3 # -n do not include custom_ scripts
4 # -v verbose mode, shows progress stuff
6 # this version of locles processes not only all required .pl files
7 # but also all parse_html_templated files.
13 unshift(@INC, 'modules/override'); # Use our own versions of various modules (e.g. YAML).
14 push (@INC, 'modules/fallback'); # Only use our own versions of modules if there's no system version.
21 use File::Slurp qw(slurp);
25 use List::MoreUtils qw(apply);
26 use List::Util qw(first);
29 $OUTPUT_AUTOFLUSH = 1;
40 my $basedir = "../..";
41 my $locales_dir = ".";
42 my $bindir = "$basedir/bin/mozilla";
43 my @progdirs = ( "$basedir/SL" );
44 my @menufiles = <${basedir}/menus/*.ini>;
45 my @javascript_dirs = ($basedir .'/js', $basedir .'/templates/webpages');
46 my $javascript_output_dir = $basedir .'/js';
47 my $submitsearch = qr/type\s*=\s*[\"\']?submit/i;
52 my %ignore_unused_templates = (
53 map { $_ => 1 } qw(ct/testpage.html generic/autocomplete.html oe/periodic_invoices_email.txt part/testpage.html t/render.html t/render.js task_server/failure_notification_email.txt
54 failed_background_jobs_report/email.txt)
57 my (%referenced_html_files, %locale, %htmllocale, %alllocales, %cached, %submit, %jslocale);
58 my ($ALL_HEADER, $MISSING_HEADER, $LOST_HEADER);
63 my ($top_dir_name) = @_;
70 tie my %dir_h, 'IO::Dir', $dir_name;
72 push @files, grep { -f } map { "${dir_name}/${_}" } keys %dir_h;
73 my @sub_dirs = grep { -d } map { "${dir_name}/${_}" } grep { ! m/^\.\.?$/ } keys %dir_h;
75 $finder->($_) for @sub_dirs;
78 $finder->($top_dir_name);
84 # overwrite existing entries with the ones from 'missing'
85 $self->{texts}->{$_} = $missing->{$_} for grep { $missing->{$_} } keys %alllocales;
87 # try to set missing entries from lost ones
88 my %lost_by_text = map { ($_->{text} => $_->{translation}) } @lost;
89 $self->{texts}->{$_} = $lost_by_text{$_} for grep { !$self->{texts}{$_} } keys %alllocales;
92 my @bindir_files = find_files($bindir);
93 my @progfiles = map { m:^(.+)/([^/]+)$:; [ $2, $1 ] } grep { /\.pl$/ && !/_custom/ } @bindir_files;
94 my @customfiles = grep /_custom/, @bindir_files;
96 push @progfiles, map { m:^(.+)/([^/]+)$:; [ $2, $1 ] } grep { /\.pm$/ } map { find_files($_) } @progdirs;
98 # put customized files into @customfiles
104 tie %dir_h, 'IO::Dir', $basedir;
105 push @menufiles, map { "$basedir/$_" } grep { /.*_menu.ini$/ } keys %dir_h;
109 foreach my $sub_dir ("Pg-upgrade2", "Pg-upgrade2-auth") {
110 my $dir = "$basedir/sql/$sub_dir";
111 tie %dir_h, 'IO::Dir', $dir;
112 push @dbplfiles, map { [ $_, $dir ] } grep { /\.pl$/ } keys %dir_h;
115 # slurp the translations in
116 if (-f "$locales_dir/all") {
117 require "$locales_dir/all";
119 if (-f "$locales_dir/missing") {
120 require "$locales_dir/missing" ;
121 unlink "$locales_dir/missing";
123 if (-f "$locales_dir/lost") {
124 require "$locales_dir/lost";
125 unlink "$locales_dir/lost";
128 my %old_texts = %{ $self->{texts} || {} };
130 handle_file(@{ $_ }) for @progfiles;
