feat: Set up Python testing infrastructure with Poetry and pytest

.gitignore

@@ -1,2 +1,55 @@
 *.pyc 
 *.bin
+
+# Testing related
+.pytest_cache/
+.coverage
+htmlcov/
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+pytest_cache/
+
+# Claude settings
+.claude/*
+
+# Virtual environments
+venv/
+.venv/
+env/
+.env/
+ENV/
+env.bak/
+venv.bak/
+
+# Poetry
+# Note: poetry.lock should be committed to ensure reproducible builds
+
+# Build artifacts
+build/
+dist/
+*.egg-info/
+.eggs/
+__pycache__/
+*.pyo
+*.pyd
+
+# IDE files
+.idea/
+.vscode/
+*.swp
+*.swo
+*~
+.DS_Store
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/

pyproject.toml

@@ -0,0 +1,84 @@
+[tool.poetry]
+name = "pointer-summarizer"
+version = "0.1.0"
+description = "PyTorch implementation of Get To The Point: Summarization with Pointer-Generator Networks"
+authors = ["Your Name <[email protected]>"]
+readme = "README.md"
+license = "MIT"
+packages = [
+    { include = "data_util" },
+    { include = "training_ptr_gen" }
+]
+
+[tool.poetry.dependencies]
+python = "^3.8"
+torch = ">=1.0.0"
+numpy = "*"
+# pyrouge = "*"  # Note: pyrouge needs to be installed separately following its instructions
+
+[tool.poetry.group.dev.dependencies]
+pytest = "^7.4.0"
+pytest-cov = "^4.1.0"
+pytest-mock = "^3.11.0"
+
+[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 = [
+    "-ra",
+    "--strict-markers",
+    "--cov=data_util",
+    "--cov=training_ptr_gen",
+    "--cov-report=html",
+    "--cov-report=xml",
+    "--cov-report=term-missing",
+    "--cov-fail-under=0",  # Set to 0 for initial setup, increase when adding real tests
+    "-v",
+    "--tb=short"
+]
+markers = [
+    "unit: marks tests as unit tests (fast)",
+    "integration: marks tests as integration tests (slower)",
+    "slow: marks tests as slow running"
+]
+
+[tool.coverage.run]
+source = ["data_util", "training_ptr_gen"]
+omit = [
+    "*/tests/*",
+    "*/__init__.py",
+    "*/setup.py",
+    "*/venv/*",
+    "*/.venv/*",
+    "data_util/data.py"  # Python 2 syntax file, needs migration
+]
+
+[tool.coverage.report]
+precision = 2
+show_missing = true
+skip_covered = false
+exclude_lines = [
+    "pragma: no cover",
+    "def __repr__",
+    "raise AssertionError",
+    "raise NotImplementedError",
+    "if __name__ == .__main__.:",
+    "if TYPE_CHECKING:"
+]
+
+[tool.coverage.html]
+directory = "htmlcov"
+
+[tool.coverage.xml]
+output = "coverage.xml"
+
+[build-system]
+requires = ["poetry-core>=1.0.0"]
+build-backend = "poetry.core.masonry.api"

