path: root/mktxt2text-1

#!/usr/bin/perl

# normalize text

use Getopt::Long;
use File::Slurp;
use Lingua::Sentence;

use strict;
use warnings;

# setup sentence splitter
my $splitter = Lingua::Sentence->new("en");

my $force;
GetOptions ("force|f");

my $stem = shift;
$_ = read_file( $stem . '.txt' );

# page header
s/^(:?24.3.2009|Official Journal of the European Union|EN)\n+//mg;

# headline
s/^(TITLE\h+(XX|XIX|XVIII|XVII|XVI|XV|XIV|XIII|XII|XI|X|IX|VIII|VII|VI|V|IV|III|II|I|))/\n#$1#\n\n/mg;
#s/^(Article\h+\d+[a-z]?)$/\n\n##$1##\n\n/mg;
s/^(\((\d+)\))(?:\s+(\S.*)(?:\n(?=\.)(\n\S.*))?)?/\n\n$1$3/mg;
#s/^(SECTION\h+\d+[a-z]?)/\n==$1==\n\n/mg;
#s/^(TITLE\h+\d+[a-z]?)/\n#$1#\n\n/mg;
#s/^(ANNEX\h+[IVX]+)$/\n===$1===\n\n/mg;
#s/^(?!\.)(Rule\h+\d+[a-z]?)\s+(\S.*)/\n====$1====\n\n$2\n\n/mg;

# join non-headline multi-lines, split into sentences, and split after comma
s/([^\n=])\n/$1 /g;
$_ = $splitter->split($_);
s/,\s(?=[^\v=]*\v)/,\n/mg;

write_file( $stem . '.mdwn', $_ );

print "DONE: $0 stem $stem\n";