1;;; devan-util.el --- Support for composing Devanagari characters 2 3;; Copyright (C) 1997, 1998, 2001, 2002, 2003, 2004, 2005, 2006, 2007 4;; Free Software Foundation, Inc. 5;; Copyright (C) 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007 6;; National Institute of Advanced Industrial Science and Technology (AIST) 7;; Registration Number H14PRO021 8 9;; Maintainer: KAWABATA, Taichi <kawabata@m17n.org> 10;; Keywords: multilingual, Devanagari 11 12;; This file is part of GNU Emacs. 13 14;; GNU Emacs is free software; you can redistribute it and/or modify 15;; it under the terms of the GNU General Public License as published by 16;; the Free Software Foundation; either version 2, or (at your option) 17;; any later version. 18 19;; GNU Emacs is distributed in the hope that it will be useful, 20;; but WITHOUT ANY WARRANTY; without even the implied warranty of 21;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 22;; GNU General Public License for more details. 23 24;; You should have received a copy of the GNU General Public License 25;; along with GNU Emacs; see the file COPYING. If not, write to the 26;; Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, 27;; Boston, MA 02110-1301, USA. 28 29;; Created: Feb. 17. 2001 30 31;;; Commentary: 32 33;; This file provides character(Unicode) to glyph(CDAC) conversion and 34;; composition of Devanagari script characters. 35 36;;; Code: 37 38;;;###autoload 39 40;; Devanagari Composable Pattern 41;; C .. Consonants 42;; V .. Vowel 43;; H .. Halant 44;; M .. Matra 45;; V .. Vowel 46;; A .. Anuswar 47;; D .. Chandrabindu 48;; (N .. Zerowidth Non Joiner) 49;; (J .. Zerowidth Joiner. ) 50;; 1. vowel 51;; V(A/D)? 52;; 2. syllable : maximum of 5 consecutive consonants. (e.g. kartsnya) 53;; ((CH)?(CH)?(CH)?CH)?C(H|M?(A|D)?)? 54 55(defconst devanagari-consonant 56 "[$,15U(B-$,15y68(B-$,16?(B]") 57 58(defconst devanagari-composable-pattern 59 (concat 60 "\\([$,15E(B-$,15T6@6A(B][$,15A5B(B]?\\)\\|[$,15C6D(B]" 61 "\\|\\(" 62 "\\(?:\\(?:[$,15U(B-$,15y68(B-$,16?(B]$,16-(B\\)?\\(?:[$,15U(B-$,15y68(B-$,16?(B]$,16-(B\\)?\\(?:[$,15U(B-$,15y68(B-$,16?(B]$,16-(B\\)?[$,15U(B-$,15y68(B-$,16?(B]$,16-(B\\)?" 63 "[$,15U(B-$,15y68(B-$,16?(B]\\(?:$,16-(B\\|[$,15~(B-$,16-6B6C(B]?[$,15B5A(B]?\\)?" 64 "\\)") 65 "Regexp matching a composable sequence of Devanagari characters.") 66 67;;;###autoload 68(defun devanagari-compose-region (from to) 69 (interactive "r") 70 (save-excursion 71 (save-restriction 72 (narrow-to-region from to) 73 (goto-char (point-min)) 74 (while (re-search-forward devanagari-composable-pattern nil t) 75 (devanagari-compose-syllable-region (match-beginning 0) 76 (match-end 0)))))) 77(defun devanagari-compose-string (string) 78 (with-temp-buffer 79 (insert (decompose-string string)) 80 (devanagari-compose-region (point-min) (point-max)) 81 (buffer-string))) 82 83;;;###autoload 84(defun