Fix missing inner_error in azure ContentPolicyViolationError

litellm/litellm_core_utils/exception_mapping_utils.py

@@ -2005,7 +2005,7 @@ def exception_type(  # type: ignore  # noqa: PLR0915
                 ):
                     exception_mapping_worked = True
                     raise ContentPolicyViolationError(
-                        message=f"litellm.ContentPolicyViolationError: AzureException - {message}",
+                        message=f"litellm.ContentPolicyViolationError: AzureException - {error_str}",
                         llm_provider="azure",
                         model=model,
                         litellm_debug_info=extra_information,

tests/test_litellm/litellm_core_utils/test_exception_mapping_utils.py

@@ -1,6 +1,10 @@
 import pytest
+import httpx
+
+from litellm.litellm_core_utils.exception_mapping_utils import ExceptionCheckers, exception_type
+from litellm import ContentPolicyViolationError
+from openai import BadRequestError
 
-from litellm.litellm_core_utils.exception_mapping_utils import ExceptionCheckers
 
 # Test cases for is_error_str_context_window_exceeded
 # Tuple format: (error_message, expected_result)
@@ -28,4 +32,58 @@ def test_is_error_str_context_window_exceeded(error_str, expected):
     """
     Tests the is_error_str_context_window_exceeded function with various error strings.
     """
-    assert ExceptionCheckers.is_error_str_context_window_exceeded(error_str) == expected
+    assert ExceptionCheckers.is_error_str_context_window_exceeded(error_str) == expected
+
+def test_mapping_azure_content_policy_exception_contains_inner_error():
+    """
+    Test mapping of Azure content policy violation error that contains 'innererror' and 'content_filter_result'
+    to the correct LiteLLM exception type.
+    """
+
+    dummy_request = httpx.Request("POST", "https://example.com/v1/completions")
+
+    response = httpx.Response(
+        status_code=400,
+        json={
+            "error": {
+                "message": "The response was filtered due to the prompt triggering Azure OpenAI's content management policy.",
+                "type": None,
+                "param": "prompt",
+                "code": "content_filter",
+                "status": 400,
+                "innererror": {
+                    "code": "ResponsibleAIPolicyViolation",
+                    "content_filter_result": {
+                        "hate": {"filtered": True, "severity": "high"},
+                        "jailbreak": {"filtered": False, "detected": False},
+                        "self_harm": {"filtered": False, "severity": "safe"},
+                        "sexual": {"filtered": False, "severity": "safe"},
+                        "violence": {"filtered": False, "severity": "medium"},
+                    },
+                },
+            }
+        },
+        request=dummy_request,
+    )
+
+    bad_request_error = BadRequestError(
+        message=f"Error code: 400 - {response.json()}",
+        response=response,
+        body=response.json(),
+    )
+
+    with pytest.raises(ContentPolicyViolationError) as excinfo:
+        exception_type('gpt-4o', Exception(bad_request_error), 'azure')
+
+    mapped_exception = excinfo.value
+
+    assert mapped_exception.__class__.__name__ == "ContentPolicyViolationError"
+
+    message = str(mapped_exception)
+    assert message.startswith("litellm.BadRequestError:")
+    assert "litellm.ContentPolicyViolationError" in message
+    assert "AzureException" in message
+    assert "The response was filtered due to the prompt triggering Azure OpenAI's content management policy." in message
+    assert "innererror" in message
+    assert "content_filter_result" in message
+    assert "ResponsibleAIPolicyViolation" in message