Extracting parts of a string

I have a text filewith information like this:

FileName; SampleFreq; Test;Modality;Channel;Description;StimIntensity; Position; RecordingTime
C:\Users\G10040419\Desktop\lp export application\Data 139\00000090_1.WAV; 22000; 2;1;1;5 CH Right;    0.00;   -10000; 40147.491374

I need to extract the sampleFreq (22000) and the position (-10000). I tried to use regular expressions, but I cannot find specific delimiter for these data.

Answer 1

[Paolo]

The following code uses regexp to extract the data you want. You can play around with the expression here .

 data = fileread('00000090Head.txt');
 expression = '(?<=WAV;\s*)(\d*)(?:;\s*\d*;\d;\d;(.*?(?=;));\s*\d*\.\d*;\s*)(-?\d*)';
 [tokens,match] = regexp(data,expression,'tokens','match');

 sampleFrequency = cellfun(@(x) x(1,1),tokens);
 position = cellfun(@(x) x(1,2),tokens);

Position and sampleFrequency are both 1x183 cell arrays and contain the data you are interested in.

position = {'-10000' '-9000' '-8000' '-7000' '-6000' '-5000' '-4500' '-4000' '-3500' '-3000' ................}

sampleFrequency = {'22000' '22000' '22000' '22000' '22000' '22000' '22000' '22000' '22000' '22000' .................}

Answer 2

[per isakson]

Is this what you are looking for?

fid = fopen( '00000090Head.txt', 'r' );
cac = textscan( fid, '%*s%f%*f%*f%*f%*s%*f%f%*f', 'Headerlines',1,'Delimiter',';' );
fclose( fid );

and inspect the result

>> cac
cac =
  1×2 cell array
    {183×1 double}    {183×1 double}

>> cac{2}(1:3)
ans =
      -10000
       -9000
       -8000

Answer 3

[Ana Maria Alzate]

Yes, but is not giving me the position, it is giving me the lst column, the recording time

[per isakson]

Did I make a mistake when counting columns? This format string works here with the sample file. (Do all files have the same format?)

'%*s%f%*f%*f%*f%*s%*f%f%*f'

However, remove one %*f and try

'%*s%f%*f%*f%*f%*s%f%*f'

[Ana Maria Alzate]

I think the problem is that the column kind of shift as you go down in the text.

[per isakson]

Shifts like this one should not be a problem. (This is the only "shift" I find in the sample file.)

• Do have problems reading the uploaded sample file?
• Why not upload a file, which causes problems.
• Do you get any error or warning messages?

[Jan]

@Ana Maria Alzate: Please do not post comments in the section for answers in the future. There is a section for comments for this job. Thanks.

[Ana Maria Alzate]

Thank you for the advice!

Answer 4

[Stephen23]

Importing the data as strings and then using regular expressions to parse them is inefficient, yet is not required because that file is very nicely formatted in delimited columns, and the required data can easily and efficiently be read directly as numeric (or char). The command textscan makes it easy specify how to read those columns, and the format string is much simpler and more intuitive that those regular expressions:

>> fmt = '%*s%f%*d%*d%*d%*s%*f%f%*f';
>> opt = {'HeaderLines',1,'Delimiter',';'};
>> [fid,msg] = fopen('00000090Head.txt','rt');
>> assert(fid>=3,msg)
>> C = textscan(fid,fmt,opt{:});
>> fclose(fid);
>> [C{:}]
ans =
   22000  -10000
   22000   -9000
   22000   -8000
   22000   -7000
   22000   -6000
   22000   -5000
   22000   -4500
   22000   -4000
   22000   -3500
   22000   -3000
   22000   -3000
   22000   -2500
   22000   -2000
   22000   -1500
   22000   -1000
   22000    -500
   22000       0
   22000     500
   22000    1000
   22000    1500
   ... lots of lines here
   22000   -3000
   22000   -2500
   22000   -2000
   22000   -1500
   22000   -1000
   22000    -500
   22000       0
   22000     500
   22000    1000
   22000    1500
   22000    2000
   22000    2500
   22000    3000
   22000    3500
   22000    4000
   22000    4000
   22000    5000