jq - select objects 并为缺失打印 null

Question

我正在尝试从 json 文件生成 CSV 文件，文件如下

cat role1.json
{
  "Tags": [
    {
      "Key": "Name",
      "Value": "Role1Name"
    },
    {
      "Key": "ID",
      "Value": "Role1ID"
    },
    {
      "Key": "Manager",
      "Value": "Role1Manager"
    },
    {
      "Key": "User",
      "Value": "Role1User"
    },
    {
      "Key": "Country",
      "Value": "USA"
    }
  ]
}

cat role2.json
{
  "Tags": [
    {
      "Key": "Name",
      "Value": "Role2Name"
    },
    {
      "Key": "ID",
      "Value": "Role2ID"
    },
    {
      "Key": "City",
      "Value": "NewYork"
    },
    {
      "Key": "Creator",
      "Value": "Role2Creator"
    },
    {
      "Key": "User",
      "Value": "Role2User"
    }
  ]
}

cat role3.json
{
  "Tags": [
    {
      "Key": "Name",
      "Value": "Role3Name"
    },
    {
      "Key": "ID",
      "Value": "Role3ID"
    },
    {
      "Key": "Creator",
      "Value": "Role3Creator"
    },
    {
      "Key": "ZIP",
      "Value": 82378
    },
    {
      "Key": "Manager",
      "Value": "Role3Manager"
    },
    {
      "Key": "User",
      "Value": "Role3User"
    }
  ]
}

我想从其中的每一行生成行以供以后用作 CSV，例如：

Role1Name, Role1ID, null, Role1Manager, Role1User
Role2Name, Role2ID, Role2Creator, null, Role2User
Role3Name, Role3ID, Role3Creator, Role3Manager, Role3User

对于header行姓名、ID、创建者、经理、用户

我能够获取所有“值”但无法打印缺少“键”的空值

$cat role1.json | jq -rc  '[.Tags[] | select(.Key == ("Name","ID","Creator","Manager","User")) | .Value]'
["Role1Name","Role1ID","Role1Manager","Role1User"]

$cat role2.json | jq -rc  '[.Tags[] | select(.Key == ("Name","ID","Creator","Manager","User")) | .Value]'
["Role2Name","Role2ID","Role2Creator","Role2User"]

$cat role3.json | jq -rc  '[.Tags[] | select(.Key == ("Name","ID","Creator","Manager","User")) | .Value]'
["Role3Name","Role3ID","Role3Creator","Role3Manager","Role3User"]

有人可以与我分享如何使用 jq 完成此操作。

另外，我们如何执行命令。

谢谢！

Answer 1

关键（哈！）是

[ .[ $keys[] ] ]

如果您查看了与 CSV 相关问题的其他答案，您可能已经注意到采取的第一步是获取密钥列表。这通常是通过收集输入 objects 的键来完成的。 (Example) 在你的例子中，你有一个 hard-coded 列表，所以它更简单。

如果您想要实际的 CSV，您可以使用

jq -sr '
   [ "Name", "ID", "Creator", "Manager", "User" ] as $keys |
   (
      $keys,
      ( .[].Tags | from_entries | [ .[ $keys[] ] ] )
   ) |
   @csv
' role*.json

这会产生

"Name","ID","Creator","Manager","User"
"Role1Name","Role1ID",,"Role1Manager","Role1User"
"Role2Name","Role2ID","Role2Creator",,"Role2User"
"Role3Name","Role3ID","Role3Creator","Role3Manager","Role3User"

jqplay

没有 header:

jq -r '.Tags | from_entries | [ .["Name","ID","Creator","Manager","User"] ] | @csv' role*.json

jqplay

要获取您发布的特定输出（不是 CSV），您可以使用

jq -sr '
   [ "Name", "ID", "Creator", "Manager", "User" ] as $keys |
   (
      $keys,
      ( .[].Tags | from_entries | [ .[ $keys[] ] | . // "null" ] )
   ) |
   join(", ")
' role*.json

这会产生

Name, ID, Creator, Manager, User
Role1Name, Role1ID, null, Role1Manager, Role1User
Role2Name, Role2ID, Role2Creator, null, Role2User
Role3Name, Role3ID, Role3Creator, Role3Manager, Role3User

jqplay

没有 header:

jq -r '.Tags | from_entries | [ .["Name","ID","Creator","Manager","User"] | . // "null" ] | join(", ")' role*.json

jqplay

Answer 2

从另一个论坛得到了答案，可能对其他人有用

$jq -rc  '.Tags | from_entries | [.Name, .ID, .Creator, .Manager, .User]' role*.json
["Role1Name","Role1ID",null,"Role1Manager","Role1User"]
["Role2Name","Role2ID","Role2Creator",null,"Role2User"]
["Role3Name","Role3ID","Role3Creator","Role3Manager","Role3User"]

jq - select objects 并为缺失打印 null

jq - select objects and print null for missing

csv

json

export-to-csv

jq