(defun indian-to-devanagari (char)
"Convert IS 13194 character CHAR to Devanagari basic characters.
If CHAR is not IS 13194, return CHAR as is."
(let ((charcodes (split-char char)))
(if (eq (car charcodes) 'indian-is13194)
(make-char 'indian-2-column ?\x21 (nth 1 charcodes))
char)))
(defun devanagari-to-indian (char)
"Convert Devanagari basic character CHAR to IS 13194 characters.
If CHAR is not Devanagari basic character, return CHAR as is."
(let ((charcodes (split-char char)))
(if (and (eq (car charcodes) 'indian-2-column)
(= (nth 1 charcodes) ?\x21))
(make-char 'indian-is13194 (nth 2 charcodes))
char)))
(defun indian-to-devanagari-region (from to)
"Convert IS 13194 characters in region to Devanagari basic characters.
When called from a program, expects two arguments,
positions (integers or markers) specifying the region."
(interactive "r")
(save-excursion
(goto-char from)
(while (< (point) to)
(let ((char (following-char)))
(if (eq (char-charset char) 'indian-is13194)
(progn
(delete-char 1)
(insert (indian-to-devanagari char)))
(forward-char 1))))))
(defun devanagari-to-indian-region (from to)
"Convert Devanagari basic characters in region to Indian characters.
When called from a program, expects two arguments,
positions (integers or markers) specifying the region."
(interactive "r")
(save-excursion
(goto-char from)
(while (< (point) to)
(let ((char (following-char)))
(if (eq (char-charset char) 'indian-2-column)
(progn
(delete-char 1)
(insert (devanagari-to-indian char)))
(forward-char 1))))))
(defun indian-to-devanagari-string (string)
"Convert Indian characters in STRING to Devanagari Basic characters."
(let* ((len (length string))
(i 0)
(vec (make-vector len 0)))
(while (< i len)
(aset vec i (indian-to-devanagari (aref string i)))
(setq i (1+ i)))
(concat vec)))
(defconst devanagari-full-cons
"\\(\\([$(5!3!4!5!:!?!@!I(B]$(5!i(B\\)\\|[$(5!3(B-$(5!X$.$E"%(B]\\)"
"Devanagari full consonant")
(defconst devanagari-pure-cons
(concat "\\(" devanagari-full-cons "$(5!h(B\\)")
"Devanagari pure consonant")
(defconst devanagari-matra
"\\(\\([$(5!_![!\(B]$(5!i(B\\)\\|[$(5!Z(B-$(5!g#K#L#M(B]\\)"
"Devanagari Matra Signs. '$(5#K#L#M(B' can also be created from the combination
of '$(5!_![!\(B' and nukta sign.")
(defconst devanagari-vowel
"\\(\\([$(5!*!&!'(B]$(5!i(B\\)\\|[$(5!$(B-$(5!2#&#'#*(B]\\)"
"Devanagari Vowels. '$(5#&#'#*(B' can also be created from the combination
of '$(5!*!&!'(B' and nukta sign.")
(defconst devanagari-vowel-syllable
(concat devanagari-vowel "[$(5!!!"(B]?")
"Devanagari vowel syllable.")
(defconst devanagari-cons-syllable
(concat devanagari-pure-cons "?" devanagari-pure-cons "?"
devanagari-pure-cons "?" devanagari-pure-cons "$")
"Devanagari consonant syllable")
(defconst devanagari-cons-vowel-syllable
(concat "\\("
devanagari-pure-cons "?" devanagari-pure-cons "?"
devanagari-pure-cons "?" devanagari-pure-cons "\\)?"
devanagari-full-cons devanagari-matra "?[$(5!!!"(B]?")
"Devanagari consonant vowel syllable.")
;;
;; Also, digits and virams should be processed other than syllables.
;;
;; In IS 13194, Avagrah is obtained by Nukta after Viram, and
;; OM is obtained by Nukta after Chandrabindu
;;
(defconst devanagari-digit-viram-visarga
"[$(5!q(B-$(5!z!j!#(B]")
(defconst devanagari-other-sign
"\\([$(5!!!j(B]$(5!i(B\\)\\|\\([$(5#!#J(B]\\)")
(defconst devanagari-composite-glyph-unit
(concat "\\(" devanagari-cons-syllable
"\\)\\|\\(" devanagari-vowel-syllable
"\\)\\|\\(" devanagari-cons-vowel-syllable
"\\)\\|\\(" devanagari-other-sign
"\\)\\|\\(" devanagari-digit-viram-visarga "\\)")
"Regexp matching to Devanagari string to be composed form one glyph.")
;;(put-charset-property charset-devanagari-1-column
;; 'char-to-glyph 'devanagari-compose-string)
;;(put-charset-property charset-devanagari-2-column
;; 'char-to-glyph 'devanagari-compose-string)
;; Sample
;;
;;(string-match devanagari-cons-vowel-syllable-examine "$(5!X![(B") => 0
;;(string-match devanagari-cons-vowel-syllable-examine "$(5!F!h!D!\(B") => 0
;;(string-match devanagari-cons-vowel-syllable-examine "$(5!X![!F!h!D!\(B") => 0
;;
;; Steps toward the composition
;; Converting Character Codes to Composite Glyph.
;;
;; Example : $(5!X![(B/$(5!F!h!D!\(B
;;
;; First, convert Characters to appropriate glyphs.
