fix: PythonCodeSplitter secondary-split pieces missing qualified name context

haystack/components/preprocessors/python_code_splitter.py

@@ -556,7 +556,9 @@ def _secondary_split(self, unit: _CodeUnit, parent_doc: Document) -> list[Docume
             meta["secondary_split"] = True
             meta["secondary_split_index"] = idx
             meta["secondary_split_total"] = len(intermediate)
+            meta["qualified_name"] = qualified_name
             results.append(Document(content=piece.content or "", meta=meta))
+
         return results
 
     @component.output_types(documents=list[Document])

releasenotes/notes/python-code-splitter-secondary-split-context-8449dc47dcce0928.yaml

@@ -0,0 +1,9 @@
+---
+fixes:
+  - |
+    Fixed ``PythonCodeSplitter`` losing identifying context for oversized
+    functions, methods, or classes. When a unit is too large and falls back
+    to line-based secondary splitting, only the first resulting piece
+    naturally retains the source ``def``/``class`` line; every piece now
+    includes a ``qualified_name`` field in ``meta`` identifying the function,
+    method, or class it came from.

test/components/preprocessors/test_python_code_splitter.py

@@ -836,6 +836,12 @@ def longer():
         assert chunks_with_short and chunks_with_long
         assert chunks_with_short[0] is not chunks_with_long[0]
 
+    def test_qualified_name_in_meta_for_all_pieces(self, oversized_function_source):
+        splitter = PythonCodeSplitter(min_effective_lines=2, max_effective_lines=5, oversized_factor=3)
+        result = splitter.run(documents=[Document(content=oversized_function_source)])
+        for piece in result["documents"]:
+            assert piece.meta.get("qualified_name") == "giant"
+
 
 class TestEdgeCases:
     def test_module_with_only_docstring(self):