how can I edit txt file?

Hello; I am a new user.Can anyone help me edit the txt file that is attached. there are eight columns in the file and I would like to make one column. The order should be from left to right. After first row has been put in order in the column, the command should do the same the for second row. Hope this is a clear explanation. Thank you. Regards, Muhsin A

[Image Analyst]

It would certainly be a lot easier if you could have the file written out with a delimiter between the numbers. Is that possible?

[Muhsin]

Yes, I can put a space between columns. Would you want me to do that? But please keep in mind It shouldn't be column by column.

Answer 1

[Image Analyst]

This works:

txt = fileread('mineral 90.txt');
numbers = cell2mat(textscan(txt, '%f')) % Convert text to doubles.

Answer 2

[Walter Roberson]

Once you have the data read in to variable DATA, then DATA.' would rearrange it with the data down the columns instead of across the rows.

[Muhsin]

Dear Walter; In attached file, I've put a space between columns. The code above didn't work. Would you help me please?

[Walter Roberson]

DATA = load('mineral.txt');
new = DATA.';

works fine for me, producing something with 8 rows and 2275 columns.

You said "After first row has been put in order in the column, the command should do the same the for second row" but you did not specify what how you wanted the second row to appear relative to the output for the first row. The above switches rows and columns, placing the outputs beside each other as rows.

If you want a single column as output, then

new = DATA.';
new = new(:);

or

new = reshape(DATA.', [], 1);

[Muhsin]

Dear Walter, The code works but the numbers in the column are not the same with txt file. the numbers are in a vertical column but not in the same order. To be clear about what I want, I will show a scheme;

1  2  3  4  5  6  7  8 
9 10 11 12 13 14 15 16 
.
.
.

However; I want it to be as following;

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
.
.
.

[Walter Roberson]

>> DATA = [1  2  3  4  5  6  7  8 
9 10 11 12 13 14 15 16 ]
DATA =
     1     2     3     4     5     6     7     8
     9    10    11    12    13    14    15    16
>> reshape(DATA .', [], 1)
ans =
     1
     2
     3
     4
     5
     6
     7
     8
     9
    10
    11
    12
    13
    14
    15
    16

[Muhsin]

Dear Walter, The code worked but as you can see from pic.

first number starts from somewhere in data not from the beginning. How can I solve it? And I need all decimals in txt file, But in matlab they are not there!! Thank you.

[Walter Roberson]

I tried with the file you supplied and it works for me.

If you were to just display the output, then because it is so long, some of it is going to scroll out of the command window, so you would not see the beginning of it. You can compare the results of

DATA(1,1:10)

and

new(1:10,:)

"And I need all decimals in txt file, But in matlab they are not there!!"

Give the command

format long g

and display the data again.

[Muhsin]

Dear Walter; I am having a warning of 'Index exceeds matrix dimensions.' since the output is so long. what would be an alternative way ?

[Walter Roberson]

What commands are you using when you get the warning?

[Muhsin]

DATA(1,1:10) #this command gives me the warning.

[Walter Roberson]

DATA(1,1:8) and compare against new(1:8,:)

[Muhsin]

Dear Walter;

DATA(1,1:8) # this command gives me only first and it is not in the format of one column

new(1:8,:) # this command says Undefined variable.

<<

>>

[Walter Roberson]

I specifically defined

new = DATA.';
new = new(:);

or

new = reshape(DATA.', [], 1);

which are equivalent.

It is not intended that DATA(1,1:8) give you back a column vector. It is intended that you check that the first row of the input, DATA(1,1:8) does indeed show up as the first 8 locations of the output, new(1:8,:)

>> DATA(1,1:8)
ans =
  Columns 1 through 6
                 -0.012813                  0.032689                   0.02469                 0.0022913                 -0.026407                 -0.019206
  Columns 7 through 8
                  0.011896                 -0.020803
>>     new = DATA.';
  new = new(:);
>> new(1:8,:)
ans =
                 -0.012813
                  0.032689
                   0.02469
                 0.0022913
                 -0.026407
                 -0.019206
                  0.011896
                 -0.020803

This shows that the command to create the output variable named "new" does work, and does not "starts from somewhere in data not from the beginning" as you had claimed.

[Muhsin]

Dear Walter;

Should't that 'new(1:8,:)' command give me the whole numbers in the columns instead of only first row of the columns??

[Walter Roberson]

It is showing you all of the columns, which is to say all 1 of them. You have been vague about what you want the output to be.

   DATA.'

would switch all the rows and columns, giving you 8 rows and over 2000 columns.

reshape(DATA.',[],1)

will give you a single column result which is the elements from the first row first, then the elements from the second row, then the elements from the third row, and so on.