;;
;; => $(5!X![(B/$(5"F!D!\(B
(defconst devanagari-char-to-glyph-rules
'(
("^\\($(5!O!h(B\\)[$(5!3(B-$(5!X(B]" "$(5"p(B")
;; Ligature Rules
("\\($(5!3!h!B!h!O!h!M(B\\)" "$(5$!(B" sanskrit)
("\\($(5!3!h!B!h!T(B\\)" "$(5$"(B" sanskrit)
("\\($(5!3!h!B!h!M(B\\)" "$(5$#(B" sanskrit)
("\\($(5!3!h!F!h!M(B\\)" "$(5$$(B")
("\\($(5!3!h!O!h!M(B\\)" "$(5$%(B")
("\\($(5!3!h!O(B\\)" "$(5"#(B") ; Post "r"
("\\($(5!3!h!T!h!M(B\\)" "$(5$&(B" sanskrit)
("\\($(5!3!h(B\\)$(5!3!h(B[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"3(B") ("\\($(5!3!h!3(B\\)" "$(5$'(B")
("\\($(5!3!h(B\\)$(5!B!h!O(B" "$(5"3(B") ; Special Rules for "k-tr"
("\\($(5!3!h!B(B\\)" "$(5$((B")
("\\($(5!3!h!F(B\\)" "$(5$)(B")
("\\($(5!3!h!L(B\\)" "$(5$*(B")
("\\($(5!3!h!M(B\\)" "$(5$+(B")
("\\($(5!3!h!Q(B\\)" "$(5$,(B")
("\\($(5!3!h!T(B\\)" "$(5$-(B")
("\\($(5!3!h!V!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"l(B") ("\\($(5$.!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"l(B") ; Half Form
("\\($(5!3!h!V(B\\)" "$(5$.(B")
("\\($(5!3!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"3(B") ("\\($(5!3!i!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"s(B") ; Nukta Half Form
("\\($(5!3!i(B\\)" "$(5#3(B") ; Nukta
("\\($(5!4!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"4(B") ("\\($(5!4!i!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"t(B") ; Nukta Half Form
("\\($(5!4!i(B\\)" "$(5#4(B") ; Nukta
("\\($(5!5!h!O!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"`(B") ("\\($(5!5!h!O(B\\)" "$(5"$(B") ; Post "r"
("\\($(5!5!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"5(B") ("\\($(5!5!i!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"u(B") ; Nukta Half Form
("\\($(5!5!i(B\\)" "$(5#5(B") ; Nukta
("\\($(5!6!h!F!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"a(B") ("\\($(5!6!h!F(B\\)" "$(5$/(B")
("\\($(5!6!h!O(B\\)" "$(5!6"q(B") ; Post "r"
("\\($(5!6!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"6(B") ("\\($(5!7!h!3!h!B!h!M(B\\)" "$(5$0(B" sanskrit)
("\\($(5!7!h!3!h!V!h!T(B\\)" "$(5$1(B" sanskrit)
("\\($(5!7!h!3!h!B(B\\)" "$(5$2(B" sanskrit)
("\\($(5!7!h!3!h!V(B\\)" "$(5$3(B" sanskrit)
("\\($(5!7!h!3!h!O(B\\)" "$(5$9"q(B") ; Special Rule. May be precomposed font needed.
("\\($(5!7!h!6!h!O(B\\)" "$(5$4(B" sanskrit)
("\\($(5!7!h!3!h!M(B\\)" "$(5$5(B" sanskrit)
("\\($(5!7!h!4!h!M(B\\)" "$(5$6(B" sanskrit)
("\\($(5!7!h!5!h!M(B\\)" "$(5$7(B" sanskrit)
("\\($(5!7!h!6!h!M(B\\)" "$(5$8(B" sanskrit)
("\\($(5!7!h!3(B\\)" "$(5$9(B")
("\\($(5!7!h!4(B\\)" "$(5$:(B")
("\\($(5!7!h!5!h!O(B\\)" "$(5$ ("\\($(5!7!h!5(B\\)" "$(5$;(B")
("\\($(5!7!h!6(B\\)" "$(5$<(B")
("\\($(5!7!h!7(B\\)" "$(5$=(B")
("\\($(5!7!h!F(B\\)" "$(5$>(B")
("\\($(5!7!h!L(B\\)" "$(5$?(B")
("\\($(5!7!h!M(B\\)" "$(5$@(B")
("\\($(5!8!h(B\\)[$(5!8!<(B]$(5!h(B" "$(5"8(B") ; Half Form
("\\($(5!8!h!8(B\\)" "$(5$A(B")
("\\($(5!8!h!<(B\\)" "$(5$B(B")
("\\($(5!8!h!O!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"8"q(B") ; Half Form Post "r"
("\\($(5!8!h!O(B\\)" "$(5!8"q(B") ("\\($(5!8!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"8(B") ; Half Form
("\\($(5!9!h!M(B\\)" "$(5$C(B")
("\\($(5!:!h!O(B\\)" "$(5$D(B")
("\\($(5!:!h!<!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"m(B") ("\\($(5!:!h!<(B\\)" "$(5$E(B")
("\\($(5!:!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5":(B") ; Half Form
("\\($(5!:!i!h!O(B\\)" "$(5"!(B") ("\\($(5!:!i!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"z(B") ; Nukta Half Form
("\\($(5!:!i(B\\)" "$(5#:(B") ; Nukta
("\\($(5! ("\\($(5!<!h(B\\)$(5!8!h(B[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"<(B") ; Special Half Form
("\\($(5!<!h!8(B\\)" "$(5$F(B")
("\\($(5!<!h(B\\)$(5!:!h(B[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"<(B") ("\\($(5!<!h!:(B\\)" "$(5$G(B")
("\\($(5!<!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"<(B") ; Half Form
