Reading a all the rows in a CSV file

Hi,

I have written the following code to read the csv file. But How can I modify to read all the rows in both columns. At the moment it only reads the first row of both columns. Also, I need to read multiple .csv files. (See the screenshot below)

clear all
folder='C:<path>Desktop';
fid = fopen('Test2.csv');
out = textscan(fid,'%s%f$f','delimiter',',');
fclose(fid);

Can somebidy advice me to improve this?

Thanks in adavance.

Answer 1

[dpb]

out = textscan(fid,'%s%f$f','delimiter',',');

What's the '$f' in the format string intended to do? Eliminating it should solve the problem; as is, there's a failure after the second value.

For multiple files, see the FAQ at http://matlab.wikia.com/wiki/FAQ#How_can_I_process_a_sequence_of_files.3F

I strongly recommend the dir solution over the alternatives.

[Damith]

Hi,

When ran the following code with 020101.csv it produced the desired outcome but when I chnaged the file name to 014501.csv (both are in the same format and has 3 columns), MATLAB produced the following error.

clear all
folder='C:\<path>\Sorted_CSV';
fid = fopen('014501.csv');
out = textscan(fid,'%s%s%f','delimiter',',');
fclose(fid);

Error using textscan
Invalid file identifier.  Use fopen to generate a valid file identifier.

I noticed when I ran the code with 020101.csv file, the fid produced was 9 but when I ran the same code after changing the same code to 014501.csv, the fid was -1. I have no idea why?

Can somebody explain why and how to correct it to read other .csv file?

See the screenshots for each file below

Thanks again.

[per isakson]

What is the purpose of

    folder='C:\<path>\Sorted_CSV';

I expected to see

    fid = fopen( fullfile( folder, '014501.csv' ) );

[Damith]

Thanks.

[dpb]

Didn't have anything to do with it since you hardcoded a filename in the fopen statement.

What did/does

[fid,msg]=fopen('014501.csv')

return? The error message will tell you (probably) "file not found" or the like which means it's either a typo and you don't actually have a file of that specific name or it's not in a directory on the matlabpath or in the working directory. Other causes are possible, but that's the most likely.

As I recommended before, if you have many of these to process, then

d=dir('*.csv');
for i=1:length(d)
  fid=fopen(d(i).name);
  ...

will let you process them all sequentially w/o having to code up the names at all. Of course, you need to do the dir call on the right subdirectory to find them.

Answer 2

[dpb]

...I ran the code with 020101.csv file, the fid produced was 9 but when I ran the same code after changing the same code to 014501.csv, the fid was -1. I have no idea why?

>> help fopen
 fopen  Open file.
  FID = fopen(FILENAME) opens the file FILENAME for read access.
  FILENAME is a string containing the name of the file to be opened.  
  ...
  FID is a scalar MATLAB integer valued double, called a file identifier. 
  You use FID as the first argument to other file input/output
  routines, such as FREAD and FCLOSE. If fopen cannot open the file, it 
  returns -1.

So, the file wasn't opened.

As Per says, the path string certainly looks funny unless you did substitute a real path in there manually. ( ADDENDUM I see you hardcoded a file name and so the path string/variable wasn't used. That means, of course, that the file must exist in the working directory or on the matlabpath variable to be found).

To see what the actual failure was in more detail use the optional return message to see just what the error was...

    ...
    [FID, MESSAGE] = fopen(FILENAME,...) returns a system dependent error 
    message if the open is not successful.
    ...

One should programmatically, always check that the returned FID value is valid before continuing.

ADDENDUM 2 That you had a returned FID of 9 indicates you've got a lot (like 6) file handles hanging around. That's not necessarily wrong but likely indicates you're not religiously closing files when done with them. Matlab will being with FID 3 for file i/o as 1 and 2 are preassigned console and standard error.