• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
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 
main(int argc,char ** argv)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 succeded. Get a pointer to the output vector, where string offsets are
202 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 /* We must guard against patterns such as /(?=.\K)/ that use \K in an assertion
221 to set the start of a match later than its end. In this demonstration program,
222 we just detect this case and give up. */
223 
224 if (ovector[0] > ovector[1])
225   {
226   printf("\\K was used in an assertion to set the match start after its end.\n"
227     "From end to start the match was: %.*s\n", (int)(ovector[0] - ovector[1]),
228       (char *)(subject + ovector[1]));
229   printf("Run abandoned\n");
230   pcre2_match_data_free(match_data);
231   pcre2_code_free(re);
232   return 1;
233   }
234 
235 /* Show substrings stored in the output vector by number. Obviously, in a real
236 application you might want to do things other than print them. */
237 
238 for (i = 0; i < rc; i++)
239   {
240   PCRE2_SPTR substring_start = subject + ovector[2*i];
241   PCRE2_SIZE substring_length = ovector[2*i+1] - ovector[2*i];
242   printf("%2d: %.*s\n", i, (int)substring_length, (char *)substring_start);
243   }
244 
245 
246 /**************************************************************************
247 * That concludes the basic part of this demonstration program. We have    *
248 * compiled a pattern, and performed a single match. The code that follows *
249 * shows first how to access named substrings, and then how to code for    *
250 * repeated matches on the same subject.                                   *
251 **************************************************************************/
252 
253 /* See if there are any named substrings, and if so, show them by name. First
254 we have to extract the count of named parentheses from the pattern. */
255 
256 (void)pcre2_pattern_info(
257   re,                   /* the compiled pattern */
258   PCRE2_INFO_NAMECOUNT, /* get the number of named substrings */
259   &namecount);          /* where to put the answer */
260 
261 if (namecount == 0) printf("No named substrings\n"); else
262   {
263   PCRE2_SPTR tabptr;
264   printf("Named substrings\n");
265 
266   /* Before we can access the substrings, we must extract the table for
267   translating names to numbers, and the size of each entry in the table. */
268 
269   (void)pcre2_pattern_info(
270     re,                       /* the compiled pattern */
271     PCRE2_INFO_NAMETABLE,     /* address of the table */
272     &name_table);             /* where to put the answer */
273 
274   (void)pcre2_pattern_info(
275     re,                       /* the compiled pattern */
276     PCRE2_INFO_NAMEENTRYSIZE, /* size of each entry in the table */
277     &name_entry_size);        /* where to put the answer */
278 
279   /* Now we can scan the table and, for each entry, print the number, the name,
280   and the substring itself. In the 8-bit library the number is held in two
281   bytes, most significant first. */
282 
283   tabptr = name_table;
284   for (i = 0; i < namecount; i++)
285     {
286     int n = (tabptr[0] << 8) | tabptr[1];
287     printf("(%d) %*s: %.*s\n", n, name_entry_size - 3, tabptr + 2,
288       (int)(ovector[2*n+1] - ovector[2*n]), subject + ovector[2*n]);
289     tabptr += name_entry_size;
290     }
291   }
292 
293 
294 /*************************************************************************
295 * If the "-g" option was given on the command line, we want to continue  *
296 * to search for additional matches in the subject string, in a similar   *
297 * way to the /g option in Perl. This turns out to be trickier than you   *
298 * might think because of the possibility of matching an empty string.    *
299 * What happens is as follows:                                            *
300 *                                                                        *
301 * If the previous match was NOT for an empty string, we can just start   *
302 * the next match at the end of the previous one.                         *
303 *                                                                        *
304 * If the previous match WAS for an empty string, we can't do that, as it *
305 * would lead to an infinite loop. Instead, a call of pcre2_match() is    *
306 * made with the PCRE2_NOTEMPTY_ATSTART and PCRE2_ANCHORED flags set. The *
307 * first of these tells PCRE2 that an empty string at the start of the    *
