Import data for arabic joining computation

diff --git a/autogen.sh b/autogen.sh
index 7cbee5cca6d966d93a53d76e143580a26a7c8ec9..f9d55318f637e44b3694882c3a039d9e3433f9ca 100755 (executable)
--- a/autogen.sh
+++ b/autogen.sh
@@ -13,7 +13,7 @@ echo timestamp > stamp-h.in
 
 python util/import_gcry.py lib/libgcrypt/ .
 
-python util/import_unicode.py unicode/UnicodeData.txt unicode/BidiMirroring.txt unidata.c
+python util/import_unicode.py unicode/UnicodeData.txt unicode/BidiMirroring.txt unicode/ArabicShaping.txt unidata.c
 
 for rmk in conf/*.rmk ${GRUB_CONTRIB}/*/conf/*.rmk; do
   if test -e $rmk ; then

diff --git a/include/grub/unicode.h b/include/grub/unicode.h
index 457bcd93b1bf8b73cefbee8e053565059bc03b0d..28c752b9e087b29b0a2b8a11bdbed375f3fd2255 100644 (file)
--- a/include/grub/unicode.h
+++ b/include/grub/unicode.h
@@ -36,6 +36,7 @@ struct grub_unicode_compact_range
   grub_uint8_t bidi_type:5;
   grub_uint8_t comb_type;
   grub_uint8_t bidi_mirror:1;
+  grub_uint8_t join_type:3;
 } __attribute__ ((packed));
 
 enum grub_bidi_type
@@ -61,6 +62,16 @@ enum grub_bidi_type
     GRUB_BIDI_TYPE_ON
   };
 
+enum grub_join_type
+  {
+    GRUB_JOIN_TYPE_NONJOINING = 0,
+    GRUB_JOIN_TYPE_LEFT = 1,
+    GRUB_JOIN_TYPE_RIGHT = 2,
+    GRUB_JOIN_TYPE_DUAL = 3,
+    GRUB_JOIN_TYPE_CAUSING = 4,
+    GRUB_JOIN_TYPE_TRANSPARENT = 5
+  };
+
 enum grub_comb_type
   {
     GRUB_UNICODE_COMB_NONE = 0,
@@ -103,7 +114,7 @@ struct grub_unicode_glyph
 {
   grub_uint32_t base;
   grub_uint16_t variant:9;
-  grub_uint8_t attributes:1;
+  grub_uint8_t attributes:5;
   grub_size_t ncomb;
   struct grub_unicode_combining {
     grub_uint32_t code;
@@ -115,12 +126,30 @@ struct grub_unicode_glyph
 };
 
 #define GRUB_UNICODE_GLYPH_ATTRIBUTE_MIRROR 0x1
+#define GRUB_UNICODE_GLYPH_ATTRIBUTES_JOIN_LEFT_TO_RIGHT_SHIFT 1
+#define GRUB_UNICODE_GLYPH_ATTRIBUTE_LEFT_JOINED 0x2
+#define GRUB_UNICODE_GLYPH_ATTRIBUTE_RIGHT_JOINED \
+  (GRUB_UNICODE_GLYPH_ATTRIBUTE_LEFT_JOINED \
+   << GRUB_UNICODE_GLYPH_ATTRIBUTES_JOIN_LEFT_TO_RIGHT_SHIFT)
+/* Set iff the corresponding joining flags come from ZWJ or ZWNJ.  */
+#define GRUB_UNICODE_GLYPH_ATTRIBUTE_LEFT_JOINED_EXPLICIT 0x8
+#define GRUB_UNICODE_GLYPH_ATTRIBUTE_RIGHT_JOINED_EXPLICIT \
+  (GRUB_UNICODE_GLYPH_ATTRIBUTE_LEFT_JOINED_EXPLICIT \
+   << GRUB_UNICODE_GLYPH_ATTRIBUTES_JOIN_LEFT_TO_RIGHT_SHIFT)
+#define GRUB_UNICODE_GLYPH_ATTRIBUTES_JOIN \
+  (GRUB_UNICODE_GLYPH_ATTRIBUTE_LEFT_JOINED \
+   | GRUB_UNICODE_GLYPH_ATTRIBUTE_RIGHT_JOINED \
+   | GRUB_UNICODE_GLYPH_ATTRIBUTE_LEFT_JOINED_EXPLICIT \
+   | GRUB_UNICODE_GLYPH_ATTRIBUTE_RIGHT_JOINED_EXPLICIT)
+
 #define GRUB_UNICODE_COMBINING_GRAPHEME_JOINER 0x34f
 #define GRUB_UNICODE_VARIATION_SELECTOR_1 0xfe00
 #define GRUB_UNICODE_VARIATION_SELECTOR_16 0xfe0f
 #define GRUB_UNICODE_VARIATION_SELECTOR_17 0xe0100
 #define GRUB_UNICODE_VARIATION_SELECTOR_256 0xe01ef
 #define GRUB_UNICODE_HEBREW_WAW  0x05d5
+#define GRUB_UNICODE_ZWNJ        0x200c
+#define GRUB_UNICODE_ZWJ         0x200d
 
 extern struct grub_unicode_compact_range grub_unicode_compact[];
 extern struct grub_unicode_bidi_pair grub_unicode_bidi_pairs[];

diff --git a/unicode/ArabicShaping.txt b/unicode/ArabicShaping.txt
new file mode 100644 (file)
index 0000000..b851d38
