How to turn data between '#' into individual matrices

A fortran model outputs the following text files :

 # 
 # 
 # done by Task#2 of TABOO on 2017.02.23 time=15.15.13
 # Observer  longitude (deg) =   145.0100
 # Observer colatitude (deg) =    27.0000
 # 
 #  Time      u_rad     u_col     u_lon     geoid       mass
 # (kyrs)      (m)       (m)       (m)       (m)        (kg)
 # --------------------------------------------------------------
    0.0000  -82.6972   -0.0055    1.4561   -3.2293    0.00000E+00
    1.0000  -23.2048   -0.0012    0.3021   -1.1764    0.00000E+00
    2.0000   -8.3823   -0.0003    0.0797   -0.5409    0.00000E+00
    3.0000   -3.6872   -0.0001    0.0266   -0.2891    0.00000E+00
    4.0000   -1.9649   -0.0000    0.0114   -0.1737    0.00000E+00
    5.0000   -1.2411   -0.0000    0.0063   -0.1141    0.00000E+00
    6.0000   -0.8946   -0.0000    0.0043   -0.0802    0.00000E+00
    7.0000   -0.7080   -0.0000    0.0034   -0.0594    0.00000E+00
    8.0000   -0.5971   -0.0000    0.0029   -0.0459    0.00000E+00
    9.0000   -0.5257   -0.0000    0.0025   -0.0368    0.00000E+00
   10.0000   -0.4765   -0.0000    0.0023   -0.0304    0.00000E+00
 # 
 # 
 # done by Task#2 of TABOO on 2017.02.23 time=15.15.13
 # Observer  longitude (deg) =   146.2500
 # Observer colatitude (deg) =    27.0000
 # 
 #  Time      u_rad     u_col     u_lon     geoid       mass
 # (kyrs)      (m)       (m)       (m)       (m)        (kg)
 # --------------------------------------------------------------
    0.0000  -82.0875   -0.0129   -2.2170   -3.2158    0.00000E+00
    1.0000  -23.0922   -0.0027   -0.4604   -1.1733    0.00000E+00
    2.0000   -8.3553   -0.0007   -0.1216   -0.5401    0.00000E+00
    3.0000   -3.6796   -0.0002   -0.0406   -0.2888    0.00000E+00
    4.0000   -1.9625   -0.0001   -0.0174   -0.1736    0.00000E+00
    5.0000   -1.2402   -0.0001   -0.0097   -0.1140    0.00000E+00
    6.0000   -0.8941   -0.0000   -0.0066   -0.0802    0.00000E+00
    7.0000   -0.7077   -0.0000   -0.0052   -0.0594    0.00000E+00
    8.0000   -0.5969   -0.0000   -0.0044   -0.0459    0.00000E+00
    9.0000   -0.5255   -0.0000   -0.0039   -0.0368    0.00000E+00
   10.0000   -0.4763   -0.0000   -0.0036   -0.0304    0.00000E+00
 # 
 # 
 # done by Task#2 of TABOO on 2017.02.23 time=15.15.13
 # Observer  longitude (deg) =   147.0000
 # Observer colatitude (deg) =    27.0000

...

I need to write a function which turns each section which does not begin with '#' into its own respective matrix or cell array. dlmread and textscan do not work well.

[per isakson]

• How large is the file compared to the available RAM?
• What about: date, time, longitude and colatitude?
• Is speed important?

Answer 1

[KSSV]

fid = fopen('your txt file') ;
S = textscan(fid,'%s','delimiter','\n') ;
fclose(fid) ;
S = S{1} ;
%%get locations where # not present
idx = strfind(S, '#');
idx = find((cellfun('isempty',idx)));
%%get indices where # breaks
% iwant = S(idx) ;
iwant = cell2mat(cellfun(@str2num,S(idx),'un',0)) ;

[Franklyn Dunbar]

Thanks! I wasn't familiar with the strfind function.

[per isakson]

Try this

fid = fopen('cssm.txt') ;
cac = textscan( fid, '%f%f%f%f%f%f', 'CollectOutput',true, 'CommentStyle','#' );
[~] = fclose(fid);

where cssm.txt contains the data of your question.

And see textscan, Read formatted data from text file or string