| 1 | /* srfi-13.c --- SRFI-13 procedures for Guile |
| 2 | * |
| 3 | * Copyright (C) 2001, 2004, 2005, 2006, 2008, 2009, 2010, 2011 Free Software Foundation, Inc. |
| 4 | * |
| 5 | * This library is free software; you can redistribute it and/or |
| 6 | * modify it under the terms of the GNU Lesser General Public License |
| 7 | * as published by the Free Software Foundation; either version 3 of |
| 8 | * the License, or (at your option) any later version. |
| 9 | * |
| 10 | * This library is distributed in the hope that it will be useful, but |
| 11 | * WITHOUT ANY WARRANTY; without even the implied warranty of |
| 12 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 13 | * Lesser General Public License for more details. |
| 14 | * |
| 15 | * You should have received a copy of the GNU Lesser General Public |
| 16 | * License along with this library; if not, write to the Free Software |
| 17 | * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA |
| 18 | * 02110-1301 USA |
| 19 | */ |
| 20 | |
| 21 | |
| 22 | #ifdef HAVE_CONFIG_H |
| 23 | # include <config.h> |
| 24 | #endif |
| 25 | |
| 26 | #include <string.h> |
| 27 | #include <unicase.h> |
| 28 | #include <unictype.h> |
| 29 | |
| 30 | #include "libguile.h" |
| 31 | |
| 32 | #include <libguile/deprecation.h> |
| 33 | #include "libguile/srfi-13.h" |
| 34 | #include "libguile/srfi-14.h" |
| 35 | |
| 36 | #define MY_VALIDATE_SUBSTRING_SPEC(pos_str, str, \ |
| 37 | pos_start, start, c_start, \ |
| 38 | pos_end, end, c_end) \ |
| 39 | do { \ |
| 40 | SCM_VALIDATE_STRING (pos_str, str); \ |
| 41 | scm_i_get_substring_spec (scm_i_string_length (str), \ |
| 42 | start, &c_start, end, &c_end); \ |
| 43 | } while (0) |
| 44 | |
| 45 | #define MY_SUBF_VALIDATE_SUBSTRING_SPEC(fname, pos_str, str, \ |
| 46 | pos_start, start, c_start, \ |
| 47 | pos_end, end, c_end) \ |
| 48 | do { \ |
| 49 | SCM_ASSERT_TYPE (scm_is_string (str), str, pos_str, fname, "string"); \ |
| 50 | scm_i_get_substring_spec (scm_i_string_length (str), \ |
| 51 | start, &c_start, end, &c_end); \ |
| 52 | } while (0) |
| 53 | |
| 54 | #define REF_IN_CHARSET(s, i, cs) \ |
| 55 | (scm_is_true (scm_char_set_contains_p ((cs), SCM_MAKE_CHAR (scm_i_string_ref (s, i))))) |
| 56 | |
| 57 | SCM_DEFINE (scm_string_null_p, "string-null?", 1, 0, 0, |
| 58 | (SCM str), |
| 59 | "Return @code{#t} if @var{str}'s length is zero, and\n" |
| 60 | "@code{#f} otherwise.\n" |
| 61 | "@lisp\n" |
| 62 | "(string-null? \"\") @result{} #t\n" |
| 63 | "y @result{} \"foo\"\n" |
| 64 | "(string-null? y) @result{} #f\n" |
| 65 | "@end lisp") |
| 66 | #define FUNC_NAME s_scm_string_null_p |
| 67 | { |
| 68 | SCM_VALIDATE_STRING (1, str); |
| 69 | return scm_from_bool (scm_i_string_length (str) == 0); |
| 70 | } |
| 71 | #undef FUNC_NAME |
| 72 | |
| 73 | #if 0 |
| 74 | static void |
| 75 | race_error () |
| 76 | { |
| 77 | scm_misc_error (NULL, "race condition detected", SCM_EOL); |
| 78 | } |
| 79 | #endif |
| 80 | |
| 81 | SCM_DEFINE (scm_string_any, "string-any-c-code", 2, 2, 0, |
| 82 | (SCM char_pred, SCM s, SCM start, SCM end), |
| 83 | "Check if @var{char_pred} is true for any character in string @var{s}.\n" |
| 84 | "\n" |
| 85 | "@var{char_pred} can be a character to check for any equal to that, or\n" |
| 86 | "a character set (@pxref{Character Sets}) to check for any in that set,\n" |
| 87 | "or a predicate procedure to call.\n" |
| 88 | "\n" |
| 89 | "For a procedure, calls @code{(@var{char_pred} c)} are made\n" |
| 90 | "successively on the characters from @var{start} to @var{end}. If\n" |
| 91 | "@var{char_pred} returns true (ie.@: non-@code{#f}), @code{string-any}\n" |
| 92 | "stops and that return value is the return from @code{string-any}. The\n" |
| 93 | "call on the last character (ie.@: at @math{@var{end}-1}), if that\n" |
| 94 | "point is reached, is a tail call.\n" |
| 95 | "\n" |
| 96 | "If there are no characters in @var{s} (ie.@: @var{start} equals\n" |
| 97 | "@var{end}) then the return is @code{#f}.\n") |
| 98 | #define FUNC_NAME s_scm_string_any |
| 99 | { |
| 100 | size_t cstart, cend; |
| 101 | SCM res = SCM_BOOL_F; |
| 102 | |
| 103 | MY_VALIDATE_SUBSTRING_SPEC (2, s, |
| 104 | 3, start, cstart, |
| 105 | 4, end, cend); |
| 106 | |
| 107 | if (SCM_CHARP (char_pred)) |
| 108 | { |
| 109 | size_t i; |
| 110 | for (i = cstart; i < cend; i ++) |
| 111 | if (scm_i_string_ref (s, i) == SCM_CHAR (char_pred)) |
| 112 | { |
| 113 | res = SCM_BOOL_T; |
| 114 | break; |
| 115 | } |
| 116 | } |
| 117 | else if (SCM_CHARSETP (char_pred)) |
| 118 | { |
| 119 | size_t i; |
| 120 | for (i = cstart; i < cend; i++) |
| 121 | if (REF_IN_CHARSET (s, i, char_pred)) |
| 122 | { |
| 123 | res = SCM_BOOL_T; |
| 124 | break; |
| 125 | } |
| 126 | } |
| 127 | else |
| 128 | { |
| 129 | SCM_ASSERT (scm_is_true (scm_procedure_p (char_pred)), |
| 130 | char_pred, SCM_ARG1, FUNC_NAME); |
| 131 | |
| 132 | while (cstart < cend) |
| 133 | { |
| 134 | res = scm_call_1 (char_pred, |
| 135 | SCM_MAKE_CHAR (scm_i_string_ref (s, cstart))); |
| 136 | if (scm_is_true (res)) |
| 137 | break; |
| 138 | cstart++; |
| 139 | } |
| 140 | } |
| 141 | |
| 142 | scm_remember_upto_here_1 (s); |
| 143 | return res; |
| 144 | } |
| 145 | #undef FUNC_NAME |
| 146 | |
| 147 | |
| 148 | SCM_DEFINE (scm_string_every, "string-every-c-code", 2, 2, 0, |
| 149 | (SCM char_pred, SCM s, SCM start, SCM end), |
| 150 | "Check if @var{char_pred} is true for every character in string\n" |
| 151 | "@var{s}.\n" |
| 152 | "\n" |
| 153 | "@var{char_pred} can be a character to check for every character equal\n" |
| 154 | "to that, or a character set (@pxref{Character Sets}) to check for\n" |
| 155 | "every character being in that set, or a predicate procedure to call.\n" |
| 156 | "\n" |
| 157 | "For a procedure, calls @code{(@var{char_pred} c)} are made\n" |
| 158 | "successively on the characters from @var{start} to @var{end}. If\n" |
| 159 | "@var{char_pred} returns @code{#f}, @code{string-every} stops and\n" |
| 160 | "returns @code{#f}. The call on the last character (ie.@: at\n" |
| 161 | "@math{@var{end}-1}), if that point is reached, is a tail call and the\n" |
| 162 | "return from that call is the return from @code{string-every}.\n" |
| 163 | "\n" |
| 164 | "If there are no characters in @var{s} (ie.@: @var{start} equals\n" |
| 165 | "@var{end}) then the return is @code{#t}.\n") |
| 166 | #define FUNC_NAME s_scm_string_every |
| 167 | { |
| 168 | size_t cstart, cend; |
| 169 | SCM res = SCM_BOOL_T; |
| 170 | |
| 171 | MY_VALIDATE_SUBSTRING_SPEC (2, s, |
| 172 | 3, start, cstart, |
| 173 | 4, end, cend); |
| 174 | if (SCM_CHARP (char_pred)) |
| 175 | { |
| 176 | size_t i; |
| 177 | for (i = cstart; i < cend; i++) |
| 178 | if (scm_i_string_ref (s, i) != SCM_CHAR (char_pred)) |
| 179 | { |
| 180 | res = SCM_BOOL_F; |
| 181 | break; |
| 182 | } |
| 183 | } |
| 184 | else if (SCM_CHARSETP (char_pred)) |
| 185 | { |
| 186 | size_t i; |
| 187 | for (i = cstart; i < cend; i++) |
| 188 | if (!REF_IN_CHARSET (s, i, char_pred)) |
| 189 | { |
| 190 | res = SCM_BOOL_F; |
| 191 | break; |
| 192 | } |
| 193 | } |
| 194 | else |
| 195 | { |
| 196 | SCM_ASSERT (scm_is_true (scm_procedure_p (char_pred)), |
| 197 | char_pred, SCM_ARG1, FUNC_NAME); |
| 198 | |
| 199 | while (cstart < cend) |
| 200 | { |
| 201 | res = scm_call_1 (char_pred, |
| 202 | SCM_MAKE_CHAR (scm_i_string_ref (s, cstart))); |
| 203 | if (scm_is_false (res)) |
| 204 | break; |
| 205 | cstart++; |
| 206 | } |
| 207 | } |
| 208 | |
| 209 | scm_remember_upto_here_1 (s); |
| 210 | return res; |
| 211 | } |
| 212 | #undef FUNC_NAME |
| 213 | |
| 214 | |
| 215 | SCM_DEFINE (scm_string_tabulate, "string-tabulate", 2, 0, 0, |
| 216 | (SCM proc, SCM len), |
| 217 | "@var{proc} is an integer->char procedure. Construct a string\n" |
| 218 | "of size @var{len} by applying @var{proc} to each index to\n" |
| 219 | "produce the corresponding string element. The order in which\n" |
| 220 | "@var{proc} is applied to the indices is not specified.") |
| 221 | #define FUNC_NAME s_scm_string_tabulate |
| 222 | { |
| 223 | size_t clen, i; |
| 224 | SCM res; |
| 225 | SCM ch; |
| 226 | |
| 227 | SCM_ASSERT (scm_is_true (scm_procedure_p (proc)), |
| 228 | proc, SCM_ARG1, FUNC_NAME); |
| 229 | |
| 230 | SCM_ASSERT_RANGE (2, len, scm_to_int (len) >= 0); |
| 231 | clen = scm_to_size_t (len); |
| 232 | |
| 233 | { |
| 234 | /* This function is more complicated than necessary for the sake |
| 235 | of speed. */ |
| 236 | scm_t_wchar *buf = scm_malloc (clen * sizeof (scm_t_wchar)); |
| 237 | int wide = 0; |
| 238 | i = 0; |
| 239 | while (i < clen) |
| 240 | { |
| 241 | ch = scm_call_1 (proc, scm_from_size_t (i)); |
| 242 | if (!SCM_CHARP (ch)) |
| 243 | { |
| 244 | SCM_MISC_ERROR ("procedure ~S returned non-char", scm_list_1 (proc)); |
| 245 | } |
| 246 | if (SCM_CHAR (ch) > 255) |
| 247 | wide = 1; |
| 248 | buf[i] = SCM_CHAR (ch); |
| 249 | i++; |
| 250 | } |
| 251 | if (wide) |
| 252 | { |
| 253 | scm_t_wchar *wbuf = NULL; |
| 254 | res = scm_i_make_wide_string (clen, &wbuf); |
| 255 | memcpy (wbuf, buf, clen * sizeof (scm_t_wchar)); |
| 256 | free (buf); |
| 257 | } |
| 258 | else |
| 259 | { |
| 260 | char *nbuf = NULL; |
| 261 | res = scm_i_make_string (clen, &nbuf); |
| 262 | for (i = 0; i < clen; i ++) |
| 263 | nbuf[i] = (unsigned char) buf[i]; |
| 264 | free (buf); |
| 265 | } |
| 266 | } |
| 267 | |
| 268 | return res; |
| 269 | } |
| 270 | #undef FUNC_NAME |
| 271 | |
| 272 | |
| 273 | SCM_DEFINE (scm_substring_to_list, "string->list", 1, 2, 0, |
| 274 | (SCM str, SCM start, SCM end), |
| 275 | "Convert the string @var{str} into a list of characters.") |
| 276 | #define FUNC_NAME s_scm_substring_to_list |
| 277 | { |
| 278 | size_t cstart, cend; |
| 279 | int narrow; |
| 280 | SCM result = SCM_EOL; |
| 281 | |
| 282 | MY_VALIDATE_SUBSTRING_SPEC (1, str, |
| 283 | 2, start, cstart, |
| 284 | 3, end, cend); |
| 285 | |
| 286 | /* This explicit narrow/wide logic (instead of just using |
| 287 | scm_i_string_ref) is for speed optimizaion. */ |
| 288 | narrow = scm_i_is_narrow_string (str); |
| 289 | if (narrow) |
| 290 | { |
| 291 | const char *buf = scm_i_string_chars (str); |
| 292 | while (cstart < cend) |
| 293 | { |
| 294 | cend--; |
| 295 | result = scm_cons (SCM_MAKE_CHAR (buf[cend]), result); |
| 296 | } |
| 297 | } |
| 298 | else |
| 299 | { |
| 300 | const scm_t_wchar *buf = scm_i_string_wide_chars (str); |
| 301 | while (cstart < cend) |
| 302 | { |
| 303 | cend--; |
| 304 | result = scm_cons (SCM_MAKE_CHAR (buf[cend]), result); |
| 305 | } |
| 306 | } |
| 307 | scm_remember_upto_here_1 (str); |
| 308 | return result; |
| 309 | } |
| 310 | #undef FUNC_NAME |
| 311 | |
| 312 | /* We export scm_substring_to_list as "string->list" since it is |
| 313 | compatible and more general. This function remains for the benefit |
| 314 | of C code that used it. |
| 315 | */ |
| 316 | |
| 317 | SCM |
| 318 | scm_string_to_list (SCM str) |
| 319 | { |
| 320 | return scm_substring_to_list (str, SCM_UNDEFINED, SCM_UNDEFINED); |
| 321 | } |
| 322 | |
| 323 | SCM_DEFINE (scm_reverse_list_to_string, "reverse-list->string", 1, 0, 0, |
| 324 | (SCM chrs), |
| 325 | "An efficient implementation of @code{(compose string->list\n" |
| 326 | "reverse)}:\n" |
| 327 | "\n" |
| 328 | "@smalllisp\n" |
| 329 | "(reverse-list->string '(#\\a #\\B #\\c)) @result{} \"cBa\"\n" |
| 330 | "@end smalllisp") |
| 331 | #define FUNC_NAME s_scm_reverse_list_to_string |
| 332 | { |
| 333 | SCM result; |
| 334 | long i = scm_ilength (chrs), j; |
| 335 | char *data; |
| 336 | |
| 337 | if (i < 0) |
| 338 | SCM_WRONG_TYPE_ARG (1, chrs); |
| 339 | result = scm_i_make_string (i, &data); |
| 340 | |
| 341 | { |
| 342 | SCM rest; |
| 343 | rest = chrs; |
| 344 | j = 0; |
| 345 | while (j < i && scm_is_pair (rest)) |
| 346 | { |
| 347 | SCM elt = SCM_CAR (rest); |
| 348 | SCM_VALIDATE_CHAR (SCM_ARGn, elt); |
| 349 | j++; |
| 350 | rest = SCM_CDR (rest); |
| 351 | } |
| 352 | rest = chrs; |
| 353 | j = i; |
| 354 | result = scm_i_string_start_writing (result); |
| 355 | while (j > 0 && scm_is_pair (rest)) |
| 356 | { |
| 357 | SCM elt = SCM_CAR (rest); |
| 358 | scm_i_string_set_x (result, j-1, SCM_CHAR (elt)); |
| 359 | rest = SCM_CDR (rest); |
| 360 | j--; |
| 361 | } |
| 362 | scm_i_string_stop_writing (); |
| 363 | } |
| 364 | |
| 365 | return result; |
| 366 | } |
| 367 | #undef FUNC_NAME |
| 368 | |
| 369 | |
| 370 | SCM_SYMBOL (scm_sym_infix, "infix"); |
| 371 | SCM_SYMBOL (scm_sym_strict_infix, "strict-infix"); |
| 372 | SCM_SYMBOL (scm_sym_suffix, "suffix"); |
| 373 | SCM_SYMBOL (scm_sym_prefix, "prefix"); |
| 374 | |
| 375 | SCM_DEFINE (scm_string_join, "string-join", 1, 2, 0, |
| 376 | (SCM ls, SCM delimiter, SCM grammar), |
| 377 | "Append the string in the string list @var{ls}, using the string\n" |
| 378 | "@var{delim} as a delimiter between the elements of @var{ls}.\n" |
| 379 | "@var{grammar} is a symbol which specifies how the delimiter is\n" |
| 380 | "placed between the strings, and defaults to the symbol\n" |
| 381 | "@code{infix}.\n" |
| 382 | "\n" |
| 383 | "@table @code\n" |
| 384 | "@item infix\n" |
| 385 | "Insert the separator between list elements. An empty string\n" |
| 386 | "will produce an empty list.\n" |
| 387 | "@item string-infix\n" |
| 388 | "Like @code{infix}, but will raise an error if given the empty\n" |
| 389 | "list.\n" |
| 390 | "@item suffix\n" |
| 391 | "Insert the separator after every list element.\n" |
| 392 | "@item prefix\n" |
| 393 | "Insert the separator before each list element.\n" |
| 394 | "@end table") |
| 395 | #define FUNC_NAME s_scm_string_join |
| 396 | { |
| 397 | #define GRAM_INFIX 0 |
| 398 | #define GRAM_STRICT_INFIX 1 |
| 399 | #define GRAM_SUFFIX 2 |
| 400 | #define GRAM_PREFIX 3 |
| 401 | SCM tmp; |
| 402 | SCM result; |
| 403 | int gram = GRAM_INFIX; |
| 404 | size_t del_len = 0; |
| 405 | long strings = scm_ilength (ls); |
| 406 | |
| 407 | /* Validate the string list. */ |
| 408 | if (strings < 0) |
| 409 | SCM_WRONG_TYPE_ARG (1, ls); |
| 410 | |
| 411 | /* Validate the delimiter and record its length. */ |
| 412 | if (SCM_UNBNDP (delimiter)) |
| 413 | { |
| 414 | delimiter = scm_from_locale_string (" "); |
| 415 | del_len = 1; |
| 416 | } |
| 417 | else |
| 418 | { |
| 419 | SCM_VALIDATE_STRING (2, delimiter); |
| 420 | del_len = scm_i_string_length (delimiter); |
| 421 | } |
| 422 | |
