Splitting Apache Vhosts into Individual Files

Question

I have several vhosts in one file that I am looking to break each into a file of their own corresponding to the ServerName. I'm looking to take this file

<VirtualHost *:80>
  ServerName host1.example.com
</VirtualHost>

<VirtualHost *:80>
  ServerName host2.example.com
</VirtualHost>

<VirtualHost *:80>
  ServerName hostN.example.com
</VirtualHost>

to subfiles host1.example.com ... hostN.example.com.

I looked at this question-answer, which is closest to what I'm looking for, but I'm unsure on how to match two lines. It tried a nested if statement,

gawk '{
  if(match($0, /(\<VirtualHost \*:80\>)/, v)){
    if(match($0, /  ServerName (.*)/, s)){
      name=s[1]
      {print name}

    }}}
' samplevhost

and

gawk '{
  if(match($0, /(\<VirtualHost \*:80\>)(\n  )(ServerName)(.*)/, k)){
      name=k[1]
      {print name}

    }}
' samplevhost

to print to screen, but I don't get any output. What am I missing?

Answer 1

If you're sure that vhost definitions are separated by an empty line, you can make use of Perl's "paragraph mode" slurp which segments its input file into chunks of lines separated by one or more empty lines and processes the file chunk-by-chunk rather than line by line (which is probably what's causing your gawk script to fail). Use -n00 (see perlrun(1)) to process the file "paragraph-by-paragraph":

perl -n00 -e '
    /ServerName\s+(.*)/;
    $file = $1;
    open $fh,">",$file or die "Failed to open file $file for writing: $!\n";
    print $fh $_;
    close $fh' your_file_here