Elliott Hughes | 5b80804 | 2021-10-01 10:56:10 -0700 | [diff] [blame] | 1 | /************************************************* |
| 2 | * PCRE2 DEMONSTRATION PROGRAM * |
| 3 | *************************************************/ |
| 4 | |
| 5 | /* This is a demonstration program to illustrate a straightforward way of |
| 6 | using the PCRE2 regular expression library from a C program. See the |
| 7 | pcre2sample documentation for a short discussion ("man pcre2sample" if you have |
| 8 | the PCRE2 man pages installed). PCRE2 is a revised API for the library, and is |
| 9 | incompatible with the original PCRE API. |
| 10 | |
| 11 | There are actually three libraries, each supporting a different code unit |
| 12 | width. This demonstration program uses the 8-bit library. The default is to |
| 13 | process each code unit as a separate character, but if the pattern begins with |
| 14 | "(*UTF)", both it and the subject are treated as UTF-8 strings, where |
| 15 | characters may occupy multiple code units. |
| 16 | |
| 17 | In Unix-like environments, if PCRE2 is installed in your standard system |
| 18 | libraries, you should be able to compile this program using this command: |
| 19 | |
| 20 | cc -Wall pcre2demo.c -lpcre2-8 -o pcre2demo |
| 21 | |
| 22 | If PCRE2 is not installed in a standard place, it is likely to be installed |
| 23 | with support for the pkg-config mechanism. If you have pkg-config, you can |
| 24 | compile this program using this command: |
| 25 | |
| 26 | cc -Wall pcre2demo.c `pkg-config --cflags --libs libpcre2-8` -o pcre2demo |
| 27 | |
| 28 | If you do not have pkg-config, you may have to use something like this: |
| 29 | |
| 30 | cc -Wall pcre2demo.c -I/usr/local/include -L/usr/local/lib \ |
| 31 | -R/usr/local/lib -lpcre2-8 -o pcre2demo |
| 32 | |
| 33 | Replace "/usr/local/include" and "/usr/local/lib" with wherever the include and |
| 34 | library files for PCRE2 are installed on your system. Only some operating |
| 35 | systems (Solaris is one) use the -R option. |
| 36 | |
| 37 | Building under Windows: |
| 38 | |
| 39 | If you want to statically link this program against a non-dll .a file, you must |
| 40 | define PCRE2_STATIC before including pcre2.h, so in this environment, uncomment |
| 41 | the following line. */ |
| 42 | |
| 43 | /* #define PCRE2_STATIC */ |
| 44 | |
| 45 | /* The PCRE2_CODE_UNIT_WIDTH macro must be defined before including pcre2.h. |
| 46 | For a program that uses only one code unit width, setting it to 8, 16, or 32 |
| 47 | makes it possible to use generic function names such as pcre2_compile(). Note |
| 48 | that just changing 8 to 16 (for example) is not sufficient to convert this |
| 49 | program to process 16-bit characters. Even in a fully 16-bit environment, where |
| 50 | string-handling functions such as strcmp() and printf() work with 16-bit |
| 51 | characters, the code for handling the table of named substrings will still need |
| 52 | to be modified. */ |
| 53 | |
| 54 | #define PCRE2_CODE_UNIT_WIDTH 8 |
| 55 | |
| 56 | #include <stdio.h> |
| 57 | #include <string.h> |
| 58 | #include <pcre2.h> |
| 59 | |
| 60 | |
| 61 | /************************************************************************** |
| 62 | * Here is the program. The API includes the concept of "contexts" for * |
| 63 | * setting up unusual interface requirements for compiling and matching, * |
| 64 | * such as custom memory managers and non-standard newline definitions. * |
| 65 | * This program does not do any of this, so it makes no use of contexts, * |
| 66 | * always passing NULL where a context could be given. * |
| 67 | **************************************************************************/ |
| 68 | |
| 69 | int main(int argc, char **argv) |
| 70 | { |
| 71 | pcre2_code *re; |