131 handle_file(@{ $_ }) for @dbplfiles;
132 scanmenu($_) for @menufiles;
134 for my $file_name (grep { /\.(?:js|html)$/i } map({find_files($_)} @javascript_dirs)) {
135 scan_javascript_file($file_name);
138 # merge entries to translate with entries from files 'missing' and 'lost'
143 file => "$locales_dir/all",
144 header => $ALL_HEADER,
145 data_name => '$self->{texts}',
146 data_sub => sub { _print_line($_, $self->{texts}{$_}, @_) for sort keys %alllocales },
149 open(my $js_file, '>:encoding(utf8)', $javascript_output_dir .'/locale/'. $locale .'.js') || die;
150 print $js_file 'namespace("kivi").setupLocale({';
152 for my $key (sort(keys(%jslocale))) {
153 print $js_file ((!$first_entry ? ',' : '') ."\n". _double_quote($key) .':'. _double_quote($self->{texts}{$key}));
156 print $js_file ("\n");
157 print $js_file ('});'."\n");
160 foreach my $text (keys %$missing) {
161 if ($locale{$text} || $htmllocale{$text}) {
162 unless ($self->{texts}{$text}) {
163 $self->{texts}{$text} = $missing->{$text};
169 # calc and generate missing
170 my @new_missing = grep { !$self->{texts}{$_} } sort keys %alllocales;
174 my %existing_lc = map { (lc $_ => $_) } grep { $self->{texts}->{$_} } keys %{ $self->{texts} };
175 foreach my $entry (@new_missing) {
176 my $other = $existing_lc{lc $entry};
177 print "W: No entry for '${entry}' exists, but there is one with different case: '${other}'\n" if $other;
182 for my $string (@new_missing) {
183 print "new string '$string' in files:\n";
186 apply { s{^(?:\.\./)+}{} }
187 grep { $cached{$_}{all}{$string} }
193 file => "$locales_dir/missing",
194 header => $MISSING_HEADER,
195 data_name => '$missing',
196 data_sub => sub { _print_line($_, '', @_) for @new_missing },
200 # calc and generate lost
201 while (my ($text, $translation) = each %old_texts) {
202 next if ($alllocales{$text});
203 push @lost, { 'text' => $text, 'translation' => $translation };
207 splice @lost, 0, (scalar @lost - 50) if (scalar @lost > 50);
209 file => "$locales_dir/lost",
210 header => $LOST_HEADER,
212 data_name => '@lost',
214 _print_line($_->{text}, $_->{translation}, @_, template => " { 'text' => %s, 'translation' => %s },\n") for @lost;
219 my $trlanguage = slurp("$locales_dir/LANGUAGE");
222 search_unused_htmlfiles() if $opt_c;
224 my $count = scalar keys %alllocales;
225 my $notext = scalar @new_missing;
226 my $per = sprintf("%.1f", ($count - $notext) / $count * 100);
227 print "\n$trlanguage - ${per}%";
228 print " - $notext/$count missing" if $notext;
237 # These are all the texts to build the translations files.
238 # The file has the form of 'english text' => 'foreign text',
239 # you can add the translation in this file or in the 'missing' file
240 # run locales.pl from this directory to rebuild the translation files
242 $MISSING_HEADER = <<EOL;
243 # add the missing texts and run locales.pl to rebuild
245 $LOST_HEADER = <<EOL;
246 # The last 50 text strings, that have been removed.
247 # This file has been auto-generated by locales.pl. Please don't edit!
