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；

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1
# 请使用A股上市公司季度营收预测数据集《Income Statement.xls》和《Company Operating.xlsx》和《Market Data.xlsx》，以Market Data为主表，将三张表中的TICKER_SYMBOL为600383和600048的信息合并在一起。只需要显示以下字段。

SELECT  INS.TICKER_SYMBOL,
				INS.END_DATE,
				INS.REVENUE,
				INS.COGS,
				INS.N_INCOME,
				MD.TICKER_SYMBOL,
				MD.END_DATE,
				MD.CLOSE_PRICE,
				CO.TICKER_SYMBOL,
				CO.INDIC_NAME_EN,
				CO.END_DATE,
				CO.VALUE
FROM (SELECT * FROM `market data` WHERE TICKER_SYMBOL IN (600383, 600048)) as MD
LEFT JOIN `income statement` AS INS
ON INS.TICKER_SYMBOL = MD.TICKER_SYMBOL
LEFT JOIN `company operating` AS CO
ON INS.TICKER_SYMBOL = CO.TICKER_SYMBOL;



# 练习题2
# 请使用 Wine Quality Data 数据集《winequality-red.csv》，找出 pH=3.03的所有红葡萄酒，然后，对其 citric acid 进行中式排名（相同排名的下一个名次应该是下一个连续的整数值。换句话说，名次之间不应该有“间隔”）


SELECT pH, 
			 `citric acid`,
			 DENSE_RANK() OVER(ORDER BY `citric acid`) AS dense_ranking
FROM `winequality-red` 
WHERE pH = 3.03;



# 练习题3
# 使用Coupon Usage Data for O2O中的数据集《ccf_offline_stage1_test_revised.csv》，试分别找出在2016年7月期间，发放优惠券总金额最多和发放优惠券张数最多的商家。这里只考虑满减的金额，不考虑打几折的优惠券

SELECT Merchant_id,
 -- SUM(SUBSTRING_INDEX(`Discount_rate`,':', 1)) AS sale_amount,
 SUM(SUBSTRING_INDEX(`Discount_rate`,':',-1)) AS discount_amount
 FROM ccf_offline_stage1_test_revised
WHERE Date_received BETWEEN '2016-07-01' AND '2016-07-31'
GROUP BY Merchant_id
ORDER BY discount_amount DESC
LIMIT 1;

SELECT Merchant_id,COUNT(1) AS cnt
FROM ccf_offline_stage1_test_revised
WHERE Date_received BETWEEN '2016-07-01' AND '2016-07-31'
GROUP BY Merchant_id
ORDER BY cnt DESC
LIMIT 1;


# 练习题4
#请使用A股上市公司季度营收预测中的数据集《Macro&Industry.xlsx》中的sheet-INDIC_DATA，请计算全社会用电量:第一产业:当月值在2015年用电最高峰是发生在哪月？并且相比去年同期增长/减少了多少个百分比

SELECT indic_id,name_cn,COUNT(1) as cnt
FROM `macro industry`
GROUP BY indic_id,name_cn;

SELECT *
FROM `macro industry`
WHERE indic_id = '2020101522'
AND YEAR(PERIOD_DATE) = '2015';


SELECT PERIOD_DATE,MAX(DATA_VALUE) AS finalvalue
FROM `macro industry`
WHERE indic_id = '2020101522'
AND YEAR(PERIOD_DATE) = '2015'
GROUP BY PERIOD_DATE
ORDER BY finalvalue
LIMIT 1;


SELECT M1.*, (M1.finalvalue-M2.finalvalue)/M2.finalvalue AS YoY
FROM ( -- 先选出2015年最高用电量
SELECT PERIOD_DATE,MAX(DATA_VALUE) AS finalvalue
			FROM `macro industry`
			WHERE indic_id = '2020101522'
			AND YEAR(PERIOD_DATE) = '2015'
			GROUP BY PERIOD_DATE
			ORDER BY finalvalue
			LIMIT 1
			) AS M1
-- 选出2015年最高用电量，然后用2014年的用电量去匹配相同月份
LEFT JOIN ( -- 选出2014年每个月的用电量
			SELECT PERIOD_DATE,MAX(DATA_VALUE) AS finalvalue
			FROM `macro industry`
			WHERE indic_id = '2020101522'
			AND YEAR(PERIOD_DATE) = '2014'
			GROUP BY PERIOD_DATE
			) AS M2
-- 约束条件：年份相差为1，月份相同 -- 选出去年同期
ON YEAR(M1.PERIOD_DATE) = YEAR(M2.PERIOD_DATE) + 1
AND MONTH(M1.PERIOD_DATE) = MONTH(M2.PERIOD_DATE);



# 练习题5
# 使用Coupon Usage Data for O2O中的数据集《ccf_online_stage1_train.csv》，试统计在2016年6月期间，线上总体优惠券弃用率为多少？并找出优惠券弃用率最高的商家。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

-- 总的被领取优惠券张数
SELECT COUNT(Date_received) as allreceived
FROM (SELECT * 
			FROM ccf_online_stage1_train
			WHERE YEAR(Date_received) = '2016'
			AND MONTH(Date_received) = '06'
) AS ccfon_201606
WHERE Date_received IS NOT NULL

UNION

-- 被领券但未使用的优惠券张数
SELECT COUNT(1) as receivenouse
FROM (SELECT * 
			FROM ccf_online_stage1_train
			WHERE YEAR(Date_received) = '2016'
			AND MONTH(Date_received) = '06'
			) AS ccfon_201606
WHERE Date IS NULL
AND Coupon_id IS NOT NULL;

# 答案
SELECT SUM(CASE WHEN Date IS NULL 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';


SELECT Merchant_id,SUM(CASE WHEN Date IS NULL 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
# HAVING discard_rate = MAX(discard_rate)  -- 还是有问题
ORDER BY discard_rate DESC;


# 练习题6
# 请使用 Wine Quality Data 数据集《winequality-white.csv》，找出 pH=3.63的所有白葡萄酒，然后，对其 residual sugar 量进行英式排名（非连续的排名）

SELECT ww.*,RANK() OVER (ORDER BY ww.`residual sugar`)
FROM `winequality-white` AS ww
WHERE ww.pH=3.63;


# 练习题7
# 请使用A股上市公司季度营收预测中的数据集《Market Data.xlsx》中的sheet-DATA，计算截止到2018年底，市值最大的三个行业是哪些？以及这三个行业里市值最大的三个公司是哪些？（每个行业找出前三大的公司，即一共要找出9个）

DESC `market data`;
SHOW COLUMNS FROM `market data`;


SELECT md.TYPE_NAME_CN,md.TYPE_NAME_EN,SUM(md.MARKET_VALUE) AS max_marketvalue
FROM `market data` AS md
WHERE YEAR(md.END_DATE) = '2018-12-31'
GROUP BY md.TYPE_NAME_CN,md.TYPE_NAME_EN
ORDER BY SUM(md.MARKET_VALUE) DESC
LIMIT 3;

SELECT BaseData.*,top3Type.*
#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+0) DESC) 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;
# 先对源表用rank_number()以type_name_cn进行分组排序，再左连结市值前三的行业


# 练习题8
# 使用Coupon Usage Data for O2O中的数据集《ccf_online_stage1_train.csv》和《ccf_offline_stage1_train.csv》，试找出在2016年6月期间，线上线下累计优惠券使用次数最多的顾客

-- 左连结union右连结实现全外连结
SELECT
FROM () -- 线上使用优惠券用户
LEFT JOIN () -- 线下使用优惠券用户
ON
WHERE 
GROUP BY 

DESC ccf_offline_stage1_train;
DESC ccf_online_stage1_train;

-- 线下使用优惠券用户
SELECT off.User_id,COUNT(*) AS use_on
FROM ccf_offline_stage1_train AS off
WHERE off.Date IS NOT NULL AND off.Coupon_id IS NOT NULL
AND DATE_FORMAT(off.Date,'%Y-%m-%d') BETWEEN '2016-06-01' AND '2016-06-30'
GROUP BY off.User_id
ORDER BY use_on DESC;

-- 线上使用优惠券用户
SELECT onl.User_id,COUNT(*) AS use_on
FROM ccf_online_stage1_train AS onl
WHERE onl.Date IS NOT NULL AND onl.Coupon_id IS NOT NULL
AND onl.Date BETWEEN '2016-06-01' AND '2016-06-30'
GROUP BY onl.User_id
ORDER BY use_on DESC;



# 答案
SELECT User_id,SUM(couponCount) couponCount
FROM (
SELECT User_id,count(*) couponCount
FROM `ccf_online_stage1_train`
WHERE (Date IS NOT NULL AND Coupon_id IS NOT NULL)
AND (Date BETWEEN '2016-06-01' AND '2016-06-30' )
GROUP BY User_id

UNION ALL

SELECT User_id,
COUNT(*) couponCount
FROM `ccf_offline_stage1_train`
WHERE (Date IS NOT NULL AND Coupon_id IS NOT NULL)
AND (DATE_FORMAT(Date,'%Y-%m-%d') BETWEEN '2016-06-01' AND '2016-06-30' )
GROUP BY User_id ) 
AS BaseData
GROUP BY User_id
ORDER BY SUM(couponCount) DESC
LIMIT 1;



# 练习题9
# 请使用A股上市公司季度营收预测数据集《Income Statement.xls》中的sheet-General Business和《Company Operating.xlsx》中的sheet-EN。找出在数据集所有年份中，按季度统计，白云机场旅客吞吐量最高的那一季度对应的净利润是多少？（注意，是单季度对应的净利润，非累计净利润。）


SELECT *
FROM (SELECT TICKER_SYMBOL,
			YEAR(END_DATE) Year,
			QUARTER(END_DATE) QUARTER,
			SUM(VALUE) 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 -- income statement
(			SELECT TICKER_SYMBOL,
			YEAR(END_DATE) Year,
			QUARTER(END_DATE) QUARTER,
			SUM(N_INCOME) Amount
			FROM `income statement`
			GROUP BY TICKER_SYMBOL,YEAR(END_DATE),QUARTER(END_DATE) ) Income
ON BaseData.TICKER_SYMBOL = Income.TICKER_SYMBOL
AND BaseData.Year = Income.Year
AND BaseData.QUARTER = Income.QUARTER;



# 练习题10
# 使用Coupon Usage Data for O2O中的数据集《ccf_online_stage1_train.csv》和《ccf_offline_stage1_train.csv》，试找出在2016年6月期间，线上线下累计被使用优惠券满减最多的前3名商家。

-- SUBSTRING_INDEX(`Discount_rate`,':',-1) 满减金额

-- 线下使用优惠券用户
SELECT Merchant_id,SUM(discount_amount)
FROM(
SELECT off.Merchant_id,SUM(SUBSTRING_INDEX(`Discount_rate`,':',-1)) AS discount_amount
FROM ccf_offline_stage1_train AS off
WHERE off.Date IS NOT NULL AND off.Coupon_id IS NOT NULL
AND DATE_FORMAT(off.Date,'%Y-%m-%d') BETWEEN '2016-06-01' AND '2016-06-30'
GROUP BY off.Merchant_id
# ORDER BY off_discount_amount DESC;

UNION ALL

-- 线上使用优惠券用户
SELECT onl.Merchant_id,SUM(SUBSTRING_INDEX(`Discount_rate`,':',-1)) AS discount_amount
FROM ccf_online_stage1_train AS onl
WHERE onl.Date IS NOT NULL AND onl.Coupon_id IS NOT NULL
AND onl.Date BETWEEN '2016-06-01' AND '2016-06-30'
GROUP BY onl.Merchant_id
# ORDER BY on_discount_amount DESC;
) AS BaseData
GROUP BY Merchant_id
ORDER BY SUM(discount_amount) DESC
LIMIT 3;