| 72 | PCRE2_SPTR pattern; /* PCRE2_SPTR is a pointer to unsigned code units of */ |
| 73 | PCRE2_SPTR subject; /* the appropriate width (in this case, 8 bits). */ |
| 74 | PCRE2_SPTR name_table; |
| 75 | |
| 76 | int crlf_is_newline; |
| 77 | int errornumber; |
| 78 | int find_all; |
| 79 | int i; |
| 80 | int rc; |
| 81 | int utf8; |
| 82 | |
| 83 | uint32_t option_bits; |
| 84 | uint32_t namecount; |
| 85 | uint32_t name_entry_size; |
| 86 | uint32_t newline; |
| 87 | |
| 88 | PCRE2_SIZE erroroffset; |
| 89 | PCRE2_SIZE *ovector; |
| 90 | PCRE2_SIZE subject_length; |
| 91 | |
| 92 | pcre2_match_data *match_data; |
| 93 | |
| 94 | |
| 95 | /************************************************************************** |
| 96 | * First, sort out the command line. There is only one possible option at * |
| 97 | * the moment, "-g" to request repeated matching to find all occurrences, * |
| 98 | * like Perl's /g option. We set the variable find_all to a non-zero value * |
| 99 | * if the -g option is present. * |
| 100 | **************************************************************************/ |
| 101 | |
| 102 | find_all = 0; |
| 103 | for (i = 1; i < argc; i++) |
| 104 | { |
| 105 | if (strcmp(argv[i], "-g") == 0) find_all = 1; |
| 106 | else if (argv[i][0] == '-') |
| 107 | { |
| 108 | printf("Unrecognised option %s\n", argv[i]); |
| 109 | return 1; |
| 110 | } |
| 111 | else break; |
| 112 | } |
| 113 | |
| 114 | /* After the options, we require exactly two arguments, which are the pattern, |
| 115 | and the subject string. */ |
| 116 | |
| 117 | if (argc - i != 2) |
| 118 | { |
| 119 | printf("Exactly two arguments required: a regex and a subject string\n"); |
| 120 | return 1; |
| 121 | } |
| 122 | |
| 123 | /* Pattern and subject are char arguments, so they can be straightforwardly |
| 124 | cast to PCRE2_SPTR because we are working in 8-bit code units. The subject |
| 125 | length is cast to PCRE2_SIZE for completeness, though PCRE2_SIZE is in fact |
| 126 | defined to be size_t. */ |
| 127 | |
| 128 | pattern = (PCRE2_SPTR)argv[i]; |
| 129 | subject = (PCRE2_SPTR)argv[i+1]; |
| 130 | subject_length = (PCRE2_SIZE)strlen((char *)subject); |
| 131 | |
| 132 | |
| 133 | /************************************************************************* |
| 134 | * Now we are going to compile the regular expression pattern, and handle * |
| 135 | * any errors that are detected. * |
| 136 | *************************************************************************/ |
| 137 | |
| 138 | re = pcre2_compile( |
| 139 | pattern, /* the pattern */ |
| 140 | PCRE2_ZERO_TERMINATED, /* indicates pattern is zero-terminated */ |
| 141 | 0, /* default options */ |
| 142 | &errornumber, /* for error number */ |
| 143 | &erroroffset, /* for error offset */ |
| 144 | NULL); /* use default compile context */ |
| 145 | |
| 146 | /* Compilation failed: print the error message and exit. */ |
| 147 | |
| 148 | if (re == NULL) |
| 149 | { |
| 150 | PCRE2_UCHAR buffer[256]; |
| 151 | pcre2_get_error_message(errornumber, buffer, sizeof(buffer)); |
| 152 | printf("PCRE2 compilation failed at offset %d: %s\n", (int)erroroffset, |
| 153 | buffer); |
| 154 | return 1; |
| 155 | } |
| 156 | |
| 157 | |
| 158 | /************************************************************************* |
| 159 | * If the compilation succeeded, we call PCRE2 again, in order to do a * |
| 160 | * pattern match against the subject string. This does just ONE match. If * |
| 161 | * further matching is needed, it will be done below. Before running the * |
| 162 | * match we must set up a match_data block for holding the result. Using * |
| 163 | * pcre2_match_data_create_from_pattern() ensures that the block is * |
| 164 | * exactly the right size for the number of capturing parentheses in the * |
