lex.h revision 280849
1/* 2 * Copyright (C) 2004, 2005, 2007, 2008 Internet Systems Consortium, Inc. ("ISC") 3 * Copyright (C) 1998-2002 Internet Software Consortium. 4 * 5 * Permission to use, copy, modify, and/or distribute this software for any 6 * purpose with or without fee is hereby granted, provided that the above 7 * copyright notice and this permission notice appear in all copies. 8 * 9 * THE SOFTWARE IS PROVIDED "AS IS" AND ISC DISCLAIMS ALL WARRANTIES WITH 10 * REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY 11 * AND FITNESS. IN NO EVENT SHALL ISC BE LIABLE FOR ANY SPECIAL, DIRECT, 12 * INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM 13 * LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE 14 * OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR 15 * PERFORMANCE OF THIS SOFTWARE. 16 */ 17 18/* $Id: lex.h,v 1.37 2008/05/30 23:47:01 tbox Exp $ */ 19 20#ifndef ISC_LEX_H 21#define ISC_LEX_H 1 22 23/***** 24 ***** Module Info 25 *****/ 26 27/*! \file isc/lex.h 28 * \brief The "lex" module provides a lightweight tokenizer. It can operate 29 * on files or buffers, and can handle "include". It is designed for 30 * parsing of DNS master files and the BIND configuration file, but 31 * should be general enough to tokenize other things, e.g. HTTP. 32 * 33 * \li MP: 34 * No synchronization is provided. Clients must ensure exclusive 35 * access. 36 * 37 * \li Reliability: 38 * No anticipated impact. 39 * 40 * \li Resources: 41 * TBS 42 * 43 * \li Security: 44 * No anticipated impact. 45 * 46 * \li Standards: 47 * None. 48 */ 49 50/*** 51 *** Imports 52 ***/ 53 54#include <stdio.h> 55 56#include <isc/lang.h> 57#include <isc/region.h> 58#include <isc/types.h> 59 60ISC_LANG_BEGINDECLS 61 62/*** 63 *** Options 64 ***/ 65 66/*@{*/ 67/*! 68 * Various options for isc_lex_gettoken(). 69 */ 70 71#define ISC_LEXOPT_EOL 0x01 /*%< Want end-of-line token. */ 72#define ISC_LEXOPT_EOF 0x02 /*%< Want end-of-file token. */ 73#define ISC_LEXOPT_INITIALWS 0x04 /*%< Want initial whitespace. */ 74#define ISC_LEXOPT_NUMBER 0x08 /*%< Recognize numbers. */ 75#define ISC_LEXOPT_QSTRING 0x10 /*%< Recognize qstrings. */ 76/*@}*/ 77 78/*@{*/ 79/*! 80 * The ISC_LEXOPT_DNSMULTILINE option handles the processing of '(' and ')' in 81 * the DNS master file format. If this option is set, then the 82 * ISC_LEXOPT_INITIALWS and ISC_LEXOPT_EOL options will be ignored when 83 * the paren count is > 0. To use this option, '(' and ')' must be special 84 * characters. 85 */ 86#define ISC_LEXOPT_DNSMULTILINE 0x20 /*%< Handle '(' and ')'. */ 87#define ISC_LEXOPT_NOMORE 0x40 /*%< Want "no more" token. */ 88 89#define ISC_LEXOPT_CNUMBER 0x80 /*%< Recognize octal and hex. */ 90#define ISC_LEXOPT_ESCAPE 0x100 /*%< Recognize escapes. */ 91#define ISC_LEXOPT_QSTRINGMULTILINE 0x200 /*%< Allow multiline "" strings */ 92#define ISC_LEXOPT_OCTAL 0x400 /*%< Expect a octal number. */ 93/*@}*/ 94/*@{*/ 95/*! 96 * Various commenting styles, which may be changed at any time with 97 * isc_lex_setcomments(). 98 */ 99 100#define ISC_LEXCOMMENT_C 0x01 101#define ISC_LEXCOMMENT_CPLUSPLUS 0x02 102#define ISC_LEXCOMMENT_SHELL 0x04 103#define ISC_LEXCOMMENT_DNSMASTERFILE 0x08 104/*@}*/ 105 106/*** 107 *** Types 108 ***/ 109 110/*! Lex */ 111 112typedef char isc_lexspecials_t[256]; 113 114/* Tokens */ 115 116typedef enum { 117 isc_tokentype_unknown = 0, 118 isc_tokentype_string = 1, 119 isc_tokentype_number = 2, 120 isc_tokentype_qstring = 3, 121 isc_tokentype_eol = 4, 122 isc_tokentype_eof = 5, 123 isc_tokentype_initialws = 6, 124 isc_tokentype_special = 7, 125 isc_tokentype_nomore = 8 126} isc_tokentype_t; 127 128typedef union { 129 char as_char; 130 unsigned long as_ulong; 131 isc_region_t as_region; 132 isc_textregion_t as_textregion; 133 void * as_pointer; 134} isc_tokenvalue_t; 135 136typedef struct isc_token { 137 isc_tokentype_t type; 138 isc_tokenvalue_t value; 139} isc_token_t; 140 141/*** 142 *** Functions 143 ***/ 144 145isc_result_t 146isc_lex_create(isc_mem_t *mctx, size_t max_token, isc_lex_t **lexp); 147/*%< 148 * Create a lexer. 