308 * subject is not a valid match; other possibilities must be tried. The   *
309 * second flag restricts PCRE2 to one match attempt at the initial string *
310 * position. If this match succeeds, an alternative to the empty string   *
311 * match has been found, and we can print it and proceed round the loop,  *
312 * advancing by the length of whatever was found. If this match does not  *
313 * succeed, we still stay in the loop, advancing by just one character.   *
314 * In UTF-8 mode, which can be set by (*UTF) in the pattern, this may be  *
315 * more than one byte.                                                    *
316 *                                                                        *
317 * However, there is a complication concerned with newlines. When the     *
318 * newline convention is such that CRLF is a valid newline, we must       *
319 * advance by two characters rather than one. The newline convention can  *
320 * be set in the regex by (*CR), etc.; if not, we must find the default.  *
321 *************************************************************************/
322 
323 if (!find_all)     /* Check for -g */
324   {
325   pcre2_match_data_free(match_data);  /* Release the memory that was used */
326   pcre2_code_free(re);                /* for the match data and the pattern. */
327   return 0;                           /* Exit the program. */
328   }
329 
330 /* Before running the loop, check for UTF-8 and whether CRLF is a valid newline
331 sequence. First, find the options with which the regex was compiled and extract
332 the UTF state. */
333 
334 (void)pcre2_pattern_info(re, PCRE2_INFO_ALLOPTIONS, &option_bits);
335 utf8 = (option_bits & PCRE2_UTF) != 0;
336 
337 /* Now find the newline convention and see whether CRLF is a valid newline
338 sequence. */
339 
340 (void)pcre2_pattern_info(re, PCRE2_INFO_NEWLINE, &newline);
341 crlf_is_newline = newline == PCRE2_NEWLINE_ANY ||
342                   newline == PCRE2_NEWLINE_CRLF ||
343                   newline == PCRE2_NEWLINE_ANYCRLF;
344 
345 /* Loop for second and subsequent matches */
346 
347 for (;;)
348   {
349   uint32_t options = 0;                   /* Normally no options */
350   PCRE2_SIZE start_offset = ovector[1];   /* Start at end of previous match */
351 
352   /* If the previous match was for an empty string, we are finished if we are
353   at the end of the subject. Otherwise, arrange to run another match at the
354   same point to see if a non-empty match can be found. */
355 
356   if (ovector[0] == ovector[1])
357     {
358     if (ovector[0] == subject_length) break;
359     options = PCRE2_NOTEMPTY_ATSTART | PCRE2_ANCHORED;
360     }
361 
362   /* If the previous match was not an empty string, there is one tricky case to
363   consider. If a pattern contains \K within a lookbehind assertion at the
364   start, the end of the matched string can be at the offset where the match
365   started. Without special action, this leads to a loop that keeps on matching
366   the same substring. We must detect this case and arrange to move the start on
367   by one character. The pcre2_get_startchar() function returns the starting
368   offset that was passed to pcre2_match(). */
369 
370   else
371     {
372     PCRE2_SIZE startchar = pcre2_get_startchar(match_data);
373     if (start_offset <= startchar)
374       {
375       if (startchar >= subject_length) break;   /* Reached end of subject.   */
376       start_offset = startchar + 1;             /* Advance by one character. */
377       if (utf8)                                 /* If UTF-8, it may be more  */
378         {                                       /*   than one code unit.     */
379         for (; start_offset < subject_length; start_offset++)
380           if ((subject[start_offset] & 0xc0) != 0x80) break;
381         }
382       }
383     }
384 
385   /* Run the next matching operation */
386 
387   rc = pcre2_match(
388     re,                   /* the compiled pattern */
389     subject,              /* the subject string */
390     subject_length,       /* the length of the subject */
391     start_offset,         /* starting offset in the subject */
392     options,              /* options */
393     match_data,           /* block for storing the result */
394     NULL);                /* use default match context */
395 
396   /* This time, a result of NOMATCH isn't an error. If the value in "options"
397   is zero, it just means we have found all possible matches, so the loop ends.
