map first field to second field in txt file

hi,

I have this txt file :

1::Toy Story (1995) ::Animation|Children's|Comedy

2::Jumanji (1995) ::Adventure|Children's|Fantasy

8::Tom and Huck (1995) ::Adventure|Children's

I want to map for example 1 into animation, and 2 into adventure 8 into adventure i.e ,i need creat txtfile has two columns , the first column contains 1,2,8 and second column contains animation,adventure,adventure

please, how do that thanks in advance

Answer 1

[per isakson]

A slight modification of the textscan command I provided to your question the other day will read the file. (You never explained how "::" should be interpreted.) What do you mean by "I read each filed alone of a one row, textscan do not work with it."? If you don't need a column add "*" after "%", e.g. "%*d" to suppress the first column.

Thus

    >> cac = txt2m
    cac = 
        [3x1 int32]    {3x1 cell}    {3x1 cell}
    >> cac{:}
    ans =
               1
               2
               8
    ans = 
        'Toy Story (1995) '
        'Jumanji (1995) '
        'Tom and Huck (1995) '
    ans = 
        'Animation|Children's|Comedy'
        'Adventure|Children's|Fantasy'
        'Adventure|Children's'
    >>

where the function, txt2m, is given by

    function    cac = txt2m()
    fid = fopen('cssm.txt');
    cac = textscan( fid, '%d%s%s'                       ...
                ,   'Delimiter'             ,   ':'     ...
                ,   'CollectOutput'         ,   false   ...
    ...         ,   'EmptyValue'            ,   -999    ...
    ...         ,   'ExpChars'              ,   ''      ...
                ,   'MultipleDelimsAsOne'   ,   true    ... 
                ,   'Whitespace'            ,   ''      );
    fclose( fid );    
    end

then regexp and str2num

    >> regexp( cac{2}, '\d{4}', 'match' )
    ans = 
        {1x1 cell}
        {1x1 cell}
        {1x1 cell}
    >> ans{:}
    ans = 
        '1995'
    ans = 
        '1995'
    ans = 
        '1995'

--- In response to the answer below ---

This modified function, txt2m, reads and parses your file. It reads the file to a string with the function, fileread (thanks Walter, I didn't know of that one), and replaces "::" by "¤" (knock on wood). I just picked a character on the keyboard.

Try

>> cac = txt2m()
cac = 
    [13x1 int32]    {13x1 cell}    {13x1 cell}
>>

where

cssm.txt contains your 13 rows

and where

function    cac = txt2m()
str = fileread( 'cssm.txt' );
str = strrep( str, '::', '¤' );
cac = textscan( str, '%d%s%s'                       ...
            ,   'Delimiter'             ,   '¤'     ...
            ,   'CollectOutput'         ,   false   ...
...         ,   'EmptyValue'            ,   -999    ...
...         ,   'ExpChars'              ,   ''      ...
            ,   'MultipleDelimsAsOne'   ,   true    ... 
            ,   'Whitespace'            ,   ''      );

end

[huda nawaf]

thanks, when write c1=cac{1} I got all data of three columns, can we get c1=cac{1};c2=cac{2}; c3=cac{3}?

many thanks

[per isakson]

If so, your textscan-command is not identical to mine. Or less likely Matlab version. I use R2012a. See above. Try

    cac2 = cac{1};

Use the Variable Editor to inspect the content of variables. Try double-click "cac" in the Workspace window.

[huda nawaf]

now i got c1=cac{1}

1
2
8

but when write c2=cac{2}

i got the other two fields together not separating.

why?

[per isakson]

Hard for me to guess. I don't know what exactly you are doing. Do you use this "line"?

    'CollectOutput'         ,   false  ...

[huda nawaf]

this is what I do

f1=fopen('d:\matlab\r2011a\bin\movielens\1m_mov\movies.txt');

    cac = textscan( f1, '%d %s %s'                       ...
                ,   'Delimiter'             ,   ':'     ...
                ,   'CollectOutput'         ,   true    ...
    ...         ,   'EmptyValue'            ,   -999    ...
                ,   'ExpChars'              ,   ''      ...
                ,   'MultipleDelimsAsOne'   ,   true    ... 
                ,   'Whitespace'            ,   ''      );
    fclose( f1 )
    c1=cac{1};c2=cac{2};c3=cac{3}

[huda nawaf]

now I got the three fields when I place false instea of true.

but I got just the 12 values for each field, while I have 3000 values for each column. why?

[per isakson]

I cannot guess

1. Any error message?
2. What do the data lines 12 to 14 look like?

[huda nawaf]

thanks there is no any error message these lines 12::Dracula: Dead and Loving It (1995)::Comedy|Horror 13::Balto (1995)::Animation|Children's 14::Nixon (1995)::Drama

I will check my code with another txtfile, then tell u

[huda nawaf]

I tried with another txt file, the same thing, where I took file with 7 rows

but when run it give me 2 row values

[per isakson]

The row

    12::Dracula: Dead and Loving It (1995)::Comedy|Horror

contains four values not three as assumed when designing the format string. The ":" after "Dracula" is interpreted as a delimiter.

There is no simple way AFAIK to prescribe "::" as delimiter.

One way is to read the complete file as one string or one string per line and replace "::" by another character and use that as delimiter. Which character would be safe to use as delimiter?

You should have inspected the result, cac, of the read operation.

[huda nawaf]

THANKS, i corrected the error, and I did what you suggested but the problem is not solved.

[per isakson]

1. What did you do? What does your new code look like?
2. How does it behave? What output? What error message?

Why do you expect me to guess?

[per isakson]

Why don't you care to respond?

Answer 2

[huda nawaf]

I just need to read txtfile with this format:

1::Toy Story (1995) ::Animation|Children's|Comedy

2::Jumanji (1995) ::Adventure|Children's|Fantasy

8::Tom and Huck (1995) ::Adventure|Children's

there is no error message , but I have 3000 rows ,when I read it use the code u sent it earlier I got just first 12 rows?

I want to map first fiels into the first word of third field

ex. 1 Animation 2 Adventure 8 Adventure

this is what I need . the first 13 rows of my file:

1::Toy Story (1995)::Animation|Children's|Comedy
2::Jumanji (1995)::Adventure|Children's|Fantasy
3::Grumpier Old Men (1995)::Comedy|Romance
4::Waiting to Exhale (1995)::Comedy|Drama
5::Father of the Bride Part II (1995)::Comedy
6::Heat (1995)::Action|Crime|Thriller
7::Sabrina (1995)::Comedy|Romance
8::Tom and Huck (1995)::Adventure|Children's
9::Sudden Death (1995)::Action
10::GoldenEye (1995)::Action|Adventure|Thriller
11::American President, The (1995)::Comedy|Drama|Romance
12::Dracula: Dead and Loving It (1995)::Comedy|Horror
13::Balto (1995)::Animation|Children's

thanks

[Walter Roberson]

textscan() will not work for this, at least not as-is. You can read the file (such as by using fileread() ) and then use regexp() to parse it.

[per isakson]

See my answer above. I hope the lines you don't show don't contain "¤".

[huda nawaf]

thanks for both walter and per. lastly, I got what I need by your efforts