From c9d11d2fc9c4a43006f44023e9b594f552518fac Mon Sep 17 00:00:00 2001
From: Xiangyu Li <3367250374@qq.com>
Date: Mon, 29 Jun 2026 22:44:59 +0800
Subject: [PATCH] Log compressed responses over warn size

---
 .../downloadermiddlewares/httpcompression.py  |  6 ++++
 ...st_downloadermiddleware_httpcompression.py | 29 +++++++++++++++++++
 2 files changed, 35 insertions(+)

diff --git a/scrapy/downloadermiddlewares/httpcompression.py b/scrapy/downloadermiddlewares/httpcompression.py
index 2b1721ced..923ae6dcb 100644
--- a/scrapy/downloadermiddlewares/httpcompression.py
+++ b/scrapy/downloadermiddlewares/httpcompression.py
@@ -123,6 +123,12 @@ class HttpCompressionMiddleware:
                     response.body, content_encoding, max_size
                 )
             except _DecompressionMaxSizeExceeded as e:
+                if warn_size and e.decompressed_size >= warn_size:
+                    logger.warning(
+                        f"{response} body size after decompression "
+                        f"({e.decompressed_size} B so far) is larger than the "
+                        f"download warning size ({warn_size} B)."
+                    )
                 raise IgnoreRequest(
                     f"Ignored response {response} because its body "
                     f"({len(response.body)} B compressed, "
diff --git a/tests/test_downloadermiddleware_httpcompression.py b/tests/test_downloadermiddleware_httpcompression.py
index 5c4085657..5eb58069b 100644
--- a/tests/test_downloadermiddleware_httpcompression.py
+++ b/tests/test_downloadermiddleware_httpcompression.py
@@ -558,6 +558,35 @@ class TestHttpCompression:
 
         self._test_compression_bomb_setting("zstd")
 
+    def test_compression_bomb_logs_warnsize_before_ignoring(self):
+        settings = {"DOWNLOAD_MAXSIZE": 1_000_000, "DOWNLOAD_WARNSIZE": 500_000}
+        crawler = get_crawler(Spider, settings_dict=settings)
+        spider = crawler._create_spider("scrapytest.org")
+        mw = HttpCompressionMiddleware.from_crawler(crawler)
+        mw.open_spider(spider)
+        response = self._getresponse("bomb-gzip")
+
+        with (
+            LogCapture(
+                "scrapy.downloadermiddlewares.httpcompression",
+                propagate=False,
+                level=WARNING,
+            ) as log,
+            pytest.raises(IgnoreRequest),
+        ):
+            mw.process_response(response.request, response)
+        log.check_present(
+            (
+                "scrapy.downloadermiddlewares.httpcompression",
+                "WARNING",
+                (
+                    "<200 http://scrapytest.org/> body size after "
+                    "decompression (1048576 B so far) is larger than the "
+                    "download warning size (500000 B)."
+                ),
+            ),
+        )
+
     def _test_compression_bomb_spider_attr(self, compression_id):
         class DownloadMaxSizeSpider(Spider):
             download_maxsize = 1_000_000