pydantic-ai 0.4.11__tar.gz → 0.5.0__tar.gz

{pydantic_ai-0.4.11 → pydantic_ai-0.5.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pydantic-ai
-Version: 0.4.11
+Version: 0.5.0
 Summary: Agent Framework / shim to use Pydantic with LLMs
 Project-URL: Homepage, https://ai.pydantic.dev
 Project-URL: Source, https://github.com/pydantic/pydantic-ai
@@ -28,11 +28,11 @@ Classifier: Topic :: Internet
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Requires-Python: >=3.9
-Requires-Dist: pydantic-ai-slim[ag-ui,anthropic,bedrock,cli,cohere,evals,google,groq,huggingface,mcp,mistral,openai,retries,vertexai]==0.4.11
+Requires-Dist: pydantic-ai-slim[ag-ui,anthropic,bedrock,cli,cohere,evals,google,groq,huggingface,mcp,mistral,openai,retries,vertexai]==0.5.0
 Provides-Extra: a2a
 Requires-Dist: fasta2a>=0.4.1; extra == 'a2a'
 Provides-Extra: examples
-Requires-Dist: pydantic-ai-examples==0.4.11; extra == 'examples'
+Requires-Dist: pydantic-ai-examples==0.5.0; extra == 'examples'
 Provides-Extra: logfire
 Requires-Dist: logfire>=3.11.0; extra == 'logfire'
 Description-Content-Type: text/markdown

{pydantic_ai-0.4.11 → pydantic_ai-0.5.0}/tests/evals/test_dataset.py

@@ -9,7 +9,7 @@ from typing import Any
 import pytest
 from dirty_equals import HasRepr, IsNumber
 from inline_snapshot import snapshot
-from pydantic import BaseModel
+from pydantic import BaseModel, TypeAdapter
 
 from ..conftest import IsStr, try_import
 from .utils import render_table
@@ -20,7 +20,7 @@ with try_import() as imports_successful:
 
     from pydantic_evals import Case, Dataset
     from pydantic_evals.dataset import increment_eval_metric, set_eval_attribute
-    from pydantic_evals.evaluators import EvaluationResult, Evaluator, EvaluatorOutput, LLMJudge, Python
+    from pydantic_evals.evaluators import EvaluationResult, Evaluator, EvaluatorOutput, EvaluatorSpec, LLMJudge, Python
     from pydantic_evals.evaluators.context import EvaluatorContext
 
     @dataclass
@@ -32,7 +32,7 @@ with try_import() as imports_successful:
         def evaluate(self, ctx: EvaluatorContext[object, object, object]) -> EvaluatorOutput:
             return self.output
 
-    from pydantic_evals.reporting import ReportCase, ReportCaseAdapter
+    from pydantic_evals.reporting import EvaluationReport, ReportCase, ReportCaseAdapter
 
 pytestmark = [pytest.mark.skipif(not imports_successful(), reason='pydantic-evals not installed'), pytest.mark.anyio]
 
@@ -456,13 +456,13 @@ async def test_repeated_name_outputs(example_dataset: Dataset[TaskInput, TaskOut
                 scores={},
                 labels={
                     'output': EvaluationResult(
-                        name='output', value='a', reason=None, source=MockEvaluator(output={'output': 'a'})
+                        name='output', value='a', reason=None, source=MockEvaluator(output={'output': 'a'}).as_spec()
                     ),
                     'output_2': EvaluationResult(
-                        name='output', value='b', reason=None, source=MockEvaluator(output={'output': 'b'})
+                        name='output', value='b', reason=None, source=MockEvaluator(output={'output': 'b'}).as_spec()
                     ),
                     'output_3': EvaluationResult(
-                        name='output', value='c', reason=None, source=MockEvaluator(output={'output': 'c'})
+                        name='output', value='c', reason=None, source=MockEvaluator(output={'output': 'c'}).as_spec()
                     ),
                 },
                 assertions={},
@@ -482,13 +482,13 @@ async def test_repeated_name_outputs(example_dataset: Dataset[TaskInput, TaskOut
                 scores={},
                 labels={
                     'output': EvaluationResult(
-                        name='output', value='a', reason=None, source=MockEvaluator(output={'output': 'a'})
+                        name='output', value='a', reason=None, source=MockEvaluator(output={'output': 'a'}).as_spec()
                     ),
                     'output_2': EvaluationResult(
-                        name='output', value='b', reason=None, source=MockEvaluator(output={'output': 'b'})
+                        name='output', value='b', reason=None, source=MockEvaluator(output={'output': 'b'}).as_spec()
                     ),
                     'output_3': EvaluationResult(
-                        name='output', value='c', reason=None, source=MockEvaluator(output={'output': 'c'})
+                        name='output', value='c', reason=None, source=MockEvaluator(output={'output': 'c'}).as_spec()
                     ),
                 },
                 assertions={},
@@ -501,6 +501,73 @@ async def test_repeated_name_outputs(example_dataset: Dataset[TaskInput, TaskOut
     )
 
