perf(parser): make get_all_text() O(nodes)

scrapling/parser.py

@@ -300,7 +300,9 @@ def get_all_text(
 
         ignored_elements: set[Any] = set()
         if ignore_tags:
-            ignored_elements.update(self._root.iter(*ignore_tags))
+            for element in self._root.iter(*ignore_tags):
+                if element not in ignored_elements:
+                    ignored_elements.update(element.iter())
 
         _all_strings = []
 
@@ -315,11 +317,7 @@ def is_visible_text_node(text_node: _ElementUnicodeResult) -> bool:
                 return False
 
             owner = parent.getparent() if text_node.is_tail else parent
-            while owner is not None:
-                if owner in ignored_elements:
-                    return False
-                owner = owner.getparent()
-            return True
+            return owner not in ignored_elements
 
         for text_node in cast(list[_ElementUnicodeResult], _find_all_text_nodes(self._root)):
             text = str(text_node)

tests/parser/test_parser_advanced.py

@@ -210,6 +210,23 @@ def test_get_all_text_preserves_interleaved_text_nodes(self):
 
         assert node.get_all_text("\n", strip=True) == "string1\nstring2\nstring3\nstring4\nstring5\nstring6\nstring7"
 
+    def test_get_all_text_keeps_text_after_comment(self):
+        """Tail text following a comment or PI node must still be collected"""
+        html = "<div>before<!-- c -->after<p>x</p></div>"
+        assert Selector(html, keep_comments=True).get_all_text() == "before\nafter\nx"
+
+    def test_get_all_text_deep_nesting(self):
+        """Deeply nested trees collect every visible text node and skip ignored subtrees"""
+        depth = 500
+        html = "<div>" + "".join(f"<span>t{i}" for i in range(depth)) + "leaf" + "</span>" * depth + "end</div>"
+        text = Selector(html).get_all_text(strip=True)
+        assert text.startswith("t0\nt1\n")
+        assert "leaf" in text and text.endswith("end")
+        assert text.count("\n") == depth
+
+        ignored = "<div>keep<script>" + "<b>x</b>" * depth + "</script>done</div>"
+        assert Selector(ignored).get_all_text(strip=True) == "keep\ndone"
+
 
 class TestTextHandlerAdvanced:
     """Test advanced TextHandler functionality"""