14.1.2 Field Groups and Expandable Groups

The GROUP_BY command is used to group one or more fields under a common name. It is one of the most time saving of all the RDML commands because it saves having to repeatedly specify a long list of field names.

Most commands that require a list of field names as a parameter also allow a group name to be specified. Consider the following example:

BEGIN_LOOP

REQUEST    FIELDS(#ORDLIN #PRODUCT #QUANTITY #PRICE)

INSERT     FIELDS(#ORDLIN #PRODUCT #QUANTITY #PRICE)

           TO_FILE(A)

INSERT     FIELDS(#ORDLIN #PRODUCT #QUANTITY #PRICE)

           TO_FILE(B)

UPRINT     FIELDS(#ORDLIN #PRODUCT #QUANTITY #PRICE)

CHANGE     FIELD(#ORDLIN #PRODUCT #QUANTITY #PRICE)

           TO(*DEFAULT)

END_LOOP

Now consider the identical RDML program written using a GROUP_BY command to group all the fields under a common name:

GROUP_BY   NAMED(#ORDERLINE) FIELDS(#ORDLIN #PRODUCT #QUANTITY

                                    #PRICE)

BEGIN_LOOP

REQUEST    FIELDS(#ORDERLINE)

INSERT     FIELDS(#ORDERLINE) TO_FILE(A)

INSERT     FIELDS(#ORDERLINE) TO_FILE(B)

UPRINT     FIELDS(#ORDERLINE)

CHANGE     FIELD(#ORDERLINE) TO(*DEFAULT)

END_LOOP

If 5 new fields were to be added to the RDML program which would be the easiest to change?

Some points to note about groups and the GROUP_BY command are:

• As many GROUP_BY commands as desired, can be declared.

• A field can be declared in the FIELDS parameter of multiple GROUP_BY commands.

• Generally a GROUP_BY name can be used wherever a list of field names can be specified.

• If a GROUP_BY name is declared like this:

GROUP_BY   NAMED(#ORDERHEAD) FIELDS(#ORDER #CUSTNO #ADDR1

                                    #ADDR2

                                    #POSTCD)

   and then used like this:

 

FETCH      FIELDS(#ORDERHEAD) FROM_FILE(ORDHED)

FETCH      FIELDS(#ORDERHEAD) FROM_FILE(CUSMST)

   then the first FETCH will only retrieve fields in the group that come from the ORDHED file. In this case only #ORDER, and #CUSTNO will be fetched. All other fields in the group are unchanged / ignored by the command because they don't come from the ORDHED file.

   The second FETCH will retrieve the fields #CUSTNO (again), #ADDR1, #ADDR2 and #POSTCD because they all come from the CUSMST file. Field #ORDER will be unchanged / ignored by this command as it does not come from the CUSMST file.

Also See

Expandable Groups

Special Considerations for Expandable Groups

Expandable Group Expressions

Expandable Group Examples