pcretest.c

php-4.4.7学习linux时下载的源代码

      new_info(re, NULL, PCRE_INFO_NAMECOUNT, &namecount);
      new_info(re, NULL, PCRE_INFO_NAMETABLE, (void *)&nametable);

#if !defined NOINFOCHECK
      old_count = pcre_info(re, &old_options, &old_first_char);
      if (count < 0) fprintf(outfile,
        "Error %d from pcre_info()\n", count);
      else
        {
        if (old_count != count) fprintf(outfile,
          "Count disagreement: pcre_fullinfo=%d pcre_info=%d\n", count,
            old_count);

        if (old_first_char != first_char) fprintf(outfile,
          "First char disagreement: pcre_fullinfo=%d pcre_info=%d\n",
            first_char, old_first_char);

        if (old_options != (int)get_options) fprintf(outfile,
          "Options disagreement: pcre_fullinfo=%ld pcre_info=%d\n",
            get_options, old_options);
        }
#endif

      if (size != regex_gotten_store) fprintf(outfile,
        "Size disagreement: pcre_fullinfo=%d call to malloc for %d\n",
        (int)size, (int)regex_gotten_store);

      fprintf(outfile, "Capturing subpattern count = %d\n", count);
      if (backrefmax > 0)
        fprintf(outfile, "Max back reference = %d\n", backrefmax);

      if (namecount > 0)
        {
        fprintf(outfile, "Named capturing subpatterns:\n");
        while (namecount-- > 0)
          {
          fprintf(outfile, "  %s %*s%3d\n", nametable + 2,
            nameentrysize - 3 - (int)strlen((char *)nametable + 2), "",
            GET2(nametable, 0));
          nametable += nameentrysize;
          }
        }

      /* The NOPARTIAL bit is a private bit in the options, so we have
      to fish it out via out back door */

      all_options = ((real_pcre *)re)->options;
      if (do_flip)
        {
        all_options = byteflip(all_options, sizeof(all_options));
         }

      if ((all_options & PCRE_NOPARTIAL) != 0)
        fprintf(outfile, "Partial matching not supported\n");

      if (get_options == 0) fprintf(outfile, "No options\n");
        else fprintf(outfile, "Options:%s%s%s%s%s%s%s%s%s%s%s%s%s\n",
          ((get_options & PCRE_ANCHORED) != 0)? " anchored" : "",
          ((get_options & PCRE_CASELESS) != 0)? " caseless" : "",
          ((get_options & PCRE_EXTENDED) != 0)? " extended" : "",
          ((get_options & PCRE_MULTILINE) != 0)? " multiline" : "",
          ((get_options & PCRE_FIRSTLINE) != 0)? " firstline" : "",
          ((get_options & PCRE_DOTALL) != 0)? " dotall" : "",
          ((get_options & PCRE_DOLLAR_ENDONLY) != 0)? " dollar_endonly" : "",
          ((get_options & PCRE_EXTRA) != 0)? " extra" : "",
          ((get_options & PCRE_UNGREEDY) != 0)? " ungreedy" : "",
          ((get_options & PCRE_NO_AUTO_CAPTURE) != 0)? " no_auto_capture" : "",
          ((get_options & PCRE_UTF8) != 0)? " utf8" : "",
          ((get_options & PCRE_NO_UTF8_CHECK) != 0)? " no_utf8_check" : "",
          ((get_options & PCRE_DUPNAMES) != 0)? " dupnames" : "");

      switch (get_options & PCRE_NEWLINE_BITS)
        {
        case PCRE_NEWLINE_CR:
        fprintf(outfile, "Forced newline sequence: CR\n");
        break;

        case PCRE_NEWLINE_LF:
        fprintf(outfile, "Forced newline sequence: LF\n");
        break;

        case PCRE_NEWLINE_CRLF:
        fprintf(outfile, "Forced newline sequence: CRLF\n");
        break;

        case PCRE_NEWLINE_ANY:
        fprintf(outfile, "Forced newline sequence: ANY\n");
        break;

