how to manipulate a cell

Hi I have 1621x1 cell and each row is something like this: ;1800742;0,0000;0,0000;0,0000;;;;;;;1,4400;0,0000;0,0000;0,0000;0,0000;

";" is delimiter between different values.

I need to find the blank places in each row and fill it with a word like "None". As a beginner I need to know how can I access the blank places?is it possible to at first store the values in an array and then find the blank places?if yes how can I do that?

I would really appreciate any helps.

Thanks,Navid

Answer 1

[Azzi Abdelmalek]

A={1800742;0;0;0;[];[];1.4400;0;0;[];0}
A(cellfun('isempty',A))={'none'}

[Navid]

Thanks for your answer,

unfortunately it did not work.

maybe because the blank places in the cell are like ;;; not like [];[];[]

could it be the reason?

[Azzi Abdelmalek]

A={1800742 ;';' ;0;0;0; ';' ;1.4400;0;0;';';0}
B=cellfun(@num2str,A,'un',0)
A(strcmp(B,';'))={'none'}

[Navid]

could you please explain about lines? I'm getting error,maybe I should change something.

Thanks

[Azzi Abdelmalek]

Can you explain how your data looks like?

[Navid]

yes sure, there is a 1621x1 cell,and this is on of the rows:

;1800742;0,0000;0,0000;0,0000;1,4400;0,0000;0,0000;0,0000;0,0000;0,0000;0,0000;0,0000;0,0000;0,0000;2,5500;;;;;;0,0000;-8,8000;12,1600;12,1600;15,0000;0,0000;-0,5000;

As you in the second line there is ;;;;;; ,it means there is no value,and I need to fill this places like this:

;1800742;0,0000;0,0000;0,0000;1,4400;0,0000;0,0000;0,0000;0,0000;0,0000;0,0000;0,0000;0,0000;0,0000;2,5500;none;none;none;none;none;0,0000;-8,8000;12,1600;12,1600;15,0000;0,0000;-0,5000;

Many thanks for your attention.

[Azzi Abdelmalek]

If A is your array, when type

A(1:20)

What do you get?

[Navid]

as I told you the cell is 1621x1,I mean there is 1621 rows and just one column. all of mentioned values are in the first row and first column.

actually I had an ASCII file and import it.this is my code for importing part:

fed = fopen('Antos_2013-11-07_07-08-58.asc');

mydata = textscan(fed,'%s');

it gives me "mydata" which is a cell with 1621 rows and one column.

[Navid]

I should add that when I type mydata(1:20) I'm getting this error:

Index exceeds matrix dimensions.

Error in plotrand (line 3)

mydata(1:20);

[Azzi Abdelmalek]

Ok then type

whos mydata
whos mydata{1}

[Navid]

I got this:

mydata      1x1             12422212  cell

[Azzi Abdelmalek]

whos mydata{1}

[Navid]

I've already written you,it gives me:

Name Size Bytes Class Attributes

mydata      1x1             12422212  cell

[Navid]

should I do something else?

could you tell me please what is the problem?

[Azzi Abdelmalek]

Type

c=mydata{1}
whos c

[Navid]

it gives:

Name Size Bytes Class Attributes

c         16201x1             12422100  cell

[Azzi Abdelmalek]

TRY THIS

B=cellfun(@num2str,mydata{1},'un',0)
mydata(~cellfun('isempty',regexp(B,';+','start')))={'none'}

[Navid]

sry for taking your time but it gives a cell which is not still correct:

 mydata      1x16200            1944000  cell

and in all 16200 elements has been written none.

I think I need to change the code to get correct result,could you just explain the lines please,I just need to understand how does it work.

Thanks

[Azzi Abdelmalek]

doc cellfun
doc regexp

Post a sample of your file

[Navid]

my file is something like this attachment.

and this is the code for importing:

fed = fopen('sample.txt');

mydata = textscan(fed,'%s');

[Azzi Abdelmalek]

Can you explain what a comma , represent in your file, and how your data are separated, and the type your data :string or double? You can read your file line by line

fid = fopen('sample.txt');
res={};
while ~feof(fid)
res{end+1,1} =fgetl(fid);
end
fclose(fid);
celldisp(res)

[Navid]

the numbers are floating point and the comma represent that,for example:(10,5 represent 10.5)

and the different numbers are separated by semicolon it could be string or double,i just need to fill the blank places(the places which there is nothing between two semicolons ) with "none".

[Azzi Abdelmalek]

Try this

clear
fid = fopen('sample.txt');
res={};
while ~feof(fid)
  res{end+1,1} =fgetl(fid);
end
fclose(fid);
A=strrep(res,',','.');
B=regexp(A,';','split');
for k=1:numel(B)
  C(k,:)=B{k};
end
C=cellfun(@str2double,C,'un',0);
idx=isnan(cell2mat(C));
C(idx)={'none'}

Answer 2

[Navid]

Dear Azzi Abdelmalek

Many thanks for your times and helps,it works great.this is exactly what I wanted. thank you so much.

Best,Navid