Convert a string cell value to representative datas

test{1,1}=['0,+,ESI,ms1,-,line,1.1-790.4,275,1.1 41,1.2 30,1.6 10,1.7 24']
test{2,1}=['0.0505163,+,ESI,ms1,-,line,1.1-834.0,254,1.1 26,1.2 16,1.3 52,1.4 30,1.5 12']
test{3,1}=['0.103382,+,ESI,ms1,-,line,1.1-895.4,252,1.1 12,1.4 21,1.5 40,1.6 29,1.7 10']

i want to convert to a struct that the value in the first comma respresents one axis(retention time), the second to sixth comma represents the analisys parameters. The ninth comma and on represents the values of the X, Y data. How i can do that Best Regards Rafael

[Oleg Komarov]

How did you get "test" in the first place?

[Rafael Freire]

Its was exported as ACSII from a Mass spectometer

[Oleg Komarov]

Then, it would be much straightforward to import it with textscan and then do the manipulations.

[Rafael Freire]

How i do that?

[Oleg Komarov]

Post the first 3 lines of your ascii file and tell me which columns you really need.

[Rafael Freire]

0,+,ESI,ms1,-,line,1.1-790.4,275,1.1 41,1.2 30,1.6 10,1.7 24,1.9 30,1.9 20,2.0 18,2.0 41

0.0505163,+,ESI,ms1,-,line,1.1-834.0,254,1.1 26,1.1 16,1.2 52,1.3 30,1.4 12,1.8 23,2.0 17,2.1 12,2.4 11

0.103382,+,ESI,ms1,-,line,1.1-895.4,252,1.1 12,1.4 21,1.5 40,1.6 29,1.7 10,1.7 28,2.0 19,2.1 11,2.1 16,2.1 21,2.2 22,2.2 14,2.4 10,2.7 13,2.8 10,3.5 21,3.6 12,3.6 11,3.7 23,3.7 18,4.2 20,4.7 14,4.9 12,4.9 11

for the first line i need the parameters

0

+

ESI

ms1

-

line

1.1

-790.4

275

than is the values for X,Y

1.1 41

1.2 30

1.6 10

1.7 24

1.9 30

1.9 20

2.0 18

2.0 41

[Oleg Komarov]

Are the fields 2-6 always 1,3,3,1,4 characters?

Also,after line you always have a float followed with by a negative float with no space in between?

[Rafael Freire]

Yes and Yes

Answer 1

[Oleg Komarov]

Con sidering that I created a temp.txt containing:

0,+,ESI,ms1,-,line,1.1-790.4,275,1.1 41,1.2 30,1.6 10,1.7 24,1.9 30,1.9 20,2.0 18,2.0 41
0.0505163,+,ESI,ms1,-,line,1.1-834.0,254,1.1 26,1.1 16,1.2 52,1.3 30,1.4 12,1.8 23,2.0 17,2.1 12,2.4 11
0.103382,+,ESI,ms1,-,line,1.1-895.4,252,1.1 12,1.4 21,1.5 40,1.6 29,1.7 10,1.7 28,2.0 19,2.1 11,2.1 16,2.1 21,2.2 22,2.2 14,2.4 10,2.7 13,2.8 10,3.5 21,3.6 12,3.6 11,3.7 23,3.7 18,4.2 20,4.7 14,4.9 12,4.9 11

Edit

% To import:

fid = fopen('temp.txt');
fmt = '%f,%1s,%3s,%3s,%1s,%4s,%f%f,%f,%s';

% Tweak 4095*n to augment the buffer

opt = {'Delimiter','','CollectOutput',1,'BufSize',4095*3};
out = textscan(fid,fmt,opt{:});
out{end} = cellfun(@(x) textscan(x,'%f %f','Delimiter',',','CollectOutput',1),out{1,end});
fclose(fid);

Now every bit of data is in the right format but it's not stored in the best way.

[Rafael Freire]

When i run the second command line it give me an error:

??? Error using ==> textscan

Buffer overflow (bufsize = 4095)

while reading string from

file (row 11, field 18). Use

'bufsize' option. See HELP TEXTSCAN.

[Oleg Komarov]

It seems to have a problem with line 11. Can you post it (should be out{1,end}{11})?

[Rafael Freire]

the problem is that the first lines has 2141 characters and the line 11 has 10777 characters

[Rafael Freire]

Oleg, you did all right only forget the value of bufsize

fid = fopen('teste2.ascii')

out = textscan(fid,'%f,%1s,%3s,%3s,%1s,%4s,%f%f,%f,%s','Delimiter','','CollectOutput',1,'bufsize',250000)

out{end} = cellfun(@(x) textscan(x,'%f %f','Delimiter',',','CollectOutput',1),out{1,end})

fclose(fid)

Thank you veru much!!!!

Rafael

[Oleg Komarov]

You can delete your comment, it's too long :).

See my edit above.

[Oleg Komarov]

Didn't see your answer because I didn't refresh. I knew about buffer size but was working also on an alternative in case your each the limit. Let me know if it works consistenlty, otherwise you can setup a loop with low level import with fscanf.

Cheers

[Rafael Freire]

Great! Thank You!

Best regards!

Rafael