|
|
from pathlib import Path |
|
|
from typing import List, Dict, Optional, Set |
|
|
from dataclasses import dataclass |
|
|
import chardet |
|
|
|
|
|
@dataclass |
|
|
class FileInfo: |
|
|
path: Path |
|
|
size: int |
|
|
extension: str |
|
|
content: Optional[str] = None |
|
|
encoding: Optional[str] = None |
|
|
|
|
|
@property |
|
|
def formatted_size(self) -> str: |
|
|
if self.size < 1024: |
|
|
return f"{self.size} B" |
|
|
elif self.size < 1024 * 1024: |
|
|
return f"{self.size/1024:.1f} KB" |
|
|
else: |
|
|
return f"{self.size/(1024*1024):.1f} MB" |
|
|
|
|
|
class FileScanner: |
|
|
EXCLUDED_DIRS = { |
|
|
'.git', '__pycache__', 'node_modules', 'venv', |
|
|
'.env', 'build', 'dist', 'target', 'bin', 'obj' |
|
|
} |
|
|
|
|
|
def __init__(self, base_dir: Path, target_extensions: Set[str]): |
|
|
self.base_dir = base_dir |
|
|
self.target_extensions = target_extensions |
|
|
|
|
|
def _should_scan_file(self, path: Path) -> bool: |
|
|
if any(excluded in path.parts for excluded in self.EXCLUDED_DIRS): |
|
|
return False |
|
|
return path.suffix.lower() in self.target_extensions |
|
|
|
|
|
def _read_file_content(self, file_path: Path) -> tuple[Optional[str], Optional[str]]: |
|
|
try: |
|
|
with file_path.open('rb') as f: |
|
|
raw_data = f.read(4096) |
|
|
result = chardet.detect(raw_data) |
|
|
encoding = result['encoding'] if result['confidence'] > 0.7 else 'utf-8' |
|
|
|
