SQL Loop to increment numbers - sql

Question

I'm struggling coming up with a way to solve this answer. I want to start at a specific value and keep increasing it by 1 every time a new line.
For example, if I have a table like so.
90
93
110
87
130
Etc..
I want to select the number 87 and then keep incrementing up from there but also read if the incremented number is there and skip it.
I am just struggling with trying to put the right logic together in my head. I know I need a while loop to keep reading through the table but I can't think of the proper way to go about it. Just looking for some suggestions to push me in the right direction.
Edit: I am using T-SQL for MSFT SQL Server 2012.
Here is an example of what the output should look like
90
93
110
87
130
88
89
91
92
94
It would skip over adding 90 and 93 because they already exist in the table.
I hope that makes sense to you guys.

Answer 1

I do it all in one recursive CTE and I make it so you can use order by and guarantee your results are returned in the correct order.
For the recursion, you can either choose and start and end number or #desiredNumberOfNewValues(keep in mind, it doesn't account for repeats). Let me know if you have any questions or need anything else.
DECLARE #yourTable TABLE (nums INT);
INSERT INTO #yourTable
VALUES (90),(93),(110),(87),(130);
DECLARE #Specific_Number INT = 87;
DECLARE #Last_Number INT = 94;
DECLARE #DesiredNumberOfNewValues INT = 7;
WITH CTE_Numbers
AS
(
SELECT 1 AS order_id,nums, 1 AS cnt
FROM #yourTable
UNION ALL
SELECT 2,
CASE
WHEN #Specific_Number + cnt NOT IN (SELECT * FROM #yourTable) --if it's not already in the table, return it
THEN #Specific_Number + cnt
ELSE NULL -- if it is in the table, return NULL
END,
cnt + 1
FROM CTE_Numbers
WHERE nums = #Specific_Number
--OR (cnt > 1 AND #Specific_Number + cnt < #Last_Number) --beginning and end(option 1)
OR (cnt > 1 AND cnt <= #DesiredNumberOfNewValues) --number of new values(option 2)
)
SELECT order_id,nums
FROM CTE_Numbers
WHERE nums IS NOT NULL
ORDER BY order_id,nums
Results:
order_id nums
----------- -----------
1 87
1 90
1 93
1 110
1 130
2 88
2 89
2 91
2 92
2 94

Answer 2

You might be able to do this with some modifications by using Numbers table. In SQL Server, you could do the following. Assuming Test has the data. To create Numbers table, please refer to the following link.
http://dataeducation.com/you-require-a-numbers-table/
SELECT n.Number,t.ColB
FROM Numbers n
LEFT JOIN Test t
ON n.Number = t.ColA
Where n.Number < (SELECT MAX(COLA) FROM Test t)

Answer 3

WITH Numbers_CTE (Number)
AS
(
SELECT 1 AS Number
UNION ALL
SELECT Number+1 FROM Numbers_CTE
)
SELECT top 10 *
FROM Numbers_CTE n
LEFT JOIN <TABLE> t ON t.ID=n.Number
WHERE t.ID IS NULL
AND n.Number BETWEEN 5 AND 100

Answer 4

There is a system table you can use for numbers called master..spt_values.
DECLARE sample TABLE
DECLARE #tbl TABLE(Id INT)
INSERT statement
INSERT INTO #tbl VALUES (90)
,(93)
,(110)
,(87)
,(130)
You can easily make a UNION to remove duplicates value you have in your table and ORDER it within a CTE
DECLARE #number INT = 87
;WITH C AS(
SELECT Id, Row_Id, ROW_NUMBER() OVER(PARTITION BY Id ORDER BY Id) AS Rn FROM (
SELECT 1 AS Row_Id, Id FROM #tbl
UNION
SELECT 2 AS Row_Id, number
FROM master..spt_values
WHERE [type] = 'P'
AND number >= #number
) t
)
SELECT Id FROM C
WHERE Rn = 1
ORDER BY Row_Id, Id
Output
Id
----
87
90
93
110
130
88
89
91
92
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
111
112
113
114
.
.
.
SQLFiddle