| 1 | /* GNU Emacs case conversion functions. |
| 2 | Copyright (C) 1985, 1994, 1997, 1998, 1999, 2001, 2002, 2003, 2004, |
| 3 | 2005, 2006, 2007, 2008 Free Software Foundation, Inc. |
| 4 | |
| 5 | This file is part of GNU Emacs. |
| 6 | |
| 7 | GNU Emacs is free software: you can redistribute it and/or modify |
| 8 | it under the terms of the GNU General Public License as published by |
| 9 | the Free Software Foundation, either version 3 of the License, or |
| 10 | (at your option) any later version. |
| 11 | |
| 12 | GNU Emacs is distributed in the hope that it will be useful, |
| 13 | but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 14 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 15 | GNU General Public License for more details. |
| 16 | |
| 17 | You should have received a copy of the GNU General Public License |
| 18 | along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>. */ |
| 19 | |
| 20 | |
| 21 | #include <config.h> |
| 22 | #include "lisp.h" |
| 23 | #include "buffer.h" |
| 24 | #include "character.h" |
| 25 | #include "commands.h" |
| 26 | #include "syntax.h" |
| 27 | #include "composite.h" |
| 28 | #include "keymap.h" |
| 29 | |
| 30 | enum case_action {CASE_UP, CASE_DOWN, CASE_CAPITALIZE, CASE_CAPITALIZE_UP}; |
| 31 | |
| 32 | Lisp_Object Qidentity; |
| 33 | \f |
| 34 | Lisp_Object |
| 35 | casify_object (flag, obj) |
| 36 | enum case_action flag; |
| 37 | Lisp_Object obj; |
| 38 | { |
| 39 | register int c, c1; |
| 40 | register int inword = flag == CASE_DOWN; |
| 41 | |
| 42 | /* If the case table is flagged as modified, rescan it. */ |
| 43 | if (NILP (XCHAR_TABLE (current_buffer->downcase_table)->extras[1])) |
| 44 | Fset_case_table (current_buffer->downcase_table); |
| 45 | |
| 46 | if (INTEGERP (obj)) |
| 47 | { |
| 48 | int flagbits = (CHAR_ALT | CHAR_SUPER | CHAR_HYPER |
| 49 | | CHAR_SHIFT | CHAR_CTL | CHAR_META); |
| 50 | int flags = XINT (obj) & flagbits; |
| 51 | int multibyte = ! NILP (current_buffer->enable_multibyte_characters); |
| 52 | |
| 53 | /* If the character has higher bits set |
| 54 | above the flags, return it unchanged. |
| 55 | It is not a real character. */ |
| 56 | if ((unsigned) XFASTINT (obj) > (unsigned) flagbits) |
| 57 | return obj; |
| 58 | |
| 59 | c1 = XFASTINT (obj) & ~flagbits; |
| 60 | /* FIXME: Even if enable-multibyte-characters is nil, we may |
| 61 | manipulate multibyte chars. This means we have a bug for latin-1 |
| 62 | chars since when we receive an int 128-255 we can't tell whether |
| 63 | it's an eight-bit byte or a latin-1 char. */ |
| 64 | if (c1 >= 256) |
| 65 | multibyte = 1; |
| 66 | if (! multibyte) |
| 67 | MAKE_CHAR_MULTIBYTE (c1); |
| 68 | c = DOWNCASE (c1); |
| 69 | if (inword) |
| 70 | XSETFASTINT (obj, c | flags); |
| 71 | else if (c == (XFASTINT (obj) & ~flagbits)) |
| 72 | { |
| 73 | if (! inword) |
| 74 | c = UPCASE1 (c1); |
| 75 | if (! multibyte) |
| 76 | MAKE_CHAR_UNIBYTE (c); |
| 77 | XSETFASTINT (obj, c | flags); |
| 78 | } |
| 79 | return obj; |
| 80 | } |
| 81 | |
| 82 | if (!STRINGP (obj)) |
| 83 | wrong_type_argument (Qchar_or_string_p, obj); |
| 84 | else if (!STRING_MULTIBYTE (obj)) |
| 85 | { |
| 86 | EMACS_INT i; |
| 87 | EMACS_INT size = SCHARS (obj); |
| 88 | |
| 89 | obj = Fcopy_sequence (obj); |
| 90 | for (i = 0; i < size; i++) |
| 91 | { |