254 my ($opt_no_c, $ignore_for_compatiblity);
257 'no-custom-files' => \$opt_n,
258 'check-files' => \$ignore_for_compatiblity,
259 'no-check-files' => \$opt_no_c,
260 'verbose' => \$opt_v,
261 'filenames' => \$opt_f,
275 pod2usage(-exitstatus => 0, -verbose => 2);
280 my $arg = shift @ARGV;
282 foreach my $dir ("../locale/$arg", "locale/$arg", "../$arg", $arg) {
283 next unless -d $dir && -f "$dir/all" && -f "$dir/LANGUAGE";
292 print "The locale directory '$arg' could not be found.\n";
296 } elsif (!-f 'all' || !-f 'LANGUAGE') {
297 print "locales.pl was not called from a locale/* subdirectory,\n"
298 . "and no locale directory name was given.\n";
302 $locale ||= (grep { $_ } split m:/:, getcwd())[-1];
307 my ($file, $dir) = @_;
308 print "\n$file" if $opt_v;
312 &scanfile("$dir/$file");
314 # scan custom_{module}.pl or {login}_{module}.pl files
315 foreach my $customfile (@customfiles) {
316 if ($customfile =~ /_$file/) {
317 if (-f "$dir/$customfile") {
318 &scanfile("$dir/$customfile");
326 sub extract_text_between_parenthesis {
327 my ($fh, $line) = @_;
328 my ($inside_string, $pos, $text, $quote_next) = (undef, 0, "", 0);
331 if (length($line) <= $pos) {
333 return ($text, "") unless ($line);
337 my $cur_char = substr($line, $pos, 1);
339 if (!$inside_string) {
340 if ((length($line) >= ($pos + 3)) && (substr($line, $pos, 2)) eq "qq") {
341 $inside_string = substr($line, $pos + 2, 1);
344 } elsif ((length($line) >= ($pos + 2)) &&
345 (substr($line, $pos, 1) eq "q")) {
346 $inside_string = substr($line, $pos + 1, 1);
349 } elsif (($cur_char eq '"') || ($cur_char eq '\'')) {
350 $inside_string = $cur_char;
352 } elsif (($cur_char eq ")") || ($cur_char eq ',')) {
353 return ($text, substr($line, $pos + 1));
358 $text .= '\\' unless $cur_char eq "'";
362 } elsif ($cur_char eq '\\') {
365 } elsif ($cur_char eq $inside_string) {
366 undef($inside_string);
379 my $dont_include_subs = shift;
380 my $scanned_files = shift;
385 $scanned_files = {} unless ($scanned_files);
386 return if ($scanned_files->{$file});
387 $scanned_files->{$file} = 1;
389 if (!defined $cached{$file}) {
391 return unless (-f "$file");
393 my $fh = new FileHandle;
394 open $fh, '<:encoding(utf8)', $file or die "$! : $file";
396 my ($is_submit, $line_no, $sub_line_no) = (0, 0, 0);
399 last if /^\s*__END__/;
403 # is this another file
404 if (/require\s+\W.*\.pl/) {
406 $newfile =~ s/require\s+\W//;
407 $newfile =~ s|bin/mozilla||;
408 $cached{$file}{scan}{"$bindir/$newfile"} = 1;
409 } elsif (/use\s+SL::([\w:]*)/) {
412 $cached{$file}{scannosubs}{"../../SL/${module}.pm"} = 1;
415 # Some calls to render() are split over multiple lines. Deal
417 while (/(?:parse_html_template2?|render)\s*\( *$/) {
422 # is this a template call?
423 if (/(?:parse_html_template2?|render)\s*\(\s*[\"\']([\w\/]+)\s*[\"\']/) {
424 my $new_file_base = "$basedir/templates/webpages/$1.";
425 if (/parse_html_template2/) {
426 print "E: " . strip_base($file) . " is still using 'parse_html_template2' for " . strip_base("${new_file_base}html") . ".\n";
430 foreach my $ext (qw(html js json)) {
431 my $new_file = "${new_file_base}${ext}";
433 $cached{$file}{scanh}{$new_file} = 1;
439 if ($opt_c && !$found_one) {
440 print "W: missing HTML template: " . strip_base($new_file_base) . "{html,json,js} (referenced from " . strip_base($file) . ")\n";
449 my ($null, $country) = split(/,/);
450 $country =~ s/^ +[\"\']//;
451 $country =~ s/[\"\'].*//;
457 # is it a submit button before $locale->
458 if (/$submitsearch/) {
460 if ($` !~ /locale->text/) {
