556 lines
23 KiB
Python
556 lines
23 KiB
Python
#!/usr/bin/env python3
|
|
# Copyright (c) 2025 Patrick Motsch
|
|
# All rights reserved.
|
|
"""
|
|
Code Generation Formats Test 11 - Tests code generation in JSON, CSV, and XML formats
|
|
Tests code generation with structured data formats including validation and formatting.
|
|
"""
|
|
|
|
import asyncio
|
|
import json
|
|
import sys
|
|
import os
|
|
import time
|
|
import csv
|
|
import io
|
|
import xml.etree.ElementTree as ET
|
|
from typing import Dict, Any, List, Optional
|
|
|
|
# Add the gateway to path (go up 2 levels from tests/functional/)
|
|
_gateway_path = os.path.abspath(os.path.join(os.path.dirname(__file__), "..", ".."))
|
|
if _gateway_path not in sys.path:
|
|
sys.path.insert(0, _gateway_path)
|
|
|
|
# Import the service initialization
|
|
from modules.services import getInterface as getServices
|
|
from modules.datamodels.datamodelChat import UserInputRequest, WorkflowModeEnum
|
|
from modules.datamodels.datamodelUam import User
|
|
from modules.features.workflow import chatStart
|
|
import modules.interfaces.interfaceDbChat as interfaceDbChat
|
|
|
|
|
|
class CodeGenerationFormatsTester11:
|
|
def __init__(self):
|
|
# Use root user for testing (has full access to everything)
|
|
from modules.interfaces.interfaceDbApp import getRootInterface
|
|
rootInterface = getRootInterface()
|
|
self.testUser = rootInterface.currentUser
|
|
|
|
# Initialize services using the existing system
|
|
self.services = getServices(self.testUser, None) # Test user, no workflow
|
|
self.workflow = None
|
|
self.testResults = {}
|
|
self.generatedDocuments = {}
|
|
|
|
async def initialize(self):
|
|
"""Initialize the test environment."""
|
|
# Enable debug file logging for tests
|
|
from modules.shared.configuration import APP_CONFIG
|
|
APP_CONFIG.set("APP_DEBUG_CHAT_WORKFLOW_ENABLED", True)
|
|
|
|
# Set logging level to INFO to see workflow progress
|
|
import logging
|
|
logging.getLogger().setLevel(logging.INFO)
|
|
|
|
print(f"Initialized test with user: {self.testUser.id}")
|
|
print(f"Mandate ID: {self.testUser.mandateId}")
|
|
print(f"Debug logging enabled: {APP_CONFIG.get('APP_DEBUG_CHAT_WORKFLOW_ENABLED', False)}")
|
|
|
|
def createTestPrompt(self, format: str) -> str:
|
|
"""Create a test prompt for code generation in the specified format.
|
|
|
|
The prompt requests 3 files for each format:
|
|
- Structured data generation appropriate for the format
|
|
- Proper formatting and validation
|
|
"""
|
|
formatPrompts = {
|
|
"json": (
|
|
"Generate 3 JSON code files for a customer management system:\n"
|
|
"1) Create a config.json file with:\n"
|
|
" - Application name: 'Customer Manager'\n"
|
|
" - Version: '1.0.0'\n"
|
|
" - Database settings: host, port, name\n"
|
|
" - API settings: baseUrl, timeout\n"
|
|
"2) Create a customers.json file with an array of customer objects:\n"
|
|
" - Each customer should have: id, name, email, phone, address\n"
|
|
" - Include at least 3 sample customers\n"
|
|
"3) Create a settings.json file with:\n"
|
|
" - Theme settings: darkMode, fontSize, language\n"
|
|
" - Notification settings: email, sms, push\n"
|
|
" - Feature flags: enableAnalytics, enableReports\n\n"
|
|
"Format all files as valid JSON with proper indentation."
