1 /* Copyright (c) 2013, Vsevolod Stakhov 2 * All rights reserved. 3 * 4 * Redistribution and use in source and binary forms, with or without 5 * modification, are permitted provided that the following conditions are met: 6 * * Redistributions of source code must retain the above copyright 7 * notice, this list of conditions and the following disclaimer. 8 * * Redistributions in binary form must reproduce the above copyright 9 * notice, this list of conditions and the following disclaimer in the 10 * documentation and/or other materials provided with the distribution. 11 * 12 * THIS SOFTWARE IS PROVIDED ''AS IS'' AND ANY 13 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 14 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE 15 * DISCLAIMED. IN NO EVENT SHALL AUTHOR BE LIABLE FOR ANY 16 * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES 17 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; 18 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND 19 * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 20 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS 21 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 22 */ 23 24 #ifndef UCL_INTERNAL_H_ 25 #define UCL_INTERNAL_H_ 26 27 #ifdef HAVE_CONFIG_H 28 #include "config.h" 29 #else 30 /* Help embedded builds */ 31 #define HAVE_SYS_TYPES_H 32 #define HAVE_SYS_MMAN_H 33 #define HAVE_SYS_STAT_H 34 #define HAVE_SYS_PARAM_H 35 #define HAVE_LIMITS_H 36 #define HAVE_FCNTL_H 37 #define HAVE_ERRNO_H 38 #define HAVE_UNISTD_H 39 #define HAVE_CTYPE_H 40 #define HAVE_STDIO_H 41 #define HAVE_STRING_H 42 #define HAVE_FLOAT_H 43 #define HAVE_LIBGEN_H 44 #define HAVE_MATH_H 45 #define HAVE_STDBOOL_H 46 #define HAVE_STDINT_H 47 #define HAVE_STDARG_H 48 #ifndef _WIN32 49 # define HAVE_REGEX_H 50 #endif 51 #endif 52 53 #ifdef HAVE_SYS_TYPES_H 54 #include <sys/types.h> 55 #endif 56 57 #ifdef HAVE_SYS_MMAN_H 58 # ifndef _WIN32 59 # include <sys/mman.h> 60 # endif 61 #endif 62 #ifdef HAVE_SYS_STAT_H 63 #include <sys/stat.h> 64 #endif 65 #ifdef HAVE_SYS_PARAM_H 66 #include <sys/param.h> 67 #endif 68 69 #ifdef HAVE_LIMITS_H 70 #include <limits.h> 71 #endif 72 #ifdef HAVE_FCNTL_H 73 #include <fcntl.h> 74 #endif 75 #ifdef HAVE_ERRNO_H 76 #include <errno.h> 77 #endif 78 #ifdef HAVE_UNISTD_H 79 #include <unistd.h> 80 #endif 81 #ifdef HAVE_CTYPE_H 82 #include <ctype.h> 83 #endif 84 #ifdef HAVE_STDIO_H 85 #include <stdio.h> 86 #endif 87 #ifdef HAVE_STRING_H 88 #include <string.h> 89 #endif 90 91 #include "utlist.h" 92 #include "utstring.h" 93 #include "uthash.h" 94 #include "ucl.h" 95 #include "ucl_hash.h" 96 #include "xxhash.h" 97 98 #ifdef HAVE_OPENSSL 99 #include <openssl/evp.h> 100 #endif 101 102 #ifndef __DECONST 103 #define __DECONST(type, var) ((type)(uintptr_t)(const void *)(var)) 104 #endif 105 106 /** 107 * @file rcl_internal.h 108 * Internal structures and functions of UCL library 109 */ 110 111 #define UCL_MAX_RECURSION 16 112 #define UCL_TRASH_KEY 0 113 #define UCL_TRASH_VALUE 1 114 115 enum ucl_parser_state { 116 UCL_STATE_INIT = 0, 117 UCL_STATE_OBJECT, 118 UCL_STATE_ARRAY, 119 UCL_STATE_KEY, 120 UCL_STATE_VALUE, 121 UCL_STATE_AFTER_VALUE, 122 UCL_STATE_ARRAY_VALUE, 123 UCL_STATE_SCOMMENT, 124 UCL_STATE_MCOMMENT, 125 UCL_STATE_MACRO_NAME, 126 UCL_STATE_MACRO, 127 UCL_STATE_ERROR 128 }; 129 130 enum ucl_character_type { 131 UCL_CHARACTER_DENIED = 0, 132 UCL_CHARACTER_KEY = 1, 133 UCL_CHARACTER_KEY_START = 1 << 1, 134 UCL_CHARACTER_WHITESPACE = 1 << 2, 135 UCL_CHARACTER_WHITESPACE_UNSAFE = 1 << 3, 136 UCL_CHARACTER_VALUE_END = 1 << 4, 137 UCL_CHARACTER_VALUE_STR = 1 << 5, 