(category
;; V : independent vowels
;; C : consonants
+ ;; R : RA
;; H : AL-LAKUNA (virama)
- ;; m : MATRA (pre)
- ;; b : MATRA (below)
- ;; u : MATRA (above)
- ;; p : MATRA (post)
- ;; t : MATRA (two-part)
+ ;; m : pre-consonant vowel signs
+ ;; b : below-consonant vowel signs
+ ;; u : above-consonant vowel signs
+ ;; p : post-consonant vowel signs
+ ;; q : AELA-PILLA
+ ;; r : GAYANUKITTA
+ ;; t : two-part vowel sign
;; a : AVUSVARAYA & VISARGAYA
- ;; h : Internal use
;; E : else
;; J : ZERO WIDTH JOINER
(0x0D80 0x0DFF ?E)
(0x0D82 0x0D83 ?a)
(0x0D85 0x0D96 ?V)
(0x0D96 0x0DC6 ?C)
+ (0x0DBB ?R)
(0x0DCA ?H)
- (0x0DCF 0x0DD1 ?p)
+ (0x0DCF ?q)
+ (0x0DD0 0x0DD1 ?p)
(0x0DD2 0x0DD3 ?u)
(0x0DD4 0x0DD6 ?b)
(0x0DD8 ?p)
(0x0DD9 ?m)
(0x0DDA ?t)
- (0x0DDB ?m)
+ (0x0DDB ?M)
(0x0DDC 0x0DDE ?t)
- (0x0DDF ?p)
+ (0x0DDF ?r)
(0x0DF2 0x0DF3 ?p)
- (0x0DFF ?h)
(0x200D ?J)
)
;; Step 0 : Syllable identification
(generator
+
(0
(cond
;; A syllable ending with a two-part vowel sign.
- ("((CHJ)*C)(t)(a)?"
+ ("(C(HJ[CR])*)(t)(a)?"
< | (1 otf:sinh=akhn,vatu,pres) (3 two-part) (4 =) | >)
- ;; A syllable ending with a vowel sign and/or a various sign.
- ;; Note that a two-part vowel may be expressed with two vowel marks.
+ ;; Decomposed two-part vowel signs.
+ ("(C(HJ[CR])*)(m(H|qH?|r)a?)"
+ < | (1 otf:sinh=akhn,vatu,pres) (3 = *) | >)
- ("((CHJ)*C)([mbup]+a?|a)"
+ ;; A syllable ending with a vowel sign and/or a various sign.
+ ("(C(HJ[CR])*)([mMbupqr]a?|a)"
< | (1 otf:sinh=akhn,vatu,pres) (3 = *) | >)
;; No signs.
- ("(CHJ)*CH?"
+ ("C(HJ[CR])*H?"
< | otf:sinh=akhn,vatu,pres | >)
+ ;; Suppress repaya explicitly.
+ ("(R)(t)(a)?"
+ < | (1 =) (2 two-part) (3 =) | >)
+ ("Rm(H|qH?|r)a?"
+ < | = * | >)
+ ("R([mMbupqr]a?|a)"
+ < | = * | >)
+ ("RHJ?"
+ < | = = | >)
+
;; Independent vowels.
("Va?"
< | = * | >)
;; Split two-part dependent vowel signs for canonicalisation.
(two-part
(cond
- ((0x0DDA) 0x0DD9 0x0DFF)
+ ((0x0DDA) 0x0DD9 0x0DCA)
((0x0DDC) 0x0DD9 0x0DCF)
- ((0x0DDD) 0x0DD9 0x0DCF 0x0DFF)
+ ((0x0DDD) 0x0DD9 0x0DCF 0x0DCA)
((0x0DDE) 0x0DD9 0x0DDF)))
)
-;; Step 1 : Matra movements.
+;; Step 1 : Move pre-consonant signs.
(generator
(0
(cond
- ;; 1 2 3 4 5 6 7
- (" ([^ mbup]+)(m)?(b)?(u)?(p)?(h)?(a)? "
- | (2 =) (1 = *) (3 =) (4 =) (5 =) (6 0x0DCA) (7 =) |)
+ ;; 1 2 3 4 5 6 7 8 9
+ (" ([^ mMbupqr]+)(m|M)?(b)?(u)?(p)?(q)?(r)?(H)?(a)? "
+ | (2 =) (1 = *) (3 =) (4 =) (5 =) (6 =) (7 =) (8 =) (9 =) |)
("." =))
*))