cldr-transforms 46.0.0-BETA2

package/transforms/my-t-my-d0-zawgyi.txt

@@ -0,0 +1,222 @@
+
+# This transform converts Unicode Burmese text into Zawgyi font encoded
+# form. Zawgyi is a popular, non-standard encoding scheme in Myanmar
+# that uses the same code range as Myanmar Unicode but assigns different
+# characters or glyphs to some codepoints. In addition to character remapping,
+# context-based reordering of codepoints is needed to give readable
+# output when the output is displayed with a Zawgyi font such as
+# ZawgyiOne.ttf or ZawgyiOne2008.ttf.
+#
+# The transform is done in two main stages:
+# (1) Map all Unicode codepoints to their Zawgyi counterparts.
+# (2) Perform reordering.
+# Modern Burmese digits and Unicode code points.
+$nondigits = [^\u1040-\u1049];
+$consonant = [\u1000-\u1021];
+$narrowconsonant = [\u1001\u1002\u1004\u1005\u1007\u100b-\u100e\u1012\u1013\u1015-\u1017\u1019\u101d\u1020\u1025\u1026\u108f];
+$wideconsonant = [\u1000\u1003\u1006\u1009\u100a\u100f\u1010\u1011\u1018\u101c\u101e\u101f\u1021];
+$widenya = [\u100a\u106b];
+$othernya = [\u1009\u106a];
+$vowelsign = [\u102B-\u1030\u1032];
+$vowelmedial = [\u102B-\u1030\u1032\u103c-\u103F];
+$ukinzi = [\u1004\u101b\u105a]\u103A\u1039;
+$medialraZ = [\u103b\u107e-\u1084];
+$lowsignZ = [\u102f\u1030\u1037\u103c\u103d\u1087-\u108a];
+$highsignZ = [\u102d\u102e\u1032\u1036\u1039\u1064];
+$subscriptitem = [\u1060-\u1063\u1064-\u1068\u106c\u106d\u1070-\u107c\u1085\u1093\u1096];
+$vowelsAndConsonants = [\u1000-\u102a];
+#### Phase 0: CODEPOINT MAPPING FROM UNICODE TO ZAWGYI
+$ukinzi ($consonant) \u103B > $1 \u103A \u1064 ;
+$ukinzi ($consonant) \u102D \u1036 > $1 \u108e ;
+$ukinzi ($consonant) \u102D > $1 \u108b ;
+$ukinzi ($consonant) \u102E > $1 \u108C ;
+$ukinzi ($consonant) \u1036 > $1 \u108D ;
+$ukinzi ($consonant) \u1031 > $1 \u1031 \u1064 ;
+$ukinzi ($consonant) \u103B \u102D \u102F > $1 \u103A \u1033 \u108B ;
+$ukinzi ($consonant) \u103B \u102D > $1 \u103A \u108b  ;
+$ukinzi ($consonant) \u103B \u102E \u102F > $1 \u103A \u108C \u1033 ;
+$ukinzi ($consonant) \u103B \u102E > $1 \u103A \u108C ;
+$ukinzi ($consonant) \u103B \u1036 > $1 \u103A \u108D ;
+$ukinzi ($consonant) \u103c > $1 \u103b \u1064; # Kinzi + medial ra
+$ukinzi \u102D > \u108B ;
+$ukinzi \u102E  > \u108C  ;
+$ukinzi \u1036 > \u108D  ;
+$ukinzi ($consonant) > $1 \u1064 ;
+\u1025 ($vowelsign) \u1038  > \u106A $1 \u1038 ;
+\u1025 \u102f \u1036  > \u1025 \u1036 \u1033 ;
+\u102D \u1036 > \u108E  ;
+# Some composed lower output
+\u103d \u103e > \u108a ;
+\u103e \u102f > \u1088 ;
+\u103E \u1030 > \u1089 ;
+\u103A > \u1039 ;
+\u103B > \u103A ;
+\u103C > \u103B ;
+\u103D > \u103C ;
+\u103E  > \u103D ;
+\u103F > \u1086 ;
+([\u1019]) \u103e \u1030 > $1 \u103d \u1034;  # A special case with signs.
+\u102B \u103A > \u105A ;
+\u1039 \u1010 \u103d > \u1096 ; # Very special case
+\u1039 \u1000 > \u1060 ;
+\u1039 \u1001 > \u1061 ;
+\u1039 \u1002 > \u1062 ;
+\u1039 \u1003 > \u1063 ;
+\u1039 \u1005 > \u1065 ;
+\u1039 \u1006 > \u1067 ;
+\u1039 \u1007 > \u1068 ;
+\u1039 \u1008 > \u1069 ;
+\u1039 \u100B > \u106C ;
+\u1039 \u100C > \u106D ;
+\u1039 \u100F > \u1070 ;
+\u1039 \u1010 > \u1072 ;
+\u1039 \u1011 > \u1074 ;
+\u1039 \u1012 > \u1075 ;
