1<html> 2<head> 3<title>pcre2demo specification</title> 4</head> 5<body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB"> 6<h1>pcre2demo man page</h1> 7<p> 8Return to the <a href="index.html">PCRE2 index page</a>. 9</p> 10<p> 11This page is part of the PCRE2 HTML documentation. It was generated 12automatically from the original man page. If there is any nonsense in it, 13please consult the man page, in case the conversion went wrong. 14<br> 15<ul> 16</ul> 17<PRE> 18/************************************************* 19* PCRE2 DEMONSTRATION PROGRAM * 20*************************************************/ 21 22/* This is a demonstration program to illustrate a straightforward way of 23using the PCRE2 regular expression library from a C program. See the 24pcre2sample documentation for a short discussion ("man pcre2sample" if you have 25the PCRE2 man pages installed). PCRE2 is a revised API for the library, and is 26incompatible with the original PCRE API. 27 28There are actually three libraries, each supporting a different code unit 29width. This demonstration program uses the 8-bit library. The default is to 30process each code unit as a separate character, but if the pattern begins with 31"(*UTF)", both it and the subject are treated as UTF-8 strings, where 32characters may occupy multiple code units. 33 34In Unix-like environments, if PCRE2 is installed in your standard system 35libraries, you should be able to compile this program using this command: 36 37cc -Wall pcre2demo.c -lpcre2-8 -o pcre2demo 38 39If PCRE2 is not installed in a standard place, it is likely to be installed 40with support for the pkg-config mechanism. If you have pkg-config, you can 41compile this program using this command: 42 43cc -Wall pcre2demo.c `pkg-config --cflags --libs libpcre2-8` -o pcre2demo 44 45If you do not have pkg-config, you may have to use something like this: 46 47cc -Wall pcre2demo.c -I/usr/local/include -L/usr/local/lib \ 48 -R/usr/local/lib -lpcre2-8 -o pcre2demo 49 50Replace "/usr/local/include" and "/usr/local/lib" with wherever the include and 51library files for PCRE2 are installed on your system. Only some operating 52systems (Solaris is one) use the -R option. 53 54Building under Windows: 55 56If you want to statically link this program against a non-dll .a file, you must 57define PCRE2_STATIC before including pcre2.h, so in this environment, uncomment 58the following line. */ 59 60/* #define PCRE2_STATIC */ 61 62/* The PCRE2_CODE_UNIT_WIDTH macro must be defined before including pcre2.h. 63For a program that uses only one code unit width, setting it to 8, 16, or 32 64makes it possible to use generic function names such as pcre2_compile(). Note 65that just changing 8 to 16 (for example) is not sufficient to convert this 66program to process 16-bit characters. Even in a fully 16-bit environment, where 67string-handling functions such as strcmp() and printf() work with 16-bit 68characters, the code for handling the table of named substrings will still need 69to be modified. */ 70 71#define PCRE2_CODE_UNIT_WIDTH 8 72 73#include <stdio.h> 74#include <string.h> 75#include <pcre2.h> 76 77 78/************************************************************************** 79* Here is the program. The API includes the concept of "contexts" for * 80* setting up unusual interface requirements for compiling and matching, * 81* such as custom memory managers and non-standard newline definitions. * 82* This program does not do any of this, so it makes no use of contexts, * 83* always passing NULL where a context could be given. * 84**************************************************************************/ 85 86int main(int argc, char **argv) 87{ 88pcre2_code *re; 89PCRE2_SPTR pattern; /* PCRE2_SPTR is a pointer to unsigned code units of */ 90PCRE2_SPTR subject; /* the appropriate width (in this case, 8 bits). */ 91PCRE2_SPTR name_table; 92 93int crlf_is_newline; 94int errornumber; 95int find_all; 96int i; 97int rc; 98int utf8; 99 100uint32_t option_bits; 101uint32_t namecount; 