Swift up and running--integers and floating-point numbers for the Park Swift series

Swift up and running--integers and floating-point numbers

The original study of the park

Parking Learning Techniques Video

In Swift , numbers are divided into integers (for example: 1/10/100/1000, etc.) and floating-point number (for example: 3.14/1.44/2.71, etc.). Let's look at integers first.

Integers-Int & UInt

According to the content space (8/16/32/64-bit) occupied by an integer variable and whether the integer is signed (unsinged),Swift defines 8 different integer types altogether:

Int8	Int16	Int32	Int64
UInt8	UInt16	UInt32	UInt64

The first row of four types, representing 8/16/32/64-bit signed integers , and the second line is the corresponding unsigned integer type.

In general, however, we do not use these specific integer types directly in our code, we only use Int to define signed integers, and we use UInt to define unsigned integers. The Swift compiler converts an Int or UInt to the corresponding integer type based on the target compilation platform. For example:

On our 64-bit platform, we use the min and max methods, respectively, to see the range of values that Int and Int64 can express:

Int.minInt.maxInt64.minInt64.max

　　

From the above comparison, we can find thatInt and Int64, they can express the range of values, is the same.

Common expressions of integers

In Swift , we can use a number of ways to express an integer. Includes 10-in ,16 -,8 -,2 -binary:

: number Literallet Fifteenindecimal = 15let Fifteeninhex = 0xFlet Fifteeninoctal = 0o17let fifteeninbinary = 0b1111

　　

Well, we can use separators in numbers:

: number Literallet million = 1_000_000

　　

"Unless we have a particularly specific requirement for our hardware platform to use UInt to define unsigned integers, we should always use Int as much as possible to express all integer types, even if we determine that an integer must be a non-negative number." This will give us the hassle of reducing many unnecessary types of conversions. ”

Best Practices

Floating point float & Double

In Swift , there are two different types of floating-point numbers depending on the range of accuracy that can be expressed:

• float: A floating-point number that expresses up to 6 bits of precision ;
• double: A floating-point number that can represent at least 15 bits of precision ;

We print Float and Doubleseparately with print:

var onethirdinfloat:float = 1/3var onethirdindouble:double = 1/3print (onethirdinfloat) print (onethirdindouble)

　　

In addition to using regular decimal expressions for floating-point numbers, we can use the scientific notation. For example, the floating-point pi is represented:

var PI = 0.314e1pi = 314e-2

　　

"If there is no clear demand, we should use Double to define floating-point numbers uniformly. ”

Best Practices

The type inference associated with the number

In Swift , we use an integer, and the compiler infers it to Int, using a floating-point number, which the compiler infers as a Double, for example:

var three = 3three.dynamictypevar Zeropointforteen = 0.14zeropointforteen.dynamictype

　　

We can use DynamicType to see the type of a variable, from the above results we can see that the type of three is Int, and the type of Zeropointforteen is Double.

In Swift , we can compute the literal values of different types of numbers directly:

PI = 3 + 0.14pi.dynamictype

　　

We can see that the integer value 3 and the floating-point number 0.14 can be added directly,andSwift Converts the added result into a Double. 

However, in Swift , we cannot directly perform arithmetic operations on variables of different numeric types , such as:

PI = three + Zeropointforteen

　　

When you do arithmetic on a variable, the types of all variables must be the same, and if the variable types are different, we must explicitly convert some of the variables into the type. Like this:

PI = Double (three) + Zeropointforteen

　　

As you can see from the code above, we use:

Double (Value)

Converts a three of type Int into a floating-point number. Here, the reason we want to quote the conversion is because we did not really convert the three type from Int to double, but instead initialized a new doublewith a value of 3 with the value of three, Add the new Double and Zeropointforteen.

Swift up and running--integers and floating-point numbers for the Park Swift series