How to reformat named.conf's zone blocks into single lines?

Question

For reasons I won't go into here, I need to make a copy of an old CentOS 5.10 server's named.conf file and perform some text manipulation on the copy. The final result should consolidate each zone's info into a single, separate line. Plus, I need to script it so it can be run as a cron job. Here's an example of what I'm starting with:

zone "." {
        type hint;
        file "named.root";
};

zone "somedomain.com" {
        type master;
        file "somedomain.net";
        allow-transfer {
                11.11.11.11;
                22.22.22.22;
                common-allow-transfer;
        };
};

zone "otherdomain.com" {
        type master;
        file "otherdomain.com";
        allow-transfer {
                33.33.33.33;
                44.44.44.44;
                55.55.55.55;
                common-allow-transfer;
        };
};

And here's an example of what should appear in the edited file:

zone "." { type hint; file "named.root"; };
zone "somedomain.com" { type master; file "somedomain.net"; allow-transfer { 11.11.11.11; 22.22.22.22; common-allow-transfer; }; };
zone "otherdomain.com" { type master; file "otherdomain.com"; allow-transfer { 33.33.33.33; 44.44.44.44; 55.55.55.55; common-allow-transfer; }; };

My CIO and I unsuccessfully searched for existing scripts that convert named.conf into one-line-per-zone format. Using tr and sed, I've managed to 1) turn all tabs into spaces and 2) compress multiple contiguous spaces into a single space. The result is closer to what I need, but I'm still having trouble deleting unwanted newlines while keeping desired newlines. Here's example output which is close to being correct:

zone "." {
 type hint;
 file "named.root"; 
};
zone "somedomain.com" {
 type master;
 file "somedomain.net";
 allow-transfer {
 11.11.11.11;
 22.22.22.22;
 common-allow-transfer;
 };
};
zone "otherdomain.com" {
 type master;
 file "otherdomain.com";
 allow-transfer {
 33.33.33.33;
 44.44.44.44;
 55.55.55.55;
 common-allow-transfer;
 };
};

Tested a few suggested sed, awk, and tr expressions to solve the newline problem, but none of the expressions gave the desired output. Any suggestions you have--even a general direction like "experiment with sed and x,y,z"--would be very appreciated. Thanks!

Wildcard · Accepted Answer · 2018-08-22 16:12:18Z

4

Open up the file in Vim (or just vi) and type:

:g/^zone/.,/^}/j

Voila, you're done.

This runs globally, on every line beginning with zone, the command:

.,/^}/j

Which means, from the current line (.) to the next line that starts with a closing brace, join all those lines together.

If you need to script it, use vi's scripted form, ex:

printf '%s\n' 'g/^zone/.,/^}/j' x | ex file.txt

answered Aug 22, 2018 at 16:12

Wildcard

37.5k30 gold badges149 silver badges284 bronze badges

Add a comment |

glenn jackman · Accepted Answer · 2018-08-22 16:31:15Z

Try this:

sed -n '
    /^zone/,/^};/ H          # append current line to hold space in this range
    /^};/ {                  # at the end of the range:
        s/.*//               #   erase the current line
        x                    #   exchange pattern space with hold space
        s/\n//               #   remove first newline (before "zone")
        s/[[:space:]]\+/ /g  #   condense all other whitespace including newlines
        p                    #   and print this zone record
    }
' named.conf

result

zone "." { type hint; file "named.root"; };
zone "somedomain.com" { type master; file "somedomain.net"; allow-transfer { 11.11.11.11; 22.22.22.22; common-allow-transfer; }; };
zone "otherdomain.com" { type master; file "otherdomain.com"; allow-transfer { 33.33.33.33; 44.44.44.44; 55.55.55.55; common-allow-transfer; }; };

Check the sed documentation (the manual, or info sed) for the commands I used. — glenn jackman
– glenn jackman, Commented Aug 22, 2018 at 16:35

Alexander · Accepted Answer · 2018-08-22 16:14:49Z

0

sed -e ':nextline' -e N -e 's/\n\s*\([^z]\)/\1/' -e 'tnextline' named.conf

answered Aug 22, 2018 at 16:14

Alexander

1,4361 gold badge9 silver badges14 bronze badges

Thank you! This almost got me there. Just have to tweak a little. Your help is very appreciated.

keyop
– keyop

2018-08-22 21:23:53 +00:00
Commented Aug 22, 2018 at 21:23

Add a comment |

Rakesh Sharma · Accepted Answer · 2018-08-23 04:30:42Z

You could do it using the paragraph-mode of awk and Perl tools and the range operator , of the sed tool:

sed -e '
    /^zone/,/^$/!d
    H;/^zone/h;/^$/!{$!d;}
    g;s/[[:space:]]\{1,\}/ /g
    s/ $//
' input.file

awk -vRS= '{$1=$1}1' input.file

perl -pl -00e 's/\s+/ /g' input.file

perl -pl -00e 'tr/\t\040\n/ /s' input.file

Output:

zone "." { type hint; file "named.root"; };
zone "somedomain.com" { type master; file "somedomain.net"; allow-transfer { 11.11.11.11; 22.22.22.22; common-allow-transfer; }; };
zone "otherdomain.com" { type master; file "otherdomain.com"; allow-transfer { 33.33.33.33; 44.44.44.44; 55.55.55.55; common-allow-transfer; }; };

Stack Exchange Network

How to reformat named.conf's zone blocks into single lines?

4 Answers 4

You must log in to answer this question.

Hot Network Questions

How to reformat named.conf's zone blocks into single lines?

4 Answers 4

You must log in to answer this question.

Related

Hot Network Questions