dwrite: Implement itemization by script id using Unicode.org data.

dlls/dwrite/Makefile.in

@@ -9,4 +9,5 @@ C_SRCS = \
 	layout.c \
 	linebreak.c \
 	main.c \
-	opentype.c
+	opentype.c \
+	scripts.c

dlls/dwrite/scripts.h

+/* Unicode Script IDs */
+/* generated from http://www.unicode.org/Public/7.0.0/ucd/Scripts.txt */
+/* DO NOT EDIT!! */
+
+enum unicode_script_id {
+    Script_Unknown = 0,
+    Script_Control = 1,
+    Script_Arabic = 2,
+    Script_Armenian = 3,
+    Script_Avestan = 4,
+    Script_Balinese = 5,
+    Script_Bamum = 6,
+    Script_Bassa = 7,
+    Script_Batak = 8,
+    Script_Bengali = 9,
+    Script_Bopomofo = 10,
+    Script_Brahmi = 11,
+    Script_Braille = 12,
+    Script_Buginese = 13,
+    Script_Buhid = 14,
+    Script_Canadian = 15,
+    Script_Carian = 16,
+    Script_Caucasian = 17,
+    Script_Chakma = 18,
+    Script_Cham = 19,
+    Script_Cherokee = 20,
+    Script_Coptic = 21,
+    Script_Cuneiform = 22,
+    Script_Cypriot = 23,
+    Script_Cyrillic = 24,
+    Script_Deseret = 25,
+    Script_Devanagari = 26,
+    Script_Duployan = 27,
+    Script_Egyptian = 28,
+    Script_Elbasan = 29,
+    Script_Ethiopic = 30,
+    Script_Georgian = 31,
+    Script_Glagolitic = 32,
+    Script_Gothic = 33,
+    Script_Grantha = 34,
+    Script_Greek = 35,
+    Script_Gujarati = 36,
+    Script_Gurmukhi = 37,
+    Script_Han = 38,
+    Script_Hangul = 39,
+    Script_Hanunoo = 40,
+    Script_Hebrew = 41,
+    Script_Hiragana = 42,
+    Script_Imperial = 43,
+    Script_Inscriptional = 44,
+    Script_Javanese = 45,
+    Script_Kaithi = 46,
+    Script_Kannada = 47,
+    Script_Katakana = 48,
+    Script_Kayah = 49,
+    Script_Kharoshthi = 50,
+    Script_Khmer = 51,
+    Script_Khojki = 52,
+    Script_Khudawadi = 53,
+    Script_Lao = 54,
+    Script_Latin = 55,
+    Script_Lepcha = 56,
+    Script_Limbu = 57,
+    Script_Linear = 58,
+    Script_Lisu = 59,
+    Script_Lycian = 60,
+    Script_Lydian = 61,
+    Script_Mahajani = 62,
+    Script_Malayalam = 63,
+    Script_Mandaic = 64,
+    Script_Manichaean = 65,
+    Script_Meetei = 66,
+    Script_Mende = 67,
+    Script_Meroitic = 68,
+    Script_Miao = 69,
+    Script_Modi = 70,
+    Script_Mongolian = 71,
+    Script_Mro = 72,
+    Script_Myanmar = 73,
+    Script_Nabataean = 74,
+    Script_New = 75,
+    Script_Nko = 76,
+    Script_Ogham = 77,
+    Script_Ol = 78,
+    Script_Old = 79,
+    Script_Oriya = 80,
+    Script_Osmanya = 81,
+    Script_Pahawh = 82,
+    Script_Palmyrene = 83,
+    Script_Pau = 84,
+    Script_Phags = 85,
+    Script_Phoenician = 86,
+    Script_Psalter = 87,
+    Script_Rejang = 88,
+    Script_Runic = 89,
+    Script_Samaritan = 90,
+    Script_Saurashtra = 91,
+    Script_Sharada = 92,
+    Script_Shavian = 93,
+    Script_Siddham = 94,
+    Script_Sinhala = 95,
+    Script_Sora = 96,
+    Script_Sundanese = 97,
+    Script_Syloti = 98,
+    Script_Syriac = 99,
+    Script_Tagalog = 100,
+    Script_Tagbanwa = 101,
+    Script_Tai = 102,
+    Script_Takri = 103,
+    Script_Tamil = 104,
+    Script_Telugu = 105,
+    Script_Thaana = 106,
+    Script_Thai = 107,
+    Script_Tibetan = 108,
+    Script_Tifinagh = 109,
+    Script_Tirhuta = 110,
+    Script_Ugaritic = 111,
+    Script_Vai = 112,
+    Script_Warang = 113,
+    Script_Yi = 114,
+};

tools/make_unicode

@@ -1172,7 +1172,7 @@ sub get_lb_ranges()
 sub dump_indic($)
 {
     my $filename = shift;
-    my @indic_table = ($indic_types{'Other'}) x 65536;;
+    my @indic_table = ($indic_types{'Other'}) x 65536;
 
     my $INPUT = open_data_file( $UNIDATA, "IndicSyllabicCategory.txt" );
     while (<$INPUT>)
@@ -1252,7 +1252,7 @@ sub dump_indic($)
 sub dump_linebreak($)
 {
     my $filename = shift;
-    my @break_table = ($break_types{'XX'}) x 65536;;
+    my @break_table = ($break_types{'XX'}) x 65536;
     my $next_group = 0;
 
     my $INPUT = open_data_file( $UNIDATA, "LineBreak.txt" );
@@ -1294,6 +1294,118 @@ sub dump_linebreak($)
     save_file($filename);
 }
 
