I have following data in a table, in table each group contain two values for ON/OFF in two rows. I want to convert rows into new columns "Col1=>Col1_Status_ON/OFF" new name based on status
here is the detail:
====================================================
| Group | DateTime | Status | Col1 | Col2 |
====================================================
| Group1 | 01-Jan-2019 | ON | 101 | 102 |
| Group1 | 01-Jan-2019 | OFF | 201 | 202 |
| Group2 | 01-Jan-2019 | ON | 301 | 302 |
| Group2 | 01-Jan-2019 | OFF | 401 | 402 |
| Group3 | 01-Jan-2019 | ON | 501 | 502 |
| Group4 | 01-Jan-2019 | OFF | 601 | 602 |
====================================================
I want query that return as follows
=======================================================================================
| Group | DateTime |Col1_Satus_ON|Col1_Satus_OFF|Col2_Status_ON|Col2_Status_OFF|
=======================================================================================
| Group1 | 01-Jan-2019| 101 | 201 | 102 | 202 |
| Group2 | 01-Jan-2019| 301 | 401 | 302 | 402 |
| Group3 | 01-Jan-2019| 501 | 601 | 502 | 602 |
=======================================================================================
below are the SQL to generate table data.
DROP TABLE IF EXISTS dbo.#MyTable;
SELECT CONVERT(VARCHAR(100),'')[Group], GETDATE()[DateTime],
CONVERT(VARCHAR(100),'')[Status],(0)[Col1],(0)[Col2]
INTO #MyTable WHERE 1=2;
INSERT INTO #MyTable ([Group],[Datetime],[Status],[Col1],[Col2])
SELECT 'Group1','01-Jan-2019 01:02:03','ON' ,101,102 UNION
SELECT 'Group1','01-Jan-2019 01:02:03','OFF',201,202 UNION
SELECT 'Group2','01-Jan-2019 01:02:03','ON' ,301,302 UNION
SELECT 'Group2','01-Jan-2019 01:02:03','OFF',401,402 UNION
SELECT 'Group3','01-Jan-2019 01:02:03','ON' ,501,502 UNION
SELECT 'Group3','01-Jan-2019 01:02:03','OFF',601,602;
SELECT * FROM #MyTable;
please suggest the query,
Here is a working dynamic pivot as folks suggested
Example
Declare #UnPiv varchar(max),#SQL varchar(max)
Set #UnPiv = Stuff(( Select ','+quotename(name)
From sys.dm_exec_describe_first_result_set(N'Select top 1 * from #MyTable',null,null ) A
Where Name not in ('Group','DateTime','Status')
For XML Path('')),1,1,'')
Set #SQL = Stuff(( Select ','+quotename(name+Suffix)
From sys.dm_exec_describe_first_result_set(N'Select top 1 * from #MyTable',null,null ) A
Cross Join ( values ('_Status_ON')
,('_Status_OFF')
) B(Suffix)
Where Name not in ('Group','DateTime','Status')
For XML Path('')),1,1,'')
Select #SQL = '
Select *
From (
Select [Group]
,[DateTime]
,Item = concat(Item,''_Status_'',Status)
,Value
From (
Select * From #MyTable Unpivot (Value for Item in ('+#UnPiv+')) UnPiv
) A
) src
Pivot (max(Value) for Item in ('+#SQL+') ) pvt
'
--Print #SQL
Exec(#SQL)
Returns
You could use conditional aggregation:
SELECT [Group], [Datetime],
Col1_Status_ON = MIN(IIF(Status='ON', Col1, NULL)),
Col1_Status_OFF = MIN(IIF(Status='OFF', Col1, NULL)),
Col2_Status_ON = MIN(IIF(Status='ON', Col2, NULL)),
Col2_Status_OFF = MIN(IIF(Status='OFF', Col2, NULL))
FROM #Mytable
GROUP BY [Group], [Datetime];
db<>fiddle demo
Related
I am trying to query a table for data conversion but I am running into issues with the structure. I am guessing I need to Pivot then Unpivot but I am not sure where to begin.
