druid json查詢：相同sql對應的json請求體

最近完成了一個新項目用druid的json查詢的方式，項目做完了在這裏總結一下：

一、

【簡單SQL】：

SELECT
 SUM(initiate_count) AS chatCount,
 SUM(online_initiate_count) AS onlineChatCount,
 SUM(initiate_chat_count) AS startChatCount,
 SUM(effective_chat_count) AS effectiveChatCount,
 SUM(info_submit_card_count) AS keywordCard,
 SUM(keyword_click_count) AS keywordClick
FROM
 f_fact_landing_page
WHERE
 __time TIMESTAMP BETWEEN '2018-08-13 00:00:00'
AND '2018-08-13 23:59:59'

【簡單SQL對應druid的json請求體】：

{
    "queryType": "timeseries",
    "dataSource": "f_fact_landing_page",
    "granularity": "all",
    "intervals": [
        "2018-08-13T00:00/2018-08-13T23:59:59"
    ],
    "aggregations": [
        {
            "type": "longSum",
            "fieldName": "initiate_count",
            "name": "chatCount"
        },
        {
            "type": "longSum",
            "fieldName": "online_initiate_count",
            "name": "onlineChatCount"
        },
        {
            "type": "longSum",
            "fieldName": "initiate_chat_count",
            "name": "startChatCount"
        },
        {
            "type": "longSum",
            "fieldName": "effective_chat_count",
            "name": "effectiveChatCount"
        },
        {
            "type": "longSum",
            "fieldName": "info_submit_card_count",
            "name": "keywordCard"
        },
        {
            "type": "longSum",
            "fieldName": "keyword_click_count",
            "name": "keywordClick"
        }
    ]
}

二、

SQL的group by/order by/limit操作：

SELECT
    SUM(keyword_click_count) AS keywordClick,
    SUM(info_submit_card_count) AS keywordCard,
    SUM(keyword_cash_amount) AS keywordCash,
    SUM(keyword_cost_amount) AS keywordCost,
    SUM(educate_amount) AS journalAccount,
    COUNT(
        DISTINCT load_cost_vv_id_unique
    ) AS visitTimes,
    COUNT(
        DISTINCT load_cost_uid_unique
    ) AS visitPeopleCount,
    SUM(load_cost_count) AS visitPageCount,
    SUM(split_screen_count) AS splitScreenCount,
    SUM(stay_time_stay_time_amount) AS stayTime SUM(load_cost_cost_amount) AS loadCost,
    SUM(click_submit_count) AS submitClickCount,
    SUM(initiate_count) AS chatCount,
    SUM(online_initiate_count) AS onlineChatCount,
    SUM(initiate_chat_count) AS startChatCount,
    SUM(effective_chat_count) AS effectiveChatCount
FROM
    f_fact_landing_page
WHERE
    advertiser_code IN (
        "360",
        "baidu",
        "shenma",
        "sogou"
    )
AND __time TIMESTAMP BETWEEN '2018-08-13 00:00:00'
AND '2018-08-13 23:59:59'
GROUP BY
    site_name,
    site_id
ORDER BY
    keywordClick
LIMIT 10000