| 92 | c = SREF (obj, i); |
| 93 | MAKE_CHAR_MULTIBYTE (c); |
| 94 | c1 = c; |
| 95 | if (inword && flag != CASE_CAPITALIZE_UP) |
| 96 | c = DOWNCASE (c); |
| 97 | else if (!UPPERCASEP (c) |
| 98 | && (!inword || flag != CASE_CAPITALIZE_UP)) |
| 99 | c = UPCASE1 (c1); |
| 100 | if ((int) flag >= (int) CASE_CAPITALIZE) |
| 101 | inword = (SYNTAX (c) == Sword); |
| 102 | if (c != c1) |
| 103 | { |
| 104 | MAKE_CHAR_UNIBYTE (c); |
| 105 | /* If the char can't be converted to a valid byte, just don't |
| 106 | change it. */ |
| 107 | if (c >= 0 && c < 256) |
| 108 | SSET (obj, i, c); |
| 109 | } |
| 110 | } |
| 111 | return obj; |
| 112 | } |
| 113 | else |
| 114 | { |
| 115 | EMACS_INT i, i_byte, size = SCHARS (obj); |
| 116 | int len; |
| 117 | USE_SAFE_ALLOCA; |
| 118 | unsigned char *dst, *o; |
| 119 | /* Over-allocate by 12%: this is a minor overhead, but should be |
| 120 | sufficient in 99.999% of the cases to avoid a reallocation. */ |
| 121 | EMACS_INT o_size = SBYTES (obj) + SBYTES (obj) / 8 + MAX_MULTIBYTE_LENGTH; |
| 122 | SAFE_ALLOCA (dst, void *, o_size); |
| 123 | o = dst; |
| 124 | |
| 125 | for (i = i_byte = 0; i < size; i++, i_byte += len) |
| 126 | { |
| 127 | if ((o - dst) + MAX_MULTIBYTE_LENGTH > o_size) |
| 128 | { /* Not enough space for the next char: grow the destination. */ |
| 129 | unsigned char *old_dst = dst; |
| 130 | o_size += o_size; /* Probably overkill, but extremely rare. */ |
| 131 | SAFE_ALLOCA (dst, void *, o_size); |
| 132 | bcopy (old_dst, dst, o - old_dst); |
| 133 | o = dst + (o - old_dst); |
| 134 | } |
| 135 | c = STRING_CHAR_AND_LENGTH (SDATA (obj) + i_byte, 0, len); |
| 136 | if (inword && flag != CASE_CAPITALIZE_UP) |
| 137 | c = DOWNCASE (c); |
| 138 | else if (!UPPERCASEP (c) |
| 139 | && (!inword || flag != CASE_CAPITALIZE_UP)) |
| 140 | c = UPCASE1 (c); |
| 141 | if ((int) flag >= (int) CASE_CAPITALIZE) |
| 142 | inword = (SYNTAX (c) == Sword); |
| 143 | o += CHAR_STRING (c, o); |
| 144 | } |
| 145 | eassert (o - dst <= o_size); |
| 146 | obj = make_multibyte_string (dst, size, o - dst); |
| 147 | SAFE_FREE (); |
| 148 | return obj; |
| 149 | } |
| 150 | } |
| 151 | |
| 152 | DEFUN ("upcase", Fupcase, Supcase, 1, 1, 0, |
| 153 | doc: /* Convert argument to upper case and return that. |
| 154 | The argument may be a character or string. The result has the same type. |
| 155 | The argument object is not altered--the value is a copy. |
| 156 | See also `capitalize', `downcase' and `upcase-initials'. */) |
| 157 | (obj) |
| 158 | Lisp_Object obj; |
| 159 | { |
| 160 | return casify_object (CASE_UP, obj); |
| 161 | } |
| 162 | |
| 163 | DEFUN ("downcase", Fdowncase, Sdowncase, 1, 1, 0, |
| 164 | doc: /* Convert argument to lower case and return that. |
| 165 | The argument may be a character or string. The result has the same type. |
| 166 | The argument object is not altered--the value is a copy. */) |
| 167 | (obj) |
| 168 | Lisp_Object obj; |
| 169 | { |
| 170 | return casify_object (CASE_DOWN, obj); |
| 171 | } |
| 172 | |
| 173 | DEFUN ("capitalize", Fcapitalize, Scapitalize, 1, 1, 0, |
| 174 | doc: /* Convert argument to capitalized form and return that. |
| 175 | This means that each word's first character is upper case |
| 176 | and the rest is lower case. |
| 177 | The argument may be a character or string. The result has the same type. |
