summaryrefslogtreecommitdiff
path: root/rt/sbin/extract-message-catalog
diff options
context:
space:
mode:
Diffstat (limited to 'rt/sbin/extract-message-catalog')
-rw-r--r--rt/sbin/extract-message-catalog141
1 files changed, 123 insertions, 18 deletions
diff --git a/rt/sbin/extract-message-catalog b/rt/sbin/extract-message-catalog
index 44f8d51..ce151bd 100644
--- a/rt/sbin/extract-message-catalog
+++ b/rt/sbin/extract-message-catalog
@@ -2,8 +2,8 @@
# BEGIN BPS TAGGED BLOCK {{{
#
# COPYRIGHT:
-#
-# This software is Copyright (c) 1996-2009 Best Practical Solutions, LLC
+#
+# This software is Copyright (c) 1996-2009 Best Practical Solutions, LLC
# <jesse@bestpractical.com>
#
# (Except where explicitly superseded by other copyright notices)
@@ -59,12 +59,17 @@ use vars qw($DEBUG $FILECAT);
$DEBUG = 1;
-@ARGV = <lib/RT/I18N/*.po> unless @ARGV;
+# po dir is for extensions
+@ARGV = (<lib/RT/I18N/*.po>, <lib/RT/I18N/*.pot>, <po/*.po>, <po/*.pot>) unless @ARGV;
$FILECAT = {};
# extract all strings and stuff them into $FILECAT
-File::Find::find( { wanted => \&extract_strings_from_code, follow => 1 }, '.' );
+# scan html dir for extensions
+File::Find::find( { wanted => \&extract_strings_from_code, follow => 1 }, qw(bin sbin lib share html etc) );
+
+# remove msgid with $ in it. XXX: perhaps give some warnings here
+$FILECAT = { map { $_ => $FILECAT->{$_} } grep { !m/\$/ } keys %$FILECAT };
# ensure proper escaping and [_1] => %1 transformation
foreach my $str ( sort keys %{$FILECAT} ) {
@@ -83,11 +88,13 @@ foreach my $str ( sort keys %{$FILECAT} ) {
# update all language dictionaries
foreach my $dict (@ARGV) {
+ $dict = "lib/RT/I18N/$dict.pot" if ( $dict eq 'rt' );
$dict = "lib/RT/I18N/$dict.po" unless -f $dict or $dict =~ m!/!;
my $lang = $dict;
$lang =~ s|.*/||;
$lang =~ s|\.po$||;
+ $lang =~ s|\.pot$||;
update($lang, $dict);
}
@@ -101,7 +108,8 @@ sub extract_strings_from_code {
local $/;
return if ( -d $_ );
return if ( $File::Find::dir =~ 'lib/blib|lib/t/autogen|var|m4|local' );
- return if ( /\.po$|\.bak$|~|,D|,B$|extract-message-catalog$/ );
+ return if ( /\.(?:pot|po|bak|gif|png|psd|jpe?g|svg|css|js)$/ );
+ return if ( /~|,D|,B$|extract-message-catalog$|tweak-template-locstring$/ );
return if ( /^[\.#]/ );
return if ( -f "$_.in" );
@@ -115,12 +123,20 @@ sub extract_strings_from_code {
return;
}
+ my $re_space_wo_nl = qr{(?!\n)\s};
+ my $re_loc_suffix = qr{$re_space_wo_nl* \# $re_space_wo_nl* loc $re_space_wo_nl* $}x;
+ my $re_loc_qw_suffix = qr{$re_space_wo_nl* \# $re_space_wo_nl* loc_qw $re_space_wo_nl* $}x;
+ my $re_loc_pair_suffix = qr{$re_space_wo_nl* \# $re_space_wo_nl* loc_pair $re_space_wo_nl* $}x;
+ my $re_loc_left_pair_suffix = qr{$re_space_wo_nl* \# $re_space_wo_nl* loc_left_pair $re_space_wo_nl* $}x;
+ my $re_delim = $RE{delimited}{-delim=>q{'"}}{-keep};
+
$_ = <_>;
# Mason filter: <&|/l>...</&>
my $line = 1;
while (m!\G.*?<&\|/l(.*?)&>(.*?)</&>!sg) {
my ( $vars, $str ) = ( $1, $2 );
+ $vars =~ s/[\n\r]//g;
$line += ( () = ( $& =~ /\n/g ) ); # cryptocontext!
