Several common questions about PHP

PHP several FAQs

One, PHP floating point number of a common problem

First, let's look at an example:

??? Why is the output 57? Is this a bug in PHP? A lot of people often ask such questions, in order to understand this problem, it is necessary to know the expression of floating point number.

For example, a floating-point number with a 64-bit length (double): a 1-bit sign bit (E), a 11 exponent (Q), and a 52-bit mantissa (M) are represented (altogether 64 bits).

The ① sign bit: The highest bit represents the positive or negative of the data, 0 represents a positive number, and 1 indicates a negative number.

???? ② Digit: Indicates that the data is based on a base of 2 emerges, and the exponent is represented by an offset code;

???? ③ Mantissa: A valid number that represents the fractional data.

The key point here is that decimals are represented in binary, about how decimals are represented in binary notation, as follows:

? ? ? 1. Let's start by explaining how the integer part is represented in binary notation, such as:

? is the integer divided by 2 to see its remainder until it is 0 position. The integral part calculation is very simple, so see how the decimal part calculates

? ? ? 2, the fractional part uses the binary expression computation method as follows:

??? for example: 5.6. Integer Part 5 binary is 101

???? 0.6, there is a formula: * * If <1 is 0, cardinality = cardinality, >1 is 1, cardinality = radix-1.

??? Then: 0.6*2=1.2>0? So that's 1? Cardinality =1.2-1=0.2

?????? 0.2*2=0.4<1? So that's 0? base =0.4

?????? 0.4*2=0.8<1? So it's 0? cardinality =0.8

?????? 0.8*2=1.6>1? So, 1? base is 1.6-1=0.6

? ? ? ? ? ?... ...

? ? ? ? ?? So 5.6 can be expressed as: 101.1001 ...

? ? ? 3, 0.58 for the binary representation, the value is infinitely long as follows:

The binary representation of 0.58 is basically (52 bits) is: The binary representation of 00101000111101011100001010001111010111000010100011110.57 is basically (52 bits) is: 0010001111010111000010100011110101110000101000111101

As for the 0.58 * 100 of the specific floating-point multiplication, we do not consider so thin, with vague intention to see: 0.58*100=57.999999999. Then intval, Nature is 57.

The key point of this problem is that " you seem to have a poor decimal number, which is infinite in the binary representation of the computer."

Ii. Merging of PHP arrays

? ? When it comes to array merging, we think of the Array_merge function or the array_merge_recursive function, in fact, there is a way, in this way we usually think very little, that is, "+" operator. Let's look at their usage separately:

? ? 1. Array_merge when the array is marked as an integer type:

?? We can see the subscript drop of the original array and then re-generate a new merged array in the order of the original array.

? ? 2. Array_merge when the array is labeled as a String type:

? ? We can see that the subscript of the array is preserved, but if there is a consistent subscript in the original array, the value of the subsequent array overrides the value of the preceding array.

? ? 3. Array_merge_recursive when the array is marked as an integer type:

? ? Found the same effect as array_merge.

?? 4, Array_merge_recursive when the subscript of an array is a string type:

? ? We find the subscript reserved, and if the original array has a consistent subscript, there is no case of numerical coverage. Instead, an array is generated to hold these values.

? 5. When the "+" operator is used:

??? We can tell whether the subscript type is a number or a string, and if there is a consistent subscript in the original array, then the data behind the array is not stored. In fact, the function of "+" is to complement the functions of the previous array of the subscript is not yet appear in the data from the back of the array to fill the surface.