("\\($(5!=!h!3(B\\)" "$(5$H(B")
("\\($(5!=!h!=(B\\)" "$(5$I(B")
("\\($(5!=!h!>(B\\)" "$(5$J(B")
("\\($(5!=!h!M(B\\)" "$(5$K(B")
("\\($(5!>!h!M(B\\)" "$(5$L(B")
("\\($(5!?!h!5!h!M(B\\)" "$(5$M(B" sanskrit)
("\\($(5!?!h!6!h!O(B\\)" "$(5$N(B" sanskrit)
("\\($(5!?!h!O!h!M(B\\)" "$(5$O(B")
("\\($(5!?!h!5(B\\)" "$(5$P(B")
("\\($(5!?!h!6(B\\)" "$(5$Q(B")
("\\($(5!?!h!?(B\\)" "$(5$R(B")
("\\($(5!?!h!L(B\\)" "$(5$S(B")
("\\($(5!?!h!M(B\\)" "$(5$T(B")
("\\($(5!?!i(B\\)" "$(5#?(B") ; Nukta
("\\($(5!@!h!M(B\\)" "$(5$`(B")
("\\($(5!@!i(B\\)" "$(5#@(B") ; Nukta
("\\($(5!A!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"A(B") ("\\($(5!B!h(B\\)$(5!B!h!O(B" "$(5"B(B") ; Special Rule for "t-tr"
("\\($(5!B!h!B!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"c(B") ("\\($(5!B!h!B(B\\)" "$(5$a(B")
("\\($(5!B!h!F(B\\)" "$(5$b(B")
("\\($(5!B!h!O!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"d(B") ; Half Form Post "r"
("\\($(5!B!h!O(B\\)" "$(5"%(B") ("\\($(5!B!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"B(B") ; Half Form
("\\($(5!C!h!O(B\\)" "$(5!C"q(B") ("\\($(5!C!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"C(B") ; Half Form
("\\($(5!D!h!D!h!M(B\\)" "$(5$c(B")
("\\($(5!D!h!E!h!M(B\\)" "$(5$d(B")
("\\($(5!D!h!K!h!M(B\\)" "$(5$e(B")
("\\($(5!D!h!K!h!O(B\\)" "$(5$r"r(B") ("\\($(5!D!h!O!h!M(B\\)" "$(5$f(B")
("\\($(5!D!h!T!h!M(B\\)" "$(5$g(B")
("\\($(5!D!h!5!h!O(B\\)" "$(5$h(B")
("\\($(5!D!h!6!h!O(B\\)" "$(5$i(B")
("\\($(5!D!h!D!h!T(B\\)" "$(5$j(B")
("\\($(5!D!h!E!h!T(B\\)" "$(5$k(B")
("\\($(5!D!h(B\\)$(5!E!h(B[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5!D!h(B") ("\\($(5!D!h!5(B\\)" "$(5$l(B")
("\\($(5!D!h!6(B\\)" "$(5$m(B")
("\\($(5!D!h!D(B\\)" "$(5$n(B")
("\\($(5!D!h!E(B\\)" "$(5$o(B")
("\\($(5!D!h!F(B\\)" "$(5$p(B")
("\\($(5!D!h(B\\)$(5!J!h(B" "$(5!D!h(B") ("\\($(5!D!h!J(B\\)" "$(5$q(B")
("\\($(5!D!h!K(B\\)" "$(5$r(B")
("\\($(5!D!h!L(B\\)" "$(5$s(B")
("\\($(5!D!h!M(B\\)" "$(5$t(B")
("\\($(5!D!h!T(B\\)" "$(5$u(B")
("\\($(5!E!h!F!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"e(B") ; Half Form
("\\($(5!E!h!F(B\\)" "$(5$v(B")
("\\($(5!E!h!O!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"f(B") ("\\($(5!E!h!O(B\\)" "$(5!E"q(B") ; Post "r"
("\\($(5!E!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"E(B") ("\\($(5!F!h!F!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"k(B") ; Half Form
("\\($(5!F!h!F(B\\)" "$(5$w(B")
("\\($(5!F!h!O(B\\)" "$(5!F"q(B")
("\\($(5!F!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"F(B") ; Half Form
("\\($(5!G!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"G(B") ("\\($(5!H!h(B\\)$(5!B!h!O(B" "$(5"H(B") ; Special Rule for "p-tr"
("\\($(5!H!h!B!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"g(B") ("\\($(5!H!h!B(B\\)" "$(5$x(B")
("\\($(5!H!h!F(B\\)" "$(5$y(B")
("\\($(5!H!h!Q(B\\)" "$(5$z(B")
("\\($(5!H!h!O(B\\)" "$(5"&(B") ; Post "r"
("\\($(5!H!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"H(B") ("\\($(5!I!h!O(B\\)" "$(5"'(B") ; Post "r"
("\\($(5!I!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"I(B") ("\\($(5!I!i!h!O(B\\)" "$(5""(B") ("\\($(5!I!i!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"y(B") ; Nukta Half Form
("\\($(5!I!i(B\\)" "$(5#I(B") ; Nukta
("\\($(5!J!h(B\\)$(5!F!h(B[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"J(B") ("\\($(5!J!h!F(B\\)" "$(5${(B")
("\\($(5!J!h(B\\)$(5!J!h(B[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"J(B") ; Special Half Form
("\\($(5!J!h!J(B\\)" "$(5$|(B")
("\\($(5!J!h(B\\)$(5!T!h(B[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"J(B") ("\\($(5!J!h!T(B\\)" "$(5$}(B")
("\\($(5!J!h!O(B\\)" "$(5!J"q(B") ; Post "r"
("\\($(5!J!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"J(B") ("\\($(5!K!h!F(B\\)" "$(5$~(B")
("\\($(5!K!h!O(B\\)" "$(5!K"q(B") ; Post "r"