--- /dev/null
+++ b/unicode/ArabicShaping.txt
@@ -0,0 +1,373 @@
+# ArabicShaping-5.2.0.txt
+# Date: 2009-08-17, 11:11:00 PDT [KW]
+#
+# This file is a normative contributory data file in the
+# Unicode Character Database.
+#
+# Copyright (c) 1991-2009 Unicode, Inc.
+# For terms of use, see http://www.unicode.org/terms_of_use.html
+#
+# This file defines the shaping classes for Arabic, Syriac, and N'Ko
+# positional shaping, repeating in machine readable form the
+# information exemplified in Tables 8-3, 8-7, 8-8, 8-11, 8-12,
+# 8-13, and 13-5 of The Unicode Standard, Version 5.2.
+#
+# See sections 8.2, 8.3, and 13.5 of The Unicode Standard, Version 5.2
+# for more information.
+#
+# Each line contains four fields, separated by a semicolon.
+#
+# Field 0: the code point, in 4-digit hexadecimal
+#   form, of an Arabic, Syriac, or N'Ko character.
+#
+# Field 1: gives a short schematic name for that character,
+#   abbreviated from the normative Unicode character name.
+#
+# Field 2: defines the joining type (property name: Joining_Type)
+#   R Right_Joining
+#   L Left_Joining
+#   D Dual_Joining
+#   C Join_Causing
+#   U Non_Joining
+#   T Transparent
+#     See Section 8.2, Arabic for more information on these types.
+#
+# Field 3: defines the joining group (property name: Joining_Group)
+#
+# The values of the joining group are based schematically on character
+# names. Where a schematic character name consists of two or more parts separated
+# by spaces, the formal Joining_Group property value, as specified in
+# PropertyValueAliases.txt, consists of the same name parts joined by
+# underscores. Hence, the entry:
+#
+#   0629; TEH MARBUTA; R; TEH MARBUTA
+#
+# corresponds to [Joining_Group = Teh_Marbuta].
+#
+# Note: For historical reasons, the property value [Joining_Group = Hamza_On_Heh_Goal]
+#   is anachronistically named. It used to apply to both of the following characters
+#   in earlier versions of the standard:
+#
+#   U+06C2 ARABIC LETTER HEH GOAL WITH HAMZA ABOVE
+#   U+06C3 ARABIC LETTER TEH MARBUTA GOAL
+#
+#   However, it currently applies only to U+06C3, and *not* to U+06C2.
+#   To avoid destabilizing existing Joining_Group property aliases, the
+#   value Hamza_On_Heh_Goal has not been changed, despite the fact that it
+#   no longer applies to Hamza On Heh Goal, but only to Teh Marbuta Goal.
+#
+# When other cursive scripts are added to the Unicode Standard in
+# the future, the joining group value of all its letters will default
+# to jg=No_Joining_Group in this data file. Other, more specific
+# joining group values will be defined only if an explicit proposal
+# to define those values exactly has been approved by the UTC. This
+# is the convention exemplified by the N'Ko script. Only the Arabic
+# and Syriac scripts currently have explicit joining group values defined.
+#
+# Note: Code points that are not explicitly listed in this file are
+# either of joining type T or U:
+#
+# - Those that not explicitly listed that are of General Category Mn, Me, or Cf
+#   have joining type T.
+# - All others not explicitly listed have joining type U.
+#
+# For an explicit listing of characters of joining type T, see
+# the derived property file DerivedJoiningType.txt.
