Thread Subject:

importdata problem !!

Hello,

I'm trying to load a .txt file in matlab for statistical manipulations (mean, plotting, etc).

When I use importdata() it creates a cell variable, and I need a structure... The problem is, that I don't have this problem all time.

For example :

If a use this text file :

Day1 Day2 Day3 Day4 Day5 Day6 Day7
95.01 76.21 61.54 40.57 5.79 20.28 1.53
23.11 45.65 79.19 93.55 35.29 19.87 74.68
60.68 1.85 92.18 91.69 81.32 60.38 44.51

the command M = importdata('myfile.txt', ' ', 1) will give me a structure :
C =
          data: [5x3 double]
      textdata: {'Day1' 'Day2' 'Day3' 'Day4' 'Day5' 'Day6' 'Day7'}
    colheaders: {'Day1' 'Day2' 'Day3' 'Day4' 'Day5' 'Day6' 'Day7'}

But !!! if I use this text file :
Temps Tension-Dev1_ai0 Tension-Dev1_ai1 Tension-Dev1_ai2 Tension-Dev1_ai3
0.0000E+0 1.3184E-1 1.4648E-1 1.2695E-1 9.7656E-2
2.0000E-3 1.2695E-1 1.3184E-1 1.2695E-1 1.1719E-1
4.0000E-3 1.2695E-1 1.3184E-1 1.2695E-1 9.7656E-2

it will give me a 1x1 cell :
C =
    'Temps Tension-Dev1_ai0 Tension-Dev1_ai1 Tension-Dev1_ai2 Tension-Dev1_ai3'

So, what I need is a structure for the second text file, and I still can't get it.
I don't see where the problem is, as I don't see any differences in the two text files.

Please, help !!!!!

Thanks very much !!!

"Egli Vorfi" <vig.notorious@gmail.com> wrote in message <ic0h40$5ti$1@fred.mathworks.com>...

> For example :
>
> If a use this text file :
>
> Day1 Day2 Day3 Day4 Day5 Day6 Day7
> 95.01 76.21 61.54 40.57 5.79 20.28 1.53
> 23.11 45.65 79.19 93.55 35.29 19.87 74.68
> 60.68 1.85 92.18 91.69 81.32 60.38 44.51
>
> the command M = importdata('myfile.txt', ' ', 1) will give me a structure :
> C =
> data: [5x3 double]
> textdata: {'Day1' 'Day2' 'Day3' 'Day4' 'Day5' 'Day6' 'Day7'}
> colheaders: {'Day1' 'Day2' 'Day3' 'Day4' 'Day5' 'Day6' 'Day7'}
>


euh there is problem even with your first test, the first answer is not expected : data should be [3x7 double] and not [5x3 double]

Aurélien

http://blog.developpez.com/matlab4geek/

yeah you're right, it should a 3x7 double, it's just a copy paste mistake i did.

does the second text file work for you ??


"Aurelien Queffurust" <tug83@yahoo.fr> wrote in message <ic0hu7$rce$1@fred.mathworks.com>...
> "Egli Vorfi" <vig.notorious@gmail.com> wrote in message <ic0h40$5ti$1@fred.mathworks.com>...
>
> > For example :
> >
> > If a use this text file :
> >
> > Day1 Day2 Day3 Day4 Day5 Day6 Day7
> > 95.01 76.21 61.54 40.57 5.79 20.28 1.53
> > 23.11 45.65 79.19 93.55 35.29 19.87 74.68
> > 60.68 1.85 92.18 91.69 81.32 60.38 44.51
> >
> > the command M = importdata('myfile.txt', ' ', 1) will give me a structure :
> > C =
> > data: [5x3 double]
> > textdata: {'Day1' 'Day2' 'Day3' 'Day4' 'Day5' 'Day6' 'Day7'}
> > colheaders: {'Day1' 'Day2' 'Day3' 'Day4' 'Day5' 'Day6' 'Day7'}
> >
>
>
> euh there is problem even with your first test, the first answer is not expected : data should be [3x7 double] and not [5x3 double]
>
> Aurélien
>
> http://blog.developpez.com/matlab4geek/

"Egli Vorfi" <vig.notorious@gmail.com> wrote in message <ic0lm9$qip$1@fred.mathworks.com>...
> yeah you're right, it should a 3x7 double, it's just a copy paste mistake i did.
>
> does the second text file work for you ??
>
ek/

Yes no problem within 10b , and to tell you the truth even if R12SP1 I get the same result:
 C=importdata('test.txt',' ',1)

C =

          data: [3x5 double]
      textdata: {'Temps' 'Tension-Dev1_ai0' 'Tension-Dev1_ai1' 'Tension-Dev1_ai2' 'Tension-Dev1_ai3'}
    colheaders: {'Temps' 'Tension-Dev1_ai0' 'Tension-Dev1_ai1' 'Tension-Dev1_ai2' 'Tension-Dev1_ai3'}

ok, thanks a lot for your answer mate !


