var_set

Add variable(s) to a dataset and initialize it/them to some value(s).

%var_set(idsn, var=, val=, odsn=, force_set=, ilib=WORK, olib=WORK);

Arguments

• idsn : input reference dataset, where variables shall be inserted;
• var : (option) list of variables that should be inserted;
• val : (option) list of values that should be used to inizialize the inserted variables; the number of arguments passed in var and val MUST be the same;

• force_set : (option) boolean flag (yes/no) set to force the initialisation of the variable passed in var:

  • yes: the variable in var is/are initialised whether it/they already exist/s in the dataset or not,
  • no: the variable in var is/are initialised only when it is added to the dataset;

  default: force_set=no is used;

• ilib : (option) name of the input library; by default: empty, i.e. WORK is used.

Returns

• odsn : (option) name of the output dataset (stored in the olib library), that will contain the data from idsn, plus the new variables inserted; default: odsn=idsn and the input dataset idsn is updated instead;
• olib : (option) name of the output library; by default: empty, and the value of ilib is used.

Examples

Let us consider test dataset #5 in WORKing directory:

f	e	d	c	b	a
.	1	2	3	.	5

then both calls to the macro below:

%var_set(_dstest5, var=FFF AA, val= 10 20);

will return the exact same dataset out1=out2 (in WORKing directory) below:

f	e	d	c	b	a	FFF	AA
.	1	2	3	.	5	10	20

Instead, the following instruction:

%var_set(_dstest5, var=k a, val= 10 20, force_set=YES);

will return the exact same dataset out1=out2 (in WORKing directory) below:

f	e	d	c	b	a	k
.	1	2	3	.	20	10

Run macro %_example_var_set for more examples.

Note

When force_set=yes, the value of existing variable/s may be replaced this way.

See also

%list_length, %var_check.