devanagari-post-read-conversion (len) 85 (save-excursion 86 (save-restriction 87 (let ((buffer-modified-p (buffer-modified-p))) 88 (narrow-to-region (point) (+ (point) len)) 89 (devanagari-compose-region (point-min) (point-max)) 90 (set-buffer-modified-p buffer-modified-p) 91 (- (point-max) (point-min)))))) 92 93(defun devanagari-range (from to) 94 "Make the list of the integers of range FROM to TO." 95 (let (result) 96 (while (<= from to) (setq result (cons to result) to (1- to))) result)) 97 98(defun devanagari-regexp-of-hashtbl-keys (hashtbl) 99 "Return a regular expression that matches all keys in hashtable HASHTBL." 100 (let ((max-specpdl-size 1000)) 101 (regexp-opt 102 (sort 103 (let (dummy) 104 (maphash (function (lambda (key val) (setq dummy (cons key dummy)))) hashtbl) 105 dummy) 106 (function (lambda (x y) (> (length x) (length y)))))))) 107 108(defun devanagari-composition-function (from to pattern &optional string) 109 "Compose Devanagari characters in REGION, or STRING if specified. 110Assume that the REGION or STRING must fully match the composable 111PATTERN regexp." 112 (if string (devanagari-compose-syllable-string string) 113 (devanagari-compose-syllable-region from to)) 114 (- to from)) 115 116;; Register a function to compose Devanagari characters. 117(mapc 118 (function (lambda (ucs) 119 (aset composition-function-table (decode-char 'ucs ucs) 120 (list (cons devanagari-composable-pattern 121 'devanagari-composition-function))))) 122 (nconc '(#x0903) (devanagari-range #x0905 #x0939) (devanagari-range #x0958 #x0961))) 123 124;; Notes on conversion steps. 125 126;; 1. chars to glyphs 127;; 128;; Rules will not be applied to the halant appeared at the end of the 129;; text. Also, the preceding/following "r" will be treated as special case. 130 131;; 2. glyphs reordering. 132;; 133;; The glyphs are split by halant, and each glyph groups are 134;; re-ordered in the following order. 135;; 136;; Note that `consonant-glyph' mentioned here does not contain the 137;; vertical bar (right modifier) attached at the right of the 138;; consonant. 139;; 140;; If the glyph-group contains right modifier, 141;; (1) consonant-glyphs/vowels, with nukta sign 142;; (2) spacing 143;; (3) right modifier (may be matra) 144;; (4) top matra 145;; (5) preceding "r" 146;; (6) anuswar 147;; (7) following "r" 148;; (8) bottom matra or halant. 149;; 150;; Otherwise, 151;; (1) consonant-glyph/vowels, with nukta sign 152;; (3) left matra 153;; (4) top matra 154;; (5) preceding "r" 155;; (6) anuswar 156;; (7) following "r" 157;; (8) bottom matra or halant. 158;; (2) spacing 159 160;; 3. glyph to glyph 161;; 162;; For better display, some glyph display would be tuned. 163 164;; 4. Composition. 165;; 166;; left modifiers will be attached at the left. 167;; others will be attached right. 168 169;; Problem:: 170;; Can we generalize this methods to other Indian scripts? 