I would like to add a counter to records in my table, that fall in the last rolling 12 months.
Below is what I've tried and what I would like to achieve. The current counter doesn't reset to 1 for a different customer ID.
select *,
case when order_date >= DATEADD(MONTH, -12, GETDATE()) then dense_rank() over (partition by customer_id order by case when order_date >= DATEADD(MONTH, -12, GETDATE()) then order_date end asc, order_no)
end as counter
from mydata
customer_id
order_no
order_date
counter
what I want
ABC
1213
Wednesday, May 12, 2021
1
1
ABC
1257
Saturday, May 15, 2021
2
2
ABC
1345
Saturday, May 22, 2021
3
3
ABC
4562
Saturday, May 22, 2021
4
4
ABC
4362
Saturday, May 29, 2021
5
5
ABC
1421
Tuesday, June 1, 2021
6
6
GHI
5525
Wednesday, January 20, 2021
NULL
NULL
GHI
2452
Friday, February 26, 2021
NULL
NULL
GHI
1452
Tuesday, March 2, 2021
3
1
GHI
3525
Wednesday, March 3, 2021
4
2
GHI
4242
Thursday, March 4, 2021
5
3
GHI
1341
Thursday, March 4, 2021
6
4
GHI
1341
Thursday, March 4, 2021
6
4
GHI
5241
Saturday, March 13, 2021
7
5
GHI
1425
Saturday, March 20, 2021
8
6
GHI
5213
Wednesday, March 31, 2021
9
7
GHI
6312
Saturday, April 17, 2021
10
8
GHI
6312
Saturday, April 17, 2021
10
8
If you want to restart at a given point, you need to partition at that point. So not only by customer_id but also your condition:
SELECT mydata.*
, CASE
WHEN calc.in_scope = 1
THEN DENSE_RANK() OVER (PARTITION BY mydata.customer_id, calc.in_scope
ORDER BY order_date, order_no)
END AS counter
FROM mydata
CROSS
APPLY (SELECT IIF(order_date >= DATEADD(MONTH, -12, GETDATE()), 1, 0) AS in_scope
) calc
I've moved the calculation into CROSS APPLY to avoid repeating it, in case it needs to change in the future you'd then only change it in one place.
Working demo on dbfiddle
Related
I want to calculate the Fiscal Week of the Year with the following rules:
The fiscal year always starts on June/01
The week always starts on Sunday
Samples are provided for 2019 but ideally it should work for any year
A few samples of correct values are provided with the screenshot attached
I tried to do something like TO_NUMBER(TO_CHAR(TO_DATE(DATE_ID + 1,'DD-Mon-YY'),'IW')) -21 but towards the end of the Calendar year I start to get negatives
SELECT
DATE_ID
, WEEK_OF_YEAR
FROM DATE_DIM
WHERE
DATE_ID IN
(
20190601
, 20190602
, 20190915
, 20191228
, 20191229
, 20200101
, 20200601
, 20200606
, 20200607
)
ORDER BY DATE_ID ASC
;
SELECT TO_CHAR(ADD_MONTHS(SYSDATE,+7),'IW')
FROM dual
Or in your case,
SELECT date_id,TO_CHAR(ADD_MONTHS(TO_DATE(date_id,'YYYYMMDD'),+7),'IW') week_of_year
FROM date_dim
If you don't like the ISO dating where week 1 is the week beginning Sunday during which the 1st of the year falls, you can try offsetting it by moving it back to the previous sunday (TRUNC(..,'D'), then advance a week, then add the 7 months. See if this works for you:
SELECT date_id, TO_CHAR(ADD_MONTHS(TRUNC(date_id,'D') + 7,7),'IW') week_of_year
FROM date_dim
You got confused with your date formats ('yyyymmdd' vs. 'DD-Mon-YY'), so I am using a real date (mydate) in my answer. Convert your string or number to a proper date and you are there :-)
The important thing is to check whether your date is >= June 1. Once this is done you can subtract that year's June 1 or the previous year's one. Well, more or less :-)
select
mydate,
to_char(mydate, 'DY') as tag,
trunc
(
case when to_char(mydate, 'mmdd') >= '0601' then
trunc(mydate + 1, 'iw') + 6 - to_date(to_char(mydate, 'yyyy') || '0601', 'yyyymmdd')
else
trunc(mydate + 1, 'iw') + 6 - to_date(to_char(extract(year from mydate) - 1) || '0601', 'yyyymmdd')
end / 7
) + 1 as fiscal_week
from ...
order by mydate;
Demo: https://dbfiddle.uk/N5pX_5cV
UPDATED ANSWER
You can try this - provide the date like 31-MAY-19 to start your calendar from 01-JUN-19 and define how many days you want. There is a shift of 5 months as the code was taken from my regular calendar and adjusted the start of week to sunday...Tested OK for periods up to 2 years per run. So if you want 6 years you'll have to run this 3 times...
