From 281828b58b86e4893cf58eb26f4dfeff91f7e74a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Mike=20F=C3=A4hrmann?= <mike_faehrmann@web.de>
Date: Fri, 24 Dec 2021 03:42:28 +0100
Subject: [PATCH] [tumblrgallery] improve search pagination (fixes #2132)

---
 gallery_dl/extractor/tumblrgallery.py | 23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)

diff --git a/gallery_dl/extractor/tumblrgallery.py b/gallery_dl/extractor/tumblrgallery.py
index cd2d0f5837..e790613fe3 100644
--- a/gallery_dl/extractor/tumblrgallery.py
+++ b/gallery_dl/extractor/tumblrgallery.py
@@ -98,7 +98,10 @@ class TumblrgallerySearchExtractor(TumblrgalleryExtractor):
     filename_fmt = "{category}_{num:>03}_{gallery_id}_{id}_{title}.{extension}"
     directory_fmt = ("{category}", "{search_term}")
     pattern = BASE_PATTERN + r"(/s\.php\?q=([^&#]+))"
-    test = ("https://tumblrgallery.xyz/s.php?q=everyday-life",)
+    test = ("https://tumblrgallery.xyz/s.php?q=everyday-life", {
+        "pattern": r"https://\d+\.media\.tumblr\.com/.+",
+        "count": "< 1000",
+    })
 
     def __init__(self, match):
         TumblrgalleryExtractor.__init__(self, match)
@@ -110,17 +113,9 @@ def metadata(self, page):
         }
 
     def images(self, _):
-        page_num = 1
+        page_url = "s.php?q=" + self.search_term
         while True:
-            url = "{}/s.php?q={}&page={}".format(
-                self.root, self.search_term, page_num)
-            response = self.request(url, allow_redirects=False, fatal=False)
-
-            if response.status_code >= 300:
-                return
-
-            page = response.text
-            page_num += 1
+            page = self.request(self.root + "/" + page_url).text
 
             for gallery_id in text.extract_iter(
                     page, '<div class="title"><a href="post/', '.html'):
@@ -135,3 +130,9 @@ def images(self, _):
                         text.extract(post_page, "<title>", "</title>")[0]
                     )).replace("_", "-")
                     yield url, data
+
+            next_url = text.extract(
+                page, '</span> <a class="btn btn-primary" href="', '"')[0]
+            if not next_url or page_url == next_url:
+                return
+            page_url = next_url