102uint32_t name_entry_size; 103uint32_t newline; 104 105PCRE2_SIZE erroroffset; 106PCRE2_SIZE *ovector; 107PCRE2_SIZE subject_length; 108 109pcre2_match_data *match_data; 110 111 112/************************************************************************** 113* First, sort out the command line. There is only one possible option at * 114* the moment, "-g" to request repeated matching to find all occurrences, * 115* like Perl's /g option. We set the variable find_all to a non-zero value * 116* if the -g option is present. * 117**************************************************************************/ 118 119find_all = 0; 120for (i = 1; i < argc; i++) 121 { 122 if (strcmp(argv[i], "-g") == 0) find_all = 1; 123 else if (argv[i][0] == '-') 124 { 125 printf("Unrecognised option %s\n", argv[i]); 126 return 1; 127 } 128 else break; 129 } 130 131/* After the options, we require exactly two arguments, which are the pattern, 132and the subject string. */ 133 134if (argc - i != 2) 135 { 136 printf("Exactly two arguments required: a regex and a subject string\n"); 137 return 1; 138 } 139 140/* Pattern and subject are char arguments, so they can be straightforwardly 141cast to PCRE2_SPTR because we are working in 8-bit code units. The subject 142length is cast to PCRE2_SIZE for completeness, though PCRE2_SIZE is in fact 143defined to be size_t. */ 144 145pattern = (PCRE2_SPTR)argv[i]; 146subject = (PCRE2_SPTR)argv[i+1]; 147subject_length = (PCRE2_SIZE)strlen((char *)subject); 148 149 150/************************************************************************* 151* Now we are going to compile the regular expression pattern, and handle * 152* any errors that are detected. * 153*************************************************************************/ 154 155re = pcre2_compile( 156 pattern, /* the pattern */ 157 PCRE2_ZERO_TERMINATED, /* indicates pattern is zero-terminated */ 158 0, /* default options */ 159 &errornumber, /* for error number */ 160 &erroroffset, /* for error offset */ 161 NULL); /* use default compile context */ 162 163/* Compilation failed: print the error message and exit. */ 164 165if (re == NULL) 166 { 167 PCRE2_UCHAR buffer[256]; 168 pcre2_get_error_message(errornumber, buffer, sizeof(buffer)); 169 printf("PCRE2 compilation failed at offset %d: %s\n", (int)erroroffset, 170 buffer); 171 return 1; 172 } 173 174 175/************************************************************************* 176* If the compilation succeeded, we call PCRE2 again, in order to do a * 177* pattern match against the subject string. This does just ONE match. If * 178* further matching is needed, it will be done below. Before running the * 179* match we must set up a match_data block for holding the result. Using * 180* pcre2_match_data_create_from_pattern() ensures that the block is * 181* exactly the right size for the number of capturing parentheses in the * 182* pattern. If you need to know the actual size of a match_data block as * 183* a number of bytes, you can find it like this: * 184* * 185* PCRE2_SIZE match_data_size = pcre2_get_match_data_size(match_data); * 186*************************************************************************/ 187 188match_data = pcre2_match_data_create_from_pattern(re, NULL); 189 190/* Now run the match. */ 191 192rc = pcre2_match( 193 re, /* the compiled pattern */ 194 subject, /* the subject string */ 195 subject_length, /* the length of the subject */ 196 0, /* start at offset 0 in the subject */ 197 0, /* default options */ 198 match_data, /* block for storing the result */ 199 NULL); /* use default match context */ 200 201/* Matching failed: handle error cases */ 202 203if (rc < 0) 204 { 205 switch(rc) 206 { 207 case PCRE2_ERROR_NOMATCH: printf("No match\n"); break; 208 /* 209 Handle other special cases if you like 210 */ 211 default: printf("Matching error %d\n", rc); break; 212 } 213 pcre2_match_data_free(match_data); /* Release memory used for the match */ 214 pcre2_code_free(re); /* data and the compiled pattern. */ 215 return 1; 216 } 