$str =~ s/\\'/\'/g;
#print "STR IS $str\n";
@@ -136,7 +152,7 @@ sub extract_strings_from_code {
my ( $vars, $str );
if ( $match =~
- /\(\s*($RE{delimited}{-delim=>q{'"}}{-keep})(.*?)\s*\)$/s ) {
+ /\(\s*($re_delim)(.*?)\s*\)$/so ) {
$str = substr( $1, 1, -1 ); # $str comes before $vars now
$vars = $9;
@@ -154,20 +170,59 @@ sub extract_strings_from_code {
# Comment-based mark: "..." # loc
$line = 1;
pos($_) = 0;
- while (m/\G.*?($RE{delimited}{-delim=>q{'"}}{-keep})[\}\)\],]*\s*\#\s*loc\s*$/smg) {
- my $str = substr($1, 1, -1);
+ while (m/\G.*?($re_delim)[\}\)\],;]*$re_loc_suffix/smgo) {
+ my $str = $1;
$line += ( () = ( $& =~ /\n/g ) ); # cryptocontext!
- $str =~ s/\\'/\'/g;
+ unless ( defined $str ) {
+ warn "Couldn't process loc at $filename:$line";
+ next;
+ }
+ $str = substr($str, 1, -1);
+ $str =~ s/\\'/\'/g;
push @{ $FILECAT->{$str} }, [ $filename, $line, '' ];
}
+ # Comment-based qw mark: "qw(...)" # loc_qw
+ $line = 1;
+ pos($_) = 0;
+ while (m/\G.*?(?:(qw\([^)]+\))[\}\)\],;]*)?$re_loc_qw_suffix/smgo) {
+ my $str = $1;
+ $line += ( () = ( $& =~ /\n/g ) ); # cryptocontext!
+ unless ( defined $str ) {
+ warn "Couldn't process loc_qw at $filename:$line";
+ next;
+ }
+ foreach my $value (eval($str)) {
+ push @{ $FILECAT->{$value} }, [ $filename, $line, '' ];
+ }
+ }
+
+ # Comment-based left pair mark: "..." => ... # loc_left_pair
+ $line = 1;
+ pos($_) = 0;
+ while (m/\G.*?(?:(\w+)\s*=>[^#\n]+?)?$re_loc_left_pair_suffix/smgo) {
+ my $key = $1;
+ $line += ( () = ( $& =~ /\n/g ) ); # cryptocontext!
+ unless ( defined $key ) {
+ warn "Couldn't process loc_left_pair at $filename:$line";
+ next;
+ }
+ $key =~ s/\\'/\'/g;
+ push @{ $FILECAT->{$key} }, [ $filename, $line, '' ];
+ }
+
# Comment-based pair mark: "..." => "..." # loc_pair
$line = 1;
pos($_) = 0;
- while (m/\G.*?(\w+)\s*=>\s*($RE{delimited}{-delim=>q{'"}}{-keep})[\}\)\],]*\s*\#\s*loc_pair\s*$/smg) {
+ while (m/\G.*?(?:(\w+)\s*=>\s*($re_delim)[\}\)\],;]*)?$re_loc_pair_suffix/smgo) {
my $key = $1;
- my $val = substr($2, 1, -1);
+ my $val = $2;
$line += ( () = ( $& =~ /\n/g ) ); # cryptocontext!
