SELECT toStartOfDaytoDateTimeDs AS datetime_day_Ds PlatformType AS PlatformType maxMergedUrlCount AS index_mergedviewcountFROM log_iMonkey_Wechat_overviewWHERE Debug = 0 AND PlatformType IN 34

您可以使用以下查询来获取每个日期和平台类型的最大MergedUrlCount和MergedViewCount的值作为变量index_mergedviewcount：

SELECT 
    toStartOfDay(toDateTime(Ds)) AS datetime_day_Ds,
    max(MergedUrlCount) FILTER (WHERE PlatformType = 3) AS MergedUrlCount_PlatformType_3,
    max(MergedUrlCount) FILTER (WHERE PlatformType = 4) AS MergedUrlCount_PlatformType_4,
    max(MergedViewCount) FILTER (WHERE PlatformType = 3) AS MergedViewCount_PlatformType_3,
    max(MergedViewCount) FILTER (WHERE PlatformType = 4) AS MergedViewCount_PlatformType_4,
    GREATEST(
        max(MergedUrlCount) FILTER (WHERE PlatformType = 3),
        max(MergedUrlCount) FILTER (WHERE PlatformType = 4),
        max(MergedViewCount) FILTER (WHERE PlatformType = 3),
        max(MergedViewCount) FILTER (WHERE PlatformType = 4)
    ) AS index_mergedviewcount
FROM log_iMonkey_Wechat_overview
WHERE Debug = 0 AND PlatformType IN (3,4) AND
    Ds >= toDateTime(#day_wechat_views_time.start#) AND Ds <= toDateTime(#day_wechat_views_time.end#) 
GROUP BY datetime_day_Ds
ORDER BY datetime_day_Ds DESC
LIMIT 5000

这个查询将为每个日期计算出PlatformType等于3和4的MergedUrlCount和MergedViewCount的最大值，并使用GREATEST函数找出这四个值中的最大值作为变量index_mergedviewcount