Genarrative/scripts/deploy/production-api-deploy.sh

#!/usr/bin/env bash

set -euo pipefail

usage() {
  cat <<'EOF'
用法:
  ./scripts/deploy/production-api-deploy.sh --source-dir build/<version> [--version <version>] [--release-root /opt/genarrative/releases] [--current-link /opt/genarrative/current] [--service genarrative-api.service] [--health-url http://127.0.0.1:8082/readyz] [--api-env-file /etc/genarrative/api-server.env] [--database genarrative-prod] [--spacetime-server-url http://127.0.0.1:3101]

说明:
  进入维护模式，校验并发布 api-server 单文件，更新 current 链接，重启 systemd 服务并执行 readiness 检查。
  若传入 --database，会在重启前把 GENARRATIVE_SPACETIME_DATABASE 写入 api-server 环境文件，避免服务继续读取旧库。
  失败时保留维护模式。
EOF
}

require_argument() {
  local value="$1"
  local label="$2"

  if [[ -z "${value}" ]]; then
    echo "[production-api-deploy] 缺少参数: ${label}" >&2
    exit 1
  fi
}

validate_spacetime_database_name() {
  local database="$1"

  if [[ ! "${database}" =~ ^[a-z0-9]+(-[a-z0-9]+)*$ ]]; then
    echo "[production-api-deploy] --database 必须匹配 SpacetimeDB 数据库名规则 ^[a-z0-9]+(-[a-z0-9]+)*$，只能使用小写字母、数字，并用单个短横线分隔: ${database}" >&2
    exit 1
  fi
}

write_env_value() {
  local file_path="$1"
  local key="$2"
  local value="$3"

  local can_write_direct=0

  if [[ -e "${file_path}" ]]; then
    [[ -w "${file_path}" ]] && can_write_direct=1
  else
    mkdir -p "$(dirname "${file_path}")"
    [[ -w "$(dirname "${file_path}")" ]] && can_write_direct=1
  fi

  # api-server 环境文件通常由 server-provision 以 root:root 0600 创建。
  # 发布流水线可能以非 root Jenkins 用户运行，因此仅在不能直接写入时使用 sudo -n，避免因为 env 文件权限导致 API 发布中断。
  if [[ "$(id -u)" -eq 0 || "${can_write_direct}" -eq 1 ]]; then
    python3 - "${file_path}" "${key}" "${value}" <<'PY'
import os
import sys
from pathlib import Path

file_path = Path(sys.argv[1])
key = sys.argv[2]
value = sys.argv[3]
was_missing = not file_path.exists()
file_path.parent.mkdir(parents=True, exist_ok=True)

lines = []
if file_path.exists():
    lines = file_path.read_text(encoding="utf-8").splitlines()

updated = False
next_lines = []
for line in lines:
    if line.startswith(f"{key}="):
        next_lines.append(f"{key}={value}")
        updated = True
    else:
        next_lines.append(line)

if not updated:
    next_lines.append(f"{key}={value}")

file_path.write_text("\n".join(next_lines) + "\n", encoding="utf-8")
if was_missing:
    os.chmod(file_path, 0o600)
PY
  else
    if ! sudo -n true >/dev/null 2>&1; then
      echo "[production-api-deploy] 当前用户无权写入 ${file_path}，且 sudo -n 不可用；请给部署用户配置免密写入该环境文件或以 root 执行发布。" >&2
      exit 1
    fi
    sudo -n python3 - "${file_path}" "${key}" "${value}" <<'PY'
import os
import sys
from pathlib import Path

file_path = Path(sys.argv[1])
key = sys.argv[2]
value = sys.argv[3]
was_missing = not file_path.exists()
file_path.parent.mkdir(parents=True, exist_ok=True)

lines = []
if file_path.exists():
    lines = file_path.read_text(encoding="utf-8").splitlines()

updated = False
next_lines = []
for line in lines:
    if line.startswith(f"{key}="):
        next_lines.append(f"{key}={value}")
        updated = True
    else:
        next_lines.append(line)

if not updated:
    next_lines.append(f"{key}={value}")

file_path.write_text("\n".join(next_lines) + "\n", encoding="utf-8")
if was_missing:
    os.chmod(file_path, 0o600)
PY
  fi
}

