Using IF ELSE statement based on Count to execute different Insert statements

Question

While I am searching through my database, I run an INSERT statement if I find that a particular item does not exist, and I run a different INSERT statement if I find one or

Answer 1

If this is in SQL Server, your syntax is correct; however, you need to reference the COUNT(*) as the Total Count from your nested query. This should give you what you need:

SELECT CASE WHEN TotalCount >0 THEN 'TRUE' ELSE 'FALSE' END FROM
(
  SELECT [Some Column], COUNT(*) TotalCount
  FROM INCIDENTS
  WHERE [Some Column] = 'Target Data'
  GROUP BY [Some Column]
) DerivedTable

Using this, you could assign TotalCount to a variable and then use an IF ELSE statement to execute your INSERT statements:

DECLARE @TotalCount int
SELECT @TotalCount = TotalCount FROM
(
  SELECT [Some Column], COUNT(*) TotalCount
  FROM INCIDENTS
  WHERE [Some Column] = 'Target Data'
  GROUP BY [Some Column]
) DerivedTable
IF @TotalCount > 0
    -- INSERT STATEMENT 1 GOES HERE
ELSE
    -- INSERT STATEMENT 2 GOES HERE

Answer 2

one obvious solution is to run 2 separate queries, first select all items that have count=1 and run your insert, then select the items with count>1 and run the second insert.

as a second step if the two inserts are similar you can probably combine them into one query.

another possibility is to use a cursor to loop thru your recordset and do whatever logic you need for each line.

Answer 3

Not very clear what you mean by

"I cant find any examples to help me understand how I can use this to run 2 different statements:"

. Is it using CASE like a SWITCH you are after?

select case when totalCount >= 0 and totalCount < 11 then '0-10'
            when tatalCount > 10 and totalCount < 101 then '10-100'
            else '>100' end as newColumn
from (
  SELECT [Some Column], COUNT(*) TotalCount
  FROM INCIDENTS
  WHERE [Some Column] = 'Target Data'
  GROUP BY [Some Column]
) A

Answer 4

There are many, many ways to code this, but here is one possible way. I'm assuming MS SQL

We'll start by getting row count (Another Quick Example) and then do if/else

-- Let's get our row count and assign it to a var that will be used
--    in our if stmt 
DECLARE @HasExistingRows int -- I'm assuming it can fit into an int
SELECT @HasExistingRows = Count(*) 
   ELSE 0 -- false
FROM
   INCIDENTS
WHERE {Your Criteria}
GROUP BY {Required Grouping}

Now we can do the If / Else Logic MSDN Docs

-- IF / Else / Begin / END Syntax
IF @HasExistingRows = 0 -- No Existing Rows
   BEGIN
      {Insert Logic for No Existing Rows}
   END
ELSE -- existing rows are found
   BEGIN
      {Insert logic for existing rows}
   END

Another faster way (inspired by Mahmoud Gamal's comment):

Forget the whole variable creation / assignment - look up "EXISTS" - MSDN Docs 2.

IF EXISTS ({SELECT Query})
   BEGIN
      {INSERT Version 1}
   END
ELSE
   BEGIN
      {INSERT version 2}
   END

Answer 5

Depending on your needs, here are a couple of ways:

IF EXISTS (SELECT * FROM TABLE WHERE COLUMN = 'SOME VALUE')
    --INSERT SOMETHING
ELSE
    --INSERT SOMETHING ELSE

Or a bit longer

DECLARE @retVal int

SELECT @retVal = COUNT(*) 
FROM TABLE
WHERE COLUMN = 'Some Value'

IF (@retVal > 0)
BEGIN
    --INSERT SOMETHING
END
ELSE
BEGIN
    --INSERT SOMETHING ELSE
END 

Answer 6

As long as you need to find it based on Count just more than 0, it is better to use EXISTS like this:

IF EXISTS (SELECT 1 FROM INCIDENTS  WHERE [Some Column] = 'Target Data')
BEGIN
    -- TRUE Procedure
END
ELSE BEGIN
    -- FALSE Procedure
END