shaguabijia-app-server/app/integrations/sms.py

"""短信验证码服务。

两种运行模式由 `SMS_MOCK` 切换:
  - **mock**(开发/测试,默认):不真发短信,验证码打到日志;校验**放行任意 N 位数字**
    (测试/开发便利)。真实校验逻辑(比对存码 / 一次性 / 防爆破)由 real 分支 + 单测覆盖。
  - **real**(生产 `SMS_MOCK=false`):本服务生成 N 位验证码 → 调极光短信 REST
    `/v1/messages` 发送(自定义验证码模式,极光只负责发,code 由本服务生成/保管/
    校验)→ 鉴权复用极光一键登录的 `JG_APP_KEY`/`JG_MASTER_SECRET`(同一极光应用)。

验证码存储:**进程内存**(单 worker uvicorn 够用)。重启丢失(用户重发即可)。多
worker / 多机时内存不共享 → 冷却、每日上限、校验都会失效,届时迁移到 DB/Redis。
见 docs/待办与技术债.md。

防刷三层(短信花钱 + `/sms/send` 在登录前无法 JWT 鉴权):
  1. 单号 `SMS_SEND_INTERVAL_SEC` 冷却(本文件)
  2. 单号每日 `SMS_DAILY_LIMIT_PER_PHONE` 条上限(本文件)
  3. 单 IP 频控(api 层 rate_limit 依赖)+ 极光控制台 IP 白名单/防轰炸(运维侧)
另:单码校验失败 `SMS_MAX_VERIFY_ATTEMPTS` 次即作废(防爆破),验过即作废(一次性)。
"""
from __future__ import annotations

import base64
import logging
import secrets
import time
from dataclasses import dataclass
from datetime import datetime
from threading import Lock

import httpx

from app.core.config import settings

logger = logging.getLogger("shagua.sms")


class SmsError(Exception):
    """业务异常。`status_code` 决定 api 层翻成哪个 HTTP 码:
    过频/每日超限 = 429(客户端等会再来),供应商不可用 = 503,手机号无效 = 400。
    """

    def __init__(self, message: str, status_code: int = 429) -> None:
        super().__init__(message)
        self.status_code = status_code


@dataclass
class _CodeRecord:
    code: str
    expires_at: float
    attempts: int = 0


# 进程内存(单 worker 有效;多 worker 不共享,见模块 docstring)
_codes: dict[str, _CodeRecord] = {}            # phone -> 当前有效验证码
_last_sent: dict[str, float] = {}              # phone -> 上次发送 epoch(冷却)
_daily_count: dict[str, tuple[str, int]] = {}  # phone -> (date_str, 当日发送数)
_lock = Lock()
_GC_THRESHOLD = 10000  # 任一内存 dict 超此阈值,send 时顺手清过期项(防无限增长,仿 ratelimit)


def _today() -> str:
    return datetime.now().strftime("%Y-%m-%d")


def _gen_code() -> str:
    """生成 N 位数字验证码(用 secrets 而非 random;允许前导 0)。"""
    return "".join(secrets.choice("0123456789") for _ in range(settings.SMS_CODE_LENGTH))


def _gc(now: float) -> None:
    """顺手清理过期内存项,防三个 dict 无限增长。仅在持锁时调用,且某 dict 超
    _GC_THRESHOLD 才扫它(低频,开销可忽略)。"""
    if len(_codes) > _GC_THRESHOLD:
        for p in [p for p, r in _codes.items() if now > r.expires_at]:
            _codes.pop(p, None)
    if len(_last_sent) > _GC_THRESHOLD:
        cutoff = now - settings.SMS_SEND_INTERVAL_SEC
        for p in [p for p, ts in _last_sent.items() if ts < cutoff]:
            _last_sent.pop(p, None)
    if len(_daily_count) > _GC_THRESHOLD:
        today = _today()
        for p in [p for p, (d, _c) in _daily_count.items() if d != today]:
            _daily_count.pop(p, None)


def send_code(phone: str) -> int:
    """发送验证码。

    Returns: 距下次可发的秒数(= SMS_SEND_INTERVAL_SEC)
    Raises: SmsError(过频 429 / 当日超限 429 / 供应商失败 503 / 手机号无效 400)
    """
    now = time.time()

    # --- lock 内:防刷检查 + 预占(防并发重复发烧钱)---
    with _lock:
        _gc(now)  # 顺手清过期内存(超阈值才扫)
        elapsed = now - _last_sent.get(phone, 0.0)
        if elapsed < settings.SMS_SEND_INTERVAL_SEC:
            remain = int(settings.SMS_SEND_INTERVAL_SEC - elapsed)
            raise SmsError(f"发送过于频繁,请 {remain}s 后再试")

        today = _today()
        day, cnt = _daily_count.get(phone, ("", 0))
        if day != today:
            cnt = 0
        if cnt >= settings.SMS_DAILY_LIMIT_PER_PHONE:
            raise SmsError("今日验证码发送次数已达上限,请明天再试")

        code = _gen_code()
        # 预占:先记冷却/计数/存码,释放锁后再发网络(发失败保留冷却+计数,见下)
        _last_sent[phone] = now
        _daily_count[phone] = (today, cnt + 1)
        _codes[phone] = _CodeRecord(code=code, expires_at=now + settings.SMS_CODE_TTL_SEC)

