pydantic-ai 0.4.0__tar.gz → 0.4.2__tar.gz

{pydantic_ai-0.4.0 → pydantic_ai-0.4.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pydantic-ai
-Version: 0.4.0
+Version: 0.4.2
 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
@@ -28,9 +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[a2a,anthropic,bedrock,cli,cohere,evals,google,groq,mcp,mistral,openai,vertexai]==0.4.0
+Requires-Dist: pydantic-ai-slim[anthropic,bedrock,cli,cohere,evals,google,groq,mcp,mistral,openai,vertexai]==0.4.2
+Provides-Extra: a2a
+Requires-Dist: fasta2a>=0.4.1; extra == 'a2a'
 Provides-Extra: examples
-Requires-Dist: pydantic-ai-examples==0.4.0; extra == 'examples'
+Requires-Dist: pydantic-ai-examples==0.4.2; extra == 'examples'
 Provides-Extra: logfire
 Requires-Dist: logfire>=3.11.0; extra == 'logfire'
 Description-Content-Type: text/markdown

{pydantic_ai-0.4.0 → pydantic_ai-0.4.2}/pyproject.toml

@@ -46,12 +46,13 @@ requires-python = ">=3.9"
 
 [tool.hatch.metadata.hooks.uv-dynamic-versioning]
 dependencies = [
-    "pydantic-ai-slim[openai,vertexai,google,groq,anthropic,mistral,cohere,bedrock,cli,mcp,evals,a2a]=={{ version }}",
+    "pydantic-ai-slim[openai,vertexai,google,groq,anthropic,mistral,cohere,bedrock,cli,mcp,evals]=={{ version }}",
 ]
 
 [tool.hatch.metadata.hooks.uv-dynamic-versioning.optional-dependencies]
 examples = ["pydantic-ai-examples=={{ version }}"]
 logfire = ["logfire>=3.11.0"]
+a2a = ["fasta2a>=0.4.1"]
 
 [project.urls]
 Homepage = "https://ai.pydantic.dev"
@@ -67,7 +68,6 @@ pydantic-ai-slim = { workspace = true }
 pydantic-evals = { workspace = true }
 pydantic-graph = { workspace = true }
 pydantic-ai-examples = { workspace = true }
-fasta2a = { workspace = true }
 
 [tool.uv.workspace]
 members = [
@@ -77,13 +77,13 @@ members = [
     "mcp-run-python",
     "clai",
     "examples",
-    "fasta2a",
 ]
 
 [dependency-groups]
 # dev dependencies are defined in `pydantic-ai-slim/pyproject.toml` to allow for minimal testing
 lint = ["mypy>=1.11.2", "pyright>=1.1.390", "ruff>=0.6.9"]
 docs = [
+    "pydantic-ai[a2a]",
     "black>=24.10.0",
     "mkdocs>=1.6.1",
     "mkdocs-glightbox>=0.4.0",
@@ -107,7 +107,6 @@ include = [
     "pydantic_evals/**/*.py",
     "pydantic_graph/**/*.py",
     "mcp-run-python/**/*.py",
-    "fasta2a/**/*.py",
     "examples/**/*.py",
     "clai/**/*.py",
     "tests/**/*.py",
@@ -126,7 +125,6 @@ extend-select = [
     "TID251",
 ]
 flake8-quotes = { inline-quotes = "single", multiline-quotes = "double" }
-isort = { combine-as-imports = true, known-first-party = ["pydantic_ai"] }
 mccabe = { max-complexity = 15 }
 ignore = [
     "D100", # ignore missing docstring in module
@@ -136,6 +134,12 @@ ignore = [
     "D107", # ignore missing docstring in __init__ methods
 ]
 
+[tool.ruff.lint.isort]
+combine-as-imports = true
+known-first-party = ["pydantic_ai"]
+# weird issue with ruff thinking fasta2a is still editable
+known-third-party = ["fasta2a"]
+
 [tool.ruff.lint.pydocstyle]
 convention = "google"
 
@@ -165,7 +169,6 @@ include = [
     "pydantic_evals",
     "pydantic_graph",
     "mcp-run-python",
-    "fasta2a",
     "tests",
     "examples",
     "clai",
@@ -189,10 +192,7 @@ files = "tests/typed_agent.py,tests/typed_graph.py"
 strict = true
 
 [tool.pytest.ini_options]
-testpaths = [
-    "tests",
-    "docs/.hooks"
-]
+testpaths = ["tests", "docs/.hooks"]
 xfail_strict = true
 filterwarnings = [
     "error",
@@ -216,8 +216,6 @@ include = [
     "pydantic_ai_slim/**/*.py",
     "pydantic_evals/**/*.py",
     "pydantic_graph/**/*.py",
-    # TODO(Marcelo): Add 100% coverage for A2A.
-    # "fasta2a/**/*.py",
     "tests/**/*.py",
 ]
 omit = [