+#
+# There are currently no characters of joining type L defined in Unicode.
+#
+# #############################################################
+ 
+# Unicode; Schematic Name; Joining Type; Joining Group
+
+# Arabic characters
+
+0600; ARABIC NUMBER SIGN; U; No_Joining_Group
+0601; ARABIC SIGN SANAH; U; No_Joining_Group
+0602; ARABIC FOOTNOTE MARKER; U; No_Joining_Group
+0603; ARABIC SIGN SAFHA; U; No_Joining_Group
+0608; ARABIC RAY; U; No_Joining_Group
+060B; AFGHANI SIGN; U; No_Joining_Group
+0621; HAMZA; U; No_Joining_Group
+0622; MADDA ON ALEF; R; ALEF
+0623; HAMZA ON ALEF; R; ALEF
+0624; HAMZA ON WAW; R; WAW
+0625; HAMZA UNDER ALEF; R; ALEF
+0626; HAMZA ON YEH; D; YEH
+0627; ALEF; R; ALEF
+0628; BEH; D; BEH
+0629; TEH MARBUTA; R; TEH MARBUTA
+062A; TEH; D; BEH
+062B; THEH; D; BEH
+062C; JEEM; D; HAH
+062D; HAH; D; HAH
+062E; KHAH; D; HAH
+062F; DAL; R; DAL
+0630; THAL; R; DAL
+0631; REH; R; REH
+0632; ZAIN; R; REH
+0633; SEEN; D; SEEN
+0634; SHEEN; D; SEEN
+0635; SAD; D; SAD
+0636; DAD; D; SAD
+0637; TAH; D; TAH
+0638; ZAH; D; TAH
+0639; AIN; D; AIN
+063A; GHAIN; D; AIN
+063B; KEHEH WITH 2 DOTS ABOVE; D; GAF
+063C; KEHEH WITH 3 DOTS BELOW; D; GAF 
+063D; FARSI YEH WITH INVERTED V; D; FARSI YEH
+063E; FARSI YEH WITH 2 DOTS ABOVE; D; FARSI YEH
+063F; FARSI YEH WITH 3 DOTS ABOVE; D; FARSI YEH
+0640; TATWEEL; C; No_Joining_Group
+0641; FEH; D; FEH
+0642; QAF; D; QAF
+0643; KAF; D; KAF
+0644; LAM; D; LAM
+0645; MEEM; D; MEEM
+0646; NOON; D; NOON
+0647; HEH; D; HEH
+0648; WAW; R; WAW
+0649; ALEF MAKSURA; D; YEH
+064A; YEH; D; YEH
+066E; DOTLESS BEH; D; BEH
+066F; DOTLESS QAF; D; QAF
+0671; HAMZAT WASL ON ALEF; R; ALEF
+0672; WAVY HAMZA ON ALEF; R; ALEF
+0673; WAVY HAMZA UNDER ALEF; R; ALEF
+0674; HIGH HAMZA; U; No_Joining_Group
+0675; HIGH HAMZA ALEF; R; ALEF
+0676; HIGH HAMZA WAW; R; WAW
+0677; HIGH HAMZA WAW WITH DAMMA; R; WAW
+0678; HIGH HAMZA YEH; D; YEH
+0679; TEH WITH SMALL TAH; D; BEH
+067A; TEH WITH 2 DOTS VERTICAL ABOVE; D; BEH
+067B; BEH WITH 2 DOTS VERTICAL BELOW; D; BEH
+067C; TEH WITH RING; D; BEH
+067D; TEH WITH 3 DOTS ABOVE DOWNWARD; D; BEH
+067E; TEH WITH 3 DOTS BELOW; D; BEH
+067F; TEH WITH 4 DOTS ABOVE; D; BEH
+0680; BEH WITH 4 DOTS BELOW; D; BEH
+0681; HAMZA ON HAH; D; HAH
+0682; HAH WITH 2 DOTS VERTICAL ABOVE; D; HAH
+0683; HAH WITH MIDDLE 2 DOTS; D; HAH
+0684; HAH WITH MIDDLE 2 DOTS VERTICAL; D; HAH
+0685; HAH WITH 3 DOTS ABOVE; D; HAH
+0686; HAH WITH MIDDLE 3 DOTS DOWNWARD; D; HAH
+0687; HAH WITH MIDDLE 4 DOTS; D; HAH
+0688; DAL WITH SMALL TAH; R; DAL
+0689; DAL WITH RING; R; DAL
+068A; DAL WITH DOT BELOW; R; DAL
+068B; DAL WITH DOT BELOW AND SMALL TAH; R; DAL
+068C; DAL WITH 2 DOTS ABOVE; R; DAL
