如何根据另一个表的 MIN/MAX 填充日期表中的缺失行

Question

I have a table of project forecasts, where each row represents one month / one financial period

select 
2210004 as ProjectID, 
count(*) as aRowCount, 
min(finperiod) as earliest, 
max(FinPeriod) as latest 
from F where ProjectID = 22100004

ProjectID   aRowCount   earliest    latest
2210004     87          2018-09-30  2026-06-30

And I have a calendar table, spanning 50 years, where each row represents one month / one financial period

DECLARE @StartDateTime DATETIME
DECLARE @EndDateTime DATETIME

SET @StartDateTime = '2010-07-01'
SET @EndDateTime = '2060-06-30';

WITH DateRange(DateData) AS 
(
    SELECT @StartDateTime as Date
    UNION ALL
    SELECT DATEADD(M,1,DateData)
    FROM DateRange 
    WHERE DateData < @EndDateTime
)
SELECT eomonth(DateData) as FinPeriod
into dbo.CalendarMonths
FROM DateRange
OPTION (MAXRECURSION 0)
GO

Now the problem I need to solve is that the forecasts sometime have gaps in them. For example, for a particular ProjectID, there may be nothing for Jan 2021 or Feb 2021. This is the end result I want to create.

FinPeriod   ProjectID   FinPeriod
2020-12-31  22100004    2020-12-31
2021-01-31  NULL        NULL
2021-02-28  NULL        NULL
2021-03-31  22100004    2021-03-31
2021-04-30  22100004    2021-04-30
2021-05-31  NULL        NULL
2021-06-30  22100004    2021-06-30
2021-07-31  NULL        NULL
2021-08-31  NULL        NULL
2021-09-30  22100004    2021-09-30

This table above is what I want to create. It's the original forecast table, but with rows for every single financial period.

This code sort of does the job

select 
    CM.FinPeriod, f.ProjectID, f.FinPeriod
from 
    dbo.CalendarMonths CM 
    
    LEFT JOIN (select ProjectID, finperiod from f where ProjectID=22100004 )  F 
        ON CM.FinPeriod = f.FinPeriod

group by 
    cm.FinPeriod, f.ProjectID, f.FinPeriod

having CM.FinPeriod between '2018-09-30' and '2026-06-30'

But the problems are:

1. When I take out the ProjectID filter on the join, and change the filter criteria in the having statement it seems to destroy the LEFT JOIN and I lose the NULLS
2. And I don't want to end up having every calendar month (50 years worth) for every ProjectID

So this doesn't work

select 
    CM.FinPeriod, f.ProjectID, f.FinPeriod
from 
    dbo.CalendarMonths CM LEFT JOIN F 
        ON CM.FinPeriod = f.FinPeriod
where 
    ProjectID=22100004
    
group by 
    cm.FinPeriod, f.ProjectID, f.FinPeriod

having 
    CM.FinPeriod >= MIN(F.FinPeriod) and CM.FinPeriod <= MAX(F.FinPeriod) 

order by 
    ProjectID, cm.FinPeriod

The issue seems to be the having statement, but I can't see what the problem is.

SQL Server 2019

Here is the sample data

CREATE TABLE [F](
    [ProjectID] [bigint] NULL,
    [FiscalYear] [bigint] NULL,
    [CalendarYear] [bigint] NULL,
    [FinPeriod] [date] NULL,
    [MonthOf] [varchar](max) NULL,
    [PlannedSpend] [float] NULL,
    [Actual] [float] NULL,
    [Forecast] [float] NULL,
    [CumulativePlannedSpend] [float] NULL,
    [CumulativeActual] [float] NULL,
    [CumulativeForecast] [float] NULL,
    [EstimatedCostAtCompletion] [float] NULL,
    [VarianceDollars] [float] NULL,
    [VariancePercentage] [float] NULL,
    [VarianceUnderControl] [float] NULL
) ON [PRIMARY] TEXTIMAGE_ON [PRIMARY]
GO

    
    
