Count number of rows for multiple tables in one query

Question

so I have a large number of tables which fall under a certain category. These tables all start with B1T(so B1T00001, B1T0002 etc). I've been tasked with providing some information on this set of tables.

Here is my initial SQL to get a count of how many B1T tables there are.

select obj_nm from od_md_obj
where proj_id = '6' and obj_nm like ('%B1T%')and obj_typ_id = '9';

This returns about 260 tables. I am attempting to get a count of the number of rows for all the tables combined. Either the query can break it the count by table, possibly use a rollup for the total count; or by just doing a single count for everything.

I tried looking at some of the solutions in my search but it requires a bit of tedious code considering the number of tables I am querying.

Any help would be greatly appreciated!!

I am using SQL developer currently.

Answer 1

Create a query that will generate a query for you.
For example this query:

select 'SELECT ''' || table_name || ''' as table_name, count(*) As con ' ||
       'FROM ' || table_name || 
        CASE WHEN lead(table_name) OVER (order by table_name ) IS NOT NULL 
        THEN ' UNION ALL ' END
from user_tables 
where table_name like 'B1T%'
;

will generate a result like this:

SELECT 'B1T00000' as table_name, count(*) As con FROM B1T00000 UNION ALL                                                                                                                                                                                                                                                  
SELECT 'B1T00001' as table_name, count(*) As con FROM B1T00001 UNION ALL                                                                                                                                                                                                                                                  
SELECT 'B1T00002' as table_name, count(*) As con FROM B1T00002 UNION ALL                                                                                                                                                                                                                                                  
SELECT 'B1T00003' as table_name, count(*) As con FROM B1T00003 UNION ALL                                                                                                                                                                                                                                                  
SELECT 'B1T00004' as table_name, count(*) As con FROM B1T00004 UNION ALL                                                                                                                                                                                                                                                  
SELECT 'B1T00005' as table_name, count(*) As con FROM B1T00005 UNION ALL                                                                                                                                                                                                                                                  
SELECT 'B1T00006' as table_name, count(*) As con FROM B1T00006 UNION ALL                                                                                                                                                                                                                                                  
SELECT 'B1T00007' as table_name, count(*) As con FROM B1T00007 UNION ALL                                                                                                                                                                                                                                                  
SELECT 'B1T00008' as table_name, count(*) As con FROM B1T00008 UNION ALL  
....
....
....
SELECT 'B1T00257' as table_name, count(*) As con FROM B1T00257 UNION ALL                                                                                                                                                                                                                                                  
SELECT 'B1T00258' as table_name, count(*) As con FROM B1T00258 UNION ALL                                                                                                                                                                                                                                                  
SELECT 'B1T00259' as table_name, count(*) As con FROM B1T00259 UNION ALL                                                                                                                                                                                                                                                  
SELECT 'B1T00260' as table_name, count(*) As con FROM B1T00260

And now just copy this result, paste it into you client (SQL-Developer etc.) and run it.
You will get desired results:

TABLE_NAME        CON
---------- ----------
B1T00000           65 
B1T00001           66 
B1T00002           67 
B1T00003           68 
B1T00004           69
...
...
...
B1T00256          321 
B1T00257          322 
B1T00258          323 
B1T00259          324 
B1T00260          325 

 261 rows selected  

Answer 2

Which version of Oracle? Oracle 12 allows functions in a with clause; so a possible solution is

WITH FUNCTION count_rows(t_name IN VARCHAR2) RETURN NUMBER  
AS  
   l_rows NUMBER;  
BEGIN  
   EXECUTE IMMEDIATE 'SELECT COUNT(*)  FROM ' || t_name INTO l_rows;  
   RETURN l_rows;  
END;  
SELECT table_name, count_rows(table_name) as row_count
   FROM user_tables 
   WHERE table_name LIKE 'B1T%'

In earlier versions, you can do the same with a PL/SQL function that you must compile first, like this:

CREATE OR REPLACE FUNCTION count_rows(t_name IN VARCHAR2) RETURN NUMBER  
AS  
  l_rows NUMBER;  
BEGIN  
  EXECUTE IMMEDIATE 'SELECT COUNT(*)  FROM ' || t_name INTO l_rows;  
  RETURN l_rows;  
END;  
/ 

Compile it, then use it in a SELECT statement identical to the one above.

If you only need the grand total, you can use

SELECT SUM(row_count) FROM (SELECT .....)

where the subquery is the query shown in the first code snippet.