Fix segfaults when accessing attributes. Resolves #135

Author: rushter

CHANGES.md

@@ -9,6 +9,7 @@
 - Add `.inner_html` property. Allows to get and set inner HTML of a node.
 - Update various docstrings.
 - Optimize performance for`css_first` in lexbor backend
+- Fix segfaults when accessing attributes. Resolves #135.
 
 ## Version 0.3.34
 

selectolax/lexbor/node.pxi

@@ -343,6 +343,9 @@ cdef class LexborNode:
         cdef size_t str_len = 0
         attributes = dict()
 
+        if self.node.type != LXB_DOM_NODE_TYPE_ELEMENT:
+            return attributes
+
         while attr != NULL:
             key = lxb_dom_attr_local_name_noi(attr, &str_len)
             value = lxb_dom_attr_value_noi(attr, &str_len)

tests/test_lexbor.py

@@ -18,3 +18,14 @@ def test_sets_inner_html():
     parser.css_first("#main").inner_html = "<span>Test</span>"
     actual = parser.css_first("#main").inner_html
     assert actual == expected
+
+
+def test_checking_attributes_does_not_segfault():
+    parser = LexborHTMLParser("")
+    root_node = parser.root
+    assert root_node is not None
+    for node in root_node.traverse():
+        print(node.parent)
+        parent = node.parent
+        assert parent is not None
+        parent = parent.attributes.get("anything")