如何处理 tde 模板中的空数组类型标量

【问题标题】：How to handle empty array-type scalar in tde template如何处理 tde 模板中的空数组类型标量
【发布时间】：2022-01-21 11:32:26
【问题描述】：

我需要使用 tde 模板处理一些有时缺少数据的文档。但仍需要在表格中显示。但是我的模板在返回带有空数据的文档时遇到问题。

 var doc1 = xdmp.toJSON(
    {
          "customer":{
            "Name": "Name1", 
            "Addr": [
              {
                "AddrTypeCd": "MailingAddress", 
                "Addr1": "911 FORBES AVE", 
                "Addr2": "SUITE XXX", 
                "CityName": "asdfasfd", 
                "StateProvCd": "PA", 
                "PostalCode": "15219"
              }, 
              {
                "AddrTypeCd": "OfficeAddress", 
                "Addr1": "911 Watson AVE", 
                "Addr2": "SUITE XXX", 
                "CityName": "asdfasfd", 
                "StateProvCd": "CT", 
                "PostalCode": "15119"
              }
            ]
          }
        });

    var doc2 = xdmp.toJSON(
        {
          "customer":{
            "Name": "Name2", 
            "Addr": [
            ]
          }
        });

  var rowtde1 = xdmp.toJSON(
  {
    "template":{
      "context":"/customer/Addr",
      "rows":[
        {
          "schemaName":"Schemas",
          "viewName":"CustomerAddress",
          "columns":[
            {
              "name":"CustomerName",
              "scalarType":"string",
              "val":"../../Name"
            },
            {
              "name":"AddrTypeCd",
              "scalarType":"string",
              "val":"AddrTypeCd",
              "nullable": true
            },
            {
              "name":"Addr1",
              "scalarType":"string",
              "val":"Addr1",
              "nullable": true
            },
            {
              "name":"Addr2",
              "scalarType":"string",
              "val":"Addr2",
              "nullable": true
            },
            {
              "name":"CityName",
              "scalarType":"string",
              "val":"CityName",
              "nullable": true
            },
            {
              "name":"StateProvCd",
              "scalarType":"string",
              "val":"StateProvCd",
              "nullable": true
            },
            {
              "name":"PostalCode",
              "scalarType":"string",
              "val":"PostalCode",
              "nullable": true
            }
          ]
        }
      ]
    }
  }
  );



tde.nodeDataExtract([doc1, doc2],[rowtde1]);

对于此数据提取，响应为：

{
  "document1": [
    {
      "row": {
        "schema": "Schemas",
        "view": "CustomerAddress",
        "data": {
          "rownum": "1",
          "CustomerName": "Name1",
          "AddrTypeCd": "MailingAddress",
          "Addr1": "911 FORBES AVE",
          "Addr2": "SUITE XXX",
          "CityName": "asdfasfd",
          "StateProvCd": "PA",
          "PostalCode": "15219"
        }
      }
    },
    {
      "row": {
        "schema": "Schemas",
        "view": "CustomerAddress",
        "data": {
          "rownum": "2",
          "CustomerName": "Name1",
          "AddrTypeCd": "OfficeAddress",
          "Addr1": "911 Watson AVE",
          "Addr2": "SUITE XXX",
          "CityName": "asdfasfd",
          "StateProvCd": "CT",
          "PostalCode": "15119"
        }
      }
    }
  ],
  "document2": []
}