        default:
        break;
        }

      if (first_char == -1)
        {
        fprintf(outfile, "First char at start or follows newline\n");
        }
      else if (first_char < 0)
        {
        fprintf(outfile, "No first char\n");
        }
      else
        {
        int ch = first_char & 255;
        const char *caseless = ((first_char & REQ_CASELESS) == 0)?
          "" : " (caseless)";
        if (PRINTHEX(ch))
          fprintf(outfile, "First char = \'%c\'%s\n", ch, caseless);
        else
          fprintf(outfile, "First char = %d%s\n", ch, caseless);
        }

      if (need_char < 0)
        {
        fprintf(outfile, "No need char\n");
        }
      else
        {
        int ch = need_char & 255;
        const char *caseless = ((need_char & REQ_CASELESS) == 0)?
          "" : " (caseless)";
        if (PRINTHEX(ch))
          fprintf(outfile, "Need char = \'%c\'%s\n", ch, caseless);
        else
          fprintf(outfile, "Need char = %d%s\n", ch, caseless);
        }

      /* Don't output study size; at present it is in any case a fixed
      value, but it varies, depending on the computer architecture, and
      so messes up the test suite. (And with the /F option, it might be
      flipped.) */

      if (do_study)
        {
        if (extra == NULL)
          fprintf(outfile, "Study returned NULL\n");
        else
          {
          uschar *start_bits = NULL;
          new_info(re, extra, PCRE_INFO_FIRSTTABLE, &start_bits);

          if (start_bits == NULL)
            fprintf(outfile, "No starting byte set\n");
          else
            {
            int i;
            int c = 24;
            fprintf(outfile, "Starting byte set: ");
            for (i = 0; i < 256; i++)
              {
              if ((start_bits[i/8] & (1<<(i&7))) != 0)
                {
                if (c > 75)
                  {
                  fprintf(outfile, "\n  ");
                  c = 2;
                  }
                if (PRINTHEX(i) && i != ' ')
                  {
                  fprintf(outfile, "%c ", i);
                  c += 2;
                  }
                else
                  {
                  fprintf(outfile, "\\x%02x ", i);
                  c += 5;
                  }
                }
              }
            fprintf(outfile, "\n");
            }
          }
        }
      }

    /* If the '>' option was present, we write out the regex to a file, and
    that is all. The first 8 bytes of the file are the regex length and then
    the study length, in big-endian order. */

    if (to_file != NULL)
      {
      FILE *f = fopen((char *)to_file, "wb");
      if (f == NULL)
        {
        fprintf(outfile, "Unable to open %s: %s\n", to_file, strerror(errno));
        }
      else
        {
        uschar sbuf[8];
        sbuf[0] = (true_size >> 24)  & 255;
        sbuf[1] = (true_size >> 16)  & 255;
        sbuf[2] = (true_size >>  8)  & 255;
        sbuf[3] = (true_size)  & 255;

        sbuf[4] = (true_study_size >> 24)  & 255;
        sbuf[5] = (true_study_size >> 16)  & 255;
        sbuf[6] = (true_study_size >>  8)  & 255;
        sbuf[7] = (true_study_size)  & 255;

        if (fwrite(sbuf, 1, 8, f) < 8 ||
            fwrite(re, 1, true_size, f) < true_size)
          {
          fprintf(outfile, "Write error on %s: %s\n", to_file, strerror(errno));
          }
        else
          {
          fprintf(outfile, "Compiled regex written to %s\n", to_file);
          if (extra != NULL)
            {
            if (fwrite(extra->study_data, 1, true_study_size, f) <
                true_study_size)
              {
              fprintf(outfile, "Write error on %s: %s\n", to_file,
                strerror(errno));
              }
            else fprintf(outfile, "Study data written to %s\n", to_file);

            }
          }
        fclose(f);
        }

      new_free(re);
      if (extra != NULL) new_free(extra);
      if (tables != NULL) new_free((void *)tables);
      continue;  /* With next regex */
      }
    }        /* End of non-POSIX compile */

