fprintf format -.123456879E-02

How can I use fprintf to print in scientific format, always starting with 0.XXXXXXE-0Y in case of a positive number, or -.XXXXXXE-0Y in the case of a negative number.

Example: a=-0.07639297; fprintf('%E',a);

This will print the number as: -7.639297E-02

However, I want to have it as: -.7639297E-01

Thanks!

Answer 1

[Star Strider]

You have to ‘adjust’ the format a bit, but it can be done as a string:

a=-0.07639297;
expstr = @(x) [x*10.^floor(-log10(abs(x)))  floor(log10(abs(x))+1)];
Result = sprintf('%.7fE%+04d', expstr(a))

Result =
-0.7639297E-001

Tweak the format in the sprintf call to your liking.

[Enrico Tapavicza]

Thank you, this is already close to what I want, but how can I remove the zero before the point to get: -.7639297E-001 ?

Thanks!

[Star Strider]

It’s possible, but the code looks really kludgy:

a=-0.07639297;
expstr = @(x) [x*10.^floor(-log10(abs(x)))  floor(log10(abs(x))+1)];
exvct = expstr(a);
sgn = '- +';
Result = sprintf('%c.%7dE%+04d', sgn(sign(a)+2),abs(exvct(1)/10^(exvct(2)-7)),exvct(2))

Result =
-.76392970E-001

If you’re going to use this frequently, it would be best to wrap it in an external function, add a precision variable to make it robust to your needs (the ‘7’ values in this code would be the precision variables to adjust), and experiment with it to get the result you want. The format descriptor itself is a string, so you can create a separate sprintf call to create it, then pass it to the sprintf call that produces ‘Result’ here. ‘Result’ is a string variable, not numeric, so print it in your own fprintf call using '%s'.

[Enrico Tapavicza]

All right, thank you very much!

[Star Strider]

My pleasure!

The sincerest expression of thanks here on MATLAB Answers is to Accept the Answer that most closely solves your problem.