| 423 | /* Validate the grammar symbol and remember the grammar. */ |
| 424 | if (SCM_UNBNDP (grammar)) |
| 425 | gram = GRAM_INFIX; |
| 426 | else if (scm_is_eq (grammar, scm_sym_infix)) |
| 427 | gram = GRAM_INFIX; |
| 428 | else if (scm_is_eq (grammar, scm_sym_strict_infix)) |
| 429 | gram = GRAM_STRICT_INFIX; |
| 430 | else if (scm_is_eq (grammar, scm_sym_suffix)) |
| 431 | gram = GRAM_SUFFIX; |
| 432 | else if (scm_is_eq (grammar, scm_sym_prefix)) |
| 433 | gram = GRAM_PREFIX; |
| 434 | else |
| 435 | SCM_WRONG_TYPE_ARG (3, grammar); |
| 436 | |
| 437 | /* Check grammar constraints. */ |
| 438 | if (strings == 0 && gram == GRAM_STRICT_INFIX) |
| 439 | SCM_MISC_ERROR ("strict-infix grammar requires non-empty list", |
| 440 | SCM_EOL); |
| 441 | |
| 442 | result = scm_i_make_string (0, NULL); |
| 443 | |
| 444 | tmp = ls; |
| 445 | switch (gram) |
| 446 | { |
| 447 | case GRAM_INFIX: |
| 448 | case GRAM_STRICT_INFIX: |
| 449 | while (scm_is_pair (tmp)) |
| 450 | { |
| 451 | result = scm_string_append (scm_list_2 (result, SCM_CAR (tmp))); |
| 452 | if (!scm_is_null (SCM_CDR (tmp)) && del_len > 0) |
| 453 | result = scm_string_append (scm_list_2 (result, delimiter)); |
| 454 | tmp = SCM_CDR (tmp); |
| 455 | } |
| 456 | break; |
| 457 | case GRAM_SUFFIX: |
| 458 | while (scm_is_pair (tmp)) |
| 459 | { |
| 460 | result = scm_string_append (scm_list_2 (result, SCM_CAR (tmp))); |
| 461 | if (del_len > 0) |
| 462 | result = scm_string_append (scm_list_2 (result, delimiter)); |
| 463 | tmp = SCM_CDR (tmp); |
| 464 | } |
| 465 | break; |
| 466 | case GRAM_PREFIX: |
| 467 | while (scm_is_pair (tmp)) |
| 468 | { |
| 469 | if (del_len > 0) |
| 470 | result = scm_string_append (scm_list_2 (result, delimiter)); |
| 471 | result = scm_string_append (scm_list_2 (result, SCM_CAR (tmp))); |
| 472 | tmp = SCM_CDR (tmp); |
| 473 | } |
| 474 | break; |
| 475 | } |
| 476 | |
| 477 | return result; |
| 478 | #undef GRAM_INFIX |
| 479 | #undef GRAM_STRICT_INFIX |
| 480 | #undef GRAM_SUFFIX |
| 481 | #undef GRAM_PREFIX |
| 482 | } |
| 483 | #undef FUNC_NAME |
| 484 | |
| 485 | |
| 486 | /* There are a number of functions to consider here for Scheme and C: |
| 487 | |
| 488 | string-copy STR [start [end]] ;; SRFI-13 variant of R5RS string-copy |
| 489 | substring/copy STR start [end] ;; Guile variant of R5RS substring |
| 490 | |
| 491 | scm_string_copy (str) ;; Old function from Guile |
| 492 | scm_substring_copy (str, [start, [end]]) |
| 493 | ;; C version of SRFI-13 string-copy |
| 494 | ;; and C version of substring/copy |
| 495 | |
| 496 | The C function underlying string-copy is not exported to C |
| 497 | programs. scm_substring_copy is defined in strings.c as the |
| 498 | underlying function of substring/copy and allows an optional START |
| 499 | argument. |
| 500 | */ |
| 501 | |
| 502 | SCM scm_srfi13_substring_copy (SCM str, SCM start, SCM end); |
| 503 | |
| 504 | SCM_DEFINE (scm_srfi13_substring_copy, "string-copy", 1, 2, 0, |
| 505 | (SCM str, SCM start, SCM end), |
| 506 | "Return a freshly allocated copy of the string @var{str}. If\n" |
| 507 | "given, @var{start} and @var{end} delimit the portion of\n" |
| 508 | "@var{str} which is copied.") |
| 509 | #define FUNC_NAME s_scm_srfi13_substring_copy |
| 510 | { |
| 511 | size_t cstart, cend; |
| 512 | |
| 513 | MY_VALIDATE_SUBSTRING_SPEC (1, str, |
| 514 | 2, start, cstart, |
| 515 | 3, end, cend); |
| 516 | return scm_i_substring_copy (str, cstart, cend); |
| 517 | } |
| 518 | #undef FUNC_NAME |
| 519 | |
| 520 | SCM |
| 521 | scm_string_copy (SCM str) |
| 522 | { |
| 523 | if (!scm_is_string (str)) |
| 524 | scm_wrong_type_arg ("scm_string_copy", 0, str); |
| 525 | |
| 526 | return scm_i_substring (str, 0, scm_i_string_length (str)); |
| 527 | } |
| 528 | |
| 529 | SCM_DEFINE (scm_string_copy_x, "string-copy!", 3, 2, 0, |
| 530 | (SCM target, SCM tstart, SCM s, SCM start, SCM end), |
| 531 | "Copy the sequence of characters from index range [@var{start},\n" |
| 532 | "@var{end}) in string @var{s} to string @var{target}, beginning\n" |
| 533 | "at index @var{tstart}. The characters are copied left-to-right\n" |
| 534 | "or right-to-left as needed -- the copy is guaranteed to work,\n" |
| 535 | "even if @var{target} and @var{s} are the same string. It is an\n" |
| 536 | "error if the copy operation runs off the end of the target\n" |
| 537 | "string.") |
| 538 | #define FUNC_NAME s_scm_string_copy_x |
| 539 | { |
| 540 | size_t cstart, cend, ctstart, dummy, len, i; |
| 541 | SCM sdummy = SCM_UNDEFINED; |
| 542 | |
| 543 | MY_VALIDATE_SUBSTRING_SPEC (1, target, |
| 544 | 2, tstart, ctstart, |
| 545 | 2, sdummy, dummy); |
| 546 | MY_VALIDATE_SUBSTRING_SPEC (3, s, |
| 547 | 4, start, cstart, |
| 548 | 5, end, cend); |
| 549 | len = cend - cstart; |
| 550 | SCM_ASSERT_RANGE (3, s, len <= scm_i_string_length (target) - ctstart); |
| 551 | |
| 552 | target = scm_i_string_start_writing (target); |
| 553 | for (i = 0; i < cend - cstart; i++) |
| 554 | { |
| 555 | scm_i_string_set_x (target, ctstart + i, |
| 556 | scm_i_string_ref (s, cstart + i)); |
| 557 | } |
| 558 | scm_i_string_stop_writing (); |
| 559 | scm_remember_upto_here_1 (target); |
| 560 | |
| 561 | return SCM_UNSPECIFIED; |
| 562 | } |
| 563 | #undef FUNC_NAME |
| 564 | |
| 565 | SCM_DEFINE (scm_substring_move_x, "substring-move!", 5, 0, 0, |
| 566 | (SCM str1, SCM start1, SCM end1, SCM str2, SCM start2), |
| 567 | "Copy the substring of @var{str1} bounded by @var{start1} and @var{end1}\n" |
| 568 | "into @var{str2} beginning at position @var{start2}.\n" |
| 569 | "@var{str1} and @var{str2} can be the same string.") |
| 570 | #define FUNC_NAME s_scm_substring_move_x |
| 571 | { |
| 572 | return scm_string_copy_x (str2, start2, str1, start1, end1); |
| 573 | } |
| 574 | #undef FUNC_NAME |
| 575 | |
| 576 | SCM_DEFINE (scm_string_take, "string-take", 2, 0, 0, |
| 577 | (SCM s, SCM n), |
| 578 | "Return the @var{n} first characters of @var{s}.") |
| 579 | #define FUNC_NAME s_scm_string_take |
| 580 | { |
| 581 | return scm_substring (s, SCM_INUM0, n); |
| 582 | } |
| 583 | #undef FUNC_NAME |
| 584 | |
| 585 | |
| 586 | SCM_DEFINE (scm_string_drop, "string-drop", 2, 0, 0, |
| 587 | (SCM s, SCM n), |
| 588 | "Return all but the first @var{n} characters of @var{s}.") |
| 589 | #define FUNC_NAME s_scm_string_drop |
| 590 | { |
| 591 | return scm_substring (s, n, SCM_UNDEFINED); |
| 592 | } |
| 593 | #undef FUNC_NAME |
| 594 | |
| 595 | |
| 596 | SCM_DEFINE (scm_string_take_right, "string-take-right", 2, 0, 0, |
| 597 | (SCM s, SCM n), |
| 598 | "Return the @var{n} last characters of @var{s}.") |
| 599 | #define FUNC_NAME s_scm_string_take_right |
| 600 | { |
| 601 | return scm_substring (s, |
| 602 | scm_difference (scm_string_length (s), n), |
| 603 | SCM_UNDEFINED); |
| 604 | } |
| 605 | #undef FUNC_NAME |
| 606 | |
| 607 | |
| 608 | SCM_DEFINE (scm_string_drop_right, "string-drop-right", 2, 0, 0, |
| 609 | (SCM s, SCM n), |
| 610 | "Return all but the last @var{n} characters of @var{s}.") |
| 611 | #define FUNC_NAME s_scm_string_drop_right |
| 612 | { |
| 613 | return scm_substring (s, |
| 614 | SCM_INUM0, |
| 615 | scm_difference (scm_string_length (s), n)); |
| 616 | } |
| 617 | #undef FUNC_NAME |
| 618 | |
| 619 | |
| 620 | SCM_DEFINE (scm_string_pad, "string-pad", 2, 3, 0, |
| 621 | (SCM s, SCM len, SCM chr, SCM start, SCM end), |
| 622 | "Take that characters from @var{start} to @var{end} from the\n" |
| 623 | "string @var{s} and return a new string, right-padded by the\n" |
| 624 | "character @var{chr} to length @var{len}. If the resulting\n" |
| 625 | "string is longer than @var{len}, it is truncated on the right.") |
| 626 | #define FUNC_NAME s_scm_string_pad |
| 627 | { |
| 628 | size_t cstart, cend, clen; |
| 629 | |
| 630 | MY_VALIDATE_SUBSTRING_SPEC (1, s, |
| 631 | 4, start, cstart, |
| 632 | 5, end, cend); |
| 633 | clen = scm_to_size_t (len); |
| 634 | |
| 635 | if (SCM_UNBNDP (chr)) |
| 636 | chr = SCM_MAKE_CHAR (' '); |
| 637 | else |
| 638 | { |
| 639 | SCM_VALIDATE_CHAR (3, chr); |
| 640 | } |
| 641 | if (clen < (cend - cstart)) |
| 642 | return scm_i_substring (s, cend - clen, cend); |
| 643 | else |
| 644 | { |
| 645 | SCM result; |
| 646 | result = (scm_string_append |
| 647 | (scm_list_2 (scm_c_make_string (clen - (cend - cstart), chr), |
| 648 | scm_i_substring (s, cstart, cend)))); |
| 649 | return result; |
| 650 | } |
| 651 | } |
| 652 | #undef FUNC_NAME |
| 653 | |
| 654 | |
| 655 | SCM_DEFINE (scm_string_pad_right, "string-pad-right", 2, 3, 0, |
| 656 | (SCM s, SCM len, SCM chr, SCM start, SCM end), |
| 657 | "Take that characters from @var{start} to @var{end} from the\n" |
| 658 | "string @var{s} and return a new string, left-padded by the\n" |
| 659 | "character @var{chr} to length @var{len}. If the resulting\n" |
| 660 | "string is longer than @var{len}, it is truncated on the left.") |
| 661 | #define FUNC_NAME s_scm_string_pad_right |
| 662 | { |
| 663 | size_t cstart, cend, clen; |
| 664 | |
| 665 | MY_VALIDATE_SUBSTRING_SPEC (1, s, |
| 666 | 4, start, cstart, |
| 667 | 5, end, cend); |
| 668 | clen = scm_to_size_t (len); |
| 669 | |
| 670 | if (SCM_UNBNDP (chr)) |
| 671 | chr = SCM_MAKE_CHAR (' '); |
| 672 | else |
| 673 | { |
| 674 | SCM_VALIDATE_CHAR (3, chr); |
| 675 | } |
| 676 | if (clen < (cend - cstart)) |
| 677 | return scm_i_substring (s, cstart, cstart + clen); |
| 678 | else |
| 679 | { |
| 680 | SCM result; |
| 681 | |
| 682 | result = (scm_string_append |
| 683 | (scm_list_2 (scm_i_substring (s, cstart, cend), |
| 684 | scm_c_make_string (clen - (cend - cstart), chr)))); |
| 685 | |
| 686 | return result; |
| 687 | } |
| 688 | } |
| 689 | #undef FUNC_NAME |
| 690 | |
| 691 | |
| 692 | SCM_DEFINE (scm_string_trim, "string-trim", 1, 3, 0, |
| 693 | (SCM s, SCM char_pred, SCM start, SCM end), |
| 694 | "Trim @var{s} by skipping over all characters on the left\n" |
| 695 | "that satisfy the parameter @var{char_pred}:\n" |
| 696 | "\n" |
| 697 | "@itemize @bullet\n" |
| 698 | "@item\n" |
| 699 | "if it is the character @var{ch}, characters equal to\n" |
| 700 | "@var{ch} are trimmed,\n" |
| 701 | "\n" |
| 702 | "@item\n" |
| 703 | "if it is a procedure @var{pred} characters that\n" |
| 704 | "satisfy @var{pred} are trimmed,\n" |
| 705 | "\n" |
| 706 | "@item\n" |
| 707 | "if it is a character set, characters in that set are trimmed.\n" |
| 708 | "@end itemize\n" |
| 709 | "\n" |
| 710 | "If called without a @var{char_pred} argument, all whitespace is\n" |
| 711 | "trimmed.") |
| 712 | #define FUNC_NAME s_scm_string_trim |
| 713 | { |
| 714 | size_t cstart, cend; |
| 715 | |
| 716 | MY_VALIDATE_SUBSTRING_SPEC (1, s, |
| 717 | 3, start, cstart, |
| 718 | 4, end, cend); |
| 719 | if (SCM_UNBNDP (char_pred)) |
| 720 | { |
| 721 | while (cstart < cend) |
| 722 | { |
| 723 | if (!uc_is_c_whitespace (scm_i_string_ref (s, cstart))) |
| 724 | break; |
| 725 | cstart++; |
| 726 | } |
| 727 | } |
| 728 | else if (SCM_CHARP (char_pred)) |
| 729 | { |
| 730 | while (cstart < cend) |
| 731 | { |
| 732 | if (scm_i_string_ref (s, cstart) != SCM_CHAR (char_pred)) |
| 733 | break; |
| 734 | cstart++; |
| 735 | } |
| 736 | } |
| 737 | else if (SCM_CHARSETP (char_pred)) |
| 738 | { |
| 739 | while (cstart < cend) |
| 740 | { |
| 741 | if (!REF_IN_CHARSET (s, cstart, char_pred)) |
| 742 | break; |
| 743 | cstart++; |
| 744 | } |
| 745 | } |
| 746 | else |
| 747 | { |
| 748 | SCM_ASSERT (scm_is_true (scm_procedure_p (char_pred)), |
| 749 | char_pred, SCM_ARG2, FUNC_NAME); |
| 750 | |
| 751 | while (cstart < cend) |
| 752 | { |
| 753 | SCM res; |
| 754 | |
| 755 | res = scm_call_1 (char_pred, SCM_MAKE_CHAR (scm_i_string_ref (s, cstart))); |
| 756 | if (scm_is_false (res)) |
| 757 | break; |
| 758 | cstart++; |
| 759 | } |
| 760 | } |
| 761 | return scm_i_substring (s, cstart, cend); |
| 762 | } |
| 763 | #undef FUNC_NAME |
| 764 | |
| 765 | |
| 766 | SCM_DEFINE (scm_string_trim_right, "string-trim-right", 1, 3, 0, |
| 767 | (SCM s, SCM char_pred, SCM start, SCM end), |
| 768 | "Trim @var{s} by skipping over all characters on the right\n" |
| 769 | "that satisfy the parameter @var{char_pred}:\n" |
| 770 | "\n" |
| 771 | "@itemize @bullet\n" |
| 772 | "@item\n" |
| 773 | "if it is the character @var{ch}, characters equal to @var{ch}\n" |
| 774 | "are trimmed,\n" |
| 775 | "\n" |
| 776 | "@item\n" |
| 777 | "if it is a procedure @var{pred} characters that satisfy\n" |
| 778 | "@var{pred} are trimmed,\n" |
| 779 | "\n" |
| 780 | "@item\n" |
| 781 | "if it is a character sets, all characters in that set are\n" |
| 782 | "trimmed.\n" |
| 783 | "@end itemize\n" |
| 784 | "\n" |
| 785 | "If called without a @var{char_pred} argument, all whitespace is\n" |
| 786 | "trimmed.") |
| 787 | #define FUNC_NAME s_scm_string_trim_right |
| 788 | { |
| 789 | size_t cstart, cend; |
| 790 | |
| 791 | MY_VALIDATE_SUBSTRING_SPEC (1, s, |
| 792 | 3, start, cstart, |
| 793 | 4, end, cend); |
| 794 | if (SCM_UNBNDP (char_pred)) |
| 795 | { |
| 796 | while (cstart < cend) |
| 797 | { |
| 798 | if (!uc_is_c_whitespace (scm_i_string_ref (s, cend - 1))) |
| 799 | break; |
| 800 | cend--; |
| 801 | } |
| 802 | } |
| 803 | else if (SCM_CHARP (char_pred)) |
| 804 | { |
| 805 | while (cstart < cend) |
| 806 | { |
| 807 | if (scm_i_string_ref (s, cend - 1) != SCM_CHAR (char_pred)) |
| 808 | break; |
| 809 | cend--; |
| 810 | } |
| 811 | } |
| 812 | else if (SCM_CHARSETP (char_pred)) |
| 813 | { |
| 814 | while (cstart < cend) |
| 815 | { |
| 816 | if (!REF_IN_CHARSET (s, cend-1, char_pred)) |
| 817 | break; |
| 818 | cend--; |
| 819 | } |
| 820 | } |
| 821 | else |
| 822 | { |
| 823 | SCM_ASSERT (scm_is_true (scm_procedure_p (char_pred)), |
| 824 | char_pred, SCM_ARG2, FUNC_NAME); |
| 825 | |
| 826 | while (cstart < cend) |
| 827 | { |
| 828 | SCM res; |
| 829 | |
| 830 | res = scm_call_1 (char_pred, SCM_MAKE_CHAR (scm_i_string_ref (s, cend - 1))); |
| 831 | if (scm_is_false (res)) |
| 832 | break; |
| 833 | cend--; |
| 834 | } |
| 835 | } |
| 836 | return scm_i_substring (s, cstart, cend); |
| 837 | } |
| 838 | #undef FUNC_NAME |
| 839 | |
| 840 | |
| 841 | SCM_DEFINE (scm_string_trim_both, "string-trim-both", 1, 3, 0, |
| 842 | (SCM s, SCM char_pred, SCM start, SCM end), |
| 843 | "Trim @var{s} by skipping over all characters on both sides of\n" |
| 844 | "the string that satisfy the parameter @var{char_pred}:\n" |
| 845 | "\n" |
| 846 | "@itemize @bullet\n" |
| 847 | "@item\n" |
