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 ucl_object_t *comments; 215 ucl_object_t *last_comment; 216 UT_string *err; 217 }; 218 219 struct ucl_object_userdata { 220 ucl_object_t obj; 221 ucl_userdata_dtor dtor; 222 ucl_userdata_emitter emitter; 223 }; 224 225 /** 226 * Unescape json string inplace 227 * @param str 228 */ 229 size_t ucl_unescape_json_string (char *str, size_t len); 230 231 /** 232 * Handle include macro 233 * @param data include data 234 * @param len length of data 235 * @param args UCL object representing arguments to the macro 236 * @param ud user data 237 * @return 238 */ 239 bool ucl_include_handler (const unsigned char *data, size_t len, 240 const ucl_object_t *args, void* ud); 241 242 /** 243 * Handle tryinclude macro 244 * @param data include data 245 * @param len length of data 246 * @param args UCL object representing arguments to the macro 247 * @param ud user data 248 * @return 249 */ 250 bool ucl_try_include_handler (const unsigned char *data, size_t len, 251 const ucl_object_t *args, void* ud); 252 253 /** 254 * Handle includes macro 255 * @param data include data 256 * @param len length of data 257 * @param args UCL object representing arguments to the macro 258 * @param ud user data 259 * @return 260 */ 261 bool ucl_includes_handler (const unsigned char *data, size_t len, 262 const ucl_object_t *args, void* ud); 263 264 /** 265 * Handle priority macro 266 * @param data include data 267 * @param len length of data 268 * @param args UCL object representing arguments to the macro 269 * @param ud user data 270 * @return 271 */ 272 bool ucl_priority_handler (const unsigned char *data, size_t len, 273 const ucl_object_t *args, void* ud); 274 275 /** 276 * Handle load macro 277 * @param data include data 278 * @param len length of data 279 * @param args UCL object representing arguments to the macro 280 * @param ud user data 281 * @return 282 */ 283 bool ucl_load_handler (const unsigned char *data, size_t len, 284 const ucl_object_t *args, void* ud); 285 /** 286 * Handle inherit macro 287 * @param data include data 288 * @param len length of data 289 * @param args UCL object representing arguments to the macro 290 * @param ctx the current context object 291 * @param ud user data 292 * @return 293 */ 294 bool ucl_inherit_handler (const unsigned char *data, size_t len, 295 const ucl_object_t *args, const ucl_object_t *ctx, void* ud); 296 297 size_t ucl_strlcpy (char *dst, const char *src, size_t siz); 298 size_t ucl_strlcpy_unsafe (char *dst, const char *src, size_t siz); 299 size_t ucl_strlcpy_tolower (char *dst, const char *src, size_t siz); 300 301 char *ucl_strnstr (const char *s, const char *find, int len); 302 char *ucl_strncasestr (const char *s, const char *find, int len); 303 304 #ifdef __GNUC__ 305 static inline void 306 ucl_create_err (UT_string **err, const char *fmt, ...) 307 __attribute__ (( format( printf, 2, 3) )); 308 #endif 309 310 #undef UCL_FATAL_ERRORS 311 312 static inline void 313 ucl_create_err (UT_string **err, const char *fmt, ...) 314 { 315 if (*err == NULL) { 316 utstring_new (*err); 317 va_list ap; 318 va_start (ap, fmt); 319 utstring_printf_va (*err, fmt, ap); 320 va_end (ap); 321 } 322 323 #ifdef UCL_FATAL_ERRORS 324 assert (0); 325 #endif 326 } 327 328 /** 329 * Check whether a given string contains a boolean value 330 * @param obj object to set 331 * @param start start of a string 332 * @param len length of a string 333 * @return true if a string is a boolean value 334 */ 335 static inline bool 336 ucl_maybe_parse_boolean (ucl_object_t *obj, const unsigned char *start, size_t len) 337 { 338 const char *p = (const char *)start; 339 bool ret = false, val = false; 340 341 if (len == 5) { 342 if ((p[0] == 'f' || p[0] == 'F') && strncasecmp (p, "false", 5) == 0) { 343 ret = true; 344 val = false; 345 } 346 } 347 else if (len == 4) { 348 if ((p[0] == 't' || p[0] == 'T') && strncasecmp (p, "true", 4) == 0) { 349 ret = true; 350 val = true; 351 } 352 } 353 else if (len == 3) { 354 if ((p[0] == 'y' || p[0] == 'Y') && strncasecmp (p, "yes", 3) == 0) { 355 ret = true; 356 val = true; 357 } 358 else if ((p[0] == 'o' || p[0] == 'O') && strncasecmp (p, "off", 3) == 0) { 359 ret = true; 360 val = false; 361 } 362 } 363 else if (len == 2) { 364 if ((p[0] == 'n' || p[0] == 'N') && strncasecmp (p, "no", 2) == 0) { 365 ret = true; 366 val = false; 367 } 368 else if ((p[0] == 'o' || p[0] == 'O') && strncasecmp (p, "on", 2) == 0) { 369 ret = true; 370 val = true; 371 } 372 } 373 374 if (ret && obj != NULL) { 375 obj->type = UCL_BOOLEAN; 376 obj->value.iv = val; 377 } 378 379 return ret; 380 } 381 382 /** 383 * Check numeric string 384 * @param obj object to set if a string is numeric 385 * @param start start of string 386 * @param end end of string 387 * @param pos position where parsing has stopped 388 * @param allow_double allow parsing of floating point values 389 * @return 0 if string is numeric and error code (EINVAL or