使用awk或bash中的sed编辑文本

Question

The content of my file.txt is something like this:

fghfg
sdfsd
thrt
wefs
dfd
htyj
dfd
gsedf
     1     sdfsdf
     3     sdfghj
     f     fgdfhtyu
dfdsfsd
sdgdfg
sdfzcxzc
fgjty

I want to change this part:

1     sdfsdf
3     sdfghj
f     fgdfhtyu

to this one: (by removing first column and spaces)

sdfsdf
sdfghj
fgdfhtyu

Then redirect the whole output of file.txt to file2.txt .

How is it possible...?

Thanks

Answer 1

Try changing your script to this:

find "$1" -print |
while IFS= read -r line; do
    awk '!a[$0]++{print $NF}' "$line" > /tmp/file.txt
done

Does anything need to change?

Answer 2

Using sed , you can do something like this

sed 's/^ \+[^ ] \+//' file1.txt > file2.txt

It will remove all sequences at the begining of the line that contain a sequence of whitespaces, followed by a single non whitespace character than a sequence of whitespaces. Using the file in your question, the results is:

fghfg
sdfsd
thrt
wefs
dfd
htyj
dfd
gsedf
sdfsdf
sdfghj
fgdfhtyu
dfdsfsd
sdgdfg
sdfzcxzc
fgjty

Answer 3

This should do:

awk '{print $NF}' file.txt > file2.txt

or shortest solution of the day:

awk '{$0=$NF}1' file.txt > file2.txt
cat file2.txt
fghfg
sdfsd
thrt
wefs
dfd
htyj
dfd
gsedf
sdfsdf
sdfghj
fgdfhtyu
dfdsfsd
sdgdfg
sdfzcxzc
fgjty

Just print the last column.

Answer 4

另一个简单的sed命令，

sed 's/.* \(.*\)/\1/g' file1.txt > file2.txt