Bash script - Text Processing

[st12iker]

Hey ...
I need to create a text processing script that can copy certain lines of a file and write it to a seperate file. So for example, copy lines 40 through 50 or 100 through 200 (the lines to be copied would be different each time, specified by the users need).

Please take a look at my comments below.

Thanks !!

[st12iker]

To copy between two defined lines I came up with:

Sed –n ‘(starting line),(ending line)p’ (input file) >> (output file)

which does the job


I tried creating a script:

#!/bin/bash

Sed –n ‘$1,$2p’ $3 >> $4

Gives me an error:
sed: -e expression #1, char 2: Unknown command: `1'

When I run the script on the command line:
i.e. ./script 8 10 input output

[Franklin52]

This would do the trick:

sed -n "$1,$2p" $3 >> $4

The reason to use double quotes is to invoke the shell to expand the parameters.

Regards

[st12iker]

I’m trying now to get the script to print between one line after the ones specified

So if.

./script 8 10 input output

It would write lines 9 – 11 instead of 8 – 10.

I tried:



#!/bin/bash

# Same as command

# invoked from the command line.



sed -n "$1+1,$2+1p" "$3" > "$4"



exit 0

[Franklin52]

Try the expr command, check the man page.

Regards

[st12iker]

#!/bin/bash
# Same as command
# invoked from the command line.

$1="expr $$1 + 1"
$2="expr $$2 + 1"
sed -n '1,1p' "$3" >> "$4"
sed -n "$1,$2p" "$3" >> "$4"

exit 0

Get a lot of different errors while playing with variants. Any more suggestions?

[Franklin52]

This should work:

FROM=`expr $1 + 1`
UNTIL=`expr $2 + 1`

sed -n "$FROM,$UNTIL p" "$3" >> "$4"

Have a read of some tutorials:

http://www.tldp.org/LDP/Bash-Beginne...tml/index.html
http://tldp.org/LDP/abs/html/

Regards

[ghostdog74]

printf "Enter start: "
read start
printf "Enter end: "
read end
awk -v start=$start -v end=$end 'NR>=start && NR<=end {print $0 }' "file" > output