🔍 Code Extractor

Integration test function that validates the AgentExecutor's ability to generate and execute data analysis projects using synthetic test data.

Purpose

This test function validates the complete workflow of the AgentExecutor system by: (1) creating a synthetic pandas DataFrame with various data types and missing values, (2) generating an analysis project based on a user query and data summary, and (3) executing the generated project to produce analysis outputs. It serves as both a functional test and a demonstration of the agent executor's capabilities for automated data analysis.

Source Code

def test_agent_executor():
    """Test the agent executor with a simple analysis"""
    
    # Initialize config and agent
    app_config = config['development']()
    agent = AgentExecutor(app_config)
    
    # Create test data summary
    test_data_summary = {
        'shape': (100, 5),
        'column_info': {
            'A': {'type': 'float64', 'non_null': 100, 'unique': 95},
            'B': {'type': 'float64', 'non_null': 100, 'unique': 88},
            'C': {'type': 'int64', 'non_null': 100, 'unique': 3},
            'D': {'type': 'object', 'non_null': 100, 'unique': 2},
            'E': {'type': 'float64', 'non_null': 95, 'unique': 85}
        }
    }
    
    test_session_id = "debug_session_001"
    test_query = "Generate descriptive statistics and basic plots for this dataset"
    
    print(f"Testing agent executor with session {test_session_id}")
    print(f"Query: {test_query}")
    print(f"Data summary: {test_data_summary}")
    print("-" * 50)
    
    # Create test data
    import pandas as pd
    import numpy as np
    
    # Generate test dataframe
    np.random.seed(42)
    test_df = pd.DataFrame({
        'A': np.random.normal(100, 15, 100),
        'B': np.random.normal(50, 10, 100),
        'C': np.random.choice([1, 2, 3], 100),
        'D': np.random.choice(['Group1', 'Group2'], 100),
        'E': np.random.normal(25, 5, 100)
    })
    
    # Add some missing values
    test_df.loc[95:99, 'E'] = np.nan
    
    # Test project generation
    print("Step 1: Generating analysis project...")
    project_result = agent.generate_analysis_project(
        session_id=test_session_id,
        user_query=test_query,
        data_summary=test_data_summary,
        analysis_config=None,
        session_data=test_df  # Pass the test dataframe
    )
    
    print(f"Project generation result: {project_result}")
    
    if project_result['success']:
        print(f"Project created at: {project_result['project_dir']}")
        
        # Test project execution
        print("\nStep 2: Executing analysis project...")
        execution_result = agent.execute_analysis_project(
            project_dir=project_result['project_dir'],
            max_iterations=2
        )
        
        print(f"Execution result: {execution_result}")
        
        # Check generated files
        project_path = Path(project_result['project_dir'])
        print(f"\nFiles in project directory {project_path}:")
        if project_path.exists():
            for file_path in project_path.rglob('*'):
                if file_path.is_file():
                    print(f"  {file_path.relative_to(project_path)} ({file_path.stat().st_size} bytes)")
        
    else:
        print(f"Project generation failed: {project_result.get('error', 'Unknown error')}")

Return Value

This function does not return any value (implicitly returns None). It prints status messages and results to stdout throughout execution, including project generation results, execution results, and a listing of generated files in the project directory.

Dependencies

• pandas
• numpy
• pathlib
• logging
• os
• sys

Required Imports

import os
import sys
import logging
from pathlib import Path
from config import config
from agent_executor import AgentExecutor
from models import DatabaseManager
import pandas as pd
import numpy as np

Usage Example

# Ensure config.py is properly set up with development configuration
# from config import config
# from agent_executor import AgentExecutor
# from models import DatabaseManager

# Run the test
test_agent_executor()

# Expected output:
# - Prints test session information
# - Creates a test DataFrame with 100 rows and 5 columns
# - Generates an analysis project in a directory
# - Executes the project with max 2 iterations
# - Lists all generated files with their sizes

Best Practices

• This is a test function and should be run in a development or testing environment, not in production
• Ensure sufficient disk space is available for generated analysis projects and outputs
• The function uses a fixed random seed (42) for reproducibility of test data
• Monitor the output directory for accumulated test projects that may need cleanup
• The max_iterations parameter is set to 2 for testing; adjust based on analysis complexity needs
• Verify that the config['development']() returns a valid configuration object before running
• The test creates a session with ID 'debug_session_001' which may conflict with existing sessions
• Consider wrapping this test in a try-except block to handle potential failures gracefully
• The function intentionally creates missing values in column 'E' (rows 95-99) to test handling of incomplete data

Tags

Similar Components

AI-powered semantic similarity - components with related functionality:

• function test_enhanced_workflow 67.3% similar

  A comprehensive test function that validates the EnhancedSQLWorkflow system by testing component initialization, request parsing, and data assessment capabilities.

  From: /tf/active/vicechatdev/full_smartstat/test_enhanced_workflow.py

• function main_v25 61.9% similar

  Orchestrates and executes a comprehensive test suite for the Vice AI Data Analysis Integration, running multiple test functions, creating test datasets, and providing detailed pass/fail reporting.

  From: /tf/active/vicechatdev/vice_ai/test_integration.py

• function test_data_analysis_service 60.7% similar

  A test function that validates the functionality of the DataAnalysisService by creating, initializing, and retrieving analysis sessions.

  From: /tf/active/vicechatdev/vice_ai/test_integration.py

• function demo_statistical_agent 60.6% similar

  Demonstrates the capabilities of a statistical agent by testing query interpretation on sample data with various statistical analysis queries.

  From: /tf/active/vicechatdev/full_smartstat/demo.py

• class AgentExecutor_v1 59.1% similar

  Agent-based script executor that generates standalone Python files, manages dependencies, and provides iterative debugging capabilities

  From: /tf/active/vicechatdev/full_smartstat/agent_executor.py