Data Types

In the CALL, REGISTER, and REGISTER.ID functions, the type_text argument specifies the data type of the return value and the data types of all arguments to the DLL function or code resource. The first character of type_text specifies the data type of the return value. The remaining characters indicate the data types of all the arguments. For example, a DLL function that returns a floating-point number and takes an integer and a floating-point number as arguments would require "BIB" for the type_text argument.

The following table contains a complete list of the data type codes that Microsoft Excel recognizes, a description of each data type, how the argument or return value is passed, and a typical declaration for the data type in the C programming language.

Code	Description	Pass by	C Declaration

A	Logical (FALSE = 0), TRUE = 1)	Value	short int
B	IEEE 8-byte floating-point number	Value (Windows) Reference (Macintosh)	double (Windows) double * (Macintosh)
C	Null-terminated string (maximum string length = 255)	Reference	char *
D	Byte-counted string (first byte contains length of string, maximum string length = 255 characters)	Reference	Unsigned char *
E	IEEE 8-byte floating-point number	Reference	double *
F	Null-terminated string (maximum string length = 255 characters)	Reference (modify in place)	char *
G	Byte-counted string (first byte contains length of string, maximum string length = 255 characters)	Reference (modify in place)	unsigned char *
H	Unsigned 2-byte integer	Value	unsigned short int
I	Signed 2-byte integer	Value	short int
J	Signed 4-byte integer	Value	long int
K	Array	Reference	FP *
L	Logical (FALSE = 0, TRUE = 1)	Reference	short int *
M	Signed 2-byte integer	Reference	short int *
N	Signed 4-byte integer	Reference	long int *
O	Array	Reference	Three arguments are passed: unsigned short int * unsigned short int * double [ ]
P	Microsoft Excel OPER data structure	Reference	OPER *
R	Microsoft Excel XLOPER data structure	Reference	XLOPER *

Remarks

The C-language declarations are based on the assumption that your compiler defaults to 8-byte doubles, 2-byte short integers, and 4-byte long integers.
In the Microsoft Windows programming environment, all pointers are far pointers. For example, you must declare the D data type code as unsigned char far * in Microsoft Windows.
All functions in DLLs and code resources are called using the Pascal calling convention. Most C compilers allow you to use the Pascal calling convention by adding the Pascal keyword to the function declaration, as shown in the following example:
```
pascal void main (rows,columns,a)
```
If a function uses a pass-by-reference data type for its return value, you can pass a null pointer as the return value. Microsoft Excel will interpret the null pointer as the #NUM! error value.