From 36a2aff363bf07f203aaed1556d8cca63987cea9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Mike=20F=C3=A4hrmann?= <mike_faehrmann@web.de>
Date: Thu, 15 Jul 2021 00:43:42 +0200
Subject: [PATCH] [vk] improve metadata extraction and URL pattern (fixes
 #1691)

- always fetch all user metadata
- use 'user[name]' for directory names if available
---
 gallery_dl/extractor/vk.py | 65 +++++++++++++++++++++++++++-----------
 1 file changed, 47 insertions(+), 18 deletions(-)

diff --git a/gallery_dl/extractor/vk.py b/gallery_dl/extractor/vk.py
index 1ce1140f..21786417 100644
--- a/gallery_dl/extractor/vk.py
+++ b/gallery_dl/extractor/vk.py
@@ -17,43 +17,60 @@ class VkPhotosExtractor(Extractor):
     """Extractor for photos from a vk user"""
     category = "vk"
     subcategory = "photos"
-    directory_fmt = ("{category}", "{user[id]}")
+    directory_fmt = ("{category}", "{user[name]|user[id]}")
     filename_fmt = "{id}.{extension}"
     archive_fmt = "{id}"
     root = "https://vk.com"
     request_interval = 1.0
-    pattern = r"(?:https://)?(?:www\.|m\.)?vk\.com/(?:albums|photos|id)(\d+)"
+    pattern = (r"(?:https://)?(?:www\.|m\.)?vk\.com/(?:"
+               r"(?:albums|photos|id)(-?\d+)|([^/?#]+))")
     test = (
         ("https://vk.com/id398982326", {
             "pattern": r"https://sun\d+-\d+\.userapi\.com/c\d+/v\d+"
                        r"/[0-9a-f]+/[\w-]+\.jpg",
             "count": ">= 35",
+            "keywords": {
+                "id": r"re:\d+",
+                "user": {
+                    "id": "398982326",
+                    "info": "Мы за Движуху! – m1ni SounD #4 [EROmusic]",
+                    "name": "",
+                    "nick": "Dobrov Kurva",
+                },
+            },
+        }),
+        ("https://vk.com/cosplayinrussia", {
+            "range": "75-100",
+            "keywords": {
+                "id": r"re:\d+",
+                "user": {
+                    "id"  : "-165740836",
+                    "info": "Предложка открыта, кидайте ваши косплейчики. При "
+                            "правильном оформлении они будут опубликованы",
+                    "name": "cosplayinrussia",
+                    "nick": "Косплей | Cosplay 18+",
+                },
+            },
         }),
         ("https://m.vk.com/albums398982326"),
         ("https://www.vk.com/id398982326?profile=1"),
+        ("https://vk.com/albums-165740836"),
     )
 
     def __init__(self, match):
         Extractor.__init__(self, match)
-        self.user_id = match.group(1)
+        self.user_id, self.user_name = match.groups()
 
     def items(self):
-        user_id = self.user_id
-
-        if self.config("metadata"):
-            url = "{}/id{}".format(self.root, user_id)
-            extr = text.extract_from(self.request(url).text)
-            data = {"user": {
-                "id"  : user_id,
-                "nick": text.unescape(extr(
-                    "<title>", " | VK<")),
-                "name": text.unescape(extr(
-                    '<h1 class="page_name">', "<")).replace("  ", " "),
-                "info": text.unescape(text.remove_html(extr(
-                    '<span class="current_text">', '</span')))
-            }}
+        if self.user_id:
+            user_id = self.user_id
+            prefix = "public" if user_id[0] == "-" else "id"
+            url = "{}/{}{}".format(self.root, prefix, user_id.lstrip("-"))
+            data = self._extract_profile(url)
         else:
-            data = {"user": {"id": user_id}}
+            url = "{}/{}".format(self.root, self.user_name)
+            data = self._extract_profile(url)
+            user_id = data["user"]["id"]
 
         photos_url = "{}/photos{}".format(self.root, user_id)
         headers = {
@@ -86,3 +103,15 @@ class VkPhotosExtractor(Extractor):
             if cnt <= 40 or offset == params["offset"]:
                 return
             params["offset"] = offset
+
+    def _extract_profile(self, url):
+        extr = text.extract_from(self.request(url).text)
+        return {"user": {
+            "name": text.unescape(extr(
+                'rel="canonical" href="https://vk.com/', '"')),
+            "nick": text.unescape(extr(
+                '<h1 class="page_name">', "<")).replace("  ", " "),
+            "info": text.unescape(text.remove_html(extr(
+                '<span class="current_text">', '</span'))),
+            "id"  : extr('<a href="/albums', '"'),
+        }}