| 848 | "if it is the character @var{ch}, characters equal to @var{ch}\n" |
| 849 | "are trimmed,\n" |
| 850 | "\n" |
| 851 | "@item\n" |
| 852 | "if it is a procedure @var{pred} characters that satisfy\n" |
| 853 | "@var{pred} are trimmed,\n" |
| 854 | "\n" |
| 855 | "@item\n" |
| 856 | "if it is a character set, the characters in the set are\n" |
| 857 | "trimmed.\n" |
| 858 | "@end itemize\n" |
| 859 | "\n" |
| 860 | "If called without a @var{char_pred} argument, all whitespace is\n" |
| 861 | "trimmed.") |
| 862 | #define FUNC_NAME s_scm_string_trim_both |
| 863 | { |
| 864 | size_t cstart, cend; |
| 865 | |
| 866 | MY_VALIDATE_SUBSTRING_SPEC (1, s, |
| 867 | 3, start, cstart, |
| 868 | 4, end, cend); |
| 869 | if (SCM_UNBNDP (char_pred)) |
| 870 | { |
| 871 | while (cstart < cend) |
| 872 | { |
| 873 | if (!uc_is_c_whitespace (scm_i_string_ref (s, cstart))) |
| 874 | break; |
| 875 | cstart++; |
| 876 | } |
| 877 | while (cstart < cend) |
| 878 | { |
| 879 | if (!uc_is_c_whitespace (scm_i_string_ref (s, cend - 1))) |
| 880 | break; |
| 881 | cend--; |
| 882 | } |
| 883 | } |
| 884 | else if (SCM_CHARP (char_pred)) |
| 885 | { |
| 886 | while (cstart < cend) |
| 887 | { |
| 888 | if (scm_i_string_ref (s, cstart) != SCM_CHAR(char_pred)) |
| 889 | break; |
| 890 | cstart++; |
| 891 | } |
| 892 | while (cstart < cend) |
| 893 | { |
| 894 | if (scm_i_string_ref (s, cend - 1) != SCM_CHAR (char_pred)) |
| 895 | break; |
| 896 | cend--; |
| 897 | } |
| 898 | } |
| 899 | else if (SCM_CHARSETP (char_pred)) |
| 900 | { |
| 901 | while (cstart < cend) |
| 902 | { |
| 903 | if (!REF_IN_CHARSET (s, cstart, char_pred)) |
| 904 | break; |
| 905 | cstart++; |
| 906 | } |
| 907 | while (cstart < cend) |
| 908 | { |
| 909 | if (!REF_IN_CHARSET (s, cend-1, char_pred)) |
| 910 | break; |
| 911 | cend--; |
| 912 | } |
| 913 | } |
| 914 | else |
| 915 | { |
| 916 | SCM_ASSERT (scm_is_true (scm_procedure_p (char_pred)), |
| 917 | char_pred, SCM_ARG2, FUNC_NAME); |
| 918 | |
| 919 | while (cstart < cend) |
| 920 | { |
| 921 | SCM res; |
| 922 | |
| 923 | res = scm_call_1 (char_pred, SCM_MAKE_CHAR (scm_i_string_ref (s, cstart))); |
| 924 | if (scm_is_false (res)) |
| 925 | break; |
| 926 | cstart++; |
| 927 | } |
| 928 | while (cstart < cend) |
| 929 | { |
| 930 | SCM res; |
| 931 | |
| 932 | res = scm_call_1 (char_pred, SCM_MAKE_CHAR (scm_i_string_ref (s, cend - 1))); |
| 933 | if (scm_is_false (res)) |
| 934 | break; |
| 935 | cend--; |
| 936 | } |
| 937 | } |
| 938 | return scm_i_substring (s, cstart, cend); |
| 939 | } |
| 940 | #undef FUNC_NAME |
| 941 | |
| 942 | |
| 943 | SCM_DEFINE (scm_substring_fill_x, "string-fill!", 2, 2, 0, |
| 944 | (SCM str, SCM chr, SCM start, SCM end), |
| 945 | "Stores @var{chr} in every element of the given @var{str} and\n" |
| 946 | "returns an unspecified value.") |
| 947 | #define FUNC_NAME s_scm_substring_fill_x |
| 948 | { |
| 949 | size_t cstart, cend; |
| 950 | size_t k; |
| 951 | |
| 952 | /* Older versions of Guile provided the function |
| 953 | scm_substring_fill_x with the following order of arguments: |
| 954 | |
| 955 | str, start, end, chr |
| 956 | |
| 957 | We accomodate this here by detecting such a usage and reordering |
| 958 | the arguments. |
| 959 | */ |
| 960 | if (SCM_CHARP (end)) |
| 961 | { |
| 962 | SCM tmp = end; |
| 963 | end = start; |
| 964 | start = chr; |
| 965 | chr = tmp; |
| 966 | } |
| 967 | |
| 968 | MY_VALIDATE_SUBSTRING_SPEC (1, str, |
| 969 | 3, start, cstart, |
| 970 | 4, end, cend); |
| 971 | SCM_VALIDATE_CHAR (2, chr); |
| 972 | |
| 973 | |
| 974 | str = scm_i_string_start_writing (str); |
| 975 | for (k = cstart; k < cend; k++) |
| 976 | scm_i_string_set_x (str, k, SCM_CHAR (chr)); |
| 977 | scm_i_string_stop_writing (); |
| 978 | |
| 979 | return SCM_UNSPECIFIED; |
| 980 | } |
| 981 | #undef FUNC_NAME |
| 982 | |
| 983 | SCM |
| 984 | scm_string_fill_x (SCM str, SCM chr) |
| 985 | { |
| 986 | return scm_substring_fill_x (str, chr, SCM_UNDEFINED, SCM_UNDEFINED); |
| 987 | } |
| 988 | |
| 989 | SCM_DEFINE (scm_string_compare, "string-compare", 5, 4, 0, |
| 990 | (SCM s1, SCM s2, SCM proc_lt, SCM proc_eq, SCM proc_gt, SCM start1, SCM end1, SCM start2, SCM end2), |
| 991 | "Apply @var{proc_lt}, @var{proc_eq}, @var{proc_gt} to the\n" |
| 992 | "mismatch index, depending upon whether @var{s1} is less than,\n" |
| 993 | "equal to, or greater than @var{s2}. The mismatch index is the\n" |
| 994 | "largest index @var{i} such that for every 0 <= @var{j} <\n" |
| 995 | "@var{i}, @var{s1}[@var{j}] = @var{s2}[@var{j}] -- that is,\n" |
| 996 | "@var{i} is the first position that does not match.") |
| 997 | #define FUNC_NAME s_scm_string_compare |
| 998 | { |
| 999 | size_t cstart1, cend1, cstart2, cend2; |
| 1000 | SCM proc; |
| 1001 | |
| 1002 | MY_VALIDATE_SUBSTRING_SPEC (1, s1, |
| 1003 | 6, start1, cstart1, |
| 1004 | 7, end1, cend1); |
| 1005 | MY_VALIDATE_SUBSTRING_SPEC (2, s2, |
| 1006 | 8, start2, cstart2, |
| 1007 | 9, end2, cend2); |
| 1008 | SCM_VALIDATE_PROC (3, proc_lt); |
| 1009 | SCM_VALIDATE_PROC (4, proc_eq); |
| 1010 | SCM_VALIDATE_PROC (5, proc_gt); |
| 1011 | |
| 1012 | while (cstart1 < cend1 && cstart2 < cend2) |
| 1013 | { |
| 1014 | if (scm_i_string_ref (s1, cstart1) |
| 1015 | < scm_i_string_ref (s2, cstart2)) |
| 1016 | { |
| 1017 | proc = proc_lt; |
| 1018 | goto ret; |
| 1019 | } |
| 1020 | else if (scm_i_string_ref (s1, cstart1) |
| 1021 | > scm_i_string_ref (s2, cstart2)) |
| 1022 | { |
| 1023 | proc = proc_gt; |
| 1024 | goto ret; |
| 1025 | } |
| 1026 | cstart1++; |
| 1027 | cstart2++; |
| 1028 | } |
| 1029 | if (cstart1 < cend1) |
| 1030 | proc = proc_gt; |
| 1031 | else if (cstart2 < cend2) |
| 1032 | proc = proc_lt; |
| 1033 | else |
| 1034 | proc = proc_eq; |
| 1035 | |
| 1036 | ret: |
| 1037 | scm_remember_upto_here_2 (s1, s2); |
| 1038 | return scm_call_1 (proc, scm_from_size_t (cstart1)); |
| 1039 | } |
| 1040 | #undef FUNC_NAME |
| 1041 | |
| 1042 | |
| 1043 | SCM_DEFINE (scm_string_compare_ci, "string-compare-ci", 5, 4, 0, |
| 1044 | (SCM s1, SCM s2, SCM proc_lt, SCM proc_eq, SCM proc_gt, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1045 | "Apply @var{proc_lt}, @var{proc_eq}, @var{proc_gt} to the\n" |
| 1046 | "mismatch index, depending upon whether @var{s1} is less than,\n" |
| 1047 | "equal to, or greater than @var{s2}. The mismatch index is the\n" |
| 1048 | "largest index @var{i} such that for every 0 <= @var{j} <\n" |
| 1049 | "@var{i}, @var{s1}[@var{j}] = @var{s2}[@var{j}] -- that is,\n" |
| 1050 | "@var{i} is the first position where the lowercased letters \n" |
| 1051 | "do not match.\n") |
| 1052 | #define FUNC_NAME s_scm_string_compare_ci |
| 1053 | { |
| 1054 | size_t cstart1, cend1, cstart2, cend2; |
| 1055 | SCM proc; |
| 1056 | |
| 1057 | MY_VALIDATE_SUBSTRING_SPEC (1, s1, |
| 1058 | 6, start1, cstart1, |
| 1059 | 7, end1, cend1); |
| 1060 | MY_VALIDATE_SUBSTRING_SPEC (2, s2, |
| 1061 | 8, start2, cstart2, |
| 1062 | 9, end2, cend2); |
| 1063 | SCM_VALIDATE_PROC (3, proc_lt); |
| 1064 | SCM_VALIDATE_PROC (4, proc_eq); |
| 1065 | SCM_VALIDATE_PROC (5, proc_gt); |
| 1066 | |
| 1067 | while (cstart1 < cend1 && cstart2 < cend2) |
| 1068 | { |
| 1069 | if (uc_tolower (uc_toupper (scm_i_string_ref (s1, cstart1))) |
| 1070 | < uc_tolower (uc_toupper (scm_i_string_ref (s2, cstart2)))) |
| 1071 | { |
| 1072 | proc = proc_lt; |
| 1073 | goto ret; |
| 1074 | } |
| 1075 | else if (uc_tolower (uc_toupper (scm_i_string_ref (s1, cstart1))) |
| 1076 | > uc_tolower (uc_toupper (scm_i_string_ref (s2, cstart2)))) |
| 1077 | { |
| 1078 | proc = proc_gt; |
| 1079 | goto ret; |
| 1080 | } |
| 1081 | cstart1++; |
| 1082 | cstart2++; |
| 1083 | } |
| 1084 | |
| 1085 | if (cstart1 < cend1) |
| 1086 | proc = proc_gt; |
| 1087 | else if (cstart2 < cend2) |
| 1088 | proc = proc_lt; |
| 1089 | else |
| 1090 | proc = proc_eq; |
| 1091 | |
| 1092 | ret: |
| 1093 | scm_remember_upto_here (s1, s2); |
| 1094 | return scm_call_1 (proc, scm_from_size_t (cstart1)); |
| 1095 | } |
| 1096 | #undef FUNC_NAME |
| 1097 | |
| 1098 | /* This function compares two substrings, S1 from START1 to END1 and |
| 1099 | S2 from START2 to END2, possibly case insensitively, and returns |
| 1100 | one of the parameters LESSTHAN, GREATERTHAN, SHORTER, LONGER, or |
| 1101 | EQUAL depending if S1 is less than S2, greater than S2, shorter, |
| 1102 | longer, or equal. */ |
| 1103 | static SCM |
| 1104 | compare_strings (const char *fname, int case_insensitive, |
| 1105 | SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2, |
| 1106 | SCM lessthan, SCM greaterthan, SCM shorter, SCM longer, SCM equal) |
| 1107 | { |
| 1108 | size_t cstart1, cend1, cstart2, cend2; |
| 1109 | SCM ret; |
| 1110 | scm_t_wchar a, b; |
| 1111 | |
| 1112 | MY_SUBF_VALIDATE_SUBSTRING_SPEC (fname, 1, s1, |
| 1113 | 3, start1, cstart1, |
| 1114 | 4, end1, cend1); |
| 1115 | MY_SUBF_VALIDATE_SUBSTRING_SPEC (fname, 2, s2, |
| 1116 | 5, start2, cstart2, |
| 1117 | 6, end2, cend2); |
| 1118 | |
| 1119 | while (cstart1 < cend1 && cstart2 < cend2) |
| 1120 | { |
| 1121 | if (case_insensitive) |
| 1122 | { |
| 1123 | a = uc_tolower (uc_toupper (scm_i_string_ref (s1, cstart1))); |
| 1124 | b = uc_tolower (uc_toupper (scm_i_string_ref (s2, cstart2))); |
| 1125 | } |
| 1126 | else |
| 1127 | { |
| 1128 | a = scm_i_string_ref (s1, cstart1); |
| 1129 | b = scm_i_string_ref (s2, cstart2); |
| 1130 | } |
| 1131 | if (a < b) |
| 1132 | { |
| 1133 | ret = lessthan; |
| 1134 | goto done; |
| 1135 | } |
| 1136 | else if (a > b) |
| 1137 | { |
| 1138 | ret = greaterthan; |
| 1139 | goto done; |
| 1140 | } |
| 1141 | cstart1++; |
| 1142 | cstart2++; |
| 1143 | } |
| 1144 | if (cstart1 < cend1) |
| 1145 | { |
| 1146 | ret = longer; |
| 1147 | goto done; |
| 1148 | } |
| 1149 | else if (cstart2 < cend2) |
| 1150 | { |
| 1151 | ret = shorter; |
| 1152 | goto done; |
| 1153 | } |
| 1154 | else |
| 1155 | { |
| 1156 | ret = equal; |
| 1157 | goto done; |
| 1158 | } |
| 1159 | |
| 1160 | done: |
| 1161 | scm_remember_upto_here_2 (s1, s2); |
| 1162 | return ret; |
| 1163 | } |
| 1164 | |
| 1165 | |
| 1166 | SCM_DEFINE (scm_string_eq, "string=", 2, 4, 0, |
| 1167 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1168 | "Return @code{#f} if @var{s1} and @var{s2} are not equal, a true\n" |
| 1169 | "value otherwise.") |
| 1170 | #define FUNC_NAME s_scm_string_eq |
| 1171 | { |
| 1172 | if (SCM_LIKELY (scm_is_string (s1) && scm_is_string (s2) && |
| 1173 | scm_i_is_narrow_string (s1) == scm_i_is_narrow_string (s2) |
| 1174 | && SCM_UNBNDP (start1) && SCM_UNBNDP (end1) |
| 1175 | && SCM_UNBNDP (start2) && SCM_UNBNDP (end2))) |
| 1176 | { |
| 1177 | /* Fast path for this common case, which avoids the repeated calls to |
| 1178 | `scm_i_string_ref'. */ |
| 1179 | size_t len1, len2; |
| 1180 | |
| 1181 | len1 = scm_i_string_length (s1); |
| 1182 | len2 = scm_i_string_length (s2); |
| 1183 | |
| 1184 | if (SCM_LIKELY (len1 == len2)) |
| 1185 | { |
| 1186 | if (!scm_i_is_narrow_string (s1)) |
| 1187 | len1 *= 4; |
| 1188 | |
| 1189 | return scm_from_bool (memcmp (scm_i_string_data (s1), |
| 1190 | scm_i_string_data (s2), |
| 1191 | len1) == 0); |
| 1192 | } |
| 1193 | } |
| 1194 | |
| 1195 | return compare_strings (FUNC_NAME, 0, |
| 1196 | s1, s2, start1, end1, start2, end2, |
| 1197 | SCM_BOOL_F, SCM_BOOL_F, SCM_BOOL_F, SCM_BOOL_F, SCM_BOOL_T); |
| 1198 | } |
| 1199 | #undef FUNC_NAME |
| 1200 | |
| 1201 | |
| 1202 | SCM_DEFINE (scm_string_neq, "string<>", 2, 4, 0, |
| 1203 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1204 | "Return @code{#f} if @var{s1} and @var{s2} are equal, a true\n" |
| 1205 | "value otherwise.") |
| 1206 | #define FUNC_NAME s_scm_string_neq |
| 1207 | { |
| 1208 | return compare_strings (FUNC_NAME, 0, |
| 1209 | s1, s2, start1, end1, start2, end2, |
| 1210 | SCM_BOOL_T, SCM_BOOL_T, SCM_BOOL_T, SCM_BOOL_T, SCM_BOOL_F); |
| 1211 | } |
| 1212 | #undef FUNC_NAME |
| 1213 | |
| 1214 | |
| 1215 | SCM_DEFINE (scm_string_lt, "string<", 2, 4, 0, |
| 1216 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1217 | "Return @code{#f} if @var{s1} is greater or equal to @var{s2}, a\n" |
| 1218 | "true value otherwise.") |
| 1219 | #define FUNC_NAME s_scm_string_lt |
| 1220 | { |
| 1221 | return compare_strings (FUNC_NAME, 0, |
| 1222 | s1, s2, start1, end1, start2, end2, |
| 1223 | SCM_BOOL_T, SCM_BOOL_F, SCM_BOOL_T, SCM_BOOL_F, SCM_BOOL_F); |
| 1224 | } |
| 1225 | #undef FUNC_NAME |
| 1226 | |
| 1227 | |
| 1228 | SCM_DEFINE (scm_string_gt, "string>", 2, 4, 0, |
| 1229 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1230 | "Return @code{#f} if @var{s1} is less or equal to @var{s2}, a\n" |
| 1231 | "true value otherwise.") |
| 1232 | #define FUNC_NAME s_scm_string_gt |
| 1233 | { |
| 1234 | return compare_strings (FUNC_NAME, 0, |
| 1235 | s1, s2, start1, end1, start2, end2, |
| 1236 | SCM_BOOL_F, SCM_BOOL_T, SCM_BOOL_F, SCM_BOOL_T, SCM_BOOL_F); |
| 1237 | } |
| 1238 | #undef FUNC_NAME |
| 1239 | |
| 1240 | |
| 1241 | SCM_DEFINE (scm_string_le, "string<=", 2, 4, 0, |
| 1242 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1243 | "Return @code{#f} if @var{s1} is greater to @var{s2}, a true\n" |
| 1244 | "value otherwise.") |
| 1245 | #define FUNC_NAME s_scm_string_le |
| 1246 | { |
| 1247 | return compare_strings (FUNC_NAME, 0, |
| 1248 | s1, s2, start1, end1, start2, end2, |
| 1249 | SCM_BOOL_T, SCM_BOOL_F, SCM_BOOL_T, SCM_BOOL_F, SCM_BOOL_T); |
| 1250 | } |
| 1251 | #undef FUNC_NAME |
| 1252 | |
| 1253 | |
| 1254 | SCM_DEFINE (scm_string_ge, "string>=", 2, 4, 0, |
| 1255 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1256 | "Return @code{#f} if @var{s1} is less to @var{s2}, a true value\n" |
| 1257 | "otherwise.") |
| 1258 | #define FUNC_NAME s_scm_string_ge |
| 1259 | { |
| 1260 | return compare_strings (FUNC_NAME, 0, |
| 1261 | s1, s2, start1, end1, start2, end2, |
| 1262 | SCM_BOOL_F, SCM_BOOL_T, SCM_BOOL_F, SCM_BOOL_T, SCM_BOOL_T); |
| 1263 | } |
| 1264 | #undef FUNC_NAME |
| 1265 | |
| 1266 | |
| 1267 | SCM_DEFINE (scm_string_ci_eq, "string-ci=", 2, 4, 0, |
| 1268 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1269 | "Return @code{#f} if @var{s1} and @var{s2} are not equal, a true\n" |
| 1270 | "value otherwise. The character comparison is done\n" |
| 1271 | "case-insensitively.") |
| 1272 | #define FUNC_NAME s_scm_string_ci_eq |
| 1273 | { |
| 1274 | return compare_strings (FUNC_NAME, 1, |
