I have a query in which I check whether an NAME _IDS of type varchar is in another table with NAME _ID column having data type as int.   Also, ID_1 have ID’s separated by comma’s example ‘1,2,3,4,5’.

Below is the query which I am trying to execute

SELECT  Tb2. _ID,Tb1.DET_ID,COUNT(Tb3. _Name) NAME_COUNT

  FROM  TABLE_1 Tb1  

  JOIN  TABLE_2 Tb2 ON Tb2.DET_ID= Tb1.DET_ID

  JOIN  Table_3 Tb3 ON Tb3. _ID= Tb1. _ID

     AND CAST(Tb3.Time AS DATE)=CURRENT_DATE   

     AND Tb3.NAME _IDS IN  (CAST(Tb2.NAME_ID AS VARCHAR(250))) /*Here Tb3.NAME _IDS is varchar & Tb2.NAME _IDS is int so casted into varchar. This line gives wrong output */

  GROUP BY Tb1.DET_ID, Tb1. NAME _ID, Tb1. _ID, Tb3._NAME

NOTE: Tb3.NAME _IDS could have multiple ids in form ‘1,2,23,15’

 Problem I am facing is that when Tb3.NAME _IDS contains multiple IDs in it then result is not shown leading to wrong result. Is there any solution to solve this query. I tried using ‘STARTING WITH’, ‘CONTAINING’,’LIKE’  but no output.

SELECT  Tb2._ID, Tb1.DET_ID, COUNT(Tb3._Name) NAME_COUNT
/* It always confuses me when I see things like COUNT(Tb3._Name), myself I always use either COUNT(*) or COUNT(DISTINCT Tb3._Name), though I think your syntax is correct if you want to do the equivalent of COUNT(*), but skip nulls */
FROM  TABLE_1 Tb1 
JOIN  TABLE_2 Tb2 ON Tb2.DET_ID = Tb1.DET_ID
JOIN  Table_3 Tb3 ON Tb3._ID    = Tb1._ID
  AND CAST(Tb3.Time AS DATE)    = CURRENT_DATE  
  AND ','||Tb2.NAME_ID||',' in ','||Tb3.NAME_IDS||','
/* Tb2 and Tb3 the other way around, and add commas to make sure things matches exactly (i.e. match even if the first or last in NAME_IDS and not match 1 to 11 etc) */
GROUP BY Tb2._ID, Tb1.DET_ID
/*It makes no sense to group by anything but the non-grouped fields that you select, most likely you would get an error with your original GROUP BY*/

HTH,
Set