tests/test_bug_fixes.py

s#!/usr/bin/env python3
"""
Test script to verify bug fixes for document editor tools
Tests that:
1. All internal functions exist and have __name__ attribute
2. tools_real contains only functions (no StructuredTools)
3. Workflow builds successfully
"""

import sys
import os
from pathlib import Path

# Add parent directory to path
sys.path.insert(0, str(Path(__file__).parent.parent))

def test_internal_functions_exist():
    """Test that all internal functions exist and have __name__"""
    print("\n" + "=" * 80)
    print("TEST 1: Internal Functions Exist and Have __name__")
    print("=" * 80)
    
    from utils.editor_tools import (
        _replace_html, _add_html, _delete_html, _inspect_document, _attempt_completion
    )
    
    functions = [
        ("_replace_html", _replace_html),
        ("_add_html", _add_html),
        ("_delete_html", _delete_html),
        ("_inspect_document", _inspect_document),
        ("_attempt_completion", _attempt_completion)
    ]
    
    all_passed = True
    for name, func in functions:
        try:
            func_name = func.__name__
            print(f"✅ {name}: __name__ = '{func_name}'")
        except AttributeError as e:
            print(f"❌ {name}: {e}")
            all_passed = False
    
    return all_passed


def test_tools_real_are_functions():
    """Test that tools_real contains only functions"""
    print("\n" + "=" * 80)
    print("TEST 2: tools_real Contains Only Functions")
    print("=" * 80)
    
    # Import agent - this will fail if tools_real has StructuredTools
    try:
        from agents.doc_editor import DocumentEditorAgent
        
        # We need a mock LLM to initialize
        class MockLLM:
            def bind_tools(self, tools):
                return self
            
            async def ainvoke(self, messages):
                from langchain_core.messages import AIMessage
                return AIMessage(content="Test response")
        
        llm = MockLLM()
        agent = DocumentEditorAgent(llm=llm)
        
        print(f"✅ Agent initialized successfully")
        print(f"📦 tools_real has {len(agent.tools_real)} items")
        
        # Check that all are functions
        all_passed = True
        for i, tool in enumerate(agent.tools_real):
            try:
                name = tool.__name__
                print(f"✅ Tool {i}: {name} (has __name__)")
            except AttributeError as e:
                print(f"❌ Tool {i}: {e}")
                all_passed = False
        
        return all_passed
    
    except Exception as e:
        print(f"❌ Failed to initialize agent: {e}")
        import traceback
        traceback.print_exc()
        return False


def test_workflow_builds():
    """Test that workflow builds successfully"""
    print("\n" + "=" * 80)
    print("TEST 3: Workflow Builds Successfully")
    print("=" * 80)
    
    try:
        from agents.doc_editor import DocumentEditorAgent
        
        class MockLLM:
            def bind_tools(self, tools):
                return self
            
            async def ainvoke(self, messages):
                from langchain_core.messages import AIMessage
                return AIMessage(content="Test response")
        
        llm = MockLLM()
        agent = DocumentEditorAgent(llm=llm)
        
        print(f"✅ Workflow built successfully")
        # Note: CompiledStateGraph doesn't expose nodes() and edges() methods directly
        # The important thing is that it built without errors
        
        return True
    
    except Exception as e:
        print(f"❌ Failed to build workflow: {e}")
        import traceback
        traceback.print_exc()
        return False


def test_tools_callable():
    """Test that internal tools are callable"""
    print("\n" + "=" * 80)
    print("TEST 4: Internal Tools Are Callable")
    print("=" * 80)
    
    from utils.editor_tools import (
        _replace_html, _add_html, _delete_html, _inspect_document, _attempt_completion
    )
    
    import asyncio
    
    # Test _replace_html
    async def test_replace():
        result = await _replace_html(
            doc_text="Hello World",
            search="World",
            replace="Universe",
            expected_matches=1
        )
        return result
    
    # Test _inspect_document
    async def test_inspect():
        result = await _inspect_document(
            doc_text="Test document"
        )
        return result
    
    # Test _attempt_completion
    async def test_attempt():
        result = await _attempt_completion(
            message="Test complete"
        )
        return result
    
    all_passed = True
    
    # Run tests
    print("\nTesting _replace_html...")
    try:
        result = asyncio.run(test_replace())
        if result.get("ok"):
            print(f"✅ _replace_html returned: {result}")
        else:
            print(f"⚠️ _replace_html failed (expected for invalid HTML): {result}")
    except Exception as e:
        print(f"❌ _replace_html error: {e}")
        all_passed = False
    
    print("\nTesting _inspect_document...")
    try:
        result = asyncio.run(test_inspect())
        if result.get("ok"):
            print(f"✅ _inspect_document returned: {result}")
        else:
            print(f"❌ _inspect_document failed: {result}")
            all_passed = False
    except Exception as e:
        print(f"❌ _inspect_document error: {e}")
        all_passed = False
    
    print("\nTesting _attempt_completion...")
    try:
        result = asyncio.run(test_attempt())
        if result.get("ok"):
            print(f"✅ _attempt_completion returned: {result}")
        else:
            print(f"❌ _attempt_completion failed: {result}")
            all_passed = False
    except Exception as e:
        print(f"❌ _attempt_completion error: {e}")
        all_passed = False
    
    return all_passed


if __name__ == "__main__":
    print("\n" + "=" * 80)
    print("BUG FIX VERIFICATION TESTS")
    print("=" * 80)
    
    results = {}
    
    # Test 1: Internal functions exist
    results["test_internal_functions_exist"] = test_internal_functions_exist()
    
    # Test 2: tools_real contains only functions
    results["test_tools_real_are_functions"] = test_tools_real_are_functions()
    
    # Test 3: Workflow builds
    results["test_workflow_builds"] = test_workflow_builds()
    
    # Test 4: Tools are callable
    results["test_tools_callable"] = test_tools_callable()
    
    # Summary
    print("\n" + "=" * 80)
    print("TEST SUMMARY")
    print("=" * 80)
    
    passed = sum(1 for v in results.values() if v)
    total = len(results)
    
    for test_name, result in results.items():
        status = "✅ PASSED" if result else "❌ FAILED"
        print(f"{status}: {test_name}")
    
    print("\n" + "=" * 80)
    if passed == total:
        print(f"✅ ALL TESTS PASSED ({passed}/{total})")
    else:
        print(f"❌ SOME TESTS FAILED ({passed}/{total} passed)")
    print("=" * 80 + "\n")
    
    sys.exit(0 if passed == total else 1)