("\\($(5!K!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"K(B") ("\\($(5!L!h!F(B\\)" "$(5#P(B")
("\\($(5!L!h!Q(B\\)" "$(5#Q(B")
("\\($(5!L!h!O(B\\)" "$(5!L"q(B") ; Post "r"
("\\($(5!L!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"L(B") ("\\($(5!M!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"M(B") ; Half Form
("\\($(5!N!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"N(B") ("\\($(5!O!](B\\)" "$(5",(B")
("\\($(5!O!^(B\\)" "$(5"-(B")
("\\($(5!P!](B\\)" "$(5".(B")
("\\($(5!P!^(B\\)" "$(5"/(B")
("\\($(5!Q!h!Q(B\\)" "$(5#`(B" sanskrit)
("\\($(5!Q!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"Q(B") ; Half Form
("\\($(5!R!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"R(B") ("\\($(5!S!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"S(B") ; Half Form
("\\($(5!T!h!F(B\\)" "$(5#a(B")
("\\($(5!T!h!T(B\\)" "$(5#b(B")
("\\($(5!T!h!O(B\\)" "$(5!T"q(B") ("\\($(5!T!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"T(B") ; Half Form
("\\($(5!U!h!8!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"h(B") ("\\($(5!U!h!8(B\\)" "$(5#c(B")
("\\($(5!U!h!F(B\\)" "$(5#d(B")
("\\($(5!U!h!J(B\\)" "$(5#e(B")
("\\($(5!U!h!Q(B\\)" "$(5#f(B")
("\\($(5!U!h(B\\)$(5!T!h!O(B" "$(5"U(B") ; Special Half Form
("\\($(5!U!h!T!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"j(B") ("\\($(5!U!h!O!h!T(B\\)" "$(5#g(B")
("\\($(5!U!h!O!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"i(B") ; Half Form
("\\($(5!U!h!O(B\\)" "$(5")(B") ("\\($(5!U!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"U(B") ; Half Form
("\\($(5!V!h!=!h!O!h!M(B\\)" "$(5#h(B")
("\\($(5!V!h!=!h!M(B\\)" "$(5#i(B")
("\\($(5!V!h!=!h!T(B\\)" "$(5#j(B")
("\\($(5!V!h!=(B\\)" "$(5#k(B")
("\\($(5!V!h!>(B\\)" "$(5#l(B")
("\\($(5!V!h!O(B\\)" "$(5!V"q(B") ("\\($(5!V!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"V(B") ; Half Form
("\\($(5!W!h!F!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"W"F(B") ; Special Half Form
("\\($(5!W!h!F(B\\)" "$(5#m(B")
("\\($(5!W!h!O(B\\)" "$(5#n(B")
("\\($(5!W!h(B\\)[$(5!3(B-$(5!N!P(B-$(5!X(B]" "$(5"W(B") ("\\($(5!X!h!A(B\\)" "$(5#p(B")
("\\($(5!X!h!F(B\\)" "$(5#q(B")
("\\($(5!X!h!L(B\\)" "$(5#r(B")
("\\($(5!X!h!M(B\\)" "$(5#s(B")
("\\($(5!X!h!O(B\\)" "$(5#t(B")
("\\($(5!X!h!Q(B\\)" "$(5#u(B")
("\\($(5!X!h!T(B\\)" "$(5#v(B")
("\\($(5!X!_(B\\)" "$(5#R(B")
("[$(5!7!9!=!>!?!@!D!O!P!R!S!X(B]\\($(5!h!O(B\\)" "$(5"r(B")
("[$(5!6!8!C!E!F!H!J!K!L!M!T!V(B]\\($(5!h!O(B\\)" "$(5"q(B")
("$(5!?!i(B\\($(5!h!O(B\\)" "$(5"r(B")
("$(5!@!i(B\\($(5!h!O(B\\)" "$(5"r(B")
("\\($(5!!!i(B\\)" "$(5#!(B")
("\\($(5!&!i(B\\)" "$(5#&(B")
("\\($(5!'!i(B\\)" "$(5#'(B")
("\\($(5!*!i(B\\)" "$(5#*(B")
("\\($(5![!i(B\\)" "$(5#L(B")
("\\($(5!\!i(B\\)" "$(5#M(B")
("\\($(5!_!i(B\\)" "$(5#K(B")
("\\($(5!j!i(B\\)" "$(5#J(B")
("\\($(5!O!_!i(B\\)" "$(5#*"p(B")
("\\($(5!O![!i(B\\)" "$(5#&"p(B")
("\\($(5!O!\!i(B\\)" "$(5#'"p(B")
("\\($(5!O!_(B\\)" "$(5!*"p(B")
("\\($(5!h!M(B\\)" "$(5"](B")
)
"Alist of regexps of Devanagari character sequences vs composed characters.")
(let ((rules devanagari-char-to-glyph-rules))
(while rules
(let ((rule (car rules))
(chars) (char) (glyphs) (glyph))
(setq rules (cdr rules))
(string-match "\\\\(\\(.+\\)\\\\)" (car rule))
(setq chars (substring (car rule) (match-beginning 1) (match-end 1)))
(setq char (string-to-char chars))
(setq glyphs (cdr rule))
(setq glyph (string-to-char (car glyphs)))
(put-char-code-property
char 'char-to-glyph
;; We don't "cons" it since priority is top to down.
(append (get-char-code-property char 'char-to-glyph) (list rule)))
(if (and (< ?(5z(B glyph) ; Glyphs only.
(null (get-char-code-property glyph 'glyph-to-char)))
; One glyph may corresponds to multiple characters,
; e.g., surrounding vowel in Tamil, etc.
; but for Devanagari, we put this restriction
; to make sure the fact that one glyph corresponds to one char.