171 172(defvar dev-char-glyph 173 '(("$,15E(B" . "$,4 K(B") 174 ("$,15F(B" . "$,4 K")(B") 175 ("$,15~(B" . "$,4")(B") 176 ("$,15G(B" . "$,4 \(B") 177 ("$,15(B" . "$,4"*(B") 178 ("$,155A(B" . "$,4"*(B\$,4"&(B") 179 ("$,15H(B" . "$,4 \"'(B") 180 ("$,15H5A(B" . "$,4 \"'"&(B") 181 ("$,16 (B" . "$,4"2(B") 182 ("$,16 5A(B" . "$,4"2"&(B") 183 ("$,15I(B" . "$,4 ](B") 184 ("$,16!(B" . "$,4"6(B") 185 ("$,15J(B" . "$,4 ^"P(B") 186 ("$,16"(B" . "$,4":(B") 187 ("$,15K(B" . "$,4 `"Q(B") 188 ("$,16#(B" . "$,4">(B") 189 ;;("$,15L(B" . nil) ; not implemented. 190 ("$,16$(B" . "$,4"?(B") 191 ("$,15M(B" . "$,4 b"L(B") 192 ("$,15M5A(B" . "$,4 b"$(B") 193 ("$,15M5B(B" . "$,4 b"$(B") 194 ("$,16%(B" . "\$,4"L(B") 195 ("$,15N(B" . "$,4 b"@(B") 196 ("$,15N5A(B" . "$,4 b"@"&(B") 197 ("$,16&(B" . "\$,4"@(B") 198 ("$,16&5A(B" . "\$,4"@(B\$,4"&(B") 199 ("$,15O(B" . "$,4 b(B") 200 ("$,16'(B" . "\$,4"D(B") 201 ("$,16'5A(B" . "\$,4"D(B\$,4"&(B") 202 ("$,15P(B" . "$,4 b"D(B") 203 ("$,15P5A(B" . "$,4 b"D"&(B") 204 ("$,16((B" . "\$,4"H(B") 205 ("$,16(5A(B" . "\$,4"H(B\$,4"&(B") 206 ("$,15Q(B" . "$,4 K")"L(B") ;; special rule for reodering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a"Q(B") 225 ;;("$,16B(B" . nil) 226 ;;("$,16A(B" . nil) 227 ;;("$,16C(B" . nil) 228 229 ;; GRUTTALS 230 ("$,15U(B" . "$,4 e"R(B") 231 ("$,15U6-(B" . "$,4 c(B") 232 ("$,15U6-5p(B" . "$,4 g"R(B") 233 ("$,15U6-5d(B" . "$,4 h"R(B") 234 ("$,15U6-5w(B" . "$,4 i")(B") 235 ("$,15U6-5w6-(B" . "$,4 i(B") 236 237 ("$,15V(B" . "$,4 j")(B") 238 ("$,15V6-(B" . "$,4 j(B") 239 ("$,15V6-5p(B" . "$,4 l")(B") 240 ("$,15V6-5p6-(B" . "$,4 l(B") 241 242 ("$,15W(B" . "$,4 m")(B") 243 ("$,15W6-(B" . "$,4 m(B") 244 ("$,15W6-5p(B" . "$,4 o")(B") 245 ("$,15W6-5p6-(B" . "$,4 o(B") 246 247 ("$,15X(B" . "$,4 p")(B") 248 ("$,15X6-(B" . "$,4 p(B") 249 ("$,15X6-5p(B" . "$,4 q")(B") 250 ("$,15X6-5p6-(B" . "$,4 q(B") 251 252 ("$,15Y(B" . "$,4 r"S(B") 253 ;; PALATALS 254 ("$,15Z(B" . "$,4 s")(B") 255 ("$,15Z6-(B" . "$,4 s(B") 256 ("$,15Z6-5p(B" . "$,4 t")(B") 257 ("$,15Z6-5p6-(B" . "$,4 t(B") 258 259 ("$,15[(B" . "$,4 u"T(B") 260 261 ("$,15\(B" . "$,4 v")(B") 262 ("$,15\6-(B" . "$,4 v(B") 263 ("$,15\6-5p(B" . "$,4 x")(B") 264 ("$,15\6-5p6-(B" . "$,4 x(B") 265 ("$,15\6-5^(B" . "$,4 y")(B") 266 ("$,15\6-5^6-(B" . "$,4 y(B") 267 268 ("$,15](B" . "$,4 z")(B") 269 ("$,15]6-(B" . "$,4 z(B") 270 ("$,15]6-5p(B" . "$,4 {")(B") 271 ("$,15]6-5p6-(B" . "$,4 {(B") 272 273 ("$,15^(B" . "$,4 |")(B") 274 ("$,15^6-(B" . "$,4 |(B") 275 ;; CEREBRALS 276 ("$,15_(B" . "$,4 }"U(B") 277 ("$,15_6-5_(B" . "$,4 ~"U(B") 278 ("$,15_6-5`(B" . "$,4 "U(B") 279 280 ("$,15`(B" . "$,4! "V(B") 281 ("$,15`6-5`(B" . "$,4!!"V(B") 282 283 ("$,15a(B" . "$,4!""W(B") 284 ("$,15a6-5a(B" . "$,4!$"W(B") 285 ("$,15a6-5b(B" . "$,4!%"W(B") 286 287 ("$,15b(B" . "$,4!&"X(B") 288 289 ("$,15c(B" . "$,4!(")(B") 290 ("$,15c6-(B" . "$,4!((B") 291 ;; DENTALS 292 ("$,15d(B" . "$,4!)")