1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
|
# © 2016 and later: Unicode, Inc. and others.
# License & terms of use: http://www.unicode.org/copyright.html
# Generated using tools/cldr/cldr-to-icu/build-icu-data.xml
#
# File: Zawgyi_my.txt
# Generated from CLDR
#
# This transform converts Zawgyi "encoded" Burmese into proper
# unicode. Zawgyi is a popular encoding scheme in Myanmar. It uses
# the Myanmar unicode range but assigns different characters or
# glyphs to some codepoints. In addition to the character mapping,
# there is reordering of codepoints needed to match the expected
# unicode order. This reordering is context-based.
#
# This transform is done in two main stages:
# (1) Map all Zawgyi codepoints to their Unicode counterpart.
# (2) Perform reordering.
# Modern Burmese digits & Unicode code points.
$nondigits = [^\u1040-\u1049];
$consonant = [\u1000-\u1021];
$vowelsign = [\u102B-\u1030\u1032]; # Unicode vowel signs except E (1031)
$vowelsAndConsonants = [\u1000-\u102a];
$umedial = [\u103B-\u103E]; # Medial codepoints in Unicode
$vowelmedial = [\u102B-\u1030\u1032\1u36\u1037\u103A-\u103F]; # Union of vowel signs and medials
$ukinzi = \u1004\u103A\u1039; # Codepoints representing kinzi in Unicode
# Zawgyi medial ra has multiple representations
$zmedialra = [\u103B\u107E-\u1084];
$wspace = [\u0020\u00a0\u1680\u2000-\u200d\u2060\u202f\u205f\u3000\ufeff];
####
#### STAGE 1: CODEPOINT MAPPING FROM ZAWGYI TO UNICODE
####
# Kinzi (predefined ligatures)
# Move base character to the right
($consonant) \u103A \u1064 → $ukinzi $1 \u103B;
($consonant) \u1064 → $ukinzi $1;
\u1064 → $ukinzi;
# Special cases moving base character to right before vowel signs
($consonant) \u108B → $ukinzi $1 \u102D;
($consonant) \u108C → $ukinzi $1 \u102E;
($consonant) \u108D → $ukinzi $1 \u1036;
# Special cases moving Kinzi block to left
($consonant) \u103A \u1033 \u108B → $ukinzi $1 \u103B \u102D \u102F;
($consonant) \u103A \u108b → $ukinzi $1 \u103B \u102D ;
($consonant) \u103A \u108C → $ukinzi $1 \u103B \u102E ;
($consonant) \u103A \u108D → $ukinzi $1 \u103B \u1036 ;
($consonant) \u103A \u108e → $1 \u103B \u102D \u1036 ;
\u108B → $ukinzi \u102D ;
\u108C → $ukinzi \u102E ;
\u108D → $ukinzi \u1036 ;
# Consonants (only the ones that have to change)
\u106A → \u1009 ; # NYA
\u106B → \u100A ;
\u108F → \u1014 ;
\u1090 → \u101B ;
\u1086 → \u103F ;
# yapin
[\u103A\u107d] → \u103B ;
# yayit
($zmedialra)+ → \u103C ;
# wasway
\u103C* \u108A → \u103D \u103E; # To avoid duplicate medials
\u103C → \u103D ;
# hatoh
[\u103D\u1087] → \u103E ;
\u1088 → \u103E \u102F ;
\u1089 → \u103E \u1030 ;
# Vowels
\u1033 → \u102F ;
\u1034 → \u1030 ;
# asat
\u1039 → \u103A ;
# lower dot
[\u1094\u1095] → \u1037 ;
# Special cases for 1025 vs 1009;
\u1025 \u1039 → \u1009 \u103a;
\u1025 \u1061 → \u1009 \u1039 \u1001;
\u1025 \u1062 → \u1009 \u1039 \u1002;
\u1025 \u1065 → \u1009 \u1039 \u1005;
\u1025 \u1068 → \u1009 \u1039 \u1007;
\u1025 \u1076 → \u1009 \u1039 \u1013;
\u1025 \u1078 → \u1009 \u1039 \u1015;
\u1025 \u107A → \u1009 \u1039 \u1017;
\u1025 \u1079 → \u1009 \u1039 \u1016;
# Stacked Consonants
\u105A → \u102B \u103A ;
\u1060 → \u1039 \u1000 ;
\u1061 → \u1039 \u1001 ;
\u1062 → \u1039 \u1002 ;
\u1063 → \u1039 \u1003 ;
\u1065 → \u1039 \u1005 ;
[\u1066\u1067] → \u1039 \u1006 ;
\u1068 → \u1039 \u1007 ;
\u1069 → \u1039 \u1008 ;
\u106C → \u1039 \u100B ;
\u106D → \u1039 \u100C ;
\u1070 → \u1039 \u100F ;
[\u1071\u1072] → \u1039 \u1010 ;
\u1096 → \u1039 \u1010 \u103D;
[\u1073\u1074] → \u1039 \u1011 ;
\u1075 → \u1039 \u1012 ;
\u1076 → \u1039 \u1013 ;
\u1077 → \u1039 \u1014 ;
\u1078 → \u1039 \u1015 ;
\u1079 → \u1039 \u1016 ;
\u107A → \u1039 \u1017 ;
[\u107B\u1093] → \u1039 \u1018 ;
\u107C → \u1039 \u1019 ;
\u1085 → \u1039 \u101C ;
\u108E → \u102D \u1036 ;
# Pre-defined ligatures
\u106E → \u100D\u1039\u100D ;
\u106F → \u100D\u1039\u100E ;
\u1091 → \u100F\u1039\u100D ;
\u1092 → \u100B\u1039\u100C ;
\u1097 → \u100B\u1039\u100B ;
\u104E → \u104E\u1004\u103A\u1038 ;
####
#### STAGE 1.01: Digits 0 and 4 used instead of letters
# Case of MYANMAR digit being used instead of a letter
# Lone digit zero and four at start
::Null;
^ \u1040 ($nondigits) → \u101D $1;
^ \u1044 ($nondigits) → | \u104E $1 ;
# Lone digit zero or four at end
($nondigits) \u1040 $ → $1 \u101D;
($nondigits) \u1044 $ → $1 \u104e;
