UNIX for Dummies Questions & Answers

hi,
there is a text like this in text.txt

DETAILS (
USERID,
USERNM
)

how to find the 'DETAILS (' and start to read next line until meet the char ')', when read next line i need to put the 'USERID' into a variable name to be use later

thanks a lot ...please help

Well what have you tried so far?

below is what i tried so far, any good idea
hasDetail='FALSE'
exec 4<$1
while read eachline <&4
do

if [ "$eachline" == ")" ]
then
hasDetail='FALSE'
fi
if [ "$hasDetail" == "TRUE" ]
then
varnm=`echo $eachline | cut -f1 -d,`
repnm=P`echo $varnm | cut -c 2-`
sed 's/$varnm/$repnm/g' $1 > $1.$$
sed 's/$repnm/$repnm/g' $1.$$> $1
fi
if [ "$eachline" == "DETAIL (" ]
then
hasDetail='TRUE'
fi
done

Change the place of

Code:

if [ "$eachline" == "DETAIL (" ] 
	then 
		hasDetail='TRUE'
	fi

This should be the first check to be done.

I dont find the approach you are using reliable, it will fail in the scenario where the data file has unmatched ")"

rishi

If you are looking to extract only USERID check this out.

Code:

sh-2.05b$ echo "DETAILS (
USERID,
USERNM
)" | sed  -n -e '/DETAILS.*/{
n
s#\(.*\)\,#\1#p
}'

It will extract and give out USERID

vino

j=0;
foreach $curline (<FILE NAME>)
{

if j=0
{
next if($cur_line !~ /DETAILS\(/);
j=1;
next;
}

if($cur_line !~ /\)/);
{
.
. put your logic here...
.
}
else
{
exit 1
}

one more way ...

echo "DETAILS (
USERID,
USERNM
)" | sed -ne 's/,$//p'