WITH
base_calendar AS
(
SELECT CurrDate AS Day_ID,
TO_CHAR(CurrDate,'Day') AS Week_Day_Full,
TO_CHAR(CurrDate,'DY') AS Week_Day_Short,
TO_NUMBER(TRIM(leading '0' FROM TO_CHAR(CurrDate,'D'))) AS Day_Num_of_Week,
--
MAX(CASE WHEN To_Char(CurrDate, 'ddmm') = '2902' THEN EXTRACT(YEAR From CurrDate) END) OVER(Order By CurrDate Rows Between Unbounded Preceding And Current Row) AS last_leap_year,
Count(CASE WHEN ( TO_CHAR(CurrDate,'DY') = 'SUN' And
CurrDate Between To_date('01.06.' || To_Char(EXTRACT(YEAR From CurrDate)), 'dd.mm.yyyy') And To_date('31.05.' || To_Char(EXTRACT(YEAR From CurrDate) + 1), 'dd.mm.yyyy') )
OR
( TO_CHAR(CurrDate,'DY') = 'SUN' And
CurrDate Between To_date('01.06.' || To_Char(EXTRACT(YEAR From CurrDate) - 1), 'dd.mm.yyyy') And To_date('31.05.' || To_Char(EXTRACT(YEAR From CurrDate) ), 'dd.mm.yyyy') )
THEN 1
END) OVER(Order By CurrDate Rows Between Unbounded Preceding And Current Row) AS cnt_sundays
FROM
(
SELECT level n,
-- Calendar starts at the day after this date
TO_DATE('31/05/2019','DD/MM/YYYY') + NUMTODSINTERVAL(level,'DAY') CurrDate
FROM dual
-- Change for the number of days to be added to the table.
CONNECT BY level <= 731
)
)
SELECT DISTINCT
day_id,
CASE WHEN (cnt_sundays + 1 ) - ( (EXTRACT(YEAR From Add_Months(Day_id, - 5)) -EXTRACT(YEAR From MIN(Day_id) OVER()) ) * 53) >= 54 THEN 1
WHEN EXTRACT(YEAR From day_id) <> last_leap_year And day_id > = To_Date('01.06.' || To_Char(EXTRACT(YEAR From day_id)), 'dd.mm.yyyy' )
THEN (cnt_sundays + 1) - ( (EXTRACT(YEAR From Add_Months(Day_id, - 5)) -EXTRACT(YEAR From MIN(Day_id) OVER()) ) * 53) + 1
ELSE (cnt_sundays + 1) - ( (EXTRACT(YEAR From Add_Months(Day_id, - 5)) -EXTRACT(YEAR From MIN(Day_id) OVER()) ) * 53)
END week_of_year,
--
week_day_full,
week_day_short,
CASE week_day_short
WHEN 'SUN' THEN 1
WHEN 'MON' THEN 2
WHEN 'TUE' THEN 3
WHEN 'WED' THEN 4
WHEN 'THU' THEN 5
WHEN 'FRI' THEN 6
WHEN 'SAT' THEN 7
END AS day_num_of_week
FROM base_calendar
ORDER BY day_id
R E S U L T S :
DAY_ID WEEK_OF_YEAR WEEK_DAY_FULL WEEK_DAY_SHORT DAY_NUM_OF_WEEK
--------- ------------ ------------- -------------- ---------------
01-JUN-19 1 Saturday SAT 7
02-JUN-19 2 Sunday SUN 1
03-JUN-19 2 Monday MON 2
04-JUN-19 2 Tuesday TUE 3
05-JUN-19 2 Wednesday WED 4
... ...
... ...
13-SEP-19 16 Friday FRI 6
14-SEP-19 16 Saturday SAT 7
15-SEP-19 17 Sunday SUN 1
16-SEP-19 17 Monday MON 2
... ...
