SQL学习打卡Task6

本笔记为阿里云天池龙珠计划SQL训练营的学习内容，链接为：https://tianchi.aliyun.com/specials/promotion/aicampsql；

练习题代码如下：

-- 练习1
SELECT MarketData.*,
OperatingData.INDIC_NAME_EN,OperatingData.VALUE,
IncomeStatement.N_INCOME,IncomeStatement.T_COGS,
IncomeStatement.T_REVENUE
FROM (
SELECT TICKER_SYMBOL,END_DATE,CLOSE_PRICE
FROM `market data`
WHERE TICKER_SYMBOL IN ('600383','600048')) MarketData
LEFT JOIN
(SELECT TICKER_SYMBOL,INDIC_NAME_EN,END_DATE,VALUE
FROM `company operating`
WHERE TICKER_SYMBOL IN ('600383','600048')) OperatingData
ON MarketData.TICKER_SYMBOL = OperatingData.TICKER_SYMBOL
AND MarketData.END_DATE = OperatingData.END_DATE
LEFT JOIN
(SELECT DISTINCT TICKER_SYMBOL,END_DATE,T_COGS,N_INCOME
FROM `income statement`
WHERE TICKER_SYMBOL IN ('600383','600048')) IncomeStatement
ON MarketData.TICKER_SYMBOL = IncomeStatement.TICKER_SYMBOL
AND MarketData.END_DATE = IncomeStatement.END_DATE
ORDER BY MarketData.TICKER_SYMBOL,MarketData.END_DATE;

-- 练习2
SELECT pH,`citric acid`,DENSE_RANK()    -- 窗口函数：中式排序
OVER (ORDER BY `citric acid`) AS rankn
FROM `winequality-red`
WHERE pH = 3.03;

-- 练习3
-- 发放优惠券总金额最多的商家
SELECT Merchant_id,
SUM(SUBSTRING_INDEX(`Discount_rate`,':',1)) AS discount_amount
FROM ccf_offline_stage1_test_revised AS ccf
WHERE Date_received BETWEEN '2016-07-01' AND '2016-07-31'
GROUP BY Merchant_id
ORDER BY discount_amount DESC
LIMIT 1;

-- 发放优惠券张数最多的商家 
SELECT Merchant_id,COUNT(1) AS cnt 
-- COUNT(1)包括忽略所有列，用1代表代码行，不好忽略null的列
FROM ccf_offline_stage1_test_revised AS ccf
WHERE Date_received BETWEEN '2016-07-01' AND '2016-07-31'
GROUP BY Merchant_id
ORDER BY cnt DESC
LIMIT 1;

-- 练习4
-- 2015年用电最高峰是发生在哪月 
SELECT PERIOD_DATE,MAX(DATA_VALUE) AS FianlValue
FROM `macro industry`
WHERE INDIC_ID = '2020101522'
AND YEAR(PERIOD_DATE) = 2015
GROUP BY PERIOD_DATE
ORDER BY FianlValue DESC
LIMIT 1;
-- 并且相比去年同期增长/减少了多少个百分比 （求同比）
-- 同比 = (本期-同期)/同期 或者 (本期/同期)-1
SELECT BaseData.*,
(BaseData.FianlValue - YOY.FianlValue) / YOY.FianlValue AS YOY
-- 也可以写成(BaseData.FianlValue / YOY.FianlValue) - 1 AS YOY
FROM(
SELECT PERIOD_DATE,MAX(DATA_VALUE) AS FianlValue
FROM `macro industry`
WHERE INDIC_ID = '2020101522'
AND YEAR(PERIOD_DATE) = 2015
GROUP BY PERIOD_DATE
ORDER BY FianlValue DESC
LIMIT 1) AS BaseData
LEFT JOIN 
(SELECT PERIOD_DATE,MAX(DATA_VALUE) AS FianlValue
FROM `macro industry`
WHERE INDIC_ID = '2020101522'
AND YEAR(PERIOD_DATE) = 2014
GROUP BY PERIOD_DATE) AS YOY
ON YEAR(BaseData.PERIOD_DATE) = YEAR(YOY.PERIOD_DATE) + 1
AND MONTH(BaseData.PERIOD_DATE) = MONTH(YOY.PERIOD_DATE);

-- 练习5
-- 弃用率为多少
-- 弃用率 = 被领券但未使用的优惠券张数/总的被领取优惠券张数
SELECT 
SUM(CASE WHEN DATE = '0000-00-00' AND Coupon_id IS NOT NULL
THEN 1 ELSE 0 END) / 
SUM(CASE WHEN Coupon_id IS NOT NULL THEN 1 ELSE 0 END) AS discard_rate
FROM ccf_online_stage1_train
WHERE Date_received BETWEEN '2016-06-01' AND '2016-06-30';

