How to read some data from a text file

Hello,

I have a text file like this:

#Parameters = {a=12.4; b=0.6; d=18; f_design=10; f_max=15; f_min=5; g=1; h=7.5; lambda=30; phi=0; react=-10000; theta=0}
#"Frequency / GHz"	"S1(1),1(1) (react=-10000) [Re]"	"S1(1),1(1) (react=-10000) [Im]"	"Ref.Imp. [Re]"	"Ref.Imp. [Im]"
#-------------------------------------------------------------------------------------------------------------------
10.000000000000	0.14429999628736	0.016848571131779	73.000000000000	0.00000000000000
#Parameters = {a=12.4; b=0.6; d=18; f_design=10; f_max=15; f_min=5; g=1; h=7.5; lambda=30; phi=0; react=-8000; theta=0}
#"Frequency / GHz"	"S1(1),1(1) (react=-8000) [Re]"	"S1(1),1(1) (react=-8000) [Im]"	"Ref.Imp. [Re]"	"Ref.Imp. [Im]"
#-----------------------------------------------------------------------------------------------------------------
10.000000000000	0.14004838730771	0.019270257352017	73.000000000000	0.00000000000000

I would like to read just the data that does not start with a " # ", therefore I would like to read:

10.000000000000 0.14429999628736 0.016848571131779 73.000000000000 0.00000000000000

10.000000000000 0.14004838730771 0.019270257352017 73.000000000000 0.00000000000000

How can I do that? because I cannot use the option of jumping the header due to the fact that those three lines that start with a # are repeated along all the file before the data I am interested in.

Thank you so much in advance,

Diego.

Answer 1

[Star Strider]

Use the textscan function:

fidi = fopen('TestFile20200709.txt','rt')
Datac = textscan(fidi, '%f%f%f%f%f', 'CommentStyle','#', 'CollectOutput',1)
fclose(fidi)
Data = cell2mat(Datac)

producing:

Data = 
10.000000000000000	 0.144299996287360	 0.016848571131779	73.000000000000000	 0.000000000000000
10.000000000000000	 0.140048387307710	 0.019270257352017	73.000000000000000	 0.000000000000000

Defining the lines that begin with ‘#’ as comments tells textscan to ignore them.

I have also attached the file I created to test this code.

.

[Diego Bermúdez Martín]

Thank you so much, this is what I was trying to do and I did not get it ^^

[Star Strider]

As always, my pleasure!

[Arthur Roué]

Much more elegant than my solution, didn't know we can do that with textscan !

[Star Strider]

I learned much about MATLAB by participating in Answers, some of which involved my learning the many options textscan offers. I doubt I would have discovered them otherwise.

Answer 2

[Arthur Roué]

Here one way to do this :

% File path
FilePath = fullfile(pwd, 'MyFile.txt');

% Open the file
FileID = fopen(FilePath, 'r');
if FileID == -1
    % End of function because file couldn't be opened
    error('File "%s" couldn''t be opened.\n', FilePath)
end

% Read all the lines
cLines = cell(1e4,1);       % Pre-allocation
strLineInFile = fgetl(FileID);    % 1st line
idxLine = 1;
while ischar(strLineInFile)
    cLines{idxLine} = strLineInFile;
    % Get next line
    strLineInFile = fgetl(FileID);
    idxLine = idxLine + 1;
end
fclose(FileID);

% Shorten the pre-allocated cell
nbLines = find(~cellfun(@ischar, cLines), 1, 'first') - 1;
cLines = cLines(1:nbLines);

% Detect line starting with '#'
vb = ~cellfun(@(line) strncmp(line, '#', 1), cLines);

% Convert data to numeric value
Values = cLines(vb);
% Values are separated with horizontal tab (\t)
Values = cellfun(@(line) str2double(strsplit(line, '\t')), Values, 'UniformOutput', 0);
Values = vertcat(Values{:});

Be aware that :

- any line starting with anythig but # will be interpret as data (and will be convert to numeric value)

- your separator on a line between value is a tab

If you change you text format, you'll have to adapt this script.

[Diego Bermúdez Martín]

Thank you so much ^^