colon operator returns different answer in command window and script

function trialpoints = fixnode(min, max, gridfine)

if min<0

min_= min - rem(min,gridfine);

else

min_= min - rem(min,gridfine) + gridfine;

end

max_= max -rem(max,gridfine);

trial1 = min_:gridfine:max_;

trialpoints = [min trial1 max];

end

When I call fixnode function with fixnode(0.0261,0.0341,0.005), it will lead to;

trial1 = 0.03:0.005:0.03; which ends up as empty trial1.

While, if I run Variable= 0.03:0.005:0.03 in command line it gives;

Variable=0.03; --> Why difference in behavor between command line and script

Moreover, When I call fixnode function with fixnode(0.0157,0.0207,0.005), it will lead to;

trial1 = 0.02:0.005:0.02;

trial1= 0.02; --> Why this works? but above not above case.

Answer 1

[Stephen23]

"colon operator returns different answer in command window and script"

Yes, because you called COLON with different values.

"Why this works? but above not above case."

When you run

0.03:0.005:0.03

ans = 0.0300

in the command windows then the first and last values are exactly the same, so the output is a scalar.

But with your function the first and last values are not the same. You cannot rely on the five or so significant figures displayed in the command window to tell you if some values are the same or not. Once we show the values to higher precision, we can see that they are clearly different:

[out,mn,mx] = fixnode(0.0261,0.0341,0.005)

out =

  1×0 empty double row vector

mn = 0.0300

mx = 0.0300

fprintf('%.40f\n',mn,mx); % forty decimal places

0.0300000000000000023592239273284576484002
0.0299999999999999988897769753748434595764

Are those the same value? NO. Lets double check:

mx - mn % same (hint: no)

ans = -3.4694e-18

The value max_ is clearly less than min_, so COLON correctly returns an empty vector for those values.

So far COLON is working exactly as expected and documented.

function [trial1,min_,max_] = fixnode(min, max, gridfine)
if min<0
    min_= min - rem(min,gridfine);
else
    min_= min - rem(min,gridfine) + gridfine;
end
max_= max -rem(max,gridfine);
trial1 = min_:gridfine:max_;
end

[Dhaval Shah]

Thanks for explanation

[Stephen23]

@Dhaval Shah: my pleasure! Please click the accept button to tell other users that your query has been answered.