UNIX for Dummies Questions & Answers

I need to change all Newline caracters (\12) to Fieldseparator(\34).
tr -A '\12' '\34' <file1> file2

Replace all delete (\177) with Newline (\12)
tr -A '\177' '\12' <file2> file3

Put the name of the file first in all rows.
awk '{printf "%s\34%s\n", FILENAME,$0} file3 > file4

So far no problem. This works fine when I just need to do it with one file. Now I need to do it on 100 files. So I cant change the filename(file1>file2), and after doing all the tr commands and the awk command I need to concatenate them with:

cat * > AllFilesAfterTrandAwk

Here is an example:

file1 contains:
row1 \12
row2 \177

file2 contains:
row1 \12
row2 \177
row3 data

I need:
file1 row1 \34
file1 row2 \12
file2 row1 \34
file2 row2 \12
file2 row3 data

Does anyone have a clue?

Thanks,
MrKlint

Try this, see if it works:

for file in *; do
(do your translation here) < $file > temp.file
mv temp.file $file
done

Sorry, I do not follow. I havent worked with Unix before. I put my commands one by one in the prompt. If you have the time can you write the commands exactly as I'm supposed to write them?

what does $file mean? all files in the folder? like * ??

You probably want to do some kind of loop.
Note that there are some simpler ways of doing this, I merely took your flow and modified.
For instance, the output of each awk could >> (append) to you final file, saving a step.

Code:

# create a listing of the files to work on
# you might not need to do this, if the list already exists
ls *.txt >tmp_file_list

while read tmp_file
  do
  tr -A '\12' '\34' <${tmp_file} | tr -A '\177' '\12' | awk -v FILENAME=${tmp_file} '{printf "%s\34%s\n", FILENAME,$0}' >${tmp_file}_2
done<tmp_file_list

# concatenate all the files
cat *_2 > AllFilesAfterTrandAwk

I guess this is scripting. I have never done that before, but I will try to find out how to do it.

Thanks.

Scripting is merely executing commands that you could type in directly. You could have put your original example into a script "as-is".

Mostly, there are exceptions, anything you type directly from

Code:

ls -l

to

Code:

awk .....

can be put inside a script file.

Note: after you save your command file, often saved with a .sh ending as in convert12.sh , you need to make the file executable with a

Code:

chmod +x convert12.sh

command.

This is the solution I came up with...

#!/usr/bin/sh
# Create list
ls F4* > tmp_file_list

# Reads the list and does the restructuring
while read tmp_file_list
do
tr -A '\12\177' '\11\12' <${tmp_file_list}> ${tmp_file_list}_1
awk '{printf "%s\11%s\n",FILENAME,$0}' ${tmp_file_list}_1 > ${tmp_file_list}_2
done < tmp_file_list

# Concatenate all the files
cat F4*_2 > AllFilesAfterTrAndAwk
rm F4*_1
rm F4*_2
rm tmp_file_list