(B") 293 ("$,15d6-(B" . "$,4!)(B") 294 ("$,15d6-5p(B" . "$,4!*")(B") 295 ("$,15d6-5p6-(B" . "$,4!*(B") 296 ("$,15d6-5d(B" . "$,4!+")(B") 297 ("$,15d6-5d6-(B" . "$,4!+(B") 298 299 ("$,15e(B" . "$,4!,")(B") 300 ("$,15e6-(B" . "$,4!,(B") 301 ("$,15e6-5p(B" . "$,4!-")(B") 302 ("$,15e6-5p6-(B" . "$,4!-(B") 303 304 ("$,15f(B" . "$,4!."Y(B") 305 ("$,15f6#(B" . "$,4!/"Y(B") 306 ("$,15f6-5p(B" . "$,4!0"Y(B") 307 ("$,15f6-5f(B" . "$,4!1"Y(B") 308 ("$,15f6-5g(B" . "$,4!2"Y(B") 309 ("$,15f6-5n(B" . "$,4!3(B") 310 ("$,15f6-5o(B" . "$,4!4(B") 311 ("$,15f6-5u(B" . "$,4!5"Y(B") 312 313 ("$,15g(B" . "$,4!6")(B") 314 ("$,15g6-(B" . "$,4!6(B") 315 ("$,15g6-5p(B" . "$,4!7")(B") 316 ("$,15g6-5p6-(B" . "$,4!7(B") 317 318 ("$,15h(B" . "$,4!8")(B") 319 ("$,15h6-(B" . "$,4!8(B") 320 ("$,15h6-5p(B" . "$,4!9")(B") 321 ("$,15h6-5p6-(B" . "$,4!9")(B") 322 ("$,15h6-5h(B" . "$,4!:")(B") 323 ("$,15h6-5h6-(B" . "$,4!:(B") 324 325 ("$,15i(B" . "$,4!8"#")(B") 326 ;; LABIALS 327 ("$,15j(B" . "$,4!;")(B") 328 ("$,15j6-(B" . "$,4!;(B") 329 ("$,15j6-5p(B" . "$,4!<")(B") 330 ("$,15j6-5p6-(B" . "$,4!<(B") 331 332 ("$,15k(B" . "$,4!a"[(B") 333 ("$,15k6-(B" . "$,4!=(B") 334 ("$,15k6-5p(B" . "$,4!c"[(B") 335 336 ("$,15l(B" . "$,4!d")(B") 337 ("$,15l6-(B" . "$,4!d(B") 338 ("$,15l6-5p(B" . "$,4!e")(B") 339 ("$,15l6-5p6-(B" . "$,4!e(B") 340 341 ("$,15m(B" . "$,4!f")(B") 342 ("$,15m6-(B" . "$,4!f(B") 343 ("$,15m6-5p(B" . "$,4!g")(B") 344 ("$,15m6-5p6-(B" . "$,4!g(B") 345 346 ("$,15n(B" . "$,4!h")(B") 347 ("$,15n6-(B" . "$,4!h(B") 348 ("$,15n6-5p(B" . "$,4!i")(B") 349 ("$,15n6-5p6-(B" . "$,4!i(B") 350 ;; SEMIVOWELS 351 ("$,15o(B" . "$,4!j")(B") 352 ("$,15o6-(B" . "$,4!j(B") 353 ("$,15o6-5p(B" . "$,4!k")(B") 354 ("$,15o6-5p6-(B" . "$,4!k(B") 355 ("$,16-5o(B" . "$,4!l(B") ;; when every ohter lig. fails. 356 357 ("$,15p(B" . "$,4!n"W(B") 358 ;; ("$,15p6-(B" . "\$,4"'(B") ;; special case. only the topmost pos. 359 ("$,15q(B" . "$,4!n"#"W(B") 360 ("$,15q6-(B" . "$,4!m(B") ;; IS 13194 speical rule. 361 ("$,15p6!(B" . "$,4!o"[(B") 362 ("$,15p6"(B" . "$,4!p"\(B") 363 364 ("$,15r(B" . "$,4!q")(B") 365 ("$,15r6-(B" . "$,4!q(B") 366 ("$,15s(B" . "$,4!s(B") 367 ("$,15s6-(B" . "$,4!r(B") 368 ("$,15t(B" . "$,4!s"#(B") 369 ("$,15t6-(B" . "$,4!r"#(B") 370 371 ("$,15u(B" . "$,4!t")(B") 372 ("$,15u6-(B" . "$,4!t(B") 373 ("$,15u6-5p(B" . "$,4!u")(B") 374 ("$,15u6-5p6-(B" . "$,4!u(B") 375 ;; SIBILANTS 376 ("$,15v(B" . "$,4!v")(B") 377 ("$,15v6-(B" . "$,4!v(B") 378 ("$,15v6-5u(B" . "$,4!w")(B") 379 ("$,15v6-5u6-(B" . "$,4!w(B") 380 ("$,15v6-5p(B" . "$,4!x")(B") 381 ("$,15v6-5p6-(B" . "$,4!x(B") 382 383 ("$,15w(B" . "$,4!y")(B") 384 ("$,15w6-(B" . "$,4!y(B") 385 ("$,15x(B" . "$,4!z")(B") 386 ("$,15x6-(B" . "$,4!z(B") 387 ("$,15x6-5p(B" . "$,4!{")(B") 388 ("$,15x6-5p6-(B" . "$,4!{(B") 389 390 ("$,15y(B" . "$,4!}(B") 391 ("$,15y6-(B" . "$,4!|(B") 392 ("$,15y6#(B" . "$,4!