【SQL的group by/order by/limit操作對應的druid的json請求體】

{
    "queryType": "groupBy",
    "dataSource": "f_fact_landing_page",
    "granularity": "all",
    "intervals": [
        "2018-08-13T00:00/2018-08-13T23:59:59"
    ],
    "dimensions": [
        "site_name",
        "site_id"
    ],
    "filter": {
        "type": "in",
        "dimension": "advertiser_code",
        "values": [
            "360",
            "baidu",
            "shenma",
            "sogou"
        ]
    },
    "aggregations": [
        {
            "type": "longSum",
            "fieldName": "keyword_click_count",
            "name": "keywordClick"
        },
        {
            "type": "longSum",
            "fieldName": "info_submit_card_count",
            "name": "keywordCard"
        },
        {
            "type": "doubleSum",
            "fieldName": "keyword_cash_amount",
            "name": "keywordCash"
        },
        {
            "type": "doubleSum",
            "fieldName": "keyword_cost_amount",
            "name": "keywordCost"
        },
        {
            "type": "doubleSum",
            "fieldName": "educate_amount",
            "name": "journalAccount"
        },
        {
            "type": "hyperUnique",
            "fieldName": "load_cost_vv_id_unique",
            "name": "visitTimes"
        },
        {
            "type": "hyperUnique",
            "fieldName": "load_cost_uid_unique",
            "name": "visitPeopleCount"
        },
        {
            "type": "longSum",
            "fieldName": "load_cost_count",
            "name": "visitPageCount"
        },
        {
            "type": "longSum",
            "fieldName": "split_screen_count",
            "name": "splitScreenCount"
        },
        {
            "type": "doubleSum",
            "fieldName": "stay_time_stay_time_amount",
            "name": "stayTime"
        },
        {
            "type": "doubleSum",
            "fieldName": "load_cost_cost_amount",
            "name": "loadCost"
        },
        {
            "type": "longSum",
            "fieldName": "click_submit_count",
            "name": "submitClickCount"
        },
        {
            "type": "longSum",
            "fieldName": "initiate_count",
            "name": "chatCount"
        },
        {
            "type": "longSum",
            "fieldName": "online_initiate_count",
            "name": "onlineChatCount"
        },
        {
            "type": "longSum",
            "fieldName": "initiate_chat_count",
            "name": "startChatCount"
        },
        {
            "type": "longSum",
            "fieldName": "effective_chat_count",
            "name": "effectiveChatCount"
        }
    ],
    "limitSpec": {
        "type": "default",
        "limit": "10000",
        "columns": [
            {
                "dimension": "keywordClick",
                "direction": "desc",
                "dimensionOrder": "numeric"
            }
        ]
    }
}

三、

【SQL的groupby/having】：

SELECT
    SUM(keyword_cost_amount) AS keywordCost
FROM
    f_fact_landing_page
WHERE
    advertiser_code IN (
        "360",
        "baidu",
        "shenma",
        "sogou"
    )
AND __time TIMESTAMP BETWEEN '2018-08-13 00:00:00'
AND '2018-08-13 23:59:59'
GROUP BY
    promote_url_decode
HAVING
    keywordCost > 0

【SQL的groupby/having對應的druid的json請求體】：

{
    "queryType": "groupBy",
    "dataSource": "f_fact_landing_page",
    "granularity": "all",
    "intervals": [
        "2018-08-13T00:00/2018-08-13T23:59:59"
    ],
    "dimensions": [
        "promote_url_decode"
    ],
    "filter": {
        "type": "in",
        "dimension": "advertiser_code",
        "values": [
            "360",
            "baidu",
            "shenma",
            "sogou"
        ]
    },
    "aggregations": [
        {
            "type": "doubleSum",
            "fieldName": "keyword_cost_amount",
            "name": "keywordCost"
        }
    ],
    "having": {
        "type": "and",
        "havingSpecs": [
            {
                "type": "greaterThan",
                "aggregation": "keywordCost",
                "value": "0"
            }
        ]
    }
}