 
+async def test_report_round_trip_serialization(example_dataset: Dataset[TaskInput, TaskOutput, TaskMetadata]):
+    """Test the increment_eval_metric function."""
+
+    async def my_task(inputs: TaskInput) -> TaskOutput:
+        return TaskOutput(answer=f'answer to {inputs.query}')
+
+    example_dataset.add_evaluator(MockEvaluator({'output': 'a'}))
+
+    report = await example_dataset.evaluate(my_task)
+    assert report == snapshot(
+        EvaluationReport(
+            name='my_task',
+            cases=[
+                ReportCase(
+                    name='case1',
+                    inputs=TaskInput(query='What is 2+2?'),
+                    metadata=TaskMetadata(difficulty='easy', category='general'),
+                    expected_output=TaskOutput(answer='4', confidence=1.0),
+                    output=TaskOutput(answer='answer to What is 2+2?', confidence=1.0),
+                    metrics={},
+                    attributes={},
+                    scores={},
+                    labels={
+                        'output': EvaluationResult(
+                            name='output',
+                            value='a',
+                            reason=None,
+                            source=EvaluatorSpec(name='MockEvaluator', arguments=({'output': 'a'},)),
+                        )
+                    },
+                    assertions={},
+                    task_duration=1.0,
+                    total_duration=6.0,
+                    trace_id='00000000000000000000000000000001',
+                    span_id='0000000000000003',
+                ),
+                ReportCase(
+                    name='case2',
+                    inputs=TaskInput(query='What is the capital of France?'),
+                    metadata=TaskMetadata(difficulty='medium', category='geography'),
+                    expected_output=TaskOutput(answer='Paris', confidence=1.0),
+                    output=TaskOutput(answer='answer to What is the capital of France?', confidence=1.0),
+                    metrics={},
+                    attributes={},
+                    scores={},
+                    labels={
+                        'output': EvaluationResult(
+                            name='output',
+                            value='a',
+                            reason=None,
+                            source=EvaluatorSpec(name='MockEvaluator', arguments=({'output': 'a'},)),
+                        )
+                    },
+                    assertions={},
+                    task_duration=1.0,
+                    total_duration=4.0,
+                    trace_id='00000000000000000000000000000001',
+                    span_id='0000000000000007',
+                ),
+            ],
+        )
+    )
+
+    report_adapter = TypeAdapter(EvaluationReport[TaskInput, TaskOutput, TaskMetadata])
+    assert report == report_adapter.validate_json(report_adapter.dump_json(report, indent=2))
+
+
 async def test_genai_attribute_collection(example_dataset: Dataset[TaskInput, TaskOutput, TaskMetadata]):
     async def my_task(inputs: TaskInput) -> TaskOutput:
         with logfire.span(

{pydantic_ai-0.4.11 → pydantic_ai-0.5.0}/tests/evals/test_evaluator_base.py

@@ -52,11 +52,11 @@ def test_evaluation_result():
     evaluator = DummyEvaluator()
 
     # Test basic result
-    result = EvaluationResult(name='test', value=True, reason='Success', source=evaluator)
+    result = EvaluationResult(name='test', value=True, reason='Success', source=evaluator.as_spec())
     assert result.name == 'test'
     assert result.value is True
     assert result.reason == 'Success'
-    assert result.source == evaluator
+    assert result.source == evaluator.as_spec()
 
     # Test downcast with matching type
     downcast = result.downcast(bool)

{pydantic_ai-0.4.11 → pydantic_ai-0.5.0}/tests/evals/test_evaluator_spec.py

@@ -6,7 +6,7 @@ from pydantic import ValidationError
 from ..conftest import try_import
 
 with try_import() as imports_successful:
-    from pydantic_evals.evaluators._spec import (
+    from pydantic_evals.evaluators.spec import (
         EvaluatorSpec,
         _SerializedEvaluatorSpec,  # pyright: ignore[reportPrivateUsage]
     )

