Removing line from text file

[Sachin Patel]

Hi All,
I have a text file tail of that is like.

zeltweg IN CNAME zeltweg.abc.co.uk.
zeltweg IN CNAME zeltweg.xyz.co.uk.
zolder IN CNAME zolder.abc.co.uk.
zolder IN CNAME zolder.xyz.co.uk.
zonta IN CNAME zonta.abc.co.uk.
zonta IN CNAME zonta.xyz.co.uk.
zuela7 IN CNAME zuela7.abc.com.
zorzi IN CNAME zorzi.abc.co.uk.
zorzi IN CNAME zorzi.xyz.co.uk.

I would like to remove these lines from bourne shell script.

zeltweg IN CNAME zeltweg.xyz.co.uk.
zolder IN CNAME zolder.xyz.co.uk.
zonta IN CNAME zonta.xyz.co.uk.
zorzi IN CNAME zorzi.xyz.co.uk.

Sachin

[Darrell Allen]

Hi Sachin,

If you want to delete all lines ending with "xyz.co.uk." use:

grep -v "xyz.co.uk.$" file >newfile

Darrell

[MANOJ SRIVASTAVA]

Hi Sachin

cat filename | grep -v xyz > new filename


Manoj Srivastava

[Jeff Schussele]

Hi Sachin,

Use
grep -v "xyz.co.uk" filename > newfilename;mv filename filename.old;mv newfilename filename

HTH,
Jeff

[A. Clay Stephenson]

If it were me, I would use sed. First create a file with the delete instructions, called myfile.sed
e.g

/zeltweg IN CNAME zeltweg.xyz.co.uk./d
/zolder IN CNAME zolder.xyz.co.uk./d
/zonta IN CNAME zonta.xyz.co.uk./d
/zorzi IN CNAME zorzi.xyz.co.uk./d

Bracket each desired pattern with /pattern/ and then end each line with 'd' - the delete command.

Now
sed -f myfile.sed oldfile > newfile

[S.K. Chan]

Another variation here ..

# grep -v -E '\.xyz.' filename > newfile
==> filter out any lines that has ".xyz." string.

[Sachin Patel]

Sorry guys I didn't explain very much
some time it is different. I had post tail of my file and so it come out xyz.co.uk but at the beginning it is different.

In sort when I had two line with same system name regarless of domain name I would like to remove second system name.

I guss i am still not explaining clearly.

Sachin

[Sachin Patel]

Well I figure it out

last=null
for LINE in `cat file`
do
current = echo $LINE |awk '{print $1}'
if current != last
output line
fi
last=current
done

Sachin

[A. Clay Stephenson · Accepted solution]

Okay, try this:

sort oldfile | sort -k1,3 -u > newfile

The use of two sorts is intentional, I want to sort first on the entire line to force exact order and then sort on the first 3 fields and only pass lines that are unique in the first 3 fields.

[Sachin Patel]

Hi Clay,
Your last answer works like champ. That is why you are Olympian and I am not....

Sachin