在T-Sql的公用表表达式中使用IF ELSE语句

Question

I have a CTE created and I am trying to delete tables using the following. Note that CTE expressions work fine but IF ELSE statements does not.

IF 
with FinalSales (time, terminal_id, count) as
(
    select time, terminal_id, count(*)
    from Final_Sales
    group by time, terminal_id
    having count(*) = 1             -- condition here
)

delete Sales
from FinalSales
inner join Sales
    on Sales.[time] = FinalSales.[time] 
    and Sales.terminal_id = FinalSales.terminal_id

ELSE IF 

  with FinalSales (time, terminal_id, count) as
  (
    select time, terminal_id, count(*)
    from Final_Sales
    group by time, terminal_id
    having count(*) > 1              -- Condition here
  )
  delete Sales
  from FinalSales
  inner join Sales
      on Sales.[time] = FinalSales.[time] 
      and Sales.terminal_id = FinalSales.terminal_id

   delete #temp1
   from FinalSales
   inner join #temp1
        on #temp1.[time] = FinalSales.[time] 
        and #temp1.terminal_id = FinalSales.terminal_id

In the first IF statement, when count(*) = 1 then I delete rows in one table, and then in ELSEIF statement when count > 1 I want to delete rows in two tables. However, my approach does not work.

How can I wrap it into a IF ELSE statement? OR is there any other ways to do this?

Whole query in simple structure,

if count(*) == 1:
   delete Sales
else if count(*) > 1:
   delete Sales, #temp1

something like this. How can I do it in t-sql?

Any ideas would be appreciated.

Answer 1

I think this might work:

declare @count_of_duplicates int

set @count_of_duplicates = 

(select max(duplicate_count)
from
(       select time, terminal_id, count(*) as duplicate_count
        from Final_Sales
        group by time, terminal_id)  t1)

We'll first set the parameter equal to the greatest value of count(*) . If there is a duplicate it will be a value greater than 1, else if there is no duplicate it will be 1.

We can then proceed to start our condition:

if @count_of_duplicates = 1

begin
with FinalSales (time, terminal_id, count) as
(
    select time, terminal_id, count(*)
    from Final_Sales
    group by time, terminal_id
)

delete Sales
from FinalSales
inner join Sales
    on Sales.[time] = FinalSales.[time] 
    and Sales.terminal_id = FinalSales.terminal_id
end

else if @count_of_duplicates > 1
begin
  with FinalSales (time, terminal_id, count) as
  (
    select time, terminal_id, count(*)
    from Final_Sales
    group by time, terminal_id
  )
  delete Sales
  from FinalSales
  inner join Sales
      on Sales.[time] = FinalSales.[time] 
      and Sales.terminal_id = FinalSales.terminal_id

   delete #temp1
   from FinalSales
   inner join #temp1
        on #temp1.[time] = FinalSales.[time] 
        and #temp1.terminal_id = FinalSales.terminal_id
end

I haven't had the chance to parse it... let me know if it works or not.

Answer 2

Why not use if exists(query)?

IF exists (
    select time, terminal_id 
    from Final_Sales 
    group by time, terminal_id 
    having count(*) = 1
) begin
    --query
end else begin
    --query
end