+068D; DAL WITH 2 DOTS BELOW; R; DAL
+068E; DAL WITH 3 DOTS ABOVE; R; DAL
+068F; DAL WITH 3 DOTS ABOVE DOWNWARD; R; DAL
+0690; DAL WITH 4 DOTS ABOVE; R; DAL
+0691; REH WITH SMALL TAH; R; REH
+0692; REH WITH SMALL V; R; REH
+0693; REH WITH RING; R; REH
+0694; REH WITH DOT BELOW; R; REH
+0695; REH WITH SMALL V BELOW; R; REH
+0696; REH WITH DOT BELOW AND DOT ABOVE; R; REH
+0697; REH WITH 2 DOTS ABOVE; R; REH
+0698; REH WITH 3 DOTS ABOVE; R; REH
+0699; REH WITH 4 DOTS ABOVE; R; REH
+069A; SEEN WITH DOT BELOW AND DOT ABOVE; D; SEEN
+069B; SEEN WITH 3 DOTS BELOW; D; SEEN
+069C; SEEN WITH 3 DOTS BELOW AND 3 DOTS ABOVE; D; SEEN
+069D; SAD WITH 2 DOTS BELOW; D; SAD
+069E; SAD WITH 3 DOTS ABOVE; D; SAD
+069F; TAH WITH 3 DOTS ABOVE; D; TAH
+06A0; AIN WITH 3 DOTS ABOVE; D; AIN
+06A1; DOTLESS FEH; D; FEH
+06A2; FEH WITH DOT MOVED BELOW; D; FEH
+06A3; FEH WITH DOT BELOW; D; FEH
+06A4; FEH WITH 3 DOTS ABOVE; D; FEH
+06A5; FEH WITH 3 DOTS BELOW; D; FEH
+06A6; FEH WITH 4 DOTS ABOVE; D; FEH
+06A7; QAF WITH DOT ABOVE; D; QAF
+06A8; QAF WITH 3 DOTS ABOVE; D; QAF
+06A9; KEHEH; D; GAF
+06AA; SWASH KAF; D; SWASH KAF
+06AB; KAF WITH RING; D; GAF
+06AC; KAF WITH DOT ABOVE; D; KAF
+06AD; KAF WITH 3 DOTS ABOVE; D; KAF
+06AE; KAF WITH 3 DOTS BELOW; D; KAF
+06AF; GAF; D; GAF
+06B0; GAF WITH RING; D; GAF
+06B1; GAF WITH 2 DOTS ABOVE; D; GAF
+06B2; GAF WITH 2 DOTS BELOW; D; GAF
+06B3; GAF WITH 2 DOTS VERTICAL BELOW; D; GAF
+06B4; GAF WITH 3 DOTS ABOVE; D; GAF
+06B5; LAM WITH SMALL V; D; LAM
+06B6; LAM WITH DOT ABOVE; D; LAM
+06B7; LAM WITH 3 DOTS ABOVE; D; LAM
+06B8; LAM WITH 3 DOTS BELOW; D; LAM
+06B9; NOON WITH DOT BELOW; D; NOON
+06BA; DOTLESS NOON; D; NOON
+06BB; DOTLESS NOON WITH SMALL TAH; D; NOON
+06BC; NOON WITH RING; D; NOON
+06BD; NYA; D; NYA
+06BE; KNOTTED HEH; D; KNOTTED HEH
+06BF; HAH WITH MIDDLE 3 DOTS DOWNWARD AND DOT ABOVE; D; HAH
+06C0; HAMZA ON HEH; R; TEH MARBUTA
+06C1; HEH GOAL; D; HEH GOAL
+06C2; HAMZA ON HEH GOAL; D; HEH GOAL
+06C3; TEH MARBUTA GOAL; R; HAMZA ON HEH GOAL
+06C4; WAW WITH RING; R; WAW
+06C5; WAW WITH BAR; R; WAW
+06C6; WAW WITH SMALL V; R; WAW
+06C7; WAW WITH DAMMA; R; WAW
+06C8; WAW WITH ALEF ABOVE; R; WAW
+06C9; WAW WITH INVERTED SMALL V; R; WAW
+06CA; WAW WITH 2 DOTS ABOVE; R; WAW
+06CB; WAW WITH 3 DOTS ABOVE; R; WAW
+06CC; FARSI YEH; D; FARSI YEH
+06CD; YEH WITH TAIL; R; YEH WITH TAIL
+06CE; FARSI YEH WITH SMALL V; D; FARSI YEH
+06CF; WAW WITH DOT ABOVE; R; WAW
+06D0; YEH WITH 2 DOTS VERTICAL BELOW; D; YEH
+06D1; YEH WITH 3 DOTS BELOW; D; YEH
+06D2; YEH BARREE; R; YEH BARREE
+06D3; HAMZA ON YEH BARREE; R; YEH BARREE
+06D5; AE; R; TEH MARBUTA
+06DD; ARABIC END OF AYAH; U; No_Joining_Group
+06EE; DAL WITH INVERTED V; R; DAL
