From 47eb18cd5635efa0319972673a5bfa90e93f7680 Mon Sep 17 00:00:00 2001
From: Battleplus <3559424769@qq.com>
Date: Mon, 24 Aug 2026 13:04:41 +0800
Subject: [PATCH] fix(filter): skip pruning nodes inside pre/code blocks
 (#2110)

PruningContentFilter was decomposing whitespace-only spans inside
<pre>/<code> elements (e.g., <span class=\"w\"> </span> in syntax
highlighting), corrupting code that raw_markdown preserves.

Add _is_in_code_block() guard mirroring the existing fix in
content_scraping_strategy.remove_empty_elements_fast (v0.7.8, #1181).

Closes #2110
---
 crawl4ai/content_filter_strategy.py | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/crawl4ai/content_filter_strategy.py b/crawl4ai/content_filter_strategy.py
index ab99a793c..40405e77f 100644
--- a/crawl4ai/content_filter_strategy.py
+++ b/crawl4ai/content_filter_strategy.py
@@ -712,6 +712,11 @@ def _prune_tree(self, node):
         if self._is_preserved(node):
             return
 
+        # Skip pruning for nodes inside <pre> or <code> — whitespace is significant there
+        # (matches the guard in content_scraping_strategy.remove_empty_elements_fast, fix for #1181)
+        if self._is_in_code_block(node):
+            return
+
         text_len = len(node.get_text(strip=True))
         tag_len = len(node.encode_contents().decode("utf-8"))
         link_text_len = sum(