| 1275 | s1, s2, start1, end1, start2, end2, |
| 1276 | SCM_BOOL_F, SCM_BOOL_F, SCM_BOOL_F, SCM_BOOL_F, SCM_BOOL_T); |
| 1277 | } |
| 1278 | #undef FUNC_NAME |
| 1279 | |
| 1280 | |
| 1281 | SCM_DEFINE (scm_string_ci_neq, "string-ci<>", 2, 4, 0, |
| 1282 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1283 | "Return @code{#f} if @var{s1} and @var{s2} are equal, a true\n" |
| 1284 | "value otherwise. The character comparison is done\n" |
| 1285 | "case-insensitively.") |
| 1286 | #define FUNC_NAME s_scm_string_ci_neq |
| 1287 | { |
| 1288 | return compare_strings (FUNC_NAME, 1, |
| 1289 | s1, s2, start1, end1, start2, end2, |
| 1290 | SCM_BOOL_T, SCM_BOOL_T, SCM_BOOL_T, SCM_BOOL_T, SCM_BOOL_F); |
| 1291 | } |
| 1292 | #undef FUNC_NAME |
| 1293 | |
| 1294 | |
| 1295 | SCM_DEFINE (scm_string_ci_lt, "string-ci<", 2, 4, 0, |
| 1296 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1297 | "Return @code{#f} if @var{s1} is greater or equal to @var{s2}, a\n" |
| 1298 | "true value otherwise. The character comparison is done\n" |
| 1299 | "case-insensitively.") |
| 1300 | #define FUNC_NAME s_scm_string_ci_lt |
| 1301 | { |
| 1302 | return compare_strings (FUNC_NAME, 1, |
| 1303 | s1, s2, start1, end1, start2, end2, |
| 1304 | SCM_BOOL_T, SCM_BOOL_F, SCM_BOOL_T, SCM_BOOL_F, SCM_BOOL_F); |
| 1305 | } |
| 1306 | #undef FUNC_NAME |
| 1307 | |
| 1308 | |
| 1309 | SCM_DEFINE (scm_string_ci_gt, "string-ci>", 2, 4, 0, |
| 1310 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1311 | "Return @code{#f} if @var{s1} is less or equal to @var{s2}, a\n" |
| 1312 | "true value otherwise. The character comparison is done\n" |
| 1313 | "case-insensitively.") |
| 1314 | #define FUNC_NAME s_scm_string_ci_gt |
| 1315 | { |
| 1316 | return compare_strings (FUNC_NAME, 1, |
| 1317 | s1, s2, start1, end1, start2, end2, |
| 1318 | SCM_BOOL_F, SCM_BOOL_T, SCM_BOOL_F, SCM_BOOL_T, SCM_BOOL_F); |
| 1319 | } |
| 1320 | #undef FUNC_NAME |
| 1321 | |
| 1322 | |
| 1323 | SCM_DEFINE (scm_string_ci_le, "string-ci<=", 2, 4, 0, |
| 1324 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1325 | "Return @code{#f} if @var{s1} is greater to @var{s2}, a true\n" |
| 1326 | "value otherwise. The character comparison is done\n" |
| 1327 | "case-insensitively.") |
| 1328 | #define FUNC_NAME s_scm_string_ci_le |
| 1329 | { |
| 1330 | return compare_strings (FUNC_NAME, 1, |
| 1331 | s1, s2, start1, end1, start2, end2, |
| 1332 | SCM_BOOL_T, SCM_BOOL_F, SCM_BOOL_T, SCM_BOOL_F, SCM_BOOL_T); |
| 1333 | } |
| 1334 | #undef FUNC_NAME |
| 1335 | |
| 1336 | |
| 1337 | SCM_DEFINE (scm_string_ci_ge, "string-ci>=", 2, 4, 0, |
| 1338 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1339 | "Return @code{#f} if @var{s1} is less to @var{s2}, a true value\n" |
| 1340 | "otherwise. The character comparison is done\n" |
| 1341 | "case-insensitively.") |
| 1342 | #define FUNC_NAME s_scm_string_ci_ge |
| 1343 | { |
| 1344 | return compare_strings (FUNC_NAME, 1, |
| 1345 | s1, s2, start1, end1, start2, end2, |
| 1346 | SCM_BOOL_F, SCM_BOOL_T, SCM_BOOL_F, SCM_BOOL_T, SCM_BOOL_T); |
| 1347 | } |
| 1348 | #undef FUNC_NAME |
| 1349 | |
| 1350 | SCM_DEFINE (scm_substring_hash, "string-hash", 1, 3, 0, |
| 1351 | (SCM s, SCM bound, SCM start, SCM end), |
| 1352 | "Compute a hash value for @var{S}. the optional argument " |
| 1353 | "@var{bound} is a non-negative exact " |
| 1354 | "integer specifying the range of the hash function. " |
| 1355 | "A positive value restricts the return value to the " |
| 1356 | "range [0,bound).") |
| 1357 | #define FUNC_NAME s_scm_substring_hash |
| 1358 | { |
| 1359 | if (SCM_UNBNDP (bound)) |
| 1360 | bound = scm_from_intmax (SCM_MOST_POSITIVE_FIXNUM); |
| 1361 | if (SCM_UNBNDP (start)) |
| 1362 | start = SCM_INUM0; |
| 1363 | return scm_hash (scm_substring_shared (s, start, end), bound); |
| 1364 | } |
| 1365 | #undef FUNC_NAME |
| 1366 | |
| 1367 | SCM_DEFINE (scm_substring_hash_ci, "string-hash-ci", 1, 3, 0, |
| 1368 | (SCM s, SCM bound, SCM start, SCM end), |
| 1369 | "Compute a hash value for @var{S}. the optional argument " |
| 1370 | "@var{bound} is a non-negative exact " |
| 1371 | "integer specifying the range of the hash function. " |
| 1372 | "A positive value restricts the return value to the " |
| 1373 | "range [0,bound).") |
| 1374 | #define FUNC_NAME s_scm_substring_hash_ci |
| 1375 | { |
| 1376 | return scm_substring_hash (scm_substring_downcase (s, start, end), |
| 1377 | bound, |
| 1378 | SCM_UNDEFINED, SCM_UNDEFINED); |
| 1379 | } |
| 1380 | #undef FUNC_NAME |
| 1381 | |
| 1382 | SCM_DEFINE (scm_string_prefix_length, "string-prefix-length", 2, 4, 0, |
| 1383 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1384 | "Return the length of the longest common prefix of the two\n" |
| 1385 | "strings.") |
| 1386 | #define FUNC_NAME s_scm_string_prefix_length |
| 1387 | { |
| 1388 | size_t cstart1, cend1, cstart2, cend2; |
| 1389 | size_t len = 0; |
| 1390 | |
| 1391 | MY_VALIDATE_SUBSTRING_SPEC (1, s1, |
| 1392 | 3, start1, cstart1, |
| 1393 | 4, end1, cend1); |
| 1394 | MY_VALIDATE_SUBSTRING_SPEC (2, s2, |
| 1395 | 5, start2, cstart2, |
| 1396 | 6, end2, cend2); |
| 1397 | |
| 1398 | while (cstart1 < cend1 && cstart2 < cend2) |
| 1399 | { |
| 1400 | if (scm_i_string_ref (s1, cstart1) |
| 1401 | != scm_i_string_ref (s2, cstart2)) |
| 1402 | goto ret; |
| 1403 | len++; |
| 1404 | cstart1++; |
| 1405 | cstart2++; |
| 1406 | } |
| 1407 | |
| 1408 | ret: |
| 1409 | scm_remember_upto_here_2 (s1, s2); |
| 1410 | return scm_from_size_t (len); |
| 1411 | } |
| 1412 | #undef FUNC_NAME |
| 1413 | |
| 1414 | |
| 1415 | SCM_DEFINE (scm_string_prefix_length_ci, "string-prefix-length-ci", 2, 4, 0, |
| 1416 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1417 | "Return the length of the longest common prefix of the two\n" |
| 1418 | "strings, ignoring character case.") |
| 1419 | #define FUNC_NAME s_scm_string_prefix_length_ci |
| 1420 | { |
| 1421 | size_t cstart1, cend1, cstart2, cend2; |
| 1422 | size_t len = 0; |
| 1423 | |
| 1424 | MY_VALIDATE_SUBSTRING_SPEC (1, s1, |
| 1425 | 3, start1, cstart1, |
| 1426 | 4, end1, cend1); |
| 1427 | MY_VALIDATE_SUBSTRING_SPEC (2, s2, |
| 1428 | 5, start2, cstart2, |
| 1429 | 6, end2, cend2); |
| 1430 | while (cstart1 < cend1 && cstart2 < cend2) |
| 1431 | { |
| 1432 | if (uc_tolower (uc_toupper (scm_i_string_ref (s1, cstart1))) |
| 1433 | != uc_tolower (uc_toupper (scm_i_string_ref (s2, cstart2)))) |
| 1434 | goto ret; |
| 1435 | len++; |
| 1436 | cstart1++; |
| 1437 | cstart2++; |
| 1438 | } |
| 1439 | |
| 1440 | ret: |
| 1441 | scm_remember_upto_here_2 (s1, s2); |
| 1442 | return scm_from_size_t (len); |
| 1443 | } |
| 1444 | #undef FUNC_NAME |
| 1445 | |
| 1446 | |
| 1447 | SCM_DEFINE (scm_string_suffix_length, "string-suffix-length", 2, 4, 0, |
| 1448 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1449 | "Return the length of the longest common suffix of the two\n" |
| 1450 | "strings.") |
| 1451 | #define FUNC_NAME s_scm_string_suffix_length |
| 1452 | { |
| 1453 | size_t cstart1, cend1, cstart2, cend2; |
| 1454 | size_t len = 0; |
| 1455 | |
| 1456 | MY_VALIDATE_SUBSTRING_SPEC (1, s1, |
| 1457 | 3, start1, cstart1, |
| 1458 | 4, end1, cend1); |
| 1459 | MY_VALIDATE_SUBSTRING_SPEC (2, s2, |
| 1460 | 5, start2, cstart2, |
| 1461 | 6, end2, cend2); |
| 1462 | while (cstart1 < cend1 && cstart2 < cend2) |
| 1463 | { |
| 1464 | cend1--; |
| 1465 | cend2--; |
| 1466 | if (scm_i_string_ref (s1, cend1) |
| 1467 | != scm_i_string_ref (s2, cend2)) |
| 1468 | goto ret; |
| 1469 | len++; |
| 1470 | } |
| 1471 | |
| 1472 | ret: |
| 1473 | scm_remember_upto_here_2 (s1, s2); |
| 1474 | return scm_from_size_t (len); |
| 1475 | } |
| 1476 | #undef FUNC_NAME |
| 1477 | |
| 1478 | |
| 1479 | SCM_DEFINE (scm_string_suffix_length_ci, "string-suffix-length-ci", 2, 4, 0, |
| 1480 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1481 | "Return the length of the longest common suffix of the two\n" |
| 1482 | "strings, ignoring character case.") |
| 1483 | #define FUNC_NAME s_scm_string_suffix_length_ci |
| 1484 | { |
| 1485 | size_t cstart1, cend1, cstart2, cend2; |
| 1486 | size_t len = 0; |
| 1487 | |
| 1488 | MY_VALIDATE_SUBSTRING_SPEC (1, s1, |
| 1489 | 3, start1, cstart1, |
| 1490 | 4, end1, cend1); |
| 1491 | MY_VALIDATE_SUBSTRING_SPEC (2, s2, |
| 1492 | 5, start2, cstart2, |
| 1493 | 6, end2, cend2); |
| 1494 | while (cstart1 < cend1 && cstart2 < cend2) |
| 1495 | { |
| 1496 | cend1--; |
| 1497 | cend2--; |
| 1498 | if (uc_tolower (uc_toupper (scm_i_string_ref (s1, cend1))) |
| 1499 | != uc_tolower (uc_toupper (scm_i_string_ref (s2, cend2)))) |
| 1500 | goto ret; |
| 1501 | len++; |
| 1502 | } |
| 1503 | |
| 1504 | ret: |
| 1505 | scm_remember_upto_here_2 (s1, s2); |
| 1506 | return scm_from_size_t (len); |
| 1507 | } |
| 1508 | #undef FUNC_NAME |
| 1509 | |
| 1510 | |
| 1511 | SCM_DEFINE (scm_string_prefix_p, "string-prefix?", 2, 4, 0, |
| 1512 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1513 | "Is @var{s1} a prefix of @var{s2}?") |
| 1514 | #define FUNC_NAME s_scm_string_prefix_p |
| 1515 | { |
| 1516 | size_t cstart1, cend1, cstart2, cend2; |
| 1517 | size_t len = 0, len1; |
| 1518 | |
| 1519 | MY_VALIDATE_SUBSTRING_SPEC (1, s1, |
| 1520 | 3, start1, cstart1, |
| 1521 | 4, end1, cend1); |
| 1522 | MY_VALIDATE_SUBSTRING_SPEC (2, s2, |
| 1523 | 5, start2, cstart2, |
| 1524 | 6, end2, cend2); |
| 1525 | len1 = cend1 - cstart1; |
| 1526 | while (cstart1 < cend1 && cstart2 < cend2) |
| 1527 | { |
| 1528 | if (scm_i_string_ref (s1, cstart1) |
| 1529 | != scm_i_string_ref (s2, cstart2)) |
| 1530 | goto ret; |
| 1531 | len++; |
| 1532 | cstart1++; |
| 1533 | cstart2++; |
| 1534 | } |
| 1535 | |
| 1536 | ret: |
| 1537 | scm_remember_upto_here_2 (s1, s2); |
| 1538 | return scm_from_bool (len == len1); |
| 1539 | } |
| 1540 | #undef FUNC_NAME |
| 1541 | |
| 1542 | |
| 1543 | SCM_DEFINE (scm_string_prefix_ci_p, "string-prefix-ci?", 2, 4, 0, |
| 1544 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1545 | "Is @var{s1} a prefix of @var{s2}, ignoring character case?") |
| 1546 | #define FUNC_NAME s_scm_string_prefix_ci_p |
| 1547 | { |
| 1548 | size_t cstart1, cend1, cstart2, cend2; |
| 1549 | size_t len = 0, len1; |
| 1550 | |
| 1551 | MY_VALIDATE_SUBSTRING_SPEC (1, s1, |
| 1552 | 3, start1, cstart1, |
| 1553 | 4, end1, cend1); |
| 1554 | MY_VALIDATE_SUBSTRING_SPEC (2, s2, |
| 1555 | 5, start2, cstart2, |
| 1556 | 6, end2, cend2); |
| 1557 | len1 = cend1 - cstart1; |
| 1558 | while (cstart1 < cend1 && cstart2 < cend2) |
| 1559 | { |
| 1560 | scm_t_wchar a = uc_tolower (uc_toupper (scm_i_string_ref (s1, cstart1))); |
| 1561 | scm_t_wchar b = uc_tolower (uc_toupper (scm_i_string_ref (s2, cstart2))); |
| 1562 | if (a != b) |
| 1563 | goto ret; |
| 1564 | len++; |
| 1565 | cstart1++; |
| 1566 | cstart2++; |
| 1567 | } |
| 1568 | |
| 1569 | ret: |
| 1570 | scm_remember_upto_here_2 (s1, s2); |
| 1571 | return scm_from_bool (len == len1); |
| 1572 | } |
| 1573 | #undef FUNC_NAME |
| 1574 | |
| 1575 | |
| 1576 | SCM_DEFINE (scm_string_suffix_p, "string-suffix?", 2, 4, 0, |
| 1577 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1578 | "Is @var{s1} a suffix of @var{s2}?") |
| 1579 | #define FUNC_NAME s_scm_string_suffix_p |
| 1580 | { |
| 1581 | size_t cstart1, cend1, cstart2, cend2; |
| 1582 | size_t len = 0, len1; |
| 1583 | |
| 1584 | MY_VALIDATE_SUBSTRING_SPEC (1, s1, |
| 1585 | 3, start1, cstart1, |
| 1586 | 4, end1, cend1); |
| 1587 | MY_VALIDATE_SUBSTRING_SPEC (2, s2, |
| 1588 | 5, start2, cstart2, |
| 1589 | 6, end2, cend2); |
| 1590 | len1 = cend1 - cstart1; |
| 1591 | while (cstart1 < cend1 && cstart2 < cend2) |
| 1592 | { |
| 1593 | cend1--; |
| 1594 | cend2--; |
| 1595 | if (scm_i_string_ref (s1, cend1) |
| 1596 | != scm_i_string_ref (s2, cend2)) |
| 1597 | goto ret; |
| 1598 | len++; |
| 1599 | } |
| 1600 | |
| 1601 | ret: |
| 1602 | scm_remember_upto_here_2 (s1, s2); |
| 1603 | return scm_from_bool (len == len1); |
| 1604 | } |
| 1605 | #undef FUNC_NAME |
| 1606 | |
| 1607 | |
| 1608 | SCM_DEFINE (scm_string_suffix_ci_p, "string-suffix-ci?", 2, 4, 0, |
| 1609 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1610 | "Is @var{s1} a suffix of @var{s2}, ignoring character case?") |
| 1611 | #define FUNC_NAME s_scm_string_suffix_ci_p |
| 1612 | { |
| 1613 | size_t cstart1, cend1, cstart2, cend2; |
| 1614 | size_t len = 0, len1; |
| 1615 | |
| 1616 | MY_VALIDATE_SUBSTRING_SPEC (1, s1, |
| 1617 | 3, start1, cstart1, |
| 1618 | 4, end1, cend1); |
| 1619 | MY_VALIDATE_SUBSTRING_SPEC (2, s2, |
| 1620 | 5, start2, cstart2, |
| 1621 | 6, end2, cend2); |
| 1622 | len1 = cend1 - cstart1; |
| 1623 | while (cstart1 < cend1 && cstart2 < cend2) |
| 1624 | { |
| 1625 | cend1--; |
| 1626 | cend2--; |
| 1627 | if (uc_tolower (uc_toupper (scm_i_string_ref (s1, cend1))) |
| 1628 | != uc_tolower (uc_toupper (scm_i_string_ref (s2, cend2)))) |
| 1629 | goto ret; |
| 1630 | len++; |
| 1631 | } |
| 1632 | |
| 1633 | ret: |
| 1634 | scm_remember_upto_here_2 (s1, s2); |
| 1635 | return scm_from_bool (len == len1); |
| 1636 | } |
| 1637 | #undef FUNC_NAME |
| 1638 | |
| 1639 | |
| 1640 | SCM_DEFINE (scm_string_index, "string-index", 2, 2, 0, |
| 1641 | (SCM s, SCM char_pred, SCM start, SCM end), |
| 1642 | "Search through the string @var{s} from left to right, returning\n" |
| 1643 | "the index of the first occurence of a character which\n" |
| 1644 | "\n" |
| 1645 | "@itemize @bullet\n" |
| 1646 | "@item\n" |
| 1647 | "equals @var{char_pred}, if it is character,\n" |
| 1648 | "\n" |
| 1649 | "@item\n" |
| 1650 | "satisifies the predicate @var{char_pred}, if it is a procedure,\n" |
| 1651 | "\n" |
| 1652 | "@item\n" |
| 1653 | "is in the set @var{char_pred}, if it is a character set.\n" |
| 1654 | "@end itemize\n\n" |
| 1655 | "Return @code{#f} if no match is found.") |
| 1656 | #define FUNC_NAME s_scm_string_index |
| 1657 | { |
| 1658 | size_t cstart, cend; |
| 1659 | |
| 1660 | MY_VALIDATE_SUBSTRING_SPEC (1, s, |
| 1661 | 3, start, cstart, |
| 1662 | 4, end, cend); |
| 1663 | if (SCM_CHARP (char_pred)) |
| 1664 | { |
| 1665 | while (cstart < cend) |
| 1666 | { |
| 1667 | if (scm_i_string_ref (s, cstart) == SCM_CHAR (char_pred)) |
| 1668 | goto found; |
| 1669 | cstart++; |
| 1670 | } |
| 1671 | } |
| 1672 | else if (SCM_CHARSETP (char_pred)) |
| 1673 | { |
| 1674 | while (cstart < cend) |
| 1675 | { |
| 1676 | if (REF_IN_CHARSET (s, cstart, char_pred)) |
