This page provides information about normal views in the database.
Most of the usual columns are absent - all the queries and I/O columns. This is because although I know which tables are in the view, I do not know which queries went to the view. I only know which tables queries go to. As such I can’t tell the difference between a query issued on the view, and a query which just happens to be issued on one of the tables used by the view. So I could produce the queries and I/O columns, for the tables which are used by the view, but I have no way of differentiating between the queries and I/O which is actually for the view, and that which is actually for tables in the view. The inability to differentiate makes the data worthless, so I’ve not produced the data.
Note the creation and age columns are missing, as Redshift does not store the creation time for a view.
Name | Type |
---|---|
schema_id | int8 |
schema | varchar |
view_id | int8 |
view | varchar |
owner_user_id | int4 |
owner | varchar |
length | int4 |
counts:columns | int8 |
counts:tables | int8 |
store:blocks:sorted | int8 |
store:blocks:unsorted | int8 |
store:blocks:total | int8 |
store:rows:sorted | int8 |
store:rows:unsorted | int8 |
store:rows:total | int8 |
related pages | varchar |
The schema ID. This column is emitted in CSV exports only.
The schema name.
The view ID. This column is emitted in CSV exports only.
Surprisingly, view IDs turn out to be unique across all databases.
The view name.
The owner user ID. This column is emitted in CSV exports only.
The owner user name.
The length of the view text, in bytes.
The number of columns. Tables have system columns (there are three), and these are not included in the count.
The number of tables the view uses. Views can use other views; this count is the number of tables from all the views which are used.
The number of sorted blocks.
The number of unsorted blocks.
The total number of blocks.
The number of sorted rows.
The number of unsorted rows.
The total number of rows.