Sometimes you want to count data in all tables on a database to get a feel for the orders of magnitude, but you cannot use the approximated example in How to speed up Count(*) in Interbase/Firebird – Stack Overflow as those require primary keys.
Then the below script can help: it generates the right queries as a union all ordering by the count in the tables.
Example generated code on C:\Program Files (x86)\Firebird\Firebird_2_5\examples\empbuild\EMPLOYEE.FDB:
SQL
--------------------------------------------------------------------------------------------------
select 128 as id, 'COUNTRY' as name, count(*) from "COUNTRY" union all
select 129 as id, 'JOB' as name, count(*) from "JOB" union all
select 130 as id, 'DEPARTMENT' as name, count(*) from "DEPARTMENT" union all
select 131 as id, 'EMPLOYEE' as name, count(*) from "EMPLOYEE" union all
select 133 as id, 'PROJECT' as name, count(*) from "PROJECT" union all
select 134 as id, 'EMPLOYEE_PROJECT' as name, count(*) from "EMPLOYEE_PROJECT" union all
select 135 as id, 'PROJ_DEPT_BUDGET' as name, count(*) from "PROJ_DEPT_BUDGET" union all
select 136 as id, 'SALARY_HISTORY' as name, count(*) from "SALARY_HISTORY" union all
select 137 as id, 'CUSTOMER' as name, count(*) from "CUSTOMER" union all
select 138 as id, 'SALES' as name, count(*) from "SALES" order by 3
Example output on C:\Program Files (x86)\Firebird\Firebird_2_5\examples\empbuild\EMPLOYEE.FDB:
ID NAME COUNT 133 PROJECT 6 128 COUNTRY 14 137 CUSTOMER 15 130 DEPARTMENT 21 135 PROJ_DEPT_BUDGET 24 134 EMPLOYEE_PROJECT 28 129 JOB 31 138 SALES 33 131 EMPLOYEE 42 136 SALARY_HISTORY 49
The generation code below uses a few tricks:
- [WayBack] How to get record count for all tables? and [WayBack] How to get a list of tables, views and columns in Firebird database?
Here’s query to list all user tables:
select rdb$relation_name
from rdb$relations
where rdb$view_blr is null
and (rdb$system_flag is null or rdb$system_flag = 0); - [WayBack] How to get record number or rank as part of dataset? (note that
IDthe negation ofSCORE)
The RANK of a row is simply the COUNT() + 1 of better-ranked rows. This can be computed with a
self-join:
SELECT p.name, p.score, COUNT(others.score) + 1 AS "rank"
FROM players p
LEFT JOIN players others ON others.score > p.score
GROUP BY 1, 2
ORDER BY "rank"
The rank helps me distinguish the last row (for the order by 3 clause) and other rows (for the union all clauses).
Generation code:
with tables(id, name) as (
-- http://www.firebirdfaq.org/faq376/
select r.RDB$RELATION_ID as id, trim(r.RDB$RELATION_NAME) as name
from RDB$RELATIONS r
where 1=1
and (r.RDB$SYSTEM_FLAG is null or r.RDB$SYSTEM_FLAG = 0)
and r.RDB$VIEW_BLR is null
order by 1
),
ranked_tables(id, rank, name) as (
-- http://www.firebirdfaq.org/faq343/
select tables.ID, count(others.id)+1 as "rank", tables.NAME
from tables
left join tables others on others.ID < tables.ID
group by "ID", "NAME"
order by "rank"
),
parts(id, rank, name, suffix) as (
select ranked_tables.id,
ranked_tables.rank,
ranked_tables.name,
case
when ranked_tables.rank = 1 then 'union all' -- first
when ranked_tables.rank = (select count(*) from tables) then 'order by 3' --last
else 'union all' -- middle
end as suffix
from ranked_tables
)
select -- parts.id, parts.rank, parts.name, parts.suffix,
'select '||parts.id||' as id, '''||parts.name||''' as name, count(*) from "'||parts.name||'" '||parts.suffix||'' as SQL
from parts
order by parts.id
–jeroen
The Wiert Corner - irregular stream of stuff 