18 /*************************************************
19 * PCRE2 DEMONSTRATION PROGRAM *
20 *************************************************/
22 /* This is a demonstration program to illustrate a straightforward way of
23 using the PCRE2 regular expression library from a C program. See the
24 pcre2sample documentation for a short discussion ("man pcre2sample" if you have
25 the PCRE2 man pages installed). PCRE2 is a revised API for the library, and is
26 incompatible with the original PCRE API.
28 There are actually three libraries, each supporting a different code unit
29 width. This demonstration program uses the 8-bit library. The default is to
30 process 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
32 characters may occupy multiple code units.
34 In Unix-like environments, if PCRE2 is installed in your standard system
35 libraries, you should be able to compile this program using this command:
37 cc -Wall pcre2demo.c -lpcre2-8 -o pcre2demo
39 If PCRE2 is not installed in a standard place, it is likely to be installed
40 with support for the pkg-config mechanism. If you have pkg-config, you can
41 compile this program using this command:
43 cc -Wall pcre2demo.c `pkg-config --cflags --libs libpcre2-8` -o pcre2demo
45 If you do not have pkg-config, you may have to use something like this:
47 cc -Wall pcre2demo.c -I/usr/local/include -L/usr/local/lib \e
48 -R/usr/local/lib -lpcre2-8 -o pcre2demo
50 Replace "/usr/local/include" and "/usr/local/lib" with wherever the include and
51 library files for PCRE2 are installed on your system. Only some operating
52 systems (Solaris is one) use the -R option.
54 Building under Windows:
56 If you want to statically link this program against a non-dll .a file, you must
57 define PCRE2_STATIC before including pcre2.h, so in this environment, uncomment
58 the following line. */
60 /* #define PCRE2_STATIC */
62 /* The PCRE2_CODE_UNIT_WIDTH macro must be defined before including pcre2.h.
63 For a program that uses only one code unit width, setting it to 8, 16, or 32
64 makes it possible to use generic function names such as pcre2_compile(). Note
65 that just changing 8 to 16 (for example) is not sufficient to convert this
66 program to process 16-bit characters. Even in a fully 16-bit environment, where
67 string-handling functions such as strcmp() and printf() work with 16-bit
68 characters, the code for handling the table of named substrings will still need
71 #define PCRE2_CODE_UNIT_WIDTH 8
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 **************************************************************************/
86 int main(int argc, char **argv)
89 PCRE2_SPTR pattern; /* PCRE2_SPTR is a pointer to unsigned code units of */
90 PCRE2_SPTR subject; /* the appropriate width (in this case, 8 bits). */
91 PCRE2_SPTR name_table;
100 uint32_t option_bits;
102 uint32_t name_entry_size;
105 PCRE2_SIZE erroroffset;
108 size_t subject_length;
109 pcre2_match_data *match_data;
113 /**************************************************************************
114 * First, sort out the command line. There is only one possible option at *
115 * the moment, "-g" to request repeated matching to find all occurrences, *
116 * like Perl's /g option. We set the variable find_all to a non-zero value *
117 * if the -g option is present. *
118 **************************************************************************/
121 for (i = 1; i < argc; i++)
123 if (strcmp(argv[i], "-g") == 0) find_all = 1;
124 else if (argv[i][0] == '-')
126 printf("Unrecognised option %s\en", argv[i]);
132 /* After the options, we require exactly two arguments, which are the pattern,
133 and the subject string. */
137 printf("Exactly two arguments required: a regex and a subject string\en");
141 /* As pattern and subject are char arguments, they can be straightforwardly
142 cast to PCRE2_SPTR as we are working in 8-bit code units. */
144 pattern = (PCRE2_SPTR)argv[i];
145 subject = (PCRE2_SPTR)argv[i+1];
146 subject_length = strlen((char *)subject);
149 /*************************************************************************
150 * Now we are going to compile the regular expression pattern, and handle *
151 * any errors that are detected. *
152 *************************************************************************/
155 pattern, /* the pattern */
156 PCRE2_ZERO_TERMINATED, /* indicates pattern is zero-terminated */
157 0, /* default options */
158 &errornumber, /* for error number */
159 &erroroffset, /* for error offset */
160 NULL); /* use default compile context */
162 /* Compilation failed: print the error message and exit. */
166 PCRE2_UCHAR buffer[256];
167 pcre2_get_error_message(errornumber, buffer, sizeof(buffer));
