使用bash脚本从文件中删除特定行

Question

I have a input file that contains DNS zone entries

; whatever.foo [000000]
$ORIGIN whatever.foo.
$TTL 8400
@       IN      SOA     ns1.whatever.foo.      admin.whatever.foo. (
2017101913 ; serial number
                    21600                    ; refresh
                    8400                     ; retry
                    1209600                 ; expire
                    8400                    ; ttl
                    )

xxx           IN    A    000.000.000.000
something-xxx IN    A    000.000.000.000
other-xxx     IN    A    000.000.000.000
else.xxx      IN    A    000.000.000.000

yyy           IN    A    000.000.000.001
something-yyy IN    A    000.000.000.001
other-yyy     IN    A    000.000.000.001
else.yyy      IN    A    000.000.000.001

How do I remove for example entries that belong to xxx subdomain domain?

Output should be

; whatever.foo [000000]
$ORIGIN whatever.foo.
$TTL 8400
@       IN      SOA     ns1.whatever.foo.      admin.whatever.foo. (
2017101913 ; serial number
                    21600                    ; refresh
                    8400                     ; retry
                    1209600                 ; expire
                    8400                    ; ttl
                    )

yyy           IN    A    000.000.000.001
something-yyy IN    A    000.000.000.001
other-yyy     IN    A    000.000.000.001
else.yyy      IN    A    000.000.000.001

I would like it to be in a bash script so I can call eg

delete_domain.sh xxx

Thanks.

Answer 1

One way:

Test file:

$ cat file
xxx           IN    A    000.000.000.000
else.xxx      IN    A    000.000.000.000
something-xxx IN    A    000.000.000.000
other-xxx     IN    A    000.000.000.000
abc           IN    A    000.000.000.000
$

Pattern file:

$ cat patterns
xxx
something-xxx
other-xxx
else.xxx
$

Awk option:

$ awk 'NR==FNR{a[$1];next}!(($1 in a) && $2=="IN" && $3=="A")' patterns file
abc           IN    A    000.000.000.000
$

First we load all the patterns to be removed into the associative array a. Then when the test file is processed, only those lines are printed whose 1st column is not a part of pattern and 2nd column is "IN" and 3rd column is "A".

Answer 2

Just a grep will do:

grep -E -v 'xxx\s+IN\s+A' file

And to get rid of multiple empty lines, pipe that into:

cat -s

Put those in a bash script:

$ cat test.sh
#!/bin/bash

grep -E -v 'xxx\s+IN\s+A' $1 | cat -s

and call it with:

./test.sh file