How to extract numeric data between string lines?

Hi MATLAB Community

I'm trying to solve this problem, which for sure is not new, but I haven't been able to find a proper solution.

I have a file with several headlines, and then a lot of information in the following way:

Binning n: 1, "De19 ", Event #: 150, Primary(s) weight 1.0000E+00

Number of hit cells: 0

Binning n: 1, "De19 ", Event #: 151, Primary(s) weight 1.0000E+00

Number of hit cells: 1

1 7.185244612628594E-05

Binning n: 1, "De19 ", Event #: 152, Primary(s) weight 1.0000E+00

Number of hit cells: 0

Binning n: 1, "De19 ", Event #: 153, Primary(s) weight 1.0000E+00

Number of hit cells: 0

As shown, sometimes after the "Number of hit cells" line, there are numbers. I would like to extract them in a matrix or array. Is there a way to do this?

I attached an example file, that usually contains a lot more of data, that I erased for weight questions.

Thank you very much in advance

Answer 1

[Stephen23]

str = fileread('02-2021-Clearance-Box005_fort72.txt');
rgx = '(?<=Number of hit cells:\s+\d+\s+)(\d+[^\n]*)';
tmp = regexp(str,rgx,'match')

tmp = 1x2 cell array
    {'1  7.185244612628594E-05'}    {'1  2.547905314713717E-04'}

vec = cellfun(@(s)sscanf(s,'%f',[1,Inf]),tmp,'uni',0) % convert to numeric

vec = 1x2 cell array
    {1×2 double}    {1×2 double}

mat = vertcat(vec{:}) % optional merge into one numeric matrix

mat = 2×2

1 7.1852e-05 1 0.00025479

[Federico Geser]

Hi Stephen!

Thank you for your answer!

Maybe I wasn't clear at my post, I'm very sorry for that!

I was aiming at the numbers in the next line, below the "Number of hit cells" line.

Thank you again!

[Stephen23]

@Federico Geser: see my edited answer.

[Federico Geser]

Hi Stephen!

I think it works, but the test file has 12 MB of info to filter, so it might take a while. I don't know if this will work when I get the real results (that may weight ca. 100 MB).

Nevertheless, very helpful solution! Thank you!

[Stephen23]

If there are always exactly two numbers on each of those lines, then this is probably more efficient:

str = fileread('02-2021-Clearance-Box005_fort72.txt');
rgx = '(?<=Number of hit cells:\s+\d+\s+)(\d+[^\n]*)'; % unchanged
tmp = regexp(str,rgx,'match');                         % unchanged
mat = sscanf(sprintf(' %s',tmp{:}),'%f',[2,Inf]).'

mat = 2×2

1 7.1852e-05 1 0.00025479