{pydantic_ai-0.4.11 → pydantic_ai-0.5.0}/tests/evals/test_evaluators.py

@@ -19,7 +19,6 @@ with try_import() as imports_successful:
     from logfire.testing import CaptureLogfire
 
     from pydantic_evals.evaluators._run_evaluator import run_evaluator
-    from pydantic_evals.evaluators._spec import EvaluatorSpec
     from pydantic_evals.evaluators.common import (
         Contains,
         Equals,
@@ -36,6 +35,7 @@ with try_import() as imports_successful:
         Evaluator,
         EvaluatorOutput,
     )
+    from pydantic_evals.evaluators.spec import EvaluatorSpec
     from pydantic_evals.otel._context_in_memory_span_exporter import context_subtree
     from pydantic_evals.otel.span_tree import SpanQuery, SpanTree
 
@@ -162,7 +162,7 @@ async def test_evaluator_call(test_context: EvaluatorContext[TaskInput, TaskOutp
     assert results[0].name == 'result'
     assert results[0].value == 'passed'
     assert results[0].reason is None
-    assert results[0].source is evaluator
+    assert results[0].source == EvaluatorSpec(name='ExampleEvaluator', arguments=None)
 
 
 async def test_is_instance_evaluator():
@@ -242,7 +242,14 @@ async def test_custom_evaluator_name(test_context: EvaluatorContext[TaskInput, T
     evaluator = CustomNameFieldEvaluator(result=123, evaluation_name='abc')
 
     assert to_jsonable_python(await run_evaluator(evaluator, test_context)) == snapshot(
-        [{'name': 'abc', 'reason': None, 'source': {'evaluation_name': 'abc', 'result': 123}, 'value': 123}]
+        [
+            {
+                'name': 'abc',
+                'reason': None,
+                'source': {'arguments': {'evaluation_name': 'abc', 'result': 123}, 'name': 'CustomNameFieldEvaluator'},
+                'value': 123,
+            }
+        ]
     )
 
     @dataclass
@@ -260,7 +267,14 @@ async def test_custom_evaluator_name(test_context: EvaluatorContext[TaskInput, T
     evaluator = CustomNamePropertyEvaluator(result=123, my_name='marcelo')
 
     assert to_jsonable_python(await run_evaluator(evaluator, test_context)) == snapshot(
-        [{'name': 'hello marcelo', 'reason': None, 'source': {'my_name': 'marcelo', 'result': 123}, 'value': 123}]
+        [
+            {
+                'name': 'hello marcelo',
+                'reason': None,
+                'source': {'arguments': {'my_name': 'marcelo', 'result': 123}, 'name': 'CustomNamePropertyEvaluator'},
+                'value': 123,
+            }
+        ]
     )
 
 

{pydantic_ai-0.4.11 → pydantic_ai-0.5.0}/tests/evals/test_reporting.py

@@ -48,7 +48,7 @@ def sample_assertion(mock_evaluator: Evaluator[TaskInput, TaskOutput, TaskMetada
         name='MockEvaluator',
         value=True,
         reason=None,
-        source=mock_evaluator,
+        source=mock_evaluator.as_spec(),
     )
 
 
@@ -57,8 +57,8 @@ def sample_score(mock_evaluator: Evaluator[TaskInput, TaskOutput, TaskMetadata])
     return EvaluationResult(
         name='MockEvaluator',
         value=2.5,
-        reason=None,
-        source=mock_evaluator,
+        reason='my reason',
+        source=mock_evaluator.as_spec(),
     )
 
 
@@ -68,7 +68,7 @@ def sample_label(mock_evaluator: Evaluator[TaskInput, TaskOutput, TaskMetadata])
         name='MockEvaluator',
         value='hello',
         reason=None,
-        source=mock_evaluator,
+        source=mock_evaluator.as_spec(),
     )
 
 
@@ -120,6 +120,7 @@ async def test_evaluation_renderer_basic(sample_report: EvaluationReport):
         label_configs={},
         metric_configs={},
         duration_config={},
+        include_reasons=False,
     )
 
     table = renderer.build_table(sample_report)
@@ -137,6 +138,43 @@ async def test_evaluation_renderer_basic(sample_report: EvaluationReport):
 """)
 