| 178 | The argument object is not altered--the value is a copy. */) |
| 179 | (obj) |
| 180 | Lisp_Object obj; |
| 181 | { |
| 182 | return casify_object (CASE_CAPITALIZE, obj); |
| 183 | } |
| 184 | |
| 185 | /* Like Fcapitalize but change only the initials. */ |
| 186 | |
| 187 | DEFUN ("upcase-initials", Fupcase_initials, Supcase_initials, 1, 1, 0, |
| 188 | doc: /* Convert the initial of each word in the argument to upper case. |
| 189 | Do not change the other letters of each word. |
| 190 | The argument may be a character or string. The result has the same type. |
| 191 | The argument object is not altered--the value is a copy. */) |
| 192 | (obj) |
| 193 | Lisp_Object obj; |
| 194 | { |
| 195 | return casify_object (CASE_CAPITALIZE_UP, obj); |
| 196 | } |
| 197 | \f |
| 198 | /* flag is CASE_UP, CASE_DOWN or CASE_CAPITALIZE or CASE_CAPITALIZE_UP. |
| 199 | b and e specify range of buffer to operate on. */ |
| 200 | |
| 201 | void |
| 202 | casify_region (flag, b, e) |
| 203 | enum case_action flag; |
| 204 | Lisp_Object b, e; |
| 205 | { |
| 206 | register int c; |
| 207 | register int inword = flag == CASE_DOWN; |
| 208 | register int multibyte = !NILP (current_buffer->enable_multibyte_characters); |
| 209 | EMACS_INT start, end; |
| 210 | EMACS_INT start_byte, end_byte; |
| 211 | EMACS_INT first = -1, last; /* Position of first and last changes. */ |
| 212 | EMACS_INT opoint = PT; |
| 213 | EMACS_INT opoint_byte = PT_BYTE; |
| 214 | |
| 215 | if (EQ (b, e)) |
| 216 | /* Not modifying because nothing marked */ |
| 217 | return; |
| 218 | |
| 219 | /* If the case table is flagged as modified, rescan it. */ |
| 220 | if (NILP (XCHAR_TABLE (current_buffer->downcase_table)->extras[1])) |
| 221 | Fset_case_table (current_buffer->downcase_table); |
| 222 | |
| 223 | validate_region (&b, &e); |
| 224 | start = XFASTINT (b); |
| 225 | end = XFASTINT (e); |
| 226 | modify_region (current_buffer, start, end, 0); |
| 227 | record_change (start, end - start); |
| 228 | start_byte = CHAR_TO_BYTE (start); |
| 229 | end_byte = CHAR_TO_BYTE (end); |
| 230 | |
| 231 | while (start < end) |
| 232 | { |
| 233 | int c2, len; |
| 234 | |
| 235 | if (multibyte) |
| 236 | { |
| 237 | c = FETCH_MULTIBYTE_CHAR (start_byte); |
| 238 | len = CHAR_BYTES (c); |
| 239 | } |
| 240 | else |
| 241 | { |
| 242 | c = FETCH_BYTE (start_byte); |
| 243 | MAKE_CHAR_MULTIBYTE (c); |
| 244 | len = 1; |
| 245 | } |
| 246 | c2 = c; |
| 247 | if (inword && flag != CASE_CAPITALIZE_UP) |
| 248 | c = DOWNCASE (c); |
| 249 | else if (!UPPERCASEP (c) |
| 250 | && (!inword || flag != CASE_CAPITALIZE_UP)) |
| 251 | c = UPCASE1 (c); |
| 252 | if ((int) flag >= (int) CASE_CAPITALIZE) |
| 253 | inword = ((SYNTAX (c) == Sword) && (inword || !SYNTAX_PREFIX (c))); |
| 254 | if (c != c2) |
| 255 | { |
| 256 | last = start; |
| 257 | if (first < 0) |
| 258 | first = start; |
| 259 | |
| 260 | if (! multibyte) |
| 261 | { |
| 262 | MAKE_CHAR_UNIBYTE (c); |
| 263 | FETCH_BYTE (start_byte) = c; |
| 264 | } |
| 265 | else if (ASCII_CHAR_P (c2) && ASCII_CHAR_P (c)) |
| 266 | FETCH_BYTE (start_byte) = c; |
| 267 | else |
| 268 | { |
| 269 | int tolen = CHAR_BYTES (c); |
| 270 | int j; |
| 271 | unsigned char str[MAX_MULTIBYTE_LENGTH]; |
| 272 | |
| 273 | CHAR_STRING (c, str); |