-- 2016年6月期间，优惠券弃用率最高的商家 
SELECT Merchant_id,
SUM(CASE WHEN DATE = '0000-00-00' AND Coupon_id IS NOT NULL
THEN 1 ELSE 0 END) / 
SUM(CASE WHEN Coupon_id IS NOT NULL THEN 1 ELSE 0 END) AS discard_rate
FROM ccf_online_stage1_train
WHERE Date_received BETWEEN '2016-06-01' AND '2016-06-30'
GROUP BY Merchant_id
ORDER BY discard_rate DESC
LIMIT 1;

-- 练习6
SELECT

pH,'residual sugar',RANK() OVER(ORDER BY 'residual sugar') AS ranking 

--  英式排序
FROM `winquality-white`
WHERE pH = 3.63;

-- 练习7
-- 截止到2018年底，市值最大的三个行业是哪些？
SELECT TYPE_NAME_CN,SUM(MARKET_VALUE)
FROM `market data`
WHERE YEAR(END_DATE) = '2018-12-31'
GROUP BY TYPE_NAME_CN
ORDER BY SUM(MARKET_VALUE) DESC
LIMIT 3;

-- 这三个行业里市值最大的三个公司是哪些？
SELECT BaseData.TYPE_NAME_CN,BaseData.TICKER_SYMBOL
FROM(
SELECT TYPE_NAME_CN,TICKER_SYMBOL,MARKET_VALUE,
ROW_NUMBER() OVER (PARTITION BY TYPE_NAME_CN ORDER BY MARKET_VALUE)

-- 顺序排序
AS CompanyRanking
FROM `market data`) AS BaseData
LEFT JOIN
(SELECT TYPE_NAME_CN,SUM(MARKET_VALUE)
FROM `market data`
WHERE YEAR(END_DATE) = '2018-12-31'
GROUP BY TYPE_NAME_CN
ORDER BY SUM(MARKET_VALUE) DESC
LIMIT 3) AS top3Type
ON BaseData.TYPE_NAME_CN = top3Type.TYPE_NAME_CN
WHERE CompanyRanking <= 3
AND top3Type.TYPE_NAME_CN IS NOT NULL;

-- 练习8
SELECT User_id,SUM(couponCount) AS couponcount
FROM(
SELECT User_id,count(*) AS couponcount
FROM ccf_online_stage1_train
WHERE (Date != 'null' AND Coupon_id != 'null')
AND (LEFT(DATE,4)=2016)           -- 取DATE的左边四位数为2016的值 
GROUP BY User_id 
UNION ALL                                     -- UNION ALL 可以保留重复行 
SELECT User_id,COUNT(*) AS couponcount
FROM ccf_offline_stage1_train
WHERE (Date != 'null' AND Coupon_id != 'null')
AND (LEFT(DATE,4)=2016)
GROUP BY User_id ) AS BaseData
GROUP BY User_id
ORDER BY SUM(couponcount) DESC
LIMIT 1;

-- 练习9
SELECT * 
FROM(

SELECT TICKER_SYMBOL,YEAR(END_DATE) AS Year,
QUARTER(END_DATE) AS QUAETER,SUM(VALUE) AS Amount
FROM `company operating`
WHERE INDIC_NAME_EN = `Baiyun Airport:Passenger throughput`
GROUP BY TICKER_SYMBOL,YEAR(END_DATE),QUARTER(END_DATE)
ORDER BY SUM(VALUE) DESC
LIMIT 1) AS BaseData
LEFT JOIN
(SELECT TICKER_SYMBOL,YEAR(END_DATE) AS Year,
QUARTER(END_DATE) AS QUAETER,SUM(N_INCOME) AS Amount
FROM `income statement`
GROUP BY TICKER_SYMBOL,YEAR(END_DATE),QUARTER(END_DATE)) AS Income
ON BaseData.TICKER_SYMBOL = Income.TICKER_SYMBOL
AND BaseData.Year = Income.Year
AND BaseDate.QUARTER = Income.QUARTER;

-- 练习10
SELECT Merchant_id, 
SUM(discount_amount) AS discount_amount
FROM(
SELECT Merchant_id,SUM(SUBSTRING_INDEX(`Discount_rate`,':',-1)) AS discount_amount
FROM `ccf_online_stage1_train`
WHERE (DATE != 'null' AND Coupon_id != 'null')
AND (LEFT(DATE,4)=2016)
AND MID(DATE,5,2) = '06'
GROUP BY Merchant_id
UNION ALL
SELECT Merchant_id,
SUM(SUBTRING_INDEX(`Discount_rate`,':',-1)) AS discount_amount
FROM `ccf_offline_stage1_train`
WHERE (DATE != 'null' AND Coupon_id != 'null')
AND (LEFT(DATE,4)=2016)
AND MID(DATE,5,2) = '06'
GROUP BY Merchant_id) AS BaseData
GROUP BY Merchant_id
ORDER BY SUM(discount_amount) DESC
LIMIT 3;