from dataclasses import dataclass from typing import Union class LexError(Exception): pass @dataclass class Token: kind: str value: Union[int, float, str, None] def __repr__(self): return f"{self.kind}({self.value!r})" _SINGLE = { '+': 'PLUS', '-': 'MINUS', '*': 'STAR', '/': 'SLASH', '(': 'LPAREN', ')': 'RPAREN', } def tokenize(src: str) -> list: tokens = [] i = 0 while i < len(src): ch = src[i] if ch in ' \t': i += 1 continue if ch in _SINGLE: tokens.append(Token(_SINGLE[ch], ch)) i += 1 continue if ch.isdigit() or ch == '.': j = i while j < len(src) and (src[j].isdigit() or src[j] == '.'): j += 1 num_str = src[i:j] if '.' in num_str: value = float(num_str) else: value = int(num_str) tokens.append(Token('NUMBER', value)) i = j continue raise LexError(f"unexpected character {ch!r} at position {i}") tokens.append(Token('EOF', None)) return tokens