Power Query 自定义列公式答案

【问题标题】：Power Query Custom Column FormulaPower Query 自定义列公式
【发布时间】：2021-10-26 19:57:52
【问题描述】：

我正在尝试汇总通过电源查询引入的数据并添加自定义列。本质上，我想按组计算数据馈送中所有前非零月份，并在单独的列中报告该计数。

例如，输入数据如下所示：

Index	UNIQUEID	PRMS Reserves Category	PRMS Reserves Sub Category	Date	Gross Oil Well Head Volume (MBBL)
36	INPTegrxW4xbF6	probable	undeveloped	6/1/2027	0
36	INPTegrxW4xbF6	probable	undeveloped	7/1/2027	0
36	INPTegrxW4xbF6	probable	undeveloped	8/1/2027	0
36	INPTegrxW4xbF6	probable	undeveloped	9/1/2027	14
36	INPTegrxW4xbF6	probable	undeveloped	10/1/2027	24
36	INPTegrxW4xbF6	probable	undeveloped	11/1/2027	27
37	INPTegrxW4xbF7	probable	undeveloped	6/1/2027	0
37	INPTegrxW4xbF7	probable	undeveloped	7/1/2027	0
37	INPTegrxW4xbF7	probable	undeveloped	8/1/2027	0
37	INPTegrxW4xbF7	probable	undeveloped	9/1/2027	14
37	INPTegrxW4xbF7	probable	undeveloped	10/1/2027	24
37	INPTegrxW4xbF7	probable	undeveloped	11/1/2027	27
38	INPTegrxW4xbF8	probable	undeveloped	6/1/2027	0
38	INPTegrxW4xbF8	probable	undeveloped	7/1/2027	0
38	INPTegrxW4xbF8	probable	undeveloped	8/1/2027	0
38	INPTegrxW4xbF8	probable	undeveloped	9/1/2027	14
38	INPTegrxW4xbF8	probable	undeveloped	10/1/2027	24
38	INPTegrxW4xbF8	probable	undeveloped	11/1/2027	27
39	INPTegrxW4xbF9	proved	undeveloped	5/1/2027	0
39	INPTegrxW4xbF9	proved	undeveloped	6/1/2027	0
39	INPTegrxW4xbF9	proved	undeveloped	7/1/2027	0
39	INPTegrxW4xbF9	proved	undeveloped	8/1/2027	14
39	INPTegrxW4xbF9	proved	undeveloped	9/1/2027	24
39	INPTegrxW4xbF9	proved	undeveloped	10/1/2027	27

我希望最终查询看起来像这样：

PRMS Reserves Category	PRMS Reserves Sub Category	Date	Sum of Gross Oil	Count
probable	undeveloped	6/1/2027	0	0
probable	undeveloped	7/1/2027	0	0
probable	undeveloped	8/1/2027	0	0
probable	undeveloped	9/1/2027	42	3
probable	undeveloped	10/1/2027	72	0
probable	undeveloped	11/1/2027	81	0
proved	undeveloped	5/1/2027	0	0
proved	undeveloped	6/1/2027	0	0
proved	undeveloped	7/1/2027	0	0
proved	undeveloped	8/1/2027	14	1
proved	undeveloped	9/1/2027	24	0
proved	undeveloped	10/1/2027	27	0

有人对如何在 power query 的自定义列中编写函数有建议吗？

【问题讨论】：

由于您的所有日期都在月初，只需按 Date 和 PMRS Reserves Category 分组，并为聚合做 SUM 的 @987654326 @.
谢谢——总油量的总和很简单。计算第一个非零总油量并按组计算是我遇到麻烦的地方。
不确定您使用的逻辑。给定 2027 年 10 月 1 日的索引 36、37、38，您对输出的期望是什么。共有三行，每行的 Volume 为 24。在输出表中，volume 显示 72，Count 显示 0。但是对于 2027 年 9 月 1 日（相同的行索引 = 36、37、38），您显示 3 表示计数
没错。 9/1/2027 有三个记录，它们是 probable undeveloped 中的第一个非零条目。 10/1/2027 在该组中有三个非零记录，但它们是第二个（而不是第一个）非零元素。

标签： dax powerquery m

【解决方案1】：

您似乎需要计算每个类别组合的最小日期MinDate，然后检查是否Date = MinDate

let
    Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("pZGxCoMwFEX/JbNgklqT/EChS+lQ6CAODYYuUkWo+PlN62DNe8vV7SVwznBPVYlDKTJxvlxv4TlM92Lyp+9HP3T+4dsQz/erCWNouz408VXmKtdSm3hKUWcwb3bydifvFl4VWwRKLga9zaD+DGY2mNRgsAgYTyNgPI2A8UwETMBFAA1MBJsaLBYB42kEjKcRMJ6JgAm4CKCBieBSg5sN449Y80cyAULTgAhN8yG0peMjuKPLI/iqXNy9/gA=", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [Index = _t, UNIQUEID = _t, #"PRMS Reserves Category" = _t, #"PRMS Reserves Sub Category" = _t, Date = _t, #"Gross Oil Well Head Volume (MBBL)" = _t]),
    #"Changed Type" = Table.TransformColumnTypes(Source,{{"Index", Int64.Type}, {"UNIQUEID", type text}, {"PRMS Reserves Category", type text}, {"PRMS Reserves Sub Category", type text}, {"Date", type date}, {"Gross Oil Well Head Volume (MBBL)", Int64.Type}}),
    #"Filtered Rows" = Table.SelectRows(#"Changed Type", each ([#"Gross Oil Well Head Volume (MBBL)"] <> 0)),
    #"Grouped Category" = Table.Group(#"Filtered Rows", {"PRMS Reserves Category", "PRMS Reserves Sub Category"}, {{"MinDate", each List.Min([Date]), type nullable date}}),
    #"Merged Queries" = Table.NestedJoin(#"Changed Type", {"PRMS Reserves Category", "PRMS Reserves Sub Category"}, #"Grouped Category", {"PRMS Reserves Category", "PRMS Reserves Sub Category"}, "Grouped Rows", JoinKind.LeftOuter),
    #"Expanded Grouped Rows" = Table.ExpandTableColumn(#"Merged Queries", "Grouped Rows", {"MinDate"}, {"MinDate"}),
    #"Added Custom" = Table.AddColumn(#"Expanded Grouped Rows", "Custom", each if [Date] = [MinDate] then 1 else 0, Int64.Type),
    #"Grouped Category and Date" = Table.Group(#"Added Custom", {"PRMS Reserves Category", "PRMS Reserves Sub Category", "Date"}, {{"Gross", each List.Sum([#"Gross Oil Well Head Volume (MBBL)"]), type nullable number}, {"Count", each List.Sum([Custom]), type number}})
in
    #"Grouped Category and Date"

这会在过滤掉0 值后通过在Date 列上取最小值来计算#"Grouped Category" 步骤中的MinDate：

然后它将这个MinDate 列合并回原始表，并使用逻辑if [Date] = [MinDate] then 1 else 0 添加一个自定义列。

上面显示的前三列中的最后一步分组，对总石油和海关求和。

【讨论】：