feat: Add comprehensive Python testing infrastructure with Poetry

.gitignore

@@ -119,3 +119,9 @@ venv.bak/
 
 # mypy
 .mypy_cache/
+
+# Claude
+.claude/*
+
+# Poetry
+poetry.lock

pyproject.toml

@@ -0,0 +1,97 @@
+[tool.poetry]
+name = "pdvc"
+version = "0.1.0"
+description = "Parallel Decoding for Video Captioning"
+authors = ["PDVC Team"]
+readme = "README.md"
+packages = [{include = "pdvc"}, {include = "video_backbone"}, {include = "data"}, {include = "misc"}, {include = "densevid_eval3"}, {include = "visualization"}]
+
+[tool.poetry.dependencies]
+python = "^3.8"
+h5py = "*"
+matplotlib = "*"
+numpy = "*"
+pandas = "*"
+Pillow = "*"
+PyYAML = "*"
+six = "*"
+tqdm = "*"
+tensorboardX = "*"
+colorlog = "*"
+scipy = "*"
+notebook = "*"
+av = "*"
+joblib = "*"
+google-trans-new = "*"
+torch = "*"
+
+[tool.poetry.group.dev.dependencies]
+pytest = "^7.4.0"
+pytest-cov = "^4.1.0"
+pytest-mock = "^3.11.1"
+
+[tool.poetry.scripts]
+test = "pytest:main"
+tests = "pytest:main"
+
+[tool.pytest.ini_options]
+minversion = "7.0"
+testpaths = ["tests"]
+python_files = ["test_*.py", "*_test.py"]
+python_classes = ["Test*"]
+python_functions = ["test_*"]
+addopts = [
+    "--strict-markers",
+    "--strict-config",
+    "--verbose",
+    "--cov=pdvc",
+    "--cov=video_backbone",
+    "--cov=data",
+    "--cov=misc",
+    "--cov=densevid_eval3",
+    "--cov=visualization",
+    "--cov-branch",
+    "--cov-report=term-missing:skip-covered",
+    "--cov-report=html",
+    "--cov-report=xml",
+    "--cov-fail-under=80",
+]
+markers = [
+    "unit: Unit tests",
+    "integration: Integration tests",
+    "slow: Slow tests",
+]
+
+[tool.coverage.run]
+source = ["pdvc", "video_backbone", "data", "misc", "densevid_eval3", "visualization"]
+omit = [
+    "*/tests/*",
+    "*/test_*.py",
+    "*/__pycache__/*",
+    "*/site-packages/*",
+    "setup.py",
+    "*/pdvc/ops/src/*",
+    "*/pdvc/ops/test.py",
+]
+
+[tool.coverage.report]
+exclude_lines = [
+    "pragma: no cover",
+    "def __repr__",
+    "if self.debug:",
+    "if __name__ == .__main__.:",
+    "raise AssertionError",
+    "raise NotImplementedError",
+    "if 0:",
+    "if False:",
+    "pass",
+]
+precision = 2
+show_missing = true
+
+[tool.coverage.html]
+directory = "htmlcov"
+
+[build-system]
+requires = ["poetry-core"]
+build-backend = "poetry.core.masonry.api"

