Deleting records from a file

To remove an existing record from an indexed or relative file, open the file as I-O and use the DELETE statement. You cannot use DELETE for a sequential or line-sequential file.

About this task

If ACCESS IS SEQUENTIAL, the record to be deleted must first be read by the COBOL program. The DELETE statement removes the record that was just read. If the DELETE statement is not preceded by a successful READ, the record is not deleted, and the file status key is set to 92.

If ACCESS IS RANDOM or ACCESS IS DYNAMIC, the record to be deleted need not be read by the COBOL program. To delete a record, move the key of the record to the RECORD KEY data item, and then issue the DELETE.

Check the file status key after each DELETE statement.

Related concepts  
File organization and access mode

Related references  
FILE STATUS clause (COBOL for Linux® on x86 Language Reference)