~(B") 393 ("$,15y6-5p(B" . "$,4!(B") 394 ("$,15y6-5n(B" . "$,4" (B") 395 ("$,15y6-5o(B" . "$,4"!(B") 396 ;; NUKTAS 397 ("$,168(B" . "$,4 f"R"S(B") 398 ("$,1686-(B" . "$,4 d(B") 399 ("$,169(B" . "$,4 k")(B") 400 ("$,1696-(B" . "$,4 k(B") 401 ("$,16:(B" . "$,4 n")(B") 402 ("$,16:6-(B" . "$,4 n(B") 403 ("$,16;(B" . "$,4 w")(B") 404 ("$,16;6-(B" . "$,4 w(B") 405 ("$,16<(B" . "$,4!#"W(B") 406 ("$,16=(B" . "$,4!'"X(B") 407 ("$,16>(B" . "$,4!b"[(B") 408 ("$,16>6-(B" . "$,4!>(B") 409 ("$,16?(B" . "$,4!j"#")(B") 410 ;; misc modifiers. 411 ("$,15A(B" . "\$,4"$(B") 412 ("$,15B(B" . "\$,4"&(B") 413 ("$,15C(B" . "$,4 F(B") 414 ("$,15|(B" . "$,4"#(B") 415 ("$,15}(B" . "$,4 E(B") 416 ("$,16-(B" . "$,4""(B") 417 ("$,16-5p(B" . "$,4"%(B") ;; following "r" 418 ;; ("$,160(B" . "$,4 D(B") 419 ("$,16D(B" . "$,4 J(B") 420 ;; ("$,16F(B" . "") 421 ;; ("$,16G(B" . "") 422 ;; ("$,16H(B" . "") 423 ;; ("$,16I(B" . "") 424 ;; ("$,16J(B" . "") 425 ;; ("$,16K(B" . "") 426 ;; ("$,16L(B" . "") 427 ;; ("$,16M(B" . "") 428 ;; ("$,16N(B" . "") 429 ;; ("$,16O(B" . "") 430 ) 431 "Devanagari characters to glyphs conversion table. 432Default value contains only the basic rules. You may add your own 433preferred rule from the sanskrit fonts." ) 434 435(defvar dev-char-glyph-hash 436 (let* ((hash (make-hash-table :test 'equal))) 437 (mapc (function (lambda (x) (puthash (car x) (cdr x) hash))) 438 dev-char-glyph) 439 hash)) 440 441(defvar dev-char-glyph-regexp 442 (devanagari-regexp-of-hashtbl-keys dev-char-glyph-hash)) 443 444;; glyph-to-glyph conversion table. 445;; it is supposed that glyphs are ordered in 446;; [consonant/nukta] - [matra/halant] - [preceding-r] - [anuswar]. 447 448(defvar dev-glyph-glyph 449 '(("\$,4"'(B\$,4"&(B" . "\$,4"((B") 450 ("\$,4"'(B\$,4"$(B" . "\$,4"((B") 451 ("$,4"*(B\$,4"&(B" . "$,4"+(B") 452 ("$,4"*(B\$,4"'(B" . "$,4",(B") 453 ("$,4"*(B\$,4"'(B\$,4"&(B" . "$,4"-(B") 454 ("$,4"2(B\$,4"&(B" . "$,4"3(B") 455 ("$,4"2(B\$,4"'(B" . "$,4"4(B") 456 ("$,4"2(B\$,4"'(B\$,4"&(B" . "$,4"5(B") 457 ("$,4"#(B\$,4"6(B" . "$,4"7(B") 458 ("$,4"%(B\$,4"6(B" . "$,4"8(B") 459 ;;("$,4"6(B" . "$,4"9(B") 460 ("$,4"#(B\$,4":(B" . "$,4";(B") 461 ("$,4"%(B\$,4":(B" . "$,4"<(B") 462 ;;("$,4":(B" . "$,4"=(B") 463 ("\$,4"@(B\$,4"&(B" . "\$,4"A(B") 464 ("\$,4"@(B\$,4"'(B" . "\$,4"B(B") 465 ("\$,4"@(B\$,4"'(B\$,4"&(B" . "\$,4"C(B") 466 ("\$,4"D(B\$,4"&(B" . "\$,4"E(B") 467 ("\$,4"D(B\$,4"'(B" . "\$,4"F(B") 468 ("\$,4"D(B\$,4"'(B\$,4"&(B" . "\$,4"G(B") 469 ("\$,4"H(B\$,4"&(B" . "\$,4"I(B") 470 ("\$,4"H(B\$,4"'(B" . "\$,4"J(B") 471 ("\$,4"H(B\$,4"'(B\$,4"&(B" . "\$,4"K(B") 472 ("\$,4"L(B\$,4"&(B" . "\$,4"M(B") 473 ("\$,4"L(B\$,4"'(B" . "\$,4"N(B") 474 ("\$,4"L(B\$,4"'(B\$,4"&(B" . "\$,4"O(B") 475 )) 476(defvar dev-glyph-glyph-hash 477 (let* ((hash (make-hash-table :test 'equal))) 478 (mapc (function (lambda (x) (puthash (car x) (cdr x) hash))) 479 dev-glyph-glyph) 480 hash)) 481(defvar dev-glyph-glyph-regexp 482 (devanagari-regexp-of-hashtbl-keys dev-glyph-glyph-hash)) 483 484 485;; yet another glyph-to-glyph conversions. 