138 UCL_CHARACTER_VALUE_DIGIT = 1 << 6, 139 UCL_CHARACTER_VALUE_DIGIT_START = 1 << 7, 140 UCL_CHARACTER_ESCAPE = 1 << 8, 141 UCL_CHARACTER_KEY_SEP = 1 << 9, 142 UCL_CHARACTER_JSON_UNSAFE = 1 << 10, 143 UCL_CHARACTER_UCL_UNSAFE = 1 << 11 144 }; 145 146 struct ucl_macro { 147 char *name; 148 union { 149 ucl_macro_handler handler; 150 ucl_context_macro_handler context_handler; 151 } h; 152 void* ud; 153 bool is_context; 154 UT_hash_handle hh; 155 }; 156 157 struct ucl_stack { 158 ucl_object_t *obj; 159 struct ucl_stack *next; 160 uint64_t level; 161 }; 162 163 struct ucl_chunk { 164 const unsigned char *begin; 165 const unsigned char *end; 166 const unsigned char *pos; 167 size_t remain; 168 unsigned int line; 169 unsigned int column; 170 unsigned priority; 171 enum ucl_duplicate_strategy strategy; 172 enum ucl_parse_type parse_type; 173 struct ucl_chunk *next; 174 }; 175 176 #ifdef HAVE_OPENSSL 177 struct ucl_pubkey { 178 EVP_PKEY *key; 179 struct ucl_pubkey *next; 180 }; 181 #else 182 struct ucl_pubkey { 183 struct ucl_pubkey *next; 184 }; 185 #endif 186 187 struct ucl_variable { 188 char *var; 189 char *value; 190 size_t var_len; 191 size_t value_len; 192 struct ucl_variable *prev, *next; 193 }; 194 195 struct ucl_parser { 196 enum ucl_parser_state state; 197 enum ucl_parser_state prev_state; 198 unsigned int recursion; 199 int flags; 200 unsigned default_priority; 201 int err_code; 202 ucl_object_t *top_obj; 203 ucl_object_t *cur_obj; 204 ucl_object_t *trash_objs; 205 ucl_object_t *includepaths; 206 char *cur_file; 207 struct ucl_macro *macroes; 208 struct ucl_stack *stack; 209 struct ucl_chunk *chunks; 210 struct ucl_pubkey *keys; 211 struct ucl_variable *variables; 212 ucl_variable_handler var_handler; 213 void *var_data; 214 UT_string *err; 215 }; 216 217 struct ucl_object_userdata { 218 ucl_object_t obj; 219 ucl_userdata_dtor dtor; 220 ucl_userdata_emitter emitter; 221 }; 222 223 /** 224 * Unescape json string inplace 225 * @param str 226 */ 227 size_t ucl_unescape_json_string (char *str, size_t len); 228 229 /** 230 * Handle include macro 231 * @param data include data 232 * @param len length of data 233 * @param args UCL object representing arguments to the macro 234 * @param ud user data 235 * @return 236 */ 237 bool ucl_include_handler (const unsigned char *data, size_t len, 238 const ucl_object_t *args, void* ud); 239 240 /** 241 * Handle tryinclude macro 242 * @param data include data 243 * @param len length of data 244 * @param args UCL object representing arguments to the macro 245 * @param ud user data 246 * @return 247 */ 248 bool ucl_try_include_handler (const unsigned char *data, size_t len, 249 const ucl_object_t *args, void* ud); 250 251 /** 252 * Handle includes macro 253 * @param data include data 254 * @param len length of data 255 * @param args UCL object representing arguments to the macro 256 * @param ud user data 257 * @return 258 */ 259 bool ucl_includes_handler (const unsigned char *data, size_t len, 260 const ucl_object_t *args, void* ud); 261 262 /** 263 * Handle priority macro 264 * @param data include data 265 * @param len length of data 266 * @param args UCL object representing arguments to the macro 267 * @param ud user data 268 * @return 269 */ 270 bool ucl_priority_handler (const unsigned char *data, size_t len, 271 const ucl_object_t *args, void* ud); 272 273 /** 274 * Handle load macro 275 * @param data include data 276 * @param len length of data 277 * @param args UCL object representing arguments to the macro 278 * @param ud user data 279 * @return 280 */ 281 bool ucl_load_handler (const