%% Average and decimate regularly scattered data with fixed defined step
% Prompt user for location of input file
% Prompt user for step size for decimation of the averaged data
% Group Y scattered data (only 1 Y column in a 2 column data set with X as first and Y as second column) in chunck and average it and decimate it according to the step size of decimation of X
% Minimum and maximum are defined according to extremas in the X data and according to the decimation step size
% Standard deviation is calculated for each step adn stored in additional column
% A new result file is generated and saved in the same folder as the input file
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% Prompt user for input file
[file,path,indx] = uigetfile('*.txt');
if isequal(file,0)
   disp('Error: the data treatment has been terminated.')
   return;
end

% get file name
filename = [path file];
[fileID,errmsg] = fopen(filename);
if isempty(errmsg)
   disp('Input file has been loaded correctly.')
else
   disp('Error: the data treatment has been terminated.')
   return;
end

% Import data
cdata = readtable(filename);
fclose(fileID);

% Check that there is at least 2 columns (because only the 2 first columns
% are used for data treatment
if length(cdata.Properties.VariableNames) < 2
    disp('Error: input file data is wrong. The data treatment has been terminated.')
   return;
end

% Check that data is longer than 1 row
if length(cdata{:,1}) > 1 && length(cdata{:,2}) > 1
   disp('Input data is valid.')
else
   disp('Error: wrong data. The data treatment has been terminated.')
   return;
end

% Prompt user for decimation step size
prompt = {'Enter a value of decimation step size'};
dlgtitle = 'Delta';
definput = {'1'};
opts.Interpreter = 'tex';
answer = inputdlg(prompt,dlgtitle,[1 40],definput,opts);
delta = str2double(answer); % convert answer into numerical

% Get first point in X data and last point in X data
first = min(cdata{:,1});
last = max(cdata{:,1});
distance = last - first;

% Check that the step size is not zero and that it is not larger than the
% distance between the first and last point
if delta > 0 && delta < distance
   disp('Step size for decimation is valid.')
else
   disp('Error: wrong step size for decimation. The data treatment has been terminated.')
   return
end

% round to the closest first data point or round lower according to
% decimation step size
starting_point = delta*floor(first/delta);
% do the same for the last data point
ending_point = delta*ceil(last/delta);
% Because of numerical truncation, there might be one step too many at the
% beginning and / or at the end. But this step will be empty or NaN

X_output = (starting_point:delta:ending_point)';
Y_output = NaN(length(X_output),1);
std_dev_output = NaN(length(X_output),1);

for i = 1:length(X_output)
% for each step in X_output, get the average of the Y data values
% at distance delta/2 from the current X_output.
   selection_array = (cdata{:,1} >= X_output(i)-delta/2).*(cdata{:,1} <= X_output(i)+delta/2); % have to fulfill 2 conditions with ".*"
   sub_table=(cdata(logical(selection_array),2));
   Y_output(i) = mean(sub_table{:,1});
   std_dev_output(i) = std(sub_table{:,1});
end

table_output = table(X_output,Y_output,std_dev_output); % Arrange all data in a table
table_output.Properties.VariableNames([1 2]) = cdata.Properties.VariableNames([1 2]); % Rename table header with header of original input file

% Write table data into text file in same folder as input file
new_filename = [path 'avrg_decimated_std_dev_' file];
writetable(table_output,new_filename,'Delimiter','tab');

disp('Result file has been created in the same file as the input file.');