_cdEPY3a5CcbZP.mp4?tag=12",
+    ),
+
+    "source_id"  : {1932079546590982508, 1930425346404274416},
+    "source_user": {
+        "name": {"Satorin69", "Derlan144p_"},
+    },
+},
+
 {
     "#url"     : "https://twitter.com/playpokemon/status/1263832915173048321/quotes",
     "#category": ("", "twitter", "quotes"),