+06EF; REH WITH INVERTED V; R; REH
+06FA; SEEN WITH DOT BELOW AND 3 DOTS ABOVE; D; SEEN
+06FB; DAD WITH DOT BELOW; D; SAD
+06FC; GHAIN WITH DOT BELOW; D; AIN
+06FF; HEH WITH INVERTED V; D; KNOTTED HEH
+
+# Syriac characters
+
+0710; ALAPH; R; ALAPH
+0712; BETH; D; BETH
+0713; GAMAL; D; GAMAL
+0714; GAMAL GARSHUNI; D; GAMAL
+0715; DALATH; R; DALATH RISH
+0716; DOTLESS DALATH RISH; R; DALATH RISH
+0717; HE; R; HE
+0718; WAW; R; SYRIAC WAW
+0719; ZAIN; R; ZAIN
+071A; HETH; D; HETH
+071B; TETH; D; TETH
+071C; TETH GARSHUNI; D; TETH
+071D; YUDH; D; YUDH
+071E; YUDH HE; R; YUDH HE
+071F; KAPH; D; KAPH
+0720; LAMADH; D; LAMADH
+0721; MIM; D; MIM
+0722; NUN; D; NUN
+0723; SEMKATH; D; SEMKATH
+0724; FINAL SEMKATH; D; FINAL SEMKATH
+0725; E; D; E
+0726; PE; D; PE
+0727; REVERSED PE; D; REVERSED PE
+0728; SADHE; R; SADHE
+0729; QAPH; D; QAPH
+072A; RISH; R; DALATH RISH
+072B; SHIN; D; SHIN
+072C; TAW; R; TAW
+072D; PERSIAN BHETH; D; BETH
+072E; PERSIAN GHAMAL; D; GAMAL
+072F; PERSIAN DHALATH; R; DALATH RISH
+074D; SOGDIAN ZHAIN; R; ZHAIN
+074E; SOGDIAN KHAPH; D; KHAPH
+074F; SOGDIAN FE; D; FE
+
+# Arabic supplement characters
+
+0750; BEH WITH 3 DOTS HORIZONTALLY BELOW; D; BEH
+0751; BEH WITH DOT BELOW AND 3 DOTS ABOVE; D; BEH
+0752; BEH WITH 3 DOTS POINTING UPWARDS BELOW; D; BEH
+0753; BEH WITH 3 DOTS POINTING UPWARDS BELOW AND 2 DOTS ABOVE; D; BEH
+0754; BEH WITH 2 DOTS BELOW AND DOT ABOVE; D; BEH
+0755; BEH WITH INVERTED SMALL V BELOW; D; BEH
+0756; BEH WITH SMALL V; D; BEH
+0757; HAH WITH 2 DOTS ABOVE; D; HAH
+0758; HAH WITH 3 DOTS POINTING UPWARDS BELOW; D; HAH
+0759; DAL WITH 2 DOTS VERTICALLY BELOW AND SMALL TAH; R; DAL
+075A; DAL WITH INVERTED SMALL V BELOW; R; DAL
+075B; REH WITH STROKE; R; REH
+075C; SEEN WITH 4 DOTS ABOVE; D; SEEN
+075D; AIN WITH 2 DOTS ABOVE; D; AIN
+075E; AIN WITH 3 DOTS POINTING DOWNWARDS ABOVE; D; AIN
+075F; AIN WITH 2 DOTS VERTICALLY ABOVE; D; AIN
+0760; FEH WITH 2 DOTS BELOW; D; FEH
+0761; FEH WITH 3 DOTS POINTING UPWARDS BELOW; D; FEH
+0762; KEHEH WITH DOT ABOVE; D; GAF
+0763; KEHEH WITH 3 DOTS ABOVE; D; GAF
+0764; KEHEH WITH 3 DOTS POINTING UPWARDS BELOW; D; GAF
+0765; MEEM WITH DOT ABOVE; D; MEEM
+0766; MEEM WITH DOT BELOW; D; MEEM
+0767; NOON WITH 2 DOTS BELOW; D; NOON
+0768; NOON WITH SMALL TAH; D; NOON
+0769; NOON WITH SMALL V; D; NOON
+076A; LAM WITH BAR; D; LAM
+076B; REH WITH 2 DOTS VERTICALLY ABOVE; R; REH
+076C; REH WITH HAMZA ABOVE; R; REH
+076D; SEEN WITH 2 DOTS VERTICALLY ABOVE; D; SEEN
+076E; HAH WITH SMALL TAH BELOW; D; HAH
+076F; HAH WITH SMALL TAH AND 2 DOTS; D; HAH
+0770; SEEN WITH SMALL TAH AND 2 DOTS; D; SEEN
+0771; REH WITH SMALL TAH AND 2 DOTS; R; REH
+0772; HAH WITH SMALL TAH ABOVE; D; HAH
