Find line containing word in a mixed format txt file

I have a FORTRAN code that generate a mixed format output text file of various lengths (depending on code run parameters). I am seeking to write a Matlab script that finds a table located somewhere in that text file, which will always contain the same number of columns and the same column variables.

The data might look something like this (this is representative of the formatting)

...

xourbwofigefifhtiryefsldtldcrpewicbqocyttbwloiopauyveapmwvkylxepftamjocccgpnybtubzhnqqnacyihnyhdxcvshiyusemrerqceebxylcqtpgksivchcrsxbcnggypsysnjdwtkbdzptffyvsesvqwpsnzhkphrftuykjogzpyqhqmuhenqulupujukqsyfqrxmtzxomaojjerczyrqmqhyoihhcoixxkyouqxumkefltqsuraaapoxourbwofigefifhtiryefsldtldcrpewicbqocyttbwloiopauyveapmwvkylxepftamjocccgpnybtubzhnqqnacyihnyhdxcvshiyusemrerqceebxylcqtpgksivchcrsxbcnggypsysnjdwtkbdzptffyvsesvqwpsnzhkphrftuykjogzpyqhqmuhenqulupujukqsyfqrxmtzxomaojjerczyrqmqhyoihhcoixxkyouqxumkefltqsuraaapo

$$$$$$$$$$$$$$$$$$$$$$$$$$$ output $$$$$$$$$$$$$$$$$$$$$$$$$$$

A B C D E F G H I J K

------ --------- ------ ------------------------------------------------------------------------------ ---------------------------------- --------------

s m kg A K mol

2000.0 7.744E-02 5.000 3.098E+01 3.098E+01 0.000E+00 0.000E+00 3.098E+01 3.87194E-01 3.87194E-01 1.000E+00

2005.0 4.646E-02 4.988 1.868E+01 1.868E+01 0.000E+00 0.000E+00 1.868E+01 6.19481E-01 6.19481E-01 1.000E+00

2010.0 6.827E-02 4.975 2.758E+01 2.758E+01 0.000E+00 0.000E+00 2.758E+01 9.60817E-01 9.60817E-01 1.000E+00

2015.0 1.038E-01 4.963 4.213E+01 4.213E+01 0.000E+00 0.000E+00 4.213E+01 1.47961E+00 1.47961E+00 1.000E+00

2020.0 9.099E-02 4.950 3.713E+01 3.713E+01 0.000E+00 0.000E+00 3.713E+01 1.93456E+00 1.93456E+00 1.000E+00

2025.0 9.283E-02 4.938 3.806E+01 3.806E+01 0.000E+00 0.000E+00 3.806E+01 2.39869E+00 2.39869E+00 1.000E+00

2030.0 5.814E-02 4.926 2.396E+01 2.396E+01 0.000E+00 0.000E+00 2.396E+01 2.68937E+00 2.68937E+00 1.000E+00

(table might continue an arbitrary number of rows down)

....

xourbwofigefifhtiryefsldtldcrpewicbqocyttbwloiopauyveapmwvkylxepftamjocccgpnybtubzhnqqnacyihnyhdxcvshiyusemrerqceebxylcqtpgksivchcrsxbcnggypsysnjdwtkbdzptffyvsesvqwpsnzhkphrftuykjogzpyqhqmuhenqulupujukqsyfqrxmtzxomaojjerczyrqmqhyoihhcoixxkyouqxumkefltqsuraaapoxourbwofigefifhtiryefsldtldcrpewicbqocyttbwloiopauyveapmwvkylxepftamjocccgpnybtubzhnqqnacyihnyhdxcvshiyusemrerqceebxylcqtpgksivchcrsxbcnggypsysnjdwtkbdzptffyvsesvqwpsnzhkphrftuykjogzpyqhqmuhenqulupujukqsyfqrxmtzxomaojjerczyrqmqhyoihhcoixxkyouqxumkefltqsuraaapo

....

[Dyuman Joshi]

Please give an example of what the data looks like, and how does the table (that is to be found) looks like.

Even better if you can attach a sample data.

[Matthew]

Dyuman, I've edited my inquiry to include representative txt file data with table format.

