kyorlin2001/code-analysis-tool

1from __future__ import annotations2 3import re4from dataclasses import dataclass5from pathlib import Path6from typing import Any7 8 9IMPORT_PATTERNS = {10    ".py": re.compile(r"^\s*(?:from\s+\S+\s+import\s+\S+|import\s+\S+)", re.MULTILINE),11    ".js": re.compile(r"^\s*import\s+.*|^\s*const\s+\w+\s*=\s*require\(", re.MULTILINE),12    ".ts": re.compile(r"^\s*import\s+.*|^\s*const\s+\w+\s*=\s*require\(", re.MULTILINE),13    ".jsx": re.compile(r"^\s*import\s+.*|^\s*const\s+\w+\s*=\s*require\(", re.MULTILINE),14    ".tsx": re.compile(r"^\s*import\s+.*|^\s*const\s+\w+\s*=\s*require\(", re.MULTILINE),15}16 17 18@dataclass19class ParsedCodeInfo:20    path: str21    language: str22    line_count: int23    imports: list[str]24    todos: list[str]25    classes: list[str]26    functions: list[str]27 28 29def read_text(path: Path) -> str:30    try:31        return path.read_text(encoding="utf-8", errors="ignore")32    except Exception:33        return ""34 35 36def detect_language(path: str) -> str:37    suffix = Path(path).suffix.lower()38 39    mapping = {40        ".py": "Python",41        ".js": "JavaScript",42        ".jsx": "JavaScript React",43        ".ts": "TypeScript",44        ".tsx": "TypeScript React",45        ".java": "Java",46        ".kt": "Kotlin",47        ".go": "Go",48        ".rs": "Rust",49        ".c": "C",50        ".cpp": "C++",51        ".h": "C/C++ Header",52        ".hpp": "C++ Header",53        ".cs": "C#",54        ".php": "PHP",55        ".rb": "Ruby",56        ".swift": "Swift",57    }58 59    return mapping.get(suffix, "Unknown")60 61 62def extract_imports(text: str, path: str) -> list[str]:63    suffix = Path(path).suffix.lower()64    pattern = IMPORT_PATTERNS.get(suffix)65    if not pattern:66        return []67 68    matches = pattern.findall(text)69    return [m.strip() for m in matches if m.strip()]70 71 72def extract_todos(text: str) -> list[str]:73    todos: list[str] = []74    for line in text.splitlines():75        stripped = line.strip()76        upper = stripped.upper()77        if "TODO" in upper or "FIXME" in upper or "XXX" in upper:78            todos.append(stripped)79    return todos80 81 82def extract_python_symbols(text: str) -> tuple[list[str], list[str]]:83    classes = re.findall(r"^\s*class\s+([A-Za-z_][A-Za-z0-9_]*)", text, re.MULTILINE)84    functions = re.findall(r"^\s*def\s+([A-Za-z_][A-Za-z0-9_]*)", text, re.MULTILINE)85    return classes, functions86 87 88def extract_js_symbols(text: str) -> tuple[list[str], list[str]]:89    classes = re.findall(r"^\s*class\s+([A-Za-z_][A-Za-z0-9_]*)", text, re.MULTILINE)90    functions = re.findall(91        r"^\s*(?:function\s+([A-Za-z_][A-Za-z0-9_]*)|const\s+([A-Za-z_][A-Za-z0-9_]*)\s*=\s*\()",92        text,93        re.MULTILINE,94    )95 96    flat_functions: list[str] = []97    for fn_a, fn_b in functions:98        name = fn_a or fn_b99        if name:100            flat_functions.append(name)101 102    return classes, flat_functions103 104 105def parse_code_file(path: str) -> ParsedCodeInfo:106    file_path = Path(path)107    text = read_text(file_path)108    language = detect_language(path)109 110    imports = extract_imports(text, path)111    todos = extract_todos(text)112 113    classes: list[str] = []114    functions: list[str] = []115 116    if file_path.suffix.lower() == ".py":117        classes, functions = extract_python_symbols(text)118    elif file_path.suffix.lower() in {".js", ".jsx", ".ts", ".tsx"}:119        classes, functions = extract_js_symbols(text)120 121    return ParsedCodeInfo(122        path=path,123        language=language,124        line_count=len(text.splitlines()),125        imports=imports,126        todos=todos,127        classes=classes,128        functions=functions,129    )130 131 132def parse_repository_files(files: list[str], root_dir: str) -> list[ParsedCodeInfo]:133    """134    Parse a list of repository-relative file paths into structured code metadata.135    """136    root = Path(root_dir).expanduser().resolve()137    parsed: list[ParsedCodeInfo] = []138 139    for rel_path in files:140        full_path = root / rel_path141        if full_path.is_file():142            parsed.append(parse_code_file(str(full_path)))143 144    return parsed