1 /* $NetBSD: tokenizer.c,v 1.21 2011/08/16 16:25:15 christos Exp $ */ 2 3 /*- 4 * Copyright (c) 1992, 1993 5 * The Regents of the University of California. All rights reserved. 6 * 7 * This code is derived from software contributed to Berkeley by 8 * Christos Zoulas of Cornell University. 9 * 10 * Redistribution and use in source and binary forms, with or without 11 * modification, are permitted provided that the following conditions 12 * are met: 13 * 1. Redistributions of source code must retain the above copyright 14 * notice, this list of conditions and the following disclaimer. 15 * 2. Redistributions in binary form must reproduce the above copyright 16 * notice, this list of conditions and the following disclaimer in the 17 * documentation and/or other materials provided with the distribution. 18 * 3. Neither the name of the University nor the names of its contributors 19 * may be used to endorse or promote products derived from this software 20 * without specific prior written permission. 21 * 22 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 23 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 24 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 25 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 26 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 27 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 28 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 29 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 30 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 31 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 32 * SUCH DAMAGE. 33 */ 34 35 #include "config.h" 36 #if !defined(lint) && !defined(SCCSID) 37 #if 0 38 static char sccsid[] = "@(#)tokenizer.c 8.1 (Berkeley) 6/4/93"; 39 #else 40 __RCSID("$NetBSD: tokenizer.c,v 1.21 2011/08/16 16:25:15 christos Exp $"); 41 #endif 42 #endif /* not lint && not SCCSID */ 43 44 /* We build this file twice, once as NARROW, once as WIDE. */ 45 /* 46 * tokenize.c: Bourne shell like tokenizer 47 */ 48 #include <string.h> 49 #include <stdlib.h> 50 #include "histedit.h" 51 #include "chartype.h" 52 53 typedef enum { 54 Q_none, Q_single, Q_double, Q_one, Q_doubleone 55 } quote_t; 56 57 #define TOK_KEEP 1 58 #define TOK_EAT 2 59 60 #define WINCR 20 61 #define AINCR 10 62 63 #define IFS STR("\t \n") 64 65 #define tok_malloc(a) malloc(a) 66 #define tok_free(a) free(a) 67 #define tok_realloc(a, b) realloc(a, b) 68 #define tok_strdup(a) Strdup(a) 69 70 71 struct TYPE(tokenizer) { 72 Char *ifs; /* In field separator */ 73 size_t argc, amax; /* Current and maximum number of args */ 74 Char **argv; /* Argument list */ 75 Char *wptr, *wmax; /* Space and limit on the word buffer */ 76 Char *wstart; /* Beginning of next word */ 77 Char *wspace; /* Space of word buffer */ 78 quote_t quote; /* Quoting state */ 79 int flags; /* flags; */ 80 }; 81 82 83 private void FUN(tok,finish)(TYPE(Tokenizer) *); 84 85 86 /* FUN(tok,finish)(): 87 * Finish a word in the tokenizer. 88 */ 89 private void 90 FUN(tok,finish)(TYPE(Tokenizer) *tok) 91 { 92 93 *tok->wptr = '\0'; 94 if ((tok->flags & TOK_KEEP) || tok->wptr != tok->wstart) { 95 tok->argv[tok->argc++] = tok->wstart; 96 tok->argv[tok->argc] = NULL; 97 tok->wstart = ++tok->wptr; 98 } 99 tok->flags &= ~TOK_KEEP; 100 } 101 102 103 /* FUN(tok,init)(): 104 * Initialize the tokenizer 105 */ 106 public TYPE(Tokenizer) * 107 FUN(tok,init)(const Char *ifs) 108 { 109 TYPE(Tokenizer) *tok = tok_malloc(sizeof(*tok)); 110 111 if (tok == NULL) 112 return NULL; 113 