I've tried use unoconv again to convert docx into html,
the final product was not really desirable, but was able to produce docx into html with couple tweeaks.
If anybody needs reference, below is the example code / unix line I've worked with.
`unoconv --stdout -f html "$docxfileloc" > "$htmfile"`;
my $t = HTML::TreeBuilder ->new_from_file("$upload_dir/$htmfile");
my $body = $t->look_down(_tag => q{body});
my @content = $body->detach_content; #grep body
my $html = $_->as_HTML for @content; #exclude <body>, </body> tag
$html = decode_entities($html); #decode special characters
i.e, If its possible, use
abiword for docx -> html. Output is better than unoconv, I just couldn't use it due to version compatibility issue.
Thank you for your replies and hope my solution help the other.