462 $sub_line_no = $line_no;
467 if (/ (?: locale->text | \b t8 ) \b .*? \(/x) {
474 ($string, $_) = extract_text_between_parenthesis($fh, $postmatch);
477 # if there is no $ in the string record it
478 unless (($string =~ /\$\D.*/) || ("" eq $string)) {
480 # this guarantees one instance of string
481 $cached{$file}{locale}{$string} = 1;
483 # this one is for all the locales
484 $cached{$file}{all}{$string} = 1;
486 # is it a submit button before $locale->
488 $cached{$file}{submit}{$string} = 1;
491 } elsif ($postmatch =~ />/) {
495 # exit loop if there are no more locales on this line
496 ($rc) = ($postmatch =~ /locale->text | \b t8/x);
498 if ( ($postmatch =~ />/)
499 || (!$found && ($sub_line_no != $line_no) && />/)) {
509 $alllocales{$_} = 1 for keys %{$cached{$file}{all}};
510 $locale{$_} = 1 for keys %{$cached{$file}{locale}};
511 $submit{$_} = 1 for keys %{$cached{$file}{submit}};
513 scanfile($_, 0, $scanned_files) for keys %{$cached{$file}{scan}};
514 scanfile($_, 1, $scanned_files) for keys %{$cached{$file}{scannosubs}};
515 scanhtmlfile($_) for keys %{$cached{$file}{scanh}};
517 $referenced_html_files{$_} = 1 for keys %{$cached{$file}{scanh}};
523 my $fh = new FileHandle;
524 open $fh, '<:encoding(utf8)', $file or die "$! : $file";
526 my @a = grep m/^\[/, <$fh>;
530 grep { s/(\[|\])//g } @a;
532 foreach my $item (@a) {
533 my @b = split /--/, $item;
534 foreach my $string (@b) {
536 $locale{$string} = 1;
537 $alllocales{$string} = 1;
543 sub unescape_template_string {
554 return if defined $cached{$file};
556 my %plugins = ( 'loaded' => { }, 'needed' => { } );
558 if (!open(IN, '<:encoding(utf8)', $file)) {
559 print "E: template file '$file' not found\n";
566 while (my $line = <IN>) {
569 while ($line =~ m/\[\%[^\w]*use[^\w]+(\w+)[^\w]*?\%\]/gi) {
570 $plugins{loaded}->{$1} = 1;
573 while ($line =~ m/\[\%[^\w]*(\w+)\.\w+\(/g) {
575 $plugins{needed}->{$plugin} = 1 if (first { $_ eq $plugin } qw(HTML LxERP JavaScript JSON L P));
578 $plugins{needed}->{T8} = 1 if $line =~ m/\[\%.*\|.*\$T8/;
580 while ($line =~ m/(?: # Start von Variante 1: LxERP.t8('...'); ohne darumliegende [% ... %]-Tags
581 (LxERP\.t8)\( # LxERP.t8( ::Parameter $1::
582 ([\'\"]) # Anfang des zu übersetzenden Strings ::Parameter $2::
583 (.*?) # Der zu übersetzende String ::Parameter $3::
584 (?<!\\)\2 # Ende des zu übersetzenden Strings
585 | # Start von Variante 2: [% '...' | $T8 %]
586 \[\% # Template-Start-Tag
587 [\-~#]? # Whitespace-Unterdrückung
588 \s* # Optional beliebig viele Whitespace
589 ([\'\"]) # Anfang des zu übersetzenden Strings ::Parameter $4::
590 (.*?) # Der zu übersetzende String ::Parameter $5::
591 (?<!\\)\4 # Ende des zu übersetzenden Strings
592 \s*\|\s* # Pipe-Zeichen mit optionalen Whitespace davor und danach
593 (\$T8) # Filteraufruf ::Parameter $6::
594 .*? # Optionale Argumente für den Filter
596 [\-~#]? # Whitespace-Unterdrückung
597 \%\] # Template-Ende-Tag
600 my $module = $1 || $6;
601 my $string = $3 || $5;
602 print "Found filter >>>$string<<<\n" if $debug;
603 substr $line, $LAST_MATCH_START[1], $LAST_MATCH_END[0] - $LAST_MATCH_START[0], '';
605 $string = unescape_template_string($string);
606 $cached{$file}{all}{$string} = 1;
607 $cached{$file}{html}{$string} = 1;
608 $cached{$file}{submit}{$string} = 1 if $PREMATCH =~ /$submitsearch/;
609 $plugins{needed}->{T8} = 1 if $module eq '$T8';
610 $plugins{needed}->{LxERP} = 1 if $module eq 'LxERP.t8';
613 while ($line =~ m/\[\% # Template-Start-Tag
614 [\-~#]? # Whitespace-Unterdrückung