但我也需要为没有地址的客户获取响应数据，如下所示：

{   "document1": [
    {
      "row": {
        "schema": "Schemas",
        "view": "CustomerAddress",
        "data": {
          "rownum": "1",
          "CustomerName": "Name1",
          "AddrTypeCd": "MailingAddress",
          "Addr1": "911 FORBES AVE",
          "Addr2": "SUITE XXX",
          "CityName": "asdfasfd",
          "StateProvCd": "PA",
          "PostalCode": "15219"
        }
      }
    },
    {
      "row": {
        "schema": "Schemas",
        "view": "CustomerAddress",
        "data": {
          "rownum": "2",
          "CustomerName": "Name1",
          "AddrTypeCd": "OfficeAddress",
          "Addr1": "911 Watson AVE",
          "Addr2": "SUITE XXX",
          "CityName": "asdfasfd",
          "StateProvCd": "CT",
          "PostalCode": "15119"
        }
      }
    }   ],   "document2": [    {
      "row": {
        "schema": "Schemas",
        "view": "CustomerAddress",
        "data": {
          "rownum": "1",
          "CustomerName": "Name2"
        }
      }
    }  ] }

我是否需要创建第二个视图来处理文档中的名称，然后将其与地址视图相结合？多个联接是否会对性能产生影响？

【问题讨论】：

    标签： marklogic marklogic-9 marklogic-10


    【解决方案1】：

    通过一点点 xPath 魔法和一些 TDE 模板设置，您实际上可以使用嵌套模板和稀疏定义来做到这一点。

    主模板

    • 无地址客户的模板
    • 带有地址的客户模板

    请注意：这适用于系统中文档的/ 上下文。但是，为此，您需要一个目录或集合范围。这不能在线欺骗，所以我添加了根元素以允许它在不保存文档的情况下作为示例工作。只要您还使用目录或集合限定它们，您就可以按定义的方式保存文档（无根）。

     var doc1 = xdmp.toJSON(
{
    "root" : {
      "customer":{
        "Name": "Name1", 
        "Addr": [
          {
            "AddrTypeCd": "MailingAddress", 
            "Addr1": "911 FORBES AVE", 
            "Addr2": "SUITE XXX", 
            "CityName": "asdfasfd", 
            "StateProvCd": "PA", 
            "PostalCode": "15219"
          }, 
          {
            "AddrTypeCd": "OfficeAddress", 
            "Addr1": "911 Watson AVE", 
            "Addr2": "SUITE XXX", 
            "CityName": "asdfasfd", 
            "StateProvCd": "CT", 
            "PostalCode": "15119"
          }
        ]
      }
    } 
    });

var doc2 = xdmp.toJSON(
  {"root": 
     {
      "customer":{
        "Name": "Name2", 
        "Addr": [
        ]
      }
  }     
    });

var rowtde1 = xdmp.toJSON(
{
"template":{
  "context":"/root",
  "templates" : [
    {
      "context":"customer[fn:count(./Addr/*) = 0]/Name",
        "rows":[
          {
            "schemaName":"Schemas",
            "viewLayout": "sparse",
            "viewName":"CustomerAddress",
            "columns":[
              {
                "name":"CustomerName",
                "scalarType":"string",
                "val":"."
              }
            ]
          }
        ]
    },
            {
      "context":"customer[fn:count(./Addr/*) > 0]/Addr",
        "rows":[
          {
            "schemaName":"Schemas",
            "viewLayout": "sparse",
            "viewName":"CustomerAddress",
            "columns":[
              {
                "name":"CustomerName",
                "scalarType":"string",
                "val":"../../Name"
              },
              {
                "name":"AddrTypeCd",
                "scalarType":"string",
                "val":"AddrTypeCd",
                "nullable": true
              },
              {
                "name":"Addr1",
                "scalarType":"string",
                "val":"Addr1",
                "nullable": true
              },
              {
                "name":"Addr2",
                "scalarType":"string",
                "val":"Addr2",
                "nullable": true
              },
              {
                "name":"CityName",
                "scalarType":"string",
                "val":"CityName",
                "nullable": true
              },
              {
                "name":"StateProvCd",
                "scalarType":"string",
                "val":"StateProvCd",
                "nullable": true
              },
              {
                "name":"PostalCode",
                "scalarType":"string",
                "val":"PostalCode",
                "nullable": true
              }
            ]
          }
        ]
    }
  ]
}
}
);
tde.nodeDataExtract([doc1, doc2],[rowtde1]);

