Skip to content

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

SELECT - ORDER BY

Short Reference

Other versions: 7.31 | 7.40 | 7.54

Syntax


... ORDER BY { {PRIMARY KEY} 
             | { {col1|a1} [ASCENDING|DESCENDING], 

                 {col2|a2} [ASCENDING|DESCENDING], ...} 

             | (column_syntax) } ...

Alternatives

1. ... ORDER BY PRIMARY KEY

2. ... ORDER BY {col1|a1} [ASCENDING|DESCENDING],
               {col2|a2} [ASCENDING|DESCENDING], ...

3. ... ORDER BY (column_syntax)

Effect

The addition ORDER BY sorts a multirow results set by the content of the specified column. The order of the rows in the results set is undefined with respect to all columns that are not specified after ORDER BY, and can be different in repeated executions of the same SELECT statement. If the addition ORDER BY is not specified, the order of all the columns in the results set is undefined.

The following restrictions apply when using the addition ORDER BY with other additions:

  • If the addition FOR ALL ENTRIES is used in the WHERE condition,, ORDER BY can only be used with the addition PRIMARY KEY and all columns of the primary key must be specified after SELECT. The addition ORDER BY cannot be used with the addition SINGLE.
  • If aggregate functions are specified after SELECT, all columns that are specified after ORDER BY and that do not have an alternative column name for an aggregation function must also be specified after SELECT and after GROUP BY.
  • If an alternative name is used for sorting, this name must be unique and cannot be the same name as a column that does not have any alternative names.
  • If the addition DISTINCT is used, only those columns can be specified after ORDER BY that are also listed after SELECT. The exception to this rule is client column when PRIMARY KEY is specified. If not, other columns can also be used, as long as there are no restrictions by other additions such as GROUP BY.


Notes

  • The data is sorted in the database system, once all other actions are completed, such as the definition of the hit list using WHERE, the calculation of aggregate functions, and grouping using GROUP BY. Only the addition UP TO n ROWS is performed once the hit list is sorted.
  • If a column specified after ORDER BY contains null values in the results set, the sort order can be platform-dependent, since null values can be sorted either before or after the other values by the database system (in accordance with the SQL standard)
  • For performance reasons, a sort should only take place in the database if supported by an index. This guaranteed only when ORDER BY PRIMARY KEY is specified. If a suitable index is not available, the results set must be sorted at runtime. This should be done using SORT on the application server and not using ORDER BY in the database system. Even if a suitable index does exist, ORDER BY col1 col2 ... should be used for large amounts of data only if the order of the database fields col1 col2 ... is the same as the order in the index.
  • :If a sorted results set is assigned to a sorted internal table, the internal table is sorted again according to the sorting instructions.

Alternative 1

... ORDER BY PRIMARY KEY

Effect

If all columns are specified in the SELECT list (using *), and a single database table is specified after FROM (rather than a view or a join expression), the addition PRIMARY KEY can be used to sort the results set in ascending order in accordance with the content of the primary key of this database table.

The addition PRIMARY KEY cannot be specified if the following is specified after FROM:

If this type of view or a join expression is specified after FROM in a dynamic source_syntax, the data is sorted by all columns of the results set.


Example

Reads the data from database table SFLIGHT for Lufthansa flight 0400, sorted by flight date.

DATA wa_sflight TYPE sflight. 

SELECT * 
       FROM sflight 
       WHERE carrid = 'LH' AND 
             connid = '0400' 
       ORDER BY PRIMARY KEY 
       INTO @wa_sflight. 
  ... 
ENDSELECT.

Alternative 2

... ORDER BY {col1|a1} [ASCENDING|DESCENDING],
             {col2|a2} [ASCENDING|DESCENDING], ...

Effect

For any columns specified in the SELECT list, a comma-separated list of columns can be specified after ORDER BY to be used as a sort criterion. Columns can be specified directly using the column names col1 col2 ..., or the alternative column names a1 a2 .... The latter is required if columns specified as aggregate expressions are to be used as sort criteria. When multiple database tables are accessed and a column name is not unique, the column must be identified using the column selector ~.

The additions ASCENDING and DESCENDING determine whether the rows are sorted in ascending or descending order. If neither addition is specified, the sort is performed in ascending order. The priority of sorting is based on the order in which the components col1 col2... or a1 a2 ... are specified.

Pooled and cluster tables cannot be sorted by all types of column. Columns specified after ORDER BY cannot be of the type LCHR, LRAW, STRING, or RAWSTRING.


Notes

  • If single columns are specified in the addition ORDER BY, the statement SELECT uses the SAP buffering only in the following cases:

  • The columns specified are a left-justified subset of the primary key in the correct order and no further columns are specified.

  • The columns specified represent the whole primary key in the correct order. Additional columns that are specified have no influence on the sorting.
In other cases, SAP buffering is ignored.
  • When a comma-separated list is used, the syntax check is performed in a strict mode, which handles the statement more strictly than the regular syntax check.
  • Instead of using commas, blanks can be used to separate columns specified in an obsolete form. Commas must be specified, however, in the strict modes of the syntax check from Release 7.40, SP05.


Example

The rows of database table sflight are grouped by the columns carrid and connid, where for each group the minimum of column seatsocc is determined. The selection is sorted in ascending order by carrid and in descending order by the minimum of occupied seats. The alternative name min is used for the aggregate expression.

TYPES: BEGIN OF wa, 
        carrid TYPE sflight-carrid, 
        connid TYPE sflight-connid, 
        min    TYPE i, 
      END OF wa. 

DATA itab TYPE TABLE OF wa WITH EMPTY KEY. 

SELECT carrid, connid, MIN( seatsocc ) AS min 
       FROM sflight 
       GROUP BY carrid, connid 
       ORDER BY carrid ASCENDING, min DESCENDING 
       INTO CORRESPONDING FIELDS OF TABLE @itab. 

cl_demo_output=>display_data( itab ).

Alternative 3

... ORDER BY (column_syntax)

Effect

As an alternative to specifying columns statically, a parenthesized data object column_syntax can be specified, which either contains the syntax of the list of columns or is initial when the statement is executed. The same applies to column_syntax as when specifying the SELECT list dynamically.

The addition PRIMARY KEY cannot be specified in column_syntax. For column_syntax, the same applies as to specifying columns dynamically after SELECT. If the content of column_syntax is initial, the addition ORDER BY is ignored.

Security Note

If used wrongly, dynamic programming techniques can present a serious security risk. Any dynamic content that is passed to a program from the outside must be checked thoroughly or escaped before being used in dynamic statements. This can be done using the system class CL_ABAP_DYN_PRG or the predefined function escape. See SQL Injections Using Dynamic Tokens.


Notes

  • The class CL_ABAP_DYN_PRG contains methods that support dynamically specified columns when created correctly and securely.

  • The literals of the dynamically specified Open SQL statements can span multiple rows of a token specified dynamically as an internal table.

  • When specified dynamically, Open SQL statements can contain the comment characters * and " as follows:

  • In a dynamic token specified as a character-like data object, all content is ignored from the first comment character ".

  • In a dynamic token specified as an internal table, all rows are ignored that start with the comment character *. In the row, all content is ignored from the first comment character ".

Comment characters placed within literals are, however, part of the literal.

Example

See Dynamic ORDER BY Clause.

Continue

SELECT, Dynamic ORDER-BY Clause - Example