|
|
),
|
|
"csv": (
|
|
"Generate 3 CSV code files for expense tracking:\n"
|
|
"1) Create an expenses.csv file with:\n"
|
|
" - Header row: Documentname, Datum, Händler, Kreditkartennummer, Gesamtbetrag, Währung, MWST-Satz\n"
|
|
" - Data rows with at least 5 expense entries\n"
|
|
" - Use consistent date format (DD.MM.YYYY)\n"
|
|
" - Use CHF as currency\n"
|
|
" - Use 7.7% as VAT rate\n"
|
|
"2) Create a categories.csv file with:\n"
|
|
" - Header row: CategoryID, CategoryName, Description, ParentCategory\n"
|
|
" - Data rows with at least 8 categories\n"
|
|
"3) Create a vendors.csv file with:\n"
|
|
" - Header row: VendorID, VendorName, ContactPerson, Email, Phone, Address\n"
|
|
" - Data rows with at least 6 vendors\n\n"
|
|
"Format all files as valid CSV with proper header row and consistent column count."
|
|
),
|
|
"xml": (
|
|
"Generate 3 XML code files for a product catalog:\n"
|
|
"1) Create a products.xml file with:\n"
|
|
" - Root element: <catalog>\n"
|
|
" - Each product as <product> element with:\n"
|
|
" - <id>, <name>, <description>, <price>, <category>\n"
|
|
" - Include at least 4 products\n"
|
|
"2) Create a categories.xml file with:\n"
|
|
" - Root element: <categories>\n"
|
|
" - Each category as <category> element with:\n"
|
|
" - <id>, <name>, <description>, <parentId>\n"
|
|
" - Include at least 5 categories\n"
|
|
"3) Create a suppliers.xml file with:\n"
|
|
" - Root element: <suppliers>\n"
|
|
" - Each supplier as <supplier> element with:\n"
|
|
" - <id>, <name>, <contact>, <address>\n"
|
|
" - Include at least 3 suppliers\n\n"
|
|
"Format all files as valid XML with proper indentation and structure."
|
|
)
|
|
}
|
|
|
|
return formatPrompts.get(format.lower(), formatPrompts["json"])
|
|
|
|
async def generateCodeInFormat(self, format: str) -> Dict[str, Any]:
|
|
"""Generate code in the specified format using workflow."""
|
|
print("\n" + "="*80)
|
|
print(f"GENERATING CODE IN {format.upper()} FORMAT")
|
|
print("="*80)
|
|
|
|
prompt = self.createTestPrompt(format)
|
|
print(f"Prompt: {prompt[:200]}...")
|
|
|
|
# Create user input request
|
|
userInput = UserInputRequest(
|
|
prompt=prompt,
|
|
listFileId=[],
|
|
userLanguage="en"
|
|
)
|
|
|
|
# Start workflow
|
|
print(f"\nStarting workflow for {format.upper()} code generation...")
|
|
workflow = await chatStart(
|
|
currentUser=self.testUser,
|
|
userInput=userInput,
|
|
workflowMode=WorkflowModeEnum.WORKFLOW_DYNAMIC,
|
|
workflowId=None
|
|
)
|
|
|
|
if not workflow:
|
|
return {
|
|
"success": False,
|
|
"error": "Failed to start workflow"
|
|
}
|
|
|
|
self.workflow = workflow
|
|
print(f"Workflow started: {workflow.id}")
|
|
|
|
# Wait for workflow completion (no timeout - wait indefinitely)
|
|
print(f"Waiting for workflow completion...")
