Improve unicode handling for malformed text. Resolved #138.

rushter · rushter · commit 6332dfac82d9 · 2025-09-28T00:37:57.000+04:00
diff --git a/CHANGES.md b/CHANGES.md
@@ -11,6 +11,7 @@
 - Optimize performance for`css_first` in lexbor backend
 - Fix segfaults when accessing attributes. Resolves #135.
 - Add new `.clone` method to lexbor backend. Resolve #117.
+- Improve unicode handling for malformed text. Resolved #138.
 
 ## Version 0.3.34
 
diff --git a/selectolax/lexbor/node.pxi b/selectolax/lexbor/node.pxi
@@ -1004,7 +1004,7 @@ cdef lexbor_action_t text_callback(lxb_dom_node_t *node, void *ctx):
         return LEXBOR_ACTION_OK
 
     try:
-        py_str = text.decode(_ENCODING)
+        py_str = text.decode(_ENCODING, "replace")
 
     except Exception as e:
         PyErr_SetNone(e)
