sed command line help

[acctman]

I'm trying to get this sed command to work correctly. I need to retrieve a specific "string" from: terms_id=string&location
below is what i'm working with so far.

sed -ne 's/^.*terms_id=\([^&]*\)&.*$/\1/'

[fris]

whats wrong with it?

test.txt

terms_id=shit&location
terms_id=sex&location
terms_id=scat&location
terms_id=porn&location
terms_id=bbw&location
terms_id=teen&location

[chris@jumbo ~]$ sed 's/^.*terms_id=\([^&]*\)&.*$/\1/' test.txt
shit
sex
scat
porn
bbw
teen

[acctman]

whats wrong with it?

test.txt

hmm I noticed that you removed -ne, it works without that. also is there a way to make it append the outputs each time the command is run? it seems to be writing a new file each time I run the command. thanks

sed 's/^.*terms_id=\([^&]*\)&.*$/\1/' test.txt > getIDs.txt

[fris]

test.txt >> getIDS.txt

[CYF]

hmm I noticed that you removed -ne, it works without that. also is there a way to make it append the outputs each time the command is run? it seems to be writing a new file each time I run the command. thanks

sed 's/^.*terms_id=\([^&]*\)&.*$/\1/' test.txt > getIDs.txt

sed 's/^.*terms_id=\([^&]*\)&.*$/\1/' test.txt >> getIDs.txt

[acctman]

thanks Fris and CYF

[acctman]

back with another question. it keeps grabbing all the html coding around the string. I can see that its finding the string in the output but its also grabbing the html coding content as well.

#! /bin/sh

if [ $# -ne 2 ]; then
    echo "usage: $0 termfile baseurl" >&2
    exit 1
fi
termfile="$1"
baseurl="$2"

while read term; do
    wget -q -O- "$baseurl/server.php?all=$term" |
		sed 's/^.*terms_id=\([^&]*\)&.*$/\1/'
done < "$termfile"