   
INSERT [F] ([ProjectID], [FiscalYear], [CalendarYear], [FinPeriod], [MonthOf], [PlannedSpend], [Actual], [Forecast], [CumulativePlannedSpend], [CumulativeActual], [CumulativeForecast], [EstimatedCostAtCompletion], [VarianceDollars], [VariancePercentage], [VarianceUnderControl]) VALUES (22100004, 2023, 2022, CAST(N'2022-08-31' AS Date), N'AUG 2022', 0, 24735, 0, 2215753.48, 1397015.48, NULL, 1397015.48, 818738, 0.37, 0)
INSERT [F] ([ProjectID], [FiscalYear], [CalendarYear], [FinPeriod], [MonthOf], [PlannedSpend], [Actual], [Forecast], [CumulativePlannedSpend], [CumulativeActual], [CumulativeForecast], [EstimatedCostAtCompletion], [VarianceDollars], [VariancePercentage], [VarianceUnderControl]) VALUES (22100004, 2023, 2022, CAST(N'2022-09-30' AS Date), N'SEP 2022', 0, 4685, 0, 2215753.48, 1401700.48, NULL, 1401700.48, 814053, 0.367, 0)
INSERT [F] ([ProjectID], [FiscalYear], [CalendarYear], [FinPeriod], [MonthOf], [PlannedSpend], [Actual], [Forecast], [CumulativePlannedSpend], [CumulativeActual], [CumulativeForecast], [EstimatedCostAtCompletion], [VarianceDollars], [VariancePercentage], [VarianceUnderControl]) VALUES (22100004, 2023, 2022, CAST(N'2022-12-31' AS Date), N'DEC 2022', 5000, 0, 5000, 2220753.48, NULL, 5000, 1406700.48, 814053, 0.367, 0)
INSERT [F] ([ProjectID], [FiscalYear], [CalendarYear], [FinPeriod], [MonthOf], [PlannedSpend], [Actual], [Forecast], [CumulativePlannedSpend], [CumulativeActual], [CumulativeForecast], [EstimatedCostAtCompletion], [VarianceDollars], [VariancePercentage], [VarianceUnderControl]) VALUES (22100004, 2023, 2023, CAST(N'2023-01-31' AS Date), N'JAN 2023', 5000, 0, 5000, 2225753.48, NULL, 10000, 1411700.48, 814053, 0.366, 0)
INSERT [F] ([ProjectID], [FiscalYear], [CalendarYear], [FinPeriod], [MonthOf], [PlannedSpend], [Actual], [Forecast], [CumulativePlannedSpend], [CumulativeActual], [CumulativeForecast], [EstimatedCostAtCompletion], [VarianceDollars], [VariancePercentage], [VarianceUnderControl]) VALUES (22100004, 2023, 2023, CAST(N'2023-02-28' AS Date), N'FEB 2023', 10000, 0, 10000, 2235753.48, NULL, 20000, 1421700.48, 814053, 0.364, 0)
INSERT [F] ([ProjectID], [FiscalYear], [CalendarYear], [FinPeriod], [MonthOf], [PlannedSpend], [Actual], [Forecast], [CumulativePlannedSpend], [CumulativeActual], [CumulativeForecast], [EstimatedCostAtCompletion], [VarianceDollars], [VariancePercentage], [VarianceUnderControl]) VALUES (22100004, 2023, 2023, CAST(N'2023-03-31' AS Date), N'MAR 2023', 150000, 0, 150000, 2385753.48, NULL, 170000, 1571700.48, 814053, 0.341, 0)
INSERT [F] ([ProjectID], [FiscalYear], [CalendarYear], [FinPeriod], [MonthOf], [PlannedSpend], [Actual], [Forecast], [CumulativePlannedSpend], [CumulativeActual], [CumulativeForecast], [EstimatedCostAtCompletion], [VarianceDollars], [VariancePercentage], [VarianceUnderControl]) VALUES (22100004, 2023, 2023, CAST(N'2023-05-31' AS Date), N'MAY 2023', 200000, 0, 200000, 2735753.48, NULL, 520000, 1921700.48, 814053, 0.298, 0)