[vsco] support '/video/' URLs (#4295 #6973)

requires yt-dlp/youtube-dl to handle m3u8 manifests
mikf · Feb 12, 2025 · f1f27eb · f1f27eb
1 parent d1a8142
commit f1f27eb
Show file tree

Hide file tree

Showing 3 changed files with 53 additions and 7 deletions.
diff --git a/docs/supportedsites.md b/docs/supportedsites.md
@@ -1018,7 +1018,7 @@ Consider all listed sites to potentially be NSFW.
 <tr>
     <td>VSCO</td>
     <td>https://vsco.co/</td>
-    <td>Avatars, Collections, Galleries, individual Images, Spaces, User Profiles</td>
+    <td>Avatars, Collections, Galleries, individual Images, Spaces, User Profiles, Videos</td>
     <td></td>
 </tr>
 <tr>

diff --git a/gallery_dl/extractor/vsco.py b/gallery_dl/extractor/vsco.py
@@ -38,7 +38,9 @@ def items(self):
             if img["is_video"]:
                 if not videos:
                     continue
-                url = "https://" + img["video_url"]
+                url = img["video_url"]
+                if not url.startswith("ytdl:"):
+                    url = "https://" + url
             else:
                 base = img["responsive_url"].partition("/")[2]
                 cdn, _, path = base.partition("/")
@@ -63,6 +65,9 @@ def items(self):
                 "height": img["height"],
                 "description": img.get("description") or "",
             })
+            if data["extension"] == "m3u8":
+                data["_ytdl_manifest"] = "hls"
+                data["extension"] = "mp4"
             yield Message.Url, url, data
 
     def images(self):
@@ -294,12 +299,33 @@ class VscoImageExtractor(VscoExtractor):
     pattern = USER_PATTERN + r"/media/([0-9a-fA-F]+)"
     example = "https://vsco.co/USER/media/0123456789abcdef"
 
-    def __init__(self, match):
-        VscoExtractor.__init__(self, match)
-        self.media_id = match.group(2)
-
     def images(self):
-        url = "{}/{}/media/{}".format(self.root, self.user, self.media_id)
+        url = "{}/{}/media/{}".format(self.root, self.user, self.groups[1])
         data = self._extract_preload_state(url)
         media = data["medias"]["byId"].popitem()[1]["media"]
         return (self._transform_media(media),)
+
+
+class VscoVideoExtractor(VscoExtractor):
+    """Extractor for vsco.co videos links"""
+    subcategory = "video"
+    pattern = USER_PATTERN + r"/video/([^/?#]+)"
+    example = "https://vsco.co/USER/video/012345678-9abc-def0"
+
+    def images(self):
+        url = "{}/{}/video/{}".format(self.root, self.user, self.groups[1])
+        data = self._extract_preload_state(url)
+        media = data["medias"]["byId"].popitem()[1]["media"]
+
+        return ({
+            "_id"           : media["id"],
+            "is_video"      : True,
+            "grid_name"     : "",
+            "upload_date"   : media["createdDate"],
+            "responsive_url": media["posterUrl"],
+            "video_url"     : "ytdl:" + media.get("playbackUrl"),
+            "image_meta"    : None,
+            "width"         : media["width"],
+            "height"        : media["height"],
+            "description"   : media["description"],
+        },)
diff --git a/test/results/vsco.py b/test/results/vsco.py
@@ -108,4 +108,24 @@
     "video": True,
 },
 
+{
+    "#url"     : "https://vsco.co/shavonec/video/c5eb34bb-dd13-4d7a-a09c-2a7cd719c9fa",
+    "#category": ("", "vsco", "video"),
+    "#class"   : vsco.VscoVideoExtractor,
+    "#pattern" : r"ytdl:https://stream\.mux\.com/3o01XbjqTI4rkeRwPzR17H9i7VSFdWX1h\.m3u8\?token=ey.+",
+
+    "date"       : "dt:2020-02-04 15:11:52",
+    "description": "Big news: 🎥 YOU CAN NOW POST VIDEOS TO VSCO ⚡️⚡️💪🏾. ⁣\n⁣\n🔗BTS from my #FashionIsActivism panel with the California African American Arts museum ✊🏾",
+    "extension"  : "mp4",
+    "filename"   : "3o01XbjqTI4rkeRwPzR17H9i7VSFdWX1h",
+    "grid"       : "",
+    "id"         : "c5eb34bb-dd13-4d7a-a09c-2a7cd719c9fa",
+    "meta"       : {},
+    "tags"       : [],
+    "user"       : "shavonec",
+    "video"      : True,
+    "width"      : 624,
+    "height"     : 1232,
+},
+
 )