168 printf("PCRE2 compilation failed at offset %d: %s\en", (int)erroroffset,
174 /*************************************************************************
175 * If the compilation succeeded, we call PCRE again, in order to do a *
176 * pattern match against the subject string. This does just ONE match. If *
177 * further matching is needed, it will be done below. Before running the *
178 * match we must set up a match_data block for holding the result. *
179 *************************************************************************/
181 /* Using this function ensures that the block is exactly the right size for
182 the number of capturing parentheses in the pattern. */
184 match_data = pcre2_match_data_create_from_pattern(re, NULL);
187 re, /* the compiled pattern */
188 subject, /* the subject string */
189 subject_length, /* the length of the subject */
190 0, /* start at offset 0 in the subject */
191 0, /* default options */
192 match_data, /* block for storing the result */
193 NULL); /* use default match context */
195 /* Matching failed: handle error cases */
201 case PCRE2_ERROR_NOMATCH: printf("No match\en"); break;
203 Handle other special cases if you like
205 default: printf("Matching error %d\en", rc); break;
207 pcre2_match_data_free(match_data); /* Release memory used for the match */
208 pcre2_code_free(re); /* data and the compiled pattern. */
212 /* Match succeded. Get a pointer to the output vector, where string offsets are
215 ovector = pcre2_get_ovector_pointer(match_data);
216 printf("Match succeeded at offset %d\en", (int)ovector[0]);
219 /*************************************************************************
220 * We have found the first match within the subject string. If the output *
221 * vector wasn't big enough, say so. Then output any substrings that were *
223 *************************************************************************/
225 /* The output vector wasn't big enough. This should not happen, because we used
226 pcre2_match_data_create_from_pattern() above. */
229 printf("ovector was not big enough for all the captured substrings\en");
231 /* We must guard against patterns such as /(?=.\eK)/ that use \eK in an assertion
232 to set the start of a match later than its end. In this demonstration program,
233 we just detect this case and give up. */
235 if (ovector[0] > ovector[1])
237 printf("\e\eK was used in an assertion to set the match start after its end.\en"
238 "From end to start the match was: %.*s\en", (int)(ovector[0] - ovector[1]),
239 (char *)(subject + ovector[1]));
240 printf("Run abandoned\en");
241 pcre2_match_data_free(match_data);
246 /* Show substrings stored in the output vector by number. Obviously, in a real
247 application you might want to do things other than print them. */
249 for (i = 0; i < rc; i++)
251 PCRE2_SPTR substring_start = subject + ovector[2*i];
252 size_t substring_length = ovector[2*i+1] - ovector[2*i];
253 printf("%2d: %.*s\en", i, (int)substring_length, (char *)substring_start);
257 /**************************************************************************
258 * That concludes the basic part of this demonstration program. We have *
259 * compiled a pattern, and performed a single match. The code that follows *
260 * shows first how to access named substrings, and then how to code for *
261 * repeated matches on the same subject. *
262 **************************************************************************/
264 /* See if there are any named substrings, and if so, show them by name. First
265 we have to extract the count of named parentheses from the pattern. */
267 (void)pcre2_pattern_info(
268 re, /* the compiled pattern */
269 PCRE2_INFO_NAMECOUNT, /* get the number of named substrings */
270 &namecount); /* where to put the answer */
272 if (namecount == 0) printf("No named substrings\en"); else
275 printf("Named substrings\en");
277 /* Before we can access the substrings, we must extract the table for
278 translating names to numbers, and the size of each entry in the table. */
280 (void)pcre2_pattern_info(
281 re, /* the compiled pattern */
282 PCRE2_INFO_NAMETABLE, /* address of the table */
283 &name_table); /* where to put the answer */
285 (void)pcre2_pattern_info(
286 re, /* the compiled pattern */
287 PCRE2_INFO_NAMEENTRYSIZE, /* size of each entry in the table */
288 &name_entry_size); /* where to put the answer */
290 /* Now we can scan the table and, for each entry, print the number, the name,
291 and the substring itself. In the 8-bit library the number is held in two
292 bytes, most significant first. */
295 for (i = 0; i < namecount; i++)
297 int n = (tabptr[0] << 8) | tabptr[1];
298 printf("(%d) %*s: %.*s\en", n, name_entry_size - 3, tabptr + 2,