Answer 1

[Star Strider]

I am not certrain how the table actually exists in the file (and it would help significantly to have the actual file rather than an imitation of it to work with). That aside, for FORTRAN files, using readtable with fixedWidthImportOptions woud likely work. You will probably need to experiment.

[Matthew]

I've attached the txt file, and I'm looking to extract this table located about midway through:

[Star Strider]

This seems to work —

% type('file.txt')                                        % Examine File (Optional)

VT = cellstr(repmat("double", 1, 11));                  % 'VariableTypes' Cell Array
opts = fixedWidthImportOptions('NumVariables',11,  'VariableWidths',[8 11 9 10 11 11 11 11 13 13 13], 'DataLines',223, 'VariableTypes',{VT{:}}, 'VariableNamesLine',220, 'VariableUnitsLine',222);

T1 = readtable('file.txt', opts)

T1 = 668×11 table
     m        rip       lam     ripl     riapl    riahl    rial     ril      riapi       rii      taua
    ____    _______    _____    _____    _____    _____    ____    _____    _______    _______    ____

    2000    0.07744        5    30.98    30.98      0       0      30.98    0.38719    0.38719     1  
    2005    0.04646    4.988    18.68    18.68      0       0      18.68    0.61948    0.61948     1  
    2010    0.06827    4.975    27.58    27.58      0       0      27.58    0.96082    0.96082     1  
    2015     0.1038    4.963    42.13    42.13      0       0      42.13     1.4796     1.4796     1  
    2020    0.09099     4.95    37.13    37.13      0       0      37.13     1.9346     1.9346     1  
    2025    0.09283    4.938    38.06    38.06      0       0      38.06     2.3987     2.3987     1  
    2030    0.05814    4.926    23.96    23.96      0       0      23.96     2.6894     2.6894     1  
    2035    0.04384    4.914    18.15    18.15      0       0      18.15     2.9085     2.9085     1  
    2040     0.1015    4.902    42.24    42.24      0       0      42.24     3.4161     3.4161     1  
    2045     0.0615     4.89    25.72    25.72      0       0      25.72     3.7236     3.7236     1  
    2050    0.07926    4.878    33.31    33.31      0       0      33.31     4.1199     4.1199     1  
    2055    0.03941    4.866    16.64    16.64      0       0      16.64     4.3169     4.3169     1  
    2060    0.05896    4.854    25.02    25.02      0       0      25.02     4.6117     4.6117     1  
    2065     0.1107    4.843    47.22    47.22      0       0      47.22     5.1654     5.1654     1  
    2070     0.0512    4.831    21.94    21.94      0       0      21.94     5.4214     5.4214     1  
    2075    0.05672    4.819    24.42    24.42      0       0      24.42      5.705      5.705     1  

VN = T1.Properties.VariableNames;

figure
plot(T1.m, real(T1{:,2:end}), '-')
hold on
% plot(T1.m, imag(T1{:,2:end}), '--')
hold off
grid
xlabel(VN{1})
set(gca, 'YScale','log')
legend(VN{2:end}, 'Location','bestoutside')

For fun (and to get some idea of what is in the file), I plotted it as well.

.

[Voss]

T1 has 668 rows, but the table in the file has 268 rows.

VT = cellstr(repmat("double", 1, 11));                  % 'VariableTypes' Cell Array
opts = fixedWidthImportOptions('NumVariables',11,  'VariableWidths',[8 11 9 10 11 11 11 11 13 13 13], 'DataLines',223, 'VariableTypes',{VT{:}}, 'VariableNamesLine',220, 'VariableUnitsLine',222);

T1 = readtable('file.txt', opts);

size(T1)

ans = 1×2

668 11

T1(264:273,:) % mostly NaNs after row 268

ans = 10×11 table
     m        rip       lam     ripl      riapl      riahl    rial     ril     riapi      rii      taua
    ____    _______    _____    _____    ________    _____    ____    _____    ______    ______    ____

