I have a cell array, where each cell is a string of the following form:
A{1} = hello, world, the, hello, the, how, are, you, world
I would like to retain only the unique words, to give me a new char array that will go into an output cell array:
B{1} = hello, world, the, how, are, you
unique() does not seem to cut it. Is there an easy way of doing this?
I agree with some previous comments on questioning why the data is in this format (if you have a choice).
However:
>> A{1} = 'hello, world, the, hello, the, how, are, you, world';
>> B{1} = strjoin(unique(strtrim(strsplit(A{1}, ','))), ', ')
B =
'are, hello, how, the, world, you'
Hope that helps.
* EDIT * another solution that removes strtrim if you know there will always be a space. Also, if you do not want the operation to sort the strings you can use a 'stable' call to unique:
>> B{1} = strjoin(unique(strsplit(A{1}, ', '),'stable'), ', ')
B =
'hello, world, the, how, are, you'
I suspect that A{1} is one long string.
A{1} = 'hello, world, the, hello, the, how, are, you, world';
In which case unique will just return the string or characters.
>> unique(A)
ans =
'hello, world, the, hello, the, how, are, you, world'
>> unique(A{:})
ans =
,adehlortuwy.
The string needs to be converted to a cell array first.
>> a = textscan(A{1},'%s','delimiter',',')
a =
{9x1 cell}
>> b = unique(a{:})
b =
'are'
'hello'
'how'
'the'
'world'
'you'
and then if you want to convert back to a long string again
>> sprintf('%s,',b{:})
ans =
are,hello,how,the,world,you,
Although to be honest if I was having to process strings in this way I wouldn't start with them all in one long string.
You probably didn't use unique() properly. It should work well. See example:
A = {'one','two','twenty-two','One','two'};
C = unique(A)
You will get
C =
'One' 'one' 'twenty-two' 'two'
unique? If used in the proper way it should do exactly what you need.