"Aurelien Queffurust" <tug83@yahoo.fr> wrote in message <ic0mid$nfd$1@fred.mathworks.com>...
> "Egli Vorfi" <vig.notorious@gmail.com> wrote in message <ic0lm9$qip$1@fred.mathworks.com>...
> > yeah you're right, it should a 3x7 double, it's just a copy paste mistake i did.
> >
> > does the second text file work for you ??
> >
> ek/
>
> Yes no problem within 10b , and to tell you the truth even if R12SP1 I get the same result:
> C=importdata('test.txt',' ',1)
>
> C =
>
> data: [3x5 double]
> textdata: {'Temps' 'Tension-Dev1_ai0' 'Tension-Dev1_ai1' 'Tension-Dev1_ai2' 'Tension-Dev1_ai3'}
> colheaders: {'Temps' 'Tension-Dev1_ai0' 'Tension-Dev1_ai1' 'Tension-Dev1_ai2' 'Tension-Dev1_ai3'}

When I copy and pasted your file from you topic into a test file I got the 1x1 cell you spoke of without any data.

After playing around it appears to be a formating issue. I went through the file and deleted all the white space and reinserted spaces. After that change the file loaded perfectly.

I would check the formating in the files you are using. I believe some of the white space was not a space it was a tab and in the depths of ASCII there is a difference. And i'm not sure if the import function supports multiple delimiters. Someone correct me if i'm wrong.

If your file is too large to edit manually you might have to go with textscan or a similar function to pull your file in.

for the second case Ex.

fid = fopen('myfile.txt')

C_header = textscan(fid, '%s', 5) %5 is the # of columns that are headers
C_data = textscan(fid, '%f)

 fclose(fid)

textscan as default uses whitespace as a delimiter any tabs, spaces, etc are used to separate the data.

Hope this help!

-Matt

thanks a lot for your help Matt !!

I remarked the space problem too, it turns to be a Tab space and when i changed it to a normal space, it worked. But my files are too large, so I tried this one :

[C1,C2,C3,C4,C5] = textread('test.txt','%f %f %f %f %f','headerlines',1);

and it works even with Tab spaces.
I don't know which is better to use, textread or textscan ?
Which do you think is better ?

Thanks again :)

"Matthew Landowski" <litho23@gmail.com> wrote in message <ic0q2s$b98$1@fred.mathworks.com>...
> When I copy and pasted your file from you topic into a test file I got the 1x1 cell you spoke of without any data.
>
> After playing around it appears to be a formating issue. I went through the file and deleted all the white space and reinserted spaces. After that change the file loaded perfectly.
>
> I would check the formating in the files you are using. I believe some of the white space was not a space it was a tab and in the depths of ASCII there is a difference. And i'm not sure if the import function supports multiple delimiters. Someone correct me if i'm wrong.
>
> If your file is too large to edit manually you might have to go with textscan or a similar function to pull your file in.
>
> for the second case Ex.
>
> fid = fopen('myfile.txt')
>
> C_header = textscan(fid, '%s', 5) %5 is the # of columns that are headers
> C_data = textscan(fid, '%f)
>
> fclose(fid)
>
> textscan as default uses whitespace as a delimiter any tabs, spaces, etc are used to separate the data.
>
> Hope this help!
>
> -Matt

"Egli Vorfi" <vig.notorious@gmail.com> wrote in message <ic0shj$nmf$1@fred.mathworks.com>...

> I don't know which is better to use, textread or textscan ?
> Which do you think is better ?
>


As of 10a , you can read here:
http://www.mathworks.com/help/techdoc/rn/br_bpq8-1.html#bsdnyvb-1

Replace all existing instances of textread with textscan, similar to strread. Open and close files with fopen and fclose.

Aurélien

http://blog.developpez.com/matlab4geek/