Definition of the seife format

This format is defined by E. Wielandt for his time series analysis programs. You can find a copy of the source code at http://www.software-for-seismometry.de

The format description taken from seife.doc

 This format was defined prior to most other presently used data formats,
 and I am still adhering to it although it is now obsolete. The structure
 of the data files is as follows:

 - one header line, arbitrary (will be echoed but not evaluated) 
 - up to 48 comment or protocol lines marked with a  % character in the
   first column. Protocol lines are automatically generated by SEIFE
   unless seife.par contains the code ‘ncl’ (no comment lines).
 - one line containing the number of samples, the FORTRAN format in which
   they are listed, the sampling interval, and an optional time stamp as
   described below.
 - data in the specified format (here 5f10.3, that is, five entries like
   123456.789 per line).

 The entries in the last header line (which is the second line if there
 are no comments)  must be in the FORTRAN format (i10,a20,f10.x). It may
 contain a time stamp, in which case the format is (i10,a20,3f10.x). The
 time stamp consists of two numbers: minutes and seconds after midnight.
 No date can be encoded, except by expressing it in minutes. The whole
 line might read:         

12000     (5f10.3)            0.01      617.      23.45

The strings begin at positions 1, 11, 31, 41, 51. The time of the first sample is 10:17:23.45.

The SEIFE format applies only to files read with the ‘fil infile outfile’ command line, and to the output files. Other ASCII data formats can be read with the asc, asl, bdf and mar commands; seife, gnuplot, and headerless single-column ASCII formats can be specified for the output. Miniseed files can be converted into ASL format with Quanterra’s CIMARRON software, and then read with SEIFE. A converter from GSE to ASCII format named CODECO written by Urs Kradolfer is also available. I am offering a WINDOWS executable of this program.

Specific format supported by this library

 Since we do not provide a full featured Fortran format parser, we limit
 the supported data format to an arbitrary sequence of floating point
 numbers separated by whitespace. This is any sequence, which can be read
 sequentially by the C++ input streams. If your data has no whitespace
 separating individual values (which is possible with fixed formats in
 Fortran), you should use the Fortran seife.f program itself to convert
 the data.

 The appropriate data format for output is

50 format (6(g12.5,1x))

This produces

0.84147E+34  0.90930E+34  0.14112E+34 -0.75680E+34 -0.95892E+34 -0.27942E+34
0.65699E+34  0.98936E+34  0.41212E+34 -0.54402E+34 -0.99999E+34 -0.53657E+34
0.42017E+34  0.99061E+34  0.65029E+34 -0.28790E+34 -0.96140E+34 -0.75099E+34
0.14988E+34  0.91295E+34

Similar output is produced by

int j=0;
for (int i=0;i<30;++i)
{
  cout << setprecision(6) << setw(12);
  cout << 1.e34*std::sin(i) << " ";
  ++j;
  if (j==5)
  {
    j=0;
    cout << endl;
  }
}
j=0;
for (int i=0;i<30;++i)
{
 cout.setf(ios::scientific,ios::floatfield);
  cout << setprecision(5) << setw(12);
  cout << 0.4*std::sin(i) << " ";
  ++j;
  if (j==5)
  {
    j=0;
    cout << endl;
  }
}

namely:

           0  8.41471e+33  9.09297e+33   1.4112e+33 -7.56802e+33 
-9.58924e+33 -2.79415e+33  6.56987e+33  9.89358e+33  4.12118e+33 
-5.44021e+33  -9.9999e+33 -5.36573e+33  4.20167e+33  9.90607e+33 
 6.50288e+33 -2.87903e+33 -9.61397e+33 -7.50987e+33  1.49877e+33 
 9.12945e+33  8.36656e+33 -8.85131e+31  -8.4622e+33 -9.05578e+33 
-1.32352e+33  7.62558e+33  9.56376e+33  2.70906e+33 -6.63634e+33 
 0.00000e+00  3.36588e-01  3.63719e-01  5.64480e-02 -3.02721e-01 
-3.83570e-01 -1.11766e-01  2.62795e-01  3.95743e-01  1.64847e-01 
-2.17608e-01 -3.99996e-01 -2.14629e-01  1.68067e-01  3.96243e-01 
 2.60115e-01 -1.15161e-01 -3.84559e-01 -3.00395e-01  5.99509e-02 
 3.65178e-01  3.34662e-01 -3.54052e-03 -3.38488e-01 -3.62231e-01 
-5.29407e-02  3.05023e-01  3.82550e-01  1.08362e-01 -2.65454e-01