    3315    0.07032    3.017    77.28    77.28+0i       0       0     77.28    71.353    71.353      1 
    3320     0.1078    3.012    118.8    118.8+0i       0       0     118.8    71.892    71.892      1 
    3325     0.1412    3.008    156.1    156.1+0i       0       0     156.1    72.599    72.599      1 
    3330    0.08589    3.003    95.24    95.24+0i       0       0     95.24    73.028    73.028      1 
    3335     0.1272    2.999    141.5    141.5+0i       0       0     141.5    73.664    73.664      1 
     NaN        NaN      NaN      NaN        0+1i     NaN     NaN       NaN       NaN       NaN    NaN 
     NaN        NaN      NaN      NaN      NaN+0i     NaN     NaN       NaN       NaN       NaN    NaN 
     NaN        NaN      NaN      NaN      NaN+0i     NaN     NaN       NaN       NaN       NaN    NaN 
     NaN        NaN      NaN      NaN      NaN+0i     NaN     NaN       NaN       NaN       NaN    NaN 
     NaN        NaN      NaN      NaN      NaN+0i     NaN     NaN       NaN       NaN       NaN    NaN 

[Matthew]

This works thank you!

[Voss]

@Matthew: It's not a problem for you that the table T1 contains 400 extra rows relative to the table in the file? See my answer for a method that produces a table without any extra rows.

[Star Strider]

@Matthew — As always, my pleasure!

Answer 2

[Mathieu NOE]

hello

you could do this

out = readcell('Doc1.txt');
eof = size(out,1);
ind1 = find(contains(out,'output'));

% extract valid portion of cell array
out = out(ind1+1:eof-1,:)
header = (split(out(1,:)))'
units = (split(out(3,:)))'
values = str2num(char(out(4:end,:)))

[Mathieu NOE]

this is the result in my command window , based on the attached txt file (copy paste from your post)

out =

10×1 cell array

{'A B C D E F G H I J K' }

{'------ --------- ------ ------------------------------------------------------------------------------ ---------------------------------- --------------' }

{'s m kg A K mol'}

{'2000.0 7.744E-02 5.000 3.098E+01 3.098E+01 0.000E+00 0.000E+00 3.098E+01 3.87194E-01 3.87194E-01 1.000E+00' }

{'2005.0 4.646E-02 4.988 1.868E+01 1.868E+01 0.000E+00 0.000E+00 1.868E+01 6.19481E-01 6.19481E-01 1.000E+00' }

{'2010.0 6.827E-02 4.975 2.758E+01 2.758E+01 0.000E+00 0.000E+00 2.758E+01 9.60817E-01 9.60817E-01 1.000E+00' }

{'2015.0 1.038E-01 4.963 4.213E+01 4.213E+01 0.000E+00 0.000E+00 4.213E+01 1.47961E+00 1.47961E+00 1.000E+00' }

{'2020.0 9.099E-02 4.950 3.713E+01 3.713E+01 0.000E+00 0.000E+00 3.713E+01 1.93456E+00 1.93456E+00 1.000E+00' }

{'2025.0 9.283E-02 4.938 3.806E+01 3.806E+01 0.000E+00 0.000E+00 3.806E+01 2.39869E+00 2.39869E+00 1.000E+00' }

{'2030.0 5.814E-02 4.926 2.396E+01 2.396E+01 0.000E+00 0.000E+00 2.396E+01 2.68937E+00 2.68937E+00 1.000E+00' }

header =

1×11 cell array

Columns 1 through 10

{'A'} {'B'} {'C'} {'D'} {'E'} {'F'} {'G'} {'H'} {'I'} {'J'}

Column 11

{'K'}

units =

1×6 cell array

{'s'} {'m'} {'kg'} {'A'} {'K'} {'mol'}

values =

1.0e+03 *

Columns 1 through 9

2.0000 0.0001 0.0050 0.0310 0.0310 0 0 0.0310 0.0004

2.0050 0.0000 0.0050 0.0187 0.0187 0 0 0.0187 0.0006

2.0100 0.0001 0.0050 0.0276 0.0276 0 0 0.0276 0.0010

2.0150 0.0001 0.0050 0.0421 0.0421 0 0 0.0421 0.0015

2.0200 0.0001 0.0050 0.0371 0.0371 0 0 0.0371 0.0019

2.0250 0.0001 0.0049 0.0381 0.0381 0 0 0.0381 0.0024

2.0300 0.0001 0.0049 0.0240 0.0240 0 0 0.0240 0.0027

Columns 10 through 11

0.0004 0.0010

0.0006 0.0010

0.0010 0.0010

0.0015 0.0010

0.0019 0.0010

0.0024 0.0010

0.0027 0.0010

[Mathieu NOE]

someone very clever may find a readtable based solution, but you may have to play with a lot of parameters to get it to work. I preferred to load everything in a cell array and then do a bit of uncomplicated post processing , but it's personnal opinion.

