Numeric operations.

Overview

Numeric operations. More…

// typedefs

typedef enum te_scalar_type te_scalar_type;

// enums

enum te_scalar_type;

// global variables

const te_enum_map te_scalar_type_names[];

// global functions

uintmax_t te_scalar_type_max(te_scalar_type type);
intmax_t te_scalar_type_min(te_scalar_type type);
size_t te_scalar_type_sizeof(te_scalar_type type);
bool te_scalar_type_is_signed(te_scalar_type type);
te_errno te_scalar_type_fit_size(bool is_signed, size_t size, te_scalar_type* type);
te_errno te_scalar_dynamic_cast(te_scalar_type src_type, const void* src, te_scalar_type dst_type, void* dst);
void te_scalar_random(te_scalar_type type, void* dst);
te_errno te_double2int_safe(long double val, intmax_t lim, intmax_t* result);
te_errno te_double2uint_safe(long double val, uintmax_t max, uintmax_t* result);

// macros

#define TE_SCALAR_TYPE_LIST(HANDLER_)

Detailed Documentation

Numeric operations.

Typedefs

typedef enum te_scalar_type te_scalar_type

Scalar types enumeration

Global Variables

const te_enum_map te_scalar_type_names[]

Maps te_scalar_type into a string name of corresponding C type

Global Functions

uintmax_t te_scalar_type_max(te_scalar_type type)

Get the maximum value of the given scalar type.

Parameters:

type	Scalar type.

Returns:

Maximum value of the scalar type.

intmax_t te_scalar_type_min(te_scalar_type type)

Get the minimum value of the given scalar type.

Parameters:

type	Scalar type.

Returns:

Minimum value of the scalar type.

size_t te_scalar_type_sizeof(te_scalar_type type)

Return the size of storage for values of given type (akin to sizeof).

Parameters:

type	Scalar type.

Returns:

Size of type.

bool te_scalar_type_is_signed(te_scalar_type type)

Checks whether a type is signed or unsigned.

Parameters:

type	Scalar type.

Returns:

true if the type is signed.

te_errno te_scalar_type_fit_size(bool is_signed, size_t size, te_scalar_type* type)

Return a type spec than can hold values of given size.

Parameters:

is_signed	Sign of value type.
size	Size of value type (the same as size of some scalar C type).
type[out]	Location for the scalar type.
TE_ERANGE	There is no scalar type of the requested size and sign.

Returns:

Status code.

te_errno te_scalar_dynamic_cast(te_scalar_type src_type, const void* src, te_scalar_type dst_type, void* dst)

Cast the source value to the destination type.

If the src value cannot be represented in the dst_type exactly an error is returned. However, if the dst is not NULL, it is always overwritten by the casted src even if src value does not fit the range of dst_type.

Parameters:

src_type	Type of source value.
src	Pointer to memory holding the source value.
dst_type	Type of destination value.
dst	Location for the casted value (may be `NULL`).
TE_EOVERFLOW	Value does not fit the destination type.

Returns:

Status code.

void te_scalar_random(te_scalar_type type, void* dst)

Generate a random value that fits the type type.

The function use te_rand_signed() or te_rand_unsigned() depending on the signedness of the type and appropriate limits for that type.

Parameters:

type	Type of a generated value.
dst	Location for a generated value.

te_errno te_double2int_safe(long double val, intmax_t lim, intmax_t* result)

Safely converts a double val to an integer.

The value is truncated towards zero, as with the regular conversion, but non-finite values and overflows are taken care of.

Parameters:

val	input value
lim	limit value: the absolute of `val` shall not be greater than `lim`
result	output value
0	Conversion is successful.
TE_ERANGE	The absolute value of `val` is greater than `lim`
TE_EDOM	`val` is non-zero and is less than `1.0`.
TE_EINVAL	`val` is not finite.

Returns:

status code

Macros

#define TE_SCALAR_TYPE_LIST(HANDLER_)

List of scalar types and their properties.

Each member of the list specifies one of the scalar types and describes its properties:

name of the scalar type: named constants are declared in the enum of scalar types;
corresponding type in C language;
minimum value of the scalar type (zero for unsigned types);
maximum value of the scalar type.

The properties are processed by a user-defined HANDLER_ which normally should be a local macro accepting at least four arguments, e.g.:

#define PRINT(scalar_type_, c_type_, min_, max_) \
    printf("- %s %jd %ju", #c_type_, (min_), (max_));
TE_SCALAR_TYPE_LIST(PRINT)
#undef PRINT

Refer https://en.wikipedia.org/wiki/X_macro for a description of the trick.

When adding new types to the list two constraints must be satisfied for te_scalar_type_fit_size() to work properly:

fixed-size types must precede semantic types;
fixed-size types must be ordered by their size.

Parameters:

HANDLER_

Name of the user-defined handler for members of the list