Delete the first five characters on any line of a text file in Linux with sed

Question

I need a one-liner to remove the first five characters on any line of a text file. How can I do that with sed?

similar question http://stackoverflow.com/questions/971879/what-is-a-unix-command-for-deleting-the-first-n-characters-of-a-line — dogbane, Sep 25 '10 at 22:00

score 157 · Answer 1 · edited Jul 17 '19 at 01:05

157

cut -c6-

This prints each line of the input starting at column 6 (the first column is 1).

edited Jul 17 '19 at 01:05

toblerone_country

answered Sep 25 '10 at 20:48

Greg Hewgill

1

Here, cut has a better performance than sed, but I got problems when used on utf-8 encoded characters. – PSchwede Nov 16 '17 at 10:13
1

Just to elaborate on @PSchwede's comment, GNU `cut` treats all characters as bytes even when you use the `-c` option. GNU `cut` does not support multi-byte characters and probably will not support multi-byte characters for the foreseeable future – Harold Fischer Jan 09 '20 at 02:22

PhilR · Accepted Answer · 2019-07-16T22:48:38.657

86

sed 's/^.....//'

means

replace ("s", substitute) beginning-of-line then 5 characters (".") with nothing.

There are more compact or flexible ways to write this using sed or cut.

edited Jul 16 '19 at 22:48

answered Sep 27 '10 at 17:11

PhilR

How would you do this for the last character of a string? – blarg Jun 23 '15 at 14:34
3

@blarg try `sed 's/.$//'` – Rodrigo Aug 25 '15 at 18:50
plus 1 for answering the 'with sed' part of the question and then mentioning that `cut` is better – s g Sep 19 '16 at 06:32

score 16 · Answer 3 · answered Sep 25 '10 at 20:48

16

sed 's/^.\{,5\}//' file.dat

answered Sep 25 '10 at 20:48

4

BSD sed does not accept the *at most* bound (`{,5}`) , this is GNU specific. An expression that works on both would be `sed 's/^.\{5\}//' file.dat` – Tristan Feb 26 '15 at 12:15

score 12 · Answer 4 · answered Sep 26 '10 at 10:18

12

awk '{print substr($0,6)}' file

answered Sep 26 '10 at 10:18

ghostdog74

4 Answers4