extract a specific line in text files

how are you best community :

I have about 6000 text files each of them has 11 lines as shown next :

COM: IFN/ENIT-database truth (label) file
COM: http://www.ifnenit.com
COM: IfN, TU-BS
COM: ae07_001.tif coming from pa021_0.tif
X_Y: 449 119
BDR: begin data record
LBL: ZIP:1251;AW1:ÇáÔÑÇíÚ;AW2:aaA|laB|shM|raE|aaA|yaB|ayE|;QUA:YB2;ADD:P4
CHA: 7
BLN: 82,78
TLN: 32,62
EDR: end of data record

i want to extract from the seventh line only the text

"aaA|laB|shM|raE|aaA|yaB|ayE|"

( the part i want to extract does not has the same size from file to file )and put all of them in a new array with same size of files numbers .

how can i do it

any answer will be appreciated .

thank you

Answer 1

[Walter Roberson]

Use fileread() to read the file into a single string. Use regexp() to match the portions of it you want. For example

regexp(TheString, '(?<=COM:\s+)\S+(?=\s+coming)', 'match')

[majed majed]

thank you sir :

i did not understand that because my experience in MATLAB is limited , could you give me the code ,or give me more explanation

thank you again and again .

[Walter Roberson]

project_dir = '/path/to/where/the/tru/files/are';
dinfo = dir( fullfile(project_dir, '*.tru') );
for K = 1 : length(dinfo)
   thisfile = fullfile(project_dir, dinfo(K).name);
   filecontent = fileread(thisfile);
   imagename_cell = regexp(filecontent, '(?<=COM:\s+)\S+(?=\s+coming)', 'match');
   lbl_cell = regexp(filecontent, '(?<=LBL:\s+)\S+', 'match');
   lbl_parts = strsplit(lbl_cell{1}, ';');

     image_names{K} = imagename_cell{1};
     image_labels{K} = lbl_parts{1};
     image_texts{K} = lbl_parts{3}(5:end);

     % fprintf('Image "%s" had label "%s" and text "%s"\n', image_names{K}, image_labels{K}, image_texts{K});
  end

The end result will be three cell arrays of strings, image_names, image_labels, and image_texts

Note: extraction of the image name relies upon the word "coming" being present after the name.

[majed majed]

thank you a lot for helping me