|
|
completed = await self.waitForWorkflowCompletion(timeout=None)
|
|
|
|
if not completed:
|
|
return {
|
|
"success": False,
|
|
"error": "Workflow did not complete",
|
|
"workflowId": workflow.id,
|
|
"status": workflow.status if workflow else "unknown"
|
|
}
|
|
|
|
# Analyze results
|
|
results = self.analyzeWorkflowResults()
|
|
|
|
# Extract documents for this format
|
|
documents = results.get("documents", [])
|
|
formatDocuments = [d for d in documents if d.get("fileName", "").endswith(f".{format.lower()}")]
|
|
|
|
return {
|
|
"success": True,
|
|
"format": format,
|
|
"workflowId": workflow.id,
|
|
"status": results.get("status"),
|
|
"documentCount": len(formatDocuments),
|
|
"documents": formatDocuments,
|
|
"results": results
|
|
}
|
|
|
|
async def waitForWorkflowCompletion(self, timeout: Optional[int] = None, checkInterval: int = 2) -> bool:
|
|
"""Wait for workflow to complete."""
|
|
if not self.workflow:
|
|
return False
|
|
|
|
startTime = time.time()
|
|
lastStatus = None
|
|
|
|
interfaceDbChat = interfaceDbChat.getInterface(self.testUser)
|
|
|
|
if timeout is None:
|
|
print("Waiting indefinitely (no timeout)")
|
|
|
|
while True:
|
|
# Check timeout only if specified
|
|
if timeout is not None and time.time() - startTime > timeout:
|
|
print(f"\n⏱️ Timeout after {timeout} seconds")
|
|
return False
|
|
|
|
# Get current workflow status
|
|
try:
|
|
currentWorkflow = interfaceDbChat.getWorkflow(self.workflow.id)
|
|
if not currentWorkflow:
|
|
print("\n❌ Workflow not found")
|
|
return False
|
|
|
|
currentStatus = currentWorkflow.status
|
|
elapsed = int(time.time() - startTime)
|
|
|
|
# Print status if it changed
|
|
if currentStatus != lastStatus:
|
|
print(f"Workflow status: {currentStatus} (elapsed: {elapsed}s)")
|
|
lastStatus = currentStatus
|
|
|
|
# Check if workflow is complete
|
|
if currentStatus in ["completed", "stopped", "failed"]:
|
|
self.workflow = currentWorkflow
|
|
statusIcon = "✅" if currentStatus == "completed" else "❌"
|
|
print(f"\n{statusIcon} Workflow finished with status: {currentStatus} (elapsed: {elapsed}s)")
|
|
return currentStatus == "completed"
|
|
|
|
# Wait before next check
|
|
await asyncio.sleep(checkInterval)
|
|
|
|
except Exception as e:
|
|
print(f"\n⚠️ Error checking workflow status: {str(e)}")
|
|
await asyncio.sleep(checkInterval)
|
|
|
|
def analyzeWorkflowResults(self) -> Dict[str, Any]:
|
|
"""Analyze workflow results and extract information."""
|
|
if not self.workflow:
|
|
return {"error": "No workflow to analyze"}
|
|
|
|
interfaceDbChat = interfaceDbChat.getInterface(self.testUser)
|
|
workflow = interfaceDbChat.getWorkflow(self.workflow.id)
|
|
|
|
if not workflow:
|
|
return {"error": "Workflow not found"}
|
|
|
|
# Get unified chat data
|
|
chatData = interfaceDbChat.getUnifiedChatData(workflow.id, None)
|
|
|
|
# Count messages
|
|
messages = chatData.get("messages", [])
|
|
userMessages = [m for m in messages if m.get("role") == "user"]
|
|
assistantMessages = [m for m in messages if m.get("role") == "assistant"]
|
|
|
|
# Count documents
|
|
documents = chatData.get("documents", [])
|
|
|
|
# Get logs
|
|
logs = chatData.get("logs", [])
|
|
|
|
results = {
|
|
"workflowId": workflow.id,
|
|
"status": workflow.status,
|
|
