gpt4 book ai didi

azure-log-analytics - 库斯托 : How to unpivot - turn columns into rows?

转载 作者:行者123 更新时间:2023-12-04 16:28:05 24 4
gpt4 key购买 nike

使用帮助集群上 Samples 数据库中的 StormEvents 表:

StormEvents
| where State startswith "AL"
| where EventType has "Wind"
| where StartTime == "2007-01-02T02:16:00Z"
| project StartTime, State, EventType, InjuriesDirect, InjuriesIndirect, DeathsDirect, DeathsIndirect

我想要表单的基于行的输出:

desired un-pivoted output

我看到了 pivot() 函数,但它似乎只向另一个方向发展,从行到列。

我一直在尝试各种 pack() 想法,但似乎无法获得所需的输出。

例子:

StormEvents
| where State startswith "AL"
| where EventType has "Wind"
| where StartTime == "2007-01-02T02:16:00Z"
| project StartTime, State, EventType, InjuriesDirect, InjuriesIndirect, DeathsDirect, DeathsIndirect
| extend Packed = pack(
"CasualtyType", "InjuriesDirect", "CasualtyCount", InjuriesDirect,
"CasualtyType", "InjuriesIndirect", "CasualtyCount", InjuriesIndirect,
"CasualtyType", "DeathsDirect", "CasualtyCount", DeathsDirect,
"CasualtyType", "DeathsIndirect", "CasualtyCount", DeathsIndirect
)
| project-away InjuriesDirect, InjuriesIndirect, DeathsDirect, DeathsIndirect
| mv-expand Packed

这给了我太多行,我不清楚如何将它们转换为列。

用于所需输出的正确模式是什么?

最佳答案

您可以尝试以下方法:

let casualty_types = dynamic(["InjuriesDirect", "DeathsDirect", "InjuriesIndirect", "DeathsIndirect"]);
StormEvents
| where State startswith "AL"
| where EventType has "Wind"
| where StartTime == "2007-01-02T02:16:00Z"
| project StartTime, State, EventType, properties = pack_all()
| mv-apply casualty_type = casualty_types to typeof(string) on (
project casualty_type, casualty_count = tolong(properties[casualty_type])
)
| project-away properties

关于azure-log-analytics - 库斯托 : How to unpivot - turn columns into rows?,我们在Stack Overflow上找到一个类似的问题: https://stackoverflow.com/questions/58820720/

24 4 0
Copyright 2021 - 2024 cfsdn All Rights Reserved 蜀ICP备2022000587号
广告合作:1813099741@qq.com 6ren.com