How to change e format?

I want to change the number's format.

I written

x=[0.00005; 0.0002; 1.00; 0.0015; 0.02; 0.01;];
txt=sprintf('%.e %.4f %.2f %.4f %.2f %.2f',x);

and the result is txt

5.e-05 0.0002 1.00 0.0015 0.02 0.01

But I want to change

x=[0.00005; 0.0002; 1.00; 0.0015; 0.02; 0.01;];

to

x=[5E-005; 0.0002; 1.00; 0.0015; 0.02; 0.01;];

What can I do for this?

Plz help me... ;(

[Walter Roberson]

x = [0.00005; 0.0002; 1.00; 0.0015; 0.02; 0.01;];
txt = sprintf('[%.0e; %.4f; %.2f; %.4f; %.2f; %.2f]',x)

txt = '[5e-05; 0.0002; 1.00; 0.0015; 0.02; 0.01]'

[hyein Sim]

Thank you for response! but I want 0.00005 to 5e-005.

Which have two zero after e ..

How can I do that?

[Walter Roberson]

What is the purpose of doing that?

What are the rules involved? absolute value less than 1e-4 is to be converted to single digit scientific? Absolute value 1e-4 up to (excluding) 1e-2 is to use 4 digit fixed format? Absolute value 1e-2 (inclusive) to (something?) is to use 2 digit fixed format?

[hyein Sim]

I connect MATLAB and other programs, and the format of substituting numbers into that program is, for example, 8.4e-005 or 5.0e-005. After e, 3 numbers.

So I have to express all numbers in the range 0.00005~0.00015 like that format.

:) Thank toy for your response again..

[Walter Roberson]

We need rules. How many digits after the decimal place, under what circumstances?

For example how should 0.000314159265358979308997711132889207874541170895099639892578125 be output? How should pi/10000 be output? How should 70, 7, 1/7, 1/70, 1/700, 1/7000 be output ?

[hyein Sim]

When I put in a number from one to four decimal places at the program(Hydrological program) which I use, such as 0.1 or 0.0004, its are written 0.1 or 0.0004 in the input file of the program. However, if I insert more than 5 decimal places such as 0.00005 or 0.0000074, its are written in input file as 5e-005 or 7.4e-006. Therefore, I change the decimal form because I want to put numbers in that program using matlab. But matlab format is like 5e-05.

You said 0.000314159...number. If I write into the program I use, it will be printed as 0.000314159. Also, I can't write special symbols such as pi/10000 or 1/7 in the program I use.I have to calculate the rest myself and insert it in decimal form. 70 or 7 is output as is. Is this enough for the rules? I'm sorry to bother you, but I've been thinking about it for a week and haven't found the result yet.

[hyein Sim]

So that I just want to know how to change the numbers which range of 0.00005 ~0.00015 to like this 5e-005 format. They will all have 5 decimal places.

[Rik]

So you just want to format the exponent with 3 digits instead of 2? (And use '%.4f' for values larger than 5e-5)

[hyein Sim]

@Rik Yes!! :)

Answer 1

[Walter Roberson]

Please check the output for pi/1000 . Is the rule intended to be "up to 9 decimal places" like it is for 0.000314159, which would give 0.003141592 if you used truncation and 0.003141593 if you used rounding. But perhaps the rule is non-zero plus up to 5 further (so up to 6 decimal places counting from the first non-zero), in which case pi/1000 would give 0.00314159 through rounding or truncation. But if the input value happened to be 0.003141596 and the rule is non-zero+5, then is truncation to 0.00314159 okay, or would you require rounding to 0.00314160 ? And in that case where the trailing digit is not "naturally" 0 but rounded to it, do you want to preserve the trailing 0 or get rid of it, to 0.0031416 ?

Note: I preserved the trailing semi-colon before the ] that you had in your output, in case it was meaningful.

Note: I did not attempt to detect row-vectors and use commas for those, and I did not attempt to deal with 2D arrays.

x = [0.00005; 0.0002; 1.00; 0.0015; 0.02; 0.01; pi/1000; pi/10000];

txt = format_vector(x);
disp(txt)

[5E-005; 0.0002; 1.00; 0.0015; 0.02; 0.01; 0.003141593; 0.000314159;]

function txt = format_vector(vec)
  txt = ['[', strjoin(arrayfun(@format_number, vec, 'uniform', 0), ' '), ']'];
end

function str = format_number(num)
    %format of each number includes a trailing semi-colon
    if abs(num) >= 1e-4
        str = sprintf('%.9f;', num);
        %Rule: 1 exactly becomes '1.00' not '1.' so preserve at least 2
        %decimal places
        %Rule: but otherwise trim trailing 0
        str = regexprep(str, '0{1,7};$', ';');
    else
        %Rule: exponent must have three digits
        %Rule: trailing 0 after period must be removed
        %Rule: if all digits after period are 0, remove period too
        %Rule: exponent character must be 'E' not 'e'
        str = sprintf('%.5E;', num);
        str = regexprep(str, {'0+E', '\.E'}, {'E', 'E'});
        str = regexprep(str, {'(E[-+])(\d);$', '(E[-+])(\d\d);$'},{'$100$2;', '$10$2;'});
    end
end

[hyein Sim]

Thank you for kind responses 👍 I will try this and write the results.