| 274 | if (len == tolen) |
| 275 | { |
| 276 | /* Length is unchanged. */ |
| 277 | for (j = 0; j < len; ++j) |
| 278 | FETCH_BYTE (start_byte + j) = str[j]; |
| 279 | } |
| 280 | else |
| 281 | { |
| 282 | /* Replace one character with the other, |
| 283 | keeping text properties the same. */ |
| 284 | replace_range_2 (start, start_byte, |
| 285 | start + 1, start_byte + len, |
| 286 | str, 1, tolen, |
| 287 | 0); |
| 288 | len = tolen; |
| 289 | } |
| 290 | } |
| 291 | } |
| 292 | start++; |
| 293 | start_byte += len; |
| 294 | } |
| 295 | |
| 296 | if (PT != opoint) |
| 297 | TEMP_SET_PT_BOTH (opoint, opoint_byte); |
| 298 | |
| 299 | if (first >= 0) |
| 300 | { |
| 301 | signal_after_change (first, last + 1 - first, last + 1 - first); |
| 302 | update_compositions (first, last + 1, CHECK_ALL); |
| 303 | } |
| 304 | } |
| 305 | |
| 306 | DEFUN ("upcase-region", Fupcase_region, Supcase_region, 2, 2, "r", |
| 307 | doc: /* Convert the region to upper case. In programs, wants two arguments. |
| 308 | These arguments specify the starting and ending character numbers of |
| 309 | the region to operate on. When used as a command, the text between |
| 310 | point and the mark is operated on. |
| 311 | See also `capitalize-region'. */) |
| 312 | (beg, end) |
| 313 | Lisp_Object beg, end; |
| 314 | { |
| 315 | casify_region (CASE_UP, beg, end); |
| 316 | return Qnil; |
| 317 | } |
| 318 | |
| 319 | DEFUN ("downcase-region", Fdowncase_region, Sdowncase_region, 2, 2, "r", |
| 320 | doc: /* Convert the region to lower case. In programs, wants two arguments. |
| 321 | These arguments specify the starting and ending character numbers of |
| 322 | the region to operate on. When used as a command, the text between |
| 323 | point and the mark is operated on. */) |
| 324 | (beg, end) |
| 325 | Lisp_Object beg, end; |
| 326 | { |
| 327 | casify_region (CASE_DOWN, beg, end); |
| 328 | return Qnil; |
| 329 | } |
| 330 | |
| 331 | DEFUN ("capitalize-region", Fcapitalize_region, Scapitalize_region, 2, 2, "r", |
| 332 | doc: /* Convert the region to capitalized form. |
| 333 | Capitalized form means each word's first character is upper case |
| 334 | and the rest of it is lower case. |
| 335 | In programs, give two arguments, the starting and ending |
| 336 | character positions to operate on. */) |
| 337 | (beg, end) |
| 338 | Lisp_Object beg, end; |
| 339 | { |
| 340 | casify_region (CASE_CAPITALIZE, beg, end); |
| 341 | return Qnil; |
| 342 | } |
| 343 | |
| 344 | /* Like Fcapitalize_region but change only the initials. */ |
| 345 | |
| 346 | DEFUN ("upcase-initials-region", Fupcase_initials_region, |
| 347 | Supcase_initials_region, 2, 2, "r", |
| 348 | doc: /* Upcase the initial of each word in the region. |
| 349 | Subsequent letters of each word are not changed. |
| 350 | In programs, give two arguments, the starting and ending |
| 351 | character positions to operate on. */) |
| 352 | (beg, end) |
| 353 | Lisp_Object beg, end; |
| 354 | { |
| 355 | casify_region (CASE_CAPITALIZE_UP, beg, end); |
| 356 | return Qnil; |
| 357 | } |
| 358 | \f |
| 359 | static Lisp_Object |
| 360 | operate_on_word (arg, newpoint) |
| 361 | Lisp_Object arg; |
| 362 | EMACS_INT *newpoint; |
| 363 | { |
| 364 | Lisp_Object val; |
| 365 | int farend; |
| 366 | int iarg; |
| 367 | |
| 368 | CHECK_NUMBER (arg); |
