如何在Druid原生GroupBy查询中统计应用所有维度后的总行数

要在你当前的GroupBy查询中获取总行数，只需添加聚合统计逻辑即可。具体做法是在查询结构中加入aggregations数组，使用count类型的聚合器来统计行数。

修改后的完整查询语句如下：

{
    "queryType": "groupBy",
    "dataSource": {
        "type": "table",
        "name": "employee"
    },
    "intervals": {
        "type": "intervals",
        "intervals": [
            "-146136543-09-08T08:23:32.096Z/146140482-04-24T15:36:27.903Z"
        ]
    },
    "virtualColumns": [],
    "granularity": {
        "type": "all"
    },
    "dimensions": [
        {
            "type": "default",
            "dimension": "department",
            "outputName": "department",
            "outputType": "STRING"
        },
        {
            "type": "default",
            "dimension": "standard",
            "outputName": "standard",
            "outputType": "STRING"
        },
        {
            "type": "default",
            "dimension": "vocational",
            "outputName": "vocational",
            "outputType": "STRING"
        },
        {
            "type": "default",
            "dimension": "lab",
            "outputName": "lab",
            "outputType": "STRING"
        }
    ],
    "aggregations": [
        {
            "type": "count",
            "name": "total_rows"
        }
    ],
    "having": null,
    "descending": false
}

关键修改说明 #

• 添加了aggregations字段，其中定义了一个名为total_rows的count聚合器，用于统计每个维度分组下的行数
• 如果需要获取全局总行数（不按维度分组），可以直接移除dimensions数组，仅保留aggregations中的count聚合器即可

内容的提问来源于stack exchange，提问作者donm