| 1677 | goto found; |
| 1678 | cstart++; |
| 1679 | } |
| 1680 | } |
| 1681 | else |
| 1682 | { |
| 1683 | SCM_ASSERT (scm_is_true (scm_procedure_p (char_pred)), |
| 1684 | char_pred, SCM_ARG2, FUNC_NAME); |
| 1685 | |
| 1686 | while (cstart < cend) |
| 1687 | { |
| 1688 | SCM res; |
| 1689 | res = scm_call_1 (char_pred, SCM_MAKE_CHAR (scm_i_string_ref (s, cstart))); |
| 1690 | if (scm_is_true (res)) |
| 1691 | goto found; |
| 1692 | cstart++; |
| 1693 | } |
| 1694 | } |
| 1695 | |
| 1696 | scm_remember_upto_here_1 (s); |
| 1697 | return SCM_BOOL_F; |
| 1698 | |
| 1699 | found: |
| 1700 | scm_remember_upto_here_1 (s); |
| 1701 | return scm_from_size_t (cstart); |
| 1702 | } |
| 1703 | #undef FUNC_NAME |
| 1704 | |
| 1705 | SCM_DEFINE (scm_string_index_right, "string-index-right", 2, 2, 0, |
| 1706 | (SCM s, SCM char_pred, SCM start, SCM end), |
| 1707 | "Search through the string @var{s} from right to left, returning\n" |
| 1708 | "the index of the last occurence of a character which\n" |
| 1709 | "\n" |
| 1710 | "@itemize @bullet\n" |
| 1711 | "@item\n" |
| 1712 | "equals @var{char_pred}, if it is character,\n" |
| 1713 | "\n" |
| 1714 | "@item\n" |
| 1715 | "satisifies the predicate @var{char_pred}, if it is a procedure,\n" |
| 1716 | "\n" |
| 1717 | "@item\n" |
| 1718 | "is in the set if @var{char_pred} is a character set.\n" |
| 1719 | "@end itemize\n\n" |
| 1720 | "Return @code{#f} if no match is found.") |
| 1721 | #define FUNC_NAME s_scm_string_index_right |
| 1722 | { |
| 1723 | size_t cstart, cend; |
| 1724 | |
| 1725 | MY_VALIDATE_SUBSTRING_SPEC (1, s, |
| 1726 | 3, start, cstart, |
| 1727 | 4, end, cend); |
| 1728 | if (SCM_CHARP (char_pred)) |
| 1729 | { |
| 1730 | while (cstart < cend) |
| 1731 | { |
| 1732 | cend--; |
| 1733 | if (scm_i_string_ref (s, cend) == SCM_CHAR (char_pred)) |
| 1734 | goto found; |
| 1735 | } |
| 1736 | } |
| 1737 | else if (SCM_CHARSETP (char_pred)) |
| 1738 | { |
| 1739 | while (cstart < cend) |
| 1740 | { |
| 1741 | cend--; |
| 1742 | if (REF_IN_CHARSET (s, cend, char_pred)) |
| 1743 | goto found; |
| 1744 | } |
| 1745 | } |
| 1746 | else |
| 1747 | { |
| 1748 | SCM_ASSERT (scm_is_true (scm_procedure_p (char_pred)), |
| 1749 | char_pred, SCM_ARG2, FUNC_NAME); |
| 1750 | |
| 1751 | while (cstart < cend) |
| 1752 | { |
| 1753 | SCM res; |
| 1754 | cend--; |
| 1755 | res = scm_call_1 (char_pred, SCM_MAKE_CHAR (scm_i_string_ref (s, cend))); |
| 1756 | if (scm_is_true (res)) |
| 1757 | goto found; |
| 1758 | } |
| 1759 | } |
| 1760 | |
| 1761 | scm_remember_upto_here_1 (s); |
| 1762 | return SCM_BOOL_F; |
| 1763 | |
| 1764 | found: |
| 1765 | scm_remember_upto_here_1 (s); |
| 1766 | return scm_from_size_t (cend); |
| 1767 | } |
| 1768 | #undef FUNC_NAME |
| 1769 | |
| 1770 | SCM_DEFINE (scm_string_rindex, "string-rindex", 2, 2, 0, |
| 1771 | (SCM s, SCM char_pred, SCM start, SCM end), |
| 1772 | "Search through the string @var{s} from right to left, returning\n" |
| 1773 | "the index of the last occurence of a character which\n" |
| 1774 | "\n" |
| 1775 | "@itemize @bullet\n" |
| 1776 | "@item\n" |
| 1777 | "equals @var{char_pred}, if it is character,\n" |
| 1778 | "\n" |
| 1779 | "@item\n" |
| 1780 | "satisifies the predicate @var{char_pred}, if it is a procedure,\n" |
| 1781 | "\n" |
| 1782 | "@item\n" |
| 1783 | "is in the set if @var{char_pred} is a character set.\n" |
| 1784 | "@end itemize\n\n" |
| 1785 | "Return @code{#f} if no match is found.") |
| 1786 | #define FUNC_NAME s_scm_string_rindex |
| 1787 | { |
| 1788 | return scm_string_index_right (s, char_pred, start, end); |
| 1789 | } |
| 1790 | #undef FUNC_NAME |
| 1791 | |
| 1792 | SCM_DEFINE (scm_string_skip, "string-skip", 2, 2, 0, |
| 1793 | (SCM s, SCM char_pred, SCM start, SCM end), |
| 1794 | "Search through the string @var{s} from left to right, returning\n" |
| 1795 | "the index of the first occurence of a character which\n" |
| 1796 | "\n" |
| 1797 | "@itemize @bullet\n" |
| 1798 | "@item\n" |
| 1799 | "does not equal @var{char_pred}, if it is character,\n" |
| 1800 | "\n" |
| 1801 | "@item\n" |
| 1802 | "does not satisify the predicate @var{char_pred}, if it is a\n" |
| 1803 | "procedure,\n" |
| 1804 | "\n" |
| 1805 | "@item\n" |
| 1806 | "is not in the set if @var{char_pred} is a character set.\n" |
| 1807 | "@end itemize") |
| 1808 | #define FUNC_NAME s_scm_string_skip |
| 1809 | { |
| 1810 | size_t cstart, cend; |
| 1811 | |
| 1812 | MY_VALIDATE_SUBSTRING_SPEC (1, s, |
| 1813 | 3, start, cstart, |
| 1814 | 4, end, cend); |
| 1815 | if (SCM_CHARP (char_pred)) |
| 1816 | { |
| 1817 | while (cstart < cend) |
| 1818 | { |
| 1819 | if (scm_i_string_ref (s, cstart) != SCM_CHAR (char_pred)) |
| 1820 | goto found; |
| 1821 | cstart++; |
| 1822 | } |
| 1823 | } |
| 1824 | else if (SCM_CHARSETP (char_pred)) |
| 1825 | { |
| 1826 | while (cstart < cend) |
| 1827 | { |
| 1828 | if (!REF_IN_CHARSET (s, cstart, char_pred)) |
| 1829 | goto found; |
| 1830 | cstart++; |
| 1831 | } |
| 1832 | } |
| 1833 | else |
| 1834 | { |
| 1835 | SCM_ASSERT (scm_is_true (scm_procedure_p (char_pred)), |
| 1836 | char_pred, SCM_ARG2, FUNC_NAME); |
| 1837 | |
| 1838 | while (cstart < cend) |
| 1839 | { |
| 1840 | SCM res; |
| 1841 | res = scm_call_1 (char_pred, SCM_MAKE_CHAR (scm_i_string_ref (s, cstart))); |
| 1842 | if (scm_is_false (res)) |
| 1843 | goto found; |
| 1844 | cstart++; |
| 1845 | } |
| 1846 | } |
| 1847 | |
| 1848 | scm_remember_upto_here_1 (s); |
| 1849 | return SCM_BOOL_F; |
| 1850 | |
| 1851 | found: |
| 1852 | scm_remember_upto_here_1 (s); |
| 1853 | return scm_from_size_t (cstart); |
| 1854 | } |
| 1855 | #undef FUNC_NAME |
| 1856 | |
| 1857 | |
| 1858 | SCM_DEFINE (scm_string_skip_right, "string-skip-right", 2, 2, 0, |
| 1859 | (SCM s, SCM char_pred, SCM start, SCM end), |
| 1860 | "Search through the string @var{s} from right to left, returning\n" |
| 1861 | "the index of the last occurence of a character which\n" |
| 1862 | "\n" |
| 1863 | "@itemize @bullet\n" |
| 1864 | "@item\n" |
| 1865 | "does not equal @var{char_pred}, if it is character,\n" |
| 1866 | "\n" |
| 1867 | "@item\n" |
| 1868 | "does not satisfy the predicate @var{char_pred}, if it is a\n" |
| 1869 | "procedure,\n" |
| 1870 | "\n" |
| 1871 | "@item\n" |
| 1872 | "is not in the set if @var{char_pred} is a character set.\n" |
| 1873 | "@end itemize") |
| 1874 | #define FUNC_NAME s_scm_string_skip_right |
| 1875 | { |
| 1876 | size_t cstart, cend; |
| 1877 | |
| 1878 | MY_VALIDATE_SUBSTRING_SPEC (1, s, |
| 1879 | 3, start, cstart, |
| 1880 | 4, end, cend); |
| 1881 | if (SCM_CHARP (char_pred)) |
| 1882 | { |
| 1883 | while (cstart < cend) |
| 1884 | { |
| 1885 | cend--; |
| 1886 | if (scm_i_string_ref (s, cend) != SCM_CHAR (char_pred)) |
| 1887 | goto found; |
| 1888 | } |
| 1889 | } |
| 1890 | else if (SCM_CHARSETP (char_pred)) |
| 1891 | { |
| 1892 | while (cstart < cend) |
| 1893 | { |
| 1894 | cend--; |
| 1895 | if (!REF_IN_CHARSET (s, cend, char_pred)) |
| 1896 | goto found; |
| 1897 | } |
| 1898 | } |
| 1899 | else |
| 1900 | { |
| 1901 | SCM_ASSERT (scm_is_true (scm_procedure_p (char_pred)), |
| 1902 | char_pred, SCM_ARG2, FUNC_NAME); |
| 1903 | |
| 1904 | while (cstart < cend) |
| 1905 | { |
| 1906 | SCM res; |
| 1907 | cend--; |
| 1908 | res = scm_call_1 (char_pred, SCM_MAKE_CHAR (scm_i_string_ref (s, cend))); |
| 1909 | if (scm_is_false (res)) |
| 1910 | goto found; |
| 1911 | } |
| 1912 | } |
| 1913 | |
| 1914 | scm_remember_upto_here_1 (s); |
| 1915 | return SCM_BOOL_F; |
| 1916 | |
| 1917 | found: |
| 1918 | scm_remember_upto_here_1 (s); |
| 1919 | return scm_from_size_t (cend); |
| 1920 | |
| 1921 | } |
| 1922 | #undef FUNC_NAME |
| 1923 | |
| 1924 | |
| 1925 | SCM_DEFINE (scm_string_count, "string-count", 2, 2, 0, |
| 1926 | (SCM s, SCM char_pred, SCM start, SCM end), |
| 1927 | "Return the count of the number of characters in the string\n" |
| 1928 | "@var{s} which\n" |
| 1929 | "\n" |
| 1930 | "@itemize @bullet\n" |
| 1931 | "@item\n" |
| 1932 | "equals @var{char_pred}, if it is character,\n" |
| 1933 | "\n" |
| 1934 | "@item\n" |
| 1935 | "satisifies the predicate @var{char_pred}, if it is a procedure.\n" |
| 1936 | "\n" |
| 1937 | "@item\n" |
| 1938 | "is in the set @var{char_pred}, if it is a character set.\n" |
| 1939 | "@end itemize") |
| 1940 | #define FUNC_NAME s_scm_string_count |
| 1941 | { |
| 1942 | size_t cstart, cend; |
| 1943 | size_t count = 0; |
| 1944 | |
| 1945 | MY_VALIDATE_SUBSTRING_SPEC (1, s, |
| 1946 | 3, start, cstart, |
| 1947 | 4, end, cend); |
| 1948 | if (SCM_CHARP (char_pred)) |
| 1949 | { |
| 1950 | while (cstart < cend) |
| 1951 | { |
| 1952 | if (scm_i_string_ref (s, cstart) == SCM_CHAR(char_pred)) |
| 1953 | count++; |
| 1954 | cstart++; |
| 1955 | } |
| 1956 | } |
| 1957 | else if (SCM_CHARSETP (char_pred)) |
| 1958 | { |
| 1959 | while (cstart < cend) |
| 1960 | { |
| 1961 | if (REF_IN_CHARSET (s, cstart, char_pred)) |
| 1962 | count++; |
| 1963 | cstart++; |
| 1964 | } |
| 1965 | } |
| 1966 | else |
| 1967 | { |
| 1968 | SCM_ASSERT (scm_is_true (scm_procedure_p (char_pred)), |
| 1969 | char_pred, SCM_ARG2, FUNC_NAME); |
| 1970 | |
| 1971 | while (cstart < cend) |
| 1972 | { |
| 1973 | SCM res; |
| 1974 | res = scm_call_1 (char_pred, SCM_MAKE_CHAR (scm_i_string_ref (s, cstart))); |
| 1975 | if (scm_is_true (res)) |
| 1976 | count++; |
| 1977 | cstart++; |
| 1978 | } |
| 1979 | } |
| 1980 | |
| 1981 | scm_remember_upto_here_1 (s); |
| 1982 | return scm_from_size_t (count); |
| 1983 | } |
| 1984 | #undef FUNC_NAME |
| 1985 | |
| 1986 | |
| 1987 | /* FIXME::martin: This should definitely get implemented more |
| 1988 | efficiently -- maybe with Knuth-Morris-Pratt, like in the reference |
| 1989 | implementation. */ |
| 1990 | SCM_DEFINE (scm_string_contains, "string-contains", 2, 4, 0, |
| 1991 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 1992 | "Does string @var{s1} contain string @var{s2}? Return the index\n" |
| 1993 | "in @var{s1} where @var{s2} occurs as a substring, or false.\n" |
| 1994 | "The optional start/end indices restrict the operation to the\n" |
| 1995 | "indicated substrings.") |
| 1996 | #define FUNC_NAME s_scm_string_contains |
| 1997 | { |
| 1998 | size_t cstart1, cend1, cstart2, cend2; |
| 1999 | size_t len2, i, j; |
| 2000 | |
| 2001 | MY_VALIDATE_SUBSTRING_SPEC (1, s1, |
| 2002 | 3, start1, cstart1, |
| 2003 | 4, end1, cend1); |
| 2004 | MY_VALIDATE_SUBSTRING_SPEC (2, s2, |
| 2005 | 5, start2, cstart2, |
| 2006 | 6, end2, cend2); |
| 2007 | len2 = cend2 - cstart2; |
| 2008 | if (cend1 - cstart1 >= len2) |
| 2009 | while (cstart1 <= cend1 - len2) |
| 2010 | { |
| 2011 | i = cstart1; |
| 2012 | j = cstart2; |
| 2013 | while (i < cend1 |
| 2014 | && j < cend2 |
| 2015 | && (scm_i_string_ref (s1, i) |
| 2016 | == scm_i_string_ref (s2, j))) |
| 2017 | { |
| 2018 | i++; |
| 2019 | j++; |
| 2020 | } |
| 2021 | if (j == cend2) |
| 2022 | { |
| 2023 | scm_remember_upto_here_2 (s1, s2); |
| 2024 | return scm_from_size_t (cstart1); |
| 2025 | } |
| 2026 | cstart1++; |
| 2027 | } |
| 2028 | |
| 2029 | scm_remember_upto_here_2 (s1, s2); |
| 2030 | return SCM_BOOL_F; |
| 2031 | } |
| 2032 | #undef FUNC_NAME |
| 2033 | |
| 2034 | |
| 2035 | /* FIXME::martin: This should definitely get implemented more |
| 2036 | efficiently -- maybe with Knuth-Morris-Pratt, like in the reference |
| 2037 | implementation. */ |
| 2038 | SCM_DEFINE (scm_string_contains_ci, "string-contains-ci", 2, 4, 0, |
| 2039 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 2040 | "Does string @var{s1} contain string @var{s2}? Return the index\n" |
| 2041 | "in @var{s1} where @var{s2} occurs as a substring, or false.\n" |
| 2042 | "The optional start/end indices restrict the operation to the\n" |
| 2043 | "indicated substrings. Character comparison is done\n" |
| 2044 | "case-insensitively.") |
| 2045 | #define FUNC_NAME s_scm_string_contains_ci |
| 2046 | { |
| 2047 | size_t cstart1, cend1, cstart2, cend2; |
| 2048 | size_t len2, i, j; |
| 2049 | |
| 2050 | MY_VALIDATE_SUBSTRING_SPEC (1, s1, |
| 2051 | 3, start1, cstart1, |
| 2052 | 4, end1, cend1); |
| 2053 | MY_VALIDATE_SUBSTRING_SPEC (2, s2, |
| 2054 | 5, start2, cstart2, |
| 2055 | 6, end2, cend2); |
| 2056 | len2 = cend2 - cstart2; |
| 2057 | if (cend1 - cstart1 >= len2) |
| 2058 | while (cstart1 <= cend1 - len2) |
| 2059 | { |
| 2060 | i = cstart1; |
| 2061 | j = cstart2; |
| 2062 | while (i < cend1 |
| 2063 | && j < cend2 |
| 2064 | && (uc_tolower (uc_toupper (scm_i_string_ref (s1, i))) |
| 2065 | == uc_tolower (uc_toupper (scm_i_string_ref (s2, j))))) |
| 2066 | { |
| 2067 | i++; |
| 2068 | j++; |
| 2069 | } |
| 2070 | if (j == cend2) |
| 2071 | { |
| 2072 | scm_remember_upto_here_2 (s1, s2); |
| 2073 | return scm_from_size_t (cstart1); |
| 2074 | } |
| 2075 | cstart1++; |
| 2076 | } |
| 2077 | |
| 2078 | scm_remember_upto_here_2 (s1, s2); |
| 2079 | return SCM_BOOL_F; |
| 2080 | } |
| 2081 | #undef FUNC_NAME |
| 2082 | |
| 2083 | |
| 2084 | /* Helper function for the string uppercase conversion functions. */ |
| 2085 | static SCM |
| 2086 | string_upcase_x (SCM v, size_t start, size_t end) |
| 2087 | { |
| 2088 | size_t k; |
| 2089 | |
| 2090 | v = scm_i_string_start_writing (v); |
| 2091 | for (k = start; k < end; ++k) |
| 2092 | scm_i_string_set_x (v, k, uc_toupper (scm_i_string_ref (v, k))); |
| 2093 | scm_i_string_stop_writing (); |
| 2094 | scm_remember_upto_here_1 (v); |
| 2095 | |
| 2096 | return v; |
| 2097 | } |
| 2098 | |
| 2099 | SCM_DEFINE (scm_substring_upcase_x, "string-upcase!", 1, 2, 0, |
| 2100 | (SCM str, SCM start, SCM end), |
| 2101 | "Destructively upcase every character in @code{str}.\n" |
| 2102 | "\n" |
| 2103 | "@lisp\n" |
| 2104 | "(string-upcase! y)\n" |
| 2105 | "@result{} \"ARRDEFG\"\n" |
| 2106 | "y\n" |
| 2107 | "@result{} \"ARRDEFG\"\n" |
| 2108 | "@end lisp") |
| 2109 | #define FUNC_NAME s_scm_substring_upcase_x |
| 2110 | { |
| 2111 | size_t cstart, cend; |
| 2112 | |
| 2113 | MY_VALIDATE_SUBSTRING_SPEC (1, str, |
| 2114 | 2, start, cstart, |
| 2115 | 3, end, cend); |
| 2116 | return string_upcase_x (str, cstart, cend); |
| 2117 | } |
| 2118 | #undef FUNC_NAME |
| 2119 | |
| 2120 | SCM |
| 2121 | scm_string_upcase_x (SCM str) |
| 2122 | { |
| 2123 | return scm_substring_upcase_x (str, SCM_UNDEFINED, SCM_UNDEFINED); |
| 2124 | } |
| 2125 | |
| 2126 | SCM_DEFINE (scm_substring_upcase, "string-upcase", 1, 2, 0, |
| 2127 | (SCM str, SCM start, SCM end), |