(put-char-code-property
glyph 'glyph-to-char
(cons (list (car glyphs) chars)
(get-char-code-property glyph 'glyph-to-char)
))))))
;;
;; Function used in both characters-to-glyphs conversion and
;; glyphs-to-characters conversion.
;;
(defun max-match-len (regexp)
"Return the maximum length of text that can match the pattern REGEXP.
Only [...] pattern of regexp is recognized."
(let ((len 0)
(index 0))
(while (string-match "\\[\\([^\]]\\)+\\]" regexp index)
(setq len (+ len (- (match-beginning 0) index) 1)
index (match-end 0)))
len))
;; Return t iff at least one member appears in both LIST1 and LIST2.
(defun intersecting-p (list1 list2)
(let ((found nil))
(while (and list1 (not found))
(if (memq (car list1) list2)
(setq found t)
(setq list1 (cdr list1))))
found))
(defun string-conversion-by-rule (source symbol &rest specs)
"Convert string SOURCE by rules stored in SYMBOL property of each character.
The remaining arguments forms a list SPECS that restricts applicable rules.
The rules has the form ((REGEXP STR RULE-SPEC ...) ...).
Each character sequence in STRING that matches REGEXP is
replaced by STR.
If SPECS is nil, only rules with no RULE-SPECs is applied. Otherwise
rules with no RULE-SPECS and rules that have at least one member of
SPECS in RULE-SPECs is applied.
Rules are tested in the order of the list, thus more specific rules
should be placed in front of less specific rules.
If rule is given in the forms of regexp '...\\(...\\)...', a character
sequence that matches the pattern inside of the parenthesis is the
subject of the match. Otherwise, the entire expression is the subject
of the match."
(let ((pos 0)
(dst-str ""))
(while (< pos (length source))
(let ((found nil)
(rules (get-char-code-property
(string-to-char
(substring source pos)) symbol)))
(while rules
(let* ((rule (car rules))
(regexp (car rule))
(replace-str (car (cdr rule)))
(rule-specs (cdr (cdr rule)))
search-pos)
(if (not (or (null rule-specs)
(intersecting-p specs rule-specs)))
(setq rules (cdr rules))
(if (null (string-match "\\\\(.+\\\\)" regexp))
(progn
(setq regexp (concat "\\(" regexp "\\)"))
(setq search-pos pos))
(setq search-pos (- pos (max-match-len
(substring regexp
(string-match "^[^\\\\]*" regexp)
(match-end 0))))))
(if (< search-pos 0) (setq search-pos 0))
(if (string-match regexp source search-pos)
(if (= (match-beginning 1) pos)
(progn
(setq dst-str (concat dst-str replace-str))
(setq rules nil) ; Get out of the loop.
(setq found t)
;; proceed `pos' for replaced characters.
(setq pos (match-end 1)))
(setq rules (cdr rules)))
(setq rules (cdr rules))))))
;; proceed to next position
(if (not found)
(setq dst-str (concat dst-str (substring source pos (1+ pos)))
pos (1+ pos)))))
dst-str))
;;
;; Convert Character Code to Glyph Code
;;
;;;###autoload
(defun char-to-glyph-devanagari (string &rest langs)