| 165 | * pattern. If you need to know the actual size of a match_data block as * |
| 166 | * a number of bytes, you can find it like this: * |
| 167 | * * |
| 168 | * PCRE2_SIZE match_data_size = pcre2_get_match_data_size(match_data); * |
| 169 | *************************************************************************/ |
| 170 | |
| 171 | match_data = pcre2_match_data_create_from_pattern(re, NULL); |
| 172 | |
| 173 | /* Now run the match. */ |
| 174 | |
| 175 | rc = pcre2_match( |
| 176 | re, /* the compiled pattern */ |
| 177 | subject, /* the subject string */ |
| 178 | subject_length, /* the length of the subject */ |
| 179 | 0, /* start at offset 0 in the subject */ |
| 180 | 0, /* default options */ |
| 181 | match_data, /* block for storing the result */ |
| 182 | NULL); /* use default match context */ |
| 183 | |
| 184 | /* Matching failed: handle error cases */ |
| 185 | |
| 186 | if (rc < 0) |
| 187 | { |
| 188 | switch(rc) |
| 189 | { |
| 190 | case PCRE2_ERROR_NOMATCH: printf("No match\n"); break; |
| 191 | /* |
| 192 | Handle other special cases if you like |
| 193 | */ |
| 194 | default: printf("Matching error %d\n", rc); break; |
| 195 | } |
| 196 | pcre2_match_data_free(match_data); /* Release memory used for the match */ |
| 197 | pcre2_code_free(re); /* data and the compiled pattern. */ |
| 198 | return 1; |
| 199 | } |
| 200 | |
| 201 | /* Match succeeded. Get a pointer to the output vector, where string offsets |
| 202 | are stored. */ |
| 203 | |
| 204 | ovector = pcre2_get_ovector_pointer(match_data); |
| 205 | printf("Match succeeded at offset %d\n", (int)ovector[0]); |
| 206 | |
| 207 | |
| 208 | /************************************************************************* |
| 209 | * We have found the first match within the subject string. If the output * |
| 210 | * vector wasn't big enough, say so. Then output any substrings that were * |
| 211 | * captured. * |
| 212 | *************************************************************************/ |
| 213 | |
| 214 | /* The output vector wasn't big enough. This should not happen, because we used |
| 215 | pcre2_match_data_create_from_pattern() above. */ |
| 216 | |
| 217 | if (rc == 0) |
| 218 | printf("ovector was not big enough for all the captured substrings\n"); |
| 219 | |
| 220 | /* Since release 10.38 PCRE2 has locked out the use of \K in lookaround |
| 221 | assertions. However, there is an option to re-enable the old behaviour. If that |
| 222 | is set, it is possible to run patterns such as /(?=.\K)/ that use \K in an |
| 223 | assertion to set the start of a match later than its end. In this demonstration |
| 224 | program, we show how to detect this case, but it shouldn't arise because the |
| 225 | option is never set. */ |
| 226 | |
| 227 | if (ovector[0] > ovector[1]) |
| 228 | { |
| 229 | printf("\\K was used in an assertion to set the match start after its end.\n" |
| 230 | "From end to start the match was: %.*s\n", (int)(ovector[0] - ovector[1]), |
| 231 | (char *)(subject + ovector[1])); |
| 232 | printf("Run abandoned\n"); |
| 233 | pcre2_match_data_free(match_data); |
| 234 | pcre2_code_free(re); |
| 235 | return 1; |
| 236 | } |
| 237 | |
| 238 | /* Show substrings stored in the output vector by number. Obviously, in a real |
| 239 | application you might want to do things other than print them. */ |
| 240 | |
| 241 | for (i = 0; i < rc; i++) |
| 242 | { |
| 243 | PCRE2_SPTR substring_start = subject + ovector[2*i]; |
| 244 | PCRE2_SIZE substring_length = ovector[2*i+1] - ovector[2*i]; |
| 245 | printf("%2d: %.*s\n", i, (int)substring_length, (char *)substring_start); |
| 246 | } |
| 247 | |
| 248 | |
| 249 | /************************************************************************** |
| 250 | * That concludes the basic part of this demonstration program. We have * |
