A script for converting docx files into dokuwiki syntax

todokuwiki.pl

#!/usr/bin/perl

# convert docx to dokuwiki
# based on https://www.dokuwiki.org/tips:docpdf2dokuwiki

# requirements: libreoffice, HTML::WikiConverter

# running: ./todokuwiki.pl path/to/file/docx

use HTML::WikiConverter;
use File::Basename;
use File::Copy;

my $wc = new HTML::WikiConverter( dialect => 'DokuWiki' );

$txtpath = "/home/ilja/pages"; # location of generated .txt files
$imgpath = "/home/ilja/media"; # location of generated images

$file=$ARGV[0]; 
print "Processing ". $file ."\n";

$dirname = dirname($file);
$basename = basename($file);
$basename = lc $basename;
$file = "$dirname/$basename";
if ( $file=~ /docx*$/ ) {
    # convert docx to html
    `soffice --headless --convert-to html:"HTML" $file`
}
$htmfile = $file;
$htmfile =~ s/\..*//;
$htmfile = "$htmfile.html";
$newfile = $file;
$newfile =~ s/\..*//;
$newfile = "$newfile.txt";
if ( -e $htmfile ) {
    open (MYFILE, ">>$newfile");
    $input = $wc->html2wiki( file => $htmfile );
    print MYFILE "$input";
    close (MYFILE);
}

# copy parsed data to the prepared folders
$txtdirpath=$txtpath;
if ( ! -e $txtdirpath) {
    system("mkdir", "$txtdirpath");
}
@txt = <$dirname/*.txt>;
    foreach $txt (@txt) {
    copy($txt, $txtdirpath);
}
$imgdirpath=$imgpath;
if ( ! -e $imgdirpath) {
    system("mkdir", "$imgdirpath");
}
@img = <$dirname/*.png>;
foreach $img (@img) {
    copy($img, $imgdirpath);
}
@img = <$dirname/*.jpg>;
foreach $img (@img) {
    copy($img, $imgdirpath);
}