This is my current table
+----+-----------+-------+---------+
| ID | field | value | date |
+----+-----------+-------+---------+
| 1 | Draw1 | 1500 | NULL |
| 1 | Draw1Date | NULL | 4/15/16 |
| 1 | Draw1Fee | 100 | NULL |
| 1 | Draw2 | 2000 | NULL |
| 1 | Draw2Date | NULL | 3/14/17 |
| 1 | Draw2Fee | 100 | NULL |
| 2 | Draw1 | 800 | NULL |
| 2 | Draw1Date | NULL | 4/16/18 |
| 2 | Draw1Fee | 150 | NULL |
| 2 | Draw2 | 760 | NULL |
| 2 | Draw2Date | NULL | 5/6/18 |
| 2 | Draw2Fee | 150 | NULL |
+----+-----------+-------+---------+
Result needed
+----+-------+---------+---------+------+
| ID | Draws | Amount | Date | Fee |
+----+-------+---------+---------+------+
| 1 | Draw1 | 1500 | 4/15/16 | 100 |
| 1 | Draw2 | 2000 | 3/14/17 | 100 |
| 2 | Draw1 | 800 | 4/16/18 | 150 |
| 2 | Draw2 | 760 | 5/6/18 | 150 |
+----+-------+---------+---------+------+
My answer works for provided data. if you are looking for more general solution, for more variety of data, you may try to find another answer. I've not used PIVOT/UNPIVOT.
Test Data:
create table #t (ID int, field varchar(20), [value] int, [date] date)
insert into #t values
(1 ,'Draw1' , 1500 , NULL ),
(1 ,'Draw1Date' , NULL , '4/15/16'),
(1 ,'Draw1Fee' , 100 , NULL ),
(1 ,'Draw2' , 2000 , NULL ),
(1 ,'Draw2Date' , NULL , '3/14/17'),
(1 ,'Draw2Fee' , 100 , NULL ),
(2 ,'Draw1' , 800 , NULL ),
(2 ,'Draw1Date' , NULL , '4/16/18'),
(2 ,'Draw1Fee' , 150 , NULL ),
(2 ,'Draw2' , 760 , NULL ),
(2 ,'Draw2Date' , NULL , '5/6/18' ),
(2 ,'Draw2Fee' , 150 , NULL )
Query:
;with ct as (
select ID, field
from #t
where field in ('Draw1', 'Draw2')
group by ID, field
)
select ct.ID, ct.field
, t1.[value] as Amount, t2.[date] as [Date], t3.[value] as Fee
from ct
inner join #t t1 on t1.ID = ct.ID and t1.field = ct.field
inner join #t t2 on t2.ID = ct.ID and t2.field = ct.field + 'Date'
inner join #t t3 on t3.ID = ct.ID and t3.field = ct.field + 'Fee'
Result:
ID field Amount Date Fee
1 Draw1 1500 2016-04-15 100
1 Draw2 2000 2017-03-14 100
2 Draw1 800 2018-04-16 150
2 Draw2 760 2018-05-06 150
Try this...
