how do i replace the first 100 characters of all lines in a file using awk

Question

How do I replace the first 100 characters of all lines in a file using awk? There is no field delimiter in this file. All fields are fixed width. And given the variation in the data, I cannot use a search and replace.

Didn't work. the -r flag is illegal with my version. sed 's/.{46}/" "/' with855st — schwartzray, Mar 04 '14 at 18:16

score 2 · Answer 1 · answered Mar 04 '14 at 16:57

2

How about sed? To replace the first 100 characters with say A:

$ sed -r 's/.{100}/A/' file

If you're happy with the results rewrite the file using -i:

$ sed -ri 's/.{100}/A/' file

answered Mar 04 '14 at 16:57

Chris Seymour

83,387
30
160
202

Didn't work. the -r flag is illegal with my version. sed 's/.{46}/" "/' with855st – schwartzray Mar 04 '14 at 18:26
Try `sed 's/.\{100\}/replacement text/'` – Scrutinizer Mar 04 '14 at 23:29
Okay this works. But I was hoping to replace the characters with an equal number of replacements--eg 100 blanks. I would just have to be literal with the replacement text I guess – schwartzray Mar 05 '14 at 23:22
In the end I did these set of commands. They worked for me. Thanks. * sed 's/.\{178\}//' with855st > with855st2 * nawk '{ sub(/^/, " ... "); print }' with855st2 > with855st3 * cut -c46-47 with855st > with855st3a * nawk 'FNR==NR{a[++i]=$0;next} {print a[FNR] "\t" $0}' with855st3a with855st3 > with855st4 * nawk '{ sub(/^/, "0000000000 "); print }' with855st4 > with855st5 – schwartzray Mar 06 '14 at 15:56

score 1 · Answer 2 · answered Mar 04 '14 at 16:57

1

awk '{print "replacing text..." substr($0,100)}'

answered Mar 04 '14 at 16:57

Sigi

4,826
1
19
23

Tried this and it didn't work. It basically deleted the whole line except for 2 blanks. nawk '{print " " substr($11,46)}' with855st – schwartzray Mar 04 '14 at 18:25
I think that you need to investigate more your file format then. This is basic gawk/nawk syntax, as you can find [here](http://heirloom.sourceforge.net/man/nawk.1.html). – Sigi Mar 04 '14 at 19:02
1

@schwartzray, why do you use `$11` ? @Sigismundo used `$0` – Scrutinizer Mar 04 '14 at 23:34
I used #11 because I ASSUMED, that was the starting column. I really wish to replace columns 11 thru 46 with blanks. – schwartzray Mar 05 '14 at 22:41

BMW · Answer 3 · 2014-03-05T05:50:40.573

1

Use pure shell.

#!/usr/bin/env bash

# read each line into shell variable REPLY
while read -r ; do
   echo "REPLACE text ... ${REPLY:100}"
done <file

Explanation

REPLY is shell variable, refer http://www.gnu.org/software/bash/manual/html_node/Bash-Variables.html. Set to the line of input read by the read builtin command when no arguments are supplied
${REPLY:100} - get the string after 100 characters.

edited Mar 05 '14 at 05:50

answered Mar 05 '14 at 01:20

BMW

42,880
12
99
116

+1: Did not know `REPLY` variable. Though you are **not** reading in an array. You are reading entire line in default variable. no? – jaypal singh Mar 05 '14 at 05:01
1

Added more explanations about REPLY – BMW Mar 05 '14 at 05:14
1

Thanks @BMW also appreciate the link! – jaypal singh Mar 05 '14 at 05:30

how do i replace the first 100 characters of all lines in a file using awk

3 Answers3

Explanation