tests/conftest.py

@@ -0,0 +1,156 @@
+"""Shared pytest fixtures and configuration for all tests."""
+
+import os
+import shutil
+import tempfile
+from pathlib import Path
+from typing import Generator, Dict, Any
+
+import pytest
+import torch
+
+
+@pytest.fixture
+def temp_dir() -> Generator[Path, None, None]:
+    """Create a temporary directory for test files."""
+    temp_path = tempfile.mkdtemp()
+    yield Path(temp_path)
+    shutil.rmtree(temp_path, ignore_errors=True)
+
+
+@pytest.fixture
+def mock_config() -> Dict[str, Any]:
+    """Provide a mock configuration dictionary for testing."""
+    return {
+        "hidden_dim": 256,
+        "emb_dim": 128,
+        "batch_size": 8,
+        "max_enc_steps": 400,
+        "max_dec_steps": 100,
+        "beam_size": 4,
+        "min_dec_steps": 35,
+        "vocab_size": 50000,
+        "lr": 0.15,
+        "adagrad_init_acc": 0.1,
+        "rand_unif_init_mag": 0.02,
+        "trunc_norm_init_std": 1e-4,
+        "max_grad_norm": 2.0,
+        "pointer_gen": True,
+        "is_coverage": True,
+        "cov_loss_wt": 1.0,
+        "eps": 1e-12,
+    }
+
+
+@pytest.fixture
+def sample_vocab() -> Dict[str, Any]:
+    """Create a sample vocabulary for testing."""
+    vocab_dict = {
+        "[PAD]": 0,
+        "[UNK]": 1,
+        "[START]": 2,
+        "[STOP]": 3,
+        "the": 4,
+        "a": 5,
+        "is": 6,
+        "was": 7,
+        "are": 8,
+        "were": 9,
+    }
+
+    class MockVocab:
+        def __init__(self, vocab_dict):
+            self._word_to_id = vocab_dict
+            self._id_to_word = {v: k for k, v in vocab_dict.items()}
+
+        def word2id(self, word):
+            return self._word_to_id.get(word, 1)  # Return UNK id for unknown words
+
+        def id2word(self, idx):
+            return self._id_to_word.get(idx, "[UNK]")
+
+        def size(self):
+            return len(self._word_to_id)
+
+    return MockVocab(vocab_dict)
+
+
+@pytest.fixture
+def sample_batch_data() -> Dict[str, Any]:
+    """Create sample batch data for model testing."""
+    batch_size = 4
+    max_enc_steps = 50
+    max_dec_steps = 20
+
+    return {
+        "enc_batch": torch.randint(0, 100, (batch_size, max_enc_steps)),
+        "enc_padding_mask": torch.ones(batch_size, max_enc_steps),
+        "enc_lens": torch.tensor([45, 48, 50, 42]),
+        "enc_batch_extend_vocab": torch.randint(0, 150, (batch_size, max_enc_steps)),
+        "extra_zeros": torch.zeros((batch_size, 50)),
+        "c_t_1": torch.zeros((batch_size, 2 * 256)),  # 2 * hidden_dim for bidirectional
+        "coverage": torch.zeros(batch_size, max_enc_steps),
+        "dec_batch": torch.randint(0, 100, (batch_size, max_dec_steps)),
+        "target_batch": torch.randint(0, 100, (batch_size, max_dec_steps)),
+        "dec_padding_mask": torch.ones(batch_size, max_dec_steps),
+        "max_art_oovs": 10,
+    }
+
+
+@pytest.fixture
+def device():
+    """Return 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."""
+    import random
+    import numpy as np
+
+    random.seed(42)
+    np.random.seed(42)
+    torch.manual_seed(42)
+    if torch.cuda.is_available():
+        torch.cuda.manual_seed_all(42)
+
+
+@pytest.fixture
+def mock_data_path(temp_dir: Path) -> Path:
+    """Create mock data directory structure."""
+    data_path = temp_dir / "data"
+    data_path.mkdir()
+
+    # Create subdirectories
+    (data_path / "finished_files").mkdir()
+    (data_path / "chunked").mkdir()
+
+    # Create a mock vocab file
+    vocab_path = data_path / "finished_files" / "vocab"
+    vocab_words = ["[PAD]", "[UNK]", "[START]", "[STOP]", "the", "a", "is", "was"]
+    vocab_path.write_text("\n".join(vocab_words))
+
+    return data_path
+
+
+@pytest.fixture
+def mock_model_path(temp_dir: Path) -> Path:
+    """Create mock model directory."""
+    model_path = temp_dir / "models"
+    model_path.mkdir()
+    return model_path
+
+
+@pytest.fixture
+def mock_log_path(temp_dir: Path) -> Path:
+    """Create mock log directory."""
+    log_path = temp_dir / "logs"
+    log_path.mkdir()
+    return log_path
+
+
+@pytest.mark.timeout(60)
+class TimeoutMarker:
+    """Marker for tests with custom timeout."""
+    pass