fix: PythonCodeSplitter emits invalid code for docstring-only bodies

haystack/components/preprocessors/python_code_splitter.py

@@ -190,10 +190,11 @@ def _strip_docstring(
             return self._slice_lines(source_lines, unit_start, unit_end), None
 
         # Skip stripping when the docstring shares a line with the def/class (would
-        # leave broken syntax) or extends past the caller's slice (e.g. class_header).
+        # leave broken syntax), extends past the caller's slice (e.g. class_header), or
+        # is the body's only statement (stripping it would leave no body at all).
         ds_start = first.lineno
         ds_end = first.end_lineno or first.lineno
-        if ds_start <= node.lineno or ds_end > unit_end:
+        if len(body) == 1 or ds_start <= node.lineno or ds_end > unit_end:
             return self._slice_lines(source_lines, unit_start, unit_end), None
 
         before = source_lines[unit_start - 1 : ds_start - 1]

releasenotes/notes/python-code-splitter-docstring-only-body-410e6c2a66da1bbf.yaml

@@ -0,0 +1,7 @@
+---
+fixes:
+  - |
+    Fix ``PythonCodeSplitter`` with ``strip_docstrings=True`` emitting a syntactically
+    invalid chunk for any function, method, or class whose body consists solely of a
+    docstring (for example a one-line custom exception class). Such units are now left
+    unstripped instead of being reduced to a header with no body.

test/components/preprocessors/test_python_code_splitter.py

@@ -2,6 +2,7 @@
 #
 # SPDX-License-Identifier: Apache-2.0
 
+import ast
 import textwrap
 
 import pytest
@@ -687,6 +688,46 @@ def method(self):
         assert "Class-level docstring." not in (header.content or "")
         assert "Class-level docstring." in " | ".join(header.meta.get("docstrings") or [])
 
+    @pytest.mark.parametrize(
+        "source",
+        [
+            pytest.param(
+                textwrap.dedent(
+                    '''
+                    class MyError(Exception):
+                        """Custom error raised when the widget explodes."""
+                    '''
+                ).lstrip(),
+                id="docstring_only_class",
+            ),
+            pytest.param(
+                textwrap.dedent(
+                    '''
+                    def foo():
+                        """This function intentionally does nothing yet."""
+                    '''
+                ).lstrip(),
+                id="docstring_only_function",
+            ),
+            pytest.param(
+                textwrap.dedent(
+                    '''
+                    class Widget:
+                        def explode(self):
+                            """Explode the widget."""
+                    '''
+                ).lstrip(),
+                id="docstring_only_method",
+            ),
+        ],
+    )
+    def test_strip_docstrings_skips_body_that_is_only_a_docstring(self, source):
+        """A def/class whose sole body statement is the docstring must not be stripped down to an empty body."""
+        splitter = PythonCodeSplitter(min_effective_lines=1, max_effective_lines=10, strip_docstrings=True)
+        result = splitter.run(documents=[Document(content=source)])
+        for chunk in result["documents"]:
+            ast.parse(chunk.content)  # raises SyntaxError if the body was stripped down to nothing
+
 
 class TestTopLevelStatements:
     @pytest.fixture