SELECT tblAmount.id AS ID,
tblAmount.field AS Draws,
Max(tblAmount.value) AS Amount,
Max(tblDate.[date]) AS [Date],
Max(tblFee.value) AS Fee
FROM tablename tblAmount
INNER JOIN (SELECT id, field, [date]
FROM tablename
WHERE [date] IS NOT NULL AND field LIKE '%Date') tblDate
ON tblAmount.id = tblDate.id
AND tblDate.field LIKE tblAmount.field + '%'
INNER JOIN (SELECT id, field, value
FROM tablename
WHERE value IS NOT NULL AND field LIKE '%Fee') tblFee
ON tblAmount.id = tblFee.id
WHERE tblAmount.value IS NOT NULL
AND tblAmount.field NOT LIKE '%Fee'
AND tblAmount.field NOT LIKE '%Date'
GROUP BY tblAmount.id, tblAmount.field
ORDER BY tblAmount.id, tblAmount.field
Output
+----+-------+--------+------------+-----+
| ID | Draws | Amount | Date | Fee |
+----+-------+--------+------------+-----+
| 1 | Draw1 | 1500 | 2016-04-15 | 100 |
| 1 | Draw2 | 2000 | 2017-03-14 | 100 |
| 2 | Draw1 | 800 | 2018-04-16 | 150 |
| 2 | Draw2 | 760 | 2018-05-06 | 150 |
+----+-------+--------+------------+-----+
Demo: http://www.sqlfiddle.com/#!18/97688/101/0
I would simply do:
select id, left(field, 5),
max(case when len(field) = 5 then value end) as value,
max(case when field like '%date' then value end) as date,
sum(case when field like '%fee' then value end) as fee
from t
group by id, left(field, 5);
If your field is really more complex, do you are looking for date and fee at the end, by anything before, then use cross apply:
select t.id, v.draws,
max(case when t.field = v.draws then value end) as value,
max(case when t.field like '%date' then value end) as date,
sum(case when t.field like '%fee' then value end) as fee
from t cross apply
(values (replace(replace(field, 'date', ''), 'fee', '')) v(draws)
group by id, v.draws;
I don't understand how PIVOT works in SQL. I have 2 tables and I would like to pivot 1 of them in order to get only 1 table with all the data together. I've attached an image with the tables I have and the result that I would like to get.
CREATE TABLE TABLE1
([serie_id] varchar(4), [Maturity] int, [Strategy] int, [Lifetime] varchar(4), [L_max] decimal(10, 5), [W_max] decimal(10, 5), [H_max] decimal(10, 5))
;
INSERT INTO TABLE1
([serie_id], [Maturity], [Strategy], [Lifetime], [L_max], [W_max], [H_max])
VALUES
('id_1', 3, 1, '2', 2.200, 1.400, 1.400),
('id_2', 3, 1, '2', 3.400, 1.800, 2.100),
('id_3', 3, 1, NULL, 24.500, 14.500, 15.000),
('id_4', 3, 1, NULL, 28.000, 24.500, 14.000)
;
CREATE TABLE TABLE2
([serie_id] varchar(4), [L_value] decimal(10, 5), [lrms] decimal(10, 5), [latTmax] decimal(10, 5), [Rdc] decimal(10, 5))
;
INSERT INTO TABLE2
([serie_id], [L_value], [lrms], [latTmax], [Rdc])
VALUES
('id_1', 67.000, 400.000, 400.000, 0.250),
('id_1', 90.000, 330.000, 330.000, 0.350),
('id_1', 120.000, 370.000, 370.000, 0.300),
('id_1', 180.000, 330.000, 300.000, 0.350),
('id_2', 260.000, 300.000, 300.000, 0.400),
('id_2', 360.000, 280.000, 280.000, 0.450),
('id_3', 90.000, 370.000, 370.000, 0.300),
('id_4', 160.000, 340.000, 340.000, 0.400)
;
SQLFiddle
If someone could help me with the SQL query I would appreciate it so much.
In order to get your final result, you are going to have to implement a variety of methods including unpivot, pivot, along with the use of a windowing function like row_number().
Since you have multiple columns in Table2 that need to be pivoted, then you will need to unpivot them first. This is the reverse of pivot, which converts your multiple columns into multiple rows. But before you unpivot, you need some value to identify the values of each row using row_number() - sounds complicated, right?
