Store (In Database)
(In-Database Processing)
Synopsis
Stores the input data in the database.Description
This operator writes the input data into the database using a CREATE TABLE AS SELECT (CTAS) statement. Proper CREATE TABLE privileges are required in the selected schema. No data is retrieved to the memory of the machine running this process unless the output of this operator is connected to a Nest output. This enables ETL processes to run fully in-database, avoiding memory issues. Custom storage parameters are only available for Impala connections.
Input
example set input
Output
- example set output
Parameters
- schema name Name of the schema.
- table name Output table.
- overwrite mode Indicates if an existing table should be overwritten or data should be appended.
- storage format Storage format for new tables. DEFAULT uses the database default.
- primary key Columns to use as primary keys for Kudu tables. Boolean, float and double Kudu types are not allowed.
- non unique primary key If enabled, creates a NON UNIQUE PRIMARY KEY for Kudu, adding an 'auto_incrementing_id' (bigint) column for uniqueness.
- customize storage Enable to specify custom storage format and partitioning options.
- external If enabled, creates an EXTERNAL table, which does not delete underlying data when dropped. For non-Kudu storage format only.
- partition by Exact PARTITION BY clause for the table. Square brackets around attribute names will be replaced by provider-specific identifier quotes.
- location HDFS location where table data will be stored. only for non-Kudu tables.
- row format ROW FORMAT clause for the table (e.g., 'DELIMITED FIELDS TERMINATED BY ,'). Only for non-Kudu tables.
- serde properties SerDe properties (key-value pairs) for the SERDEPROPERTIES clause. Only for non-Kudu tables.
- table properties Table properties (key-value pairs) for the TBLPROPERTIES clause.
- table comment Table comment.