486(defvar dev-glyph-glyph-2 487 '(("$,4"*(B" . "$,4".(B") 488 ("$,4"+(B" . "$,4"/(B") 489 ("$,4",(B" . "$,4"0(B") 490 ("$,4"-(B" . "$,4"1(B"))) 491(defvar dev-glyph-glyph-2-hash 492 (let* ((hash (make-hash-table :test 'equal))) 493 (mapc (function (lambda (x) (puthash (car x) (cdr x) hash))) 494 dev-glyph-glyph-2) 495 hash)) 496(defvar dev-glyph-glyph-2-regexp 497 (devanagari-regexp-of-hashtbl-keys dev-glyph-glyph-2-hash)) 498 499 500(defun dev-charseq (from &optional to) 501 (if (null to) (setq to from)) 502 (mapcar (function (lambda (x) (indian-glyph-char x 'devanagari))) 503 (devanagari-range from to))) 504 505(defvar dev-glyph-cvn 506 (append 507 (dev-charseq #x2b) 508 (dev-charseq #x3c #xc1) 509 (dev-charseq #xc3)) 510 "Devanagari Consonants/Vowels/Nukta Glyphs") 511 512(defvar dev-glyph-space 513 (dev-charseq #xf0 #xfe) 514 "Devanagari Spacing Glyphs") 515 516(defvar dev-glyph-right-modifier 517 (append 518 (dev-charseq #xc9) 519 (dev-charseq #xd2 #xd5)) 520 "Devanagari Modifiers attached at the right side.") 521 522(defvar dev-glyph-right-modifier-regexp 523 (concat "[" dev-glyph-right-modifier "]")) 524 525(defvar dev-glyph-left-matra 526 (dev-charseq #xca #xd1) 527 "Devanagari Matras attached at the left side.") 528 529(defvar dev-glyph-top-matra 530 (dev-charseq #xe0 #xef) 531 "Devanagari Matras attached at the top side.") 532 533(defvar dev-glyph-bottom-modifier 534 (append 535 (dev-charseq #xd6 #xdf) 536 (dev-charseq #xc2)) 537 "Devanagari Modifiers attached at the bottom.") 538 539(defvar dev-glyph-order 540 `((,dev-glyph-cvn . 1) 541 (,dev-glyph-space . 2) 542 (,dev-glyph-right-modifier . 3) 543 (,dev-glyph-left-matra . 3) ;; processed by reference point. 544 (,dev-glyph-top-matra . 4) 545 (,(dev-charseq #xc7 #xc8) . 5) 546 (,(dev-charseq #xc4) . 6) 547 (,(dev-charseq #xc6) . 6) 548 (,(dev-charseq #xc5) . 7) 549 (,dev-glyph-bottom-modifier . 8))) 550 551(mapc 552 (function (lambda (x) 553 (mapc 554 (function (lambda (y) 555 (put-char-code-property y 'composition-order (cdr x)))) 556 (car x)))) 557 dev-glyph-order) 558 559(mapc 560 (function (lambda (x) 561 (put-char-code-property x 'reference-point '(3 . 5)))) 562 dev-glyph-left-matra) 563 564(defun devanagari-compose-syllable-string (string) 565 (with-temp-buffer 566 (insert (decompose-string string)) 567 (devanagari-compose-syllable-region (point-min) (point-max)) 568 (buffer-string))) 569 570(defun devanagari-compose-syllable-region (from to) 571 "Compose devanagari syllable in region FROM to TO." 572 (let ((glyph-str nil) (cons-num 0) glyph-str-list 573 (last-halant nil) (preceding-r nil) (last-modifier nil) 574 (last-char (char-before to)) match-str 575 glyph-block split-pos) 576 (save-excursion 577 (save-restriction 578 ;;; *** char-to-glyph conversion *** 579 ;; Special rule 1. -- Last halant must be preserved. 