标准值类型与本体函数：把字段口径管成规范

1. 打开"值类型目录"（GET /api/v1/ontology/value-types）
2. 逐类看 label / description / sql_type / sql_type_pg / validate_rules
3. 对照 order 对象的 total / status 属性，标记可升级项
4. 确认旧值兼容：varchar→text、float→number 等历史别名读取时规范化，不写回存储

{
  "code": 0,
  "data": [
    { "type": "text", "label": "文本", "description": "任意字符串",
      "sql_type": "TEXT", "sql_type_pg": "TEXT", "validate_rules": ["任意字符串"] },
    { "type": "decimal", "label": "小数", "description": "十进制小数，可用 constraints.precision 指定精度",
      "sql_type": "NUMERIC", "sql_type_pg": "NUMERIC",
      "validate_rules": ["数值", "constraints.precision 指定小数位"] },
    { "type": "currency", "label": "金额", "description": "货币金额",
      "sql_type": "NUMERIC", "sql_type_pg": "NUMERIC",
      "validate_rules": ["数值", "constraints.unit 必填（币种，如 CNY/USD）"] }
  ]
}

1. POST /api/v1/ontology/value-types/normalize，body 传 {"dry_run": true}
2. 逐条看 diff 的 reason：text+enum→enum、number+百分比单位→percentage、number+币种 unit→currency、precision 0→integer、precision>0→decimal
3. 确认建议无误后 POST body 传 {"dry_run": false}
4. 到对象详情核对 total / status 的 data_type 已更新

POST /api/v1/ontology/value-types/normalize
{ "dry_run": true }
->
{ "code": 0, "data": {
    "dry_run": true,
    "diff": [
      { "property_id": 1, "object_type_name": "order",
        "property_name": "total", "old_type": "number",
        "new_type": "decimal", "reason": "precision>0（小数值域）" },
      { "property_id": 2, "object_type_name": "order",
        "property_name": "status", "old_type": "text",
        "new_type": "enum", "reason": "text+constraints.enum→enum" }
    ],
    "applied": 0 } }

1. POST /api/v1/ontology/functions 创建 compute_tax：params 声明 x / rate（必选），body=round({{x}} * {{rate}}, 2)
2. 创建时校验 name 正则与 body 可编译（CheckExpressionSyntax）
3. POST /api/v1/ontology/functions/compute_tax/test，body 传 {"args":{"x":100,"rate":0.06}}
4. 看返回的 result 即时求值结果

POST /api/v1/ontology/functions
{ "name": "compute_tax", "description": "按税率计算税额",
  "params": [ { "name": "x", "type": "number", "required": true },
              { "name": "rate", "type": "number", "required": true } ],
  "return_type": "number", "body": "round({{x}} * {{rate}}, 2)" }
->
{ "code": 0, "data": { "id": "7f3a...", "name": "compute_tax",
    "return_type": "number", "body": "round({{x}} * {{rate}}, 2)",
    "pure": 1, "enabled": 1 } }

POST /api/v1/ontology/functions/compute_tax/test
{ "args": { "x": 100, "rate": 0.06 } }
->
{ "code": 0, "data": { "name": "compute_tax", "result": 6 } }

1. 先注册并测试 compute_tax 函数
2. 在 order 草稿中加计算属性 tax_amount：data_type=number、is_calculated=true、formula=fn_compute_tax(amount, 0.06)（不能与 mapped_column 同设）
3. 提交 review → merge，到对象查询看该列
4. 在 workflow 条件里写 if(fn_compute_tax({{amount}}, 0.06) > 100, 1, 0) 并运行验证

{
  "name": "tax_amount", "display_name": "税额",
  "data_type": "number", "is_calculated": true,
  "formula": "fn_compute_tax(amount, 0.06)",
  "mapped_column": ""
}

{ "code": 0, "data": {
  "columns": ["order_id", "tax_amount"],
  "rows": [[1, null]] } }