... FOR ALL ENTRIES IN @itab WHERE ... col operator @itab-comp ...
Effect
If the addition FOR ALL ENTRIES is specified in front of the language element
WHERE of the statement SELECT of a
main query, the components
comp of the internal table itab specified here
can be used within sql_cond on the right side of comparisons of a
relational operator in comparisons with a
columncol. The specified
component comp must be compatible with the column col.
The internal table itab can have a structured or an elementary row type. For an elementary row type, the
pseudo componenttable_line must be specified for comp. The name of the host variable
dbcur should be prefixed with the escape character @.
The entire logical expression sql_cond
is evaluated for each individual row of the internal table itab. The results
set of the SELECT statement is the union set of the results sets produced
by the individual evaluations. Rows that occur more than once are removed from the results set automatically. The full content of a row is considered here.
If the internal table itab is empty, the entire WHERE
condition is ignored. This means that none of the rows in the database table are skipped and are placed in the results set (once any duplicate rows are removed).
The logical expression sql_cond
of the WHERE condition can comprise multiple logical expressions using
AND and OR. However, if FOR ALL ENTRIES is specified, there must be at least one
comparison with a column of the internal table itab that can be specified statically or dynamically.
The following restrictions apply when using the addition FOR ALL ENTRIES with other additions:
The addition FOR ALL ENTRIES is only possible in front of WHERE
conditions in a standalone SELECT statement or in the
main query after OPEN CURSOR, if no common table expressions are defined using WITH.
The addition FOR ALL ENTRIES cannot be used with the addition SINGLE.
The addition FOR ALL ENTRIES cannot be combined with SQL expressions.
The addition FOR ALL ENTRIES cannot be combined with UNION.
The addition FOR ALL ENTRIES should not be used with the addition GROUP BY. The addition GROUP BY has no effect if FOR ALL ENTRIES is used.
In a SELECT statement with FOR ALL ENTRIES,
the addition ORDER BY can
only be used with the addition PRIMARY KEY and can only be used to access
a single table or view. In this case, all columns of the primary key (except for the client column in
client-specific tables) must be in the SELECT list.
No path expressions can be used in a SELECT statement with FOR ALL ENTRIES.
If the addition FOR ALL ENTRIES is used, no database fields of the built-in
types STRING and RAWSTRING plus LCHR and LRAW should occur in the
SELECT list. These data types prevent rows that occur more than once on the database system
from being removed. These rows are only removed from the results set on the application server. If specified
in the SELECT list, a syntax check warning is raised that can be hidden by a pragma.
The internal table itab is evaluated once for each query. Any changes made
to the content of the internal table in a SELECT loop or WITH loop are ignored by the logical expression.
Notes
The same internal table can be specified after FOR ALL ENTRIES and after
INTO. The content of the table
is evaluated by FOR ALL ENTRIES and then overwritten by the INTO clause.
A comparison with a column of an internal table can also be performed using the WHERE condition of a
subquery for the same data source.
With respect to rows occurring more than once in the results set, the addition FOR
ALL ENTRIES has the same effect as when the addition
DISTINCT is specified in the definition of the selection set. Unlike DISTINCT,
the rows are not always deleted from the database system but instead are sometimes first deleted from
the results set on the application server. The duplicate rows are then removed from the database system
if the SELECT statement can be passed to the database system as a single
SQL statement. The addition DISTINCT is supported here. If the SELECT
statement needs to be distributed to multiple SQL statements before it is passed or if columns of the
types STRING and RAWSTRING plus LCHR and LRAW are specified in the SELECT list, the rows are aggregated on the application server.
If duplicate rows are first removed from the application server, all rows specified by the WHERE condition (in some cases) are passed to an internal system table and then aggregated. The
maximum size of this system table is restricted
to that of normal internal tables. More specifically, the system table is always required if one of
the additions PACKAGE SIZE
or UP TO, OFFSET
is used simultaneously. These then have no effect on the number of rows passed from the database server
to the application server, but are only used when the rows are passed from the system table to the actual target area. If the maximum size of the internal system table is exceeded, a runtime error occurs.
The addition FOR ALL ENTRIES bypasses
table buffering for tables
with generic buffering if the condition after FOR ALL ENTRIES prevents a single generic area from being specified exactly.
In all other cases, table buffering is used and the addition FOR ALL ENTRIES can be a more efficient alternative to
join expressions.
Before using an internal table itab after FOR ALL
ENTRIES, always check that the internal table is not initial. In an initial internal tables,
all rows are read from the database regardless of any further conditions specified after WHERE. This is not usually the required behavior.
If the full WHERE condition is ignored because the internal table
itab is empty, the implicit WHERE condition for the current client
or the client specified using USING CLIENT is not affected (if automatic
client handling is switched on). This means
that all data is only read from the current client. If automatic client handling is switched off using
CLIENT SPECIFIED, no implicit
WHERE condition exists for the client. Any WHERE
condition specified explicitly for the client column is ignored with the full condition if the internal table itab is empty and the data from all clients is read.
Gets all flight data for a specified departure city. The relevant airlines and flight numbers are first
passed to an internal table entry_tab, which is evaluated in the WHERE
condition of the subsequent SELECT statement. This selection could also be
carried out in a single SELECT statement by using a join in the FROM clause. Make sure that the table entry_tab
is not initial before the SELECT statement is executed using FOR ALL ENTRIES.
DATA city TYPE spfli-cityfrom VALUE 'FRANKFURT'. cl_demo_input=>request( CHANGING field = city ).
SELECT carrid, connid FROM spfli WHERE cityfrom = @( to_upper( city ) )
INTO TABLE @DATA(entry_tab).
IF entry_tab IS NOT INITIAL.
SELECT carrid, connid, fldate FROM sflight
FOR ALL ENTRIES IN @entry_tab WHERE carrid = @entry_tab-carrid AND
connid = @entry_tab-connid
ORDER BY PRIMARY KEY INTO TABLE @DATA(result_tab). cl_demo_output=>display( result_tab ). ENDIF.
Example
Uses FOR ALL ENTRIES with an empty internal table. All rows of the database
table are respected. The number of read rows is usually, however, smaller in the first SELECT
statement than in the second statement. This is because only one column is read and hence more duplicate
rows can be removed. The second SELECT statement, on the other hand, moves all rows of the database table to the results set, since their structure covers the full table key.
DATA carriers TYPE TABLE OF scarr.
SELECT carrid, connid FROM spfli
FOR ALL ENTRIES IN @carriers WHERE carrid = @carriers-carrid
INTO TABLE @DATA(result1). cl_demo_output=>write( result1 ).
SELECT carrid FROM spfli FOR ALL ENTRIES IN @carriers
WHERE carrid = @carriers-carrid INTO TABLE @DATA(result2). cl_demo_output=>display( result2 ).