gateway/scripts/script_analyze_imports.py

# Copyright (c) 2025 Patrick Motsch
# All rights reserved.
"""
Analyze all imports in the gateway codebase and generate a CSV report.
Helps identify which imports need to be cleaned up after refactoring.
"""

import os
import ast
import csv
import sys
from pathlib import Path
from typing import List, Tuple, Optional

# Gateway root directory
GATEWAY_ROOT = Path(__file__).parent.parent
OUTPUT_FILE = GATEWAY_ROOT / "scripts" / "import_analysis.csv"


def getModuleName(filePath: Path) -> str:
    """Convert file path to module name format."""
    relPath = filePath.relative_to(GATEWAY_ROOT.parent)
    # Remove .py extension and convert to module format
    modulePath = str(relPath).replace(os.sep, ".").replace("/", ".")
    if modulePath.endswith(".py"):
        modulePath = modulePath[:-3]
    return modulePath


def getImportedModuleName(importNode: ast.AST) -> List[str]:
    """Extract imported module names from import node."""
    modules = []
    
    if isinstance(importNode, ast.Import):
        for alias in importNode.names:
            modules.append(alias.name)
    elif isinstance(importNode, ast.ImportFrom):
        if importNode.module:
            # For relative imports, we'll mark them specially
            if importNode.level > 0:
                modules.append(f"{'.' * importNode.level}{importNode.module or ''}")
            else:
                modules.append(importNode.module)
        elif importNode.level > 0:
            # Pure relative import like "from . import x"
            modules.append("." * importNode.level)
    
    return modules


def findEnclosingFunction(node: ast.AST, tree: ast.Module) -> Optional[str]:
    """Find the function name that contains this import, if any."""
    for item in ast.walk(tree):
        if isinstance(item, (ast.FunctionDef, ast.AsyncFunctionDef)):
            for child in ast.walk(item):
                if child is node:
                    return item.name
    return None


def checkModuleExists(moduleName: str, currentFile: Path) -> bool:
    """Check if an imported module exists."""
    if moduleName.startswith("."):
        # Relative import - check relative to current file's directory
        currentDir = currentFile.parent
        relativePath = moduleName.lstrip(".")
        
        if not relativePath:
            return True  # "from . import x" - current package
        
        # Convert module path to file path
        parts = relativePath.split(".")
        checkPath = currentDir
        for part in parts:
            checkPath = checkPath / part
        
        # Check if it's a package or module
        if checkPath.is_dir() and (checkPath / "__init__.py").exists():
            return True
        if (checkPath.parent / f"{checkPath.name}.py").exists():
            return True
        if checkPath.with_suffix(".py").exists():
            return True
        
        return False
    
    # Absolute import
    if moduleName.startswith("modules."):
        # Internal module - check in gateway
        parts = moduleName.split(".")
        checkPath = GATEWAY_ROOT
        for part in parts:
            checkPath = checkPath / part
        
        # Check if it's a package or module
        if checkPath.is_dir() and (checkPath / "__init__.py").exists():
            return True
        if checkPath.with_suffix(".py").exists():
            return True
        
        return False
    
    # External module - assume it exists (can't easily verify without importing)
    return True


def analyzeFile(filePath: Path) -> List[Tuple[str, str, str, str]]:
    """Analyze imports in a single file."""
    results = []
    moduleName = getModuleName(filePath)
    
    try:
        with open(filePath, "r", encoding="utf-8") as f:
            content = f.read()
        
        tree = ast.parse(content, filename=str(filePath))
    except (SyntaxError, UnicodeDecodeError) as e:
        print(f"Error parsing {filePath}: {e}")
        return results
    
    # Find all imports and their positions
    for node in ast.walk(tree):
        if isinstance(node, (ast.Import, ast.ImportFrom)):
            importedModules = getImportedModuleName(node)
            
            # Determine position
            position = "header"
            
            # Check if import is inside a function by examining parent nodes
            # We need to traverse the tree structure
            for item in ast.walk(tree):
                if isinstance(item, (ast.FunctionDef, ast.AsyncFunctionDef)):
                    # Check if this import node is within the function's body
                    for bodyNode in ast.walk(item):
                        if bodyNode is node:
                            position = f"function {item.name}"
                            break
                    if position != "header":
                        break
            
            for importedModule in importedModules:
                # Check if module exists
                exists = checkModuleExists(importedModule, filePath)
                validStr = "Yes" if exists else "No"
                
                # Format imported module name
                if importedModule.startswith("."):
                    # Make relative import absolute for clarity
                    importedModuleDisplay = f"(relative) {importedModule}"
                else:
                    importedModuleDisplay = importedModule
                
                results.append((moduleName, importedModuleDisplay, position, validStr))
    
    return results


def main():
    """Main function to analyze all imports."""
    allResults = []
    
    # Find all Python files in gateway
    for root, dirs, files in os.walk(GATEWAY_ROOT):
        # Skip __pycache__ directories
        dirs[:] = [d for d in dirs if d != "__pycache__"]
        
        for file in files:
            if file.endswith(".py"):
                filePath = Path(root) / file
                results = analyzeFile(filePath)
                allResults.extend(results)
    
    # Sort results
    allResults.sort(key=lambda x: (x[0], x[1]))
    
    # Write CSV
    with open(OUTPUT_FILE, "w", newline="", encoding="utf-8") as f:
        writer = csv.writer(f)
        writer.writerow(["module_name", "imported_module_name", "position", "import_valid"])
        writer.writerows(allResults)
    
    print(f"Analysis complete. Found {len(allResults)} imports.")
    print(f"Output written to: {OUTPUT_FILE}")
    
    # Print summary of invalid imports
    invalidImports = [r for r in allResults if r[3] == "No"]
    if invalidImports:
        print(f"\nFound {len(invalidImports)} potentially invalid imports:")
        for moduleName, importedModule, position, _ in invalidImports[:20]:
            print(f"  {moduleName} -> {importedModule} ({position})")
        if len(invalidImports) > 20:
            print(f"  ... and {len(invalidImports) - 20} more")


if __name__ == "__main__":
    main()