read_env_value() {
  local file_path="$1"
  local key="$2"

  if [[ ! -f "${file_path}" ]]; then
    return 0
  fi

  local python_script='
import sys
from pathlib import Path

path = Path(sys.argv[1])
key = sys.argv[2]
if not path.exists():
    raise SystemExit(0)
for raw_line in path.read_text(encoding="utf-8").splitlines():
    line = raw_line.strip()
    if not line or line.startswith("#") or "=" not in line:
        continue
    current_key, value = line.split("=", 1)
    if current_key == key:
        value = value.strip()
        if len(value) >= 2 and value[0] == value[-1] and value[0] in ("\"", "'\''"):
            value = value[1:-1]
        print(value)
        raise SystemExit(0)
'

  if [[ -r "${file_path}" ]]; then
    python3 -c "${python_script}" "${file_path}" "${key}"
  else
    if ! sudo -n true >/dev/null 2>&1; then
      echo "[production-api-deploy] 当前用户无权读取 ${file_path}，且 sudo -n 不可用；无法检查运行态环境变量。" >&2
      exit 1
    fi
    sudo -n python3 -c "${python_script}" "${file_path}" "${key}"
  fi
}

ensure_env_value() {
  local file_path="$1"
  local key="$2"
  local default_value="$3"
  local current_value

  current_value="$(read_env_value "${file_path}" "${key}")"
  if [[ -n "${current_value}" ]]; then
    return
  fi

  echo "[production-api-deploy] 补齐 api-server 环境变量: ${key} -> ${file_path}"
  write_env_value "${file_path}" "${key}" "${default_value}"
}

run_privileged() {
  if [[ "$(id -u)" -eq 0 ]]; then
    "$@"
    return
  fi
  if ! sudo -n true >/dev/null 2>&1; then
    echo "[production-api-deploy] 当前用户不是 root，且 sudo -n 不可用；无法执行: $*" >&2
    exit 1
  fi
  sudo -n "$@"
}

ensure_runtime_dir() {
  local path="$1"
  local mode="$2"

  if [[ -z "${path}" ]]; then
    return
  fi
  if [[ "${path}" != /* ]]; then
    echo "[production-api-deploy] 运行态目录必须使用绝对路径，避免写入只读发布目录: ${path}" >&2
    exit 1
  fi

  echo "[production-api-deploy] 确保运行态目录可写: ${path}"
  run_privileged install -d -o genarrative -g genarrative -m "${mode}" "${path}"
}

ensure_runtime_env_and_dirs() {
  local api_env_file="$1"
  local tracking_enabled tracking_outbox_dir

  # 旧生产环境文件会被 server-provision 保留，不一定包含新增的运行态写入路径。
  # 发布前只补缺省值，不覆盖线上已经定制过的目录或开关。
  ensure_env_value "${api_env_file}" "GENARRATIVE_API_SHUTDOWN_OUTBOX_FLUSH_TIMEOUT_MS" "5000"
  ensure_env_value "${api_env_file}" "GENARRATIVE_TRACKING_OUTBOX_ENABLED" "true"
  ensure_env_value "${api_env_file}" "GENARRATIVE_TRACKING_OUTBOX_DIR" "/var/lib/genarrative/tracking-outbox"
  ensure_env_value "${api_env_file}" "GENARRATIVE_TRACKING_OUTBOX_BATCH_SIZE" "500"
  ensure_env_value "${api_env_file}" "GENARRATIVE_TRACKING_OUTBOX_FLUSH_INTERVAL_MS" "1000"
  ensure_env_value "${api_env_file}" "GENARRATIVE_TRACKING_OUTBOX_MAX_BYTES" "268435456"

  tracking_enabled="$(read_env_value "${api_env_file}" "GENARRATIVE_TRACKING_OUTBOX_ENABLED")"
  tracking_outbox_dir="$(read_env_value "${api_env_file}" "GENARRATIVE_TRACKING_OUTBOX_DIR")"
  if [[ "$(printf "%s" "${tracking_enabled}" | tr '[:upper:]' '[:lower:]')" != "false" ]]; then
    ensure_runtime_dir "${tracking_outbox_dir}" "0750"
  fi
}