"Convert Devanagari characters in STRING to Devanagari glyphs.
Ligatures and special rules are processed."
(apply
'string-conversion-by-rule
(append (list string 'char-to-glyph) langs)))
;; Example:
;;(char-to-glyph-devanagari "$(5!X![!F!h!D!\(B") => "$(5!X!["F!D!\(B"
(defconst devanagari-composition-rules
'((?$(5!!(B 0 (tr . br))
(?$(5!"(B 0 (mr . mr))
(?$(5!#(B 0)
(?$(5!$(B 0)
(?$(5!%(B 0)
(?$(5!&(B 0)
(?$(5!'(B 0)
(?$(5!((B 0)
(?$(5!)(B 0)
(?$(5!*(B 0)
(?$(5!+(B 0)
(?$(5!,(B 0)
(?$(5!-(B 0)
(?$(5!.(B 0)
(?$(5!/(B 0)
(?$(5!0(B 0)
(?$(5!1(B 0)
(?$(5!2(B 0)
(?$(5!3(B 0)
(?$(5!4(B 0)
(?$(5!5(B 0)
(?$(5!6(B 0)
(?$(5!7(B 0)
(?$(5!8(B 0)
(?$(5!9(B 0)
(?$(5!:(B 0)
(?$(5!;(B 0)
(?$(5!<(B 0)
(?$(5!=(B 0)
(?$(5!>(B 0)
(?$(5!?(B 0)
(?$(5!@(B 0)
(?$(5!A(B 0)
(?$(5!B(B 0)
(?$(5!C(B 0)
(?$(5!D(B 0)
(?$(5!E(B 0)
(?$(5!F(B 0)
(?$(5!G(B 0)
(?$(5!H(B 0)
(?$(5!I(B 0)
(?$(5!J(B 0)
(?$(5!K(B 0)
(?$(5!L(B 0)
(?$(5!M(B 0)
(?$(5!N(B 0)
(?$(5!O(B 0)
(?$(5!P(B 0)
(?$(5!Q(B 0)
(?$(5!R(B 0)
(?$(5!S(B 0)
(?$(5!T(B 0)
(?$(5!U(B 0)
(?$(5!V(B 0)
(?$(5!W(B 0)
(?$(5!X(B 0)
(?$(5!Y(B 0)
(?$(5!Z(B 0)
(?$(5![(B 0 (ml . mr))
(?$(5!\(B 0)
(?$(5!](B 0 (br . tr))
(?$(5!^(B 0 (br . tr))
(?$(5!_(B 0 (br . tr))
(?$(5!`(B 0 (mr . mr)) ; (tc . bc)
(?$(5!a(B 0 (mr . mr))
(?$(5!b(B 0 (mr . mr))
(?$(5!c(B 0 (mr . mr))
(?$(5!d(B 0)
(?$(5!e(B 0)
(?$(5!f(B 0)
(?$(5!g(B 0)
(?$(5!h(B 0 (br . tr))
(?$(5!i(B 0 (br . tr))
(?$(5!j(B 0)
(nil 0)
(nil 0)
(nil 0)
(nil 0)
(nil 0)
(nil 0)
(?$(5!q(B 0)
(?$(5!r(B 0)
(?$(5!s(B 0)
(?$(5!t(B 0)
(?$(5!u(B 0)
(?$(5!v(B 0)
(?$(5!w(B 0)
(?$(5!x(B 0)
(?$(5!y(B 0)
(?$(5!z(B 0)
(nil 0)
(nil 0)
(nil 0)
(nil 0)
(?$(5"!(B 0)
(?$(5""(B 0)
(?$(5"#(B 0)
(?$(5"$(B 0)
(?$(5"%(B 0)
(?$(5"&(B 0)
(?$(5"'(B 0)
(?$(5"((B 0)
(?$(5")(B 0)
(?$(5"*(B 0)
(?$(5"+(B 0)
(?$(5",(B 0)
(?$(5"-(B 0)
(?$(5".(B 0)
(?$(5"/(B 0)
(?$(5"0(B 0)
(?$(5"1(B 0)
(?$(5"2(B 0)
(?$(5"3(B 0)
(?$(5"4(B 0)
(?$(5"5(B 0)
(?$(5"6(B 0)
(?$(5"7(B 0)
(?$(5"8(B 0)
(?$(5"9(B 0)
(?$(5":(B 0)
(?$(5";(B 0)
(?$(5"<(B 0)
(?$(5"=(B 0)
(?$(5">(B 0)
(?$(5"?(B 0)
(?$(5"@(B 0)
(?$(5"A(B 0)
(?$(5"B(B 0)
(?$(5"C(B 0)
(?$(5"D(B 0)
(?$(5"E(B 0)
(?$(5"F(B 0)
(?$(5"G(B 0)
(?$(5"H(B 0)
(?$(5"I(B 0)
(?$(5"J(B 0)
(?$(5"K(B 0)
(?$(5"L(B 0)
(?$(5"M(B 0)
(?$(5"N(B 0)
(?$(5"O(B 0)
(?$(5"P(B 0)
(?$(5"Q(B 0)
(?$(5"R(B 0)
(?$(5"S(B 0)
(?$(5"T(B 0)
(?$(5"U(B 0)
(?$(5"V(B 0)
(?$(5"W(B 0)
(?$(5"X(B 0)
(?$(5"Y(B 0)
(?$(5"Z(B 0)
(?$(5"[(B 0)
(?$(5"\(B 0)
(?$(5"](B 0)
(?$(5"^(B 0)
(?$(5"_(B 0)
(?$(5"`(B 0)
(?$(5"a(B 0)
(?$(5"b(B 0)
(?$(5"c(B 0)
(?$(5"d(B 0)
(?$(5"e(B 0)
(?$(5"f(B 0)
(?$(5"g(B 0)
(?$(5"h(B 0)
(?$(5"i(B 0)
(?$(5"j(B 0)
(?$(5"k(B 0)
(?$(5"l(B 0)
(?$(5"m(B 0)
(?$(5"n(B 0)
(?$(5"o(B 0)
(?$(5"p(B 10 (mr . mr))
(?$(5"q(B 0 (br . br))
(?$(5"r(B 0 (br . tr))
(?$(5"s(B 0)
(?$(5"t(B 0)
(?$(5"u(B 0)
(?$(5"v(B 0)
(?$(5"w(B 0)
(?$(5"x(B 0)
(?$(5"y(B 0)
(?$(5"z(B 0)
(?$(5"{(B 0)
(?$(5"|(B 0)
(?$(5"}(B 0)
