如何检查JSON并添加/删除密钥

Question

我有一个JSON文件作为NOSQL数据库的导出，以下示例是1xxx记录中的2个。

{
    "task": {
        "id1": {
            "completed": true,
            "title": "Testing",
            "desc": "short desc",
            "status": "completed",
            "dueDate": {
                "_seconds": 1622607310,
                "_nanoseconds": 867000000
            },
            "citizenID": "Uvr0vZqIPON5UZsgJsMYPe2Qfsi2",
            "datePosted": {
                "_seconds": 1622607408,
                "_nanoseconds": 365000000
            },
            "genre": "Errands"
        },
        "id2": {
            "completed": true,
            "title": "Testing 2",
            "status": "completed",
            "dueDate": {
                "_seconds": 1622608576,
                "_nanoseconds": 476999000
            },
            "citizenID": "Uvr0vZqIPON5UZsgJsMYPe2Qfsi2",
            "datePosted": {
                "_seconds": 1622608592,
                "_nanoseconds": 628999000
            },
            "genre": "Errands",
            "latLng": null
        }
    }
}

由于ID1和ID2中的键不一致，因此我试图找到一种方法/工具来确保在使用它们之前键保持一致。另外，当导出将日期推到秒时，我还试图将其转换为时间戳/字符串，

最终输出理想情况应该是：

{
"task": {
    "id1": {
        "completed": true,
        "title": "Testing",
        "desc": "short desc",
        "status": "completed",
        "dueDate": "Wednesday, 2 June 2021 12:15:10.867 GMT+08:00",
        "citizenID": "Uvr0vZqIPON5UZsgJsMYPe2Qfsi2",
        "datePosted": "Wednesday, 2 June 2021 12:16:48.365 GMT+08:00",
        "genre": "Errands"
    },
    "id2": {
        "completed": true,
        "title": "Testing 2",
        "desc": "",
        "status": "completed",
        "dueDate": "Wednesday, 2 June 2021 12:36:16.476 GMT+08:00",
        "citizenID": "Uvr0vZqIPON5UZsgJsMYPe2Qfsi2",
        "datePosted": "Wednesday, 2 June 2021 12:36:32.628 GMT+08:00",
        "genre": "Errands"
    }
}

任何帮助都会受到赞赏

1：我找到了很棒的JQ，但我无法写入一个足够好的查询：

    . as $data | 
paths(scalars) | . as $path | 
"\($path[1]):{\($path[2]):\($data | getpath($path))}"

这给了我：

"id1:{completed:true}"
"id1:{title:Testing}"
"id1:{desc:short desc}"
"id1:{status:completed}"
"id1:{dueDate:1622607310}"
"id1:{dueDate:867000000}"
"id1:{citizenID:Uvr0vZqIPON5UZsgJsMYPe2Qfsi2}"
"id1:{datePosted:1622607408}"
"id1:{datePosted:365000000}"
"id1:{genre:Errands}"
"id2:{completed:true}"
"id2:{title:Testing 2}"
"id2:{status:completed}"
"id2:{dueDate:1622608576}"
"id2:{dueDate:476999000}"
"id2:{citizenID:Uvr0vZqIPON5UZsgJsMYPe2Qfsi2}"
"id2:{datePosted:1622608592}"
"id2:{datePosted:628999000}"
"id2:{genre:Errands}"

它离理想的输出还很遥远

Answer 1

给定输入文件作为input.json和以下JQ脚本为transform.jq：

def stddate:
    ._seconds | todate;

.task
| map_values(
    .dueDate |= stddate
    | .datePosted |= stddate
)

此命令

jq -f transform.jq input.json

返回此输出：

{
  "id1": {
    "completed": true,
    "title": "Testing",
    "desc": "short desc",
    "status": "completed",
    "dueDate": "2021-06-02T04:15:10Z",
    "citizenID": "Uvr0vZqIPON5UZsgJsMYPe2Qfsi2",
    "datePosted": "2021-06-02T04:16:48Z",
    "genre": "Errands"
  },
  "id2": {
    "completed": true,
    "title": "Testing 2",
    "status": "completed",
    "dueDate": "2021-06-02T04:36:16Z",
    "citizenID": "Uvr0vZqIPON5UZsgJsMYPe2Qfsi2",
    "datePosted": "2021-06-02T04:36:32Z",
    "genre": "Errands",
    "latLng": null
  }
}

当然您可以将完整的脚本挤入一个命令行中争论。

缺少事物：

• _nanoseconds被忽略。您可以读取 jq 的日期/时间处理链接到使用的C函数。

• 缺少不必要的任务级别。如果您真的需要它，请用.task | .task | =将其恢复。

• 什么“一致的键”意味着您可以在map_values中完成。