reading out a particular column with a header name from a large text file with multiple rows and columns. of different dimensions. Attached the .txt file.

I couldnot find a solution for reading out a particular column with a header name in text file with multiple row and columns of different dimensions. Attached the .txt file for reference. The intention is to read out all the numbers under the column header name N482.

I would appreciate the help!

Smruti

Answer 1

[Mathieu NOE]

hello

I am not sure if you ask for the data in the N482 rows (and not column) as we can see in the first 49,407 lines of your data file

or if we are talking about some columns in the very lower end of the data file where N482 also sometimes appear but seems asociated with many different variables

in case the good answer is correponding to the first scenario, here a little code that will retrieve the data :

clc
clearvars

%%% main code %%%
[ data_out ]  = retrieve_data('PerformanceResults.txt');

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function [ data_out ] = retrieve_data(Filename)

fid = fopen(Filename);
tline = fgetl(fid);
% initialization
k = 0;
p = 0;
data_line = [];

while ischar(tline)
    
    k = k+1;    % loop over line index
    tmp1 = contains(tline,'N482:'); 
    if tmp1  &&   ~isempty(tline) 
        p = p+1;
        ind1 = strfind(tline,':');
        ind2 = strfind(tline,'[');
        data_out(p,1) = str2double(tline(ind1+1:ind2-1));    % 
    end
    tline = fgetl(fid);
end

fclose(fid);
end

[smruti sahoo]

Hello Mathieu,

This is really a prompt reply. My bad, my question was not clear. I was referring to the data in the very lower part of the text file. Would I get help for that as well?

Regards,

Smruti

[Mathieu NOE]

Hello Smruti

I will do my best to further help you . Do you need all occurences of columns where N482 appears or only some specific ones ? can you do a list of which variabes you're interested in ?

[smruti sahoo]

Hello Mathieu,

I need to read data from the row 49428 untill 49574 for "N481 " and "PWX481". It would be great if the data column also accompanies the respective subset as mentioned between the tags such as "Top of Climb" , "Cruise" etc. .

Best Regards,

Smruti

[Mathieu NOE]

hello

ok but I still have a few questions , like

N481 , no more N482 ?

when i search for N481 in the specified rows , I can find it under different variables / units like :

EF1.Ncrel

N481

[-]

________________________

0.1000000000000000E+01

0.9162897600284080E+00

0.7497359606957799E+00

0.8814135438000129E+00

EFAN1Shaft.N

N481

[rpm]

_______________________

0.8000000000000000E+04

0.7168221667751842E+04

0.6176755811549974E+04

0.7745871554961489E+04

EF1.Ncrel

N481

[-]

_______________________

0.1000000000000000E+01

0.9162897600284080E+00

0.7497359606957799E+00

0.8814135438000129E+00

EFAN1Shaft.N

N481

[rpm]

________________________

0.8000000000000000E+04

0.7168221667751842E+04

0.6176755811549974E+04

0.7745871554961489E+04

EF1.Ncrel

N481

[-]

_______________________

0.1000000000000000E+01

0.9162897600284080E+00

0.7497359606957799E+00

0.8814135438000129E+00

EFAN1Shaft.N

N481

[rpm]

______________________

0.8000000000000000E+04

0.7168221667751842E+04

0.6176755811549974E+04

0.7745871554961489E+04

EF1.Ncrel

N481

[-]

_______________________

0.1000000000000000E+01

0.9162897600284080E+00

0.7497359606957799E+00

0.8814135438000129E+00

and so on...

how do I have to understand how we organize the retrieval of data ? we cannot simply put together what is different topic

for PWX481 same question

Answer 2

[smruti sahoo]

Hello,

N481 and N482 have the same value per se. As you rightly mentioned I would be intrested in extarcting the real value of N481 which is folowed by "rpm". The PWX481 with 'W' as in watt is desired to be retrieved. I wish to save the retrived data in matrix for further compilation. So that I can read the particular column.

thanks,

Smruti

[Mathieu NOE]

After some good sweat , I believe I finally managed to get it work !

at the end all datas are concatenated together - see the excel output file (attached)

code :

clc
clearvars

% initialization
start_line = 49421 ;
last_line = start_line;
string_out_all = [];
data_out_all = [];
EOF = 0;

%%% main code %%%
while 1
    if EOF == 1
        break
    end
    [ string_out,data_out,last_line,EOF ]  = retrieve_data('PerformanceResults.txt',max(start_line,last_line));
    string_out_all = [string_out_all; string_out]; % string concatenation all blocks
    data_out_all = [data_out_all; data_out];  % data concatenation all blocks
   
end

% export to excel
out = [{'COND' ' N481 [RPM]' 'PWX481 [W]'}; string_out_all num2cell(data_out_all)];
writecell(out,'results.xlsx');

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function [ string_out,data_out,last_line,EOF ] = retrieve_data(Filename,start_line)
EOF = 0;
fid = fopen(Filename);
tline = fgetl(fid);
% initialization
k = 0;
p = 0;
rpm_line = start_line ;
rpm_detect_line = [];
ind_rpm = [];

while ischar(tline)
    
    k = k+1;    % loop over line index
    
    if k>=start_line  &&  ~isempty(tline)
        
        if contains(tline,'N481');  
            rpm_detect_line = k+1;
            ind_N481 = strfind(tline,'N481');
            ind_PWX481 = strfind(tline,'PWX481');
        end
        
        if ~isempty(rpm_detect_line) && k  == rpm_detect_line
            rpm_line = k;
            ind_rpm = strfind(tline,'[rpm]');
            ind_w = strfind(tline,'[W]');
            
        end

        if ~isempty(ind_rpm) && k  >= rpm_line + 3  
            if contains(tline,'END FILE')
                EOF = 1;
                last_line = k;
                break
            end
            
            if contains(tline,'________')  
                last_line = k;
                break
            end
            
            p = p+1;
            ind_str = strfind(tline,'"');
            string_out{p,1} = tline(ind_str(1)+1:ind_str(2)-1);
            
            % N481  / RPM : find closest value between ind_N481 and ind_rpm
            dist = abs(ind_N481'-ind_rpm);
            [~,ind] = min(dist,[],1);
            ind2 = ind_N481(ind(1));
            data_out1(p,1) = str2double(tline(ind2-20:ind2+5));    % 

            % PWX481 / W  : find closest value between ind_PWX481 and ind_w
            dist = abs(ind_PWX481'-ind_w);
            [~,ind] = min(dist,[],1);
            ind2 = ind_PWX481(ind(1));
            data_out2(p,1) = str2double(tline(ind2-20:ind2+5));
            
        end
    end
    
    tline = fgetl(fid); 
    
end

fclose(fid);

data_out = [data_out1 data_out2];

end

[smruti sahoo]

Hello Mathieu,

That was a great work, To say thank you would be less!

Best regards,

Smruti

[Mathieu NOE]

My pleasure !