30-DEC-19 32 Monday MON 2
31-DEC-19 32 Tuesday TUE 3
01-JAN-20 32 Wednesday WED 4
02-JAN-20 32 Thursday THU 5
...
27-FEB-20 40 Thursday THU 5
28-FEB-20 40 Friday FRI 6
29-FEB-20 40 Saturday SAT 7
01-MAR-20 41 Sunday SUN 1
02-MAR-20 41 Monday MON 2
...
29-MAY-20 53 Friday FRI 6
30-MAY-20 53 Saturday SAT 7
31-MAY-20 1 Sunday SUN 1
01-JUN-20 1 Monday MON 2
02-JUN-20 1 Tuesday TUE 3
03-JUN-20 1 Wednesday WED 4
04-JUN-20 1 Thursday THU 5
05-JUN-20 1 Friday FRI 6
06-JUN-20 1 Saturday SAT 7
07-JUN-20 2 Sunday SUN 1
... ...
... ...
22-MAY-21 51 Saturday SAT 7
23-MAY-21 52 Sunday SUN 1
24-MAY-21 52 Monday MON 2
25-MAY-21 52 Tuesday TUE 3
26-MAY-21 52 Wednesday WED 4
27-MAY-21 52 Thursday THU 5
28-MAY-21 52 Friday FRI 6
29-MAY-21 52 Saturday SAT 7
30-MAY-21 53 Sunday SUN 1
31-MAY-21 53 Monday MON 2
731 rows selected
... with year 2025 - part of the result mentioned in comments above ...
... ...
29-MAY-25 53 Thursday THU 5
30-MAY-25 53 Friday FRI 6
31-MAY-25 53 Saturday SAT 7
01-JUN-25 1 Sunday SUN 1
02-JUN-25 1 Monday MON 2
03-JUN-25 1 Tuesday TUE 3
04-JUN-25 1 Wednesday WED 4
05-JUN-25 1 Thursday THU 5
06-JUN-25 1 Friday FRI 6
07-JUN-25 1 Saturday SAT 7
08-JUN-25 2 Sunday SUN 1
... ...
I have a list of date, fiscal week, and fiscal year:
DATE_VALUE FISCAL_WEEK FISCAL_YEAR_VALUE
14-Dec-20 51 2020
15-Dec-20 51 2020
16-Dec-20 51 2020
17-Dec-20 51 2020
18-Dec-20 51 2020
19-Dec-20 51 2020
20-Dec-20 51 2020
21-Dec-20 52 2020
22-Dec-20 52 2020
23-Dec-20 52 2020
24-Dec-20 52 2020
25-Dec-20 52 2020
26-Dec-20 52 2020
27-Dec-20 52 2020
28-Dec-20 1 2021
29-Dec-20 1 2021
30-Dec-20 1 2021
31-Dec-20 1 2021
1-Jan-21 1 2021
2-Jan-21 1 2021
3-Jan-21 1 2021
4-Jan-21 2 2021
5-Jan-21 2 2021
6-Jan-21 2 2021
7-Jan-21 2 2021
8-Jan-21 2 2021
9-Jan-21 2 2021
10-Jan-21 2 2021
11-Jan-21 3 2021
12-Jan-21 3 2021
13-Jan-21 3 2021
14-Jan-21 3 2021
15-Jan-21 3 2021
16-Jan-21 3 2021
17-Jan-21 3 2021
18-Jan-21 4 2021
19-Jan-21 4 2021
20-Jan-21 4 2021
21-Jan-21 4 2021
22-Jan-21 4 2021
23-Jan-21 4 2021
24-Jan-21 4 2021
20-Dec-21 52 2021
21-Dec-21 52 2021
22-Dec-21 52 2021
23-Dec-21 52 2021
24-Dec-21 52 2021
25-Dec-21 52 2021
26-Dec-21 52 2021
27-Dec-21 53 2021
28-Dec-21 53 2021
29-Dec-21 53 2021
30-Dec-21 53 2021
31-Dec-21 53 2021
1-Jan-22 53 2021
2-Jan-22 53 2021
3-Jan-22 1 2022
4-Jan-22 1 2022
5-Jan-22 1 2022
6-Jan-22 1 2022
7-Jan-22 1 2022
8-Jan-22 1 2022
9-Jan-22 1 2022
10-Jan-22 2 2022
11-Jan-22 2 2022
12-Jan-22 2 2022
13-Jan-22 2 2022
14-Jan-22 2 2022
15-Jan-22 2 2022
16-Jan-22 2 2022
17-Jan-22 3 2022
18-Jan-22 3 2022
19-Jan-22 3 2022
20-Jan-22 3 2022
21-Jan-22 3 2022
22-Jan-22 3 2022
23-Jan-22 3 2022
24-Jan-22 4 2022
25-Jan-22 4 2022
26-Jan-22 4 2022
27-Jan-22 4 2022
28-Jan-22 4 2022
29-Jan-22 4 2022
30-Jan-22 4 2022
I want to pull the last 4 weeks prior to the current week AND the same 4 weeks of the year before. Please see example 1. This works fine when all 4 weeks are within the same year. But when it comes to the beginning of a year when 1 or more weeks are in the current year but the other are in the previous year, I am not able to get the desired output below:
FISCAL_YEAR_VALUE FISCAL_WEEK
2020 51
2020 52
2021 2
2021 1
2021 52
2021 53
2022 1
