icuSources/data/translit/Zawgyi_my.txt

   1 # © 2016 and later: Unicode, Inc. and others.
   2 # License & terms of use: http://www.unicode.org/copyright.html#License
   3 #
   4 # File: Zawgyi_my.txt
   5 # Generated from CLDR
   6 #
   7
   8 # This transform converts Zawgyi "encoded" Burmese into proper
   9 # unicode. Zawgyi is a popular encoding scheme in Myanmar. It uses
  10 # the Myanmar unicode range but assigns different characters or
  11 # glyphs to some codepoints. In addition to the character mapping,
  12 # there is reordering of codepoints needed to match the expected
  13 # unicode order. This reordering is context-based.
  14 #
  15 # This transform is done in two main stages:
  16 # (1) Map all Zawgyi codepoints to their Unicode counterpart.
  17 # (2) Perform reordering.
  18 # Modern Burmese digits & Unicode code points.
  19 $nondigits = [^\u1040-\u1049];
  20 $consonant = [\u1000-\u1021];
  21 $vowelsign = [\u102B-\u1030\u1032];  # Unicode vowel signs except E (1031)
  22 $vowelsAndConsonants = [\u1000-\u102a];
  23 $umedial = [\u103B-\u103E];    # Medial codepoints in Unicode
  24 $vowelmedial = [\u102B-\u1030\u1032\1u36\u1037\u103A-\u103F];  # Union of vowel signs and medials
  25 $ukinzi = \u1004\u103A\u1039;  # Codepoints representing kinzi in Unicode
  26 # Zawgyi medial ra has multiple representations
  27 $zmedialra = [\u103B\u107E-\u1084];
  28 $wspace = [\u0020\u00a0\u1680\u2000-\u200d\u2060\u202f\u205f\u3000\ufeff];
  29 ####
  30 #### STAGE 1: CODEPOINT MAPPING FROM ZAWGYI TO UNICODE
  31 ####
  32 # Kinzi (predefined ligatures)
  33 # Move base character to the right
  34 ($consonant) \u103A \u1064 → $ukinzi $1 \u103B;
  35 ($consonant) \u1064 → $ukinzi $1;
  36 \u1064 → $ukinzi;
  37 # Special cases moving base character to right before vowel signs
  38 ($consonant) \u108B → $ukinzi $1 \u102D;
  39 ($consonant) \u108C → $ukinzi $1 \u102E;
  40 ($consonant) \u108D → $ukinzi $1 \u1036;
  41 # Special cases moving Kinzi block to left
  42 ($consonant) \u103A \u1033 \u108B → $ukinzi $1 \u103B \u102D \u102F;
  43 ($consonant) \u103A \u108b → $ukinzi $1 \u103B \u102D ;
  44 ($consonant) \u103A \u108C → $ukinzi $1 \u103B \u102E ;
  45 ($consonant) \u103A \u108D → $ukinzi $1 \u103B \u1036 ;
  46 ($consonant) \u103A \u108e → $1 \u103B \u102D \u1036 ;
  47 \u108B → $ukinzi \u102D ;
  48 \u108C → $ukinzi \u102E ;
  49 \u108D → $ukinzi \u1036 ;
  50 # Consonants (only the ones that have to change)
  51 \u106A → \u1009 ;  # NYA
  52 \u106B → \u100A ;
  53 \u108F → \u1014 ;
  54 \u1090 → \u101B ;
  55 \u1086 → \u103F ;
  56 # yapin
  57 [\u103A|\u107d] → \u103B ;
  58 # yayit
  59 ($zmedialra)+ → \u103C ;
  60 # wasway
  61 \u103C* \u108A → \u103D \u103E;  # To avoid duplicate medials
  62 \u103C → \u103D ;
  63 # hatoh
  64 [\u103D|\u1087] → \u103E ;
  65 \u1088 → \u103E \u102F ;
  66 \u1089 → \u103E \u1030 ;
  67 # Vowels
  68 \u1033 → \u102F ;
  69 \u1034 → \u1030 ;
  70 # asat
  71 \u1039 → \u103A ;
  72 # lower dot
  73 [\u1094\u1095] → \u1037 ;
  74 # Special cases for 1025 vs 1009;
  75 \u1025 \u1039 → \u1009 \u103a;
  76 \u1025 \u1061 → \u1009 \u1039 \u1001;