First, query table2 using the windowing function row_number(). This creates a unique identifier for each row and allows you to easily be able to associate the values for id_1 from any of the others.
select serie_id, l_value, lrms, latTmax, Rdc,
rn = cast(row_number() over(partition by serie_id order by serie_id)
as varchar(10))
from table2;
See Demo. Once you've created this unique identifier, then you will unpivot the L_value, lrms, latTmax, and rdc. You can unpivot the data using several different methods, including the unpivot function, CROSS APPLY, or UNION ALL.
select serie_id,
col, value
from
(
select serie_id, l_value, lrms, latTmax, Rdc,
rn = cast(row_number() over(partition by serie_id order by serie_id)
as varchar(10))
from table2
) d
cross apply
(
select 'L_value_'+rn, L_value union all
select 'lrms_'+rn, lrms union all
select 'latTmax_'+rn, latTmax union all
select 'Rdc_'+rn, Rdc
) c (col, value)
See SQL Fiddle with Demo. The data from table2 is not in a completely different format that can be pivoted into the new columns:
| SERIE_ID | COL | VALUE |
|----------|-----------|-------|
| id_1 | L_value_1 | 67 |
| id_1 | lrms_1 | 400 |
| id_1 | latTmax_1 | 400 |
| id_1 | Rdc_1 | 0.25 |
| id_1 | L_value_2 | 90 |
| id_1 | lrms_2 | 330 |
| id_1 | latTmax_2 | 330 |
| id_1 | Rdc_2 | 0.35 |
The final step would be to PIVOT the data above into the final result:
select serie_id, maturity, strategy, lifetime, l_max, w_max, h_max,
L_value_1, lrms_1, latTmax_1, Rdc_1,
L_value_2, lrms_2, latTmax_2, Rdc_2,
L_value_3, lrms_3, latTmax_3, Rdc_3,
L_value_4, lrms_4, latTmax_4, Rdc_4
from
(
select t1.serie_id, t1.maturity, t1.strategy, t1.lifetime,
t1.l_max, t1.w_max, t1.h_max,
t2.col, t2.value
from table1 t1
inner join
(
select serie_id,
col, value
from
(
select serie_id, l_value, lrms, latTmax, Rdc,
rn = cast(row_number() over(partition by serie_id order by serie_id)
as varchar(10))
from table2
) d
cross apply
(
select 'L_value_'+rn, L_value union all
select 'lrms_'+rn, lrms union all
select 'latTmax_'+rn, latTmax union all
select 'Rdc_'+rn, Rdc
) c (col, value)
) t2
on t1.serie_id = t2.serie_id
) d
pivot
(
max(value)
for col in (L_value_1, lrms_1, latTmax_1, Rdc_1,
L_value_2, lrms_2, latTmax_2, Rdc_2,
L_value_3, lrms_3, latTmax_3, Rdc_3,
L_value_4, lrms_4, latTmax_4, Rdc_4)
) p;
See SQL Fiddle with Demo.