| 251 | * compiled a pattern, and performed a single match. The code that follows * |
| 252 | * shows first how to access named substrings, and then how to code for * |
| 253 | * repeated matches on the same subject. * |
| 254 | **************************************************************************/ |
| 255 | |
| 256 | /* See if there are any named substrings, and if so, show them by name. First |
| 257 | we have to extract the count of named parentheses from the pattern. */ |
| 258 | |
| 259 | (void)pcre2_pattern_info( |
| 260 | re, /* the compiled pattern */ |
| 261 | PCRE2_INFO_NAMECOUNT, /* get the number of named substrings */ |
| 262 | &namecount); /* where to put the answer */ |
| 263 | |
| 264 | if (namecount == 0) printf("No named substrings\n"); else |
| 265 | { |
| 266 | PCRE2_SPTR tabptr; |
| 267 | printf("Named substrings\n"); |
| 268 | |
| 269 | /* Before we can access the substrings, we must extract the table for |
| 270 | translating names to numbers, and the size of each entry in the table. */ |
| 271 | |
| 272 | (void)pcre2_pattern_info( |
| 273 | re, /* the compiled pattern */ |
| 274 | PCRE2_INFO_NAMETABLE, /* address of the table */ |
| 275 | &name_table); /* where to put the answer */ |
| 276 | |
| 277 | (void)pcre2_pattern_info( |
| 278 | re, /* the compiled pattern */ |
| 279 | PCRE2_INFO_NAMEENTRYSIZE, /* size of each entry in the table */ |
| 280 | &name_entry_size); /* where to put the answer */ |
| 281 | |
| 282 | /* Now we can scan the table and, for each entry, print the number, the name, |
| 283 | and the substring itself. In the 8-bit library the number is held in two |
| 284 | bytes, most significant first. */ |
| 285 | |
| 286 | tabptr = name_table; |
| 287 | for (i = 0; i < namecount; i++) |
| 288 | { |
| 289 | int n = (tabptr[0] << 8) | tabptr[1]; |
| 290 | printf("(%d) %*s: %.*s\n", n, name_entry_size - 3, tabptr + 2, |
| 291 | (int)(ovector[2*n+1] - ovector[2*n]), subject + ovector[2*n]); |
| 292 | tabptr += name_entry_size; |
| 293 | } |
| 294 | } |
| 295 | |
| 296 | |
| 297 | /************************************************************************* |
| 298 | * If the "-g" option was given on the command line, we want to continue * |
| 299 | * to search for additional matches in the subject string, in a similar * |
| 300 | * way to the /g option in Perl. This turns out to be trickier than you * |
| 301 | * might think because of the possibility of matching an empty string. * |
| 302 | * What happens is as follows: * |
| 303 | * * |
| 304 | * If the previous match was NOT for an empty string, we can just start * |
| 305 | * the next match at the end of the previous one. * |
| 306 | * * |
| 307 | * If the previous match WAS for an empty string, we can't do that, as it * |
| 308 | * would lead to an infinite loop. Instead, a call of pcre2_match() is * |
| 309 | * made with the PCRE2_NOTEMPTY_ATSTART and PCRE2_ANCHORED flags set. The * |
| 310 | * first of these tells PCRE2 that an empty string at the start of the * |
| 311 | * subject is not a valid match; other possibilities must be tried. The * |
| 312 | * second flag restricts PCRE2 to one match attempt at the initial string * |
| 313 | * position. If this match succeeds, an alternative to the empty string * |
| 314 | * match has been found, and we can print it and proceed round the loop, * |
| 315 | * advancing by the length of whatever was found. If this match does not * |
| 316 | * succeed, we still stay in the loop, advancing by just one character. * |
| 317 | * In UTF-8 mode, which can be set by (*UTF) in the pattern, this may be * |
| 318 | * more than one byte. * |
| 319 | * * |
| 320 | * However, there is a complication concerned with newlines. When the * |
| 321 | * newline convention is such that CRLF is a valid newline, we must * |