+################################################################
+# dump Script IDs table
+sub dump_scripts($)
+{
+    my $filename = shift;
+    my $header = $filename;
+    my @scripts_table = (0) x 65536; # 0 means unknown script
+    my $next_group = 0;
+    my %scripts;
+    my $i;
+
+    my $INPUT = open_data_file( $UNIDATA, "Scripts.txt" );
+    while (<$INPUT>)
+    {
+        my $type = "";
+
+        next if /^\#/;  # skip comments
+        next if /^\s*$/;  # skip empty lines
+        next if /\x1a/;  # skip ^Z
+        if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z]+)\s*/)
+        {
+            $type = $2;
+        }
+        elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z]+)\s*/)
+        {
+            $type = $3;
+        }
+        else
+        {
+            die "malformed line $_";
+        }
+
+        # ignore some scripts
+        if ($type eq "" || $type eq "Common" || $type eq "Inherited")
+        {
+            next;
+        }
+
+        $scripts{$type} = -1;
+    }
+
+    $i = 2;
+    foreach my $script (sort keys %scripts) {
+        $scripts{$script} = $i;
+        $i++;
+    }
+
+    # now fill a table
+    seek $INPUT, 0, 0;
+    while (<$INPUT>)
+    {
+        my $type = "";
+
+        next if /^\#/;  # skip comments
+        next if /^\s*$/;  # skip empty lines
+        next if /\x1a/;  # skip ^Z
+        if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z]+)\s*/)
+        {
+            $type = $2;
+            if (defined $scripts{$type})
+            {
+                $scripts_table[hex $1] = $scripts{$type};
+            }
+            next;
+        }
+        elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z]+)\s*/)
+        {
+            $type = $3;
+            if (defined $scripts{$type})
+            {
+                foreach my $i (hex $1 .. hex $2)
+                {
+                    $scripts_table[$i] = $scripts{$type};
+                }
+            }
+            next;
+        }
+    }
+
+    close $INPUT;
+
+    $header = "$filename.h";
+    open OUTPUT,">$header.new" or die "Cannot create $header";
+    print "Building $header\n";
+    print OUTPUT "/* Unicode Script IDs */\n";
+    print OUTPUT "/* generated from $UNIDATA/Scripts.txt */\n";
+    print OUTPUT "/* DO NOT EDIT!! */\n\n";
+
+    # reserve Unknown and Control ids
+    print OUTPUT "enum unicode_script_id {\n";
+    print OUTPUT "    Script_Unknown = 0,\n";
+    print OUTPUT "    Script_Control = 1,\n";
+    foreach my $script (sort keys %scripts)
+    {
+        print OUTPUT "    Script_$script = $scripts{$script},\n";
+    }
+    print OUTPUT "};\n";
+
+    close OUTPUT;
+    save_file($header);
+
+    $filename = "$filename.c";
+    open OUTPUT,">$filename.new" or die "Cannot create $header";
+    print "Building $filename\n";
+    print OUTPUT "/* Unicode Script IDs */\n";
+    print OUTPUT "/* generated from $UNIDATA/Scripts.txt */\n";
+    print OUTPUT "/* DO NOT EDIT!! */\n\n";
+
+    dump_two_level_mapping( "wine_scripts_table", @scripts_table);
+    close OUTPUT;
+    save_file($filename);
+}
 
 ################################################################
 # dump the BiDi mirroring table
@@ -1333,7 +1445,7 @@ sub dump_mirroring($)
 sub dump_bracket($)
 {
     my $filename = shift;
-    my @bracket_table = (0) x 65536;;
+    my @bracket_table = (0) x 65536;
 
     my $INPUT = open_data_file( $UNIDATA, "BidiBrackets.txt" );
     while (<$INPUT>)
@@ -2275,6 +2387,7 @@ dump_bracket( "dlls/usp10/bracket.c" );
 dump_shaping( "dlls/usp10/shaping.c" );
 dump_linebreak( "dlls/usp10/linebreak.c" );
 dump_linebreak( "dlls/dwrite/linebreak.c" );
+dump_scripts( "dlls/dwrite/scripts" );
 dump_indic( "dlls/usp10/indicsyllable.c" );
 dump_intl_nls("loader/l_intl.nls");
 dump_vertical( "dlls/gdi32/vertical.c" );