+0773; ALEF WITH DIGIT TWO ABOVE; R; ALEF
+0774; ALEF WITH DIGIT THREE ABOVE; R; ALEF
+0775; FARSI YEH WITH DIGIT TWO ABOVE; D; FARSI YEH
+0776; FARSI YEH WITH DIGIT THREE ABOVE; D; FARSI YEH
+0777; YEH WITH DIGIT FOUR BELOW; D; YEH
+0778; WAW WITH DIGIT TWO ABOVE; R; WAW
+0779; WAW WITH DIGIT THREE ABOVE; R; WAW
+077A; YEH BARREE WITH DIGIT TWO ABOVE; D; BURUSHASKI YEH BARREE
+077B; YEH BARREE WITH DIGIT THREE ABOVE; D; BURUSHASKI YEH BARREE
+077C; HAH WITH DIGIT FOUR BELOW; D; HAH
+077D; SEEN WITH DIGIT FOUR ABOVE; D; SEEN
+077E; SEEN WITH INVERTED V; D; SEEN
+077F; KAF WITH 2 DOTS ABOVE; D; KAF
+
+# N'Ko Characters
+
+07CA; NKO A; D; No_Joining_Group
+07CB; NKO EE; D; No_Joining_Group
+07CC; NKO I; D; No_Joining_Group
+07CD; NKO E; D; No_Joining_Group
+07CE; NKO U; D; No_Joining_Group
+07CF; NKO OO; D; No_Joining_Group
+07D0; NKO O; D; No_Joining_Group
+07D1; NKO DAGBASINNA; D; No_Joining_Group
+07D2; NKO N; D; No_Joining_Group
+07D3; NKO BA; D; No_Joining_Group
+07D4; NKO PA; D; No_Joining_Group
+07D5; NKO TA; D; No_Joining_Group
+07D6; NKO JA; D; No_Joining_Group
+07D7; NKO CHA; D; No_Joining_Group
+07D8; NKO DA; D; No_Joining_Group
+07D9; NKO RA; D; No_Joining_Group
+07DA; NKO RRA; D; No_Joining_Group
+07DB; NKO SA; D; No_Joining_Group
+07DC; NKO GBA; D; No_Joining_Group
+07DD; NKO FA; D; No_Joining_Group
+07DE; NKO KA; D; No_Joining_Group
+07DF; NKO LA; D; No_Joining_Group
+07E0; NKO NA WOLOSO; D; No_Joining_Group
+07E1; NKO MA; D; No_Joining_Group
+07E2; NKO NYA; D; No_Joining_Group
+07E3; NKO NA; D; No_Joining_Group
+07E4; NKO HA; D; No_Joining_Group
+07E5; NKO WA; D; No_Joining_Group
+07E6; NKO YA; D; No_Joining_Group
+07E7; NKO NYA WOLOSO; D; No_Joining_Group
+07E8; NKO JONA JA; D; No_Joining_Group
+07E9; NKO JONA CHA; D; No_Joining_Group
+07EA; NKO JONA RA; D; No_Joining_Group
+07FA; NKO LAJANYALAN; C; No_Joining_Group
+
+# Other
+
+200C; ZERO WIDTH NON-JOINER; U; No_Joining_Group
+200D; ZERO WIDTH JOINER; C; No_Joining_Group
+
+# EOF

diff --git a/util/import_unicode.py b/util/import_unicode.py
index e0e53b65531a61edf898371d5d1b99b6b4c223e0..653d7bae23d53c8bdedd476a00dbeadbbb1093a6 100644 (file)
--- a/util/import_unicode.py
+++ b/util/import_unicode.py
@@ -24,8 +24,33 @@ import datetime
 if len (sys.argv) < 3:
     print ("Usage: %s SOURCE DESTINATION" % sys.argv[0])
     exit (0)
+infile = open (sys.argv[3], "r")
+joining = {}
+for line in infile:
+    line = re.sub ("#.*$", "", line)
+    line = line.replace ("\n", "")
+    line = line.replace (" ", "")
+    if len (line) == 0 or line[0] == '\n':
+        continue
+    sp = line.split (";")
+    curcode = int (sp[0], 16)
+    if sp[2] == "U":
