[instagram] download videos from 'video_dash_manifest' data

(#6379 #7006)

requires yt-dlp / youtube-dl

docs/configuration.rst

@@ -2997,11 +2997,19 @@ Description
 extractor.instagram.videos
 --------------------------
 Type
-    ``bool``
+    * ``bool``
+    * ``string``
 Default
     ``true``
 Description
-    Download video files.
+    Controls video download behavior.
+
+    ``true`` | ``"dash"`` | ``"ytdl"``
+        Download videos from ``video_dash_manifest`` data using |ytdl|
+    ``"merged"``
+        Download pre-merged video formats
+    ``false``
+        Do not download videos
 
 
 extractor.itaku.videos
@@ -4448,19 +4456,19 @@ Description
 
     Possible formats include
 
-    * ``"gif"``
-    * ``"gif_transparent"``
-    * ``"gifpreview"``
-    * ``"mediumgif"``
-    * ``"tinygif"``
-    * ``"tinygif_transparent"``
-    * ``"mp4"``
-    * ``"tinymp4"``
-    * ``"webm"``
-    * ``"webp"``
-    * ``"webp_transparent"``
-    * ``"tinywebp"``
-    * ``"tinywebp_transparent"``
+    * ``gif``
+    * ``gif_transparent``
+    * ``mediumgif``
+    * ``gifpreview``
+    * ``tinygif``
+    * ``tinygif_transparent``
+    * ``mp4``
+    * ``tinymp4``
+    * ``webm``
+    * ``webp``
+    * ``webp_transparent``
+    * ``tinywebp``
+    * ``tinywebp_transparent``
 
 
 extractor.tiktok.audio

gallery_dl/extractor/instagram.py

@@ -56,9 +56,11 @@ class InstagramExtractor(Extractor):
 
         data = self.metadata()
         videos = self.config("videos", True)
+        if videos:
+            videos_dash = (videos != "merged")
+            videos_headers = {"User-Agent": "Mozilla/5.0"}
         previews = self.config("previews", False)
         max_posts = self.config("max-posts")
-        video_headers = {"User-Agent": "Mozilla/5.0"}
 
         order = self.config("order-files")
         reverse = order[0] in ("r", "d") if order else False
@@ -92,8 +94,12 @@ class InstagramExtractor(Extractor):
                 url = file.get("video_url")
                 if url:
                     if videos:
-                        file["_http_headers"] = video_headers
+                        file["_http_headers"] = videos_headers
                         text.nameext_from_url(url, file)
+                        if videos_dash:
+                            file["_fallback"] = (url,)
+                            file["_ytdl_manifest"] = "dash"
+                            url = "ytdl:dash"
                         yield Message.Url, url, file
                     if previews:
                         file["media_id"] += "p"
@@ -246,6 +252,7 @@ class InstagramExtractor(Extractor):
                 "video_url"  : video["url"] if video else None,
                 "width"      : media["width"],
                 "height"     : media["height"],
+                "_ytdl_manifest_data": item.get("video_dash_manifest"),
             }
 
             if "expiring_at" in item: