第三模块： 窗口练习 使用到的表： website_url_info 、website_pv_info -- 13、求停留时间最长的url，并给出停留时长，按照停留时长倒叙排序 -- 14、求2018-04-11到2018-04-15 之间，每天的pv 和 uv 数量，并按照日期升序，pv降序排序 -- 15、求2018年4月每个时段的，pv、uv 数量、按照uv转化率取top5. -- uv 转化率 = 转化人数/uv数量 ，假设现在转化人数是 1

13、求停留时间最长的url，并给出停留时长，按照停留时长倒叙排序

SELECT 
    website_url_info.url,
    SUM(website_pv_info.stay_time) AS total_stay_time 
FROM 
    website_pv_info 
    JOIN website_url_info 
        ON website_pv_info.url_id = website_url_info.id 
GROUP BY 
    website_url_info.url 
ORDER BY 
    total_stay_time DESC 
LIMIT 1;

14、求2018-04-11到2018-04-15 之间，每天的pv 和 uv 数量，并按照日期升序，pv降序排序

SELECT 
    DATE(website_pv_info.create_time) AS date,
    COUNT(DISTINCT website_pv_info.user_id) AS uv,
    COUNT(*) AS pv 
FROM 
    website_pv_info 
WHERE 
    website_pv_info.create_time BETWEEN '2018-04-11' AND '2018-04-15' 
GROUP BY 
    DATE(website_pv_info.create_time) 
ORDER BY 
    date ASC, 
    pv DESC;

15、求2018年4月每个时段的，pv、uv 数量、按照uv转化率取top5

SELECT 
    DATE_FORMAT(website_pv_info.create_time, '%H:00-%H:59') AS time_range,
    COUNT(*) AS pv,
    COUNT(DISTINCT website_pv_info.user_id) AS uv,
    CAST(COUNT(CASE WHEN website_pv_info.is_conversion = 1 THEN 1 ELSE NULL END) / COUNT(DISTINCT website_pv_info.user_id) AS DECIMAL(10,2)) AS uv_conversion_rate
FROM 
    website_pv_info 
WHERE 
    website_pv_info.create_time BETWEEN '2018-04-01' AND '2018-04-30' 
GROUP BY 
    time_range 
ORDER BY 
    uv_conversion_rate DESC 
LIMIT 5;

注：上述SQL语句仅供参考，具体情况下需根据实际数据表结构和业务需求进行调整。