+ unless ( defined $key && defined $val ) {
+ warn "Couldn't process loc_pair at $filename:$line";
+ next;
+ }
+ $val = substr($val, 1, -1);
$key =~ s/\\'/\'/g;
$val =~ s/\\'/\'/g;
push @{ $FILECAT->{$key} }, [ $filename, $line, '' ];
@@ -196,19 +251,52 @@ sub update {
@lines = grep { !/^(#(:|\.)\s*|$)/ } @lines;
while (@lines) {
my $msghdr = "";
- $msghdr .= shift @lines while ( $lines[0] && $lines[0] !~ /^msgid/ );
+ $msghdr .= shift @lines while ( $lines[0] && $lines[0] !~ /^(#~ )?msgid/ );
my $msgid = "";
- $msgid .= shift @lines while ( $lines[0] && $lines[0] =~ /^(msgid|")/ );
+
+# '#~ ' is the prefix of launchpad for msg that's not found the the source
+# we'll remove the prefix later so we can still show them with our own mark
+
+ $msgid .= shift @lines while ( $lines[0] && $lines[0] =~ /^(#~ )?(msgid|")/ );
my $msgstr = "";
- $msgstr .= shift @lines while ( $lines[0] && $lines[0] =~ /^(msgstr|")/ );
+ $msgstr .= shift @lines while ( $lines[0] && $lines[0] =~ /^(#~ )?(msgstr|")/ );
last unless $msgid;
chomp $msgid;
chomp $msgstr;
- $msgid =~ s/^msgid "(.*)"\s*?$/$1/ms or warn "$msgid in $file";
- $msgstr =~ s/^msgstr "(.*)"\s*?$/$1/ms or warn "$msgstr in $file";
+
+ $msgid =~ s/^#~ //mg;
+ $msgstr =~ s/^#~ //mg;
+
+ $msgid =~ s/^msgid "(.*)"\s*?$/$1/m or warn "$msgid in $file";
+
+ if ( $msgid eq '' ) {
+ # null msgid, msgstr will have head info
+ $msgstr =~ s/^msgstr "(.*)"\s*?$/$1/ms or warn "$msgstr in $file";
+ }
+ else {
+ $msgstr =~ s/^msgstr "(.*)"\s*?$/$1/m or warn "$msgstr in $file";
+ }
+
+ if ( $msgid ne '' ) {
+ for my $msg ( \$msgid, \$msgstr ) {
+ if ( $$msg =~ /\n/ ) {
+ my @lines = split /\n/, $$msg;
+ $$msg =
+ shift @lines; # first line don't need to handle any more
+ for (@lines) {
+ if (/^"(.*)"\s*$/) {
+ $$msg .= $1;
+ }
+ }
+ }
+
+ # convert \\n back to \n
+ $$msg =~ s/(?!\\)\\n/\n/g;
+ }
+ }
$Lexicon{$msgid} = $msgstr;
$Header{$msgid} = $msghdr;
@@ -217,7 +305,7 @@ sub update {
my $is_english = ( $lang =~ /^en(?:[^A-Za-z]|$)/ );
foreach my $str ( sort keys %{$FILECAT} ) {
- $Lexicon{$str} ||= '';;
+ $Lexicon{$str} ||= '';
}
foreach ( sort keys %Lexicon ) {
my $f = join ( ' ', sort map $_->[0].":".$_->[1], @{ $FILECAT->{$_} } );
@@ -250,7 +338,7 @@ sub update {
$var =~ s/\s*$//;
$out .= "#. ($var)\n" unless $seen{$var}++;
}
- $out .= "msgid \"$_\"\nmsgstr \"$Lexicon{$_}\"\n\n";
+ $out .= 'msgid ' . fmt($_) . "msgstr \"$Lexicon{$_}\"\n\n";
}
open PO, ">$file" or die $!;
@@ -266,6 +354,23 @@ sub escape {
return $text;
}
+sub fmt {
+ my $str = shift;
+ return "\"$str\"\n" unless $str =~ /\n/;
+
+ my $multi_line = ($str =~ /\n(?!\z)/);
+ $str =~ s/\n/\\n"\n"/g;
+
+ if ($str =~ /\n"$/) {
+ chop $str;
+ }
+ else {
+ $str .= "\"\n";
+ }
+ return $multi_line ? qq(""\n"$str) : qq("$str);
+}
+
+
__END__
# Local variables:
# c-indentation-style: bsd
generated by cgit v1.1 at 2024-06-22 18:05:36 +0000