(?$(5"~(B 0)
(?$(5#!(B 0)
(?$(5#"(B 0)
(?$(5##(B 0)
(?$(5#$(B 0)
(?$(5#%(B 0)
(?$(5#&(B 0)
(?$(5#'(B 0)
(?$(5#((B 0)
(?$(5#)(B 0)
(?$(5#*(B 0)
(?$(5#+(B 0)
(?$(5#,(B 0)
(?$(5#-(B 0)
(?$(5#.(B 0)
(?$(5#/(B 0)
(?$(5#0(B 0)
(?$(5#1(B 0)
(?$(5#2(B 0)
(?$(5#3(B 0)
(?$(5#4(B 0)
(?$(5#5(B 0)
(?$(5#6(B 0)
(?$(5#7(B 0)
(?$(5#8(B 0)
(?$(5#9(B 0)
(?$(5#:(B 0)
(?$(5#;(B 0)
(?$(5#<(B 0)
(?$(5#=(B 0)
(?$(5#>(B 0)
(?$(5#?(B 0)
(?$(5#@(B 0)
(?$(5#A(B 0)
(?$(5#B(B 0)
(?$(5#C(B 0)
(?$(5#D(B 0)
(?$(5#E(B 0)
(?$(5#F(B 0)
(?$(5#G(B 0)
(?$(5#H(B 0)
(?$(5#I(B 0)
(?$(5#J(B 0)
(?$(5#K(B 0 (br . tr))
(?$(5#L(B 0 (br . tr))
(?$(5#M(B 0 (br . tr))
(?$(5#N(B 0)
(?$(5#O(B 0)
(?$(5#P(B 0)
(?$(5#Q(B 0)
(?$(5#R(B 0)
(?$(5#S(B 0)
(?$(5#T(B 0)
(?$(5#U(B 0)
(?$(5#V(B 0)
(?$(5#W(B 0)
(?$(5#X(B 0)
(?$(5#Y(B 0)
(?$(5#Z(B 0)
(?$(5#[(B 0)
(?$(5#\(B 0)
(?$(5#](B 0)
(?$(5#^(B 0)
(?$(5#_(B 0)
(?$(5#`(B 0)
(?$(5#a(B 0)
(?$(5#b(B 0)
(?$(5#c(B 0)
(?$(5#d(B 0)
(?$(5#e(B 0)
(?$(5#f(B 0)
(?$(5#g(B 0)
(?$(5#h(B 0)
(?$(5#i(B 0)
(?$(5#j(B 0)
(?$(5#k(B 0)
(?$(5#l(B 0)
(?$(5#m(B 0)
(?$(5#n(B 0)
(?$(5#o(B 0)
(?$(5#p(B 0)
(?$(5#q(B 0)
(?$(5#r(B 0)
(?$(5#s(B 0)
(?$(5#t(B 0)
(?$(5#u(B 0)
(?$(5#v(B 0)
(?$(5#w(B 0)
(?$(5#x(B 0)
(?$(5#y(B 0)
(?$(5#z(B 0)
(?$(5#{(B 0)
(?$(5#|(B 0)
(?$(5#}(B 0)
(?$(5#~(B 0)
(?$(5$!(B 0)
(?$(5$"(B 0)
(?$(5$#(B 0)
(?$(5$$(B 0)
(?$(5$%(B 0)
(?$(5$&(B 0)
(?$(5$'(B 0)
(?$(5$((B 0)
(?$(5$)(B 0)
(?$(5$*(B 0)
(?$(5$+(B 0)
(?$(5$,(B 0)
(?$(5$-(B 0)
(?$(5$.(B 0)
(?$(5$/(B 0)
(?$(5$0(B 0)
(?$(5$1(B 0)
(?$(5$2(B 0)
(?$(5$3(B 0)
(?$(5$4(B 0)
(?$(5$5(B 0)
(?$(5$6(B 0)
(?$(5$7(B 0)
(?$(5$8(B 0)
(?$(5$9(B 0)
(?$(5$:(B 0)
(?$(5$ (?$(5$<(B 0)
(?$(5$=(B 0)
(?$(5$>(B 0)
(?$(5$?(B 0)
(?$(5$@(B 0)
(?$(5$A(B 0)
(?$(5$B(B 0)
(?$(5$C(B 0)
(?$(5$D(B 0)
(?$(5$E(B 0)
(?$(5$F(B 0)
(?$(5$G(B 0)
(?$(5$H(B 0)
(?$(5$I(B 0)
(?$(5$J(B 0)
(?$(5$K(B 0)
(?$(5$L(B 0)
(?$(5$M(B 0)
(?$(5$N(B 0)
(?$(5$O(B 0)
(?$(5$P(B 0)
(?$(5$Q(B 0)
(?$(5$R(B 0)
(?$(5$S(B 0)
(?$(5$T(B 0)
(?$(5$U(B 0)
(?$(5$V(B 0)
(?$(5$W(B 0)
(?$(5$X(B 0)
(?$(5$Y(B 0)
(?$(5$Z(B 0)
(?$(5$[(B 0)
(?$(5$\(B 0)
(?$(5$](B 0)
(?$(5$^(B 0)
(?$(5$_(B 0)
(?$(5$`(B 0)
(?$(5$a(B 0)
(?$(5$b(B 0)
(?$(5$c(B 0)
(?$(5$d(B 0)
(?$(5$e(B 0)
(?$(5$f(B 0)
(?$(5$g(B 0)
(?$(5$h(B 0)
(?$(5$i(B 0)
(?$(5$j(B 0)
(?$(5$k(B 0)
(?$(5$l(B 0)
(?$(5$m(B 0)
(?$(5$n(B 0)
(?$(5$o(B 0)
(?$(5$p(B 0)
(?$(5$q(B 0)
(?$(5$r(B 0)
(?$(5$s(B 0)
(?$(5$t(B 0)
(?$(5$u(B 0)
(?$(5$v(B 0)
(?$(5$w(B 0)
(?$(5$x(B 0)
(?$(5$y(B 0)
(?$(5$z(B 0)
(?$(5${(B 0)
(?$(5$|(B 0)
(?$(5$}(B 0)
(?$(5$~(B 0)
))
(defun devanagari-reorder-glyphs-for-composition (string start end)
(let ((pos start)
(ordered-glyphs nil))
(while (< pos end)
(let ((glyph (aref string pos)))
(setq pos (1+ pos))
(setq ordered-glyphs
(append ordered-glyphs
(list (assq glyph devanagari-composition-rules))))))
(sort ordered-glyphs '(lambda (x y) (< (car (cdr x)) (car (cdr y)))))))
!
