Reading A matrix created in MATLAB ?

[Pushhead]

Hi All!

I'm trying 2 read a text file containing a N X 2 Matrix created in matlab.
The file was created in matlab using :

 n=1:3;
a=exp(j*n);
y=[real(a);imag(a)];
fid = fopen('cplx1.txt','w');
fprintf(fid,'%g %g\n',y);
fclose(fid);

I tried 2 read the file using (TClite):

....
}
 while (!feof(fid))
 {
  fscanf(fid,"%g %g",&array[i][0],&array[i][1]);
  printf("x[%d]= %g+j%g \n",i,array[i][0],array[i][1]);
  i++;
  }
 }...

I get the error msg: "scanf: Floating point formats not linked, Abnormal program termination".
Where is the problem?
is it C or MATLAB?
Thanks 4 your time.

 

[Salem]

Your compiler is too dumb to spot that you need floating point support

http://www.eskimo.com/~scs/C-faq/q14.13.html

Various "hacks" are needed to make it realise this, eg create the following global
float dummy = 1.0;

> while (!feof(fid))
This is the wrong way - you'll find it processes the last line "twice"

Should be

while ( fscanf(fid,"%g %g",&array[i][0],&array[i][1]) == 2 ) {
    printf("x[%d]= %g+j%g \n",i,array[i][0],array[i][1]);
    i++;
}

 

[Pushhead]

I tried your suggestion, but i still get
the same error.

can u think of something else?

thanks.

 

[Salem]

Did you follow the FAQs to....

http://www.stratoware.com/msdos/dos-faq.html#Q0304

 

[butthead]

If you can handle the bounce at the end of the file,
"while (!feof(fid))" will work fine.

I have good luck with something like this. Sometimes you have to fiddle with the if statement below to exit properly, depending upon your intentions.

bool done = false;

while (!done)
{
fscanf(fid,"%g %g",&array[0],&array[1]);
printf("x[%d]= %g+j%g \n",i,array[0],array[1]);
i++;

if (feof (fid))
{
done = true;
}
}

tomcruz.net

 

[Pushhead]

first of all thanks alot 4 ur help,
it just shows people are SERIOUS here, in comparison 2 other forums i've visited....


but:
the error messege has disappeared, but the values i get r not the same.
the data file( text format) contains the matrix:
0.540302 0.841471
-0.416147 0.909297
-0.989992 0.14112

and what i get is:
x[0]=-9.25526e+061+j-9.25526e+061
x[1]=-9.25526e+061+j-9.25526e+061
x[2]=-9.25526e+061+j-9.25526e+061
x[3]=6.25526e-314+j-9.25526e-314

can u help?
10x alot.

 

[Salem]

I'm guessing array is an array of double, not an array of float

In which case, use "%lg" (not "%g&quot in the scanf() conversion string.