3.4. Columns of Analytical Tables #
For more information about analytical tables, refer to Section 3.3.
3.4.1. Creating a Column #
Execute the following query:
SELECT metastore.add_column('schema_name.table_name' or 'table_name', 'column_name', 'data_type');
Where:
schema_name: The name of the analytical schema where the table is created.Optional parameter. If you do not specify it, the current schema is used.
table_name: The name of the analytical table where the column will be created.column_name: The unique column name.data_type: The data type of the column.For more information about supported data types, refer to Appendix E.
Example 3.23.
SELECT metastore.add_column('testtable', 'third_col', 'uint8');
Once the query is executed, pgpro_metastore performs the following actions:
Verifies user privileges.
Creates a new column in
pga_columnmetadata table.Updates the
view_sqlvalue for the analytical table.
3.4.2. Renaming a Column #
Execute the following query:
SELECT metastore.rename_column('schema_name.table_name' or 'table_name', 'current_column_name', 'new_column_name');
Where:
schema_name: The name of the analytical schema where the table is created.Optional parameter. If you do not specify it, the current schema is used.
table_name: The name of the analytical table where the column is created.current_column_name: The current column name.new_column_name: The new unique column name.
Example 3.24.
SELECT metastore.rename_column('testtable', 'first_col', 'eleventh_col');
Once the query is executed, pgpro_metastore performs the following actions:
Verifies user privileges.
Updates the
column_namevalue for the column in thepga_columnmetadata table.Updates the
view_sqlvalue for the analytical table.
3.4.3. Changing the Data Type of a Column #
Execute the following query:
SELECT metastore.change_column_type('schema_name.table_name' or 'table_name', 'column_name', 'new_data_type');
Where:
schema_name: The name of the analytical schema where the table is created.Optional parameter. If you do not specify it, the current schema is used.
table_name: The name of the analytical table where the column is created.column_name: The name of the column whose data type will be changed.data_type: The new data type of the column.For more information about supported data types, refer to Appendix E.
Example 3.25.
SELECT metastore.change_column_type('testtable', 'third_col', 'uint8');
Once the query is executed, pgpro_metastore performs the following actions:
Verifies user privileges.
Ensures that the new column data type does not match the previous one.
Ensures that the analytical table is empty or that the new data type is compatible with the previous one:
int8can be changed toint16,int32, orint64int16can be changed toint32orint64int32can be changed toint64uint8can be changed touint16,uint32, oruint64uint16can be changed touint32oruint64uint32can be changed touint64float32can be changed tofloat64
Updates the
column_typevalue for the column in thepga_columnmetadata table.Updates the
view_sqlvalue for the analytical table.
3.4.4. Deleting a Column #
Execute the following query:
SELECT metastore.delete_column('schema_name.table_name' or 'table_name', 'column_name');
Where:
schema_name: The name of the analytical schema where the table is created.Optional parameter. If you do not specify it, the current schema is used.
table_name: The name of the analytical table where the column is created.column_name: The name of the column that will be deleted.
Example 3.26.
SELECT metastore.delete_column('my_schema.testtable', 'second_col');
Once the query is executed, pgpro_metastore performs the following actions:
Verifies user privileges.
Creates a new snapshot in the
pga_snapshotmetadata table and specifies for the column theend_snapshotvalue in thepga_columnmetadata table.Updates the
view_sqlvalue for the analytical table.