57 lines
1.6 KiB
Python
57 lines
1.6 KiB
Python
"""OCR 面单识别响应模型 + 国家代码映射。"""
|
|
|
|
from __future__ import annotations
|
|
|
|
from typing import Any
|
|
|
|
from pydantic import BaseModel, ConfigDict
|
|
|
|
|
|
class OcrResult(BaseModel):
|
|
"""单张面单 OCR 识别结果。"""
|
|
|
|
model_config = ConfigDict(extra="allow")
|
|
|
|
file: str | None = None
|
|
layout: str | None = None
|
|
courier: str | None = None
|
|
courier_name: str | None = None
|
|
country: str | None = None
|
|
error: str | None = None
|
|
tracking: str | None = None
|
|
s_name: str | None = None
|
|
s_street: str | None = None
|
|
s_city: str | None = None
|
|
s_state: str | None = None
|
|
s_zip: str | None = None
|
|
r_name: str | None = None
|
|
r_street: str | None = None
|
|
r_city: str | None = None
|
|
r_state: str | None = None
|
|
r_zip: str | None = None
|
|
marker: str | None = None
|
|
|
|
@property
|
|
def ok(self) -> bool:
|
|
return not self.error
|
|
|
|
|
|
# OCR 服务已有分类器模型的国家(版面见 api.md §4)
|
|
SUPPORTED_COUNTRIES = {"US", "UK", "DE", "MX", "SA"}
|
|
|
|
# ISO 国家代码 / 中文国名 → OCR 国家代码
|
|
_COUNTRY_MAP: dict[str, str] = {
|
|
# 代码
|
|
"GB": "UK", "UK": "UK", "US": "US", "DE": "DE", "MX": "MX", "SA": "SA",
|
|
# 中文名
|
|
"英国": "UK", "美国": "US", "德国": "DE", "墨西哥": "MX", "沙特": "SA",
|
|
}
|
|
|
|
|
|
def country_to_ocr_code(country: Any) -> str | None:
|
|
"""将 ISO 代码或中文国名转为 OCR 服务支持的国家代码,不支持则返回 None。"""
|
|
if not country or not isinstance(country, str):
|
|
return None
|
|
code = _COUNTRY_MAP.get(country.strip().upper()) or _COUNTRY_MAP.get(country.strip())
|
|
return code if code in SUPPORTED_COUNTRIES else None
|