You can use SQL to update, insert data into, and delete data from nonversioned tables in the geodatabase if they do not participate in geodatabase behavior. See What type of data can be edited using SQL? for information on the types of data and geodatabase behavior you cannot edit with SQL.
All data that is registered with the geodatabase has a system-maintained, unique, not-null ObjectID (Row ID) field. When you use SQL to insert records into nonversioned tables in the geodatabase, you must provide a unique value for the ObjectID. Include the Next_RowID function in an insert statement to populate the ObjectID field with the next valid value.
This set of instructions describes updating one row at a time in a nonversioned table and inserting an ObjectID. You would most likely write a routine or client program to retrieve ObjectIDs and update your data.
- Log in to the database from an SQL editor such as psql.
Be sure to log in to the database as a user who has permission to edit the data.
- Use the Next_RowID function to populate the ObjectID field.
In this example, the site_id column is the ObjectID field, the table is named outbreak, and it is stored in the hmoot schema.
INSERT INTO hmoot.outbreak (site_id, med_code,num_affected, shape) VALUES ( sde.next_rowid('hmoot', 'outbreak'), 'v-22a', 3, ST_Point('point (12 36))', 4326 );
- You can continue editing or, if you are done editing, commit your edits to the database.
If the table also contains a GUID or Global ID field, you must provide a unique value for this field as well. See Next_GlobalID for more information.