#! /usr/bin/env perl
eval 'exec perl -S $0 ${1+"$@"}'
  if $running_under_some_shell;

# po4a-normalize -- normalize documentation files
#
# Copyright 2002-2023 by SPI, inc.
#
# This program is free software; you can redistribute it and/or modify it
# under the terms of GPL v2.0 or later (see COPYING).

=encoding UTF-8

=head1 NAME

po4a-normalize - normalize a documentation file by parsing it in po4a, and writing it back

=head1 SYNOPSIS

B<po4a-normalize> B<-f> I<fmt> I<master.doc>

=head1 DESCRIPTION

The po4a (PO for anything) project goal is to ease translations (and more
interestingly, the maintenance of translations) using gettext tools on
areas where they were not expected like documentation.

The B<po4a-normalize> script is a debugging tool used to make sure that
po4a don't change the document when it's not supposed to. Only use it if
you're developing a new module, or if you doubt the sanity of the tools.

The generated document will be written to F<po4a-normalize.output> while the
generated POT file will be written to F<po4a-normalize.po> by default, but 
you can use the B<--localized> and B<--pot> options to change that.

=head1 OPTIONS

=over 4

=item B<-o>, B<--option>

Extra option(s) to pass to the format plugin. See the documentation of each
plugin for more information about the valid options and their meanings. For 
example, you could pass '-o tablecells' to the AsciiDoc parser, while the 
text parser would accept '-o tabs=split'.

=item B<-w>, B<--width>

Column at which we should wrap the resulting file if the format supports it (default: 76).
If the provided value is 0 or negative, then the resulting file will not be wraped at all.
This is useful with some formats that could break when rewraped.

=item B<-b>, B<--blank>

Create a blank translated document.
The generated translated document will be generated assuming all messages
are translated by a space or new line.

This is useful to check what parts of the document cannot be translated.

Both B<--blank> and B<--capitalize> can't be specified at the same time.

=item B<-C>, B<--capitalize>

Create a translated document and corresponding po-file with original strings
translated as their capitalized versions.

This is useful to check what parts of the document cannot be translated and
generate test data for po4a.

Both B<--blank> and B<--capitalize> can't be specified at the same time.

=item B<-h>, B<--help>

Show a short help message.

=item B<--help-format>

List the documentation formats understood by po4a.

=item B<-f>, B<--format>

Format of the documentation you want to handle. Use the B<--help-format>
option to see the list of available formats.

=item B<-M>, B<--master-charset>

Charset of the file containing the document to translate.

=item B<-l>, B<--localized>

Name of the normalized file to produce (I<po4a-normalize.output> by default). 

=item B<-p>, B<--pot>

Pot file to produce (I<po4a-normalize.po> by default).

=item B<-V>, B<--version>

Display the version of the script and exit.

=back

=head1 SEE ALSO

L<po4a-gettextize(1)>,
L<po4a-translate(1)>,
L<po4a-updatepo(1)>,
L<po4a(7)>

=head1 AUTHORS

 Denis Barbier <barbier@linuxfr.org>
 Nicolas François <nicolas.francois@centraliens.net>
 Martin Quinson (mquinson#debian.org)

=head1 COPYRIGHT AND LICENSE

Copyright 2002-2023 by SPI, inc.

This program is free software; you may redistribute it and/or modify it
under the terms of GPL v2.0 or later (see the COPYING file).

=cut

use 5.16.0;
use strict;
use warnings;

use Locale::Po4a::Chooser;
use Locale::Po4a::TransTractor;
use Locale::Po4a::Common qw(wrap_msg gettext dgettext);

use Getopt::Long qw(GetOptions);

use Pod::Usage qw(pod2usage);

Locale::Po4a::Common::textdomain('po4a');

sub show_version {
    Locale::Po4a::Common::show_version("po4a-normalize");
    exit 0;
}

my ( $help_fmt, $help, $type, $debug, $verbose, $quiet, $wrappo, @options, $no_deprecation );
my ( $blank, $capitalize );
my ($mastchar);
my ( $localized, $potfile, $width ) = ( 'po4a-normalize.output', 'po4a-normalize.po', 76 );
Getopt::Long::Configure( 'no_auto_abbrev', 'no_ignore_case' );
GetOptions(
    'help|h'        => \$help,
    'help-format'   => \$help_fmt,
    'format|f=s'    => \$type,
    'localized|l=s' => \$localized,
    'pot|p=s'       => \$potfile,

    'blank|b'      => \$blank,
    'capitalize|C' => \$capitalize,

    'master-charset|M=s' => \$mastchar,

    # undocumented option to silence the warning about po4a-normalize not being intended for users
    'no-deprecation' => \$no_deprecation,

    'option|o=s' => \@options,
    'wrap-po=s'  => \$wrappo,

    'width|w=s' => \$width,
    'verbose|v' => \$verbose,
    'quiet|q'   => \$quiet,
    'debug|d'   => \$debug,
    'version|V' => \&show_version
) or pod2usage();

$help     && pod2usage( -verbose => 1, -exitval => 0 );
$help_fmt && Locale::Po4a::Chooser::list(0);
pod2usage() unless scalar @ARGV == 1;

my %options = (
    "verbose" => $verbose,
    "debug"   => $debug,
    "wrapcol" => $width
);
if ( $blank and $capitalize ) {
    die wrap_msg( gettext("Both B<--blank> and B<--capitalize> can't be specified at the same time.") );
}
foreach (@options) {
    if (m/^([^=]*)=(.*)$/) {
        $options{$1} = "$2";
    } else {
        $options{$_} = 1;
    }
}

unless ($no_deprecation) {
    print wrap_msg(
        gettext(
                "po4a-normalize is mostly useful to debug po4a itself. "
              . "Users are advised to use the po4a(1) program instead."
        )
    );
}
my $parser = Locale::Po4a::Chooser::new( $type, %options );

my $filename = shift || pod2usage(1);
$filename eq '-' || -e $filename || die wrap_msg( gettext("File %s does not exist."), $filename );
die wrap_msg( gettext("Options %s and %s cannot be provided together"), "quiet", "verbose" ) if $verbose && $quiet;

$parser->read( $filename, $filename, $mastchar // '' );
my %pot_options = ( 'wrap-po' => $wrappo );
$parser->setpoout( Locale::Po4a::Po->new( \%pot_options ) );
$parser->parse();
if ( $blank or $capitalize ) {
    foreach my $msgid ( keys %{ $parser->{TT}{po_out}{po} } ) {
        if ($blank) {
            if ( $msgid =~ m/\n$/s ) {
                $parser->{TT}{po_out}{po}{$msgid}{'msgstr'} = "\n";
            } else {
                $parser->{TT}{po_out}{po}{$msgid}{'msgstr'} = " ";
            }
        } else {    # $capitalize
            ( $parser->{TT}{po_out}{po}{$msgid}{'msgstr'} = $msgid ) =~ s/(?<!\\)(.)/\U$1/g;
        }
    }
    my $modified_po = $parser->{TT}{po_out};
    $parser = Locale::Po4a::Chooser::new( $type, %options );
    $parser->{TT}{po_in} = $modified_po;
    $parser->read( $filename, $filename, $mastchar // '' );
    $parser->{TT}{file_in_charset} = $mastchar;
    $parser->parse();
    $parser->{TT}{po_out} = $modified_po;
}
print wrap_msg( dgettext( "po4a", "Write the normalized document to %s." ), $localized )
  unless $quiet;
$parser->write( $localized, '' );
print wrap_msg( dgettext( "po4a", "Write the normalized PO file to %s." ), $potfile )
  unless $quiet;
$parser->writepo($potfile);

__END__