Type System

Cyrus features a strong and expressive type system. The compiler performs implicit type conversions when it is safe, and requires explicit casts when conversions could be unsafe or lossy.

Safe Implicit Casting

If a type can be safely converted to another type without losing information, Cyrus will perform the conversion automatically.

#a: int32 = 42;
#b: int64 = a; // Implicit cast from int32 to int64

int32 (32-bit signed integer) can always fit into int64 (64-bit signed integer).
The compiler handles this implicitly, so no special syntax is required.
Safe casts also apply to numeric literals, e.g., #x: float64 = 10.0 is allowed.

Unsafe / Explicit Casting

If a conversion could be unsafe, the compiler requires an explicit cast using the cast keyword.

#u: uint32 = 10;
#i: int64 = cast(int64, u); // Explicit unsafe cast from uint32 to int64

Basic Types

Cyrus provides a set of built-in types for numbers, characters, booleans, and special-purpose values. These are the fundamental building blocks of your programs.

Type	Category	Bit Width	Signedness
int	Signed Integer (default)	Target-dependent (usually 32 or 64)	Signed
int8	Signed Integer	8	Signed
int16	Signed Integer	16	Signed
int32	Signed Integer	32	Signed
int64	Signed Integer	64	Signed
int128	Signed Integer	128	Signed
uint	Unsigned Integer (default)	Target-dependent (usually 32 or 64)	Unsigned
uint8	Unsigned Integer	8	Unsigned
uint16	Unsigned Integer	16	Unsigned
uint32	Unsigned Integer	32	Unsigned
uint64	Unsigned Integer	64	Unsigned
uint128	Unsigned Integer	128	Unsigned
intptr	Pointer-sized Integer	Pointer width (32 or 64)	Signed
uintptr	Pointer-sized Integer	Pointer width (32 or 64)	Unsigned
size_t	Size / Index Type	Pointer width (32 or 64)	Unsigned

Floating-Point Types

Type	Description	Precision
float16	16-bit floating-point number (half precision)	~3 decimal digits
float32	32-bit floating-point number (single precision)	~7 decimal digits
float64	64-bit floating-point number (double precision)	~15–16 decimal digits
float128	128-bit floating-point number (quadruple precision)	~34 decimal digits

Character and Boolean Types

Type	Description	Notes
char	A single character unit (one code unit; multi-byte Unicode not allowed in `'...'`).	Represents ASCII/byte chars only
bool	Boolean values, either `true` or `false`.	1-bit storage (typically padded to 8 bits)

Special Types

Type	Description	Notes
void	Represents no value, commonly used for functions that do not return anything.	Return type only