+\u1039 \u1013 > \u1076 ;
+\u1039 \u1014 > \u1077 ;
+\u1039 \u1015 > \u1078 ;
+\u1039 \u1016 > \u1079 ;
+\u1039 \u1017 > \u107A ;
+\u1039 \u1018 > \u1093 ;
+\u1039 \u1019 > \u107C ;
+\u1039 \u101C > \u1085 ;
+\u100d\u1039\u100D > \u106E ;
+\u100d\u1039\u100E > \u106F ;
+\u100F\u1039\u100D > \u1091 ;
+\u100B\u1039\u100C > \u1092 ;
+\u100B\u1039\u100B > \u1097 ;
+\u104E\u1004\u103A\u1038 > \u104E ;
+#### PHASE 1: Everything is now in Zawgyi code points. REORDERING RULES.
+::Null;
+# Handle Na with lower modifiers, medial ra.
+\u1014 ($subscriptitem) ($highsignZ*) \u103b > \u103b \u108f $1 $2;
+\u1014 \u103b ([\u103c\u103d]*) (\u1031*) > $2 \u103b \u108f $1;
+# E Vowel + medial ra. Move the e vowel
+($consonant) \u103b ([\u103c\u103d]*) \u1031 > \u1031 \u103b $1 $2;
+($consonant) \u103b > \u103b $1 ;
+($consonant) \u103d \u1031 \u1037 > \u1031 $1 \u1094 \u103D ;
+($consonant) (\u108a) \u1031 > \u1031 $1 $2 ;
+# Ra + kinzi
+($consonant) \u1064 \u103b > \u103b $1 \u1064 ;
+# E vowel plus medials
+($consonant) ([\u103a\u103c-\u103d]+) \u1031 > \u1031 $1 $2 ;
+# Handle consonant, subscripted consonant, medial ra
+($consonant) ($subscriptitem) ($highsignZ*) \u103b > \u103b $1 $2 $3 ;
+# No medials intervening.
+($vowelsAndConsonants) \u1031 > \u1031 $1 ;
+# Handle Na with lower modifiers.
+\u1014 ($subscriptitem) > \u108f $1 ;
+\u1014 ($lowsignZ) ($highsignZ) \u1037 > \u108f $1 $2 \u1094;
+\u1014 ($highsignZ) ($lowsignZ) \u1037 > \u108f $1 $2 \u1094;
+\u1014 ($highsignZ) \u1037 > \u1014 $1 \u1094;
+# a special case
+\u1014 \u1032 \u1037 > \u1014 \u1032 \u1094;
+\u1014 \u1037 > \u1014 \u1094;
+\u1014 \u1032 ($lowsignZ) \u1037 > \u108f $1 \u1032 \u1094;
+\u1014 ($highsignZ) ($lowsignZ) > \u108f $1 $2;
+\u1014 ($lowsignZ) ($highsignZ) > \u108f $1 $2;
+\u1014 ($lowsignZ) \u1037 > \u108f $1 \u1094;
+\u1014 ($lowsignZ) > \u108f $1;
+# Move 1037 dot to right with other descenders.
+($lowsignZ) ($highsignZ*) \u1037 > $1 $2 \u1094;
+($nondigits) \u1040 ([\u102B-\u103F]) > $1 \u101D $2;
+# Handle lack of 104E ၎ MYANMAR SYMBOL AFOREMENTIONED
+($nondigits) \u104e > $1 \u1044;
+\u1031 \u1040 ($nondigits) > \u1031 \u101D $1;
+\u1009 \u103A > \u1025 \u103A;
+\u1025 \u102E > \u1026;
+\u1037 \u103A > \u103A \u1037;
+([\u102B\u102C\u102F\u1030]) ([\u102D\u102E\u1032]) > $2 $1;
+## Phase 2: Further adjustments
+::Null;
+# Two medials
+\u103a \u103c > \u103c \u107d;
+\u103c \u1094 > \u103c \u1095 ;
+# Medial ra variations, context dependent
+$medialraZ ($narrowconsonant) ($subscriptitem) ($highsignZ) > \u1083 $1 $2 $3 ;
+$medialraZ ($wideconsonant) ($subscriptitem) ($highsignZ) > \u1084 $1 $2 $3;
+$medialraZ ($narrowconsonant) ($subscriptitem) > \u1081 $1 $2 ;
+$medialraZ ($wideconsonant) ($subscriptitem) > \u1082 $1 $2 ;
+$medialraZ ($narrowconsonant) ([\u103c\u108a]) ($highsignZ) > \u1083 $1 $2 $3 ;
+$medialraZ ($wideconsonant) ([\u103c\u108a]) ($highsignZ) > \u1084 $1 $2 $3 ;
+$medialraZ ($narrowconsonant) \u103d ($highsignZ) > \u107f $1 \u1087 $2;
+$medialraZ ($wideconsonant) \u103d ($highsignZ) > \u1080 $1 \u1087 $2;
+$medialraZ ($narrowconsonant) \u102f ($highsignZ) > \u107f $1 \u1033 $2;
+$medialraZ ($wideconsonant) \u102f ($highsignZ) > \u1080 $1 \u1033 $2;