SCRIPT_DIR="$(cd -- "$(dirname -- "${BASH_SOURCE[0]}")" && pwd)"
SOURCE_DIR=""
VERSION=""
RELEASE_ROOT="/opt/genarrative/releases"
CURRENT_LINK="/opt/genarrative/current"
SERVICE_NAME="genarrative-api.service"
HEALTH_URL="http://127.0.0.1:8082/readyz"
API_ENV_FILE="/etc/genarrative/api-server.env"
DATABASE=""
SPACETIME_SERVER_URL=""
DEPLOY_COMPLETED=0

while [[ $# -gt 0 ]]; do
  case "$1" in
    -h|--help)
      usage
      exit 0
      ;;
    --source-dir)
      SOURCE_DIR="${2:?缺少 --source-dir 的值}"
      shift 2
      ;;
    --version)
      VERSION="${2:?缺少 --version 的值}"
      shift 2
      ;;
    --release-root)
      RELEASE_ROOT="${2:?缺少 --release-root 的值}"
      shift 2
      ;;
    --current-link)
      CURRENT_LINK="${2:?缺少 --current-link 的值}"
      shift 2
      ;;
    --service)
      SERVICE_NAME="${2:?缺少 --service 的值}"
      shift 2
      ;;
    --health-url)
      HEALTH_URL="${2:?缺少 --health-url 的值}"
      shift 2
      ;;
    --api-env-file)
      API_ENV_FILE="${2:?缺少 --api-env-file 的值}"
      shift 2
      ;;
    --database)
      DATABASE="${2:?缺少 --database 的值}"
      shift 2
      ;;
    --spacetime-server-url)
      SPACETIME_SERVER_URL="${2:?缺少 --spacetime-server-url 的值}"
      shift 2
      ;;
    *)
      echo "[production-api-deploy] 未知参数: $1" >&2
      usage >&2
      exit 1
      ;;
  esac
done

require_argument "${SOURCE_DIR}" "--source-dir"

if [[ -n "${DATABASE}" ]]; then
  validate_spacetime_database_name "${DATABASE}"
fi

if [[ ! -d "${SOURCE_DIR}" ]]; then
  echo "[production-api-deploy] 发布目录不存在: ${SOURCE_DIR}" >&2
  exit 1
fi

SOURCE_DIR="$(cd "${SOURCE_DIR}" && pwd)"
VERSION="${VERSION:-$(basename "${SOURCE_DIR}")}"

if [[ ! "${VERSION}" =~ ^[0-9A-Za-z._-]+$ ]]; then
  echo "[production-api-deploy] --version 只能包含数字、字母、点、下划线和短横线: ${VERSION}" >&2
  exit 1
fi

if [[ ! -f "${SOURCE_DIR}/api-server" || ! -f "${SOURCE_DIR}/api-server.sha256" ]]; then
  echo "[production-api-deploy] 缺少 api-server 或 api-server.sha256: ${SOURCE_DIR}" >&2
  exit 1
fi

on_exit() {
  local exit_code=$?
  if [[ "${exit_code}" -ne 0 && "${DEPLOY_COMPLETED}" -ne 1 ]]; then
    echo "[production-api-deploy] 部署失败，保持维护模式。" >&2
  fi
  exit "${exit_code}"
}

trap on_exit EXIT

"${SCRIPT_DIR}/maintenance-on.sh" "api deploy ${VERSION}"

echo "[production-api-deploy] 校验 api-server"
(
  cd "${SOURCE_DIR}"
  sha256sum -c api-server.sha256
)

RELEASE_DIR="${RELEASE_ROOT}/${VERSION}"
mkdir -p "${RELEASE_DIR}"
cp "${SOURCE_DIR}/api-server" "${RELEASE_DIR}/api-server"
chmod +x "${RELEASE_DIR}/api-server"

