From 46b6b71159eaf86d4b2bc850494a57aa2d8e8d1a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Mike=20F=C3=A4hrmann?= <mike_faehrmann@web.de>
Date: Mon, 6 Jan 2025 17:18:04 +0100
Subject: [PATCH] [wallhaven] extract 'search[tags]' and 'search[tag_id]'
 metadata

(#6772)
---
 gallery_dl/extractor/wallhaven.py | 23 +++++++++++++++++++----
 test/results/wallhaven.py         | 19 +++++++++++++++++++
 2 files changed, 38 insertions(+), 4 deletions(-)

diff --git a/gallery_dl/extractor/wallhaven.py b/gallery_dl/extractor/wallhaven.py
index 479e8a85..e5b764ab 100644
--- a/gallery_dl/extractor/wallhaven.py
+++ b/gallery_dl/extractor/wallhaven.py
@@ -54,7 +54,7 @@ class WallhavenExtractor(Extractor):
 class WallhavenSearchExtractor(WallhavenExtractor):
     """Extractor for search results on wallhaven.cc"""
     subcategory = "search"
-    directory_fmt = ("{category}", "{search[q]}")
+    directory_fmt = ("{category}", "{search[tags]}")
     archive_fmt = "s_{search[q]}_{id}"
     pattern = r"(?:https?://)?wallhaven\.cc/search(?:/?\?([^#]+))?"
     example = "https://wallhaven.cc/search?q=QUERY"
@@ -64,7 +64,7 @@ class WallhavenSearchExtractor(WallhavenExtractor):
         self.params = text.parse_query(match.group(1))
 
     def wallpapers(self):
-        return self.api.search(self.params.copy())
+        return self.api.search(self.params)
 
     def metadata(self):
         return {"search": self.params}
@@ -141,7 +141,7 @@ class WallhavenUploadsExtractor(WallhavenExtractor):
 
     def wallpapers(self):
         params = {"q": "@" + self.username}
-        return self.api.search(params.copy())
+        return self.api.search(params)
 
     def metadata(self):
         return {"username": self.username}
@@ -215,20 +215,35 @@ class WallhavenAPI():
 
     def _pagination(self, endpoint, params=None, metadata=None):
         if params is None:
+            params_ptr = None
             params = {}
+        else:
+            params_ptr = params
+            params = params.copy()
         if metadata is None:
             metadata = self.extractor.config("metadata")
 
         while True:
             data = self._call(endpoint, params)
 
+            meta = data.get("meta")
+            if params_ptr is not None:
+                if meta and "query" in meta:
+                    query = meta["query"]
+                    if isinstance(query, dict):
+                        params_ptr["tags"] = query.get("tag")
+                        params_ptr["tag_id"] = query.get("id")
+                    else:
+                        params_ptr["tags"] = query
+                        params_ptr["tag_id"] = 0
+                params_ptr = None
+
             if metadata:
                 for wp in data["data"]:
                     yield self.info(str(wp["id"]))
             else:
                 yield from data["data"]
 
-            meta = data.get("meta")
             if not meta or meta["current_page"] >= meta["last_page"]:
                 return
             params["page"] = meta["current_page"] + 1
diff --git a/test/results/wallhaven.py b/test/results/wallhaven.py
index 47a8ba77..cf942aaf 100644
--- a/test/results/wallhaven.py
+++ b/test/results/wallhaven.py
@@ -12,6 +12,14 @@ __tests__ = (
     "#url"     : "https://wallhaven.cc/search?q=touhou",
     "#category": ("", "wallhaven", "search"),
     "#class"   : wallhaven.WallhavenSearchExtractor,
+    "#pattern" : r"https://w\.wallhaven\.cc/full/\w\w/wallhaven-\w+\.\w+",
+    "#range"   : "1-10",
+
+    "search": {
+        "q"     : "touhou",
+        "tags"  : "touhou",
+        "tag_id": 0,
+    },
 },
 
 {
@@ -20,6 +28,17 @@ __tests__ = (
     "#class"   : wallhaven.WallhavenSearchExtractor,
     "#pattern" : r"https://w\.wallhaven\.cc/full/\w\w/wallhaven-\w+\.\w+",
     "#count"   : "<= 30",
+
+    "search": {
+        "categories": "111",
+        "order"     : "asc",
+        "page"      : "3",
+        "purity"    : "100",
+        "sorting"   : "date_added",
+        "q"         : "id:87",
+        "tags"      : "Fujibayashi Kyou",
+        "tag_id"    : 87,
+    },
 },
 
 {