    # --- lock 外:真正发送(网络 IO 不持锁)---
    try:
        if settings.SMS_MOCK:
            logger.info("[SMS-MOCK] to %s**** code=%s (不真发)", phone[:3], code)
        else:
            _send_via_jiguang(phone, code)
            logger.info("[SMS] sent to %s****", phone[:3])
    except Exception as e:
        # 发送失败:**保留冷却 + 每日计数**(失败也限速,挡住余额不足/签名失效时
        # 前端重试狂打极光),只清掉没发出去的码(用户收不到,留着无意义且占内存)。
        with _lock:
            _codes.pop(phone, None)
        if isinstance(e, SmsError):
            raise
        logger.exception("[SMS] send failed phone=%s****", phone[:3])
        raise SmsError("验证码发送失败,请稍后重试", status_code=503) from e

    return settings.SMS_SEND_INTERVAL_SEC


def verify_code(phone: str, code: str) -> bool:
    """校验验证码。

    - **mock 模式**:放行任意 N 位数字(测试/开发便利,不真校验)。
    - **real 模式**:比对本服务存的码,匹配即作废(一次性);失败累计到上限也作废(防爆破)。
    """
    if settings.SMS_MOCK:
        ok = len(code) == settings.SMS_CODE_LENGTH and code.isdigit()
        logger.info("[SMS-MOCK] verify %s for %s****", "ok" if ok else "fail", phone[:3])
        return ok

    with _lock:
        rec = _codes.get(phone)
        if rec is None:
            return False
        if time.time() > rec.expires_at:
            _codes.pop(phone, None)
            return False
        if rec.attempts >= settings.SMS_MAX_VERIFY_ATTEMPTS:
            _codes.pop(phone, None)  # 试错过多,作废
            return False
        if secrets.compare_digest(code.encode("utf-8"), rec.code.encode("utf-8")):
            _codes.pop(phone, None)  # 验过即作废
            return True
        rec.attempts += 1
        return False


def _send_via_jiguang(phone: str, code: str) -> None:
    """调极光短信 REST /v1/messages 发送(自定义验证码模式)。失败抛 SmsError。"""
    if not settings.JG_APP_KEY or not settings.JG_MASTER_SECRET:
        raise SmsError("短信服务未配置(缺 JG_APP_KEY/JG_MASTER_SECRET)", status_code=503)
    if not settings.SMS_SIGN_ID or not settings.SMS_TEMPLATE_ID:
        raise SmsError("短信服务未配置(缺 SMS_SIGN_ID/SMS_TEMPLATE_ID)", status_code=503)

    auth_b64 = base64.b64encode(
        f"{settings.JG_APP_KEY}:{settings.JG_MASTER_SECRET}".encode()
    ).decode()
    body = {
        "mobile": phone,
        "sign_id": settings.SMS_SIGN_ID,
        "temp_id": settings.SMS_TEMPLATE_ID,
        "temp_para": {"code": code},
    }
    try:
        resp = httpx.post(
            settings.SMS_SEND_ENDPOINT,
            json=body,
            headers={
                "Content-Type": "application/json",
                "Authorization": f"Basic {auth_b64}",
            },
            timeout=settings.JG_REQUEST_TIMEOUT_SEC,
        )
    except httpx.HTTPError as e:
        raise SmsError(f"短信网关网络错误: {e}", status_code=503) from e

    if resp.status_code == 200:
        return  # {"msg_id": ...}

    # 极光错误码映射(节选常见;完整码表见 docs/integrations/sms.md)
    try:
        err = resp.json().get("error", {})
        ecode, emsg = err.get("code"), err.get("message", "")
    except Exception:
        ecode, emsg = None, resp.text[:200]
    logger.error("[SMS] jiguang error http=%s code=%s msg=%s", resp.status_code, ecode, emsg)

    if ecode == 50014:  # 余额不足:运维要立即告警充值
        logger.critical("[SMS] 极光短信余额不足(50014),需充值!")
        raise SmsError("短信服务暂不可用,请稍后重试", status_code=503)
    if ecode == 50009:  # 极光侧超频
        raise SmsError("发送过于频繁,请稍后再试", status_code=429)
    if ecode == 50006:  # 手机号无效(schema 已挡格式,这里多是空号/停机)
        raise SmsError("手机号无效", status_code=400)
    raise SmsError(f"短信发送失败(code={ecode})", status_code=503)