How are uint8 calculated? And how can I convert them back to Integers in Codesys?

Question

Tim 2023-12-20

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2062542-how-are-uint8-calculated-and-how-can-i-convert-them-back-to-integers-in-codesys

编辑： Varun 2023-12-26

I am converting a Struct containing integers to a uint8

data = struct('Header', [], 'Pos', []);
data.Header = 1010;
data.Pos = double([81, -17, 8651]);
uintHeader = typecast(data.Header, 'uint8');
uintPos = typecast(data.Pos, 'uint8');
dataArray = [uintHeader uintPos];

dataArray = 0 0 0 0 0 144 143 64 0 0 0 0 0 64 84 64 0 0 0 0 0 0 49 192 0 0 0 0 128 229 192 64

How do i get from '1010' to '0 0 0 0 0 144 143 64' etc.?

I then send this array to a system that runs on Codesys.

Is there a function in Codesys, which can convert it back to the original integers?

3 个评论
显示 1更早的评论隐藏 1更早的评论

Tim 2023-12-20

Hi, data.Header is 64b float.

Michael 2023-12-20

在 MATLAB Online 中打开

Since 'Header' is a 64 bit float number the representation in hexadecimal format is

format hex
1010
ans = 
   408f900000000000

and each byte in ascending order

format short
uint8([0x0 0x0 0x0 0x0 0x0 0x90 0x8f 0x40])
ans = 1×8
     0     0     0     0     0   144   143    64

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Varun 2023-12-26

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2062542-how-are-uint8-calculated-and-how-can-i-convert-them-back-to-integers-in-codesys#answer_1377812

编辑：Varun 2023-12-26

在 MATLAB Online 中打开

Hi Tim,

I understand that you are using the MATLAB functions “typecast” as below:

uintHeader = typecast(data.Header, 'uint8');

To know about working of “typecast”, consider an example:

x=unit32(256);
y=typecast(x,'unit8');

To represent decimal 256, 8 bits are not enough, you need minimum 9 bits. It can be easily stored in 32 bits but how can it be stored in 8 bits. You can see that we are trying to typecast 256 into “uint8” i.e., only 8 bits.

So, the “typecast” function splits this 32-bit representation of 256 into 4 chunks each with 8 bits or 1 byte. It then uses little-endian convention to produce the following row vector where each element is of type “uint8”.

y =
1×4 uint8 row vector
0   1   0   0

If “x” was of 64 bits, then it would have split into 64/8 = 8 chunks and calculate each chunk value independently and store them in the row vector using the little-endian convention.

Please refer to the following MATLAB documentation for “typecast”:

https://www.mathworks.com/help/matlab/ref/typecast.html

Now, answering your 2nd part of the question where you want to convert back from this “uint8” vector notation back to original “uint32” integer in CODESYS. You can refer to the following example:

VAR
Y: ARRAY[0..11] OF BYTE; // Your array of bytes
X: ARRAY[0..2] OF UDINT; // Resulting array of 32-bit unsigned integers
END_VAR
// Assume Y is your 12-element array of bytes
X[0] := BYTE_TO_UDINT(Y[0 TO 3]);
X[1] := BYTE_TO_UDINT(Y[4 TO 7]);
X[2] := BYTE_TO_UDINT(Y[8 TO 11]);