580 (if (eq last-char ?$,16-(B) 581 (progn 582 (setq last-halant t) 583 (narrow-to-region from (1- to))) 584 (narrow-to-region from to) 585 ;; note if the last char is modifier. 586 (if (or (eq last-char ?$,15A(B) (eq last-char ?$,15B(B)) 587 (setq last-modifier t))) 588 (goto-char (point-min)) 589 ;; Special rule 2. -- preceding "r halant" must be modifier. 590 (when (looking-at "$,15p6-(B.") 591 (setq preceding-r t) 592 (goto-char (+ 2 (point)))) 593 ;; translate the rest characters into glyphs 594 (while (re-search-forward dev-char-glyph-regexp nil t) 595 (setq match-str (match-string 0)) 596 (setq glyph-str 597 (concat glyph-str 598 (gethash match-str dev-char-glyph-hash))) 599 ;; count the number of consonant-glyhs. 600 (if (string-match devanagari-consonant match-str) 601 (setq cons-num (1+ cons-num)))) 602 ;; preceding-r must be attached before the anuswar if exists. 603 (if preceding-r 604 (if last-modifier 605 (setq glyph-str (concat (substring glyph-str 0 -1) 606 "$,4"'(B" (substring glyph-str -1))) 607 (setq glyph-str (concat glyph-str "$,4"'(B")))) 608 (if last-halant (setq glyph-str (concat glyph-str "$,4""(B"))) 609 ;;; *** glyph-to-glyph conversion *** 610 (when (string-match dev-glyph-glyph-regexp glyph-str) 611 (setq glyph-str 612 (replace-match (gethash (match-string 0 glyph-str) 613 dev-glyph-glyph-hash) 614 nil t glyph-str)) 615 (if (and (> cons-num 1) 616 (string-match dev-glyph-glyph-2-regexp glyph-str)) 617 (setq glyph-str 618 (replace-match (gethash (match-string 0 glyph-str) 619 dev-glyph-glyph-2-hash) 620 nil t glyph-str)))) 621 ;;; *** glyph reordering *** 622 (while (setq split-pos (string-match "$,4""(B\\|.$" glyph-str)) 623 (setq glyph-block (substring glyph-str 0 (1+ split-pos))) 624 (setq glyph-str (substring glyph-str (1+ split-pos))) 625 (setq 626 glyph-block 627 (if (string-match dev-glyph-right-modifier-regexp glyph-block) 628 (sort (string-to-list glyph-block) 629 (function (lambda (x y) 630 (< (get-char-code-property x 'composition-order) 631 (get-char-code-property y 'composition-order))))) 632 (sort (string-to-list glyph-block) 633 (function (lambda (x y) 634 (let ((xo (get-char-code-property x 'composition-order)) 635 (yo (get-char-code-property y 'composition-order))) 636 (if (= xo 2) nil (if (= yo 2) t (< xo yo))))))))) 637 (setq glyph-str-list (nconc glyph-str-list glyph-block))) 638 ;; concatenate and attach reference-points. 639 (setq glyph-str 640 (cdr 641 (apply 642 'nconc 643 (mapcar 644 (function (lambda (x) 645 (list 646 (or (get-char-code-property x 'reference-point) 647 '(5 . 3) ;; default reference point. 648 ) 649 x))) 650 glyph-str-list)))))) 651 (compose-region from to glyph-str))) 652 653(provide 'devan-util) 654 655;;; arch-tag: 9bc4d6e3-f2b9-4110-886e-ff9b66b7eebc 656;;; devan-util.el ends here 657