Shell Programming and Scripting

Hi all,
I am just trying to read the contents of a file. basically this file has a list of dat files. then i want to access these dat files n execute a script on them one by one using a loop.

i hav e written like this

Code:

ls -l | cut -c 58-88 > file1.txt
while [ $j -lt 928 ]
do
 
arr1[j]="$( sed -n '1p' file1.txt )"
con.ksh ${arr1[j]};
j=`expr $j + 1`
 
echo "working"
done

Please help me out.

edit by scottn: Moved from AIX sub-forum (it's not an AIX-specific question

If we don't want file names to be in a file

Code:

 
ls -1 | while read file
do
   con.ksh $file
done

OR

Code:

 
ls -1 > file.txt
while read file
do
  con.ksh $file
done < file.txt

Thanx anurag 4 ur help.
No basically I am trying to delete the control M characters from around 100 files in a directory.I am using the con.ksh script to delete cntrlM characters in a single file. But how to do it in a loop.
I have tried the following script too. but its not working also

If u can help me out ..

Regards

It looks like you are just passing all files names to convert.ksh script one by one in a loop. Here variables a and arr1[j] have file name value in it.
If so, same is happening in my script. each file ($file variable) is being passed to con.ksh (change it to convert.ksh).

Code:

ls -1 | while read file
do
   echo "processing file: $file"
   convert.ksh $file
done

But the issue is not resolved.
I think you could not get it. I am trying to pass the filenames but those filenames are the files of a directory. So i am trying to extract the files names only by using

ls -l | cut -c 58-88 > file1.txt .

Now this file1.txt contains all the filenames only in column manner.
Now I am trying to execute the files by selecting the names of the files from file1.txt.

Any ways thanks for ur support..

Code:

 
ls | while read file
do
   if [ -f $file ]; then
        echo "processing file: $file"
        convert.ksh $file
   fi
done

Here all files under current dir will be passed to convert.ksh one by one.
If condition can be removed if current dir contains only FILES, not subdirectories.

anurag.singh is correct in what is said. The question we have to be clear on is why you want to

Code:

ls -l|cut -c 58-88 > file1.txt

I assume that this is so you get just the names of the files from an ls -l, but this is just the same as output from ls sent to a file or ls -1 (flag is minus one) to be used in a loop as described.

So, what are you trying to get? If it is just files from a directory, you might also confuse yourself if you write the temporary file there too. Editing the file with con.ksh mid program when that is the input file might have unpredictable results too.

You also run the risk of having a file that is longer than the 31 characters you are limiting this to. If you have a file called this_is_a_very_long_name_for_a_file, then you would try to run

Code:

con.ksh this_is_a_very_long_name_for_a_

and presumably get an error meaning file not found.

Another way to approach this is to use the find command. As such, you can exclude links, directories, pipe files etc. by specifying you only want regular files:

Code:

find . -type f -exec con.ksh {} \;

This looks in the current directory (the dot) for a regular files (-type f) and then executes con.ksh with each as an arguement (-exec con.ksh {} \ The trailing \; marks the end of the -exec part of the command and the find will error if it is missing.



I hope that this is useful, but please write back if this I have missed the point.


Robin
Liverpool/Blackburn
UK