If you have an unknown number of values in Table2 then you will need to use dynamic SQL to create a sql string that will be executed. Converting the above code to dynamic sql is pretty easy once you have the logic correct. The code will be:
DECLARE #cols AS NVARCHAR(MAX),
#query AS NVARCHAR(MAX)
select #cols
= STUFF((SELECT ',' + QUOTENAME(col+cast(rn as varchar(10)))
from
(
select rn = cast(row_number() over(partition by serie_id order by serie_id)
as varchar(10))
from table2
) d
cross apply
(
select 'L_value_', 0 union all
select 'lrms_', 1 union all
select 'latTmax_', 2 union all
select 'Rdc_', 3
) c (col, so)
group by col, rn, so
order by rn, so
FOR XML PATH(''), TYPE
).value('.', 'NVARCHAR(MAX)')
,1,1,'')
set #query = N'SELECT serie_id, maturity, strategy, lifetime, l_max,
w_max, h_max,' + #cols + N'
from
(
select t1.serie_id, t1.maturity, t1.strategy, t1.lifetime,
t1.l_max, t1.w_max, t1.h_max,
t2.col, t2.value
from table1 t1
inner join
(
select serie_id,
col, value
from
(
select serie_id, l_value, lrms, latTmax, Rdc,
rn = cast(row_number() over(partition by serie_id order by serie_id)
as varchar(10))
from table2
) d
cross apply
(
select ''L_value_''+rn, L_value union all
select ''lrms_''+rn, lrms union all
select ''latTmax_''+rn, latTmax union all
select ''Rdc_''+rn, Rdc
) c (col, value)
) t2
on t1.serie_id = t2.serie_id
) x
pivot
(
max(value)
for col in (' + #cols + N')
) p '
exec sp_executesql #query
See SQL Fiddle with Demo
Both versions will give a result of:
| SERIE_ID | MATURITY | STRATEGY | LIFETIME | L_MAX | W_MAX | H_MAX | L_VALUE_1 | LRMS_1 | LATTMAX_1 | RDC_1 | L_VALUE_2 | LRMS_2 | LATTMAX_2 | RDC_2 | L_VALUE_3 | LRMS_3 | LATTMAX_3 | RDC_3 | L_VALUE_4 | LRMS_4 | LATTMAX_4 | RDC_4 |
|----------|----------|----------|----------|-------|-------|-------|-----------|--------|-----------|-------|-----------|--------|-----------|--------|-----------|--------|-----------|--------|-----------|--------|-----------|--------|
| id_1 | 3 | 1 | 2 | 2.2 | 1.4 | 1.4 | 67 | 400 | 400 | 0.25 | 90 | 330 | 330 | 0.35 | 120 | 370 | 370 | 0.3 | 180 | 330 | 300 | 0.35 |
| id_2 | 3 | 1 | 2 | 3.4 | 1.8 | 2.1 | 260 | 300 | 300 | 0.4 | 360 | 280 | 280 | 0.45 | (null) | (null) | (null) | (null) | (null) | (null) | (null) | (null) |
| id_3 | 3 | 1 | (null) | 24.5 | 14.5 | 15 | 90 | 370 | 370 | 0.3 | (null) | (null) | (null) | (null) | (null) | (null) | (null) | (null) | (null) | (null) | (null) | (null) |
| id_4 | 3 | 1 | (null) | 28 | 24.5 | 14 | 160 | 340 | 340 | 0.4 | (null) | (null) | (null) | (null) | (null) | (null) | (null) | (null) | (null) | (null) | (null) | (null) |
Ive an SQL table looks like this
ID | Pattern
---------------------
1 | 2345
2 | 3345
3 | 2345
4 | 5647
5 | 1472
6 | 4817
7 | 5871
8 | 2345
9 | 5871
10 | 1472
Here the pattern is just some random patterns. But I need a kinda grouping based on that pattern. So the output should be something like this
Pattern | ConnectedIDs
-----------------------------
2345 | 1,3,8
3345 | 2
5647 | 4
1472 | 5,10
4817 | 6
5871 | 7,9
-----------------------------------------
Hope you got the scenario..
