osamu2001/po2tmx.pl

## po2tmx.pl
#!/usr/bin/perl
use strict;
use warnings;
use utf8;
use Encode;

my $fcode = "";
my $mode  = "";
my $msgid = "";
my $msgstr = "";

&print_tmx_header();

while(<STDIN>){
  if($fcode eq ""){
    $fcode = $1  if (/charset=([-a-zA-Z0-9]+)/);
    next;
  }
  my $line = decode($fcode, $_);
  chomp $line;
  if((0 == length($line  ))&&(0 != length($msgstr))&&(0 != length($msgid ))){
    &print_tuv_en_b();
    print encode("utf8",&HTMLSanitize($msgid));
    &print_tuv_en_e();
    &print_tuv_ja_b();
    print encode("utf8",&HTMLSanitize($msgstr));
    &print_tuv_ja_e();
    next;
  }
  if($mode eq "msgid"){
    if ($line =~ /^"(.*)"$/){
      my $str = $1;
      $str =~ s/\\n/\n/;
      if(0==length($msgid)){
        $msgid = $str;
      }else{
        $msgid = $msgid . $str;
      }
      next;
    }
  }
  if($mode eq "msgstr"){
    if ($line =~ /^"(.*)"$/){
      my $str = $1;
      $str =~ s/\\n/\n/;
      if(0==length($msgstr)){
        $msgstr = $str;
      }else{
        $msgstr = $msgstr . $str;
      }
      next;
    }
  }
  if ($line =~ /^msgid "(.*)"$/){
    $mode = "msgid";
    $msgid = $1;
    next;
  }
  if($line =~ /^msgstr "(.*)"$/){
    $mode = "msgstr";
    $msgstr = $1;
    next;
  }
}

&print_tmx_footer();

sub print_tuv{
  my ($ens,$jas) = @_;
  &print_tuv_en_b();
  print encode("utf8",$ens);
  &print_tuv_en_e();
  &print_tuv_ja_b();
  print encode("utf8",$jas);
  &print_tuv_ja_e();
}
sub print_tuv_en_b(){
  print << "TUV_EN";
<tu>
<tuv lang="EN-US">
TUV_EN
  print "<seg>";
}
sub print_tuv_en_e(){
  print << "TUV_EN";
</seg>
</tuv>
TUV_EN
}

sub print_tuv_ja_b(){
  print << "TUV_JA";
<tuv lang="JA">
TUV_JA
  print "<seg>";
}
sub print_tuv_ja_e(){
  print << "TUV_JA";
</seg>
</tuv>
</tu>
TUV_JA
}

sub print_tmx_footer(){
  print << "TMX_FOOTER";
</body>
</tmx>
TMX_FOOTER
}

sub print_tmx_header(){
print << "TMX_HEADER";
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE tmx SYSTEM "tmx11.dtd">
<tmx version="1.1">
  <header
    creationtool="OmegaT"
    creationtoolversion="2.0.2_1"
    segtype="sentence"
    o-tmf="OmegaT TMX"
    adminlang="EN-US"
    srclang="EN-US"
    datatype="plaintext"
  >
  </header>
  <body>
TMX_HEADER
}

sub HTMLSanitize {
    my $str = shift;
    $str =~ s/&/&amp;/g;
    $str =~ s/</&lt;/g;
    $str =~ s/>/&gt;/g;
    $str =~ s/"/&quot;/g;
    return $str;
}
	#!/usr/bin/perl
	use strict;
	use warnings;
	use utf8;
	use Encode;

	my $fcode = "";
	my $mode = "";
	my $msgid = "";
	my $msgstr = "";

	&print_tmx_header();

	while(<STDIN>){
	if($fcode eq ""){
	$fcode = $1 if (/charset=([-a-zA-Z0-9]+)/);
	next;
	}
	my $line = decode($fcode, $_);
	chomp $line;
	if((0 == length($line ))&&(0 != length($msgstr))&&(0 != length($msgid ))){
	&print_tuv_en_b();
	print encode("utf8",&HTMLSanitize($msgid));
	&print_tuv_en_e();
	&print_tuv_ja_b();
	print encode("utf8",&HTMLSanitize($msgstr));
	&print_tuv_ja_e();
	next;
	}
	if($mode eq "msgid"){
	if ($line =~ /^"(.*)"$/){
	my $str = $1;
	$str =~ s/\\n/\n/;
	if(0==length($msgid)){
	$msgid = $str;
	}else{
	$msgid = $msgid . $str;
	}
	next;
	}
	}
	if($mode eq "msgstr"){
	if ($line =~ /^"(.*)"$/){
	my $str = $1;
	$str =~ s/\\n/\n/;
	if(0==length($msgstr)){
	$msgstr = $str;
	}else{
	$msgstr = $msgstr . $str;
	}
	next;
	}
	}
	if ($line =~ /^msgid "(.*)"$/){
	$mode = "msgid";
	$msgid = $1;
	next;
	}
	if($line =~ /^msgstr "(.*)"$/){
	$mode = "msgstr";
	$msgstr = $1;
	next;
	}
	}

	&print_tmx_footer();

	sub print_tuv{
	my ($ens,$jas) = @_;
	&print_tuv_en_b();
	print encode("utf8",$ens);
	&print_tuv_en_e();
	&print_tuv_ja_b();
	print encode("utf8",$jas);
	&print_tuv_ja_e();
	}
	sub print_tuv_en_b(){
	print << "TUV_EN";
	<tu>
	<tuv lang="EN-US">
	TUV_EN
	print "<seg>";
	}
	sub print_tuv_en_e(){
	print << "TUV_EN";
	</seg>
	</tuv>
	TUV_EN
	}

	sub print_tuv_ja_b(){
	print << "TUV_JA";
	<tuv lang="JA">
	TUV_JA
	print "<seg>";
	}
	sub print_tuv_ja_e(){
	print << "TUV_JA";
	</seg>
	</tuv>
	</tu>
	TUV_JA
	}

	sub print_tmx_footer(){
	print << "TMX_FOOTER";
	</body>
	</tmx>
	TMX_FOOTER
	}

	sub print_tmx_header(){
	print << "TMX_HEADER";
	<?xml version="1.0" encoding="UTF-8"?>
	<!DOCTYPE tmx SYSTEM "tmx11.dtd">
	<tmx version="1.1">
	<header
	creationtool="OmegaT"
	creationtoolversion="2.0.2_1"
	segtype="sentence"
	o-tmf="OmegaT TMX"
	adminlang="EN-US"
	srclang="EN-US"
	datatype="plaintext"
	>
	</header>
	<body>
	TMX_HEADER
	}

	sub HTMLSanitize {
	my $str = shift;
	$str =~ s/&/&/g;
	$str =~ s/</</g;
	$str =~ s/>/>/g;
	$str =~ s/"/"/g;
	return $str;
	}