Amazon Redshift will no longer support the creation of new Python UDFs starting November 1, 2025.
If you would like to use Python UDFs, create the UDFs prior to that date.
Existing Python UDFs will continue to function as normal. For more information, see the
blog post
SVV_DATASHARE_CONSUMERS
Use SVV_DATASHARE_CONSUMERS to view a list of consumers for a datashare created on a cluster.
SVV_DATASHARE_CONSUMERS is visible to the following users:
Superusers
Datashare owners
Users with ALTER or USAGE permissions on a datashare
Other users can't see any rows. For information on the ALTER and USAGE permissions, see GRANT.
Table columns
Column name | Data type | Description |
---|---|---|
share_name | varchar(128) | The name of the datashare. |
consumer_account | varchar(16) | The account ID for the datashare consumer. |
consumer_namespace | varchar(64) | The unique cluster identifier of the datashare consumer cluster. |
share_date | timestamp without time zone | The date that the datashare was shared. |
Sample query
The following example returns the output for SVV_DATASHARE_CONSUMERS.
SELECT count(*) FROM svv_datashare_consumers WHERE share_name LIKE 'tickit_datashare%'; 1