"workflowMode": str(workflow.workflowMode) if hasattr(workflow, 'workflowMode') else None,
|
|
"currentRound": workflow.currentRound,
|
|
"totalTasks": workflow.totalTasks,
|
|
"totalActions": workflow.totalActions,
|
|
"messageCount": len(messages),
|
|
"userMessageCount": len(userMessages),
|
|
"assistantMessageCount": len(assistantMessages),
|
|
"documentCount": len(documents),
|
|
"logCount": len(logs),
|
|
"documents": documents,
|
|
"logs": logs
|
|
}
|
|
|
|
print(f"\nWorkflow Results:")
|
|
print(f" Status: {results['status']}")
|
|
print(f" Tasks: {results['totalTasks']}")
|
|
print(f" Actions: {results['totalActions']}")
|
|
print(f" Messages: {results['messageCount']}")
|
|
print(f" Documents: {results['documentCount']}")
|
|
|
|
# Print document details
|
|
if documents:
|
|
print(f"\nGenerated Documents:")
|
|
for doc in documents:
|
|
fileName = doc.get("fileName", "unknown")
|
|
fileSize = doc.get("fileSize", 0)
|
|
mimeType = doc.get("mimeType", "unknown")
|
|
print(f" - {fileName} ({fileSize} bytes, {mimeType})")
|
|
|
|
return results
|
|
|
|
|
|
def verifyCodeFormat(self, document: Dict[str, Any], expectedFormat: str) -> Dict[str, Any]:
|
|
"""Verify that a code file matches the expected format and is valid."""
|
|
fileName = document.get("fileName", "")
|
|
mimeType = document.get("mimeType", "")
|
|
fileSize = document.get("fileSize", 0)
|
|
|
|
# Expected MIME types
|
|
expectedMimeTypes = {
|
|
"json": ["application/json"],
|
|
"csv": ["text/csv"],
|
|
"xml": ["application/xml", "text/xml"]
|
|
}
|
|
|
|
# Expected file extensions
|
|
expectedExtensions = {
|
|
"json": [".json"],
|
|
"csv": [".csv"],
|
|
"xml": [".xml"]
|
|
}
|
|
|
|
formatLower = expectedFormat.lower()
|
|
expectedMimes = expectedMimeTypes.get(formatLower, [])
|
|
expectedExts = expectedExtensions.get(formatLower, [])
|
|
|
|
# Check file extension
|
|
hasCorrectExtension = any(fileName.lower().endswith(ext) for ext in expectedExts)
|
|
|
|
# Check MIME type
|
|
hasCorrectMimeType = any(mimeType.lower() == mime.lower() for mime in expectedMimes)
|
|
|
|
# Check file size (should be > 0)
|
|
hasValidSize = fileSize > 0
|
|
|
|
# Try to read and validate content
|
|
isValidContent = False
|
|
validationError = None
|
|
|
|
try:
|
|
# Get file content from fileId
|
|
fileId = document.get("fileId")
|
|
if fileId and hasattr(self.services, 'interfaceDbComponent'):
|
|
fileData = self.services.interfaceDbComponent.getFileData(fileId)
|
|
if fileData:
|
|
content = fileData.decode('utf-8') if isinstance(fileData, bytes) else fileData
|
|
|
|
# Validate format-specific syntax
|
|
if formatLower == "json":
|
|
try:
|
|
json.loads(content)
|
|
isValidContent = True
|
|
except json.JSONDecodeError as e:
|
|
validationError = f"Invalid JSON: {str(e)}"
|
|
|
|
elif formatLower == "csv":
|
|
try:
|
|
reader = csv.reader(io.StringIO(content))
|
|
rows = list(reader)
|
|
if len(rows) > 0:
|
|
# Check header row exists
|
|
headerCount = len(rows[0])
|
|
# Check all rows have same column count
|
|
allRowsValid = all(len(row) == headerCount for row in rows)
|
|
isValidContent = allRowsValid
|
|
if not allRowsValid:
|
|
validationError = "CSV rows have inconsistent column counts"
|
|
else:
|
|
validationError = "CSV file is empty"
|
|
except Exception as e:
|
|
validationError = f"CSV parsing error: {str(e)}"