tests/conftest.py

@@ -0,0 +1,164 @@
+import os
+import tempfile
+import shutil
+from pathlib import Path
+from unittest.mock import Mock
+
+import pytest
+import torch
+import numpy as np
+import yaml
+
+
+@pytest.fixture
+def temp_dir():
+    """Create a temporary directory for test files."""
+    temp_path = tempfile.mkdtemp()
+    yield Path(temp_path)
+    shutil.rmtree(temp_path)
+
+
+@pytest.fixture
+def mock_config():
+    """Create a mock configuration dictionary for testing."""
+    return {
+        'train': {
+            'batch_size': 32,
+            'learning_rate': 1e-4,
+            'num_epochs': 10,
+            'grad_clip': 1.0,
+            'weight_decay': 1e-4,
+        },
+        'model': {
+            'visual_feature_dim': 2048,
+            'hidden_dim': 512,
+            'num_heads': 8,
+            'num_layers': 6,
+            'dropout': 0.1,
+        },
+        'data': {
+            'num_workers': 4,
+            'max_caption_len': 50,
+            'video_feature_path': '/data/features',
+        },
+        'eval': {
+            'beam_size': 5,
+            'max_pred_length': 50,
+        }
+    }
+
+
+@pytest.fixture
+def sample_video_features():
+    """Create sample video features for testing."""
+    batch_size = 2
+    num_frames = 10
+    feature_dim = 2048
+
+    features = torch.randn(batch_size, num_frames, feature_dim)
+    mask = torch.ones(batch_size, num_frames, dtype=torch.bool)
+
+    return {
+        'features': features,
+        'mask': mask,
+        'video_ids': ['video_001', 'video_002'],
+    }
+
+
+@pytest.fixture
+def sample_captions():
+    """Create sample captions for testing."""
+    return [
+        {
+            'video_id': 'video_001',
+            'captions': [
+                {'sentence': 'A person walks into the room', 'timestamp': [0.0, 3.5]},
+                {'sentence': 'They sit down on a chair', 'timestamp': [3.5, 7.0]},
+            ]
+        },
+        {
+            'video_id': 'video_002',
+            'captions': [
+                {'sentence': 'Someone opens a door', 'timestamp': [0.0, 2.0]},
+                {'sentence': 'They turn on the lights', 'timestamp': [2.0, 4.5]},
+            ]
+        }
+    ]
+
+
+@pytest.fixture
+def mock_model():
+    """Create a mock model for testing."""
+    model = Mock()
+    model.train = Mock(return_value=None)
+    model.eval = Mock(return_value=None)
+    model.parameters = Mock(return_value=[torch.randn(10, 10)])
+    model.forward = Mock(return_value={
+        'loss': torch.tensor(1.0),
+        'predictions': torch.randn(2, 10, 100)
+    })
+    return model
+
+
+@pytest.fixture
+def mock_dataloader():
+    """Create a mock dataloader for testing."""
+    dataloader = Mock()
+    dataloader.__iter__ = Mock(return_value=iter([
+        {
+            'video_features': torch.randn(2, 10, 2048),
+            'captions': ['caption1', 'caption2'],
+            'video_ids': ['vid1', 'vid2']
+        }
+    ]))
+    dataloader.__len__ = Mock(return_value=10)
+    return dataloader
+
+
+@pytest.fixture
+def sample_yaml_config(temp_dir):
+    """Create a sample YAML configuration file."""
+    config_path = temp_dir / 'config.yaml'
+    config = {
+        'experiment_name': 'test_experiment',
+        'model_type': 'pdvc',
+        'dataset': 'activitynet',
+        'checkpoint_dir': str(temp_dir / 'checkpoints'),
+        'log_dir': str(temp_dir / 'logs'),
+    }
+
+    with open(config_path, 'w') as f:
+        yaml.dump(config, f)
+
+    return config_path
+
+
+@pytest.fixture
+def mock_h5_file(temp_dir):
+    """Create a mock HDF5 file structure for testing."""
+    import h5py
+
+    h5_path = temp_dir / 'features.h5'
+    with h5py.File(h5_path, 'w') as f:
+        # Add some sample video features
+        for i in range(5):
+            video_id = f'video_{i:03d}'
+            features = np.random.randn(20, 2048).astype(np.float32)
+            f.create_dataset(video_id, data=features)
+
+    return h5_path
+
+
+@pytest.fixture
+def device():
+    """Get the appropriate device for testing."""
+    return torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+
+
+@pytest.fixture(autouse=True)
+def reset_random_seeds():
+    """Reset random seeds before each test for reproducibility."""
+    np.random.seed(42)
+    torch.manual_seed(42)
+    if torch.cuda.is_available():
+        torch.cuda.manual_seed_all(42)