pcretest.c

PHP v6.0 For Linux 运行环境：Win9X/ WinME/ WinNT/ Win2K/ WinXP

            fprintf(outfile, "no parentheses with name \"%s\"\n", name);
          else copystrings |= 1 << n;
          }
        else if (*p == '+')
          {
          callout_extra = 1;
          p++;
          }
        else if (*p == '-')
          {
          pcre_callout = NULL;
          p++;
          }
        else if (*p == '!')
          {
          callout_fail_id = 0;
          p++;
          while(isdigit(*p))
            callout_fail_id = callout_fail_id * 10 + *p++ - '0';
          callout_fail_count = 0;
          if (*p == '!')
            {
            p++;
            while(isdigit(*p))
              callout_fail_count = callout_fail_count * 10 + *p++ - '0';
            }
          }
        else if (*p == '*')
          {
          int sign = 1;
          callout_data = 0;
          if (*(++p) == '-') { sign = -1; p++; }
          while(isdigit(*p))
            callout_data = callout_data * 10 + *p++ - '0';
          callout_data *= sign;
          callout_data_set = 1;
          }
        continue;

#if !defined NODFA
        case 'D':
#if !defined NOPOSIX
        if (posix || do_posix)
          printf("** Can't use dfa matching in POSIX mode: \\D ignored\n");
        else
#endif
          use_dfa = 1;
        continue;

        case 'F':
        options |= PCRE_DFA_SHORTEST;
        continue;
#endif

        case 'G':
        if (isdigit(*p))
          {
          while(isdigit(*p)) n = n * 10 + *p++ - '0';
          getstrings |= 1 << n;
          }
        else if (isalnum(*p))
          {
          uschar name[256];
          uschar *npp = name;
          while (isalnum(*p)) *npp++ = *p++;
          *npp = 0;
          n = pcre_get_stringnumber(re, (char *)name);
          if (n < 0)
            fprintf(outfile, "no parentheses with name \"%s\"\n", name);
          else getstrings |= 1 << n;
          }
        continue;

        case 'L':
        getlist = 1;
        continue;

        case 'M':
        find_match_limit = 1;
        continue;

        case 'N':
        options |= PCRE_NOTEMPTY;
        continue;

        case 'O':
        while(isdigit(*p)) n = n * 10 + *p++ - '0';
        if (n > size_offsets_max)
          {
          size_offsets_max = n;
          free(offsets);
          use_offsets = offsets = (int *)malloc(size_offsets_max * sizeof(int));
          if (offsets == NULL)
            {
            printf("** Failed to get %d bytes of memory for offsets vector\n",
              size_offsets_max * sizeof(int));
            yield = 1;
            goto EXIT;
            }
          }
        use_size_offsets = n;
        if (n == 0) use_offsets = NULL;   /* Ensures it can't write to it */
        continue;

        case 'P':
        options |= PCRE_PARTIAL;
        continue;

#if !defined NODFA
        case 'R':
        options |= PCRE_DFA_RESTART;
        continue;
#endif

        case 'S':
        show_malloc = 1;
        continue;

        case 'Z':
        options |= PCRE_NOTEOL;
        continue;

        case '?':
        options |= PCRE_NO_UTF8_CHECK;
        continue;
        }
      *q++ = c;
      }
    *q = 0;
    len = q - dbuffer;

    if ((all_use_dfa || use_dfa) && find_match_limit)
      {
      printf("**Match limit not relevant for DFA matching: ignored\n");
      find_match_limit = 0;
      }

    /* Handle matching via the POSIX interface, which does not
    support timing or playing with the match limit or callout data. */

#if !defined NOPOSIX
    if (posix || do_posix)
      {
      int rc;
      int eflags = 0;
      regmatch_t *pmatch = NULL;
      if (use_size_offsets > 0)
        pmatch = (regmatch_t *)malloc(sizeof(regmatch_t) * use_size_offsets);
      if ((options & PCRE_NOTBOL) != 0) eflags |= REG_NOTBOL;
      if ((options & PCRE_NOTEOL) != 0) eflags |= REG_NOTEOL;

      rc = regexec(&preg, (const char *)bptr, use_size_offsets, pmatch, eflags);

      if (rc != 0)
        {
        (void)regerror(rc, &preg, (char *)buffer, BUFFER_SIZE);
        fprintf(outfile, "No match: POSIX code %d: %s\n", rc, buffer);
        }
      else if ((((const pcre *)preg.re_pcre)->options & PCRE_NO_AUTO_CAPTURE)
              != 0)
        {
        fprintf(outfile, "Matched with REG_NOSUB\n");
        }
      else
        {
        size_t i;
        for (i = 0; i < (size_t)use_size_offsets; i++)
          {
          if (pmatch[i].rm_so >= 0)
            {
            fprintf(outfile, "%2d: ", (int)i);
            (void)pchars(dbuffer + pmatch[i].rm_so,
              pmatch[i].rm_eo - pmatch[i].rm_so, outfile);
            fprintf(outfile, "\n");
            if (i == 0 && do_showrest)
              {
              fprintf(outfile, " 0+ ");
              (void)pchars(dbuffer + pmatch[i].rm_eo, len - pmatch[i].rm_eo,
                outfile);
              fprintf(outfile, "\n");
              }
            }
          }
        }
      free(pmatch);
      }