Try this (Fiddle demo):
SELECT t1.Pattern,
Ids = REPLACE( (SELECT Id AS [data()]
FROM Table1 t2
WHERE t2.Pattern = t1.Pattern
ORDER BY t2.Pattern
FOR XML PATH('')
), ' ', ',')
FROM Table1 t1
GROUP BY Pattern;
Results:
| PATTERN | IDS |
|---------|-------|
| 1472 | 5,10 |
| 2345 | 1,3,8 |
| 3345 | 2 |
| 4817 | 6 |
| 5647 | 4 |
| 5871 | 7,9 |
You could use this scalar valued function:
CREATE FUNCTION [dbo].[GetPatternIDs]
(
#pattern varchar(4),
#delimiter varchar(4)
)
RETURNS VARCHAR(8000)
AS
BEGIN
DECLARE #IDs VARCHAR(8000)
SELECT #IDs = COALESCE(#IDs + #delimiter, '') + CAST(Patterns.ID AS VARCHAR(10))
FROM Patterns
WHERE Pattern=#pattern
ORDER By Patterns.ID
RETURN #IDs
END
and this query:
SELECT Pattern,
ConnectedIDs = dbo.GetPatternIDs(Pattern, ',')
FROM Patterns
GROUP BY Pattern
ORDER BY MIN(ID)
DEMO
I have the following dataset on a sql database
----------------------------------
| ID | NAME | AGE | STATUS |
-----------------------------------
| 1ASDF | Brenda | 21 | Single |
-----------------------------------
| 2FDSH | Ging | 24 | Married|
-----------------------------------
| 3SDFD | Judie | 18 | Widow |
-----------------------------------
| 4GWWX | Sophie | 21 | Married|
-----------------------------------
| 5JDSI | Mylene | 24 | Singe |
-----------------------------------
I want to query that dataset so that i can have this structure in my result
--------------------------------------
| AGE | SINGLE | MARRIED | WIDOW |
--------------------------------------
| 21 | 1 | 1 | 0 |
--------------------------------------
| 24 | 1 | 1 | 0 |
--------------------------------------
| 18 | 0 | 0 | 1 |
--------------------------------------
And the status column can be dynamic so there will be more columns to come.
Is this possible?
Since you are using SQL Server, you can use the PIVOT table operator like this:
SELECT *
FROM
(
SELECT Age, Name, Status FROM tablename
) AS t
PIVOT
(
COUNT(Name)
FOR Status IN(Single, Married, Widow)
) AS p;
SQL Fiddle Demo
To do it dynamically you have to use dynamic sql like this:
DECLARE #cols AS NVARCHAR(MAX);
DECLARE #query AS NVARCHAR(MAX);
select #cols = STUFF((SELECT distinct ',' +
QUOTENAME(status)
FROM tablename
FOR XML PATH(''), TYPE
).value('.', 'NVARCHAR(MAX)')
, 1, 1, '');
SELECT #query = '
SELECT *
FROM
(
SELECT Age, Name, Status FROM tablename
) AS t
PIVOT
(
COUNT(Name)
FOR Status IN( ' +#cols + ')
) AS p;';
execute(#query);
Updated SQL Fiddle Demo
I have a table like the following:
+----+-----+-----+
| ID | GRP | NR |
+----+-----+-----+
| 1 | 1 | 101 |
| 2 | 1 | 102 |
| 3 | 1 | 103 |
| 4 | 1 | 105 |
| 5 | 1-2 | 106 |
| 6 | 1-2 | 109 |
| 7 | 1-2 | 110 |
| 8 | 2 | 201 |
| 9 | 2 | 202 |
| 10 | 3 | 300 |
| 11 | 3 | 350 |
| 12 | 3 | 351 |
| 13 | 3 | 352 |
+----+-----+-----+
I wanted to create a view which groups this list by GRP and concatenates values in NR.
Is it possible to dynamically detect sequences and shorten them into ranges?
Like 1, 2, 3, 5 would become 1-3, 5.