| 2128 | "Upcase every character in @code{str}.") |
| 2129 | #define FUNC_NAME s_scm_substring_upcase |
| 2130 | { |
| 2131 | size_t cstart, cend; |
| 2132 | |
| 2133 | MY_VALIDATE_SUBSTRING_SPEC (1, str, |
| 2134 | 2, start, cstart, |
| 2135 | 3, end, cend); |
| 2136 | return string_upcase_x (scm_string_copy (str), cstart, cend); |
| 2137 | } |
| 2138 | #undef FUNC_NAME |
| 2139 | |
| 2140 | SCM |
| 2141 | scm_string_upcase (SCM str) |
| 2142 | { |
| 2143 | return scm_substring_upcase (str, SCM_UNDEFINED, SCM_UNDEFINED); |
| 2144 | } |
| 2145 | |
| 2146 | /* Helper function for the string lowercase conversion functions. |
| 2147 | * No argument checking is performed. */ |
| 2148 | static SCM |
| 2149 | string_downcase_x (SCM v, size_t start, size_t end) |
| 2150 | { |
| 2151 | size_t k; |
| 2152 | |
| 2153 | v = scm_i_string_start_writing (v); |
| 2154 | for (k = start; k < end; ++k) |
| 2155 | scm_i_string_set_x (v, k, uc_tolower (scm_i_string_ref (v, k))); |
| 2156 | scm_i_string_stop_writing (); |
| 2157 | scm_remember_upto_here_1 (v); |
| 2158 | |
| 2159 | return v; |
| 2160 | } |
| 2161 | |
| 2162 | SCM_DEFINE (scm_substring_downcase_x, "string-downcase!", 1, 2, 0, |
| 2163 | (SCM str, SCM start, SCM end), |
| 2164 | "Destructively downcase every character in @var{str}.\n" |
| 2165 | "\n" |
| 2166 | "@lisp\n" |
| 2167 | "y\n" |
| 2168 | "@result{} \"ARRDEFG\"\n" |
| 2169 | "(string-downcase! y)\n" |
| 2170 | "@result{} \"arrdefg\"\n" |
| 2171 | "y\n" |
| 2172 | "@result{} \"arrdefg\"\n" |
| 2173 | "@end lisp") |
| 2174 | #define FUNC_NAME s_scm_substring_downcase_x |
| 2175 | { |
| 2176 | size_t cstart, cend; |
| 2177 | |
| 2178 | MY_VALIDATE_SUBSTRING_SPEC (1, str, |
| 2179 | 2, start, cstart, |
| 2180 | 3, end, cend); |
| 2181 | return string_downcase_x (str, cstart, cend); |
| 2182 | } |
| 2183 | #undef FUNC_NAME |
| 2184 | |
| 2185 | SCM |
| 2186 | scm_string_downcase_x (SCM str) |
| 2187 | { |
| 2188 | return scm_substring_downcase_x (str, SCM_UNDEFINED, SCM_UNDEFINED); |
| 2189 | } |
| 2190 | |
| 2191 | SCM_DEFINE (scm_substring_downcase, "string-downcase", 1, 2, 0, |
| 2192 | (SCM str, SCM start, SCM end), |
| 2193 | "Downcase every character in @var{str}.") |
| 2194 | #define FUNC_NAME s_scm_substring_downcase |
| 2195 | { |
| 2196 | size_t cstart, cend; |
| 2197 | |
| 2198 | MY_VALIDATE_SUBSTRING_SPEC (1, str, |
| 2199 | 2, start, cstart, |
| 2200 | 3, end, cend); |
| 2201 | return string_downcase_x (scm_string_copy (str), cstart, cend); |
| 2202 | } |
| 2203 | #undef FUNC_NAME |
| 2204 | |
| 2205 | SCM |
| 2206 | scm_string_downcase (SCM str) |
| 2207 | { |
| 2208 | return scm_substring_downcase (str, SCM_UNDEFINED, SCM_UNDEFINED); |
| 2209 | } |
| 2210 | |
| 2211 | /* Helper function for the string capitalization functions. |
| 2212 | * No argument checking is performed. */ |
| 2213 | static SCM |
| 2214 | string_titlecase_x (SCM str, size_t start, size_t end) |
| 2215 | { |
| 2216 | SCM ch; |
| 2217 | size_t i; |
| 2218 | int in_word = 0; |
| 2219 | |
| 2220 | str = scm_i_string_start_writing (str); |
| 2221 | for(i = start; i < end; i++) |
| 2222 | { |
| 2223 | ch = SCM_MAKE_CHAR (scm_i_string_ref (str, i)); |
| 2224 | if (scm_is_true (scm_char_alphabetic_p (ch))) |
| 2225 | { |
| 2226 | if (!in_word) |
| 2227 | { |
| 2228 | scm_i_string_set_x (str, i, uc_totitle (SCM_CHAR (ch))); |
| 2229 | in_word = 1; |
| 2230 | } |
| 2231 | else |
| 2232 | { |
| 2233 | scm_i_string_set_x (str, i, uc_tolower (SCM_CHAR (ch))); |
| 2234 | } |
| 2235 | } |
| 2236 | else |
| 2237 | in_word = 0; |
| 2238 | } |
| 2239 | scm_i_string_stop_writing (); |
| 2240 | scm_remember_upto_here_1 (str); |
| 2241 | |
| 2242 | return str; |
| 2243 | } |
| 2244 | |
| 2245 | |
| 2246 | SCM_DEFINE (scm_string_titlecase_x, "string-titlecase!", 1, 2, 0, |
| 2247 | (SCM str, SCM start, SCM end), |
| 2248 | "Destructively titlecase every first character in a word in\n" |
| 2249 | "@var{str}.") |
| 2250 | #define FUNC_NAME s_scm_string_titlecase_x |
| 2251 | { |
| 2252 | size_t cstart, cend; |
| 2253 | |
| 2254 | MY_VALIDATE_SUBSTRING_SPEC (1, str, |
| 2255 | 2, start, cstart, |
| 2256 | 3, end, cend); |
| 2257 | return string_titlecase_x (str, cstart, cend); |
| 2258 | } |
| 2259 | #undef FUNC_NAME |
| 2260 | |
| 2261 | |
| 2262 | SCM_DEFINE (scm_string_titlecase, "string-titlecase", 1, 2, 0, |
| 2263 | (SCM str, SCM start, SCM end), |
| 2264 | "Titlecase every first character in a word in @var{str}.") |
| 2265 | #define FUNC_NAME s_scm_string_titlecase |
| 2266 | { |
| 2267 | size_t cstart, cend; |
| 2268 | |
| 2269 | MY_VALIDATE_SUBSTRING_SPEC (1, str, |
| 2270 | 2, start, cstart, |
| 2271 | 3, end, cend); |
| 2272 | return string_titlecase_x (scm_string_copy (str), cstart, cend); |
| 2273 | } |
| 2274 | #undef FUNC_NAME |
| 2275 | |
| 2276 | SCM_DEFINE (scm_string_capitalize_x, "string-capitalize!", 1, 0, 0, |
| 2277 | (SCM str), |
| 2278 | "Upcase the first character of every word in @var{str}\n" |
| 2279 | "destructively and return @var{str}.\n" |
| 2280 | "\n" |
| 2281 | "@lisp\n" |
| 2282 | "y @result{} \"hello world\"\n" |
| 2283 | "(string-capitalize! y) @result{} \"Hello World\"\n" |
| 2284 | "y @result{} \"Hello World\"\n" |
| 2285 | "@end lisp") |
| 2286 | #define FUNC_NAME s_scm_string_capitalize_x |
| 2287 | { |
| 2288 | return scm_string_titlecase_x (str, SCM_UNDEFINED, SCM_UNDEFINED); |
| 2289 | } |
| 2290 | #undef FUNC_NAME |
| 2291 | |
| 2292 | |
| 2293 | SCM_DEFINE (scm_string_capitalize, "string-capitalize", 1, 0, 0, |
| 2294 | (SCM str), |
| 2295 | "Return a freshly allocated string with the characters in\n" |
| 2296 | "@var{str}, where the first character of every word is\n" |
| 2297 | "capitalized.") |
| 2298 | #define FUNC_NAME s_scm_string_capitalize |
| 2299 | { |
| 2300 | return scm_string_capitalize_x (scm_string_copy (str)); |
| 2301 | } |
| 2302 | #undef FUNC_NAME |
| 2303 | |
| 2304 | |
| 2305 | /* Reverse the portion of @var{str} between str[cstart] (including) |
| 2306 | and str[cend] excluding. */ |
| 2307 | static void |
| 2308 | string_reverse_x (SCM str, size_t cstart, size_t cend) |
| 2309 | { |
| 2310 | SCM tmp; |
| 2311 | |
| 2312 | str = scm_i_string_start_writing (str); |
| 2313 | if (cend > 0) |
| 2314 | { |
| 2315 | cend--; |
| 2316 | while (cstart < cend) |
| 2317 | { |
| 2318 | tmp = SCM_MAKE_CHAR (scm_i_string_ref (str, cstart)); |
| 2319 | scm_i_string_set_x (str, cstart, scm_i_string_ref (str, cend)); |
| 2320 | scm_i_string_set_x (str, cend, SCM_CHAR (tmp)); |
| 2321 | cstart++; |
| 2322 | cend--; |
| 2323 | } |
| 2324 | } |
| 2325 | scm_i_string_stop_writing (); |
| 2326 | } |
| 2327 | |
| 2328 | |
| 2329 | SCM_DEFINE (scm_string_reverse, "string-reverse", 1, 2, 0, |
| 2330 | (SCM str, SCM start, SCM end), |
| 2331 | "Reverse the string @var{str}. The optional arguments\n" |
| 2332 | "@var{start} and @var{end} delimit the region of @var{str} to\n" |
| 2333 | "operate on.") |
| 2334 | #define FUNC_NAME s_scm_string_reverse |
| 2335 | { |
| 2336 | size_t cstart, cend; |
| 2337 | SCM result; |
| 2338 | |
| 2339 | MY_VALIDATE_SUBSTRING_SPEC (1, str, |
| 2340 | 2, start, cstart, |
| 2341 | 3, end, cend); |
| 2342 | result = scm_string_copy (str); |
| 2343 | string_reverse_x (result, cstart, cend); |
| 2344 | scm_remember_upto_here_1 (str); |
| 2345 | return result; |
| 2346 | } |
| 2347 | #undef FUNC_NAME |
| 2348 | |
| 2349 | |
| 2350 | SCM_DEFINE (scm_string_reverse_x, "string-reverse!", 1, 2, 0, |
| 2351 | (SCM str, SCM start, SCM end), |
| 2352 | "Reverse the string @var{str} in-place. The optional arguments\n" |
| 2353 | "@var{start} and @var{end} delimit the region of @var{str} to\n" |
| 2354 | "operate on. The return value is unspecified.") |
| 2355 | #define FUNC_NAME s_scm_string_reverse_x |
| 2356 | { |
| 2357 | size_t cstart, cend; |
| 2358 | |
| 2359 | MY_VALIDATE_SUBSTRING_SPEC (1, str, |
| 2360 | 2, start, cstart, |
| 2361 | 3, end, cend); |
| 2362 | |
| 2363 | string_reverse_x (str, cstart, cend); |
| 2364 | scm_remember_upto_here_1 (str); |
| 2365 | return SCM_UNSPECIFIED; |
| 2366 | } |
| 2367 | #undef FUNC_NAME |
| 2368 | |
| 2369 | |
| 2370 | SCM_DEFINE (scm_string_append_shared, "string-append/shared", 0, 0, 1, |
| 2371 | (SCM rest), |
| 2372 | "Like @code{string-append}, but the result may share memory\n" |
| 2373 | "with the argument strings.") |
| 2374 | #define FUNC_NAME s_scm_string_append_shared |
| 2375 | { |
| 2376 | /* If "rest" contains just one non-empty string, return that. |
| 2377 | If it's entirely empty strings, then return scm_nullstr. |
| 2378 | Otherwise use scm_string_concatenate. */ |
| 2379 | |
| 2380 | SCM ret = scm_nullstr; |
| 2381 | int seen_nonempty = 0; |
| 2382 | SCM l, s; |
| 2383 | |
| 2384 | SCM_VALIDATE_REST_ARGUMENT (rest); |
| 2385 | |
| 2386 | for (l = rest; scm_is_pair (l); l = SCM_CDR (l)) |
| 2387 | { |
| 2388 | s = SCM_CAR (l); |
| 2389 | if (!scm_is_string (s)) |
| 2390 | scm_wrong_type_arg (FUNC_NAME, 0, s); |
| 2391 | if (scm_i_string_length (s) != 0) |
| 2392 | { |
| 2393 | if (seen_nonempty) |
| 2394 | /* two or more non-empty strings, need full concat */ |
| 2395 | return scm_string_append (rest); |
| 2396 | |
| 2397 | seen_nonempty = 1; |
| 2398 | ret = s; |
| 2399 | } |
| 2400 | } |
| 2401 | return ret; |
| 2402 | } |
| 2403 | #undef FUNC_NAME |
| 2404 | |
| 2405 | |
| 2406 | SCM_DEFINE (scm_string_concatenate, "string-concatenate", 1, 0, 0, |
| 2407 | (SCM ls), |
| 2408 | "Append the elements of @var{ls} (which must be strings)\n" |
| 2409 | "together into a single string. Guaranteed to return a freshly\n" |
| 2410 | "allocated string.") |
| 2411 | #define FUNC_NAME s_scm_string_concatenate |
| 2412 | { |
| 2413 | SCM_VALIDATE_LIST (SCM_ARG1, ls); |
| 2414 | return scm_string_append (ls); |
| 2415 | } |
| 2416 | #undef FUNC_NAME |
| 2417 | |
| 2418 | |
| 2419 | SCM_DEFINE (scm_string_concatenate_reverse, "string-concatenate-reverse", 1, 2, 0, |
| 2420 | (SCM ls, SCM final_string, SCM end), |
| 2421 | "Without optional arguments, this procedure is equivalent to\n" |
| 2422 | "\n" |
| 2423 | "@smalllisp\n" |
| 2424 | "(string-concatenate (reverse ls))\n" |
| 2425 | "@end smalllisp\n" |
| 2426 | "\n" |
| 2427 | "If the optional argument @var{final_string} is specified, it is\n" |
| 2428 | "consed onto the beginning to @var{ls} before performing the\n" |
| 2429 | "list-reverse and string-concatenate operations. If @var{end}\n" |
| 2430 | "is given, only the characters of @var{final_string} up to index\n" |
| 2431 | "@var{end} are used.\n" |
| 2432 | "\n" |
| 2433 | "Guaranteed to return a freshly allocated string.") |
| 2434 | #define FUNC_NAME s_scm_string_concatenate_reverse |
| 2435 | { |
| 2436 | if (!SCM_UNBNDP (end)) |
| 2437 | final_string = scm_substring (final_string, SCM_INUM0, end); |
| 2438 | |
| 2439 | if (!SCM_UNBNDP (final_string)) |
| 2440 | ls = scm_cons (final_string, ls); |
| 2441 | |
| 2442 | return scm_string_concatenate (scm_reverse (ls)); |
| 2443 | } |
| 2444 | #undef FUNC_NAME |
| 2445 | |
| 2446 | |
| 2447 | SCM_DEFINE (scm_string_concatenate_shared, "string-concatenate/shared", 1, 0, 0, |
| 2448 | (SCM ls), |
| 2449 | "Like @code{string-concatenate}, but the result may share memory\n" |
| 2450 | "with the strings in the list @var{ls}.") |
| 2451 | #define FUNC_NAME s_scm_string_concatenate_shared |
| 2452 | { |
| 2453 | SCM_VALIDATE_LIST (SCM_ARG1, ls); |
| 2454 | return scm_string_append_shared (ls); |
| 2455 | } |
| 2456 | #undef FUNC_NAME |
| 2457 | |
| 2458 | |
| 2459 | SCM_DEFINE (scm_string_concatenate_reverse_shared, "string-concatenate-reverse/shared", 1, 2, 0, |
| 2460 | (SCM ls, SCM final_string, SCM end), |
| 2461 | "Like @code{string-concatenate-reverse}, but the result may\n" |
| 2462 | "share memory with the the strings in the @var{ls} arguments.") |
| 2463 | #define FUNC_NAME s_scm_string_concatenate_reverse_shared |
| 2464 | { |
| 2465 | /* Just call the non-sharing version. */ |
| 2466 | return scm_string_concatenate_reverse (ls, final_string, end); |
| 2467 | } |
| 2468 | #undef FUNC_NAME |
| 2469 | |
| 2470 | |
| 2471 | SCM_DEFINE (scm_string_map, "string-map", 2, 2, 0, |
| 2472 | (SCM proc, SCM s, SCM start, SCM end), |
| 2473 | "@var{proc} is a char->char procedure, it is mapped over\n" |
| 2474 | "@var{s}. The order in which the procedure is applied to the\n" |
| 2475 | "string elements is not specified.") |
| 2476 | #define FUNC_NAME s_scm_string_map |
| 2477 | { |
| 2478 | size_t p; |
| 2479 | size_t cstart, cend; |
| 2480 | SCM result; |
| 2481 | |
| 2482 | SCM_ASSERT (scm_is_true (scm_procedure_p (proc)), |
| 2483 | proc, SCM_ARG1, FUNC_NAME); |
| 2484 | MY_VALIDATE_SUBSTRING_SPEC (2, s, |
| 2485 | 3, start, cstart, |
| 2486 | 4, end, cend); |
| 2487 | result = scm_i_make_string (cend - cstart, NULL); |
| 2488 | p = 0; |
| 2489 | while (cstart < cend) |
| 2490 | { |
| 2491 | SCM ch = scm_call_1 (proc, scm_c_string_ref (s, cstart)); |
| 2492 | if (!SCM_CHARP (ch)) |
| 2493 | SCM_MISC_ERROR ("procedure ~S returned non-char", scm_list_1 (proc)); |
| 2494 | cstart++; |
| 2495 | result = scm_i_string_start_writing (result); |
| 2496 | scm_i_string_set_x (result, p, SCM_CHAR (ch)); |
| 2497 | scm_i_string_stop_writing (); |
| 2498 | p++; |
| 2499 | } |
| 2500 | |
| 2501 | return result; |
| 2502 | } |
| 2503 | #undef FUNC_NAME |
| 2504 | |
| 2505 | |
| 2506 | SCM_DEFINE (scm_string_map_x, "string-map!", 2, 2, 0, |
| 2507 | (SCM proc, SCM s, SCM start, SCM end), |
| 2508 | "@var{proc} is a char->char procedure, it is mapped over\n" |
| 2509 | "@var{s}. The order in which the procedure is applied to the\n" |
| 2510 | "string elements is not specified. The string @var{s} is\n" |
| 2511 | "modified in-place, the return value is not specified.") |
| 2512 | #define FUNC_NAME s_scm_string_map_x |
| 2513 | { |
| 2514 | size_t cstart, cend; |
| 2515 | |
| 2516 | SCM_ASSERT (scm_is_true (scm_procedure_p (proc)), |
| 2517 | proc, SCM_ARG1, FUNC_NAME); |
| 2518 | MY_VALIDATE_SUBSTRING_SPEC (2, s, |
| 2519 | 3, start, cstart, |
| 2520 | 4, end, cend); |
| 2521 | while (cstart < cend) |
| 2522 | { |
| 2523 | SCM ch = scm_call_1 (proc, scm_c_string_ref (s, cstart)); |
| 2524 | if (!SCM_CHARP (ch)) |
| 2525 | SCM_MISC_ERROR ("procedure ~S returned non-char", scm_list_1 (proc)); |
| 2526 | s = scm_i_string_start_writing (s); |
| 2527 | scm_i_string_set_x (s, cstart, SCM_CHAR (ch)); |
| 2528 | scm_i_string_stop_writing (); |
