fix: normalize empty content-filter responses instead of raising

[andrewwhitecdw]

fix: content-filter responses raise instead of normalizing

Summary

_response() raised ValueError("LiteLLM returned no text content") for any
response with empty content, including content-filter responses where empty
output is expected. This caused legitimate content-filter results to fail
instead of being normalized.

Root cause

The guard at the end of _response() did not distinguish between a truly
empty/invalid response and a content_filter finish reason, which by design
returns no content.

Fix

Allow empty content when the finish reason is content_filter; keep the
existing raise for all other empty-content cases so unexpected responses still
fail fast.

-    if not content:
-        raise ValueError("LiteLLM returned no text content")
+    if not content and choice.finish_reason != "content_filter":
+        raise ValueError("LiteLLM returned no text content")

Testing

Added test_response_content_filter_empty_content to
examples/experimental/litellm/tests/test_client.py and verified the full
non-e2e suite passes:

uv run --project examples/experimental/litellm --python 3.12 \
  pytest examples/experimental/litellm/tests/test_client.py -m "not e2e" -v
# 34 passed

Contributor guidelines

• DCO sign-off included.
• One focused commit per PR.

Signed-off-by: andrewwhitecdw andrewwhitecdw@users.noreply.github.com

Summary by CodeRabbit

• Bug Fixes

  • Responses blocked by content filtering are now handled correctly when they contain no text.
  • Other unexpected empty responses continue to be reported as errors.

• Tests

  • Added coverage to verify content-filtered responses preserve their stop reason and return an empty output.

[coderabbitai]

Walkthrough

The LiteLLM response parser now accepts empty content when the finish reason is content_filter. A regression test verifies the normalized empty output and preserved stop reason.

Changes

LiteLLM content filter handling

Layer / File(s)	Summary
Content-filter response normalization and regression coverage examples/experimental/litellm/src/switchyard_litellm/client.py, examples/experimental/litellm/tests/test_client.py	_response accepts missing content for content_filter responses. The test verifies empty output content and the preserved stop reason.

Estimated code review effort: 1 (Trivial) | ~5 minutes

Poem

A bunny found a filtered reply,
With empty content drifting by.
The parser kept its stopping sign,
And tests now guard the altered line.
Hop, hop—clean output in a row!

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly summarizes the fix for empty content-filter responses and matches the main change.

---

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

Actionable comments posted: 1

🧹 Nitpick comments (1)

examples/experimental/litellm/tests/test_client.py (1)

557-576: 🎯 Functional Correctness | 🔵 Trivial | ⚡ Quick win

Add a regression test for the preserved non-filter error path.

This test covers the allowed content_filter path. It does not verify that empty content with another finish_reason still raises ValueError("LiteLLM returned no text content"). Add a companion test for finish_reason="stop" with content=None.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@examples/experimental/litellm/tests/test_client.py` around lines 557 - 576,
Add a companion test alongside test_response_content_filter_empty_content using
finish_reason="stop" and content=None, then assert that _response raises
ValueError with the exact message "LiteLLM returned no text content". Keep the
existing content_filter test unchanged.

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@examples/experimental/litellm/tests/test_client.py`:
- Around line 563-576: Update the _response regression test fixture to construct
a typed ModelResponse instead of SimpleNamespace, preserving the content_filter
response values. Narrow or validate the returned outputs collection before
indexing it so the dict[str, object] result passes strict mypy checks.

---

Nitpick comments:
In `@examples/experimental/litellm/tests/test_client.py`:
- Around line 557-576: Add a companion test alongside
test_response_content_filter_empty_content using finish_reason="stop" and
content=None, then assert that _response raises ValueError with the exact
message "LiteLLM returned no text content". Keep the existing content_filter
test unchanged.

🪄 Autofix

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: ee5a76f1-690f-44cf-81a4-ce4df4182125

📥 Commits

Reviewing files that changed from the base of the PR and between fb3fc30 and 8692e42.

📒 Files selected for processing (2)

• examples/experimental/litellm/src/switchyard_litellm/client.py
• examples/experimental/litellm/tests/test_client.py

[andrewwhitecdw]

Completed all CodeRabbit review recommendations:

1. Major inline (typed fixture) — addressed in d2c4535: fixture now builds a typed litellm.ModelResponse using Choices/Message/Usage from litellm.types.utils; outputs is narrowed with isinstance before indexing. The two strict-mypy [index] errors this PR introduced are gone (base: 11 errors → now 9; the remaining 9 are all pre-existing on main, verified via git stash).
2. Nitpick (companion test) — added test_response_empty_content_without_content_filter_raises: finish_reason="stop" + content=None must still raise ValueError("LiteLLM returned no text content"). Existing content_filter test unchanged in behavior.
3. Title warning — PR retitled to fix: normalize empty content-filter responses instead of raising.

Validation: pytest tests/test_client.py -m "not e2e" → 35 passed; ruff check clean.

[nachiketb-nvidia]

Looks good, needs a rebase pls! Thank you!