{pydantic_ai-0.4.0 → pydantic_ai-0.4.2}/tests/evals/test_dataset.py

@@ -1,6 +1,5 @@
 from __future__ import annotations as _annotations
 
-import asyncio
 import json
 import sys
 from dataclasses import dataclass
@@ -8,7 +7,7 @@ from pathlib import Path
 from typing import Any
 
 import pytest
-from dirty_equals import HasRepr
+from dirty_equals import HasRepr, IsNumber
 from inline_snapshot import snapshot
 from pydantic import BaseModel
 
@@ -178,21 +177,21 @@ async def test_add_evaluator(
     }
 
 
-async def test_evaluate(
+async def test_evaluate_async(
     example_dataset: Dataset[TaskInput, TaskOutput, TaskMetadata],
     simple_evaluator: type[Evaluator[TaskInput, TaskOutput, TaskMetadata]],
 ):
     """Test evaluating a dataset."""
     example_dataset.add_evaluator(simple_evaluator())
 
-    async def mock_task(inputs: TaskInput) -> TaskOutput:
+    async def mock_async_task(inputs: TaskInput) -> TaskOutput:
         if inputs.query == 'What is 2+2?':
             return TaskOutput(answer='4')
         elif inputs.query == 'What is the capital of France?':
             return TaskOutput(answer='Paris')
         return TaskOutput(answer='Unknown')  # pragma: no cover
 
-    report = await example_dataset.evaluate(mock_task)
+    report = await example_dataset.evaluate(mock_async_task)
 
     assert report is not None
     assert len(report.cases) == 2
@@ -230,6 +229,58 @@ async def test_evaluate(
     )
 
 
+async def test_evaluate_sync(
+    example_dataset: Dataset[TaskInput, TaskOutput, TaskMetadata],
+    simple_evaluator: type[Evaluator[TaskInput, TaskOutput, TaskMetadata]],
+):
+    """Test evaluating a dataset."""
+    example_dataset.add_evaluator(simple_evaluator())
+
+    def mock_sync_task(inputs: TaskInput) -> TaskOutput:
+        if inputs.query == 'What is 2+2?':
+            return TaskOutput(answer='4')
+        elif inputs.query == 'What is the capital of France?':
+            return TaskOutput(answer='Paris')
+        return TaskOutput(answer='Unknown')  # pragma: no cover
+
+    report = await example_dataset.evaluate(mock_sync_task)
+
+    assert report is not None
+    assert len(report.cases) == 2
+    assert ReportCaseAdapter.dump_python(report.cases[0]) == snapshot(
+        {
+            'assertions': {
+                'correct': {
+                    'name': 'correct',
+                    'reason': None,
+                    'source': {'name': 'SimpleEvaluator', 'arguments': None},
+                    'value': True,
+                }
+            },
+            'attributes': {},
+            'expected_output': {'answer': '4', 'confidence': 1.0},
+            'inputs': {'query': 'What is 2+2?'},
+            'labels': {},
+            'metadata': {'category': 'general', 'difficulty': 'easy'},
+            'metrics': {},
+            'name': 'case1',
+            'output': {'answer': '4', 'confidence': 1.0},
+            'scores': {
+                'confidence': {
+                    'name': 'confidence',
+                    'reason': None,
+                    'source': {'name': 'SimpleEvaluator', 'arguments': None},
+                    'value': 1.0,
+                }
+            },
+            'span_id': '0000000000000003',
+            'task_duration': IsNumber(),  # the runtime behavior is not deterministic due to threading
+            'total_duration': IsNumber(),  # the runtime behavior is not deterministic due to threading
+            'trace_id': '00000000000000000000000000000001',
+        }
+    )
+
+
 async def test_evaluate_with_concurrency(
     example_dataset: Dataset[TaskInput, TaskOutput, TaskMetadata],
     simple_evaluator: type[Evaluator[TaskInput, TaskOutput, TaskMetadata]],
@@ -828,8 +879,8 @@ async def test_dataset_evaluate_with_sync_task(example_dataset: Dataset[TaskInpu
     def sync_task(inputs: TaskInput) -> TaskOutput:
         return TaskOutput(answer=inputs.query.upper())
 
-    report = await example_dataset.evaluate(lambda x: asyncio.sleep(0, sync_task(x)))
-    assert report.name == '<lambda>'
+    report = await example_dataset.evaluate(sync_task)
+    assert report.name == 'sync_task'
     assert len(report.cases) == 2