#!/usr/bin/awk -f
# extract elements from XML
# usage: awk -f tag.awk tag=tag.class file
# BUG: May not handle weird nesting well,
#      because end tags do not have to have the class attribute.

BEGIN {
  tag = "pre";
  printing = 0;
}

# This should happen before the first line,
# but after command line definitions.
NR < 2 {
  class = "";
  n = split( tag, a, "." );
  if ( 1 < n ) {
    tag = a[1];
    class = a[2];
  }
  startTag = "<" tag ">";
  if ( 0 < length( class ) ) {
    startTag = "<" tag "[^>]*class *= *\"" class "\"[^>]*>";
  }
  endTag = "<\/" tag ">";
}

# For a "one liner":
$0 ~ startTag && $0 ~ endTag {
  print "one liner";
  print startTag, endTag;
  print;
  n = split($0, a, startTag);
  print n, a[2];
  n = split(a[2], b, endTag);
  print b[1];
  next;
}

$0 ~ endTag {
  if ( 0 < printing ) {
    printing--;
  }
}

{
  if ( 0 < printing ) {
    print;
  }
}

$0 ~ startTag {
  printing++;
}