# Evowel and dependent vowel signs before 0 or 4 only
# -> convert to the consonant.
([\u102b-\u103f]) \u1040 ($nondigits) → $1 \u101d $2;
([\u102b-\u103f]) \u1044 ($nondigits) → $1 \u104E $2;
####
#### STAGE 1.1: Strip spaces immediately before combining characters.
#### Move e-vowel after consonants and medials
#### Now every codepoint is Unicode. This starts conversion
#### from semi-visual order to logical order.
####
::Null;
# Don't remove spaces before E vowel or medial Ra at this stage
($wspace) \u1037 > \u1037 $1;
($wspace+) ([\u102b-\u1030\u1032-\u103b\u103d\u103e]) → $2;
# Remove a duplicate early
\u1037+ → \u1037;
# Move e-vowel after medials and consonants.
\u1031+ $ukinzi ($consonant) > $ukinzi $1 \u1031;
\u1031+ \u1037+ ($consonant) > $1 \u1031 \u1037 ;
\u1031+ \u103c ($consonant) > $1 \u103c \u1031;
# Move medials other than 103c before the 1031. Leave 103c for
# the next consonant.
\u1031+ ($consonant) ([\u103b\u103d\u103e]+) > $1 $2 \u1031;
\u1031+ ($vowelsAndConsonants) > $1 \u1031;
####
#### STAGE 2: POST REORDERING RULES FOR UNICODE RENDERING
####
::Null;
\u103b \u103a > \u103a \u103b;
# Simpler replacements for Zawgyi 1025
\u1025 \u102E → \u1026;
# Asat and dot below reordering, to Unicode NFC.
\u103A\u1037 → \u1037\u103A;
# Reorder some vowel signs
\u1036 ($umedial*) ($vowelsign+) → $1 $2 \u1036 ;
([\u102B\u102C\u102F\u1030]) ([\u102D\u102E\u1032]) → $2 $1;
# Move ra medial which precedes consonant, but not other medials.
\u103C ($consonant) → $1 \u103C;
####
#### Stage 3
#### Move \u1036, and \u103C after consonants.
::Null;
($umedial) \u1039 ($consonant) > \u1039 $2 $1;
\u103C \u103A \u1039 ($consonant) → \u103A \u1039 $1 \u103C;
\u1036 ($umedial+) → $1 \u1036;
####
#### Stage 4
#### Reordering medials, dot below, contractions, E sign, and asat.
::Null;
# Reorder the medials
([\u103C\u103D\u103E]+) \u103B → \u103B $1;
([\u103D\u103E]+) \u103C → \u103C $1;
\u103E\u103D → \u103D\u103E ;
# Contractions with vowel signs
([\u1031]+) ($vowelsign*) \u1039 ($consonant) → \u1039 $3 $1 $2;
($vowelsign+) \u1039 ($consonant) → \u1039 $2 $1;
# Move vowel sign E \u1031 after medials, but not across consonants
($umedial*) ([\u1031]+) ($umedial*) → $1 $3 $2;
# Reorder dot below after medials and vowel diacritics
\u1037 ([\u102D-\u1030\u1032\u1036\u103b-\u103e]+) → $1 \u1037;
# Move vowel signs after medials
($vowelsign+) ($umedial+) → $2 $1;
# Reorder modifiers and asat
($consonant) ([\u102B-\u1032\u1036\u103B-\u103E]) \u103A ($consonant) → $1 \u103A $2 $3;
####
#### Stage 5. More reorderings
#### Vowel signs after medials, sort medials,
####
::Null;
# Replace CA + YA with JHA after moving other things beyond the medials.
\u1005 \u103b → \u1008;
# More moving vowel signs after medials
([\u102b-\u1032]) ($umedial) → $2 $1;
# Sort the medials
([\u103C\u103D\u103E]) \u103B → \u103B $1;
([\u103D\u103E]) \u103C → \u103C $1;
\u103E\u103D → \u103D\u103E ;
# Move visarga after other signs
\u1038 ($vowelmedial) → $1 \u1038;
# Reorder
\u1036 \u102f → \u102f \u1036;
###
### Stage 6
### Finish conflicting and extra diacritics. Remove some white space
###
::Null;
# Fix duplicate combiners
\u102D \u102D+ → \u102D;
\u102E \u102E+ → \u102E;
\u102F \u102F+ → \u102F;
\u1030 \u1030+ → \u1030;
\u1032 \u1032+ → \u1032;
\u1036 \u1036+ → \u1036;
\u1037 \u1037+ → \u1037;
\u1039 \u1039+ → \u1039;
\u103a \u103a+ → \u103a;
\u103b \u103b+ → \u103b;
\u103c \u103c+ → \u103c;
\u103d \u103d+ → \u103d;
\u103e \u103e+ → \u103e; # http://unicode.org/cldr/trac/ticket/10386
# Fix overlapping signs
\u102F [\u1030\u103a] → \u102F;
\u102D \u102E → \u102E;
# Remove space directly before diacritics.
($wspace)+ ([\u102b-\u1032\u1036-\u103e]) → $2;
# Remove ZWSP at start and end
^ \u200b+ → ;
\u200b+ $ → ;
# Fix multiple spaces around ZWSP to single ZWSP.
$wspace* \u200b $wspace* → \u200b;
|
