In order to provide consistency between calls from different
programming languages, the following rules apply to all parameters:
Parameter Types |
|
The data types that are used in VPLUS intrinsics are shown
in Table 6-3 “Data Types Used for Various Languages” Note that not all types are allowed
for all languages.
Table 6-3 Data Types Used for Various Languages
Data Type | COBOL | FORTRAN | BASIC | PASCAL | SPL |
---|
Character | DISPLAY PIC X(n) | Character | String | Packed Array of Char | Byte Array |
Two-byte Integer | COMP PIC S9 thru PIC S9(4) | Integer*2 | Integer | Subrange -32768.. 32767 | Integer |
Unsigned Two-byte Integer | COMP PIC 9 thru PIC 9(4) | Logical | Integer (with value <32767) | Subrange 0..65535 | Logical |
Four-byte Integer | COMP PIC S9(5) thru PIC S9(9) | Integer*4 | Integer Integer1 | Integer | Double |
Real Four-byte | — | Real | Real | Real | Real |
Long 8-byte | — | Double Precision | Long | Longreal | Long |
1In
BASIC, a double integer can be represented by two consecutive integers: the first contains the high-order digits of values
above 32767 or is zero, the second contains the low-order digits of values
above 32767 or the entire value up to 32767.
|
The VPLUS parameters use only data types that are available
in all programming languages: character, integer, logical, and double
integer. The only exceptions are the transfer and conversion intrinsics, VPUTtype and VGETtype, which use real and long.
Each parameter is described according to its generic type
(character, integer, logical, or double integer). This table is
provided for those languages that do not call their data types by
these particular names. For example, if you are coding in COBOL
and a parameter is specified as logical, you can determine from
this table that it is an unsigned computational item that uses from
one to four digits.