如何在Mule Dataweave中循环并组合成一个整体

Question

我有json的请求和预期的响应，如下所述。它需要group pByclientItemCode，我在中途卡住了在同一个地方循环。使用了MapObject和减少函数的组合。任何帮助都将不胜感激。

[
{
  "ClientCode": "1",
  "ClientItemCode": "245",
  "LocationId": "CLOSED"
 },
 {
  "ClientCode": "1",
  "ClientItemCode": "245",
  "LocationId": "OPEN"
 },
    {
  "ClientCode": "2",
  "ClientItemCode": "245",
  "LocationId": "CHECKOUT"
 },
 {
  "ClientCode": "2",
  "ClientItemCode": "245",
  "LocationId": "TEST"
 },
 {
  "ClientCode": "1",
  "ClientItemCode": "123",
  "LocationId": "OPEN"
 },
 {
  "ClientCode": "1",
  "ClientItemCode": "123",
  "LocationId": "CLOSED"
 }
 ]

预期响应：

  <Results>
  <Result>
    <ClientItemCode>123<ClientItemCode>
    <ResultLines>
      <ResultLine>
        <ClientCode>1</ClientCode>
        <From>
          <LocationId>OPEN</LocationId>
        </From>
        <To>
          <LocationId>CLOSED</LocationId>
        </To>
      </ResultLine>
       <ResultLine>
        <ClientCode>2</ClientCode>
        <From>
          <LocationId>CHECKOUT</LocationId>
        </From>
        <To>
          <LocationId>TEST</LocationId>
        </To>
      </ResultLine>
    </ResultLines>
  </Result>
  <Result>
   <CientItemCode>245<ClientItemCode>
   <ResultLines>
      <ResultLine>
        <ClientCode>1</ClientCode>
        <From>
          <LocationId>CLOSED</LocationId>
        </From>
        <To>
          <LocationId>OPEN</LocationId>
        </To>
     </ResultLine>
    </ResultLines>
  </Result>
</Results>

Answer 1

假设打开和/或关闭可能丢失:

%dw 2.0
output application/xml
---
Results: payload groupBy $.ClientItemCode
    mapObject ((value, key, index) -> result: {
            ClientItemCode: key,
            ResultLines: {
                From: if (value.LocationId contains "OPEN") LocationId: "OPEN" else null,
                To: if (value.LocationId contains "CLOSED") LocationId: "CLOSED" else null

            }
        } 
    )

输出:

<?xml version='1.0' encoding='UTF-8'?>
<Results>
  <result>
    <ClientItemCode>245</ClientItemCode>
    <ResultLines>
      <From>
        <LocationId>OPEN</LocationId>
      </From>
      <To>
        <LocationId>CLOSED</LocationId>
      </To>
    </ResultLines>
  </result>
  <result>
    <ClientItemCode>123</ClientItemCode>
    <ResultLines>
      <From>
        <LocationId>OPEN</LocationId>
      </From>
      <To>
        <LocationId>CLOSED</LocationId>
      </To>
    </ResultLines>
  </result>
</Results>

Answer 2

这是我想出来的，如果我有更多的时间投入，可能会简化一点。试一试:

%dw 2.0
output application/xml
var data = [
 {
  "ClientCode": "1",
  "ClientItemCode": "245",
  "LocationId": "CLOSED"
 },
 {
  "ClientCode": "1",
  "ClientItemCode": "245",
  "LocationId": "OPEN"
 },
    {
  "ClientCode": "2",
  "ClientItemCode": "245",
  "LocationId": "CHECKOUT"
 },
 {
  "ClientCode": "2",
  "ClientItemCode": "245",
  "LocationId": "TEST"
 },
 {
  "ClientCode": "1",
  "ClientItemCode": "123",
  "LocationId": "OPEN"
 },
 {
  "ClientCode": "1",
  "ClientItemCode": "123",
  "LocationId": "CLOSED"
 }
]
---
// I assume that your data are ordered and all the records that will be From and To
// are paired with one another. It is doable without making such assumption but the
// algorithm will get complex.
results: do {
    // Group by the data
    var groupedData = data map {($),(From: true) if (isEven($$))} groupBy $.ClientItemCode
    // Order the client Ids
    var orderedClientIds = groupedData pluck $$ orderBy $ as Number
    ---
    orderedClientIds reduce (cId, results={}) -> do {
        var clientItemCode = cId
        var groupedByClientICode = groupedData[cId] groupBy $.ClientCode pluck $
        ---
        results ++ {result: {
            ClientItemCode: clientItemCode,
            ResultLines: groupedByClientICode reduce (cliCode, lines={}) -> do {
                var clientCode = cliCode[0].ClientCode
                ---
                lines ++ {
                    ClientCode: clientCode,
                    ResultLine: cliCode reduce (e, acc={}) -> do {
                        var locRec = {LocationId: e.LocationId}
                        ---
                        acc ++ (if (e.From?) {From: locRec } else {To: locRec})
                    }
                }
            }
        }}
    }
}

当我在注释中复制时，我还做了一个假设：我假设您的数据是有序的，并且所有将成为From和To的记录都彼此配对。

编辑:再次编辑代码，以强制对< code>ClientItemCode进行排序，然后在创建< code>result标记的所有转换之前按顺序访问每个值。剩下的代码几乎和以前一样。不知道为什么一个简单的< code>orderBy对你没有用，但对我有用。