+$medialraZ ($narrowconsonant) \u1030 ($highsignZ) > \u107f $1 \u1034 $2;
+$medialraZ ($wideconsonant) \u1030 ($highsignZ) > \u1080 $1 \u1034 $2;
+$medialraZ ($narrowconsonant) ($lowsignZ*) ($highsignZ) > \u107f $1 $2 $3;
+$medialraZ ($wideconsonant) ($lowsignZ*) ($highsignZ) > \u1080 $1 $2 $3;
+$medialraZ ($narrowconsonant) ([\u103c\u108a]) > \u1081 $1 $2 ;
+$medialraZ ($wideconsonant) ([\u103c\u108a]) > \u1082 $1 $2 ;
+$medialraZ ($narrowconsonant) \u103d > \u103b $1 \u1087;
+$medialraZ ($wideconsonant) \u103d > \u107e $1 \u1087;
+$medialraZ ($narrowconsonant) \u102f > \u103b $1 \u1033;
+$medialraZ ($wideconsonant) \u102f > \u107e $1 \u1033;
+$medialraZ ($narrowconsonant) \u1030 > \u103b $1 \u1034;
+$medialraZ ($wideconsonant) \u1030 > \u107e $1 \u1034;
+$medialraZ ($widenya) > \u1082 $1 ;
+$medialraZ ($othernya) > \u103b \u106a ;
+$medialraZ ($narrowconsonant) > \u103b $1 ;
+$medialraZ ($wideconsonant) > \u107e $1 ;
+\u1009 ($lowsignZ) > \u106a $1;
+\u100A ($lowsignZ)> \u106B $1  ;  ## NYA and NNYA
+\u103d \u102d > \u102d \u103d;
+\u103a ($highsignZ) \u102f [\u1037\u1094\u1095] > \u103a $1 \u1033 \u1095;
+\u103a \u102f [\u1037\u1094\u1095] > \u103a \u1033 \u1095;
+\u103a \u102f > \u103a \u1033;
+# Kinzi combo
+\u1064 \u102e > \u108c ;
+##### Phase 3
+::Null;
+\u1037 ([\u102D-\u1030\u1032\u1036]) > $1 \u1037;
+($consonant) ([\u102B-\u1032\u1036\u103B-\u103E]) \u103A ($consonant)> $1 \u103A $2 $3;
+# Combine vowel and consonant signs
+\u103d \u102f > \u1088;
+\u1033 \u1094 > \u1033 \u1095; # Wider spacing on lower dot
+($medialraZ) ($narrowconsonant) ($lowsignZ*) ($highsignZ*) \u102f > $1 $2 $3 $4 \u1033;
+($medialraZ) ($wideconsonant) ($lowsignZ*) ($highsignZ*) \u102f > $1 $2 $3 $4 \u1033;
+($medialraZ) ($narrowconsonant) ($lowsignZ*) ($highsignZ*) \u1030 > $1 $2 $3 $4 \u1034;
+($medialraZ) ($wideconsonant) ($lowsignZ*) ($highsignZ*) \u1030 > $1 $2 $3 $4 \u1034;
+##### Phase 4.  More reorderings of medials
+::Null;
+([\u103D\u103E]) \u103C > \u103C $1;
+\u103E\u103D > \u103D\u103E ;
+\u1038 ($vowelmedial) > $1 \u1038;
+\u1038 ([\u1036\u1037\u103A]) > $1 \u1038;
+\u103a ([\u1064\u108b-\u108e]) \u102d \u102f > \u103a $1 \u102d \u1033;
+\u103a \u102d \u102f > \u103a \u102d \u1033;
+#### Phase 5
+::Null;
+($consonant) \u103B \u103A > $1 \u103A \u103B;
+([\u103C\u103D\u103E]) \u103B > \u103B $1;
+([\u103D\u103E]) \u103C > \u103C $1;
+\u103E\u103D > \u103D\u103E ;
+([\u102D-\u1030\u1032]) \u103A ($consonant) \u103A > $1 $2 \u103A;
+\u102D \u103A > \u102D;
+\u102E \u103A > \u102E;
+\u102F \u103A > \u102F;
+\u102D \u102E > \u102E;
+\u102F \u1030 > \u102F;
+\u102B \u102B+ > \u102B;
+\u102C \u102C+ > \u102C;
+\u102D \u102D+ > \u102D;
+\u102E \u102E+ > \u102E;
+\u102F \u102F+ > \u102F;
+\u1030 \u1030+ > \u1030;
+\u1031 \u1031+ > \u1031;
+\u1032 \u1032+ > \u1032;
+\u1036 \u1036+ > \u1036;
+\u103A \u103A+ > \u103A;
+\u103B \u103B+ > \u103B;
+\u103C \u103C+ > \u103C;
+\u103D \u103D+ > \u103D;
+\u103E \u103E+ > \u103E;
+# Visually identical orderings - standardize
+([\u102f\u1033]) \u102D > \u102D $1 ;
+([\u102f\u1033]) \u1036 > \u1036 $1 ;
+\u1037 \u1039 > \u1039 \u1037;
+\u1032 \u103c > \u103c \u1032 ;
+\u102e \u103c > \u103c \u102e ;
+\u103d \u1088 > \u1088 ;
+