149 * 150 * 'max_token' is a hint of the number of bytes in the largest token. 151 * 152 * Requires: 153 *\li '*lexp' is a valid lexer. 154 * 155 *\li max_token > 0. 156 * 157 * Ensures: 158 *\li On success, *lexp is attached to the newly created lexer. 159 * 160 * Returns: 161 *\li #ISC_R_SUCCESS 162 *\li #ISC_R_NOMEMORY 163 */ 164 165void 166isc_lex_destroy(isc_lex_t **lexp); 167/*%< 168 * Destroy the lexer. 169 * 170 * Requires: 171 *\li '*lexp' is a valid lexer. 172 * 173 * Ensures: 174 *\li *lexp == NULL 175 */ 176 177unsigned int 178isc_lex_getcomments(isc_lex_t *lex); 179/*%< 180 * Return the current lexer commenting styles. 181 * 182 * Requires: 183 *\li 'lex' is a valid lexer. 184 * 185 * Returns: 186 *\li The commenting sytles which are currently allowed. 187 */ 188 189void 190isc_lex_setcomments(isc_lex_t *lex, unsigned int comments); 191/*%< 192 * Set allowed lexer commenting styles. 193 * 194 * Requires: 195 *\li 'lex' is a valid lexer. 196 * 197 *\li 'comments' has meaningful values. 198 */ 199 200void 201isc_lex_getspecials(isc_lex_t *lex, isc_lexspecials_t specials); 202/*%< 203 * Put the current list of specials into 'specials'. 204 * 205 * Requires: 206 *\li 'lex' is a valid lexer. 207 */ 208 209void 210isc_lex_setspecials(isc_lex_t *lex, isc_lexspecials_t specials); 211/*!< 212 * The characters in 'specials' are returned as tokens. Along with 213 * whitespace, they delimit strings and numbers. 214 * 215 * Note: 216 *\li Comment processing takes precedence over special character 217 * recognition. 218 * 219 * Requires: 220 *\li 'lex' is a valid lexer. 221 */ 222 223isc_result_t 224isc_lex_openfile(isc_lex_t *lex, const char *filename); 225/*%< 226 * Open 'filename' and make it the current input source for 'lex'. 227 * 228 * Requires: 229 *\li 'lex' is a valid lexer. 230 * 231 *\li filename is a valid C string. 232 * 233 * Returns: 234 *\li #ISC_R_SUCCESS 235 *\li #ISC_R_NOMEMORY Out of memory 236 *\li #ISC_R_NOTFOUND File not found 237 *\li #ISC_R_NOPERM No permission to open file 238 *\li #ISC_R_FAILURE Couldn't open file, not sure why 239 *\li #ISC_R_UNEXPECTED 240 */ 241 242isc_result_t 243isc_lex_openstream(isc_lex_t *lex, FILE *stream); 244/*%< 245 * Make 'stream' the current input source for 'lex'. 246 * 247 * Requires: 248 *\li 'lex' is a valid lexer. 249 * 250 *\li 'stream' is a valid C stream. 251 * 252 * Returns: 253 *\li #ISC_R_SUCCESS 254 *\li #ISC_R_NOMEMORY Out of memory 255 */ 256 257isc_result_t 258isc_lex_openbuffer(isc_lex_t *lex, isc_buffer_t *buffer); 259/*%< 260 * Make 'buffer' the current input source for 'lex'. 261 * 262 * Requires: 263 *\li 'lex' is a valid lexer. 264 * 265 *\li 'buffer' is a valid buffer. 266 * 267 * Returns: 268 *\li #ISC_R_SUCCESS 269 *\li #ISC_R_NOMEMORY Out of memory 270 */ 271 272isc_result_t 273isc_lex_close(isc_lex_t *lex); 274/*%< 275 * Close the most recently opened object (i.e. file or buffer). 276 * 277 * Returns: 278 *\li #ISC_R_SUCCESS 279 *\li #ISC_R_NOMORE No more input sources 280 */ 281 282isc_result_t 283isc_lex_gettoken(isc_lex_t *lex, unsigned int options, isc_token_t *tokenp); 284/*%< 285 * Get the next token. 