BACKUP_SCRIPT_SOURCE="${SOURCE_DIR}/scripts/database-backup-to-oss.mjs"
WORKSPACE_BACKUP_SCRIPT_SOURCE="$(cd "${SCRIPT_DIR}/../.." && pwd)/scripts/database-backup-to-oss.mjs"
HEALTH_PATROL_SCRIPT_SOURCE="${SOURCE_DIR}/scripts/ops/production-health-patrol.mjs"
WORKSPACE_HEALTH_PATROL_SCRIPT_SOURCE="$(cd "${SCRIPT_DIR}/../.." && pwd)/scripts/ops/production-health-patrol.mjs"
mkdir -p "${RELEASE_DIR}/scripts" "${RELEASE_DIR}/scripts/ops"
if [[ ! -f "${BACKUP_SCRIPT_SOURCE}" ]]; then
  if [[ -f "${WORKSPACE_BACKUP_SCRIPT_SOURCE}" ]]; then
    echo "[production-api-deploy] 发布产物缺少 scripts/database-backup-to-oss.mjs，回退使用部署工作区脚本；请重新触发包含该脚本的 API 构建。" >&2
    BACKUP_SCRIPT_SOURCE="${WORKSPACE_BACKUP_SCRIPT_SOURCE}"
  else
    echo "[production-api-deploy] 缺少数据库备份脚本: ${SOURCE_DIR}/scripts/database-backup-to-oss.mjs" >&2
    exit 1
  fi
fi
cp "${BACKUP_SCRIPT_SOURCE}" "${RELEASE_DIR}/scripts/database-backup-to-oss.mjs"
chmod 0644 "${RELEASE_DIR}/scripts/database-backup-to-oss.mjs"
if [[ ! -f "${HEALTH_PATROL_SCRIPT_SOURCE}" ]]; then
  if [[ -f "${WORKSPACE_HEALTH_PATROL_SCRIPT_SOURCE}" ]]; then
    echo "[production-api-deploy] 发布产物缺少 scripts/ops/production-health-patrol.mjs，回退使用部署工作区脚本；请重新触发包含该脚本的 API 构建。" >&2
    HEALTH_PATROL_SCRIPT_SOURCE="${WORKSPACE_HEALTH_PATROL_SCRIPT_SOURCE}"
  else
    echo "[production-api-deploy] 未找到生产健康巡检脚本，跳过复制；genarrative-health-patrol.service 会因脚本缺失而跳过执行。" >&2
    HEALTH_PATROL_SCRIPT_SOURCE=""
  fi
fi
if [[ -n "${HEALTH_PATROL_SCRIPT_SOURCE}" ]]; then
  cp "${HEALTH_PATROL_SCRIPT_SOURCE}" "${RELEASE_DIR}/scripts/ops/production-health-patrol.mjs"
  chmod 0644 "${RELEASE_DIR}/scripts/ops/production-health-patrol.mjs"
fi

if [[ -f "${SOURCE_DIR}/release-manifest.json" ]]; then
  cp "${SOURCE_DIR}/release-manifest.json" "${RELEASE_DIR}/release-manifest.api-server.json"
fi

if [[ -n "${DATABASE}" ]]; then
  echo "[production-api-deploy] 写入 api-server SpacetimeDB database: ${DATABASE} -> ${API_ENV_FILE}"
  write_env_value "${API_ENV_FILE}" "GENARRATIVE_SPACETIME_DATABASE" "${DATABASE}"
fi

if [[ -n "${SPACETIME_SERVER_URL}" ]]; then
  echo "[production-api-deploy] 写入 api-server SpacetimeDB server: ${SPACETIME_SERVER_URL} -> ${API_ENV_FILE}"
  write_env_value "${API_ENV_FILE}" "GENARRATIVE_SPACETIME_SERVER_URL" "${SPACETIME_SERVER_URL}"
fi

ensure_runtime_env_and_dirs "${API_ENV_FILE}"

mkdir -p "$(dirname "${CURRENT_LINK}")"
ln -sfn "${RELEASE_DIR}" "${CURRENT_LINK}"

echo "[production-api-deploy] 重启服务: ${SERVICE_NAME}"
systemctl restart "${SERVICE_NAME}"

echo "[production-api-deploy] 等待 readiness: ${HEALTH_URL}"
for _ in {1..30}; do
  if curl -fsS "${HEALTH_URL}" >/dev/null; then
    "${SCRIPT_DIR}/maintenance-off.sh"
    DEPLOY_COMPLETED=1
    echo "[production-api-deploy] 完成: ${RELEASE_DIR}/api-server"
    exit 0
  fi
  sleep 2
done

echo "[production-api-deploy] readiness 检查超时: ${HEALTH_URL}" >&2
exit 1