Best solution to finding repeating characters on a line.

I am looking for any instances of two characters (e/d) being repeated in a row greater then or equal to 10. I just want to either print every line that this occurs to the command line or stop and print the location of the stop everytime it is detected. Basically I am trying to find when e and d show up over ten times grouped together in a large data file. For example:

asdfsdfsdfsasdfsdfsdfsasdfsdfsdfs

asseefadfefeeedddeeedddasdfsdf

asdfsdfsdfsasdfsdfsdfsasdfsdfsdfs

asseefadfefeeedddeeedddasdfsdf

The script would then print out line 2 and line 4 in the command line.

Thank you for your help

[Rena Berman]

(Answers Dev) Restored edit

Answer 1

[Stephen23]

inp = {'asdfsdfsdfsasdfsdfsdfsasdfsdfsdfs';'asseefadfefaaadddaaadddasdfsdf';'asdfsdfsdfsasdfsdfsdfsasdfsdfsdfs';'asseefadfefaaadddaaadddasdfsdf'};
rgx = '(.)(??$1*)(.?)(??[$1$2]*)';
spl = regexp(inp,rgx,'match');
idx = cellfun(@(c)any(cellfun(@numel,c)>9),spl);
find(idx)

ans = 2×1

2 4

[Stephen23]

Matthew Worker: this solves (my interpretation of) your original question. You have edited your question and changed the example data: you should check that my answer does what you require.

[Stephen23]

"However if it is possible to do this with the original text file I would be extremely appreciative. "

Yes it can be done with the original text file. But as I do not have the original text file, I cannot develop code for it.

If you want further help, please explain more precisely the requirements for detecting those characters, for example:

• which two characters?
• in what sequence?
• with what repititions? (of characters or groups of characters)

The more vague your explanation is, the more likely you are to get answers that do not do what you need.

[Dan Sargent]

Matthew Worker, I am having the same problem, there doesn't seem to be a way to do this with MATLAB from what I've seen it may be easier to just use python.

[Walter Roberson]

The bold text does not represent repetitions this time, not unless you mean repetition between lines. In the previous example there was two halves, with the second being the same as the first.

If the task is to find places where there is a string of at least 10 d or e characters then

'[de]{10,}'

can find that, and the 'once' and isempty and indexing from my Answer gives you the rest. It just depends on your having used readlines() on the file.

[Stephen23]

Matthew Worker: are the specific characters known in advance? Or do you want to detect them automatically? (i.e. detect any two characters that are repeated more than 10 times contiguously)

Are there any particular patterns that you need to include/exclude? (e.g. does 10 'e' characters in a row count, or does the sequence have to include at least one 'd' character?).

Answer 2

[Walter Roberson]

You say "10 or over", so is it correct that the program needs to all possible patterns? For example,

'adadadadaaaadadadadaaa'

ans = 'adadadadaaaadadadadaaa'

(length 22) should be located if it exists?

S = {'asseefadfefaaadddaaadddasdfsdf', 'asseeadadadadaaaadadadadaaadfsdf'}

S = 1×2 cell array
    {'asseefadfefaaadddaaadddasdfsdf'}    {'asseeadadadadaaaadadadadaaadfsdf'}

matches = regexp(S, '([ad]{5,})\1', 'match');
celldisp(matches)

 
matches{1}{1} =
 
aaadddaaaddd
 
 
matches{2}{1} =
 
adadadadaaaadadadadaaa
 

[Matthew Worker]

I am sorry let me rephrase my question.

[Walter Roberson]

S = {'asseefadfefaaadddaaadddasdfsdf', 'asseeadadadadaaaadadadadaaadfsdf', 'asdfsdfsdfsasdfsdfsdfsasdfsdfsdfs', 'asseefadfefaaadddaaadddasdfsdf', 'asdfsdfsdfsasdfsdfsdfsasdfsdfsdfs', 'asseefadfefaaadddaaadddasdfsdf'}

S = 1×6 cell array
    {'asseefadfefaaadddaaadddasdfsdf'}    {'asseeadadadadaaaadadadadaaadfsdf'}    {'asdfsdfsdfsasdfsdfsdfsasdfsdfsdfs'}    {'asseefadfefaaadddaaadddasdfsdf'}    {'asdfsdfsdfsasdfsdfsdfsasdfsdfsdfs'}    {'asseefadfefaaadddaaadddasdfsdf'}

matchidx = regexp(S, '([ad]{5,})\1', 'once')

matchidx = 1×6 cell array
    {[12]}    {[6]}    {0×0 double}    {[12]}    {0×0 double}    {[12]}

S(~cellfun(@isempty, matchidx))

ans = 1×4 cell array
    {'asseefadfefaaadddaaadddasdfsdf'}    {'asseeadadadadaaaadadadadaaadfsdf'}    {'asseefadfefaaadddaaadddasdfsdf'}    {'asseefadfefaaadddaaadddasdfsdf'}

[Walter Roberson]

... Wait, any two characters, or two specific characters?

[Matthew Worker]

@Walter Roberson Two specific characters (d/e), sorry should've gone more in depth. The problem is already solved as I can convert the data to an array. But if you know a way of going through a block of text data (in a txt file) and stating the lines it is detected, I would be very appreciative.

Thank you for you previous answer as it does give a good explanation.

[Walter Roberson]

Example of reading from file:

%create a file for demonstration purposes only
tname = [tempname() '.txt'];
fid = fopen(tname, 'w');
T = regexprep('asseefadfefaaadddaaadddasdfsdf\nasseeadadadadaaaadadadadaaadfsdf\nasdfsdfsdfsasdfsdfsdfsasdfsdfsdfs\nasseefadfefaaadddaaadddasdfsdf\nasdfsdfsdfsasdfsdfsdfsasdfsdfsdfs\nasseefadfefaaadddaaadddasdfsdf\n', 'a', 'e');
fprintf(fid, T);
fclose(fid);

%okay, main function
filename = tname;

%okay, main function
S = readlines(filename);
matches = S(~cellfun(@isempty, regexp(S, '[de]{10}', 'once')));

matches

matches = 4×1 string array
    "esseefedfefeeedddeeedddesdfsdf"
    "esseeededededeeeededededeeedfsdf"
    "esseefedfefeeedddeeedddesdfsdf"
    "esseefedfefeeedddeeedddesdfsdf"

%alternative without readlines
S = regexp(fileread(filename), '\r?\n', 'split');
matches = S(~cellfun(@isempty, regexp(S, '[de]{10}', 'once')));
matches

matches = 1×4 cell array
    {'esseefedfefeeedddeeedddesdfsdf'}    {'esseeededededeeeededededeeedfsdf'}    {'esseefedfefeeedddeeedddesdfsdf'}    {'esseefedfefeeedddeeedddesdfsdf'}

%alternative without splitting

S = fileread(filename);
matches = regexp(S, '^.*[de]{10}.*$', 'match', 'dotexceptnewline', 'lineanchors');
matches

matches = 1×4 cell array
    {'esseefedfefeeedddeeedddesdfsdf'}    {'esseeededededeeeededededeeedfsdf'}    {'esseefedfefeeedddeeedddesdfsdf'}    {'esseefedfefeeedddeeedddesdfsdf'}