299 (int)(ovector[2*n+1] - ovector[2*n]), subject + ovector[2*n]);
300 tabptr += name_entry_size;
305 /*************************************************************************
306 * If the "-g" option was given on the command line, we want to continue *
307 * to search for additional matches in the subject string, in a similar *
308 * way to the /g option in Perl. This turns out to be trickier than you *
309 * might think because of the possibility of matching an empty string. *
310 * What happens is as follows: *
312 * If the previous match was NOT for an empty string, we can just start *
313 * the next match at the end of the previous one. *
315 * If the previous match WAS for an empty string, we can't do that, as it *
316 * would lead to an infinite loop. Instead, a call of pcre2_match() is *
317 * made with the PCRE2_NOTEMPTY_ATSTART and PCRE2_ANCHORED flags set. The *
318 * first of these tells PCRE2 that an empty string at the start of the *
319 * subject is not a valid match; other possibilities must be tried. The *
320 * second flag restricts PCRE2 to one match attempt at the initial string *
321 * position. If this match succeeds, an alternative to the empty string *
322 * match has been found, and we can print it and proceed round the loop, *
323 * advancing by the length of whatever was found. If this match does not *
324 * succeed, we still stay in the loop, advancing by just one character. *
325 * In UTF-8 mode, which can be set by (*UTF) in the pattern, this may be *
326 * more than one byte. *
328 * However, there is a complication concerned with newlines. When the *
329 * newline convention is such that CRLF is a valid newline, we must *
330 * advance by two characters rather than one. The newline convention can *
331 * be set in the regex by (*CR), etc.; if not, we must find the default. *
332 *************************************************************************/
334 if (!find_all) /* Check for -g */
336 pcre2_match_data_free(match_data); /* Release the memory that was used */
337 pcre2_code_free(re); /* for the match data and the pattern. */
338 return 0; /* Exit the program. */
341 /* Before running the loop, check for UTF-8 and whether CRLF is a valid newline
342 sequence. First, find the options with which the regex was compiled and extract
345 (void)pcre2_pattern_info(re, PCRE2_INFO_ALLOPTIONS, &option_bits);
346 utf8 = (option_bits & PCRE2_UTF) != 0;
348 /* Now find the newline convention and see whether CRLF is a valid newline
351 (void)pcre2_pattern_info(re, PCRE2_INFO_NEWLINE, &newline);
352 crlf_is_newline = newline == PCRE2_NEWLINE_ANY ||
353 newline == PCRE2_NEWLINE_CRLF ||
354 newline == PCRE2_NEWLINE_ANYCRLF;
356 /* Loop for second and subsequent matches */
360 uint32_t options = 0; /* Normally no options */
361 PCRE2_SIZE start_offset = ovector[1]; /* Start at end of previous match */
363 /* If the previous match was for an empty string, we are finished if we are
364 at the end of the subject. Otherwise, arrange to run another match at the
365 same point to see if a non-empty match can be found. */
367 if (ovector[0] == ovector[1])
369 if (ovector[0] == subject_length) break;
370 options = PCRE2_NOTEMPTY_ATSTART | PCRE2_ANCHORED;
373 /* If the previous match was not an empty string, there is one tricky case to
374 consider. If a pattern contains \eK within a lookbehind assertion at the
375 start, the end of the matched string can be at the offset where the match
376 started. Without special action, this leads to a loop that keeps on matching
377 the same substring. We must detect this case and arrange to move the start on
378 by one character. The pcre2_get_startchar() function returns the starting
379 offset that was passed to pcre2_match(). */
383 PCRE2_SIZE startchar = pcre2_get_startchar(match_data);
384 if (start_offset <= startchar)
386 if (startchar >= subject_length) break; /* Reached end of subject. */
387 start_offset = startchar + 1; /* Advance by one character. */
388 if (utf8) /* If UTF-8, it may be more */
389 { /* than one code unit. */
390 for (; start_offset < subject_length; start_offset++)
391 if ((subject[start_offset] & 0xc0) != 0x80) break;
396 /* Run the next matching operation */
399 re, /* the compiled pattern */
400 subject, /* the subject string */
401 subject_length, /* the length of the subject */
402 start_offset, /* starting offset in the subject */
403 options, /* options */
404 match_data, /* block for storing the result */
405 NULL); /* use default match context */
407 /* This time, a result of NOMATCH isn't an error. If the value in "options"
408 is zero, it just means we have found all possible matches, so the loop ends.