tok->ifs = tok_strdup(ifs ? ifs : IFS); 114 if (tok->ifs == NULL) { 115 tok_free(tok); 116 return NULL; 117 } 118 tok->argc = 0; 119 tok->amax = AINCR; 120 tok->argv = tok_malloc(sizeof(*tok->argv) * tok->amax); 121 if (tok->argv == NULL) { 122 tok_free(tok->ifs); 123 tok_free(tok); 124 return NULL; 125 } 126 tok->argv[0] = NULL; 127 tok->wspace = tok_malloc(WINCR * sizeof(*tok->wspace)); 128 if (tok->wspace == NULL) { 129 tok_free(tok->argv); 130 tok_free(tok->ifs); 131 tok_free(tok); 132 return NULL; 133 } 134 tok->wmax = tok->wspace + WINCR; 135 tok->wstart = tok->wspace; 136 tok->wptr = tok->wspace; 137 tok->flags = 0; 138 tok->quote = Q_none; 139 140 return tok; 141 } 142 143 144 /* FUN(tok,reset)(): 145 * Reset the tokenizer 146 */ 147 public void 148 FUN(tok,reset)(TYPE(Tokenizer) *tok) 149 { 150 151 tok->argc = 0; 152 tok->wstart = tok->wspace; 153 tok->wptr = tok->wspace; 154 tok->flags = 0; 155 tok->quote = Q_none; 156 } 157 158 159 /* FUN(tok,end)(): 160 * Clean up 161 */ 162 public void 163 FUN(tok,end)(TYPE(Tokenizer) *tok) 164 { 165 166 tok_free(tok->ifs); 167 tok_free(tok->wspace); 168 tok_free(tok->argv); 169 tok_free(tok); 170 } 171 172 173 174 /* FUN(tok,line)(): 175 * Bourne shell (sh(1)) like tokenizing 176 * Arguments: 177 * tok current tokenizer state (setup with FUN(tok,init)()) 178 * line line to parse 179 * Returns: 180 * -1 Internal error 181 * 3 Quoted return 182 * 2 Unmatched double quote 183 * 1 Unmatched single quote 184 * 0 Ok 185 * Modifies (if return value is 0): 186 * argc number of arguments 187 * argv argument array 188 * cursorc if !NULL, argv element containing cursor 189 * cursorv if !NULL, offset in argv[cursorc] of cursor 190 */ 191 public int 192 FUN(tok,line)(TYPE(Tokenizer) *tok, const TYPE(LineInfo) *line, 193 int *argc, const Char ***argv, int *cursorc, int *cursoro) 194 { 195 const Char *ptr; 196 int cc, co; 197 198 cc = co = -1; 199 ptr = line->buffer; 200 for (ptr = line->buffer; ;ptr++) { 201 if (ptr >= line->lastchar) 202 ptr = STR(""); 203 if (ptr == line->cursor) { 204 cc = (int)tok->argc; 205 co = (int)(tok->wptr - tok->wstart); 206 } 207 switch (*ptr) { 208 case '\'': 209 tok->flags |= TOK_KEEP; 210 tok->flags &= ~TOK_EAT; 211 switch (tok->quote) { 212 case Q_none: 213 tok->quote = Q_single; /* Enter single quote 214 * mode */ 215 break; 216 217 case Q_single: /* Exit single quote mode */ 218 tok->quote = Q_none; 219 break; 220 221 case Q_one: /* Quote this ' */ 222 tok->quote = Q_none; 223 *tok->wptr++ = *ptr; 224 break; 225 226 case Q_double: /* Stay in double quote mode */ 227 *tok->wptr++ = *ptr; 228 break; 229 230 case Q_doubleone: /* Quote this ' */ 231 tok->quote = Q_double; 232 *tok->wptr++ = *ptr; 233 break; 234 235 default: 236 return -1; 237 } 238 break; 239 240 case '"': 241 tok->flags &= ~TOK_EAT; 242 tok->flags |= TOK_KEEP; 243 switch (tok->quote) { 244 case Q_none: /* Enter double quote mode */ 245 tok->quote = Q_double; 246 break; 247 248 case Q_double: /* Exit double quote mode */ 249 tok->quote = Q_none; 250 break; 251 252 case Q_one: /* Quote this " */ 253 tok->quote = Q_none; 254 *tok->wptr++ = *ptr; 255 break; 256 257 case Q_single: /* Stay in single quote mode */ 258 *tok->wptr++ = *ptr; 259 break; 260 261 case Q_doubleone: /* Quote this " */ 262 tok->quote = Q_double; 263 *tok->wptr++ = *ptr; 264 break; 265 266 default: 267 return -1; 268 } 269 break; 270 271 case '\\': 272 tok->flags |= TOK_KEEP; 273 tok->flags &= ~TOK_EAT; 274 switch (tok->quote) { 275 case Q_none: /* Quote next