| 369 | iarg = XINT (arg); |
| 370 | farend = scan_words (PT, iarg); |
| 371 | if (!farend) |
| 372 | farend = iarg > 0 ? ZV : BEGV; |
| 373 | |
| 374 | *newpoint = PT > farend ? PT : farend; |
| 375 | XSETFASTINT (val, farend); |
| 376 | |
| 377 | return val; |
| 378 | } |
| 379 | |
| 380 | DEFUN ("upcase-word", Fupcase_word, Supcase_word, 1, 1, "p", |
| 381 | doc: /* Convert following word (or ARG words) to upper case, moving over. |
| 382 | With negative argument, convert previous words but do not move. |
| 383 | See also `capitalize-word'. */) |
| 384 | (arg) |
| 385 | Lisp_Object arg; |
| 386 | { |
| 387 | Lisp_Object beg, end; |
| 388 | EMACS_INT newpoint; |
| 389 | XSETFASTINT (beg, PT); |
| 390 | end = operate_on_word (arg, &newpoint); |
| 391 | casify_region (CASE_UP, beg, end); |
| 392 | SET_PT (newpoint); |
| 393 | return Qnil; |
| 394 | } |
| 395 | |
| 396 | DEFUN ("downcase-word", Fdowncase_word, Sdowncase_word, 1, 1, "p", |
| 397 | doc: /* Convert following word (or ARG words) to lower case, moving over. |
| 398 | With negative argument, convert previous words but do not move. */) |
| 399 | (arg) |
| 400 | Lisp_Object arg; |
| 401 | { |
| 402 | Lisp_Object beg, end; |
| 403 | EMACS_INT newpoint; |
| 404 | XSETFASTINT (beg, PT); |
| 405 | end = operate_on_word (arg, &newpoint); |
| 406 | casify_region (CASE_DOWN, beg, end); |
| 407 | SET_PT (newpoint); |
| 408 | return Qnil; |
| 409 | } |
| 410 | |
| 411 | DEFUN ("capitalize-word", Fcapitalize_word, Scapitalize_word, 1, 1, "p", |
| 412 | doc: /* Capitalize the following word (or ARG words), moving over. |
| 413 | This gives the word(s) a first character in upper case |
| 414 | and the rest lower case. |
| 415 | With negative argument, capitalize previous words but do not move. */) |
| 416 | (arg) |
| 417 | Lisp_Object arg; |
| 418 | { |
| 419 | Lisp_Object beg, end; |
| 420 | EMACS_INT newpoint; |
| 421 | XSETFASTINT (beg, PT); |
| 422 | end = operate_on_word (arg, &newpoint); |
| 423 | casify_region (CASE_CAPITALIZE, beg, end); |
| 424 | SET_PT (newpoint); |
| 425 | return Qnil; |
| 426 | } |
| 427 | \f |
| 428 | void |
| 429 | syms_of_casefiddle () |
| 430 | { |
| 431 | Qidentity = intern ("identity"); |
| 432 | staticpro (&Qidentity); |
| 433 | defsubr (&Supcase); |
| 434 | defsubr (&Sdowncase); |
| 435 | defsubr (&Scapitalize); |
| 436 | defsubr (&Supcase_initials); |
| 437 | defsubr (&Supcase_region); |
| 438 | defsubr (&Sdowncase_region); |
| 439 | defsubr (&Scapitalize_region); |
| 440 | defsubr (&Supcase_initials_region); |
| 441 | defsubr (&Supcase_word); |
| 442 | defsubr (&Sdowncase_word); |
| 443 | defsubr (&Scapitalize_word); |
| 444 | } |
| 445 | |
| 446 | void |
| 447 | keys_of_casefiddle () |
| 448 | { |
| 449 | initial_define_key (control_x_map, Ctl('U'), "upcase-region"); |
| 450 | Fput (intern ("upcase-region"), Qdisabled, Qt); |
| 451 | initial_define_key (control_x_map, Ctl('L'), "downcase-region"); |
| 452 | Fput (intern ("downcase-region"), Qdisabled, Qt); |
| 453 | |
| 454 | initial_define_key (meta_map, 'u', "upcase-word"); |
| 455 | initial_define_key (meta_map, 'l', "downcase-word"); |
| 456 | initial_define_key (meta_map, 'c', "capitalize-word"); |
| 457 | } |
| 458 | |
| 459 | /* arch-tag: 60a73c66-5489-47e7-a81f-cead4057c526 |
| 460 | (do not change this comment) */ |