Skip to content

ABAP Keyword Documentation →  ABAP - Reference →  Processing External Data →  ABAP - Database Accesses →  Open SQL →  Open SQL - Read Accesses →  OPEN CURSOR 


Short Reference

Other versions: 7.31 | 7.40 | 7.54




This statement extracts the requested rows (using the addition INTO or APPENDING) from the results set of the database cursor (associated with the cursor variable dbcur) from the current cursor position and assigns these rows to the data objects specified in the results set.

The cursor variable dbcur must be a variable declared by the special predefined data type cursor, which was opened with the statement OPEN CURSOR, or to which an opened cursor was assigned. Otherwise, a handleable exception occurs.

The syntax and meaning of the addition INTO or APPENDING target are completely synonymous with the identical additions of the SELECT statement, with the exception that no LOB handles can currently be created.

If you specify non-table-like data objects after INTO, then one row is extracted. If an internal table is specified after INTO or APPENDING, then either all rows are extracted, or as many as specified in the addition PACKAGE SIZE.

The statement FETCH moves the position of the database cursor (which is associated with dbcur) by the amount of extracted lines to the next line to be extracted. If the last row of the results set was extracted in a FETCH statement, each subsequent FETCH statement in which
dbcur is associated with the same database cursor sets sy-subrc to 4, without affecting the data objects specified after INTO or APPENDING.

System Fields

sy-subrc Meaning
0 At least one row was extracted from the results set.
4 No row was extracted.

After every row extraction, the statement FETCH sets sy-dbcnt to the amount of rows lines extracted so far from the relevant results set. If an overflow occurs because the number or rows is greater than 2,147,483,647, sy-dbcnt is set to -1. If no row can be extracted, then sy-dbcnt is set to 0.


  • Consecutive FETCH statements that access the same results set can have the different additions INTO or APPENDING: You can specify work areas together with internal tables. In doing so, the addition CORRESPONDING FIELDS is either not listed at all in any of the FETCH statements involved, or has to be listed in every statement. Moreover, the data types of all work areas wa involved or the row types of the internal tables itab must be identical. A bracketed list of data objects after INTO cannot be specified together with work areas or internal tables, but every involved FETCH statement must contain a list of this type.
  • It depends on the database system whether the database cursor in the database is closed implicitly after the extraction of the final row of the results set or not. For this reason, it is always better to use the CLOSE CURSOR statement explicitly.


Reading of data from the database table SPFLI in packets of varying size using two parallel cursors. The packet size is determined by the first cursor using the aggregation functioncount( * ) and using the second cursor for access. Variable control of the addition PACKAGE SIZE is not possible within a single SELECT statement.

DATA: BEGIN OF count_line, 
        carrid TYPE spfli-carrid, 
        count  TYPE i, 
      END OF count_line, 
      spfli_tab TYPE TABLE OF spfli. 

DATA: dbcur1 TYPE cursor, 
      dbcur2 TYPE cursor. 

  SELECT carrid count(*) AS count 
         FROM spfli 
         GROUP BY carrid 
         ORDER BY carrid. 

         FROM spfli 
         ORDER BY carrid. 

  FETCH NEXT CURSOR dbcur1 INTO count_line. 
  IF sy-subrc <> 0. 
    INTO TABLE spfli_tab PACKAGE SIZE count_line-count. 

CLOSE CURSOR: dbcur1,