| 2529 | cstart++; |
| 2530 | } |
| 2531 | return SCM_UNSPECIFIED; |
| 2532 | } |
| 2533 | #undef FUNC_NAME |
| 2534 | |
| 2535 | |
| 2536 | SCM_DEFINE (scm_string_fold, "string-fold", 3, 2, 0, |
| 2537 | (SCM kons, SCM knil, SCM s, SCM start, SCM end), |
| 2538 | "Fold @var{kons} over the characters of @var{s}, with @var{knil}\n" |
| 2539 | "as the terminating element, from left to right. @var{kons}\n" |
| 2540 | "must expect two arguments: The actual character and the last\n" |
| 2541 | "result of @var{kons}' application.") |
| 2542 | #define FUNC_NAME s_scm_string_fold |
| 2543 | { |
| 2544 | size_t cstart, cend; |
| 2545 | SCM result; |
| 2546 | |
| 2547 | SCM_VALIDATE_PROC (1, kons); |
| 2548 | MY_VALIDATE_SUBSTRING_SPEC (3, s, |
| 2549 | 4, start, cstart, |
| 2550 | 5, end, cend); |
| 2551 | result = knil; |
| 2552 | while (cstart < cend) |
| 2553 | { |
| 2554 | result = scm_call_2 (kons, SCM_MAKE_CHAR (scm_i_string_ref (s, cstart)), result); |
| 2555 | cstart++; |
| 2556 | } |
| 2557 | |
| 2558 | scm_remember_upto_here_1 (s); |
| 2559 | return result; |
| 2560 | } |
| 2561 | #undef FUNC_NAME |
| 2562 | |
| 2563 | |
| 2564 | SCM_DEFINE (scm_string_fold_right, "string-fold-right", 3, 2, 0, |
| 2565 | (SCM kons, SCM knil, SCM s, SCM start, SCM end), |
| 2566 | "Fold @var{kons} over the characters of @var{s}, with @var{knil}\n" |
| 2567 | "as the terminating element, from right to left. @var{kons}\n" |
| 2568 | "must expect two arguments: The actual character and the last\n" |
| 2569 | "result of @var{kons}' application.") |
| 2570 | #define FUNC_NAME s_scm_string_fold_right |
| 2571 | { |
| 2572 | size_t cstart, cend; |
| 2573 | SCM result; |
| 2574 | |
| 2575 | SCM_VALIDATE_PROC (1, kons); |
| 2576 | MY_VALIDATE_SUBSTRING_SPEC (3, s, |
| 2577 | 4, start, cstart, |
| 2578 | 5, end, cend); |
| 2579 | result = knil; |
| 2580 | while (cstart < cend) |
| 2581 | { |
| 2582 | result = scm_call_2 (kons, SCM_MAKE_CHAR (scm_i_string_ref (s, cend-1)), result); |
| 2583 | cend--; |
| 2584 | } |
| 2585 | |
| 2586 | scm_remember_upto_here_1 (s); |
| 2587 | return result; |
| 2588 | } |
| 2589 | #undef FUNC_NAME |
| 2590 | |
| 2591 | |
| 2592 | SCM_DEFINE (scm_string_unfold, "string-unfold", 4, 2, 0, |
| 2593 | (SCM p, SCM f, SCM g, SCM seed, SCM base, SCM make_final), |
| 2594 | "@itemize @bullet\n" |
| 2595 | "@item @var{g} is used to generate a series of @emph{seed}\n" |
| 2596 | "values from the initial @var{seed}: @var{seed}, (@var{g}\n" |
| 2597 | "@var{seed}), (@var{g}^2 @var{seed}), (@var{g}^3 @var{seed}),\n" |
| 2598 | "@dots{}\n" |
| 2599 | "@item @var{p} tells us when to stop -- when it returns true\n" |
| 2600 | "when applied to one of these seed values.\n" |
| 2601 | "@item @var{f} maps each seed value to the corresponding\n" |
| 2602 | "character in the result string. These chars are assembled\n" |
| 2603 | "into the string in a left-to-right order.\n" |
| 2604 | "@item @var{base} is the optional initial/leftmost portion\n" |
| 2605 | "of the constructed string; it default to the empty\n" |
| 2606 | "string.\n" |
| 2607 | "@item @var{make_final} is applied to the terminal seed\n" |
| 2608 | "value (on which @var{p} returns true) to produce\n" |
| 2609 | "the final/rightmost portion of the constructed string.\n" |
| 2610 | "It defaults to @code{(lambda (x) "")}.\n" |
| 2611 | "@end itemize") |
| 2612 | #define FUNC_NAME s_scm_string_unfold |
| 2613 | { |
| 2614 | SCM res, ans; |
| 2615 | |
| 2616 | SCM_VALIDATE_PROC (1, p); |
| 2617 | SCM_VALIDATE_PROC (2, f); |
| 2618 | SCM_VALIDATE_PROC (3, g); |
| 2619 | if (!SCM_UNBNDP (base)) |
| 2620 | { |
| 2621 | SCM_VALIDATE_STRING (5, base); |
| 2622 | ans = base; |
| 2623 | } |
| 2624 | else |
| 2625 | ans = scm_i_make_string (0, NULL); |
| 2626 | if (!SCM_UNBNDP (make_final)) |
| 2627 | SCM_VALIDATE_PROC (6, make_final); |
| 2628 | |
| 2629 | res = scm_call_1 (p, seed); |
| 2630 | while (scm_is_false (res)) |
| 2631 | { |
| 2632 | SCM str; |
| 2633 | size_t i = 0; |
| 2634 | SCM ch = scm_call_1 (f, seed); |
| 2635 | if (!SCM_CHARP (ch)) |
| 2636 | SCM_MISC_ERROR ("procedure ~S returned non-char", scm_list_1 (f)); |
| 2637 | str = scm_i_make_string (1, NULL); |
| 2638 | str = scm_i_string_start_writing (str); |
| 2639 | scm_i_string_set_x (str, i, SCM_CHAR (ch)); |
| 2640 | scm_i_string_stop_writing (); |
| 2641 | i++; |
| 2642 | |
| 2643 | ans = scm_string_append (scm_list_2 (ans, str)); |
| 2644 | seed = scm_call_1 (g, seed); |
| 2645 | res = scm_call_1 (p, seed); |
| 2646 | } |
| 2647 | if (!SCM_UNBNDP (make_final)) |
| 2648 | { |
| 2649 | res = scm_call_1 (make_final, seed); |
| 2650 | return scm_string_append (scm_list_2 (ans, res)); |
| 2651 | } |
| 2652 | else |
| 2653 | return ans; |
| 2654 | } |
| 2655 | #undef FUNC_NAME |
| 2656 | |
| 2657 | |
| 2658 | SCM_DEFINE (scm_string_unfold_right, "string-unfold-right", 4, 2, 0, |
| 2659 | (SCM p, SCM f, SCM g, SCM seed, SCM base, SCM make_final), |
| 2660 | "@itemize @bullet\n" |
| 2661 | "@item @var{g} is used to generate a series of @emph{seed}\n" |
| 2662 | "values from the initial @var{seed}: @var{seed}, (@var{g}\n" |
| 2663 | "@var{seed}), (@var{g}^2 @var{seed}), (@var{g}^3 @var{seed}),\n" |
| 2664 | "@dots{}\n" |
| 2665 | "@item @var{p} tells us when to stop -- when it returns true\n" |
| 2666 | "when applied to one of these seed values.\n" |
| 2667 | "@item @var{f} maps each seed value to the corresponding\n" |
| 2668 | "character in the result string. These chars are assembled\n" |
| 2669 | "into the string in a right-to-left order.\n" |
| 2670 | "@item @var{base} is the optional initial/rightmost portion\n" |
| 2671 | "of the constructed string; it default to the empty\n" |
| 2672 | "string.\n" |
| 2673 | "@item @var{make_final} is applied to the terminal seed\n" |
| 2674 | "value (on which @var{p} returns true) to produce\n" |
| 2675 | "the final/leftmost portion of the constructed string.\n" |
| 2676 | "It defaults to @code{(lambda (x) "")}.\n" |
| 2677 | "@end itemize") |
| 2678 | #define FUNC_NAME s_scm_string_unfold_right |
| 2679 | { |
| 2680 | SCM res, ans; |
| 2681 | |
| 2682 | SCM_VALIDATE_PROC (1, p); |
| 2683 | SCM_VALIDATE_PROC (2, f); |
| 2684 | SCM_VALIDATE_PROC (3, g); |
| 2685 | if (!SCM_UNBNDP (base)) |
| 2686 | { |
| 2687 | SCM_VALIDATE_STRING (5, base); |
| 2688 | ans = base; |
| 2689 | } |
| 2690 | else |
| 2691 | ans = scm_i_make_string (0, NULL); |
| 2692 | if (!SCM_UNBNDP (make_final)) |
| 2693 | SCM_VALIDATE_PROC (6, make_final); |
| 2694 | |
| 2695 | res = scm_call_1 (p, seed); |
| 2696 | while (scm_is_false (res)) |
| 2697 | { |
| 2698 | SCM str; |
| 2699 | size_t i = 0; |
| 2700 | SCM ch = scm_call_1 (f, seed); |
| 2701 | if (!SCM_CHARP (ch)) |
| 2702 | SCM_MISC_ERROR ("procedure ~S returned non-char", scm_list_1 (f)); |
| 2703 | str = scm_i_make_string (1, NULL); |
| 2704 | str = scm_i_string_start_writing (str); |
| 2705 | scm_i_string_set_x (str, i, SCM_CHAR (ch)); |
| 2706 | scm_i_string_stop_writing (); |
| 2707 | i++; |
| 2708 | |
| 2709 | ans = scm_string_append (scm_list_2 (str, ans)); |
| 2710 | seed = scm_call_1 (g, seed); |
| 2711 | res = scm_call_1 (p, seed); |
| 2712 | } |
| 2713 | if (!SCM_UNBNDP (make_final)) |
| 2714 | { |
| 2715 | res = scm_call_1 (make_final, seed); |
| 2716 | return scm_string_append (scm_list_2 (res, ans)); |
| 2717 | } |
| 2718 | else |
| 2719 | return ans; |
| 2720 | } |
| 2721 | #undef FUNC_NAME |
| 2722 | |
| 2723 | |
| 2724 | SCM_DEFINE (scm_string_for_each, "string-for-each", 2, 2, 0, |
| 2725 | (SCM proc, SCM s, SCM start, SCM end), |
| 2726 | "@var{proc} is mapped over @var{s} in left-to-right order. The\n" |
| 2727 | "return value is not specified.") |
| 2728 | #define FUNC_NAME s_scm_string_for_each |
| 2729 | { |
| 2730 | size_t cstart, cend; |
| 2731 | |
| 2732 | SCM_ASSERT (scm_is_true (scm_procedure_p (proc)), |
| 2733 | proc, SCM_ARG1, FUNC_NAME); |
| 2734 | MY_VALIDATE_SUBSTRING_SPEC (2, s, |
| 2735 | 3, start, cstart, |
| 2736 | 4, end, cend); |
| 2737 | while (cstart < cend) |
| 2738 | { |
| 2739 | scm_call_1 (proc, SCM_MAKE_CHAR (scm_i_string_ref (s, cstart))); |
| 2740 | cstart++; |
| 2741 | } |
| 2742 | |
| 2743 | scm_remember_upto_here_1 (s); |
| 2744 | return SCM_UNSPECIFIED; |
| 2745 | } |
| 2746 | #undef FUNC_NAME |
| 2747 | |
| 2748 | SCM_DEFINE (scm_string_for_each_index, "string-for-each-index", 2, 2, 0, |
| 2749 | (SCM proc, SCM s, SCM start, SCM end), |
| 2750 | "Call @code{(@var{proc} i)} for each index i in @var{s}, from\n" |
| 2751 | "left to right.\n" |
| 2752 | "\n" |
| 2753 | "For example, to change characters to alternately upper and\n" |
| 2754 | "lower case,\n" |
| 2755 | "\n" |
| 2756 | "@example\n" |
| 2757 | "(define str (string-copy \"studly\"))\n" |
| 2758 | "(string-for-each-index\n" |
| 2759 | " (lambda (i)\n" |
| 2760 | " (string-set! str i\n" |
| 2761 | " ((if (even? i) char-upcase char-downcase)\n" |
| 2762 | " (string-ref str i))))\n" |
| 2763 | " str)\n" |
| 2764 | "str @result{} \"StUdLy\"\n" |
| 2765 | "@end example") |
| 2766 | #define FUNC_NAME s_scm_string_for_each_index |
| 2767 | { |
| 2768 | size_t cstart, cend; |
| 2769 | |
| 2770 | SCM_ASSERT (scm_is_true (scm_procedure_p (proc)), |
| 2771 | proc, SCM_ARG1, FUNC_NAME); |
| 2772 | MY_VALIDATE_SUBSTRING_SPEC (2, s, |
| 2773 | 3, start, cstart, |
| 2774 | 4, end, cend); |
| 2775 | |
| 2776 | while (cstart < cend) |
| 2777 | { |
| 2778 | scm_call_1 (proc, scm_from_size_t (cstart)); |
| 2779 | cstart++; |
| 2780 | } |
| 2781 | |
| 2782 | scm_remember_upto_here_1 (s); |
| 2783 | return SCM_UNSPECIFIED; |
| 2784 | } |
| 2785 | #undef FUNC_NAME |
| 2786 | |
| 2787 | SCM_DEFINE (scm_xsubstring, "xsubstring", 2, 3, 0, |
| 2788 | (SCM s, SCM from, SCM to, SCM start, SCM end), |
| 2789 | "This is the @emph{extended substring} procedure that implements\n" |
| 2790 | "replicated copying of a substring of some string.\n" |
| 2791 | "\n" |
| 2792 | "@var{s} is a string, @var{start} and @var{end} are optional\n" |
| 2793 | "arguments that demarcate a substring of @var{s}, defaulting to\n" |
| 2794 | "0 and the length of @var{s}. Replicate this substring up and\n" |
| 2795 | "down index space, in both the positive and negative directions.\n" |
| 2796 | "@code{xsubstring} returns the substring of this string\n" |
| 2797 | "beginning at index @var{from}, and ending at @var{to}, which\n" |
| 2798 | "defaults to @var{from} + (@var{end} - @var{start}).") |
| 2799 | #define FUNC_NAME s_scm_xsubstring |
| 2800 | { |
| 2801 | size_t p; |
| 2802 | size_t cstart, cend; |
| 2803 | int cfrom, cto; |
| 2804 | SCM result; |
| 2805 | |
| 2806 | MY_VALIDATE_SUBSTRING_SPEC (1, s, |
| 2807 | 4, start, cstart, |
| 2808 | 5, end, cend); |
| 2809 | |
| 2810 | cfrom = scm_to_int (from); |
| 2811 | if (SCM_UNBNDP (to)) |
| 2812 | cto = cfrom + (cend - cstart); |
| 2813 | else |
| 2814 | cto = scm_to_int (to); |
| 2815 | if (cstart == cend && cfrom != cto) |
| 2816 | SCM_MISC_ERROR ("start and end indices must not be equal", SCM_EOL); |
| 2817 | |
| 2818 | result = scm_i_make_string (cto - cfrom, NULL); |
| 2819 | result = scm_i_string_start_writing (result); |
| 2820 | |
| 2821 | p = 0; |
| 2822 | while (cfrom < cto) |
| 2823 | { |
| 2824 | size_t t = ((cfrom < 0) ? -cfrom : cfrom) % (cend - cstart); |
| 2825 | if (cfrom < 0) |
| 2826 | scm_i_string_set_x (result, p, |
| 2827 | scm_i_string_ref (s, (cend - cstart) - t)); |
| 2828 | else |
| 2829 | scm_i_string_set_x (result, p, scm_i_string_ref (s, t)); |
| 2830 | cfrom++; |
| 2831 | p++; |
| 2832 | } |
| 2833 | scm_i_string_stop_writing (); |
| 2834 | |
| 2835 | scm_remember_upto_here_1 (s); |
| 2836 | return result; |
| 2837 | } |
| 2838 | #undef FUNC_NAME |
| 2839 | |
| 2840 | |
| 2841 | SCM_DEFINE (scm_string_xcopy_x, "string-xcopy!", 4, 3, 0, |
| 2842 | (SCM target, SCM tstart, SCM s, SCM sfrom, SCM sto, SCM start, SCM end), |
| 2843 | "Exactly the same as @code{xsubstring}, but the extracted text\n" |
| 2844 | "is written into the string @var{target} starting at index\n" |
| 2845 | "@var{tstart}. The operation is not defined if @code{(eq?\n" |
| 2846 | "@var{target} @var{s})} or these arguments share storage -- you\n" |
| 2847 | "cannot copy a string on top of itself.") |
| 2848 | #define FUNC_NAME s_scm_string_xcopy_x |
| 2849 | { |
| 2850 | size_t p; |
| 2851 | size_t ctstart, cstart, cend; |
| 2852 | int csfrom, csto; |
| 2853 | SCM dummy = SCM_UNDEFINED; |
| 2854 | size_t cdummy; |
| 2855 | |
| 2856 | MY_VALIDATE_SUBSTRING_SPEC (1, target, |
| 2857 | 2, tstart, ctstart, |
| 2858 | 2, dummy, cdummy); |
| 2859 | MY_VALIDATE_SUBSTRING_SPEC (3, s, |
| 2860 | 6, start, cstart, |
| 2861 | 7, end, cend); |
| 2862 | csfrom = scm_to_int (sfrom); |
| 2863 | if (SCM_UNBNDP (sto)) |
| 2864 | csto = csfrom + (cend - cstart); |
| 2865 | else |
| 2866 | csto = scm_to_int (sto); |
| 2867 | if (cstart == cend && csfrom != csto) |
| 2868 | SCM_MISC_ERROR ("start and end indices must not be equal", SCM_EOL); |
| 2869 | SCM_ASSERT_RANGE (1, tstart, |
| 2870 | ctstart + (csto - csfrom) <= scm_i_string_length (target)); |
| 2871 | |
| 2872 | p = 0; |
| 2873 | target = scm_i_string_start_writing (target); |
| 2874 | while (csfrom < csto) |
| 2875 | { |
| 2876 | size_t t = ((csfrom < 0) ? -csfrom : csfrom) % (cend - cstart); |
| 2877 | if (csfrom < 0) |
| 2878 | scm_i_string_set_x (target, p + cstart, scm_i_string_ref (s, (cend - cstart) - t)); |
| 2879 | else |
| 2880 | scm_i_string_set_x (target, p + cstart, scm_i_string_ref (s, t)); |
| 2881 | csfrom++; |
| 2882 | p++; |
| 2883 | } |
| 2884 | scm_i_string_stop_writing (); |
| 2885 | |
| 2886 | scm_remember_upto_here_2 (target, s); |
| 2887 | return SCM_UNSPECIFIED; |
| 2888 | } |
| 2889 | #undef FUNC_NAME |
| 2890 | |
| 2891 | |
| 2892 | SCM_DEFINE (scm_string_replace, "string-replace", 2, 4, 0, |
| 2893 | (SCM s1, SCM s2, SCM start1, SCM end1, SCM start2, SCM end2), |
| 2894 | "Return the string @var{s1}, but with the characters\n" |
| 2895 | "@var{start1} @dots{} @var{end1} replaced by the characters\n" |
| 2896 | "@var{start2} @dots{} @var{end2} from @var{s2}.") |
| 2897 | #define FUNC_NAME s_scm_string_replace |
| 2898 | { |
| 2899 | size_t cstart1, cend1, cstart2, cend2; |
| 2900 | SCM result; |
| 2901 | |
| 2902 | MY_VALIDATE_SUBSTRING_SPEC (1, s1, |
| 2903 | 3, start1, cstart1, |
| 2904 | 4, end1, cend1); |
| 2905 | MY_VALIDATE_SUBSTRING_SPEC (2, s2, |
| 2906 | 5, start2, cstart2, |
