如何在sql中的ARRAY中获取最频繁的值？

Question

I´ma newbie in SQL.

I have built an array where I have been storing the values ​​that I have been collecting in a loop and dividing them by other value.

The array has the correct values.

Now I would need to get the most repeated value in this array.

Is this possible with SQL? My environment is Oracle 11.

This is the array:

type array_type is varray(100) of NUMBER(10);
loop_results array_type := array_type();

The values:

 1: 906450
 2: 906450
 3: 306449
 4: 906446
 5: 306450
 6: 906447
 7: 306449
 8: 306448
 9: 306448
10: 306450

Thanks for your time.

Answer 1

First of all arrays are a part of pl/sql. Likewise try the below. The below would be somewhat easy using a temporary table.

  Create table sample(value1 
   number(10));
  Declare
  type array_type is varray(100) of 
  NUMBER(10);
  loop_results array_type := array_type();
   Max varchar2(20);

    begin
     for i in 1..loop_results.length
     loop
     Insert into sample values
     (loop_results(i)) ;

     End loop
     Select value1 into max from(Select 
      Value1, 
      count(*) 
       from sample order by count(*) desc
      Group by value1) where rownum=1;

   dbms_output.put_line(max) ;

     End

Answer 2

You can do it in SQL if the type is declared in the SQL scope:

CREATE TYPE array_type is varray(100) of NUMBER(10);

Then (Oracle 12+ solution)

SELECT COLUMN_VALUE,
       COUNT(*)
FROM   TABLE(
         array_type(
           906450,
           906450,
           306449,
           906446,
           306450,
           906447,
           306449,
           306448,
           306448,
           306450
         )
       )
GROUP BY COLUMN_VALUE
ORDER BY COUNT(*) DESC
FETCH FIRST 1 ROW WITH TIES;

or (for earlier versions, you can use analytic functions):

SELECT value,
       cnt
FROM   (
  SELECT COLUMN_VALUE AS value,
         COUNT(*) As cnt,
         DENSE_RANK() OVER ( ORDER BY COUNT(*) DESC ) AS rnk
  FROM   TABLE(
           array_type(
             906450,
             906450,
             306449,
             906446,
             306450,
             906447,
             306449,
             306448,
             306448,
             306450
           )
         )
  GROUP BY COLUMN_VALUE
)
WHERE  rnk = 1;

Outputs all the values that are tied for the highest frequency:

\nCOLUMN_VALUE |  COUNT(*)\n-----------: |  -------: \n      306449 |  2 \n      906450 |  2 \n      306450 |  2 \n      306448 |  2 \n

db<>fiddle Oracle 18 or Oracle 11