  77 \u1025 \u1062 → \u1009 \u1039 \u1002;
  78 \u1025 \u1065 → \u1009 \u1039 \u1005;
  79 \u1025 \u1068 → \u1009 \u1039 \u1007;
  80 \u1025 \u1076 → \u1009 \u1039 \u1013;
  81 \u1025 \u1078 → \u1009 \u1039 \u1015;
  82 \u1025 \u107A → \u1009 \u1039 \u1017;
  83 \u1025 \u1079 → \u1009 \u1039 \u1016;
  84 # Stacked Consonants
  85 \u105A → \u102B \u103A ;
  86 \u1060 → \u1039 \u1000 ;
  87 \u1061 → \u1039 \u1001 ;
  88 \u1062 → \u1039 \u1002 ;
  89 \u1063 → \u1039 \u1003 ;
  90 \u1065 → \u1039 \u1005 ;
  91 [\u1066\u1067] → \u1039 \u1006 ;
  92 \u1068 → \u1039 \u1007 ;
  93 \u1069 → \u1039 \u1008 ;
  94 \u106C → \u1039 \u100B ;
  95 \u106D → \u1039 \u100C ;
  96 \u1070 → \u1039 \u100F ;
  97 [\u1071\u1072] → \u1039 \u1010 ;
  98 \u1096 → \u1039 \u1010 \u103D;
  99 [\u1073\u1074] → \u1039 \u1011 ;
 100 \u1075 → \u1039 \u1012 ;
 101 \u1076 → \u1039 \u1013 ;
 102 \u1077 → \u1039 \u1014 ;
 103 \u1078 → \u1039 \u1015 ;
 104 \u1079 → \u1039 \u1016 ;
 105 \u107A → \u1039 \u1017 ;
 106 [\u107B\u1093] → \u1039 \u1018 ;
 107 \u107C → \u1039 \u1019 ;
 108 \u1085 → \u1039 \u101C ;
 109 \u108E → \u102D \u1036 ;
 110 # Pre-defined ligatures
 111 \u106E → \u100D\u1039\u100D ;
 112 \u106F → \u100D\u1039\u100E ;
 113 \u1091 → \u100F\u1039\u100D ;
 114 \u1092 → \u100B\u1039\u100C ;
 115 \u1097 → \u100B\u1039\u100B ;
 116 \u104E → \u104E\u1004\u103A\u1038 ;
 117 ####
 118 #### STAGE 1.01: Digits 0 and 4 used instead of letters
 119 # Case of MYANMAR digit being used instead of a letter
 120 # Lone digit zero and four at start
 121 ::Null;
 122 ^ \u1040 ($nondigits) → \u101D $1;
 123 ^ \u1044 ($nondigits) → | \u104E $1 ;
 124 # Lone digit zero or four at end
 125 ($nondigits) \u1040 $ → $1 \u101D;
 126 ($nondigits) \u1044 $ → $1 \u104e;
 127 # Evowel and dependent vowel signs before 0 or 4 only
 128 #   -> convert to the consonant.
 129 ([\u102b-\u103f]) \u1040 ($nondigits) → $1 \u101d $2;
 130 ([\u102b-\u103f]) \u1044 ($nondigits) → $1 \u104E $2;
 131 ####
 132 #### STAGE 1.1: Strip spaces immediately before combining characters.
 133 ####   Move e-vowel after consonants and medials
 134 ####   Now every codepoint is Unicode. This starts conversion
 135 ####   from semi-visual order to logical order.
 136 ####
 137 ::Null;
 138 # Don't remove spaces before E vowel or medial Ra at this stage
 139 ($wspace) \u1037 > \u1037 $1;
 140 ($wspace+) ([\u102b-\u1030\u1032-\u103b\u103d\u103e]) → $2;
 141 # Remove a duplicate early
 142 \u1037+ → \u1037;
 143 # Move e-vowel after medials and consonants.
 144 \u1031+ $ukinzi ($consonant) > $ukinzi $1 \u1031;
 145 \u1031+ \u1037+ ($consonant) > $1 \u1031 \u1037 ;
 146 \u1031+ \u103c ($consonant) > $1 \u103c \u1031;
 147 # Move medials other than 103c before the 1031. Leave 103c for
 148 # the next consonant.
 149 \u1031+ ($consonant) ([\u103b\u103d\u103e]+) > $1 $2 \u1031;
 150 \u1031+ ($vowelsAndConsonants) > $1 \u1031;
 151 ####
 152 #### STAGE 2: POST REORDERING RULES FOR UNICODE RENDERING
 153 ####
 154 ::Null;
 155 \u103b \u103a > \u103a \u103b;
 156 # Simpler replacements for Zawgyi 1025
 157 \u1025 \u102E → \u1026;
 158 # Asat and dot below reordering, to Unicode NFC.
