merge from main.

deepeval/_version.py

@@ -1 +1 @@
-__version__: str = "0.20.30"
+__version__: str = "0.20.33"

deepeval/evaluate.py

@@ -22,6 +22,7 @@ class TestResult:
     actual_output: str
     expected_output: str
     context: List[str]
+    retrieval_context: List[str]
 
 
 def create_test_result(
@@ -37,6 +38,7 @@ def create_test_result(
             actual_output=test_case.actual_output,
             expected_output=test_case.expected_output,
             context=test_case.context,
+            retrieval_context=test_case.retrieval_context,
         )
     else:
         raise ValueError("TestCase not supported yet.")
@@ -139,3 +141,4 @@ def print_test_result(test_result: TestResult):
     print(f"  - actual output: {test_result.actual_output}")
     print(f"  - expected output: {test_result.expected_output}")
     print(f"  - context: {test_result.context}")
+    print(f"  - retrieval context: {test_result.retrieval_context}")

deepeval/test_run.py

@@ -72,6 +72,7 @@ class APITestCase(BaseModel):
     run_duration: float = Field(..., alias="runDuration")
     traceStack: Optional[dict] = Field(None)
     context: Optional[list] = Field(None)
+    retrieval_context: Optional[list] = Field(None, alias="retrievalContext")
     id: Optional[str] = None
 
 
@@ -133,6 +134,7 @@ def add_llm_test_case(
                 metricsMetadata=[metrics_metadata],
                 runDuration=run_duration,
                 context=test_case.context,
+                retrievalContext=test_case.retrieval_context,
                 traceStack=get_trace_stack(),
                 id=test_case.id,
             )