Find table that DON'T have a column with specific name in Redshift

Databases often have standard columns. Examples of such standard columns can be id, modified_date, created_by or row_version.

Query below finds all tables that do not have a 'id' column.

Query

select t.table_schema,
       t.table_name
from information_schema.tables t
left join (select table_schema, table_name
           from information_schema.columns
           where column_name = 'id') c
      on c.table_name = t.table_name
      and c.table_schema = t.table_schema
where c.table_name is null
      and t.table_schema not in ('information_schema', 'pg_catalog')
      and t.table_type = 'BASE TABLE'
order by t.table_schema,
         t.table_name;

Columns

  • table_schema - name of schema of found table
  • table_name - name of found table

Rows

  • One row represents a table
  • Scope of rows: all found tables
  • Ordered by schema name, table name

Sample results

These tables do not have a 'id' column.