 159 \u103A\u1037 → \u1037\u103A;
 160 # Reorder some vowel signs
 161 \u1036 ($umedial*) ($vowelsign+) → $1 $2 \u1036 ;
 162 ([\u102B\u102C\u102F\u1030]) ([\u102D\u102E\u1032]) → $2 $1;
 163 # Move ra medial which precedes consonant, but not other medials.
 164 \u103C ($consonant) → $1 \u103C;
 165 ####
 166 #### Stage 3
 167 #### Move \u1036, and \u103C after consonants.
 168 ::Null;
 169 ($umedial) \u1039 ($consonant) > \u1039 $2 $1;
 170 \u103C \u103A \u1039 ($consonant) → \u103A \u1039 $1 \u103C;
 171 \u1036 ($umedial+) → $1 \u1036;
 172 ####
 173 #### Stage 4
 174 #### Reordering medials, dot below, contractions, E sign, and asat.
 175 ::Null;
 176 # Reorder the medials
 177 ([\u103C\u103D\u103E]+) \u103B → \u103B $1;
 178 ([\u103D\u103E]+) \u103C → \u103C $1;
 179 \u103E\u103D → \u103D\u103E ;
 180 # Contractions with vowel signs
 181 ([\u1031]+) ($vowelsign*) \u1039 ($consonant) → \u1039 $3 $1 $2;
 182 ($vowelsign+) \u1039 ($consonant) → \u1039 $2 $1;
 183 # Move vowel sign E \u1031 after medials, but not across consonants
 184 ($umedial*) ([\u1031]+) ($umedial*) → $1 $3 $2;
 185 # Reorder dot below after medials and vowel diacritics
 186 \u1037 ([\u102D-\u1030\u1032\u1036\u103b-\u103e]+) → $1 \u1037;
 187 # Move vowel signs after medials
 188 ($vowelsign+) ($umedial+) → $2 $1;
 189 # Reorder modifiers and asat
 190 ($consonant) ([\u102B-\u1032\u1036\u103B-\u103E]) \u103A ($consonant) → $1 \u103A $2 $3;
 191 ####
 192 #### Stage 5.  More reorderings
 193 #### Vowel signs after medials, sort medials,
 194 ####
 195 ::Null;
 196 # Replace CA + YA with JHA after moving other things beyond the medials.
 197 \u1005 \u103b → \u1008;
 198 # More moving vowel signs after medials
 199 ([\u102b-\u1032]) ($umedial) → $2 $1;
 200 # Sort the medials
 201 ([\u103C\u103D\u103E]) \u103B → \u103B $1;
 202 ([\u103D\u103E]) \u103C → \u103C $1;
 203 \u103E\u103D → \u103D\u103E ;
 204 # Move visarga after other signs
 205 \u1038 ($vowelmedial) → $1 \u1038;
 206 # Reorder
 207 \u1036 \u102f → \u102f \u1036;
 208 ###
 209 ### Stage 6
 210 ### Finish conflicting and extra diacritics. Remove some white space
 211 ###
 212 ::Null;
 213 # Fix duplicate combiners
 214 \u102D \u102D+ → \u102D;
 215 \u102E \u102E+ → \u102E;
 216 \u102F \u102F+ → \u102F;
 217 \u1030 \u1030+ → \u1030;
 218 \u1032 \u1032+ → \u1032;
 219 \u1036 \u1036+ → \u1036;
 220 \u1037 \u1037+ → \u1037;
 221 \u1039 \u1039+ → \u1039;
 222 \u103a \u103a+ → \u103a;
 223 \u103b \u103b+ → \u103b;
 224 \u103c \u103c+ → \u103c;
 225 \u103d \u103d+ → \u103d;
 226 \u103e \u103e+ → \u103e; # http://unicode.org/cldr/trac/ticket/10386
 227 # Fix overlapping signs
 228 \u102F [\u1030\u103a] → \u102F;
 229 \u102D \u102E → \u102E;
 230 # Remove space directly before diacritics.
 231 ($wspace)+ ([\u102b-\u1032\u1036-\u103e]) → $2;
 232 # Remove ZWSP at start and end
 233 ^ \u200b+ → ;
 234 \u200b+ $ → ;
 235 # Fix multiple spaces around ZWSP to single ZWSP.
 236 $wspace* \u200b $wspace* → \u200b;
 237