|
|
|
|
elif formatLower == "xml":
|
|
try:
|
|
ET.fromstring(content)
|
|
isValidContent = True
|
|
except ET.ParseError as e:
|
|
validationError = f"Invalid XML: {str(e)}"
|
|
else:
|
|
validationError = "Could not read file data"
|
|
else:
|
|
validationError = "No fileId available"
|
|
|
|
except Exception as e:
|
|
validationError = f"Error reading/validating file: {str(e)}"
|
|
|
|
verification = {
|
|
"format": expectedFormat,
|
|
"fileName": fileName,
|
|
"mimeType": mimeType,
|
|
"fileSize": fileSize,
|
|
"hasCorrectExtension": hasCorrectExtension,
|
|
"hasCorrectMimeType": hasCorrectMimeType,
|
|
"hasValidSize": hasValidSize,
|
|
"isValidContent": isValidContent,
|
|
"validationError": validationError,
|
|
"isValid": hasCorrectExtension and hasValidSize and hasCorrectMimeType,
|
|
"isComplete": hasCorrectExtension and hasValidSize and hasCorrectMimeType and isValidContent
|
|
}
|
|
|
|
return verification
|
|
|
|
async def testAllFormats(self) -> Dict[str, Any]:
|
|
"""Test code generation in JSON, CSV, and XML formats."""
|
|
print("\n" + "="*80)
|
|
print("TESTING CODE GENERATION IN ALL FORMATS")
|
|
print("="*80)
|
|
|
|
# Test all code formats
|
|
formats = ["json", "csv", "xml"]
|
|
results = {}
|
|
|
|
for format in formats:
|
|
try:
|
|
print(f"\n{'='*80}")
|
|
print(f"Testing {format.upper()} format...")
|
|
print(f"{'='*80}")
|
|
|
|
result = await self.generateCodeInFormat(format)
|
|
results[format] = result
|
|
|
|
if result.get("success"):
|
|
documents = result.get("documents", [])
|
|
if documents:
|
|
# Verify all documents (expecting 3 files per format)
|
|
verifications = []
|
|
for doc in documents:
|
|
verification = self.verifyCodeFormat(doc, format)
|
|
verifications.append(verification)
|
|
|
|
result["verifications"] = verifications
|
|
|
|
# Count valid documents
|
|
validCount = sum(1 for v in verifications if v.get("isValid"))
|
|
contentValidCount = sum(1 for v in verifications if v.get("isValidContent"))
|
|
|
|
print(f"\n✅ {format.upper()} generation successful!")
|
|
print(f" Documents: {len(documents)} (expected: 3)")
|
|
print(f" Valid Format: {validCount}/{len(documents)}")
|
|
print(f" Valid Content: {contentValidCount}/{len(documents)}")
|
|
|
|
# Print details for each file
|
|
for i, verification in enumerate(verifications, 1):
|
|
statusIcon = "✅" if verification.get("isValid") else "❌"
|
|
contentIcon = "✅" if verification.get("isValidContent") else "❌"
|
|
print(f" File {i}: {statusIcon} Format, {contentIcon} Content - {verification.get('fileName', 'unknown')}")
|
|
if verification.get("validationError"):
|
|
print(f" Error: {verification['validationError']}")
|
|
else:
|
|
print(f"\n⚠️ {format.upper()} generation completed but no documents found")
|
|
else:
|
|
error = result.get("error", "Unknown error")
|
|
print(f"\n❌ {format.upper()} generation failed: {error}")
|
|
|
|
# Small delay between tests
|
|
await asyncio.sleep(2)
|
|
|
|
except Exception as e:
|
|
import traceback
|
|
print(f"\n❌ Error testing {format.upper()}: {str(e)}")
|
|
print(traceback.format_exc())
|
|
results[format] = {
|
|
"success": False,
|
|
"error": str(e),
|
|
"traceback": traceback.format_exc()
|
|
}
|
|
|
|
return results
|
|
|
|
async def runTest(self):
|
|
"""Run the complete test."""
