Finding the indexes of multiple substrings within a larger string.
2 views (last 30 days)
Show older comments
I’m trying to find the indexes of all two digit pairs in a very long string of numbers, say “c”. I can easily find all occurrences of one string at a time; for example strfind(c, ’00’)…strfind (c, ’01’). But I want a way to do this for all sets one hundred sets; 00 to 99. I tried this:
x=0:99;
dig=sprintf('%02d ',x);
%converts the vector 0to99 into a string with two digits, space between numbers
dub_dig=strsplit(dig);
%splits each pair into cells
dub_dig_str=string(dub_dig);
%converts to a string
How do I get this sequence of strings (dub_dig_str) to work in something like a for loop using the strfind function? When I try this it crashes. I would like to output a matrix of indexes of where each pair occurs, for all pairs.
Thanks
0 Comments
Accepted Answer
Stephen23
on 24 Mar 2023
Edited: Stephen23
on 24 Mar 2023
idx = regexp(c,'\d\d') % no overlaps
idx = regexp(c,'\d(?=\d)') % with overlaps
7 Comments
Stephen23
on 26 Mar 2023
Edited: Stephen23
on 27 Mar 2023
"My goal is to output a separate row of indexes for each pair of numbers (one hundred total, 00to99), stating where each appears in c."
Aaah, so you actually want to compare the pairs against another set with a specific order, which is what you were achieving with the loop. Here is an alternative approach:
c = char(randi(+'09',1,123)) % random data
% Character pairs:
[T,U] = meshgrid('0':'9'); % all pairs
P = cellstr([T(:),U(:)]) % all pairs
Q = cellstr(c([1:end-1;2:end]).'); % data pairs
% Find indices of data pairs:
[~,X] = ismember(Q,P);
% Place indices into cell array:
Y = (1:numel(Q)).';
Z = accumarray(X,Y,[100,1],@(a){a})
Checking the indices of '00' and some random pair:
Z{1}
Z{strcmp(P,'23')}
You can probably do something simiar with table operations. Lets try it now:
D = cell2table(Q, 'VariableNames',"Pair");
D.Index = (1:numel(Q)).';
G = groupsummary(D,"Pair",@(a){a})
More Answers (1)
Walter Roberson
on 24 Mar 2023
c = 'a91bb48353'
mask = ismember(c, '0':'9');
odd_pair = find(mask(1:2:end-1) & mask(2:2:end)) * 2 - 1
even_pair = find(mask(2:2:end-1) & mask(3:2:end)) * 2
pair_starts_at = union(odd_pair, even_pair)
2 Comments
Walter Roberson
on 26 Mar 2023
c = char(randi([0 9], 1, 30) + '0')
C = c - '0';
odds = C(1:2:end-1) * 10 + C(2:2:end);
evens = C(2:2:end-1) * 10 + C(3:2:end);
odd_idx = (1:numel(odds)) * 2 - 1;
even_idx = (1:numel(evens)) * 2;
indices = accumarray([odds(:); evens(:)] + 1, [odd_idx(:); even_idx(:)], [], @(locs){locs});
populated = find(~cellfun(@isempty, indices));
[num2cell(populated-1), indices(populated)]
See Also
Categories
Find more on Characters and Strings in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!