2022 2
The code I have is below. I am using the date of 21-JAN-22 as an example:
SELECT
FISCAL_YEAR_VALUE,
FISCAL_WEEK
FROM TABLE_NAME
WHERE FISCAL_YEAR_VALUE IN (SELECT *
FROM (WITH T AS (
SELECT DISTINCT FISCAL_YEAR_VALUE
FROM TABLE_NAME
WHERE TRUNC(DATE_VALUE) <= TRUNC(TO_DATE('21-JAN-22'))--TEST DATE
ORDER BY FISCAL_YEAR_VALUE DESC
FETCH NEXT 2 ROWS ONLY
)
SELECT FISCAL_YEAR_VALUE
FROM T ORDER BY FISCAL_YEAR_VALUE
)
)
AND FISCAL_WEEK IN (SELECT *
FROM (WITH T AS (
SELECT DISTINCT FISCAL_WEEK, FISCAL_YEAR_VALUE
FROM TABLE_NAME
WHERE TRUNC(DATE_VALUE) <= TRUNC(TO_DATE('21-JAN-22'))--TEST DATE
ORDER BY FISCAL_YEAR_VALUE DESC, FISCAL_WEEK DESC
OFFSET 1 ROWS
FETCH NEXT 4 ROWS ONLY
)
SELECT FISCAL_WEEK
FROM T ORDER BY FISCAL_YEAR_VALUE, FISCAL_WEEK
)
)
GROUP BY FISCAL_YEAR_VALUE, FISCAL_WEEK
ORDER BY FISCAL_YEAR_VALUE, FISCAL_WEEK
Output of the code is:
FISCAL_YEAR_VALUE FISCAL_WEEK
2021 2
2021 1
2021 52
2021 53
2022 1
2022 2
As you can see, the last 2 weeks of year 2020 are not included. Please see example 2. How can I also include this exception in the code to make it dynamic? Any help would be greatly appreciated!
To find the values this year, you can use:
SELECT DISTINCT fiscal_year_value, fiscal_week
FROM table_name
WHERE date_value < TRUNC(SYSDATE, 'IW')
AND date_value >= TRUNC(SYSDATE, 'IW') - INTERVAL '28' DAY
To find the values from the previous year, you can find the maximum fiscal week from this year and subtract 1 from the year and then use that to find the upper bound of the date_value for last fiscal year and, given that can use a similar range for last year:
WITH this_year (fiscal_year_value, fiscal_week) AS (
SELECT fiscal_year_value, fiscal_week
FROM table_name
WHERE date_value < TRUNC(SYSDATE, 'IW')
AND date_value >= TRUNC(SYSDATE, 'IW') - INTERVAL '28' DAY
),
max_last_year (max_date_value) AS (
SELECT MAX(date_value) + INTERVAL '1' DAY
FROM table_name
WHERE (fiscal_year_value, fiscal_week) IN (
SELECT fiscal_year_value - 1, fiscal_week
FROM this_year
ORDER BY fiscal_year_value DESC, fiscal_week DESC
FETCH FIRST ROW ONLY
)
)
SELECT fiscal_year_value, fiscal_week
FROM this_year
UNION
SELECT t.fiscal_year_value, t.fiscal_week
FROM table_name t
INNER JOIN max_last_year m
ON ( t.date_value < m.max_date_value
AND t.date_value >= m.max_date_value - INTERVAL '28' DAY);
Which, for the sample data:
Create Table table_name(DATE_VALUE DATE, FISCAL_WEEK INT, FISCAL_YEAR_VALUE INT);
INSERT INTO table_name (date_value, fiscal_week, fiscal_year_value)
SELECT DATE '2019-12-30' + LEVEL - 1, CEIL(LEVEL/7), 2020
FROM DUAL
CONNECT BY LEVEL <= 7 * 52
UNION ALL
SELECT DATE '2020-12-28' + LEVEL - 1, CEIL(LEVEL/7), 2021
FROM DUAL
CONNECT BY LEVEL <= 7 * 53
UNION ALL
SELECT DATE '2022-01-03' + LEVEL - 1, CEIL(LEVEL/7), 2022
FROM DUAL
CONNECT BY LEVEL <= 7 * 52;
Outputs:
FISCAL_YEAR_VALUE
FISCAL_WEEK
2022
38
2022
39
2022
40
2022
41
2021
38
2021
39
2021
40
2021
41
And if today's date was 2022-01-01, would output:
FISCAL_YEAR_VALUE
FISCAL_WEEK
2021
52
2021
53
2022
1
2022
2
2020
51
2020
52
2021
1
2021
2
There may be a simpler method but without any knowledge of how you calculate a fiscal year that is not immediately possible.