(defun devanagari-compose-to-one-glyph (devanagari-string)
(let* ((o-glyph-list (devanagari-reorder-glyphs-for-composition
devanagari-string 0 (length devanagari-string)))
(cmp-glyph-list (list (car (car o-glyph-list))))
(o-glyph-list (cdr o-glyph-list)))
(while o-glyph-list
(let* ((o-glyph (car o-glyph-list))
(glyph (if (< 2 (length o-glyph))
(list (car (cdr (cdr o-glyph))) (car o-glyph))
(list '(mr . ml) (car o-glyph)))))
(setq o-glyph-list (cdr o-glyph-list))
(setq cmp-glyph-list (append cmp-glyph-list glyph))))
(setq cmp-glyph-list (devanagari-wide-to-narrow cmp-glyph-list))
(if (= (length cmp-glyph-list) 1) (char-to-string (car cmp-glyph-list))
(apply 'compose-chars cmp-glyph-list))))
(defun devanagari-composition-component (string &optional start end)
(or start (setq start 0))
(or end (setq end (length string)))
(let* ((o-glyph-list (devanagari-reorder-glyphs-for-composition
string start end))
(cmp-glyph-list (list (car (car o-glyph-list)))))
(setq o-glyph-list (cdr o-glyph-list))
(while o-glyph-list
(let* ((o-glyph (car o-glyph-list))
(glyph (if (< 2 (length o-glyph))
(list (car (cdr (cdr o-glyph))) (car o-glyph))
(list '(mr . ml) (car o-glyph)))))
(setq o-glyph-list (cdr o-glyph-list))
(setq cmp-glyph-list (append cmp-glyph-list glyph))))
(devanagari-wide-to-narrow cmp-glyph-list)))
(defun devanagari-vertical-modifier-p (glyph)
(string-match (char-to-string glyph)
"[$(5!"!]!^!_!`!a!b!c!h!i"p"q"r#K#L#M(B]"))
(defun devanagari-non-vertical-modifier-p (glyph)
(string-match (char-to-string glyph)
"[$(5![(B]"))
(defun devanagari-wide-to-narrow-char (char)
"Convert Devanagari character CHAR to the corresponding narrow character.
If there's no corresponding narrow character, return CHAR as is."
(let ((narrow (cdr (assq char devanagari-1-column-char))))
(or narrow char)))
(defun devanagari-wide-to-narrow (src-list)
(devanagari-wide-to-narrow-iter src-list t))
(defun devanagari-wide-to-narrow-iter (src-list 2-col-glyph)
(let ((glyph (car src-list)))
(cond ((null src-list) '())
((not (numberp glyph))
(cons glyph
(devanagari-wide-to-narrow-iter (cdr src-list) 2-col-glyph)))
((devanagari-non-vertical-modifier-p glyph)
(cons (devanagari-wide-to-narrow-char glyph)
(devanagari-wide-to-narrow-iter (cdr src-list) 2-col-glyph)))
((devanagari-vertical-modifier-p glyph)
(if 2-col-glyph
(cons glyph
(devanagari-wide-to-narrow-iter (cdr src-list) t))
(cons (devanagari-wide-to-narrow-char glyph)
(devanagari-wide-to-narrow-iter (cdr src-list)
2-col-glyph))))
(t
(if (cdr (assq glyph devanagari-1-column-char))
(cons (devanagari-wide-to-narrow-char glyph)
(devanagari-wide-to-narrow-iter (cdr src-list) nil))
(cons glyph
(devanagari-wide-to-narrow-iter (cdr src-list) t)))))))
(defun devanagari-decompose-string (str)
"Decompose Devanagari string STR"
(decompose-string (copy-sequence str)))
(defun devanagari-decompose-region (from to)
(interactive "r")
(decompose-region from to))
(defun devanagari-compose-string (str &rest langs)
(setq str (copy-sequence str))
(let ((idx 0)
rest match-b match-e)
(while (string-match devanagari-composite-glyph-unit str idx)
(let* ((match-b (match-beginning 0))
(match-e (match-end 0))
(cmps (devanagari-composition-component
(apply
'char-to-glyph-devanagari
(cons (substring str match-b match-e) langs)))))
(compose-string str match-b match-e cmps)
(setq idx match-e))))
str)
(defun devanagari-compose-region (from to &rest langs)
(interactive "r")
(save-excursion
(save-restriction
(narrow-to-region from to)
(goto-char (point-min))
(while (re-search-forward devanagari-composite-glyph-unit nil t)
(let* ((match-b (match-beginning 0)) (match-e (match-end 0))
(cmps (devanagari-composition-component
(apply
'char-to-glyph-devanagari
(cons (buffer-substring match-b match-e) langs)))))
(compose-region match-b match-e cmps))))))
(defun devanagari-compose-from-is13194-region (from to)
"Compose IS 13194 characters in the region to Devanagari characters."
(interactive "r")
(save-excursion
(save-restriction
(narrow-to-region from to)
(indian-to-devanagari-region (point-min) (point-max))
(devanagari-compose-region (point-min) (point-max))
(- (point-max) (point-min)))))
(defun in-is13194-devanagari-post-read-conversion (len)
(let ((pos (point)))
(devanagari-compose-from-is13194-region pos (+ pos len))))
(defun devanagari-decompose-to-is13194-region (from to)
"Decompose Devanagari characters in the region to IS 13194 characters."
(interactive "r")
(save-excursion
(save-restriction
(narrow-to-region from to)
(devanagari-decompose-region (point-min) (point-max))
(devanagari-to-indian-region (point-min) (point-max)))))
(defun in-is13194-devanagari-pre-write-conversion (from to)
(let ((old-buf (current-buffer)))
(set-buffer (generate-new-buffer " *temp*"))
(if (stringp from)
(insert from)
(insert-buffer-substring old-buf from to))
(devanagari-decompose-to-is13194-region (point-min) (point-max))
nil))
(defun devanagari-encode-itrans-region (from to)
(interactive "r")
(save-restriction
(narrow-to-region from to)
(devanagari-decompose-to-is13194-region (point-min) (point-max))
(indian-encode-itrans-region (point-min) (point-max))))
(defun devanagari-decode-itrans-region (from to)
(interactive "r")
(save-restriction
(narrow-to-region from to)
(indian-decode-itrans-region (point-min) (point-max))
(devanagari-compose-from-is13194-region (point-min) (point-max))))
(provide 'devan-util)