| 322 | * advance by two characters rather than one. The newline convention can * |
| 323 | * be set in the regex by (*CR), etc.; if not, we must find the default. * |
| 324 | *************************************************************************/ |
| 325 | |
| 326 | if (!find_all) /* Check for -g */ |
| 327 | { |
| 328 | pcre2_match_data_free(match_data); /* Release the memory that was used */ |
| 329 | pcre2_code_free(re); /* for the match data and the pattern. */ |
| 330 | return 0; /* Exit the program. */ |
| 331 | } |
| 332 | |
| 333 | /* Before running the loop, check for UTF-8 and whether CRLF is a valid newline |
| 334 | sequence. First, find the options with which the regex was compiled and extract |
| 335 | the UTF state. */ |
| 336 | |
| 337 | (void)pcre2_pattern_info(re, PCRE2_INFO_ALLOPTIONS, &option_bits); |
| 338 | utf8 = (option_bits & PCRE2_UTF) != 0; |
| 339 | |
| 340 | /* Now find the newline convention and see whether CRLF is a valid newline |
| 341 | sequence. */ |
| 342 | |
| 343 | (void)pcre2_pattern_info(re, PCRE2_INFO_NEWLINE, &newline); |
| 344 | crlf_is_newline = newline == PCRE2_NEWLINE_ANY || |
| 345 | newline == PCRE2_NEWLINE_CRLF || |
| 346 | newline == PCRE2_NEWLINE_ANYCRLF; |
| 347 | |
| 348 | /* Loop for second and subsequent matches */ |
| 349 | |
| 350 | for (;;) |
| 351 | { |
| 352 | uint32_t options = 0; /* Normally no options */ |
| 353 | PCRE2_SIZE start_offset = ovector[1]; /* Start at end of previous match */ |
| 354 | |
| 355 | /* If the previous match was for an empty string, we are finished if we are |
| 356 | at the end of the subject. Otherwise, arrange to run another match at the |
| 357 | same point to see if a non-empty match can be found. */ |
| 358 | |
| 359 | if (ovector[0] == ovector[1]) |
| 360 | { |
| 361 | if (ovector[0] == subject_length) break; |
| 362 | options = PCRE2_NOTEMPTY_ATSTART | PCRE2_ANCHORED; |
| 363 | } |
| 364 | |
| 365 | /* If the previous match was not an empty string, there is one tricky case to |
| 366 | consider. If a pattern contains \K within a lookbehind assertion at the |
| 367 | start, the end of the matched string can be at the offset where the match |
| 368 | started. Without special action, this leads to a loop that keeps on matching |
| 369 | the same substring. We must detect this case and arrange to move the start on |
| 370 | by one character. The pcre2_get_startchar() function returns the starting |
| 371 | offset that was passed to pcre2_match(). */ |
| 372 | |
| 373 | else |
| 374 | { |
| 375 | PCRE2_SIZE startchar = pcre2_get_startchar(match_data); |
| 376 | if (start_offset <= startchar) |
| 377 | { |
| 378 | if (startchar >= subject_length) break; /* Reached end of subject. */ |
| 379 | start_offset = startchar + 1; /* Advance by one character. */ |
| 380 | if (utf8) /* If UTF-8, it may be more */ |
| 381 | { /* than one code unit. */ |
| 382 | for (; start_offset < subject_length; start_offset++) |
| 383 | if ((subject[start_offset] & 0xc0) != 0x80) break; |
| 384 | } |
| 385 | } |
| 386 | } |
| 387 | |
| 388 | /* Run the next matching operation */ |
| 389 | |
| 390 | rc = pcre2_match( |
| 391 | re, /* the compiled pattern */ |
| 392 | subject, /* the subject string */ |
| 393 | subject_length, /* the length of the subject */ |
| 394 | start_offset, /* starting offset in the subject */ |
| 395 | options, /* options */ |
| 396 | match_data, /* block for storing the result */ |
| 397 | NULL); /* use default match context */ |
| 398 | |
| 399 | /* This time, a result of NOMATCH isn't an error. If the value in "options" |
| 400 | is zero, it just means we have found all possible matches, so the loop ends. |
| 401 | Otherwise, it means we have failed to find a non-empty-string match at a |
| 402 | point where there was a previous empty-string match. In this case, we do what |