package/transforms/my-t-my-s0-zawgyi.json

@@ -0,0 +1,8 @@
+{
+  "_visibility": "external",
+  "_aliasBcp47": "my-t-my-s0-zawgyi",
+  "_source": "Zawgyi",
+  "_direction": "forward",
+  "_target": "my",
+  "_rulesFile": "my-t-my-s0-zawgyi.txt"
+}

package/transforms/my-t-my-s0-zawgyi.txt

@@ -0,0 +1,231 @@
+
+# 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;
+

package/transforms/nl-Title.json

@@ -0,0 +1,8 @@
+{
+  "_visibility": "external",
+  "_aliasBcp47": "nl-t-d0-title",
+  "_source": "nl",
+  "_direction": "forward",
+  "_target": "Title",
+  "_rulesFile": "nl-Title.txt"
+}

package/transforms/nl-Title.txt

@@ -0,0 +1,7 @@
+
+# Copyright (C) 2011-2013, Apple Inc. and others. All Rights Reserved.
+# Special titlecasing for Dutch initial "ij".
+::Any-Title();
+# Fix up Ij at the beginning of a "word" (per Any-Title, notUAX #29)
+[:^WB=ALetter:] [:WB=Extend:]* [[:WB=MidLetter:][:WB=MidNumLet:]]? { Ij } → IJ ;
+

package/transforms/nv-nv_FONIPA.json

@@ -0,0 +1,8 @@
+{
+  "_visibility": "external",
+  "_aliasBcp47": "nv-fonipa-t-nv",
+  "_source": "nv",
+  "_direction": "forward",
+  "_target": "nv_FONIPA",
+  "_rulesFile": "nv-nv_FONIPA.txt"
+}