286 * 287 * Requires: 288 *\li 'lex' is a valid lexer. 289 * 290 *\li 'lex' has an input source. 291 * 292 *\li 'options' contains valid options. 293 * 294 *\li '*tokenp' is a valid pointer. 295 * 296 * Returns: 297 *\li #ISC_R_SUCCESS 298 *\li #ISC_R_UNEXPECTEDEND 299 *\li #ISC_R_NOMEMORY 300 * 301 * These two results are returned only if their corresponding lexer 302 * options are not set. 303 * 304 *\li #ISC_R_EOF End of input source 305 *\li #ISC_R_NOMORE No more input sources 306 */ 307 308isc_result_t 309isc_lex_getmastertoken(isc_lex_t *lex, isc_token_t *token, 310 isc_tokentype_t expect, isc_boolean_t eol); 311/*%< 312 * Get the next token from a DNS master file type stream. This is a 313 * convenience function that sets appropriate options and handles quoted 314 * strings and end of line correctly for master files. It also ungets 315 * unexpected tokens. 316 * 317 * Requires: 318 *\li 'lex' is a valid lexer. 319 * 320 *\li 'token' is a valid pointer 321 * 322 * Returns: 323 * 324 * \li any return code from isc_lex_gettoken(). 325 */ 326 327isc_result_t 328isc_lex_getoctaltoken(isc_lex_t *lex, isc_token_t *token, isc_boolean_t eol); 329/*%< 330 * Get the next token from a DNS master file type stream. This is a 331 * convenience function that sets appropriate options and handles end 332 * of line correctly for master files. It also ungets unexpected tokens. 333 * 334 * Requires: 335 *\li 'lex' is a valid lexer. 336 * 337 *\li 'token' is a valid pointer 338 * 339 * Returns: 340 * 341 * \li any return code from isc_lex_gettoken(). 342 */ 343 344void 345isc_lex_ungettoken(isc_lex_t *lex, isc_token_t *tokenp); 346/*%< 347 * Unget the current token. 348 * 349 * Requires: 350 *\li 'lex' is a valid lexer. 351 * 352 *\li 'lex' has an input source. 353 * 354 *\li 'tokenp' points to a valid token. 355 * 356 *\li There is no ungotten token already. 357 */ 358 359void 360isc_lex_getlasttokentext(isc_lex_t *lex, isc_token_t *tokenp, isc_region_t *r); 361/*%< 362 * Returns a region containing the text of the last token returned. 363 * 364 * Requires: 365 *\li 'lex' is a valid lexer. 366 * 367 *\li 'lex' has an input source. 368 * 369 *\li 'tokenp' points to a valid token. 370 * 371 *\li A token has been gotten and not ungotten. 372 */ 373 374char * 375isc_lex_getsourcename(isc_lex_t *lex); 376/*%< 377 * Return the input source name. 378 * 379 * Requires: 380 *\li 'lex' is a valid lexer. 381 * 382 * Returns: 383 * \li source name or NULL if no current source. 384 *\li result valid while current input source exists. 385 */ 386 387 388unsigned long 389isc_lex_getsourceline(isc_lex_t *lex); 390/*%< 391 * Return the input source line number. 392 * 393 * Requires: 394 *\li 'lex' is a valid lexer. 395 * 396 * Returns: 397 *\li Current line number or 0 if no current source. 398 */ 399 400isc_result_t 401isc_lex_setsourcename(isc_lex_t *lex, const char *name); 402/*%< 403 * Assigns a new name to the input source. 404 * 405 * Requires: 406 * 407 * \li 'lex' is a valid lexer. 408 * 409 * Returns: 410 * \li #ISC_R_SUCCESS 411 * \li #ISC_R_NOMEMORY 412 * \li #ISC_R_NOTFOUND - there are no sources. 413 */ 414 415isc_boolean_t 416isc_lex_isfile(isc_lex_t *lex); 417/*%< 418 * Return whether the current input source is a file. 419 * 420 * Requires: 421 *\li 'lex' is a valid lexer. 422 * 423 * Returns: 424 * \li #ISC_TRUE if the current input is a file, 425 *\li #ISC_FALSE otherwise. 426 */ 427 428 429ISC_LANG_ENDDECLS 430 431#endif /* ISC_LEX_H */ 432