Python需要将花括号替换为（花括号和逗号）

Question

我有一个包含数十万行的JSON响应，所有的解析都来自python字典，并使用JSON.load转换为JSON

现在获取带有无效响应的JSON响应，因此解决方法如下

响应

{
    "ComplianceType": "",
    "Details": {
        "InstalledTime": "",
        "PatchBaselineId": "",
        "PatchState": ""
    },
    "ExecutionSummary": {
        "ExecutionId": "",
        "ExecutionTime": "",
        "ExecutionType": ""
    },
    "Id": "",
    "ResourceId": "",
    "ResourceType": "",
    "Severity": "",
    "Status": "",
    "Title": ""
}
{
    "ComplianceType": "",
    "Details": {
        "InstalledTime": "",
        "PatchBaselineId": "",
        "PatchState": ""
    },
    "ExecutionSummary": {
        "ExecutionId": "",
        "ExecutionTime": "",
        "ExecutionType": ""
    },
    "Id": "",
    "ResourceId": "",
    "ResourceType": "",
    "Severity": "",
    "Status": "",
    "Title": ""
}

设法使其有效

[{
    "ComplianceType": "",
    "Details": {
        "InstalledTime": "",
        "PatchBaselineId": "",
        "PatchState": ""
    },
    "ExecutionSummary": {
        "ExecutionId": "",
        "ExecutionTime": "",
        "ExecutionType": ""
    },
    "Id": "",
    "ResourceId": "",
    "ResourceType": "",
    "Severity": "",
    "Status": "",
    "Title": ""
},
{
    "ComplianceType": "",
    "Details": {
        "InstalledTime": "",
        "PatchBaselineId": "",
        "PatchState": ""
    },
    "ExecutionSummary": {
        "ExecutionId": "",
        "ExecutionTime": "",
        "ExecutionType": ""
    },
    "Id": "",
    "ResourceId": "",
    "ResourceType": "",
    "Severity": "",
    "Status": "",
    "Title": ""
}
]

当我尝试使用sed shell时，我需要通过程序执行这些操作-->#sed-i的#}、#test.json<；--。此shell命令替换}，in}，并返回为}

[{
    "ComplianceType": "",
    "Details": {
        "InstalledTime": "",
        "PatchBaselineId": "",
        "PatchState": ""
    },,
    "ExecutionSummary": {
        "ExecutionId": "",
        "ExecutionTime": "",
        "ExecutionType": ""
    },,
    "Id": "",
    "ResourceId": "",
    "ResourceType": "",
    "Severity": "",
    "Status": "",
    "Title": ""
},
{
    "ComplianceType": "",
    "Details": {
        "InstalledTime": "",
        "PatchBaselineId": "",
        "PatchState": ""
    },,
    "ExecutionSummary": {
        "ExecutionId": "",
        "ExecutionTime": "",
        "ExecutionType": ""
    },,
    "Id": "",
    "ResourceId": "",
    "ResourceType": "",
    "Severity": "",
    "Status": "",
    "Title": ""
},
]

请提供一个适用于shell或python的解决方案

Answer 1

根据您的示例输入，这可能适用于您：

$ cat input.json \
    | grep -v -e '^[[:space:]]*$' \
    | sed -e '1s/^/[/'   \
          -e '$s/$/]/'   \
          -e 's/^}$/},/'

• grep删除任何空行（如果存在）。这很重要，因为下面的sed需要在最后一个非空行上操作
• 第一个sed表达式将[添加到第一行的开头
• 第二个sed表达式将]添加到最后一行的末尾
• 最后的sed表达式在一行中的任何}后面添加一个逗号

通过在第三个sed之前应用第二个}，最后一个}变为}]，因此}不再单独在一行上