unsigned char *data, size_t len, 282 const ucl_object_t *args, void* ud); 283 /** 284 * Handle inherit macro 285 * @param data include data 286 * @param len length of data 287 * @param args UCL object representing arguments to the macro 288 * @param ctx the current context object 289 * @param ud user data 290 * @return 291 */ 292 bool ucl_inherit_handler (const unsigned char *data, size_t len, 293 const ucl_object_t *args, const ucl_object_t *ctx, void* ud); 294 295 size_t ucl_strlcpy (char *dst, const char *src, size_t siz); 296 size_t ucl_strlcpy_unsafe (char *dst, const char *src, size_t siz); 297 size_t ucl_strlcpy_tolower (char *dst, const char *src, size_t siz); 298 299 char *ucl_strnstr (const char *s, const char *find, int len); 300 char *ucl_strncasestr (const char *s, const char *find, int len); 301 302 #ifdef __GNUC__ 303 static inline void 304 ucl_create_err (UT_string **err, const char *fmt, ...) 305 __attribute__ (( format( printf, 2, 3) )); 306 #endif 307 308 #undef UCL_FATAL_ERRORS 309 310 static inline void 311 ucl_create_err (UT_string **err, const char *fmt, ...) 312 { 313 if (*err == NULL) { 314 utstring_new (*err); 315 va_list ap; 316 va_start (ap, fmt); 317 utstring_printf_va (*err, fmt, ap); 318 va_end (ap); 319 } 320 321 #ifdef UCL_FATAL_ERRORS 322 assert (0); 323 #endif 324 } 325 326 /** 327 * Check whether a given string contains a boolean value 328 * @param obj object to set 329 * @param start start of a string 330 * @param len length of a string 331 * @return true if a string is a boolean value 332 */ 333 static inline bool 334 ucl_maybe_parse_boolean (ucl_object_t *obj, const unsigned char *start, size_t len) 335 { 336 const char *p = (const char *)start; 337 bool ret = false, val = false; 338 339 if (len == 5) { 340 if ((p[0] == 'f' || p[0] == 'F') && strncasecmp (p, "false", 5) == 0) { 341 ret = true; 342 val = false; 343 } 344 } 345 else if (len == 4) { 346 if ((p[0] == 't' || p[0] == 'T') && strncasecmp (p, "true", 4) == 0) { 347 ret = true; 348 val = true; 349 } 350 } 351 else if (len == 3) { 352 if ((p[0] == 'y' || p[0] == 'Y') && strncasecmp (p, "yes", 3) == 0) { 353 ret = true; 354 val = true; 355 } 356 else if ((p[0] == 'o' || p[0] == 'O') && strncasecmp (p, "off", 3) == 0) { 357 ret = true; 358 val = false; 359 } 360 } 361 else if (len == 2) { 362 if ((p[0] == 'n' || p[0] == 'N') && strncasecmp (p, "no", 2) == 0) { 363 ret = true; 364 val = false; 365 } 366 else if ((p[0] == 'o' || p[0] == 'O') && strncasecmp (p, "on", 2) == 0) { 367 ret = true; 368 val = true; 369 } 370 } 371 372 if (ret && obj != NULL) { 373 obj->type = UCL_BOOLEAN; 374 obj->value.iv = val; 375 } 376 377 return ret; 378 } 379 380 /** 381 * Check numeric string 382 * @param obj object to set if a string is numeric 383 * @param start start of string 384 * @param end end of string 385 * @param pos position where parsing has stopped 386 * @param allow_double allow parsing of floating point values 387 * @return 0 if string is numeric and error code (EINVAL or ERANGE) in case of conversion error 388 */ 389 int ucl_maybe_parse_number (ucl_object_t *obj, 390 const char *start, const char *end, const char **pos, 391 bool allow_double, bool number_bytes, bool allow_time); 392 393 394 static inline const ucl_object_t * 395 ucl_hash_search_obj (ucl_hash_t* hashlin, ucl_object_t *obj) 396 { 397 return (const ucl_object_t *)ucl_hash_search (hashlin, obj->key, obj->keylen); 398 } 399 400 static inline ucl_hash_t * ucl_hash_insert_object (ucl_hash_t *hashlin, 401 const ucl_object_t *obj, 402 bool ignore_case) UCL_WARN_UNUSED_RESULT; 403 