409 Otherwise, it means we have failed to find a non-empty-string match at a
410 point where there was a previous empty-string match. In this case, we do what
411 Perl does: advance the matching position by one character, and continue. We
412 do this by setting the "end of previous match" offset, because that is picked
413 up at the top of the loop as the point at which to start again.
415 There are two complications: (a) When CRLF is a valid newline sequence, and
416 the current position is just before it, advance by an extra byte. (b)
417 Otherwise we must ensure that we skip an entire UTF character if we are in
420 if (rc == PCRE2_ERROR_NOMATCH)
422 if (options == 0) break; /* All matches found */
423 ovector[1] = start_offset + 1; /* Advance one code unit */
424 if (crlf_is_newline && /* If CRLF is a newline & */
425 start_offset < subject_length - 1 && /* we are at CRLF, */
426 subject[start_offset] == '\er' &&
427 subject[start_offset + 1] == '\en')
428 ovector[1] += 1; /* Advance by one more. */
429 else if (utf8) /* Otherwise, ensure we */
430 { /* advance a whole UTF-8 */
431 while (ovector[1] < subject_length) /* character. */
433 if ((subject[ovector[1]] & 0xc0) != 0x80) break;
437 continue; /* Go round the loop again */
440 /* Other matching errors are not recoverable. */
444 printf("Matching error %d\en", rc);
445 pcre2_match_data_free(match_data);
452 printf("\enMatch succeeded again at offset %d\en", (int)ovector[0]);
454 /* The match succeeded, but the output vector wasn't big enough. This
455 should not happen. */
458 printf("ovector was not big enough for all the captured substrings\en");
460 /* We must guard against patterns such as /(?=.\eK)/ that use \eK in an
461 assertion to set the start of a match later than its end. In this
462 demonstration program, we just detect this case and give up. */
464 if (ovector[0] > ovector[1])
466 printf("\e\eK was used in an assertion to set the match start after its end.\en"
467 "From end to start the match was: %.*s\en", (int)(ovector[0] - ovector[1]),
468 (char *)(subject + ovector[1]));
469 printf("Run abandoned\en");
470 pcre2_match_data_free(match_data);
475 /* As before, show substrings stored in the output vector by number, and then
476 also any named substrings. */
478 for (i = 0; i < rc; i++)
480 PCRE2_SPTR substring_start = subject + ovector[2*i];
481 size_t substring_length = ovector[2*i+1] - ovector[2*i];
482 printf("%2d: %.*s\en", i, (int)substring_length, (char *)substring_start);
485 if (namecount == 0) printf("No named substrings\en"); else
487 PCRE2_SPTR tabptr = name_table;
488 printf("Named substrings\en");
489 for (i = 0; i < namecount; i++)
491 int n = (tabptr[0] << 8) | tabptr[1];
492 printf("(%d) %*s: %.*s\en", n, name_entry_size - 3, tabptr + 2,
493 (int)(ovector[2*n+1] - ovector[2*n]), subject + ovector[2*n]);
494 tabptr += name_entry_size;
497 } /* End of loop to find second and subsequent matches */
500 pcre2_match_data_free(match_data);
505 /* End of pcre2demo.c */