    仅名称行的结果只是单个值。稀疏模板允许这样做。您也可以将整个列列表定义为 null，但我没有看到该值。

    {
"document1": [{
        "row": {
            "schema": "Schemas",
            "view": "CustomerAddress",
            "data": {
                "rownum": "1",
                "CustomerName": "Name1",
                "AddrTypeCd": "MailingAddress",
                "Addr1": "911 FORBES AVE",
                "Addr2": "SUITE XXX",
                "CityName": "asdfasfd",
                "StateProvCd": "PA",
                "PostalCode": "15219"
            }
        }
    },
    {
        "row": {
            "schema": "Schemas",
            "view": "CustomerAddress",
            "data": {
                "rownum": "2",
                "CustomerName": "Name1",
                "AddrTypeCd": "OfficeAddress",
                "Addr1": "911 Watson AVE",
                "Addr2": "SUITE XXX",
                "CityName": "asdfasfd",
                "StateProvCd": "CT",
                "PostalCode": "15119"
            }
        }
    }
],
"document2": [{
    "row": {
        "schema": "Schemas",
        "view": "CustomerAddress",
        "data": {
            "rownum": "1",
            "CustomerName": "Name2"
        }
    }
}]
}

    【讨论】：

    • 加入不一定便宜。在分布式数据库中，它们可能更昂贵。在森林上进行过滤、连接等，然后是节点，然后是本地无法解决的问题，传输到 E 节点。根据连接的不同，这可能会导致大量内容跨节点传输，只是为了解决查询。在像 MarkLogic 这样的数据库中，按照您的意图来构建您的内容。我会尝试尽可能接近相同的 TDE 结构。试图在结构化内容的视图上达到 3NF 将是一个倒退
    【解决方案2】：

    由于有一个项目数组，您应该创建第二个模板来表达这些行。您将看到主要客户的视图和地址的视图。最好有一个唯一的密钥来加入。拥有这两个视图后，您可以使用 SQL 或 Optic API（MarkLogic 的多模型查询库）进行查询和连接。

    {
    "template": {
        "description": "test template",
        "context": "customer",
        "rows": [
            {
                "schemaName": "Schemas",
                "viewName": "Customer",
                "columns": [
                    {
                        "name": "id",
                        "scalarType": "int",
                        "val": "./id"
                    },
                    {
                        "name": "CustomerName",
                        "scalarType": "string",
                        "val": "./Name"
                    }
                ]
            }
        ],
        "templates": [
            {
                "context": "./Addr",
                "rows": [
                    {
                        "schemaName": "Schemas",
                        "viewName": "CustomerAddr",
                        "viewLayout": "sparse",
                        "columns": [
                            {
                                "name": "customerId",
                                "scalarType": "int",
                                "val": "../../id"
                            },
                            {
                                "name": "AddrTypeCd",
                                "scalarType": "string",
                                "val": "AddrTypeCd",
                                "nullable": true
                            },
                            {
                                "name": "Addr1",
                                "scalarType": "string",
                                "val": "Addr1",
                                "nullable": true
                            },
                            {
                                "name": "Addr2",
                                "scalarType": "string",
                                "val": "Addr2",
                                "nullable": true
                            },
                            {
                                "name": "CityName",
                                "scalarType": "string",
                                "val": "CityName",
                                "nullable": true
                            },
                            {
                                "name": "StateProvCd",
                                "scalarType": "string",
                                "val": "StateProvCd",
                                "nullable": true
                            },
                            {
                                "name": "PostalCode",
                                "scalarType": "string",
                                "val": "PostalCode",
                                "nullable": true
                            }
                        ]
                    }
                ]
            }
        ]
    }
}

    Optic API 查询示例

    'use strict';

const op = require('/MarkLogic/optic')

let customer = op.fromView('Schemas', 'Customer');
let addr = op.fromView('Schemas', 'CustomerAddr');

customer.joinLeftOuter(addr, op.on(op.col('id'), op.col('customerId'))).limit(10).result()

