Amazon Redshift query console
Run Redshift SQL with named parameters, an EXPLAIN plan tree, a read-only transaction for reads, query telemetry and session control.
The Redshift console highlights SQL in the Redshift dialect and shows the result in the same grid as the other engines. You can use named parameters written as :name. Tablyne asks for the values before it runs the statement.
One statement at a time
Each run sends a single statement, up to 100 KiB of SQL. If the text could be read in two ways by the server, for example a backslash inside a string or a nested comment, Tablyne asks you to split the commands. A connection can have two runs in flight. A third waits until one finishes or is cancelled. A read stops after 60 seconds unless you set another timeout, between one second and 24 hours.
Plans
EXPLAIN shows the Redshift plan as a tree. If the text is not a plan Tablyne recognizes, it falls back to plain text.
Query activity and sessions
After a run, you can load the queue time, execution time, cache hit and WLM class of that query. A server history view reads SYS_QUERY_HISTORY only when you click, because that query runs on your cluster and can cost money. The Sessions view lists open sessions and can end one with pg_terminate_backend after confirmation.
When the outcome is unknown
On the Data API, a lost response can leave a statement in doubt. Tablyne keeps a record and lets you reconcile it with the same token before you run it again. Abandoning the record warns that the statement may already have been applied.
Limits
The console shows a banner that manual transactions are unavailable with the transport in use. The Data API also rejects NULL as a parameter value; write NULL in the SQL itself or use the direct connection.