 
+async def test_evaluation_renderer_with_reasons(sample_report: EvaluationReport):
+    """Test basic functionality of EvaluationRenderer."""
+    renderer = EvaluationRenderer(
+        include_input=True,
+        include_output=True,
+        include_metadata=True,
+        include_expected_output=True,
+        include_durations=True,
+        include_total_duration=True,
+        include_removed_cases=False,
+        include_averages=True,
+        input_config={},
+        metadata_config={},
+        output_config={},
+        score_configs={},
+        label_configs={},
+        metric_configs={},
+        duration_config={},
+        include_reasons=True,
+    )
+
+    table = renderer.build_table(sample_report)
+    assert render_table(table) == snapshot("""\
+                                                                                     Evaluation Summary: test_report
+┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
+┃ Case ID   ┃ Inputs                    ┃ Metadata               ┃ Expected Output ┃ Outputs         ┃ Scores              ┃ Labels                 ┃ Metrics         ┃ Assertions       ┃    Durations ┃
+┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
+│ test_case │ {'query': 'What is 2+2?'} │ {'difficulty': 'easy'} │ {'answer': '4'} │ {'answer': '4'} │ score1: 2.50        │ label1: hello          │ accuracy: 0.950 │ MockEvaluator: ✔ │  task: 0.100 │
+│           │                           │                        │                 │                 │   Reason: my reason │                        │                 │                  │ total: 0.200 │
+│           │                           │                        │                 │                 │                     │                        │                 │                  │              │
+├───────────┼───────────────────────────┼────────────────────────┼─────────────────┼─────────────────┼─────────────────────┼────────────────────────┼─────────────────┼──────────────────┼──────────────┤
+│ Averages  │                           │                        │                 │                 │ score1: 2.50        │ label1: {'hello': 1.0} │ accuracy: 0.950 │ 100.0% ✔         │  task: 0.100 │
+│           │                           │                        │                 │                 │                     │                        │                 │                  │ total: 0.200 │
+└───────────┴───────────────────────────┴────────────────────────┴─────────────────┴─────────────────┴─────────────────────┴────────────────────────┴─────────────────┴──────────────────┴──────────────┘
+""")
+
+
 async def test_evaluation_renderer_with_baseline(sample_report: EvaluationReport):
     """Test EvaluationRenderer with baseline comparison."""
     baseline_report = EvaluationReport(
@@ -191,20 +229,21 @@ async def test_evaluation_renderer_with_baseline(sample_report: EvaluationReport
         label_configs={},
         metric_configs={},
         duration_config={},
+        include_reasons=False,
     )
 
     table = renderer.build_diff_table(sample_report, baseline_report)
     assert render_table(table) == snapshot("""\
-                                                                                                                               Evaluation Diff: baseline_report → test_report
-┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓
-┃ Case ID   ┃ Inputs                    ┃ Metadata               ┃ Expected Output ┃ Outputs         ┃ Scores       ┃ Labels                                                                              ┃ Metrics                                 ┃ Assertions   ┃                             Durations ┃
-┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩
-│ test_case │ {'query': 'What is 2+2?'} │ {'difficulty': 'easy'} │ {'answer': '4'} │ {'answer': '4'} │ score1: 2.50 │ label1: EvaluationResult(name='MockEvaluator', value='hello', reason=None,          │ accuracy: 0.900 → 0.950 (+0.05 / +5.6%) │  → ✔         │  task: 0.150 → 0.100 (-0.05 / -33.3%) │
-│           │                           │                        │                 │                 │              │ source=mock_evaluator.<locals>.MockEvaluator())                                     │                                         │              │ total: 0.250 → 0.200 (-0.05 / -20.0%) │
-├───────────┼───────────────────────────┼────────────────────────┼─────────────────┼─────────────────┼──────────────┼─────────────────────────────────────────────────────────────────────────────────────┼─────────────────────────────────────────┼──────────────┼───────────────────────────────────────┤
-│ Averages  │                           │                        │                 │                 │ score1: 2.50 │ label1: {'hello': 1.0}                                                              │ accuracy: 0.900 → 0.950 (+0.05 / +5.6%) │ - → 100.0% ✔ │  task: 0.150 → 0.100 (-0.05 / -33.3%) │
-│           │                           │                        │                 │                 │              │                                                                                     │                                         │              │ total: 0.250 → 0.200 (-0.05 / -20.0%) │
-└───────────┴───────────────────────────┴────────────────────────┴─────────────────┴─────────────────┴──────────────┴─────────────────────────────────────────────────────────────────────────────────────┴─────────────────────────────────────────┴──────────────┴───────────────────────────────────────┘
+                                                                                                Evaluation Diff: baseline_report → test_report
+┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓
+┃ Case ID   ┃ Inputs                    ┃ Metadata               ┃ Expected Output ┃ Outputs         ┃ Scores       ┃ Labels                 ┃ Metrics                                 ┃ Assertions   ┃                             Durations ┃
+┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩
+│ test_case │ {'query': 'What is 2+2?'} │ {'difficulty': 'easy'} │ {'answer': '4'} │ {'answer': '4'} │ score1: 2.50 │ label1: hello          │ accuracy: 0.900 → 0.950 (+0.05 / +5.6%) │  → ✔         │  task: 0.150 → 0.100 (-0.05 / -33.3%) │
+│           │                           │                        │                 │                 │              │                        │                                         │              │ total: 0.250 → 0.200 (-0.05 / -20.0%) │
+├───────────┼───────────────────────────┼────────────────────────┼─────────────────┼─────────────────┼──────────────┼────────────────────────┼─────────────────────────────────────────┼──────────────┼───────────────────────────────────────┤
+│ Averages  │                           │                        │                 │                 │ score1: 2.50 │ label1: {'hello': 1.0} │ accuracy: 0.900 → 0.950 (+0.05 / +5.6%) │ - → 100.0% ✔ │  task: 0.150 → 0.100 (-0.05 / -33.3%) │
+│           │                           │                        │                 │                 │              │                        │                                         │              │ total: 0.250 → 0.200 (-0.05 / -20.0%) │
+└───────────┴───────────────────────────┴────────────────────────┴─────────────────┴─────────────────┴──────────────┴────────────────────────┴─────────────────────────────────────────┴──────────────┴───────────────────────────────────────┘
 """)
 