fiddle
I have to calculate sales based on WeekNo(Yearly) and Month.
Table1: Sales
ID SalDate Amount Region
1 2020-12-27 1000 USA
2 2020-12-28 1000 EU
3 2020-12-29 1000 AUS
4 2021-01-01 1000 USA
5 2021-01-02 1000 EU
6 2021-01-05 1000 AUS
7 2020-09-30 1000 EU
8 2020-10-01 1000 AUS
Select DateName(Month,SalDate)+' - '+Convert(Varchar(10),Year(SalDate)) Months,
'Week '+ Convert(Varchar(50), DATEPART(WEEK, SalDate)) As Weeks,
Sum(Amount) OrderValue
From [Sales]
Group By DateName(Month,SalDate)+' - '+Convert(Varchar(10),Year(SalDate)),
Convert(Varchar(50), DATEPART(WEEK, SalDate))
Months Weeks Total
January - 2021 Week 1 2000.00
January - 2021 Week 2 1000.00
December - 2020 Week 53 3000.00
October - 2020 Week 40 1000.00
September - 2020 Week 40 1000.00
But client want to merge Week1 Total with Week 53
And Week2 show as Week1.
And Week 40 Merge with Sep 2020.
I want result like below
Months Weeks Total
January - 2021 Week 1 1000.00
December - 2020 Week 53 5000.00 (2000+3000)
September - 2020 Week 40 2000.00 (1000+100)
Kindly help me to fix this problem.
Looking for an Oracle SQL query to show Month and Year starting from the current year- 1y and current year+1y.
Eg: December 2019, January 2020, February 2020,......December 2021
You can use the hierarchy query as follows:
SQL> SELECT trunc(ADD_MONTHS(ADD_MONTHS(sysdate,-12), LEVEL-1), 'Mon') as month_year
2 FROM DUAL CONNECT BY LEVEL <= 24 + 1;
MONTH_YEAR
--------------
December 2019
January 2020
February 2020
March 2020
April 2020
May 2020
June 2020
July 2020
August 2020
September 2020
October 2020
November 2020
December 2020
January 2021
February 2021
March 2021
April 2021
May 2021
June 2021
July 2021
August 2021
September 2021
October 2021
November 2021
December 2021
25 rows selected.
SQL>
There are multiple methods for doing this. I think simple examples like this are a good opportunity to learn about recursive CTEs:
with dates(yyyymm, n) as (
select trunc(sysdate, 'Mon') as yyyymm, 1 as n
from dual
union all
select add_months(yyyymm, -1), n + 1
from dates
where n <= 12
)
select yyyymm
from dates;
WITH d AS (
SELECT
'JAN' m,
2021 y
FROM
dual
), d1 AS (
SELECT
to_date(m || y, 'MONYYYY') first_day,
last_day(to_date(m || y, 'MONYYYY')) last_day1,
last_day(to_date(m || y, 'MONYYYY')) - to_date(m || y, 'MONYYYY') no_of_days
FROM
d
)
SELECT
level - 1 + first_day dates
FROM
d1
CONNECT BY
level <= no_of_days + 1;
I have an ASP.NET Core application, through controller endpoint I pass #by and #period string values to the SQL query.