|
|
print("\n" + "="*80)
|
|
print("CODE GENERATION FORMATS TEST 11 - JSON, CSV, XML")
|
|
print("="*80)
|
|
|
|
try:
|
|
# Initialize
|
|
await self.initialize()
|
|
|
|
# Test all formats
|
|
formatResults = await self.testAllFormats()
|
|
|
|
# Summary
|
|
print("\n" + "="*80)
|
|
print("TEST SUMMARY")
|
|
print("="*80)
|
|
|
|
# Format tests summary
|
|
print("\nFormat Tests:")
|
|
successCount = 0
|
|
failCount = 0
|
|
completeCount = 0 # Files with valid content
|
|
|
|
for format, result in formatResults.items():
|
|
if result.get("success"):
|
|
successCount += 1
|
|
verifications = result.get("verifications", [])
|
|
docCount = result.get("documentCount", 0)
|
|
|
|
# Count valid files
|
|
validCount = sum(1 for v in verifications if v.get("isValid"))
|
|
contentValidCount = sum(1 for v in verifications if v.get("isValidContent"))
|
|
completeCount += contentValidCount
|
|
|
|
# Overall status (all files valid)
|
|
allValid = len(verifications) > 0 and all(v.get("isValid") for v in verifications)
|
|
allContentValid = len(verifications) > 0 and all(v.get("isValidContent") for v in verifications)
|
|
|
|
statusIcon = "✅" if allValid else "⚠️"
|
|
contentIcon = "✅" if allContentValid else "❌"
|
|
|
|
print(f"{statusIcon} {format.upper():6s}: {'PASS' if allValid else 'PARTIAL'} - {docCount} file(s) ({validCount} valid format, {contentValidCount} valid content)")
|
|
|
|
# Print errors if any
|
|
for v in verifications:
|
|
if v.get("validationError"):
|
|
print(f" {v.get('fileName', 'unknown')}: {v['validationError']}")
|
|
else:
|
|
failCount += 1
|
|
error = result.get("error", "Unknown error")
|
|
print(f"❌ {format.upper():6s}: FAIL - {error}")
|
|
|
|
print(f"\nFormat Tests: {successCount} passed, {failCount} failed out of {len(formatResults)} formats")
|
|
print(f"Valid Content Files: {completeCount} total files with valid content")
|
|
|
|
self.testResults = {
|
|
"success": failCount == 0,
|
|
"formatTests": {
|
|
"successCount": successCount,
|
|
"failCount": failCount,
|
|
"completeCount": completeCount,
|
|
"totalFormats": len(formatResults),
|
|
"results": formatResults
|
|
},
|
|
"totalSuccess": successCount,
|
|
"totalFail": failCount
|
|
}
|
|
|
|
return self.testResults
|
|
|
|
except Exception as e:
|
|
import traceback
|
|
print(f"\n❌ Test failed with error: {type(e).__name__}: {str(e)}")
|
|
print(f"Traceback:\n{traceback.format_exc()}")
|
|
self.testResults = {
|
|
"success": False,
|
|
"error": str(e),
|
|
"traceback": traceback.format_exc()
|
|
}
|
|
return self.testResults
|
|
|
|
|
|
async def main():
|
|
"""Run code generation formats test 11."""
|
|
tester = CodeGenerationFormatsTester11()
|
|
results = await tester.runTest()
|
|
|
|
# Print final results as JSON for easy parsing
|
|
print("\n" + "="*80)
|
|
print("FINAL RESULTS (JSON)")
|
|
print("="*80)
|
|
print(json.dumps(results, indent=2, default=str))
|
|
|
|
|
|
if __name__ == "__main__":
|
|
asyncio.run(main())
|