#!/usr/bin/env perl 

use strict;

use utf8::all;

my $fl=0;

my $rom= qr{(?i:[ivx]+)};
my $no= qr{(\d+(?:\.?[oºaª]\.?))};

while(<>){
  # print "#";
  s{\s+$}{\n};
  unicode_fix();
  s{^[ \t]+}{} if $fl;
  $fl=0;
  s{^\s*(\(?(\d+|$rom|[a-z])\))}{\n==item($1)}i;
  s{^\s*([.*►•]|[‐‑–—\-])(\s)}{\n==item($1)$2}i;
  s{^\s*(\d+\.)(\s)}{\n==item($1)$2}i;
  s{^\s*(\d\.\d\.)(\s)}{\n==item($1)$2}i;
  s{^\s*(\d\.\d\.\d\.)(\s)}{\n==item($1)$2}i;
  s{^(\s*((
      C(?i:hapter)
     |C(?i:apítulo)
     |P(?i:arte?)
     |S(?i:ection)
     |S(?i:ec?ção)
     |A(?i:rticle)
     |A(?i:rtigo)
     |A(?i:nnex)
     |A(?i:nexo)
     |A(?i:ppendix)
     |A(?i:pêndice)
     |C(?i:onsiderando)
     |W(?i:hereas)
         ) \s+
     (\d+|$rom|$no|[A-Z]))
     )
     \s*$}{\n==SUBDIV= $2\n\n}x;
  s{\b[\-]\n}{} and $fl=1 ;
  print $_;
}

sub unicode_fix{
  s/eˆ/ê/g;
  s/aˆ/â/g;
  s/oˆ/ô/g;
  s/e´/é/g;
  s/a´/á/g;
  s/o´/ó/g;
  s/u´/ú/g;
  s/a˜/ã/g;
  s/o˜/õ/g;
  s/n˜/ñ/g;
  s/ı´/í/g;
}