From 2715fed8f83bea6add5b00894e0878a066a82c2d Mon Sep 17 00:00:00 2001
From: Alessandro <155005371+3clyp50@users.noreply.github.com>
Date: Wed, 30 Sep 2026 16:47:56 +0200
Subject: [PATCH] test(models): cover Chat Completions reasoning fallback

Cover streaming and complete messages, legacy-field precedence, empty or missing reasoning, and encrypted-reasoning filtering. Document the transport contract.

Validation: transport, Responses architecture, encrypted-reasoning, and refusal suites passed (126 tests); git diff --check passed.
---
 helpers/litellm_transport.py.dox.md  |  1 +
 tests/test_stream_tool_early_stop.py | 20 ++++++++++++++++++++
 2 files changed, 21 insertions(+)

diff --git a/helpers/litellm_transport.py.dox.md b/helpers/litellm_transport.py.dox.md
index 551b41259..ce9af15e1 100644
--- a/helpers/litellm_transport.py.dox.md
+++ b/helpers/litellm_transport.py.dox.md
@@ -45,6 +45,7 @@
 - Serialize synthesized Responses function-call JSON with literal Unicode so streamed raw-response logs preserve tool arguments.
 - Preserve provider-state metadata when Responses API calls succeed, and fall back to local replay when provider state is unsupported.
 - Keep prompt-cache markers only for providers that accept them.
+- Chat Completions parsing prefers `reasoning_content` and falls back to `reasoning` for both streaming deltas and complete messages, preserving response content and applying the same encrypted-reasoning filter. Verify with `test_chat_reasoning_field_fallback` in `tests/test_stream_tool_early_stop.py`.
 - Automatic `cache_control` marker placement is provider-gated; the OpenAI/Codex Responses configuration bypasses it. Skipping a marker target never removes its message, tool call, or reasoning content. Responses `phase: commentary`, encrypted reasoning, native replay, and request flags remain owned by their Responses paths.
 - For supported Chat providers, mark the last leading system/developer message and the last two nonempty assistant messages. Retain the previous assistant boundary as the conversation grows; leave user/tool tails containing rebuilt extras unmarked. Before the first assistant message, cache only the leading context. Do not add markers to empty assistant content or thinking/redacted-thinking blocks. Automatic placement uses at most three message breakpoints, leaving one for tool definitions; this is not a limit on conversation messages or native tool calls. Do not mutate caller messages or stored history.
 
diff --git a/tests/test_stream_tool_early_stop.py b/tests/test_stream_tool_early_stop.py
index 65d9310c6..988962009 100644
--- a/tests/test_stream_tool_early_stop.py
+++ b/tests/test_stream_tool_early_stop.py
@@ -1444,6 +1444,26 @@ def test_responses_fallback_on_untyped_input_item_rejection():
     assert policy.mode is litellm_transport.TransportMode.CHAT_COMPLETIONS
 
 
+@pytest.mark.parametrize("source", ["delta", "message"])
+@pytest.mark.parametrize(
+    ("fields", "expected"),
+    [
+        ({"reasoning_content": "legacy"}, "legacy"),
+        ({"reasoning": "fallback"}, "fallback"),
+        ({"reasoning_content": "legacy", "reasoning": "fallback"}, "legacy"),
+        ({"reasoning_content": "", "reasoning": "fallback"}, "fallback"),
+        ({"reasoning": "summary__ENCRYPTED_REASONING__opaque"}, "summary"),
+        ({}, ""),
+    ],
+)
+def test_chat_reasoning_field_fallback(source, fields, expected):
+    parsed = litellm_transport.ChatCompletionsTransport.parse(
+        {"choices": [{source: {"content": "answer", **fields}}]}
+    )
+
+    assert parsed == {"reasoning_delta": expected, "response_delta": "answer"}
+
+
 def test_responses_response_parser_extracts_text_reasoning_and_function_calls():
     text_response = {
         "output": [