[Matthew]

Mathieu,

When I try to run this, I get "Error using contains; First argument must be text." Do you have any advice?

Thanks.

[Mathieu NOE]

hello Matthew

maybe I should work with your data file and not something I generated myself

can you share one file ?

[Matthew]

Hello Mathieu,

I have uploaded my file. I am looking to extract this table data, located about midway through the txt file.

Can you help?

[Mathieu NOE]

hello again

I have a one line code solution for you (with the help of the attached function, of course)

[outdata,head] = readclm('file.txt',11,219);

outdata 268x11 double

head = 4×119 char array

' m rip lam ripl riapl riahl rial ril riapi rii taua '

' ------ --------- ------ ----------------------------------------------------- ------------------------ ---------'

' m m/J/m A m/J/A m/J t '

' '

[Mathieu NOE]

if you want to access the individual labels and units from head , you can do that

header = (split(strtrim(head(1,:))))'
units = (split(strtrim(head(3,:))))'

header = 1×11 cell array

{'m'} {'rip'} {'lam'} {'ripl'} {'riapl'} {'riahl'} {'rial'} {'ril'} {'riapi'} {'rii'} {'taua'}

units = 1×6 cell array

{'m'} {'m/J/m'} {'A'} {'m/J/A'} {'m/J'} {'t'}

Answer 3

[Voss]

filename = 'file.txt';

fid = fopen(filename,'r');
data = fread(fid,[1 Inf],'*char');
fclose(fid);

idx = strfind(data,'$ output $');
idx = idx(1);

n = 1 + nnz(data(1:idx) == newline());
T = readtable(filename,'NumHeaderLines',n+2);
T(any(isnan(T{:,:}),2),:) = [];

head(T)

     m        rip       lam     ripl     riapl    riahl    rial     ril      riapi       rii      taua
    ____    _______    _____    _____    _____    _____    ____    _____    _______    _______    ____

    2000    0.07744        5    30.98    30.98      0       0      30.98    0.38719    0.38719     1  
    2005    0.04646    4.988    18.68    18.68      0       0      18.68    0.61948    0.61948     1  
    2010    0.06827    4.975    27.58    27.58      0       0      27.58    0.96082    0.96082     1  
    2015     0.1038    4.963    42.13    42.13      0       0      42.13     1.4796     1.4796     1  
    2020    0.09099     4.95    37.13    37.13      0       0      37.13     1.9346     1.9346     1  
    2025    0.09283    4.938    38.06    38.06      0       0      38.06     2.3987     2.3987     1  
    2030    0.05814    4.926    23.96    23.96      0       0      23.96     2.6894     2.6894     1  
    2035    0.04384    4.914    18.15    18.15      0       0      18.15     2.9085     2.9085     1  

tail(T)

     m        rip       lam     ripl     riapl    riahl    rial     ril     riapi      rii      taua
    ____    _______    _____    _____    _____    _____    ____    _____    ______    ______    ____

    3300     0.1136     3.03    123.7    123.7      0       0      123.7    70.035    70.035     1  
    3305    0.07579    3.026    82.79    82.79      0       0      82.79    70.414    70.414     1  
    3310     0.1176    3.021    128.9    128.9      0       0      128.9    71.002    71.002     1  
    3315    0.07032    3.017    77.28    77.28      0       0      77.28    71.353    71.353     1  
    3320     0.1078    3.012    118.8    118.8      0       0      118.8    71.892    71.892     1  
    3325     0.1412    3.008    156.1    156.1      0       0      156.1    72.599    72.599     1  
    3330    0.08589    3.003    95.24    95.24      0       0      95.24    73.028    73.028     1  
    3335     0.1272    2.999    141.5    141.5      0       0      141.5    73.664    73.664     1  

[Matthew]

Voss, thank you, this also works for me!