| 403 | Perl does: advance the matching position by one character, and continue. We |
| 404 | do this by setting the "end of previous match" offset, because that is picked |
| 405 | up at the top of the loop as the point at which to start again. |
| 406 | |
| 407 | There are two complications: (a) When CRLF is a valid newline sequence, and |
| 408 | the current position is just before it, advance by an extra byte. (b) |
| 409 | Otherwise we must ensure that we skip an entire UTF character if we are in |
| 410 | UTF mode. */ |
| 411 | |
| 412 | if (rc == PCRE2_ERROR_NOMATCH) |
| 413 | { |
| 414 | if (options == 0) break; /* All matches found */ |
| 415 | ovector[1] = start_offset + 1; /* Advance one code unit */ |
| 416 | if (crlf_is_newline && /* If CRLF is a newline & */ |
| 417 | start_offset < subject_length - 1 && /* we are at CRLF, */ |
| 418 | subject[start_offset] == '\r' && |
| 419 | subject[start_offset + 1] == '\n') |
| 420 | ovector[1] += 1; /* Advance by one more. */ |
| 421 | else if (utf8) /* Otherwise, ensure we */ |
| 422 | { /* advance a whole UTF-8 */ |
| 423 | while (ovector[1] < subject_length) /* character. */ |
| 424 | { |
| 425 | if ((subject[ovector[1]] & 0xc0) != 0x80) break; |
| 426 | ovector[1] += 1; |
| 427 | } |
| 428 | } |
| 429 | continue; /* Go round the loop again */ |
| 430 | } |
| 431 | |
| 432 | /* Other matching errors are not recoverable. */ |
| 433 | |
| 434 | if (rc < 0) |
| 435 | { |
| 436 | printf("Matching error %d\n", rc); |
| 437 | pcre2_match_data_free(match_data); |
| 438 | pcre2_code_free(re); |
| 439 | return 1; |
| 440 | } |
| 441 | |
| 442 | /* Match succeeded */ |
| 443 | |
| 444 | printf("\nMatch succeeded again at offset %d\n", (int)ovector[0]); |
| 445 | |
| 446 | /* The match succeeded, but the output vector wasn't big enough. This |
| 447 | should not happen. */ |
| 448 | |
| 449 | if (rc == 0) |
| 450 | printf("ovector was not big enough for all the captured substrings\n"); |
| 451 | |
| 452 | /* We must guard against patterns such as /(?=.\K)/ that use \K in an |
| 453 | assertion to set the start of a match later than its end. In this |
| 454 | demonstration program, we just detect this case and give up. */ |
| 455 | |
| 456 | if (ovector[0] > ovector[1]) |
| 457 | { |
| 458 | printf("\\K was used in an assertion to set the match start after its end.\n" |
| 459 | "From end to start the match was: %.*s\n", (int)(ovector[0] - ovector[1]), |
| 460 | (char *)(subject + ovector[1])); |
| 461 | printf("Run abandoned\n"); |
| 462 | pcre2_match_data_free(match_data); |
| 463 | pcre2_code_free(re); |
| 464 | return 1; |
| 465 | } |
| 466 | |
| 467 | /* As before, show substrings stored in the output vector by number, and then |
| 468 | also any named substrings. */ |
| 469 | |
| 470 | for (i = 0; i < rc; i++) |
| 471 | { |
| 472 | PCRE2_SPTR substring_start = subject + ovector[2*i]; |
| 473 | size_t substring_length = ovector[2*i+1] - ovector[2*i]; |
| 474 | printf("%2d: %.*s\n", i, (int)substring_length, (char *)substring_start); |
| 475 | } |
| 476 | |
| 477 | if (namecount == 0) printf("No named substrings\n"); else |
| 478 | { |
| 479 | PCRE2_SPTR tabptr = name_table; |
| 480 | printf("Named substrings\n"); |
| 481 | for (i = 0; i < namecount; i++) |
| 482 | { |
| 483 | int n = (tabptr[0] << 8) | tabptr[1]; |
| 484 | printf("(%d) %*s: %.*s\n", n, name_entry_size - 3, tabptr + 2, |
| 485 | (int)(ovector[2*n+1] - ovector[2*n]), subject + ovector[2*n]); |
| 486 | tabptr += name_entry_size; |
| 487 | } |
| 488 | } |
| 489 | } /* End of loop to find second and subsequent matches */ |
| 490 | |
| 491 | printf("\n"); |
| 492 | pcre2_match_data_free(match_data); |
| 493 | pcre2_code_free(re); |
| 494 | return 0; |
| 495 | } |
| 496 | |
| 497 | /* End of pcre2demo.c */ |