So the result should look like this:
+-----+--------------------+
| GRP | NRS |
+-----+--------------------+
| 1 | 101 - 103, 105 |
| 1-2 | 106, 109 - 110 |
| 2 | 201 - 202 |
| 3 | 300, 350 - 352 |
+-----+--------------------+
What i got now is simply concatenate values, so the table above would become this:
+-----+--------------------+
| GRP | NRS |
+-----+--------------------+
| 1 | 101, 102, 103, 105 |
| 1-2 | 106, 109, 110 |
| 2 | 201, 202 |
| 3 | 300, 350, 351, 352 |
+-----+--------------------+
Here's the actual statement:
DECLARE #T TABLE
(
ID INT IDENTITY(1, 1)
, GRP VARCHAR(10)
, NR INT
)
INSERT INTO #T
VALUES ('1',101),('1',102),('1',103),('1',105)
,('1-2',106),('1-2',109), ('1-2',110)
,('2',201),('2',202)
,('3',300),('3',350),('3',351),('3',352)
SELECT * FROM #T
;WITH GROUPNUMS (RN, GRP, NR, NRS) AS
(
SELECT 1, GRP, MIN(NR), CAST(MIN(NR) AS VARCHAR(MAX))
FROM #T
GROUP BY GRP
UNION ALL
SELECT CT.RN + 1, T.GRP, T.NR, CT.NRS + ', ' + CAST(T.NR AS VARCHAR(MAX))
FROM #T T
INNER JOIN GROUPNUMS CT ON CT.GRP = T.GRP
WHERE T.NR > CT.NR
)
SELECT NRS.GRP, NRS.NRS
FROM GROUPNUMS NRS
INNER JOIN (
SELECT GRP, MAX(RN) AS MRN
FROM GROUPNUMS
GROUP BY GRP
) R
ON NRS.RN = R.MRN AND NRS.GRP = R.GRP
ORDER BY NRS.GRP
Can anyone tell me if it's easily possible to do something like that?
Would be great if anyone has an idea and would like to share it.
SQLFiddle demo
with TRes
as
(
select T.GRP,T.NR NR,
CASE WHEN T1.NR IS NULL and T2.NR is null
THEN CAST(T.NR as VARCHAR(MAX))
WHEN T1.NR IS NULL and T2.NR IS NOT NULL
THEN '-'+CAST(T.NR as VARCHAR(MAX))
WHEN T1.NR IS NOT NULL and T2.NR IS NULL
THEN CAST(T.NR as VARCHAR(MAX))+'-'
END AS NR_GRP
from T
left join T T1 on T.Grp=T1.Grp and t.Nr+1=t1.Nr
left join T T2 on T.Grp=T2.Grp and t.Nr-1=t2.Nr
WHERE T1.NR IS NULL or T2.NR IS NULL
)
SELECT
GRP,
REPLACE(
substring((SELECT ( ',' + NR_GRP)
FROM TRes t2
WHERE t1.GRP = t2.GRP
ORDER BY
GRP,
NR
FOR XML PATH( '' )
), 2, 10000 )
,'-,-','-')
FROM TRes t1
GROUP BY GRP
Please check my try:
DECLARE #T TABLE
(
ID INT IDENTITY(1, 1)
, GRP VARCHAR(10)
, NR INT
)
INSERT INTO #T
VALUES ('1',101),('1',102),('1',103),('1',105)
,('1-2',106),('1-2',109), ('1-2',110)
,('2',201),('2',202)
,('3',300),('3',350),('3',351),('3',352)
SELECT * FROM #T
;WITH T1 as
(
SELECT GRP, NR, ROW_NUMBER() over(order by GRP, NR) ID FROM #T
)
,T as (
SELECT *, 1 CNT FROM T1 where ID=1
union all
SELECT b.*, (case when T.NR+1=b.NR and T.GRP=b.GRP then t.CNT
else T.CNT+1 end)
from T1 b INNER JOIN T on b.ID=T.ID+1
)
, TN as(
select *,
MIN(NR) over(partition by GRP, CNT) MinVal,
MAX(NR) over(partition by GRP, CNT) MaxVal
From T
)
SELECT GRP, STUFF(
(SELECT distinct ','+(CASE WHEN MinVal=MaxVal THEN CAST(MinVal as nvarchar(10)) ELSE CAST(MinVal as nvarchar(10))+'-'+cast(MaxVal as nvarchar(10)) END)
FROM TN b where b.GRP=a.GRP
FOR XML PATH(''),type).value('.','nvarchar(max)'),1,1,'') AS [ACCOUNT NAMES]
FROM TN a GROUP BY GRP