test(viewer): add comprehensive test suite (82 tests)

- test_data.py: Data models (ExecutionStep, TaskExecution, BenchmarkRun)
- test_generator.py: HTML generation, validation, XSS prevention
- test_cli.py: CLI commands (demo, benchmark)
- conftest.py: Shared fixtures

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

Author: abrichr

tests/__init__.py

@@ -0,0 +1 @@
+"""Tests for openadapt-viewer package."""

tests/conftest.py

@@ -0,0 +1,202 @@
+"""Shared pytest fixtures for openadapt-viewer tests."""
+
+import pytest
+from datetime import datetime, timedelta
+from pathlib import Path
+import tempfile
+import json
+
+from openadapt_viewer.core.types import (
+    BenchmarkRun,
+    BenchmarkTask,
+    TaskExecution,
+    ExecutionStep,
+)
+
+
+@pytest.fixture
+def sample_execution_step():
+    """Create a sample ExecutionStep for testing."""
+    return ExecutionStep(
+        step_number=0,
+        timestamp=datetime.now(),
+        screenshot_path="tasks/task_001/screenshots/step_000.png",
+        action_type="click",
+        action_details={"x": 500, "y": 300},
+        reasoning="Clicking on the target button",
+        raw_output="Action: CLICK(500, 300)",
+    )
+
+
+@pytest.fixture
+def sample_task_execution(sample_execution_step):
+    """Create a sample TaskExecution for testing."""
+    steps = [
+        sample_execution_step,
+        ExecutionStep(
+            step_number=1,
+            timestamp=datetime.now() + timedelta(seconds=2),
+            action_type="type",
+            action_details={"text": "Hello World"},
+            reasoning="Typing the required text",
+        ),
+        ExecutionStep(
+            step_number=2,
+            timestamp=datetime.now() + timedelta(seconds=4),
+            action_type="click",
+            action_details={"x": 800, "y": 600},
+            reasoning="Clicking the submit button",
+        ),
+    ]
+    return TaskExecution(
+        task_id="task_001",
+        start_time=datetime.now(),
+        end_time=datetime.now() + timedelta(seconds=10),
+        steps=steps,
+        success=True,
+        error=None,
+    )
+
+
+@pytest.fixture
+def failed_task_execution():
+    """Create a failed TaskExecution for testing."""
+    return TaskExecution(
+        task_id="task_002",
+        start_time=datetime.now(),
+        end_time=datetime.now() + timedelta(seconds=5),
+        steps=[
+            ExecutionStep(
+                step_number=0,
+                action_type="click",
+                action_details={"x": 100, "y": 100},
+                reasoning="Attempting to click target",
+            )
+        ],
+        success=False,
+        error="Task failed: Could not locate target element",
+    )
+
+
+@pytest.fixture
+def sample_benchmark_task():
+    """Create a sample BenchmarkTask for testing."""
+    return BenchmarkTask(
+        task_id="task_001",
+        instruction="Open Notepad and type 'Hello World'",
+        domain="office",
+        difficulty="easy",
+        time_limit=300,
+        metadata={"source": "test"},
+    )
+
+
+@pytest.fixture
+def sample_benchmark_run(sample_benchmark_task, sample_task_execution, failed_task_execution):
+    """Create a sample BenchmarkRun with multiple tasks for testing."""
+    tasks = [
+        sample_benchmark_task,
+        BenchmarkTask(
+            task_id="task_002",
+            instruction="Navigate to google.com in Chrome",
+            domain="browser",
+            difficulty="easy",
+            time_limit=300,
+        ),
+        BenchmarkTask(
+            task_id="task_003",
+            instruction="Create a new folder on Desktop",
+            domain="file_management",
+            difficulty="medium",
+            time_limit=300,
+        ),
+    ]
+
+    # Add a third successful execution
+    third_execution = TaskExecution(
+        task_id="task_003",
+        start_time=datetime.now(),
+        end_time=datetime.now() + timedelta(seconds=8),
+        steps=[
+            ExecutionStep(
+                step_number=0,
+                action_type="click",
+                action_details={"x": 200, "y": 400},
+            ),
+            ExecutionStep(
+                step_number=1,
+                action_type="type",
+                action_details={"text": "New Folder"},
+            ),
+        ],
+        success=True,
+    )
+
+    executions = [sample_task_execution, failed_task_execution, third_execution]
+
+    return BenchmarkRun(
+        run_id="test_run_001",
+        benchmark_name="Test Benchmark",
+        model_id="test-model-v1",
+        start_time=datetime.now() - timedelta(hours=1),
+        end_time=datetime.now(),
+        tasks=tasks,
+        executions=executions,
+        config={"max_steps": 10, "timeout": 300},
+    )
+
+
+@pytest.fixture
+def temp_dir():
+    """Create a temporary directory for test files."""
+    with tempfile.TemporaryDirectory() as tmpdir:
+        yield Path(tmpdir)
+
+
+@pytest.fixture
+def benchmark_data_dir(temp_dir):
+    """Create a mock benchmark data directory with test data."""
+    # Create directory structure
+    tasks_dir = temp_dir / "tasks"
+    task_001_dir = tasks_dir / "task_001"
+    screenshots_dir = task_001_dir / "screenshots"
+    screenshots_dir.mkdir(parents=True)
+
+    # Create metadata.json
+    metadata = {
+        "run_id": "test_run",
+        "benchmark_name": "Test Benchmark",
+        "model_id": "test-model",
+        "start_time": datetime.now().isoformat(),
+        "config": {"max_steps": 10},
+    }
+    with open(temp_dir / "metadata.json", "w") as f:
+        json.dump(metadata, f)
+
+    # Create task.json
+    task_data = {
+        "task_id": "task_001",
+        "instruction": "Test task instruction",
+        "domain": "test",
+        "difficulty": "easy",
+    }
+    with open(task_001_dir / "task.json", "w") as f:
+        json.dump(task_data, f)
+
+    # Create execution.json
+    execution_data = {
+        "task_id": "task_001",
+        "start_time": datetime.now().isoformat(),
+        "steps": [
+            {
+                "action_type": "click",
+                "action_details": {"x": 100, "y": 200},
+                "reasoning": "Test reasoning",
+            }
+        ],
+        "success": True,
+    }
+    with open(task_001_dir / "execution.json", "w") as f:
+        json.dump(execution_data, f)
+
+    return temp_dir