| 1 | from __future__ import annotations |
| 2 | |
| 3 | from abc import ABC, abstractmethod |
| 4 | from dataclasses import dataclass, field |
| 5 | |
| 6 | from app.services.normalize import RawRead |
| 7 | |
| 8 | |
| 9 | @dataclass |
| 10 | class ParseResult: |
| 11 | headers: list[str] |
| 12 | rows: list[RawRead] = field(default_factory=list) |
| 13 | errors: list[str] = field(default_factory=list) |
| 14 | |
| 15 | |
| 16 | class BaseParser(ABC): |
| 17 | """Every parser turns raw bytes into a ParseResult of RawRead rows + discovered headers. |
| 18 | Normalization happens downstream so parsers stay format-specific and dumb.""" |
| 19 | |
| 20 | name: str = "base" |
| 21 | |
| 22 | @abstractmethod |
| 23 | def parse(self, data: bytes) -> ParseResult: |
| 24 | ... |