    /* Handle matching via the native interface - repeats for /g and /G */

    else
#endif  /* !defined NOPOSIX */

    for (;; gmatched++)    /* Loop for /g or /G */
      {
      if (timeit)
        {
        register int i;
        clock_t time_taken;
        clock_t start_time = clock();

#if !defined NODFA
        if (all_use_dfa || use_dfa)
          {
          int workspace[1000];
          for (i = 0; i < LOOPREPEAT; i++)
            count = pcre_dfa_exec(re, NULL, (char *)bptr, len, start_offset,
              options | g_notempty, use_offsets, use_size_offsets, workspace,
              sizeof(workspace)/sizeof(int));
          }
        else
#endif

        for (i = 0; i < LOOPREPEAT; i++)
          count = pcre_exec(re, extra, (char *)bptr, len,
            start_offset, options | g_notempty, use_offsets, use_size_offsets);

        time_taken = clock() - start_time;
        fprintf(outfile, "Execute time %.3f milliseconds\n",
          (((double)time_taken * 1000.0) / (double)LOOPREPEAT) /
            (double)CLOCKS_PER_SEC);
        }

      /* If find_match_limit is set, we want to do repeated matches with
      varying limits in order to find the minimum value for the match limit and
      for the recursion limit. */

      if (find_match_limit)
        {
        if (extra == NULL)
          {
          extra = (pcre_extra *)malloc(sizeof(pcre_extra));
          extra->flags = 0;
          }

        count = check_match_limit(re, extra, bptr, len, start_offset,
          options|g_notempty, use_offsets, use_size_offsets,
          PCRE_EXTRA_MATCH_LIMIT, &(extra->match_limit),
          PCRE_ERROR_MATCHLIMIT, "match()");

        count = check_match_limit(re, extra, bptr, len, start_offset,
          options|g_notempty, use_offsets, use_size_offsets,
          PCRE_EXTRA_MATCH_LIMIT_RECURSION, &(extra->match_limit_recursion),
          PCRE_ERROR_RECURSIONLIMIT, "match() recursion");
        }

      /* If callout_data is set, use the interface with additional data */

      else if (callout_data_set)
        {
        if (extra == NULL)
          {
          extra = (pcre_extra *)malloc(sizeof(pcre_extra));
          extra->flags = 0;
          }
        extra->flags |= PCRE_EXTRA_CALLOUT_DATA;
        extra->callout_data = &callout_data;
        count = pcre_exec(re, extra, (char *)bptr, len, start_offset,
          options | g_notempty, use_offsets, use_size_offsets);
        extra->flags &= ~PCRE_EXTRA_CALLOUT_DATA;
        }

      /* The normal case is just to do the match once, with the default
      value of match_limit. */

#if !defined NODFA
      else if (all_use_dfa || use_dfa)
        {
        int workspace[1000];
        count = pcre_dfa_exec(re, NULL, (char *)bptr, len, start_offset,
          options | g_notempty, use_offsets, use_size_offsets, workspace,
          sizeof(workspace)/sizeof(int));
        if (count == 0)
          {
          fprintf(outfile, "Matched, but too many subsidiary matches\n");
          count = use_size_offsets/2;
          }
        }
#endif

      else
        {
        count = pcre_exec(re, extra, (char *)bptr, len,
          start_offset, options | g_notempty, use_offsets, use_size_offsets);
        if (count == 0)
          {
          fprintf(outfile, "Matched, but too many substrings\n");
          count = use_size_offsets/3;
          }
        }

