*Byte Data Type
Estimated reading time: 3 minutes
*Byte variables are stored as unsigned 8-bit (1-byte) numbers ranging in value from 0 through 255.
The *Byte data type is used for containing binary data. The equivalent .NET data type is System.Byte.
*BYTE is not *CHAR
The *CHAR is a text data type. It is used to store characters. *BYTE is a binary data type. It is used to store numbers. *CHAR and *BYTE are not interchangeable. They are not implicitly convertible to each other.
On IBM’s RPG, *CHAR is used to hold Text with its byte content interpreted as EBCDIC. Sometimes, a CHAR(1) is used to hold a single byte of binary data, but in order to use it in numeric operations it is moved using a Data Structure to the low byte of a 2-byte integer field. This practice assumes that the numeric values use a Big-Endian encoding, which is true on the /400, but that is not true on all platforms, particularly those using x86 processors.
Encore RPG provides the additional type *BYTE to facilitate certain migration operations. *BYTE is a 1-byte nuemric field which has no implicit conversion to *CHAR. *BYTE cannot be used in a MOVE operation.
Non *BYTE Numeric Fields and *CHAR on IBM’s RPG and Encore RPG
On Encore RPG and on IBM’s RPG, when *CHARs interact with Numeric types using the MOVE opeartion, numeric data is converted to its decimal representation as a Zoned value but without a decimal point, and the resulting text is moved to or from the *CHAR field.
Some examples
DclFld CharBits Type(*Char) Len(1)
DclFld ByteBits Type(*Byte)
DclFld Int2 Type(*Integer2)
DclArray ByteArray Type(*Byte) Rank(1) New(1)
DclDs ByteFiddling
DclDsFld CharIn1 Type(*Char) Len(1)
DclDsFld ByteIn1 Type(*Byte) StartAt(1)
DclDsFld ByteIn2 Type(*Byte) StartAt(2)
BegSr DontMixCharWithBytes
ByteBits = H"4E" // Hex 4E, Decimal 78, ASCII 'N', EBCDIC '+'
// *BYTE can never be the Source or Target of a MOVE operation
// Move ByteBits CharBits <<< Produces a compile error
// Move CharBits ByteBits <<< Produces a compile error
// *BYTE cannot be implicitly converted to or form *CHAR
// CharBits = ByteBits <<< Produces a compile error
// ByteBits = CharBits <<< Also produces an error
// The Built-in %CHAR() function can be used to convert a *BYTE to a *CHAR.
// %CHAR() converts a number to its corresponding Text. In this case "78"
CharBits = %CHAR(ByteBits) // CharBits = "7"
Move %CHAR(ByteBits) CharBits // CharBits = "8"
// To 'convert' a *BYTE to its encoded character value, use a DataStructure to
// reinterpret the underlying bits.
ByteIn2 = *Zeros
ByteIn1 = ByteBits
// CharIn1 is 'N'; On the /400 it would have been a '+'.
// This is because the /400 uses EBCDIC encoding, not Unicode.
// You could also use one of the System.Text.Encoding classes to achive the save result
ByteArray[0] = ByteBits
CharIn1 = (*new System.Text.ASCIIEncoding()).GetString(ByteArray)
// Byte fields can be asigned to/from other numeric types
Int2 = ByteBits // Int2's value is now 78
EndSr