398   Otherwise, it means we have failed to find a non-empty-string match at a
399   point where there was a previous empty-string match. In this case, we do what
400   Perl does: advance the matching position by one character, and continue. We
401   do this by setting the "end of previous match" offset, because that is picked
402   up at the top of the loop as the point at which to start again.
403 
404   There are two complications: (a) When CRLF is a valid newline sequence, and
405   the current position is just before it, advance by an extra byte. (b)
406   Otherwise we must ensure that we skip an entire UTF character if we are in
407   UTF mode. */
408 
409   if (rc == PCRE2_ERROR_NOMATCH)
410     {
411     if (options == 0) break;                    /* All matches found */
412     ovector[1] = start_offset + 1;              /* Advance one code unit */
413     if (crlf_is_newline &&                      /* If CRLF is a newline & */
414         start_offset < subject_length - 1 &&    /* we are at CRLF, */
415         subject[start_offset] == '\r' &&
416         subject[start_offset + 1] == '\n')
417       ovector[1] += 1;                          /* Advance by one more. */
418     else if (utf8)                              /* Otherwise, ensure we */
419       {                                         /* advance a whole UTF-8 */
420       while (ovector[1] < subject_length)       /* character. */
421         {
422         if ((subject[ovector[1]] & 0xc0) != 0x80) break;
423         ovector[1] += 1;
424         }
425       }
426     continue;    /* Go round the loop again */
427     }
428 
429   /* Other matching errors are not recoverable. */
430 
431   if (rc < 0)
432     {
433     printf("Matching error %d\n", rc);
434     pcre2_match_data_free(match_data);
435     pcre2_code_free(re);
436     return 1;
437     }
438 
439   /* Match succeded */
440 
441   printf("\nMatch succeeded again at offset %d\n", (int)ovector[0]);
442 
443   /* The match succeeded, but the output vector wasn't big enough. This
444   should not happen. */
445 
446   if (rc == 0)
447     printf("ovector was not big enough for all the captured substrings\n");
448 
449   /* We must guard against patterns such as /(?=.\K)/ that use \K in an
450   assertion to set the start of a match later than its end. In this
451   demonstration program, we just detect this case and give up. */
452 
453   if (ovector[0] > ovector[1])
454     {
455     printf("\\K was used in an assertion to set the match start after its end.\n"
456       "From end to start the match was: %.*s\n", (int)(ovector[0] - ovector[1]),
457         (char *)(subject + ovector[1]));
458     printf("Run abandoned\n");
459     pcre2_match_data_free(match_data);
460     pcre2_code_free(re);
461     return 1;
462     }
463 
464   /* As before, show substrings stored in the output vector by number, and then
465   also any named substrings. */
466 
467   for (i = 0; i < rc; i++)
468     {
469     PCRE2_SPTR substring_start = subject + ovector[2*i];
470     size_t substring_length = ovector[2*i+1] - ovector[2*i];
471     printf("%2d: %.*s\n", i, (int)substring_length, (char *)substring_start);
472     }
473 
474   if (namecount == 0) printf("No named substrings\n"); else
475     {
476     PCRE2_SPTR tabptr = name_table;
477     printf("Named substrings\n");
478     for (i = 0; i < namecount; i++)
479       {
480       int n = (tabptr[0] << 8) | tabptr[1];
481       printf("(%d) %*s: %.*s\n", n, name_entry_size - 3, tabptr + 2,
482         (int)(ovector[2*n+1] - ovector[2*n]), subject + ovector[2*n]);
483       tabptr += name_entry_size;
484       }
485     }
486   }      /* End of loop to find second and subsequent matches */
487 
488 printf("\n");
489 pcre2_match_data_free(match_data);
490 pcre2_code_free(re);
491 return 0;
492 }
493 
494 /* End of pcre2demo.c */
495