217 218/* Match succeeded. Get a pointer to the output vector, where string offsets 219are stored. */ 220 221ovector = pcre2_get_ovector_pointer(match_data); 222printf("Match succeeded at offset %d\n", (int)ovector[0]); 223 224 225/************************************************************************* 226* We have found the first match within the subject string. If the output * 227* vector wasn't big enough, say so. Then output any substrings that were * 228* captured. * 229*************************************************************************/ 230 231/* The output vector wasn't big enough. This should not happen, because we used 232pcre2_match_data_create_from_pattern() above. */ 233 234if (rc == 0) 235 printf("ovector was not big enough for all the captured substrings\n"); 236 237/* Since release 10.38 PCRE2 has locked out the use of \K in lookaround 238assertions. However, there is an option to re-enable the old behaviour. If that 239is set, it is possible to run patterns such as /(?=.\K)/ that use \K in an 240assertion to set the start of a match later than its end. In this demonstration 241program, we show how to detect this case, but it shouldn't arise because the 242option is never set. */ 243 244if (ovector[0] > ovector[1]) 245 { 246 printf("\\K was used in an assertion to set the match start after its end.\n" 247 "From end to start the match was: %.*s\n", (int)(ovector[0] - ovector[1]), 248 (char *)(subject + ovector[1])); 249 printf("Run abandoned\n"); 250 pcre2_match_data_free(match_data); 251 pcre2_code_free(re); 252 return 1; 253 } 254 255/* Show substrings stored in the output vector by number. Obviously, in a real 256application you might want to do things other than print them. */ 257 258for (i = 0; i < rc; i++) 259 { 260 PCRE2_SPTR substring_start = subject + ovector[2*i]; 261 PCRE2_SIZE substring_length = ovector[2*i+1] - ovector[2*i]; 262 printf("%2d: %.*s\n", i, (int)substring_length, (char *)substring_start); 263 } 264 265 266/************************************************************************** 267* That concludes the basic part of this demonstration program. We have * 268* compiled a pattern, and performed a single match. The code that follows * 269* shows first how to access named substrings, and then how to code for * 270* repeated matches on the same subject. * 271**************************************************************************/ 272 273/* See if there are any named substrings, and if so, show them by name. First 274we have to extract the count of named parentheses from the pattern. */ 275 276(void)pcre2_pattern_info( 277 re, /* the compiled pattern */ 278 PCRE2_INFO_NAMECOUNT, /* get the number of named substrings */ 279 &namecount); /* where to put the answer */ 280 281if (namecount == 0) printf("No named substrings\n"); else 282 { 283 PCRE2_SPTR tabptr; 284 printf("Named substrings\n"); 285 286 /* Before we can access the substrings, we must extract the table for 287 translating names to numbers, and the size of each entry in the table. */ 288 289 (void)pcre2_pattern_info( 290 re, /* the compiled pattern */ 291 PCRE2_INFO_NAMETABLE, /* address of the table */ 292 &name_table); /* where to put the answer */ 293 294 (void)pcre2_pattern_info( 295 re, /* the compiled pattern */ 296 PCRE2_INFO_NAMEENTRYSIZE, /* size of each entry in the table */ 297 &name_entry_size); /* where to put the answer */ 298 299 /* Now we can scan the table and, for each entry, print the number, the name, 300 and the substring itself. In the 8-bit library the number is held in two 301 bytes, most significant first. */ 302 303 tabptr = name_table; 304 for (i = 0; i < namecount; i++) 305 { 306 int n = (tabptr[0] << 8) | tabptr[1]; 307 printf("(%d) %*s: %.