Extreact Information from a String

Hello, I do have the following Code

d = dir('/home/user/Documents/Data*.inp');
names = {d.name}

which gives me an Array of Strings.

names(1) is like = "Data_c11_t3.322111_id01111_Units.inp"

names(2) is like = "Data_c101_t4.32111_id01112_Units.inp"

etc.

Now I want to extract the time, which is in the name after "t". How can I extract the time in such a format?

Any suggestions?

Answer 1

[KL]

regexp would be the best idea, I've no big experiences with it, I'll give it a shot anyway,

 names = {'Data_c11_t3.322111_id01111_Units.inp';'Data_c101_t4.32111_id01112_Units.inp'};
 t = regexp(names,'[0-9]+\.+[0-9*]+','match')

[Guillaume]

regexp (regular expressions) is indeed the easiest and best way to solve the problem.

There are a few problems with KL regex. It will extract all numbers in the string not just the one required, the '\.+' part should be just '\.| (only one dot is acceptable), and there's no reason to have a * in the [0-9*].

To be guaranteed to extract the number after the _t, the following regular expression would be much better:

t = regexp(names, (?<=_t)\d+\.\d+', 'match', 'once')

This will extract the expression immediately following '_t' that consists of at least one digit (the \d+'), followed by a dot (the |\.) followed by at least one digit (the second \d+).

There are always many ways to extract something with regular expressions, depending on what you want to allow or disallow. In the examples provided, the following would also work:

t = regexp(names, '(?<=_t)[^_]+', 'match', 'once')

which extract everything following _t up to but not including the next _.

[Jos (10584)]

regular expressions are indeed very powerful, but are, in my humble opinion, way over the top for such a simple problem :)

[Guillaume]

To each their own opinion of course. Regular expressions are designed to extract specific patterns in strings, which, in my opinion is exactly the problem at hand.

[Christian Berwanger]

Ok. This worked for me. Thank you. I do have still a question:

Can you please explain me how this worked? I searched up regexp on Mathworks documentation. But honestly I still don't really understand why this one works.

To Guilaume: It extraced only the time. Even I have more numbers in the name. Thats why I am asking how this works as I don't see any reason to just take the numbers from time.

Answer 2

[Jos (10584)]

Use a sscanf to read the number between the _t and _id, ignoring the varying number after the _c. Use cellfun to apply this to all cells:

names = {'Data_c11_t3.322111_id01111_Units.inp' ; 'Data_c101_t4.32111_id01112_Units.inp'} ;
T = cellfun(@(N) sscanf(N,'Data_c%*d_t%f_id'), names) % "%*d" = the * means skip

[Stephen23]

+1 a very neat solution.

[Coco Newton]

@Jos, this is a great answer

How would I adapt the sscanf formatSpec to extract 130 as a numeric value from this string? I have attached the string array variable that this is example string is taken from.

'Position: (19.59862|1.8|53.84677)Rotation: (0 | 130 | 0)'

Many thanks!

[Jos (10584)]

Thanks :-) Something along these lines should work for your input

str = 'Position: (19.59862|1.8|53.84677)Rotation: (0 | 130 | 0)'
A = sscanf(str, 'Position: (%*f|%*f|%*f)Rotation: (%*d | %d | %*d)')

Answer 3

[KSSV]

str = 'Data_c11_t3.322111_id01111_Units.inp'  ;
 idx = strfind(str,'_') ;
 str2num(str(idx(2)+2:idx(3)-1))

Using regexp:

str = 'Data_c11_t3.322111_id01111_Units.inp'  ;
N = str2num( regexprep( str, {'\D*([\d\.]+\d)[^\d]*', '[^\d\.]*'}, {'$1 ', ' '} ) ) ;
t = N(2)

[Christian Berwanger]

Now I get the following error: Index exceeds matrix dimensions in the line:

time = str2num(fileName(idx(2)+2:idx(3)-1));

Where

fileName = names(i);

[KSSV]

What is filename? Can you copy it?

[Christian Berwanger]

I do have it in a for loop. its like

for i = 1:length(names)
  fileName = names(i);
  idx = strfind(fileName,'_') ;
  time = str2DOUBLLE(fileName(idx(2)+2:idx(3)-1));
end

so filename is in the first iteration "Data_c11_t3.322111_id01111_Units.inp"

in the second iteration: "Data_c101_t4.32111_id01112_Units.inp"

and so on.