从每一行中删除部分字符串

Question

I have a text file, where each line is a single string of the format

/home/usr1/284.txt

The whole file is like

/home/usr1/284.txt
/home/usr1/361.txt

What I want is to remove /home/usr1/ and keep the file name, eg, 284.txt

How to do that using linux/unix command?

Answer 1

sed -e 's!/home/usr1/!!' filename.txt

or

awk -F\/ {print $NF} filename.txt

should do the trick. Note the use of ! instead of the more usual / as pattern delimiters in the sed example - it means you don't have to escape literal / characters in your pattern.

Answer 2

Since the fields in the file are fixed, you can simply do:

cut -b 12-

To skip the first 11 bytes of the input.

Answer 3

您还可以使用Perl，如下所示：

perl -pe 's,.*/,,' file.txt

Answer 4

Try this:

while read line; do basename "$line"; done < filename

The reciprocal of basename is dirname , in case you need the other part eventually.

Answer 5

There's more than one way to do it (TIMTOWTDI). In addition to the existing answers, I can think of two ways of going about this:

1. Use slash "/" as a field delimiter: reverse, cut the first field, then reverse again:

    < filename.txt rev | cut -d/ -f 1 | rev 

2. These are filenames, hence you can use GNU basename in combination with xargs (or GNU parallel ) to, as you say, "keep the file name":

    < filename.txt xargs basename -a 

   or

    < filename.txt parallel -X basename -a 

Answer 6

Got bash?

read -d '' -a lines < input.txt
echo "${lines[@]##*/}"