PostgreSQL: get count of occurrences of array1 values in array2

I have a table:

I need to count the proportion of items from the items_next_order contained in the items_current_order.

For example, for order_id = 4254645, the share should be 50% (or 0.5), since only 2 of the 4 items_next_order items are contained in items_current_order.

I tried to calculate the amount (not the share) like this::

SELECT
    client_id ,
    COUNT(1) AS count_values
FROM
    (SELECT
         client_id ,
         unnest(items_next_order) AS value
     FROM
         orders
     GROUP BY
         1, 2) AS a
WHERE
    value = ANY (SELECT unnest(items_current_order) FROM orders as rj WHERE rj.client_id = a.client_id )
GROUP BY
    client_id 

but I need to count in the grouping by client_id , order_id

The result that I expect:

Here's a working query which accomplishes what you are looking to do:

SELECT 
 order_id, 
 items_current_order,
 items_next_order,
 (
    SELECT COUNT(*) FROM unnest(items_current_order) AS i
   WHERE i = ANY(items_next_order)
 )::FLOAT / array_length(items_current_order,1)::FLOAT AS share
 FROM T

Explanation:

• We are using a subquery combined with the unnest to iterate over the items of items_current_order and count how many of them are contained in items_next_order.
• We then divide that by the length of the items_current_order array to get the ratio - we use the array_length function to count the length of the array.
• In both cases we convert the integer value to a float to get a float response.

You can see a working SQL Fiddle with this query here: https://www.db-fiddle.com/f/5opYDXhZu5zvK8An8YyWRs/0