character */ 276 tok->quote = Q_one; 277 break; 278 279 case Q_double: /* Quote next character */ 280 tok->quote = Q_doubleone; 281 break; 282 283 case Q_one: /* Quote this, restore state */ 284 *tok->wptr++ = *ptr; 285 tok->quote = Q_none; 286 break; 287 288 case Q_single: /* Stay in single quote mode */ 289 *tok->wptr++ = *ptr; 290 break; 291 292 case Q_doubleone: /* Quote this \ */ 293 tok->quote = Q_double; 294 *tok->wptr++ = *ptr; 295 break; 296 297 default: 298 return -1; 299 } 300 break; 301 302 case '\n': 303 tok->flags &= ~TOK_EAT; 304 switch (tok->quote) { 305 case Q_none: 306 goto tok_line_outok; 307 308 case Q_single: 309 case Q_double: 310 *tok->wptr++ = *ptr; /* Add the return */ 311 break; 312 313 case Q_doubleone: /* Back to double, eat the '\n' */ 314 tok->flags |= TOK_EAT; 315 tok->quote = Q_double; 316 break; 317 318 case Q_one: /* No quote, more eat the '\n' */ 319 tok->flags |= TOK_EAT; 320 tok->quote = Q_none; 321 break; 322 323 default: 324 return 0; 325 } 326 break; 327 328 case '\0': 329 switch (tok->quote) { 330 case Q_none: 331 /* Finish word and return */ 332 if (tok->flags & TOK_EAT) { 333 tok->flags &= ~TOK_EAT; 334 return 3; 335 } 336 goto tok_line_outok; 337 338 case Q_single: 339 return 1; 340 341 case Q_double: 342 return 2; 343 344 case Q_doubleone: 345 tok->quote = Q_double; 346 *tok->wptr++ = *ptr; 347 break; 348 349 case Q_one: 350 tok->quote = Q_none; 351 *tok->wptr++ = *ptr; 352 break; 353 354 default: 355 return -1; 356 } 357 break; 358 359 default: 360 tok->flags &= ~TOK_EAT; 361 switch (tok->quote) { 362 case Q_none: 363 if (Strchr(tok->ifs, *ptr) != NULL) 364 FUN(tok,finish)(tok); 365 else 366 *tok->wptr++ = *ptr; 367 break; 368 369 case Q_single: 370 case Q_double: 371 *tok->wptr++ = *ptr; 372 break; 373 374 375 case Q_doubleone: 376 *tok->wptr++ = '\\'; 377 tok->quote = Q_double; 378 *tok->wptr++ = *ptr; 379 break; 380 381 case Q_one: 382 tok->quote = Q_none; 383 *tok->wptr++ = *ptr; 384 break; 385 386 default: 387 return -1; 388 389 } 390 break; 391 } 392 393 if (tok->wptr >= tok->wmax - 4) { 394 size_t size = (size_t)(tok->wmax - tok->wspace + WINCR); 395 Char *s = tok_realloc(tok->wspace, 396 size * sizeof(*s)); 397 if (s == NULL) 398 return -1; 399 400 if (s != tok->wspace) { 401 size_t i; 402 for (i = 0; i < tok->argc; i++) { 403 tok->argv[i] = 404 (tok->argv[i] - tok->wspace) + s; 405 } 406 tok->wptr = (tok->wptr - tok->wspace) + s; 407 tok->wstart = (tok->wstart - tok->wspace) + s; 408 tok->wspace = s; 409 } 410 tok->wmax = s + size; 411 } 412 if (tok->argc >= tok->amax - 4) { 413 Char **p; 414 tok->amax += AINCR; 415 p = tok_realloc(tok->argv, tok->amax * sizeof(*p)); 416 if (p == NULL) 417 return -1; 418 tok->argv = p; 419 } 420 } 421 tok_line_outok: 422 if (cc == -1 && co == -1) { 423 cc = (int)tok->argc; 424 co = (int)(tok->wptr - tok->wstart); 425 } 426 if (cursorc != NULL) 427 *cursorc = cc; 428 if (cursoro != NULL) 429 *cursoro = co; 430 FUN(tok,finish)(tok); 431 *argv = (const Char **)tok->argv; 432 *argc = (int)tok->argc; 433 return 0; 434 } 435 436 /* FUN(tok,str)(): 437 * Simpler version of tok_line, taking a NUL terminated line 438 * and splitting into words, ignoring cursor state. 439 */ 440 public int 441 FUN(tok,str)(TYPE(Tokenizer) *tok, const Char *line, int *argc, 442 const Char ***argv) 443 { 444 TYPE(LineInfo) li; 445 446 memset(&li, 0, sizeof(li)); 447 li.buffer = line; 448 li.cursor = li.lastchar = Strchr(line, '\0'); 449 return FUN(tok,line(tok, &li, argc, argv, NULL, NULL)); 450 } 451