#by takes one of the following values: day, week
#period takes one of the following values: week, month, year
When the #period is month or year, then #by is a week, else it's a day.
I have the following working query when the #period is a month or a year:
SELECT
l.region_id AS region_id,
'Region ' + r.region_desc AS region_name,
MIN(DATEADD(D, -(DATEPART(WEEKDAY, s.pos_date) - 1), s.pos_date)) AS date_pos,
CONVERT(VARCHAR(20), MIN(DATEADD(D, -(DATEPART(WEEKDAY, s.pos_date) - 1), s.pos_date)), 107) AS display_date_pos
FROM
incent_summary s
INNER JOIN
location l ON s.store_num = l.store_num
INNER JOIN
region r ON l.region_id = r.region_id
WHERE
s.pos_date >= DATEADD(day, #period , CONVERT(date, GETDATE()))
AND s.pos_date <= GETDATE()
GROUP BY
DATEPART (#by, s.pos_date),
l.region_id, r.region_desc
ORDER BY
DATEPART (#by, pos_date),
l.region_id, r.region_desc
The issue is when the #period is a week, #by is day, and the statement
MIN(DATEADD(D, -(DATEPART(WEEKDAY, s.pos_date) - 1), s.pos_date)) AS date_pos
returns the same day for all the 7 days.
Sample output when #period = year and #by = week:
region_id region_name date_pos display_date_pos
---------------------------------------------------------------------
34 Region 43 2019-12-29 00:00:00.000 Dec 29, 2019
50 Region 22 2019-12-29 00:00:00.000 Dec 29, 2019
34 Region 43 2020-01-05 00:00:00.000 Jan 05, 2020
50 Region 22 2020-01-05 00:00:00.000 Jan 05, 2020
34 Region 43 2020-01-12 00:00:00.000 Jan 12, 2020
50 Region 22 2020-01-12 00:00:00.000 Jan 12, 2020
34 Region 43 2020-01-19 00:00:00.000 Jan 19, 2020
50 Region 22 2020-01-19 00:00:00.000 Jan 19, 2020
34 Region 43 2020-01-26 00:00:00.000 Jan 26, 2020
50 Region 22 2020-01-26 00:00:00.000 Jan 26, 2020
Sample output when #period = week and #by = day:
region_id region_name date_pos display_date_pos
--------------------------------------------------------------------
34 Region 43 2020-07-12 00:00:00.000 Jul 12, 2020
50 Region 22 2020-07-12 00:00:00.000 Jul 12, 2020
34 Region 43 2020-07-12 00:00:00.000 Jul 12, 2020
50 Region 22 2020-07-12 00:00:00.000 Jul 12, 2020
34 Region 43 2020-07-19 00:00:00.000 Jul 19, 2020
50 Region 22 2020-07-19 00:00:00.000 Jul 19, 2020
34 Region 43 2020-07-19 00:00:00.000 Jul 19, 2020
50 Region 22 2020-07-19 00:00:00.000 Jul 19, 2020
34 Region 43 2020-07-19 00:00:00.000 Jul 19, 2020
50 Region 22 2020-07-19 00:00:00.000 Jul 19, 2020
How can I fix this?
SELECT
DATEADD(D, -(DATEPART(WEEKDAY, s.pos_date) - 1), s.pos_date)
Will always return the first day of the week because the logic is: "subtract from my date the number of days from sunday and add 1."
Sunday: 1 - 1 + 1 = 1 = Sunday
Monday: 2 - 2 + 1 = 1 = Sunday
.
.
.
Saturday: 7 - 7 + 1 = Sunday
That's fine when you want the first Sunday of the year/month/whatever. But the first sunday of every week is always... sunday. But in this case you really just need to take the MIN(s.pos_date) if #period is week.
There's probably some crazy way to do this in a single statement using quaternions or something else super mathy, but it's easiest to just use a case statement:
MIN
(
CASE
WHEN '#by' = 'day' THEN s.pos_date
ELSE DATEADD(D, -(DATEPART(WEEKDAY, s.pos_date) - 1), s.pos_date)
END
)
I'm not a C# programmer so I can't tell you the exact way to make sure the string DAY is passed to the query as "DAY" but I'm sure you can handle that part.
ALSO IMPORTANT The datepart "day" is day of month, so if you're going to possibly have a span greater than one month (but under a year), use dayofyear.