luci-base: zh_CN: update Simplified Chinese translation

[project/luci.git] / build / i18n-scan.pl

diff --git a/build/i18n-scan.pl b/build/i18n-scan.pl
index 102ae5f..899d90d 100755 (executable)
--- a/build/i18n-scan.pl
+++ b/build/i18n-scan.pl
@@ -2,54 +2,38 @@
 
 use strict;
 use warnings;
 
 use strict;
 use warnings;

-use Text::Balanced qw(extract_codeblock);
+use Text::Balanced qw(extract_bracketed extract_delimited extract_tagged);

 
 

-@ARGV == 1 || die "Usage: $0 <source direcory>\n";
+@ARGV >= 1 || die "Usage: $0 <source direcory>\n";

 
 
 
 

-sub _parse
-{
-       my ( $code ) = @_;
-       my ( $k, $v );
-       
-       if( $code =~ s/^<%:-?\s*(.+)\s*%>/$1/s )
-       {
-               my ( $key, @text ) = split /[\n\s]+/, $code;
+my %stringtable;

 
 

-               $k = $key;
-               $v = join ' ', @text;
-       }
-       elsif( $code =~ s/^\(\s*(.+)\s*\)/$1/s )
-       {
-               if( $code =~ /^(?:"(\w+)"|'(\w+)')\s*,\s*(?:"(.+?)"|'(.+?)')/s )
-               {
-                       $k = $1 || $2;
-                       $v = $3 || $4 || '';
-                       $v =~ s/\s+/ /sg;
-               }
-               elsif( $code =~ /^(?:"(\w+)"|'(\w+)')/ )
-               {
-                       $k = $1 || $2;
-                       $v = '';
-               }
-               else
-               {
-                       return ();
-               }
-       }
-       else
-       {
-               return ();
-       }
+sub dec_lua_str
+{
+       my $s = shift;
+       $s =~ s/[\s\n]+/ /g;
+       $s =~ s/\\n/\n/g;
+       $s =~ s/\\t/\t/g;
+       $s =~ s/\\(.)/$1/g;
+       $s =~ s/^ //;
+       $s =~ s/ $//;
+       return $s;
+}

 
 

-       $v =~ s/\\"/"/g;
-       $v =~ s/"/\\"/g;
-       
-       return ( $k, $v );
+sub dec_tpl_str
+{
+       my $s = shift;
+       $s =~ s/-$//;
+       $s =~ s/[\s\n]+/ /g;
+       $s =~ s/^ //;
+       $s =~ s/ $//;
+       $s =~ s/\\/\\\\/g;
+       return $s;

 }
 
 
 }
 
 

-if( open F, "find $ARGV[0] -type f -name '*.htm' -or -name '*.lua' |" )
+if( open F, "find @ARGV -type f '(' -name '*.htm' -o -name '*.lua' ')' |" )

 {
        while( defined( my $file = readline F ) )
        {
 {
        while( defined( my $file = readline F ) )
        {


@@ -57,58 +41,88 @@ if( open F, "find $ARGV[0] -type f -name '*.htm' -or -name '*.lua' |" )
 
                if( open S, "< $file" )
                {
 
                if( open S, "< $file" )
                {

-                       my $text = '';
-                       $text .= $_ foreach( readline S );
-
-                       while(
-                               $text =~ s/
-                                       ^ .*?
-                                       (?:
-                                               (?: translate f? | i18n )
-                                               [\s\n]* ( \( )
-                                       |
-                                               ( \<%: -? )
-                                       )
-                               /$1 || $2/segx
-                       ) {
-                               #$text =~ s/^ .*? (?: (?: translate f? | i18n ) [\s\n]* ( \( ) | ( <%: -? ) ) / $1 || $2 /sex;
-                               #warn "T[$text]";
-                               my $code;
-
-                               reparse:
-                               ( $code, $text ) = extract_codeblock( $text, '', '^', '()' );
-                               if( ! $code ) {
-                                       ( $code, $text ) = extract_codeblock( $text, '', '^', '<>' );
-                               }
+                       local $/ = undef;
+                       my $raw = <S>;
+                       close S;

 
 

-                               if( ! $code ) {
-                                       # Corner case:
-                                       $text =~ s/(#[^\n]*)%>/$1\n%>/;
-                                       ( $code, $text ) = extract_codeblock( $text, '<>', '^' );
-                                       if( ! $code ) {
-                                               last;
-                                       }
-                               }

 
 

-                               my ( $k, $v ) = _parse( $code );
-                               #warn "M[$code]";
-                               #last;
-               
-                               if( $k && defined($v) )
+                       my $text = $raw;
+
+                       while( $text =~ s/ ^ .*? (?:translate|translatef|i18n|_) [\n\s]* \( /(/sgx )
+                       {
+                               ( my $code, $text ) = extract_bracketed($text, q{('")});
+
+                               $code =~ s/\\\n/ /g;
+                               $code =~ s/^\([\n\s]*//;
+                               $code =~ s/[\n\s]*\)$//;
+
+                               my $res = "";
+                               my $sub = "";
+
+                               if( $code =~ /^['"]/ )

                                {
                                {

-                                       if( $v )
+                                       while( defined $sub )

                                        {
                                        {

-                                               printf "#. %s\n", $v || $k;
+                                               ( $sub, $code ) = extract_delimited($code, q{'"}, q{\s*(?:\.\.\s*)?});
+
+                                               if( defined $sub && length($sub) > 2 )
+                                               {
+                                                       $res .= substr $sub, 1, length($sub) - 2;
+                                               }
+                                               else
+                                               {
+                                                       undef $sub;
+                                               }

                                        }
                                        }

+                               }
+                               elsif( $code =~ /^(\[=*\[)/ )
+                               {
+                                       my $stag = quotemeta $1;
+                                       my $etag = $stag;
+                                          $etag =~ s/\[/]/g;

 
 

-                                       printf "msgid \"%s\"\nmsgstr \"%s\"\n\n",
-                                               $k, $v;
+                                       ( $res ) = extract_tagged($code, $stag, $etag);
+
+                                       $res =~ s/^$stag//;
+                                       $res =~ s/$etag$//;

                                }
                                }

+
+                               $res = dec_lua_str($res);
+                               $stringtable{$res}++ if $res;

                        }
 
                        }
 

-                       close S;
+
+                       $text = $raw;
+
+                       while( $text =~ s/ ^ .*? <% -? [:_] /<%/sgx )
+                       {
+                               ( my $code, $text ) = extract_tagged($text, '<%', '%>');
+
+                               if( defined $code )
+                               {
+                                       $code = dec_tpl_str(substr $code, 2, length($code) - 4);
+                                       $stringtable{$code}++;
+                               }
+                       }

                }
        }
 
        close F;
 }
                }
        }
 
        close F;
 }

+
+
+if( open C, "| msgcat -" )
+{
+       printf C "msgid \"\"\nmsgstr \"Content-Type: text/plain; charset=UTF-8\"\n\n";
+
+       foreach my $key ( sort keys %stringtable )
+       {
+               if( length $key )
+               {
+                       $key =~ s/"/\\"/g;
+                       printf C "msgid \"%s\"\nmsgstr \"\"\n\n", $key;
+               }
+       }
+
+       close C;
+}