使用 jq 将 JSON 解析为 CSV，但在多个记录中拆分子列表

Question

我正在解析从 Azure RateAPI 获得的 JSON 结果，并希望将其转换为标准 CSV 文件。

下一行是我用来将它转换为 CSV 的内容，它可以工作，但由于其中一个属性是一个列表，它没有提供我正在寻找的结果。对于 "sub list" 中的每一项，我需要在我的 csv 文件中创建另一条记录。

cat myfile.json | jq -r '.Meters[] | [ .EffectiveDate, .IncludedQuantity, .MeterCategory, .MeterId, .MeterName, .MeterRates[], .MeterRegion, .MeterStatus, .MeterSubCategory, .MeterTags[], .Units] | @csv'

这是我要解析的 3 条记录。我在处理记录 2 时遇到问题，因为 MeterRates 实际上是我同时需要属性和值的列表。我需要记录 2，一旦解析，对应于 CSV 文件中的 3 条记录，其中每条记录包含 MeterRates 列表中的一项。预期结果的例子在最后

  "OfferTerms": [],
  "Meters": [
    {
      "EffectiveDate": "2019-03-01T00:00:00Z",
      "IncludedQuantity": 0,
      "MeterCategory": "Virtual Machines",
      "MeterId": "d0bf9053-17c4-4fec-8502-4eb8376343a7",
      "MeterName": "F2/F2s Low Priority",
      "MeterRates": {
        "0": 0.0766
      },
      "MeterRegion": "US West 2",
      "MeterStatus": "Active",
      "MeterSubCategory": "F/FS Series Windows",
      "MeterTags": [],
      "Unit": "1 Hour"
    },
    {
      "EffectiveDate": "2014-11-01T00:00:00Z",
      "IncludedQuantity": 0,
      "MeterCategory": "Azure DevOps",
      "MeterId": "c4d6fa88-0df9-4680-867a-b13c960a875f",
      "MeterName": "Virtual User Minute",
      "MeterRates": {
        "0": 0.0004,
        "1980000": 0.0002,
        "9980000": 0.0001
      },
      "MeterRegion": "",
      "MeterStatus": "Active",
      "MeterSubCategory": "Cloud-Based Load Testing",
      "MeterTags": [],
      "Unit": "1/Month"
    },
    {
      "EffectiveDate": "2017-04-01T00:00:00Z",
      "IncludedQuantity": 0,
      "MeterCategory": "SQL Database",
      "MeterId": "cb770eab-d5c8-45fd-ac56-8c35069f5a29",
      "MeterName": "P4 DTUs",
      "MeterRates": {
        "0": 68.64
      },
      "MeterRegion": "IN West",
      "MeterStatus": "Active",
      "MeterSubCategory": "Single Premium",
      "MeterTags": [],
      "Unit": "1/Day"
    }
    ]
}

使用我提供的代码的实际结果如下：

"2019-03-01T00:00:00Z",0,"Virtual Machines","d0bf9053-17c4-4fec-8502-4eb8376343a7","F2/F2s Low Priority",0.0766,"US West 2","Active","F/FS Series Windows",
"2014-11-01T00:00:00Z",0,"Azure DevOps","c4d6fa88-0df9-4680-867a-b13c960a875f","Virtual User Minute",0.0004,0.0002,0.0001,"","Active","Cloud-Based Load Testing",
"2017-04-01T00:00:00Z",0,"SQL Database","cb770eab-d5c8-45fd-ac56-8c35069f5a29","P4 DTUs",68.64,"IN West","Active","Single Premium",

但我期望的结果是（记录 2 对应于基于 MeterRates 的 CSV 文件中的 3 条记录）：

"2019-03-01T00:00:00Z",0,"Virtual Machines","d0bf9053-17c4-4fec-8502-4eb8376343a7","F2/F2s Low Priority",0,0.0766,"US West 2","Active","F/FS Series Windows",
"2014-11-01T00:00:00Z",0,"Azure DevOps","c4d6fa88-0df9-4680-867a-b13c960a875f","Virtual User Minute",0,0.0004,"","Active","Cloud-Based Load Testing",
"2014-11-01T00:00:00Z",0,"Azure DevOps","c4d6fa88-0df9-4680-867a-b13c960a875f","Virtual User Minute",1980000,0.0002,"","Active","Cloud-Based Load Testing",
"2014-11-01T00:00:00Z",0,"Azure DevOps","c4d6fa88-0df9-4680-867a-b13c960a875f","Virtual User Minute",9980000,0.0001"","Active","Cloud-Based Load Testing",
"2017-04-01T00:00:00Z",0,"SQL Database","cb770eab-d5c8-45fd-ac56-8c35069f5a29","P4 DTUs",0,68.64,"IN West","Active","Single Premium",

感谢您的帮助。

Answer 1

您需要在获取 Meters 项和输出行之间添加一个步骤，以输出具有不同比率的 Meters 项的各种组合。正如您现在所拥有的那样，您正在将费率作为该行的其他项目输出，这并不是您真正想要的。

在这种情况下，您可以只添加一个新的属性来保存相应 MeterRate 的值。

.Meters[] | .MeterRate = (.MeterRates | to_entries[])
    | [.EffectiveDate, .IncludedQuantity, .MeterCategory, .MeterId , .MeterName,
       .MeterRate.key, .MeterRate.value,
       .MeterRegion, .MeterStatus, .MeterSubCategory, .MeterTags[], .Units]
    | @csv

您可能需要考虑对 MeterTags 项执行类似的操作，这样您就不会得到可能随机的列数。

使用 jq 将 JSON 解析为 CSV，但在多个记录中拆分子列表

Parse JSON to CSV using jq but split sub list in multiple records

csv

bash

json

jq

ratecard-api