 
@@ -248,6 +287,7 @@ async def test_evaluation_renderer_with_removed_cases(sample_report: EvaluationR
         label_configs={},
         metric_configs={},
         duration_config={},
+        include_reasons=False,
     )
 
     table = renderer.build_diff_table(sample_report, baseline_report)
@@ -311,6 +351,7 @@ async def test_evaluation_renderer_with_custom_configs(sample_report: Evaluation
             'diff_increase_style': 'bold red',
             'diff_decrease_style': 'bold green',
         },
+        include_reasons=False,
     )
 
     table = renderer.build_table(sample_report)
@@ -350,7 +391,7 @@ async def test_report_case_aggregate_average():
                     name='MockEvaluator',
                     value=0.8,
                     reason=None,
-                    source=MockEvaluator(),
+                    source=MockEvaluator().as_spec(),
                 )
             },
             labels={
@@ -358,7 +399,7 @@ async def test_report_case_aggregate_average():
                     name='MockEvaluator',
                     value='good',
                     reason=None,
-                    source=MockEvaluator(),
+                    source=MockEvaluator().as_spec(),
                 )
             },
             assertions={
@@ -366,7 +407,7 @@ async def test_report_case_aggregate_average():
                     name='MockEvaluator',
                     value=True,
                     reason=None,
-                    source=MockEvaluator(),
+                    source=MockEvaluator().as_spec(),
                 )
             },
             task_duration=0.1,
@@ -387,7 +428,7 @@ async def test_report_case_aggregate_average():
                     name='MockEvaluator',
                     value=0.7,
                     reason=None,
-                    source=MockEvaluator(),
+                    source=MockEvaluator().as_spec(),
                 )
             },
             labels={
@@ -395,7 +436,7 @@ async def test_report_case_aggregate_average():
                     name='MockEvaluator',
                     value='good',
                     reason=None,
-                    source=MockEvaluator(),
+                    source=MockEvaluator().as_spec(),
                 )
             },
             assertions={
@@ -403,7 +444,7 @@ async def test_report_case_aggregate_average():
                     name='MockEvaluator',
                     value=False,
                     reason=None,
-                    source=MockEvaluator(),
+                    source=MockEvaluator().as_spec(),
                 )
             },
             task_duration=0.15,

{pydantic_ai-0.4.11 → pydantic_ai-0.5.0}/tests/evals/test_reports.py

@@ -57,7 +57,7 @@ def sample_evaluation_result(
         name='MockEvaluator',
         value=True,
         reason=None,
-        source=mock_evaluator,
+        source=mock_evaluator.as_spec(),
     )
 
 
@@ -177,7 +177,7 @@ async def test_report_with_error(mock_evaluator: Evaluator[TaskInput, TaskOutput
         name='error_evaluator',
         value=False,  # No result
         reason='Test error message',
-        source=mock_evaluator,
+        source=mock_evaluator.as_spec(),
     )
 
     # Create a case