| 2907 | 6, end2, cend2); |
| 2908 | return (scm_string_append |
| 2909 | (scm_list_3 (scm_i_substring (s1, 0, cstart1), |
| 2910 | scm_i_substring (s2, cstart2, cend2), |
| 2911 | scm_i_substring (s1, cend1, scm_i_string_length (s1))))); |
| 2912 | return result; |
| 2913 | } |
| 2914 | #undef FUNC_NAME |
| 2915 | |
| 2916 | |
| 2917 | SCM_DEFINE (scm_string_tokenize, "string-tokenize", 1, 3, 0, |
| 2918 | (SCM s, SCM token_set, SCM start, SCM end), |
| 2919 | "Split the string @var{s} into a list of substrings, where each\n" |
| 2920 | "substring is a maximal non-empty contiguous sequence of\n" |
| 2921 | "characters from the character set @var{token_set}, which\n" |
| 2922 | "defaults to @code{char-set:graphic}.\n" |
| 2923 | "If @var{start} or @var{end} indices are provided, they restrict\n" |
| 2924 | "@code{string-tokenize} to operating on the indicated substring\n" |
| 2925 | "of @var{s}.") |
| 2926 | #define FUNC_NAME s_scm_string_tokenize |
| 2927 | { |
| 2928 | size_t cstart, cend; |
| 2929 | SCM result = SCM_EOL; |
| 2930 | |
| 2931 | MY_VALIDATE_SUBSTRING_SPEC (1, s, |
| 2932 | 3, start, cstart, |
| 2933 | 4, end, cend); |
| 2934 | |
| 2935 | if (SCM_UNBNDP (token_set)) |
| 2936 | token_set = scm_char_set_graphic; |
| 2937 | |
| 2938 | if (SCM_CHARSETP (token_set)) |
| 2939 | { |
| 2940 | size_t idx; |
| 2941 | |
| 2942 | while (cstart < cend) |
| 2943 | { |
| 2944 | while (cstart < cend) |
| 2945 | { |
| 2946 | if (REF_IN_CHARSET (s, cend-1, token_set)) |
| 2947 | break; |
| 2948 | cend--; |
| 2949 | } |
| 2950 | if (cstart >= cend) |
| 2951 | break; |
| 2952 | idx = cend; |
| 2953 | while (cstart < cend) |
| 2954 | { |
| 2955 | if (!REF_IN_CHARSET (s, cend-1, token_set)) |
| 2956 | break; |
| 2957 | cend--; |
| 2958 | } |
| 2959 | result = scm_cons (scm_i_substring (s, cend, idx), result); |
| 2960 | } |
| 2961 | } |
| 2962 | else |
| 2963 | SCM_WRONG_TYPE_ARG (2, token_set); |
| 2964 | |
| 2965 | scm_remember_upto_here_1 (s); |
| 2966 | return result; |
| 2967 | } |
| 2968 | #undef FUNC_NAME |
| 2969 | |
| 2970 | SCM_DEFINE (scm_string_split, "string-split", 2, 0, 0, |
| 2971 | (SCM str, SCM chr), |
| 2972 | "Split the string @var{str} into the a list of the substrings delimited\n" |
| 2973 | "by appearances of the character @var{chr}. Note that an empty substring\n" |
| 2974 | "between separator characters will result in an empty string in the\n" |
| 2975 | "result list.\n" |
| 2976 | "\n" |
| 2977 | "@lisp\n" |
| 2978 | "(string-split \"root:x:0:0:root:/root:/bin/bash\" #\\:)\n" |
| 2979 | "@result{}\n" |
| 2980 | "(\"root\" \"x\" \"0\" \"0\" \"root\" \"/root\" \"/bin/bash\")\n" |
| 2981 | "\n" |
| 2982 | "(string-split \"::\" #\\:)\n" |
| 2983 | "@result{}\n" |
| 2984 | "(\"\" \"\" \"\")\n" |
| 2985 | "\n" |
| 2986 | "(string-split \"\" #\\:)\n" |
| 2987 | "@result{}\n" |
| 2988 | "(\"\")\n" |
| 2989 | "@end lisp") |
| 2990 | #define FUNC_NAME s_scm_string_split |
| 2991 | { |
| 2992 | long idx, last_idx; |
| 2993 | int narrow; |
| 2994 | SCM res = SCM_EOL; |
| 2995 | |
| 2996 | SCM_VALIDATE_STRING (1, str); |
| 2997 | SCM_VALIDATE_CHAR (2, chr); |
| 2998 | |
| 2999 | /* This is explicit wide/narrow logic (instead of using |
| 3000 | scm_i_string_ref) is a speed optimization. */ |
| 3001 | idx = scm_i_string_length (str); |
| 3002 | narrow = scm_i_is_narrow_string (str); |
| 3003 | if (narrow) |
| 3004 | { |
| 3005 | const char *buf = scm_i_string_chars (str); |
| 3006 | while (idx >= 0) |
| 3007 | { |
| 3008 | last_idx = idx; |
| 3009 | while (idx > 0 && buf[idx-1] != (char) SCM_CHAR(chr)) |
| 3010 | idx--; |
| 3011 | if (idx >= 0) |
| 3012 | { |
| 3013 | res = scm_cons (scm_i_substring (str, idx, last_idx), res); |
| 3014 | idx--; |
| 3015 | } |
| 3016 | } |
| 3017 | } |
| 3018 | else |
| 3019 | { |
| 3020 | const scm_t_wchar *buf = scm_i_string_wide_chars (str); |
| 3021 | while (idx >= 0) |
| 3022 | { |
| 3023 | last_idx = idx; |
| 3024 | while (idx > 0 && buf[idx-1] != SCM_CHAR(chr)) |
| 3025 | idx--; |
| 3026 | if (idx >= 0) |
| 3027 | { |
| 3028 | res = scm_cons (scm_i_substring (str, idx, last_idx), res); |
| 3029 | idx--; |
| 3030 | } |
| 3031 | } |
| 3032 | } |
| 3033 | scm_remember_upto_here_1 (str); |
| 3034 | return res; |
| 3035 | } |
| 3036 | #undef FUNC_NAME |
| 3037 | |
| 3038 | |
| 3039 | SCM_DEFINE (scm_string_filter, "string-filter", 2, 2, 0, |
| 3040 | (SCM char_pred, SCM s, SCM start, SCM end), |
| 3041 | "Filter the string @var{s}, retaining only those characters\n" |
| 3042 | "which satisfy @var{char_pred}.\n" |
| 3043 | "\n" |
| 3044 | "If @var{char_pred} is a procedure, it is applied to each\n" |
| 3045 | "character as a predicate, if it is a character, it is tested\n" |
| 3046 | "for equality and if it is a character set, it is tested for\n" |
| 3047 | "membership.") |
| 3048 | #define FUNC_NAME s_scm_string_filter |
| 3049 | { |
| 3050 | size_t cstart, cend; |
| 3051 | SCM result; |
| 3052 | size_t idx; |
| 3053 | |
| 3054 | if (scm_is_string (char_pred)) |
| 3055 | { |
| 3056 | SCM tmp; |
| 3057 | |
| 3058 | scm_c_issue_deprecation_warning |
| 3059 | ("Guile used to use the wrong argument order for string-filter.\n" |
| 3060 | "This call to string-filter had the arguments in the wrong order.\n" |
| 3061 | "See SRFI-13 for more details. At some point we will remove this hack."); |
| 3062 | |
| 3063 | tmp = char_pred; |
| 3064 | char_pred = s; |
| 3065 | s = tmp; |
| 3066 | } |
| 3067 | |
| 3068 | MY_VALIDATE_SUBSTRING_SPEC (2, s, |
| 3069 | 3, start, cstart, |
| 3070 | 4, end, cend); |
| 3071 | |
| 3072 | /* The explicit loops below stripping leading and trailing non-matches |
| 3073 | mean we can return a substring if those are the only deletions, making |
| 3074 | string-filter as efficient as string-trim-both in that case. */ |
| 3075 | |
| 3076 | if (SCM_CHARP (char_pred)) |
| 3077 | { |
| 3078 | size_t count; |
| 3079 | |
| 3080 | /* strip leading non-matches by incrementing cstart */ |
| 3081 | while (cstart < cend && scm_i_string_ref (s, cstart) != SCM_CHAR (char_pred)) |
| 3082 | cstart++; |
| 3083 | |
| 3084 | /* strip trailing non-matches by decrementing cend */ |
| 3085 | while (cend > cstart && scm_i_string_ref (s, cend-1) != SCM_CHAR (char_pred)) |
| 3086 | cend--; |
| 3087 | |
| 3088 | /* count chars to keep */ |
| 3089 | count = 0; |
| 3090 | for (idx = cstart; idx < cend; idx++) |
| 3091 | if (scm_i_string_ref (s, idx) == SCM_CHAR (char_pred)) |
| 3092 | count++; |
| 3093 | |
| 3094 | if (count == cend - cstart) |
| 3095 | { |
| 3096 | /* whole of cstart to cend is to be kept, return a copy-on-write |
| 3097 | substring */ |
| 3098 | result_substring: |
| 3099 | result = scm_i_substring (s, cstart, cend); |
| 3100 | } |
| 3101 | else |
| 3102 | result = scm_c_make_string (count, char_pred); |
| 3103 | } |
| 3104 | else if (SCM_CHARSETP (char_pred)) |
| 3105 | { |
| 3106 | size_t count; |
| 3107 | |
| 3108 | /* strip leading non-matches by incrementing cstart */ |
| 3109 | while (cstart < cend && ! REF_IN_CHARSET (s, cstart, char_pred)) |
| 3110 | cstart++; |
| 3111 | |
| 3112 | /* strip trailing non-matches by decrementing cend */ |
| 3113 | while (cend > cstart && ! REF_IN_CHARSET (s, cend-1, char_pred)) |
| 3114 | cend--; |
| 3115 | |
| 3116 | /* count chars to be kept */ |
| 3117 | count = 0; |
| 3118 | for (idx = cstart; idx < cend; idx++) |
| 3119 | if (REF_IN_CHARSET (s, idx, char_pred)) |
| 3120 | count++; |
| 3121 | |
| 3122 | /* if whole of start to end kept then return substring */ |
| 3123 | if (count == cend - cstart) |
| 3124 | goto result_substring; |
| 3125 | else |
| 3126 | { |
| 3127 | size_t dst = 0; |
| 3128 | result = scm_i_make_string (count, NULL); |
| 3129 | result = scm_i_string_start_writing (result); |
| 3130 | |
| 3131 | /* decrement "count" in this loop as well as using idx, so that if |
| 3132 | another thread is simultaneously changing "s" there's no chance |
| 3133 | it'll make us copy more than count characters */ |
| 3134 | for (idx = cstart; idx < cend && count != 0; idx++) |
| 3135 | { |
| 3136 | if (REF_IN_CHARSET (s, idx, char_pred)) |
| 3137 | { |
| 3138 | scm_i_string_set_x (result, dst, scm_i_string_ref (s, idx)); |
| 3139 | dst ++; |
| 3140 | count--; |
| 3141 | } |
| 3142 | } |
| 3143 | scm_i_string_stop_writing (); |
| 3144 | } |
| 3145 | } |
| 3146 | else |
| 3147 | { |
| 3148 | SCM ls = SCM_EOL; |
| 3149 | |
| 3150 | SCM_ASSERT (scm_is_true (scm_procedure_p (char_pred)), |
| 3151 | char_pred, SCM_ARG1, FUNC_NAME); |
| 3152 | idx = cstart; |
| 3153 | while (idx < cend) |
| 3154 | { |
| 3155 | SCM res, ch; |
| 3156 | ch = SCM_MAKE_CHAR (scm_i_string_ref (s, idx)); |
| 3157 | res = scm_call_1 (char_pred, ch); |
| 3158 | if (scm_is_true (res)) |
| 3159 | ls = scm_cons (ch, ls); |
| 3160 | idx++; |
| 3161 | } |
| 3162 | result = scm_reverse_list_to_string (ls); |
| 3163 | } |
| 3164 | |
| 3165 | scm_remember_upto_here_1 (s); |
| 3166 | return result; |
| 3167 | } |
| 3168 | #undef FUNC_NAME |
| 3169 | |
| 3170 | |
| 3171 | SCM_DEFINE (scm_string_delete, "string-delete", 2, 2, 0, |
| 3172 | (SCM char_pred, SCM s, SCM start, SCM end), |
| 3173 | "Delete characters satisfying @var{char_pred} from @var{s}.\n" |
| 3174 | "\n" |
| 3175 | "If @var{char_pred} is a procedure, it is applied to each\n" |
| 3176 | "character as a predicate, if it is a character, it is tested\n" |
| 3177 | "for equality and if it is a character set, it is tested for\n" |
| 3178 | "membership.") |
| 3179 | #define FUNC_NAME s_scm_string_delete |
| 3180 | { |
| 3181 | size_t cstart, cend; |
| 3182 | SCM result; |
| 3183 | size_t idx; |
| 3184 | |
| 3185 | if (scm_is_string (char_pred)) |
| 3186 | { |
| 3187 | SCM tmp; |
| 3188 | |
| 3189 | scm_c_issue_deprecation_warning |
| 3190 | ("Guile used to use the wrong argument order for string-delete.\n" |
| 3191 | "This call to string-filter had the arguments in the wrong order.\n" |
| 3192 | "See SRFI-13 for more details. At some point we will remove this hack."); |
| 3193 | |
| 3194 | tmp = char_pred; |
| 3195 | char_pred = s; |
| 3196 | s = tmp; |
| 3197 | } |
| 3198 | |
| 3199 | MY_VALIDATE_SUBSTRING_SPEC (2, s, |
| 3200 | 3, start, cstart, |
| 3201 | 4, end, cend); |
| 3202 | |
| 3203 | /* The explicit loops below stripping leading and trailing matches mean we |
| 3204 | can return a substring if those are the only deletions, making |
| 3205 | string-delete as efficient as string-trim-both in that case. */ |
| 3206 | |
| 3207 | if (SCM_CHARP (char_pred)) |
| 3208 | { |
| 3209 | size_t count; |
| 3210 | |
| 3211 | /* strip leading matches by incrementing cstart */ |
| 3212 | while (cstart < cend && scm_i_string_ref (s, cstart) == SCM_CHAR(char_pred)) |
| 3213 | cstart++; |
| 3214 | |
| 3215 | /* strip trailing matches by decrementing cend */ |
| 3216 | while (cend > cstart && scm_i_string_ref (s, cend-1) == SCM_CHAR (char_pred)) |
| 3217 | cend--; |
| 3218 | |
| 3219 | /* count chars to be kept */ |
| 3220 | count = 0; |
| 3221 | for (idx = cstart; idx < cend; idx++) |
| 3222 | if (scm_i_string_ref (s, idx) != SCM_CHAR (char_pred)) |
| 3223 | count++; |
| 3224 | |
| 3225 | if (count == cend - cstart) |
| 3226 | { |
| 3227 | /* whole of cstart to cend is to be kept, return a copy-on-write |
| 3228 | substring */ |
| 3229 | result_substring: |
| 3230 | result = scm_i_substring (s, cstart, cend); |
| 3231 | } |
| 3232 | else |
| 3233 | { |
| 3234 | int i = 0; |
| 3235 | /* new string for retained portion */ |
| 3236 | result = scm_i_make_string (count, NULL); |
| 3237 | result = scm_i_string_start_writing (result); |
| 3238 | /* decrement "count" in this loop as well as using idx, so that if |
| 3239 | another thread is simultaneously changing "s" there's no chance |
| 3240 | it'll make us copy more than count characters */ |
| 3241 | for (idx = cstart; idx < cend && count != 0; idx++) |
| 3242 | { |
| 3243 | scm_t_wchar c = scm_i_string_ref (s, idx); |
| 3244 | if (c != SCM_CHAR (char_pred)) |
| 3245 | { |
| 3246 | scm_i_string_set_x (result, i, c); |
| 3247 | i++; |
| 3248 | count--; |
| 3249 | } |
| 3250 | } |
| 3251 | scm_i_string_stop_writing (); |
| 3252 | } |
| 3253 | } |
| 3254 | else if (SCM_CHARSETP (char_pred)) |
| 3255 | { |
| 3256 | size_t count; |
| 3257 | |
| 3258 | /* strip leading matches by incrementing cstart */ |
| 3259 | while (cstart < cend && REF_IN_CHARSET (s, cstart, char_pred)) |
| 3260 | cstart++; |
| 3261 | |
| 3262 | /* strip trailing matches by decrementing cend */ |
| 3263 | while (cend > cstart && REF_IN_CHARSET (s, cend-1, char_pred)) |
| 3264 | cend--; |
| 3265 | |
| 3266 | /* count chars to be kept */ |
| 3267 | count = 0; |
| 3268 | for (idx = cstart; idx < cend; idx++) |
| 3269 | if (!REF_IN_CHARSET (s, idx, char_pred)) |
| 3270 | count++; |
| 3271 | |
| 3272 | if (count == cend - cstart) |
| 3273 | goto result_substring; |
| 3274 | else |
| 3275 | { |
| 3276 | size_t i = 0; |
| 3277 | /* new string for retained portion */ |
| 3278 | result = scm_i_make_string (count, NULL); |
| 3279 | result = scm_i_string_start_writing (result); |
| 3280 | |
| 3281 | /* decrement "count" in this loop as well as using idx, so that if |
| 3282 | another thread is simultaneously changing "s" there's no chance |
| 3283 | it'll make us copy more than count characters */ |
| 3284 | for (idx = cstart; idx < cend && count != 0; idx++) |
| 3285 | { |
| 3286 | if (!REF_IN_CHARSET (s, idx, char_pred)) |
| 3287 | { |
| 3288 | scm_i_string_set_x (result, i, scm_i_string_ref (s, idx)); |
| 3289 | i++; |
| 3290 | count--; |
| 3291 | } |
| 3292 | } |
| 3293 | scm_i_string_stop_writing (); |
| 3294 | } |
| 3295 | } |
| 3296 | else |
| 3297 | { |
| 3298 | SCM ls = SCM_EOL; |
| 3299 | SCM_ASSERT (scm_is_true (scm_procedure_p (char_pred)), |
| 3300 | char_pred, SCM_ARG1, FUNC_NAME); |
| 3301 | |
| 3302 | idx = cstart; |
| 3303 | while (idx < cend) |
| 3304 | { |
| 3305 | SCM res, ch = SCM_MAKE_CHAR (scm_i_string_ref (s, idx)); |
| 3306 | res = scm_call_1 (char_pred, ch); |
| 3307 | if (scm_is_false (res)) |
| 3308 | ls = scm_cons (ch, ls); |
| 3309 | idx++; |
| 3310 | } |
| 3311 | result = scm_reverse_list_to_string (ls); |
| 3312 | } |
| 3313 | |
| 3314 | scm_remember_upto_here_1 (s); |
| 3315 | return result; |
| 3316 | } |
| 3317 | #undef FUNC_NAME |
| 3318 | |
| 3319 | void |
| 3320 | scm_init_srfi_13 (void) |
| 3321 | { |
| 3322 | #include "libguile/srfi-13.x" |
| 3323 | } |
| 3324 | |
| 3325 | /* End of srfi-13.c. */ |