*s\n", n, name_entry_size - 3, tabptr + 2, 308 (int)(ovector[2*n+1] - ovector[2*n]), subject + ovector[2*n]); 309 tabptr += name_entry_size; 310 } 311 } 312 313 314/************************************************************************* 315* If the "-g" option was given on the command line, we want to continue * 316* to search for additional matches in the subject string, in a similar * 317* way to the /g option in Perl. This turns out to be trickier than you * 318* might think because of the possibility of matching an empty string. * 319* What happens is as follows: * 320* * 321* If the previous match was NOT for an empty string, we can just start * 322* the next match at the end of the previous one. * 323* * 324* If the previous match WAS for an empty string, we can't do that, as it * 325* would lead to an infinite loop. Instead, a call of pcre2_match() is * 326* made with the PCRE2_NOTEMPTY_ATSTART and PCRE2_ANCHORED flags set. The * 327* first of these tells PCRE2 that an empty string at the start of the * 328* subject is not a valid match; other possibilities must be tried. The * 329* second flag restricts PCRE2 to one match attempt at the initial string * 330* position. If this match succeeds, an alternative to the empty string * 331* match has been found, and we can print it and proceed round the loop, * 332* advancing by the length of whatever was found. If this match does not * 333* succeed, we still stay in the loop, advancing by just one character. * 334* In UTF-8 mode, which can be set by (*UTF) in the pattern, this may be * 335* more than one byte. * 336* * 337* However, there is a complication concerned with newlines. When the * 338* newline convention is such that CRLF is a valid newline, we must * 339* advance by two characters rather than one. The newline convention can * 340* be set in the regex by (*CR), etc.; if not, we must find the default. * 341*************************************************************************/ 342 343if (!find_all) /* Check for -g */ 344 { 345 pcre2_match_data_free(match_data); /* Release the memory that was used */ 346 pcre2_code_free(re); /* for the match data and the pattern. */ 347 return 0; /* Exit the program. */ 348 } 349 350/* Before running the loop, check for UTF-8 and whether CRLF is a valid newline 351sequence. First, find the options with which the regex was compiled and extract 352the UTF state. */ 353 354(void)pcre2_pattern_info(re, PCRE2_INFO_ALLOPTIONS, &option_bits); 355utf8 = (option_bits & PCRE2_UTF) != 0; 356 357/* Now find the newline convention and see whether CRLF is a valid newline 358sequence. */ 359 360(void)pcre2_pattern_info(re, PCRE2_INFO_NEWLINE, &newline); 361crlf_is_newline = newline == PCRE2_NEWLINE_ANY || 362 newline == PCRE2_NEWLINE_CRLF || 363 newline == PCRE2_NEWLINE_ANYCRLF; 364 365/* Loop for second and subsequent matches */ 366 367for (;;) 368 { 369 uint32_t options = 0; /* Normally no options */ 370 PCRE2_SIZE start_offset = ovector[1]; /* Start at end of previous match */ 371 372 /* If the previous match was for an empty string, we are finished if we are 373 at the end of the subject. Otherwise, arrange to run another match at the 374 same point to see if a non-empty match can be found. */ 375 376 if (ovector[0] == ovector[1]) 377 { 378 if (ovector[0] == subject_length) break; 379 options = PCRE2_NOTEMPTY_ATSTART | PCRE2_ANCHORED; 380 } 381 382 /* If the previous match was not an empty string, there is one tricky case to 383 consider. If a pattern contains \K within a lookbehind assertion at the 384 start, the end of the matched string can be at the offset where the match 385 started. Without special action, this leads to a loop that keeps on matching 386 the same substring. We must detect this case and arrange to move the start on 387 by one character. The pcre2_get_startchar() function returns the starting 388 offset that was passed to pcre2_match(). */ 389 390 else 391 { 392 PCRE2_SIZE startchar = pcre2_get_startchar(match_data); 393 if (start_offset <= startchar) 394 { 395 if (startchar >= subject_length) break; /* Reached end of subject. */ 396 start_offset = startchar + 1; /* Advance by one character. */ 397 if (utf8) /* If UTF-8, it may be more */ 398 { /* than one code unit. */ 399 for (; start_offset < subject_length; start_offset++) 400 if ((subject[start_offset] & 