615 \s* # Optional beliebig viele Whitespace
616 (?: # Die erkannten Template-Direktiven
621 \s+ # Mindestens ein Whitespace
622 [\'\"]? # Anfang des Dateinamens
623 ([^\s]+) # Beliebig viele Nicht-Whitespaces -- Dateiname
624 \.(html|js) # Endung ".html" oder ".js", ansonsten kann es der Name eines Blocks sein
626 my $new_file_name = "$basedir/templates/webpages/$1.$2";
627 $cached{$file}{scanh}{$new_file_name} = 1;
628 substr $line, $LAST_MATCH_START[1], $LAST_MATCH_END[0] - $LAST_MATCH_START[0], '';
634 foreach my $plugin (keys %{ $plugins{needed} }) {
635 next if ($plugins{loaded}->{$plugin});
636 print "E: " . strip_base($file) . " requires the Template plugin '$plugin', but is not loaded with '[\% USE $plugin \%]'.\n";
639 # copy back into global arrays
640 $alllocales{$_} = 1 for keys %{$cached{$file}{all}};
641 $locale{$_} = 1 for keys %{$cached{$file}{html}};
642 $submit{$_} = 1 for keys %{$cached{$file}{submit}};
644 scanhtmlfile($_) for keys %{$cached{$file}{scanh}};
646 $referenced_html_files{$_} = 1 for keys %{$cached{$file}{scanh}};
649 sub scan_javascript_file {
652 open(my $fh, '<:encoding(utf8)', $file) || die('can not open file: '. $file);
654 while( my $line = readline($fh) ) {
665 my $text = unescape_template_string($2);
667 $jslocale{$text} = 1;
668 $alllocales{$text} = 1;
674 sub search_unused_htmlfiles {
675 my @unscanned_dirs = ('../../templates/webpages');
677 while (scalar @unscanned_dirs) {
678 my $dir = shift @unscanned_dirs;
680 foreach my $entry (<$dir/*>) {
682 push @unscanned_dirs, $entry;
684 } elsif (!$ignore_unused_templates{strip_base($entry)} && -f $entry && !$referenced_html_files{$entry}) {
685 print "W: unused HTML template: " . strip_base($entry) . "\n";
693 my $s = "$_[0]"; # Create a copy of the string.
696 $s =~ s|templates/webpages/||;
703 $val =~ s/(\'|\\$)/\\$1/g;
704 return "'" . $val . "'";
709 $val =~ s/(\"|\\$)/\\$1/g;
710 return '"'. $val .'"';
714 my $key = _single_quote(shift);
715 my $text = _single_quote(shift);
717 my $template = $params{template} || qq| %-29s => %s,\n|;
718 my $fh = $params{fh} || croak 'need filehandle in _print_line';
720 print $fh sprintf $template, $key, $text;
726 my $file = $params{file} || croak 'need filename in generate_file';
727 my $header = $params{header};
728 my $lines = $params{data_sub};
729 my $data_name = $params{data_name};
730 my @delim = split //, ($params{delim} || '{}');
732 open my $fh, '>:encoding(utf8)', $file or die "$! : $file";
734 print $fh "#!/usr/bin/perl\n# -*- coding: utf-8; -*-\n# vim: fenc=utf-8\n\nuse utf8;\n\n";
735 print $fh $header, "\n" if $header;
736 print $fh "$data_name = $delim[0]\n" if $data_name;
740 print $fh qq|$delim[1];\n\n1;\n|;
748 locales.pl - Collect strings for translation in kivitendo
752 locales.pl [options] lang_code
755 -n, --no-custom-files Do not process files whose name contains "_"
756 -c, --check-files Run extended checks on HTML files
757 -f, --filenames Show the filenames where new strings where found
758 -v, --verbose Be more verbose
759 -h, --help Show this help
765 =item B<-n>, B<--no-custom-files>
767 Do not process files whose name contains "_", e.g. "custom_io.pl".
769 =item B<-c>, B<--check-files>
771 Run extended checks on the usage of templates. This can be used to
772 discover HTML templates that are never used as well as the usage of
773 non-existing HTML templates.
775 =item B<-v>, B<--verbose>
783 This script collects strings from Perl files, the menu.ini file and
784 HTML templates and puts them into the file "all" for translation.