Column labels with source names

Hi, i have a file like attached. I'd like to show each source name in each column of results (For example If I calculate mean value). Could you tell me how to do? I know I can use fprintf but don't know how Thanks, regards

[Stephen23]

@Julio Martín: you uploaded a PDF, but a PDF is useless for us. We cannot import PDF data, so we cannot test any code using your PDF. No one will sit and copy out those values by hand, so if you want help then please upload the original text file.

[Jan]

"Each source name in each coumn of results" - but what are the "results"? What is the connection to fprintf? Do you want to pint something to a file?

What do you want to achieve? Please explain this exactly.

[Julio Martín]

Hi, I have already changed the file. I mean when I calculate results (mean, max, etc.). If I have eight channel results, I'd like to show source name Hope I've explained better. Thanks, regards

[Walter Roberson]

You have 8 numeric columns. Should we assume that each column is one channel? At the top you have 8 lines under the source name header: should we assume that those 8 lines are the channel names?

Will the files always have 8 source names and 8 columns?

[Julio Martín]

Hi yes, each column is one channel, and yes, those 8 lines are the channel names Thanks

Answer 1

[Walter Roberson]

filename = 'BES30_1.txt';
nvars = 8;

fid = fopen(filename, 'rt');
if fid < 0; error('Could not open file "%s"', filename); end

failed_step = false;
while true
  thisline = fgetl(fid);
  if ~ischar(thisline); failed_step = true; break; end    %end of file
  if strcmpi(thisline, '[SOURCE NAMES]'); break; end
end
if failed_step; fclose(fid); error('No [SOURCE NAMES] in file'); end

channel_names = cell(1, nvars);
failed_step = false;
for K = 1 : nvars
  thisline = fgetl(fid);
  if ~ischar(thisline); failed_step = true; break; end    %end of file
  if isempty(thisline); failed_step = true; break; end    %unexpected blank line
  channel_names{K} = thisline;
end
if failed_step; fclose(fid); error('Not enough column names in file'); end

failed_step = false;
while true
  thisline = fgetl(fid);
  if ~ischar(thisline); failed_step = true; break; end    %end of file
  if strcmpi(thisline, '[DATA]'); break; end
end
if failed_step; fclose(fid); error('No [DATA] in file'); end

fmt = repmat('%f', 1, nvars);
data = cell2mat( textscan(fid, fmt, 'CollectOutput', 1) );

fclose(fid);

data_mean = mean(data);
data_max = max(data);
data_min = min(data);

channel_name_width = max( cellfun(@length, channel_names) );
name_fmt = repmat( sprintf('%%%ds  ', channel_name_width), 1, nvars );
name_fmt(end-1:end) = '\n';
data_fmt = repmat( sprintf('%%%.dg  ', channel_name_width), 1, nvars );
data_fmt(end-1:end) = '\n';

fprintf(name_fmt, channel_names{:});
fprintf(data_fmt, data_min);
fprintf(data_fmt, data_max);
fprintf(data_fmt, data_mean);

[Julio Martín]

I thought it was easier Thank you!

[Walter Roberson]

If the number of lines before each major section is fixed then this could be made notably shorter. Also, part of the length is in error checking.