      /* Matched */

      if (count >= 0)
        {
        int i;
        for (i = 0; i < count * 2; i += 2)
          {
          if (use_offsets[i] < 0)
            fprintf(outfile, "%2d: <unset>\n", i/2);
          else
            {
            fprintf(outfile, "%2d: ", i/2);
            (void)pchars(bptr + use_offsets[i],
              use_offsets[i+1] - use_offsets[i], outfile);
            fprintf(outfile, "\n");
            if (i == 0)
              {
              if (do_showrest)
                {
                fprintf(outfile, " 0+ ");
                (void)pchars(bptr + use_offsets[i+1], len - use_offsets[i+1],
                  outfile);
                fprintf(outfile, "\n");
                }
              }
            }
          }

        for (i = 0; i < 32; i++)
          {
          if ((copystrings & (1 << i)) != 0)
            {
            char copybuffer[16];
            int rc = pcre_copy_substring((char *)bptr, use_offsets, count,
              i, copybuffer, sizeof(copybuffer));
            if (rc < 0)
              fprintf(outfile, "copy substring %d failed %d\n", i, rc);
            else
              fprintf(outfile, "%2dC %s (%d)\n", i, copybuffer, rc);
            }
          }

        for (i = 0; i < 32; i++)
          {
          if ((getstrings & (1 << i)) != 0)
            {
            const char *substring;
            int rc = pcre_get_substring((char *)bptr, use_offsets, count,
              i, &substring);
            if (rc < 0)
              fprintf(outfile, "get substring %d failed %d\n", i, rc);
            else
              {
              fprintf(outfile, "%2dG %s (%d)\n", i, substring, rc);
              /* free((void *)substring); */
              pcre_free_substring(substring);
              }
            }
          }

        if (getlist)
          {
          const char **stringlist;
          int rc = pcre_get_substring_list((char *)bptr, use_offsets, count,
            &stringlist);
          if (rc < 0)
            fprintf(outfile, "get substring list failed %d\n", rc);
          else
            {
            for (i = 0; i < count; i++)
              fprintf(outfile, "%2dL %s\n", i, stringlist[i]);
            if (stringlist[i] != NULL)
              fprintf(outfile, "string list not terminated by NULL\n");
            /* free((void *)stringlist); */
            pcre_free_substring_list(stringlist);
            }
          }
        }

      /* There was a partial match */

      else if (count == PCRE_ERROR_PARTIAL)
        {
        fprintf(outfile, "Partial match");
#if !defined NODFA
        if ((all_use_dfa || use_dfa) && use_size_offsets > 2)
          fprintf(outfile, ": %.*s", use_offsets[1] - use_offsets[0],
            bptr + use_offsets[0]);
#endif
        fprintf(outfile, "\n");
        break;  /* Out of the /g loop */
        }

      /* Failed to match. If this is a /g or /G loop and we previously set
      g_notempty after a null match, this is not necessarily the end.
      We want to advance the start offset, and continue. In the case of UTF-8
      matching, the advance must be one character, not one byte. Fudge the
      offset values to achieve this. We won't be at the end of the string -
      that was checked before setting g_notempty. */

      else
        {
        if (g_notempty != 0)
          {
          int onechar = 1;
          use_offsets[0] = start_offset;
          if (use_utf8)
            {
            while (start_offset + onechar < len)
              {
              int tb = bptr[start_offset+onechar];
              if (tb <= 127) break;
              tb &= 0xc0;
              if (tb != 0 && tb != 0xc0) onechar++;
              }
            }
          use_offsets[1] = start_offset + onechar;
          }
        else
          {
          if (count == PCRE_ERROR_NOMATCH)
            {
            if (gmatched == 0) fprintf(outfile, "No match\n");
            }
          else fprintf(outfile, "Error %d\n", count);
          break;  /* Out of the /g loop */
          }
        }

      /* If not /g or /G we are done */

      if (!do_g && !do_G) break;

      /* If we have matched an empty string, first check to see if we are at
      the end of the subject. If so, the /g loop is over. Otherwise, mimic
      what Perl's /g options does. This turns out to be rather cunning. First
      we set PCRE_NOTEMPTY and PCRE_ANCHORED and try the match again at the
      same point. If this fails (picked up above) we advance to the next
      character. */

      g_notempty = 0;
      if (use_offsets[0] == use_offsets[1])
        {
        if (use_offsets[0] == len) break;
        g_notempty = PCRE_NOTEMPTY | PCRE_ANCHORED;
        }

      /* For /g, update the start offset, leaving the rest alone */

      if (do_g) start_offset = use_offsets[1];

      /* For /G, update the pointer and length */

      else
        {
        bptr += use_offsets[1];
        len -= use_offsets[1];
        }
      }  /* End of loop for /g and /G */
    }    /* End of loop for data lines */

  CONTINUE:

#if !defined NOPOSIX
  if (posix || do_posix) regfree(&preg);
#endif

  if (re != NULL) new_free(re);
  if (extra != NULL) new_free(extra);
  if (tables != NULL)
    {
    new_free((void *)tables);
    setlocale(LC_CTYPE, "C");
    }
  }

if (infile == stdin) fprintf(outfile, "\n");

EXIT:

if (infile != NULL && infile != stdin) fclose(infile);
if (outfile != NULL && outfile != stdout) fclose(outfile);

free(buffer);
free(dbuffer);
free(pbuffer);
free(offsets);

return yield;
}

/* End of pcretest.c */