+        joining[curcode] = "GRUB_JOIN_TYPE_NONJOINING"
+    elif sp[2] == "L":
+        joining[curcode] = "GRUB_JOIN_TYPE_LEFT"
+    elif sp[2] == "R":
+        joining[curcode] = "GRUB_JOIN_TYPE_RIGHT"
+    elif sp[2] == "D":
+        joining[curcode] = "GRUB_JOIN_TYPE_DUAL"
+    elif sp[2] == "C":
+        joining[curcode] = "GRUB_JOIN_TYPE_CAUSING"
+    else:
+        print ("Unknown joining type '%s'" % sp[2])
+        exit (1)
+infile.close ()
+
 infile = open (sys.argv[1], "r")
-outfile = open (sys.argv[3], "w")
+outfile = open (sys.argv[4], "w")
 outfile.write ("#include <grub/unicode.h>\n")
 outfile.write ("\n")
 outfile.write ("struct grub_unicode_compact_range grub_unicode_compact[] = {\n")
@@ -74,23 +99,32 @@ for line in infile:
                 curcombtype != 240 and curcombtype != 253 and \
                 curcombtype != 254 and curcombtype != 255):
         print ("WARNING: Unknown combining type %d" % curcombtype)
+    if curcode in joining:
+        curjoin = joining[curcode]
+    elif sp[2] == "Me" or sp[2] == "Mn" or sp[2] == "Cf":
+        curjoin = "GRUB_JOIN_TYPE_TRANSPARENT"
+    else:
+        curjoin = "GRUB_JOIN_TYPE_NONJOINING"
     if lastcode + 1 != curcode or curbiditype != lastbiditype \
-            or curcombtype != lastcombtype or curmirrortype != lastmirrortype:
+            or curcombtype != lastcombtype or curmirrortype != lastmirrortype \
+            or curjoin != lastjoin:
         if begincode != -2 and (lastbiditype != "L" or lastcombtype != 0 or \
                                     lastmirrortype):
-            outfile.write (("{0x%x, 0x%x, GRUB_BIDI_TYPE_%s, %d, %d},\n" \
+            outfile.write (("{0x%x, 0x%x, GRUB_BIDI_TYPE_%s, %d, %d, %s},\n" \
                                 % (begincode, lastcode, lastbiditype, \
-                                       lastcombtype, lastmirrortype)))
+                                       lastcombtype, lastmirrortype, \
+                                       lastjoin)))
         begincode = curcode
     lastcode = curcode
+    lastjoin = curjoin
     lastbiditype = curbiditype
     lastcombtype = curcombtype
     lastmirrortype = curmirrortype
 if lastbiditype != "L" or lastcombtype != 0 or lastmirrortype:
-    outfile.write (("{0x%x, 0x%x, GRUB_BIDI_TYPE_%s, %d, %d},\n" \
+    outfile.write (("{0x%x, 0x%x, GRUB_BIDI_TYPE_%s, %d, %d, %s},\n" \
                         % (begincode, lastcode, lastbiditype, lastcombtype, \
-                               lastmirrortype)))
-outfile.write ("{0, 0, 0, 0, 0},\n")
+                               lastmirrortype, lastjoin)))
+outfile.write ("{0, 0, 0, 0, 0, 0},\n")
 
 outfile.write ("};\n")