404 static inline ucl_hash_t * 405 ucl_hash_insert_object (ucl_hash_t *hashlin, 406 const ucl_object_t *obj, 407 bool ignore_case) 408 { 409 if (hashlin == NULL) { 410 hashlin = ucl_hash_create (ignore_case); 411 } 412 ucl_hash_insert (hashlin, obj, obj->key, obj->keylen); 413 414 return hashlin; 415 } 416 417 /** 418 * Get standard emitter context for a specified emit_type 419 * @param emit_type type of emitter 420 * @return context or NULL if input is invalid 421 */ 422 const struct ucl_emitter_context * 423 ucl_emit_get_standard_context (enum ucl_emitter emit_type); 424 425 /** 426 * Serialize string as JSON string 427 * @param str string to emit 428 * @param buf target buffer 429 */ 430 void ucl_elt_string_write_json (const char *str, size_t size, 431 struct ucl_emitter_context *ctx); 432 433 /** 434 * Write multiline string using `EOD` as string terminator 435 * @param str 436 * @param size 437 * @param ctx 438 */ 439 void ucl_elt_string_write_multiline (const char *str, size_t size, 440 struct ucl_emitter_context *ctx); 441 442 /** 443 * Emit a single object to string 444 * @param obj 445 * @return 446 */ 447 unsigned char * ucl_object_emit_single_json (const ucl_object_t *obj); 448 449 /** 450 * Check whether a specified string is long and should be likely printed in 451 * multiline mode 452 * @param obj 453 * @return 454 */ 455 bool ucl_maybe_long_string (const ucl_object_t *obj); 456 457 /** 458 * Print integer to the msgpack output 459 * @param ctx 460 * @param val 461 */ 462 void ucl_emitter_print_int_msgpack (struct ucl_emitter_context *ctx, 463 int64_t val); 464 /** 465 * Print integer to the msgpack output 466 * @param ctx 467 * @param val 468 */ 469 void ucl_emitter_print_double_msgpack (struct ucl_emitter_context *ctx, 470 double val); 471 /** 472 * Print double to the msgpack output 473 * @param ctx 474 * @param val 475 */ 476 void ucl_emitter_print_bool_msgpack (struct ucl_emitter_context *ctx, 477 bool val); 478 /** 479 * Print string to the msgpack output 480 * @param ctx 481 * @param s 482 * @param len 483 */ 484 void ucl_emitter_print_string_msgpack (struct ucl_emitter_context *ctx, 485 const char *s, size_t len); 486 487 /** 488 * Print binary string to the msgpack output 489 * @param ctx 490 * @param s 491 * @param len 492 */ 493 void ucl_emitter_print_binary_string_msgpack (struct ucl_emitter_context *ctx, 494 const char *s, size_t len); 495 496 /** 497 * Print array preamble for msgpack 498 * @param ctx 499 * @param len 500 */ 501 void ucl_emitter_print_array_msgpack (struct ucl_emitter_context *ctx, 502 size_t len); 503 504 /** 505 * Print object preamble for msgpack 506 * @param ctx 507 * @param len 508 */ 509 void ucl_emitter_print_object_msgpack (struct ucl_emitter_context *ctx, 510 size_t len); 511 /** 512 * Print NULL to the msgpack output 513 * @param ctx 514 */ 515 void ucl_emitter_print_null_msgpack (struct ucl_emitter_context *ctx); 516 /** 517 * Print object's key if needed to the msgpack output 518 * @param print_key 519 * @param ctx 520 * @param obj 521 */ 522 void ucl_emitter_print_key_msgpack (bool print_key, 523 struct ucl_emitter_context *ctx, 524 const ucl_object_t *obj); 525 526 /** 527 * Add new element to an object using the current merge strategy and priority 528 * @param parser 529 * @param nobj 530 * @return 531 */ 532 bool ucl_parser_process_object_element (struct ucl_parser *parser, 533 ucl_object_t *nobj); 534 535 /** 536 * Parse msgpack chunk 537 * @param parser 538 * @return 539 */ 540 bool ucl_parse_msgpack (struct ucl_parser *parser); 541 542 #endif /* UCL_INTERNAL_H_ */ 543