    结果

    {
    "Schemas.Customer.id": 1,
    "Schemas.CustomerAddr.customerId": 1,
    "Schemas.Customer.CustomerName": "Name1",
    "Schemas.CustomerAddr.AddrTypeCd": "MailingAddress",
    "Schemas.CustomerAddr.Addr1": "911 FORBES AVE",
    "Schemas.CustomerAddr.Addr2": "SUITE XXX",
    "Schemas.CustomerAddr.CityName": "asdfasfd",
    "Schemas.CustomerAddr.StateProvCd": "PA",
    "Schemas.CustomerAddr.PostalCode": "15219"
}

{
    "Schemas.Customer.id": 1,
    "Schemas.CustomerAddr.customerId": 1,
    "Schemas.Customer.CustomerName": "Name1",
    "Schemas.CustomerAddr.AddrTypeCd": "OfficeAddress",
    "Schemas.CustomerAddr.Addr1": "911 Watson AVE",
    "Schemas.CustomerAddr.Addr2": "SUITE XXX",
    "Schemas.CustomerAddr.CityName": "asdfasfd",
    "Schemas.CustomerAddr.StateProvCd": "CT",
    "Schemas.CustomerAddr.PostalCode": "15119"
} 

{
    "Schemas.Customer.id": 2,
    "Schemas.CustomerAddr.customerId": null,
    "Schemas.Customer.CustomerName": "Name2",
    "Schemas.CustomerAddr.AddrTypeCd": null,
    "Schemas.CustomerAddr.Addr1": null,
    "Schemas.CustomerAddr.Addr2": null,
    "Schemas.CustomerAddr.CityName": null,
    "Schemas.CustomerAddr.StateProvCd": null,
    "Schemas.CustomerAddr.PostalCode": null
}

{
    "Schemas.Customer.id": 3,
    "Schemas.CustomerAddr.customerId": null,
    "Schemas.Customer.CustomerName": "Name3",
    "Schemas.CustomerAddr.AddrTypeCd": null,
    "Schemas.CustomerAddr.Addr1": null,
    "Schemas.CustomerAddr.Addr2": null,
    "Schemas.CustomerAddr.CityName": null,
    "Schemas.CustomerAddr.StateProvCd": null,
    "Schemas.CustomerAddr.PostalCode": null
}

    【讨论】：

      【解决方案3】：

      如果您没有自然连接键，则可以在两个视图中添加类似下面的内容..

      {
 "name": "joinKey",
 "scalarType": "string",
 "val": "xdmp:node-uri(.) || '#' || fn:position()"
}

      安德鲁的反应仍然正确。需要的是创建两个视图，这就是如何使用 TDE 展平嵌入式阵列。 您也可以使用 SQL 语句进行快速测试，如下所示

      select * from Schemas.Customer cus
LEFT  JOIN Schemas.CustomerAddr addr ON cus.joinKey = addr.joinKey

      和

      select * from Schemas.Customer cus,
Schemas.CustomerAddr addr 
where cus.joinKey = addr.joinKey

      【讨论】：

        猜你喜欢
        • 1970-01-01
        • 1970-01-01
        • 1970-01-01
        • 2011-12-07
        • 1970-01-01
        • 2010-12-22
        • 1970-01-01
        • 1970-01-01
        • 1970-01-01
        相关资源
        最近更新 更多
        热门标签
        Java Python linux javascript Mysql C# Docker 算法 前端 SpringBoot Redis Vue spring 设计模式 .net core .net kubernetes c++ 数据库 数据结构 大数据 js 机器学习 微服务 Android Go 程序员 面试 JVM ASP.net core 云原生 人工智能 后端 PHP git CSS golang k8s Nginx Django mybatis 深度学习 多线程 React 架构 devops 爬虫 云计算 Spring Boot LeetCode