四、

【有二級指標，以一級指標相除爲例】：

SELECT
    SUM(keyword_click_count) AS keywordClick,
    SUM(info_submit_card_count) AS keywordCard,
    (SUM(keyword_click_count) / SUM(info_submit_card_count)) AS netSale,
    SUM(keyword_cash_amount) AS keywordCash,
    SUM(keyword_cost_amount) AS keywordCost,
    SUM(educate_amount) AS journalAccount,
    COUNT(DISTINCT load_cost_vv_id_unique) AS visitTimes,
    COUNT( DISTINCT load_cost_uid_unique ) AS visitPeopleCount,
    SUM(load_cost_count) AS visitPageCount,
    SUM(split_screen_count) AS splitScreenCount,
    SUM(stay_time_stay_time_amount) AS stayTime,
    SUM(load_cost_cost_amount) AS loadCost,
    SUM(click_submit_count) AS submitClickCount,
    SUM(initiate_count) AS chatCount,
    SUM(online_initiate_count) AS onlineChatCount,
    SUM(initiate_chat_count) AS startChatCount,
    SUM(effective_chat_count) AS effectiveChatCount
FROM
    f_fact_landing_page
WHERE
    advertiser_code IN ( "360", "baidu", "shenma", "sogou" )
    AND __time TIMESTAMP BETWEEN '2018-08-13 00:00:00' AND '2018-08-13 23:59:59'
GROUP BY
    site_name, site_id
ORDER BY
    netSale ASC 
LIMIT 10

【以上對應的druid的json請求體】：

{
    "queryType": "groupBy",
    "dataSource": "f_fact_landing_page",
    "granularity": "all",
    "intervals": [
        "2018-08-13T00:00/2018-08-13T23:59:59"
    ],
    "dimensions": [ "site_name", "site_id" ],
    "filter": {
        "type": "in",
        "dimension": "advertiser_code",
        "values": [ "360", "baidu", "shenma", "sogou" ]
    },
    "aggregations": [
        {
            "type": "longSum",
            "fieldName": "keyword_click_count",
            "name": "keywordClick"
        },
        { "type":"longSum", "fieldName":"info_submit_card_count", "name":"keywordCard" },
        { "type":"doubleSum", "fieldName":"keyword_cash_amount", "name":"keywordCash" },
        { "type":"doubleSum", "fieldName":"keyword_cost_amount", "name":"keywordCost" },
        { "type":"doubleSum", "fieldName":"educate_amount", "name":"journalAccount" },
        {
            "type": "hyperUnique",
            "fieldName": "load_cost_vv_id_unique",
            "name": "visitTimes"
        },
        { "type":"hyperUnique", "fieldName":"load_cost_uid_unique", "name":"visitPeopleCount" },
        { "type":"longSum", "fieldName":"load_cost_count", "name":"visitPageCount" },
        { "type":"longSum", "fieldName":"split_screen_count", "name":"splitScreenCount" },
        { "type":"doubleSum", "fieldName":"stay_time_stay_time_amount", "name":"stayTime" },
        { "type":"doubleSum", "fieldName":"load_cost_cost_amount", "name":"loadCost" },
        {"type":"longSum", "fieldName":"click_submit_count" ,"name":"submitClickCount" },
        { "type":"longSum", "fieldName":"initiate_count", "name":"chatCount" },
        { "type":"longSum", "fieldName":"online_initiate_count", "name":"onlineChatCount" },
        { "type":"longSum", "fieldName":"initiate_chat_count", "name":"startChatCount" },
        { "type":"longSum", "fieldName":"effective_chat_count", "name":"effectiveChatCount" }
    ],
    "having": {
        "type": "and",
        "havingSpecs": [
            {
                "type": "greaterThan",
                "aggregation": "keywordCost",
                "value": "0"
            }
        ]
    },
    "postAggregations": [
        {
            "type": "arithmetic",
            "name": "netSale",
            "fn": "/",
            "fields": [
                {
                    "type": "fieldAccess",
                    "fieldName": "keywordCard"
                },
                {
                    "type": "fieldAccess",
                    "fieldName": "keywordClick"
                }
            ]
        }
    ],
    "limitSpec": {
        "type": "default",
        "limit": "10",
        "columns": [
            {
                "dimension": "netSale",
                "direction": "asc",
                "dimensionOrder": "numeric"
            }
        ]
    }
}

其實這些都是在druid的官網裏面有的，只要細心查都能夠找到，提供druid官網的Doc：http://druid.io/docs/latest/design/

如有問題，請多指教！！！