UPDATE statement updates an arbitrary number of rows in a Kudu
table. This statement only works for Impala tables that use the Kudu storage engine.
UPDATE [database_name.]table_name SET col = val [, col = val ... ] [ FROM joined_table_refs ] [ WHERE where_conditions ]
None of the columns that make up the primary key can be updated by the
The conditions in the
WHERE clause are the same ones allowed for the
WHERE clause is omitted, all rows in the table are updated.
The conditions in the
WHERE clause can refer to
any combination of primary key columns or other columns. Referring to
primary key columns in the
WHERE clause is more efficient
than referring to non-primary key columns.
Because Kudu currently does not enforce strong consistency during concurrent DML operations, be aware that the results after this statement finishes might be different than you intuitively expect:
If some rows cannot be updated because their some primary key columns are not found, due to their being deleted by a concurrent
DELETEoperation, the statement succeeds but returns a warning.
UPDATEstatement might also overlap with
UPSERTstatements running concurrently on the same table. After the statement finishes, there might be more or fewer matching rows than expected in the table because it is undefined whether the
UPDATEapplies to rows that are inserted or updated while the
UPDATEis in progress.
The number of affected rows is reported in an impala-shell message and in the query profile.
FROM clause lets you restrict the
updates to only the rows in the specified table that are part
of the result set for a join query. The join clauses can include
non-Kudu tables, but the table from which the rows are deleted
must be a Kudu table.
Statement type: DML
The following examples show how to perform a simple update
on a table, with or without a
-- Set all rows to the same value for column c3. -- In this case, c1 and c2 are primary key columns -- and so cannot be updated. UPDATE kudu_table SET c3 = 'not applicable'; -- Update only the rows that match the condition. UPDATE kudu_table SET c3 = NULL WHERE c1 > 100 AND c3 IS NULL; -- Does not update any rows, because the WHERE condition is always false. UPDATE kudu_table SET c3 = 'impossible' WHERE 1 = 0; -- Change the values of multiple columns in a single UPDATE statement. UPDATE kudu_table SET c3 = upper(c3), c4 = FALSE, c5 = 0 WHERE c6 = TRUE;
The following examples show how to perform an update using the
FROM keyword with a join clause:
-- Uppercase a column value, only for rows that have -- an ID that matches the value from another table. UPDATE kudu_table SET c3 = upper(c3) FROM kudu_table JOIN non_kudu_table ON kudu_table.id = non_kudu_table.id; -- Same effect as previous statement. -- Assign table aliases in FROM clause, then refer to -- short names elsewhere in the statement. UPDATE t1 SET c3 = upper(c3) FROM kudu_table t1 JOIN non_kudu_table t2 ON t1.id = t2.id; -- Same effect as previous statements, but more efficient. -- Use WHERE clause to skip updating values that are -- already uppercase. UPDATE t1 SET c3 = upper(c3) FROM kudu_table t1 JOIN non_kudu_table t2 ON t1.id = t2.id WHERE c3 != upper(c3);