extracting numbers with different decimal extention

Hi! I have a line in my file:

OutputY:  35,40  50  57,80  57  54,71  58,73  61,01  61,67  60,02  57,57  57,51  61,11  64,55  67,74  68,31  66,26  60,52  60,47  62,29  43,38

There some numbers have 2 decimal places and some numbers don't have it at all. When I use:

A1 = regexp(TRGT_a,'[\d*\.]*\d*','match') %TRGT_a is my variable with many columns like OutputY

it sees every number separately and instead of 20 values with decimal extension, I have 38 values... I would appreciate if you could help me to solve this? Also I will have to change all comas to fulstops to do further calculations.

Thank you in advance!

Answer 1

[Stephen23]

>> S = 'OutputY:        35,40   50      57,80   57      54,71   58,73   61,01   61,67   60,02   57,57   57,51   61,11   64,55   67,74   68,31   66,26   60,52   60,47   62,29   43,38';
>> C = regexp(S,'\d+,?\d*','match');
>> V = str2double(strrep(C,',','.'))
V =
   35.400   50.000   57.800   57.000   54.710   58.730   61.010   61.670   60.020   57.570   57.510   61.110   64.550   67.740   68.310   66.260   60.520   60.470   62.290   43.380

[KDRA]

Sorry, I forgot to add that my A1 is 1x64 cell. It gives me this:

Error using strrep Cell elements must be character vectors.

Error in loadTargets (line 53) TRGT_full = str2double(strrep(A1,',','.'))

[KDRA]

This is how my TRGT_a looks like...

and this is a bit more of my code:

if true
  % code
end
for kk=1:nfiles
  idx{kk} = find(contains(s{1,kk}{1,1},'OutputVsFreq_Normal_Target_dB_SPL'));
  idx2 = cell2mat(idx);
  idx3 = idx2(2,:)
  TRGT_a{kk} = s{1,kk}{1,1}{idx3(kk),1};
end
A1 = regexp(TRGT_a,'[\d*\.]*\d*','match')
TRGT_full = str2double(strrep(A1,',','.'))

I understand why this is not working but again, I am not sure how to fix this...

[Stephen23]

>> S{1} = 'OutputY:  35,40  50  57,80  57  54,71  58,73  61,01  61,67  60,02  57,57  57,51  61,11  64,55  67,74  68,31  66,26  60,52  60,47  62,29  43,38';
>> S{2} = 'HelloWorld 12.34 56 7.89';
>> C = regexp(S,'\d+,?\d*','match');
>> F = @(c)str2double(strrep(c,',','.'));
>> V = cellfun(F,C,'uni',0);
>> V{:}
ans =
   35.400   50.000   57.800   57.000   54.710   58.730   61.010   61.670   60.020   57.570   57.510   61.110   64.550   67.740   68.310   66.260   60.520   60.470   62.290   43.380
ans =
   12   34   56    7   89

Note that your regular expression does not do what you think it does. In regular expressions, [] has a special meaning: it does NOT simply group some characters together, as you seem to be trying to do. Also there is no point in specifying to match the period character if your data uses commas.

[KDRA]

This worked! Thanks a lot also for the explanation:) very helpful for learning!

Answer 2

[dpb]

I'm sure there is regexp but I'm just not adept-enough w/o a lot of time and effort...the "dead-ahead" solution--

>> fmt=['Output Y:' repmat('%f',1,sum(s==9))];
>> dat=cell2mat(textscan(strrep(s,',','.'),fmt))
dat =
  Columns 1 through 17
   35.4000   50.0000   57.8000   57.0000   54.7100   58.7300   61.0100   61.6700   60.0200   57.5700   57.5100   61.1100   64.5500   67.7400   68.3100   66.2600   60.5200
  Columns 18 through 20
   60.4700   62.2900   43.3800
>>