ERANGE) in case of conversion error 390 */ 391 int ucl_maybe_parse_number (ucl_object_t *obj, 392 const char *start, const char *end, const char **pos, 393 bool allow_double, bool number_bytes, bool allow_time); 394 395 396 static inline const ucl_object_t * 397 ucl_hash_search_obj (ucl_hash_t* hashlin, ucl_object_t *obj) 398 { 399 return (const ucl_object_t *)ucl_hash_search (hashlin, obj->key, obj->keylen); 400 } 401 402 static inline ucl_hash_t * ucl_hash_insert_object (ucl_hash_t *hashlin, 403 const ucl_object_t *obj, 404 bool ignore_case) UCL_WARN_UNUSED_RESULT; 405 406 static inline ucl_hash_t * 407 ucl_hash_insert_object (ucl_hash_t *hashlin, 408 const ucl_object_t *obj, 409 bool ignore_case) 410 { 411 if (hashlin == NULL) { 412 hashlin = ucl_hash_create (ignore_case); 413 } 414 ucl_hash_insert (hashlin, obj, obj->key, obj->keylen); 415 416 return hashlin; 417 } 418 419 /** 420 * Get standard emitter context for a specified emit_type 421 * @param emit_type type of emitter 422 * @return context or NULL if input is invalid 423 */ 424 const struct ucl_emitter_context * 425 ucl_emit_get_standard_context (enum ucl_emitter emit_type); 426 427 /** 428 * Serialize string as JSON string 429 * @param str string to emit 430 * @param buf target buffer 431 */ 432 void ucl_elt_string_write_json (const char *str, size_t size, 433 struct ucl_emitter_context *ctx); 434 435 /** 436 * Write multiline string using `EOD` as string terminator 437 * @param str 438 * @param size 439 * @param ctx 440 */ 441 void ucl_elt_string_write_multiline (const char *str, size_t size, 442 struct ucl_emitter_context *ctx); 443 444 /** 445 * Emit a single object to string 446 * @param obj 447 * @return 448 */ 449 unsigned char * ucl_object_emit_single_json (const ucl_object_t *obj); 450 451 /** 452 * Check whether a specified string is long and should be likely printed in 453 * multiline mode 454 * @param obj 455 * @return 456 */ 457 bool ucl_maybe_long_string (const ucl_object_t *obj); 458 459 /** 460 * Print integer to the msgpack output 461 * @param ctx 462 * @param val 463 */ 464 void ucl_emitter_print_int_msgpack (struct ucl_emitter_context *ctx, 465 int64_t val); 466 /** 467 * Print integer to the msgpack output 468 * @param ctx 469 * @param val 470 */ 471 void ucl_emitter_print_double_msgpack (struct ucl_emitter_context *ctx, 472 double val); 473 /** 474 * Print double to the msgpack output 475 * @param ctx 476 * @param val 477 */ 478 void ucl_emitter_print_bool_msgpack (struct ucl_emitter_context *ctx, 479 bool val); 480 /** 481 * Print string to the msgpack output 482 * @param ctx 483 * @param s 484 * @param len 485 */ 486 void ucl_emitter_print_string_msgpack (struct ucl_emitter_context *ctx, 487 const char *s, size_t len); 488 489 /** 490 * Print binary string to the msgpack output 491 * @param ctx 492 * @param s 493 * @param len 494 */ 495 void ucl_emitter_print_binary_string_msgpack (struct ucl_emitter_context *ctx, 496 const char *s, size_t len); 497 498 /** 499 * Print array preamble for msgpack 500 * @param ctx 501 * @param len 502 */ 503 void ucl_emitter_print_array_msgpack (struct ucl_emitter_context *ctx, 504 size_t len); 505 506 /** 507 * Print object preamble for msgpack 508 * @param ctx 509 * @param len 510 */ 511 void ucl_emitter_print_object_msgpack (struct ucl_emitter_context *ctx, 512 size_t len); 513 /** 514 * Print NULL to the msgpack output 515 * @param ctx 516 */ 517 void ucl_emitter_print_null_msgpack (struct ucl_emitter_context *ctx); 518 /** 519 * Print object's key if needed to the msgpack output 520 * @param print_key 521 * @param ctx 522 * @param obj 523 */ 524 void ucl_emitter_print_key_msgpack (bool print_key, 525 struct ucl_emitter_context *ctx, 526 const ucl_object_t *obj); 527 528 /** 529 * Fetch URL into a buffer 530 * @param url url to fetch 531 * @param buf pointer to buffer (must be freed by callee) 532 * @param buflen pointer to buffer length 533 * @param err pointer to error argument 534 * @param must_exist fail if cannot find a url 535 */ 536 bool ucl_fetch_url (const unsigned char *url, 537 unsigned char **buf, 538 size_t *buflen, 539 UT_string **err, 540 bool must_exist); 541 542 /** 543 * Fetch a file and save results to the memory buffer 544 * @param filename filename to fetch 545 * @param len length of filename 546 * @param buf target buffer 547 * @param buflen target length 548 * @return 549 */ 550 bool ucl_fetch_file (const unsigned char *filename, 551 unsigned char **buf, 552 size_t *buflen, 553 UT_string **err, 554 bool must_exist); 555 556 /** 557 * Add new element to an object using the current merge strategy and priority 558 * @param parser 559 * @param nobj 560 * @return 561 */ 562 bool ucl_parser_process_object_element (struct ucl_parser *parser, 563 ucl_object_t *nobj); 564 565 /** 566 * Parse msgpack chunk 567 * @param parser 568 * @return 569 */ 570 bool ucl_parse_msgpack (struct ucl_parser *parser); 571 572 #endif /* UCL_INTERNAL_H_ */ 573