fix: keep docstring when PythonCodeSplitter would leave an empty body

[Harsh23Kashyap]

Related Issues

• fixes PythonCodeSplitter with strip_docstrings=True emits invalid Python for docstring-only function/method/class bodies #12330

Proposed Changes:

PythonCodeSplitter with strip_docstrings=True reduced a unit whose body was only the docstring to a header with no body (def foo():\n, class Foo:\n), which is a SyntaxError. The same bug applied to a class whose class-header slice contained only the class-level docstring and no other body statement before the first method: stripping the docstring from the header left class Foo:\n, also invalid.

The fix is in _strip_docstring: after confirming the first body statement is the docstring, also check that there is at least one other body statement within the slice. If not, return the original slice unchanged and None for the docstring, so the chunk stays valid Python and the docstring is not moved to meta["docstrings"] (consistent with how units that have nothing to strip are handled today). Functions, methods, and classes whose body has both a docstring and other statements are unaffected: the docstring is still stripped and moved to meta["docstrings"] as before.

How did you test it?

• Added test_strip_docstrings_keeps_docstring_when_body_is_only_docstring in TestDocstringStripping, covering a function, a class, a nested class, and a method whose body is only the docstring. Each case asserts the emitted content (a) parses as valid Python via ast.parse, (b) still contains the docstring inline, and (c) did not move anything to meta["docstrings"].
• Added test_strip_docstrings_keeps_class_header_docstring_when_no_other_body covering the class-header-with-only-docstring case (docstring + methods, but no other body statement before the first method).
• The pre-existing test_strip_docstrings_moves_them_to_meta, test_strip_docstrings_preserves_module_docstring, and test_strip_class_header_docstring_moves_to_meta all still pass — the change only affects the empty-body edge case.
• Full module: 71 passed (was 69). The wider preprocessor surface (test/components/preprocessors/) is 358 passed with no regressions. hatch run fmt clean. hatch run test:types reports 17 pre-existing errors in 9 unrelated files (azure, openai_counter, etc.) — confirmed pre-existing via git stash of this branch.
• Verified the bug end-to-end with a standalone reproducer before the fix (4 cases produce SyntaxError; all 4 produce valid Python after the fix).

Checklist

• I have read the contributors guidelines and the code of conduct
• I have updated the release note (reno note in releasenotes/notes/)
• I have run hatch run fmt and hatch run test:types locally
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• Any dependent changes have been merged and published in downstream modules

[vercel]

@Harsh23Kashyap is attempting to deploy a commit to the deepset Team on Vercel.

A member of the Team first needs to authorize it.

[github-actions]

Hi @Harsh23Kashyap, thanks for your interest in contributing to Haystack! 🙏

⚠️ Issue #12330 is already being addressed by open pull request(s) #12331, #12346. Before opening a PR for an issue, please check whether a PR is already linked to it, and consider contributing to the existing PR instead. We may close duplicate PRs to keep the review queue manageable.

This is an automated message to help us keep the review queue healthy.

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[HaystackBot]

Hi @Harsh23Kashyap, thanks a lot for your contribution! 🙏

We noticed that the Contributor License Agreement (CLA) check (license/cla) hasn't passed yet, so we've temporarily moved this PR to draft and paused the review assignment.

To get your PR reviewed, please sign the CLA via the link in the license/cla check below (or in the CLA bot comment). As soon as the check turns green, this PR will automatically be marked ready for review again and a reviewer will be re-assigned.