0xc0) != 0x80) break; 401 } 402 } 403 } 404 405 /* Run the next matching operation */ 406 407 rc = pcre2_match( 408 re, /* the compiled pattern */ 409 subject, /* the subject string */ 410 subject_length, /* the length of the subject */ 411 start_offset, /* starting offset in the subject */ 412 options, /* options */ 413 match_data, /* block for storing the result */ 414 NULL); /* use default match context */ 415 416 /* This time, a result of NOMATCH isn't an error. If the value in "options" 417 is zero, it just means we have found all possible matches, so the loop ends. 418 Otherwise, it means we have failed to find a non-empty-string match at a 419 point where there was a previous empty-string match. In this case, we do what 420 Perl does: advance the matching position by one character, and continue. We 421 do this by setting the "end of previous match" offset, because that is picked 422 up at the top of the loop as the point at which to start again. 423 424 There are two complications: (a) When CRLF is a valid newline sequence, and 425 the current position is just before it, advance by an extra byte. (b) 426 Otherwise we must ensure that we skip an entire UTF character if we are in 427 UTF mode. */ 428 429 if (rc == PCRE2_ERROR_NOMATCH) 430 { 431 if (options == 0) break; /* All matches found */ 432 ovector[1] = start_offset + 1; /* Advance one code unit */ 433 if (crlf_is_newline && /* If CRLF is a newline & */ 434 start_offset < subject_length - 1 && /* we are at CRLF, */ 435 subject[start_offset] == '\r' && 436 subject[start_offset + 1] == '\n') 437 ovector[1] += 1; /* Advance by one more. */ 438 else if (utf8) /* Otherwise, ensure we */ 439 { /* advance a whole UTF-8 */ 440 while (ovector[1] < subject_length) /* character. */ 441 { 442 if ((subject[ovector[1]] & 0xc0) != 0x80) break; 443 ovector[1] += 1; 444 } 445 } 446 continue; /* Go round the loop again */ 447 } 448 449 /* Other matching errors are not recoverable. */ 450 451 if (rc < 0) 452 { 453 printf("Matching error %d\n", rc); 454 pcre2_match_data_free(match_data); 455 pcre2_code_free(re); 456 return 1; 457 } 458 459 /* Match succeeded */ 460 461 printf("\nMatch succeeded again at offset %d\n", (int)ovector[0]); 462 463 /* The match succeeded, but the output vector wasn't big enough. This 464 should not happen. */ 465 466 if (rc == 0) 467 printf("ovector was not big enough for all the captured substrings\n"); 468 469 /* We must guard against patterns such as /(?=.\K)/ that use \K in an 470 assertion to set the start of a match later than its end. In this 471 demonstration program, we just detect this case and give up. */ 472 473 if (ovector[0] > ovector[1]) 474 { 475 printf("\\K was used in an assertion to set the match start after its end.\n" 476 "From end to start the match was: %.*s\n", (int)(ovector[0] - ovector[1]), 477 (char *)(subject + ovector[1])); 478 printf("Run abandoned\n"); 479 pcre2_match_data_free(match_data); 480 pcre2_code_free(re); 481 return 1; 482 } 483 484 /* As before, show substrings stored in the output vector by number, and then 485 also any named substrings. */ 486 487 for (i = 0; i < rc; i++) 488 { 489 PCRE2_SPTR substring_start = subject + ovector[2*i]; 490 size_t substring_length = ovector[2*i+1] - ovector[2*i]; 491 printf("%2d: %.*s\n", i, (int)substring_length, (char *)substring_start); 492 } 493 494 if (namecount == 0) printf("No named substrings\n"); else 495 { 496 PCRE2_SPTR tabptr = name_table; 497 printf("Named substrings\n"); 498 for (i = 0; i < namecount; i++) 499 { 500 int n = (tabptr[0] << 8) | tabptr[1]; 501 printf("(%d) %*s: %.*s\n", n, name_entry_size - 3, tabptr + 2, 502 (int)(ovector[2*n+1] - ovector[2*n]), subject + ovector[2*n]); 503 tabptr += name_entry_size; 504 } 505 } 506 } /* End of loop to find second and subsequent matches */ 507 508printf("\n"); 509pcre2_match_data_free(match_data); 510pcre2_code_free(re); 511return 0; 512} 513 514/* End of pcre2demo.c */ 515<p> 516Return to the <a href="index.html">PCRE2 index page</a>. 517</p> 518