package/transforms/nv-nv_FONIPA.txt

@@ -0,0 +1,74 @@
+
+::Lower;
+::NFC;
+# References
+# [1] https://en.wikipedia.org/wiki/Navajo_language#Orthography
+# [2] https://en.wikipedia.org/wiki/Navajo_phonology
+$apostrophe = [’ ʼ \'];
+ą́ą́ → ɑ̃́ː;
+áá → ɑ́ː;
+ąą → ɑ̃ː;
+aa → ɑː;
+ą́ → ɑ̃́;
+á → ɑ́;
+ą → ɑ̃;
+a → ɑ;
+ę́ę́ → ẽ́ː;
+éé → éː;
+ęę → ẽː;
+ee → eː;
+ę́ → ẽ́;
+é → é;
+ę → ẽ;
+e → e;
+į́į́ → ɪ̃́ː;
+íí → ɪ́ː;
+įį → ɪ̃ː;
+ii → ɪː;
+į́ → ɪ̃́;
+í → ɪ́;
+į → ɪ̃;
+i → ɪ;
+ǫ́ǫ́ → ṍː;
+óó → óː;
+ǫǫ → õː;
+oo → oː;
+ǫ́ → ṍ;
+ó → ó;
+ǫ → õ;
+o → o;
+$apostrophe → ʔ;
+b → p;
+ch $apostrophe → t͡ʃʼ;
+ch → t͡ʃʰ;
+dl → tˡ;
+dz → t͡s;
+d → t;
+gh → ɣ;
+g → k;
+hw → xʷ;
+h → h;
+j → t͡ʃ;
+k $apostrophe → kʼ;
+kw → k͡xʷ;
+k → k͡x;
+l → l;
+ł → ɬ;
+m → m;
+n → n;
+sh → ʃ;
+s → s;
+tł $apostrophe → t͡ɬʼ;
+tł → t͡ɬʰ;
+ts $apostrophe → t͡sʼ;
+ts → t͡sʰ;
+t $apostrophe → tʼ;
+t → t͡x;
+w → w;
+x → x;
+y → j;
+zh → ʒ;
+z → z;
+::NULL;
+{ɣ} [{ṍ} {ó} {õ} {o}] → ɣʷ;
+

package/transforms/pl-am.json

@@ -0,0 +1,8 @@
+{
+  "_visibility": "external",
+  "_aliasBcp47": "am-t-pl",
+  "_source": "pl",
+  "_direction": "forward",
+  "_target": "am",
+  "_rulesFile": "pl-am.txt"
+}

package/transforms/pl-am.txt

@@ -0,0 +1,4 @@
+
+::pl-pl_FONIPA;
+::am_FONIPA-am;
+

package/transforms/pl-ar.json

@@ -0,0 +1,8 @@
+{
+  "_visibility": "external",
+  "_aliasBcp47": "ar-t-pl",
+  "_source": "pl",
+  "_direction": "forward",
+  "_target": "ar",
+  "_rulesFile": "pl-ar.txt"
+}

package/transforms/pl-ar.txt

@@ -0,0 +1,4 @@
+
+::pl-pl_FONIPA;
+::und_FONIPA-ar;
+

package/transforms/pl-chr.json

@@ -0,0 +1,8 @@
+{
+  "_visibility": "external",
+  "_aliasBcp47": "chr-t-pl",
+  "_source": "pl",
+  "_direction": "forward",
+  "_target": "chr",
+  "_rulesFile": "pl-chr.txt"
+}

package/transforms/pl-chr.txt

@@ -0,0 +1,4 @@
+
+::pl-pl_FONIPA;
+::und_FONIPA-chr;
+

package/transforms/pl-fa.json

@@ -0,0 +1,8 @@
+{
+  "_visibility": "external",
+  "_aliasBcp47": "fa-t-pl",
+  "_source": "pl",
+  "_direction": "forward",
+  "_target": "fa",
+  "_rulesFile": "pl-fa.txt"
+}

package/transforms/pl-fa.txt

@@ -0,0 +1,4 @@
+
+::pl-pl_FONIPA;
+::und_FONIPA-fa;
+

package/transforms/pl-ja.json

@@ -0,0 +1,8 @@
+{
+  "_visibility": "external",
+  "_aliasBcp47": "ja-t-pl",
+  "_source": "pl",
+  "_direction": "forward",
+  "_target": "ja",
+  "_rulesFile": "pl-ja.txt"
+}

package/transforms/pl-ja.txt

@@ -0,0 +1,4 @@
+
+::pl-pl_FONIPA;
+::pl_FONIPA-ja;
+

package/transforms/pl-pl_FONIPA.json

@@ -0,0 +1,8 @@
+{
+  "_visibility": "external",
+  "_aliasBcp47": "pl-fonipa-t-pl",
+  "_source": "pl",
+  "_direction": "forward",
+  "_target": "pl_FONIPA",
+  "_rulesFile": "pl-pl_FONIPA.txt"
+}