  /* Read data lines and test them */

  for (;;)
    {
    uschar *q;
    uschar *bptr = dbuffer;
    int *use_offsets = offsets;
    int use_size_offsets = size_offsets;
    int callout_data = 0;
    int callout_data_set = 0;
    int count, c;
    int copystrings = 0;
    int find_match_limit = 0;
    int getstrings = 0;
    int getlist = 0;
    int gmatched = 0;
    int start_offset = 0;
    int g_notempty = 0;
    int use_dfa = 0;

    options = 0;

    *copynames = 0;
    *getnames = 0;

    copynamesptr = copynames;
    getnamesptr = getnames;

    pcre_callout = callout;
    first_callout = 1;
    callout_extra = 0;
    callout_count = 0;
    callout_fail_count = 999999;
    callout_fail_id = -1;
    show_malloc = 0;

    if (extra != NULL) extra->flags &=
      ~(PCRE_EXTRA_MATCH_LIMIT|PCRE_EXTRA_MATCH_LIMIT_RECURSION);

    len = 0;
    for (;;)
      {
      if (infile == stdin) printf("data> ");
      if (extend_inputline(infile, buffer + len) == NULL)
        {
        if (len > 0) break;
        done = 1;
        goto CONTINUE;
        }
      if (infile != stdin) fprintf(outfile, "%s", (char *)buffer);
      len = (int)strlen((char *)buffer);
      if (buffer[len-1] == '\n') break;
      }

    while (len > 0 && isspace(buffer[len-1])) len--;
    buffer[len] = 0;
    if (len == 0) break;

    p = buffer;
    while (isspace(*p)) p++;

    q = dbuffer;
    while ((c = *p++) != 0)
      {
      int i = 0;
      int n = 0;

      if (c == '\\') switch ((c = *p++))
        {
        case 'a': c =    7; break;
        case 'b': c = '\b'; break;
        case 'e': c =   27; break;
        case 'f': c = '\f'; break;
        case 'n': c = '\n'; break;
        case 'r': c = '\r'; break;
        case 't': c = '\t'; break;
        case 'v': c = '\v'; break;

        case '0': case '1': case '2': case '3':
        case '4': case '5': case '6': case '7':
        c -= '0';
        while (i++ < 2 && isdigit(*p) && *p != '8' && *p != '9')
          c = c * 8 + *p++ - '0';

#if !defined NOUTF8
        if (use_utf8 && c > 255)
          {
          unsigned char buff8[8];
          int ii, utn;
          utn = ord2utf8(c, buff8);
          for (ii = 0; ii < utn - 1; ii++) *q++ = buff8[ii];
          c = buff8[ii];   /* Last byte */
          }
#endif
        break;

        case 'x':

        /* Handle \x{..} specially - new Perl thing for utf8 */

#if !defined NOUTF8
        if (*p == '{')
          {
          unsigned char *pt = p;
          c = 0;
          while (isxdigit(*(++pt)))
            c = c * 16 + tolower(*pt) - ((isdigit(*pt))? '0' : 'W');
          if (*pt == '}')
            {
            unsigned char buff8[8];
            int ii, utn;
            utn = ord2utf8(c, buff8);
            for (ii = 0; ii < utn - 1; ii++) *q++ = buff8[ii];
            c = buff8[ii];   /* Last byte */
            p = pt + 1;
            break;
            }
          /* Not correct form; fall through */
          }
#endif

        /* Ordinary \x */

        c = 0;
        while (i++ < 2 && isxdigit(*p))
          {
          c = c * 16 + tolower(*p) - ((isdigit(*p))? '0' : 'W');
          p++;
          }
        break;

        case 0:   /* \ followed by EOF allows for an empty line */
        p--;
        continue;

        case '>':
        while(isdigit(*p)) start_offset = start_offset * 10 + *p++ - '0';
        continue;

        case 'A':  /* Option setting */
        options |= PCRE_ANCHORED;
        continue;

        case 'B':
        options |= PCRE_NOTBOL;
        continue;

        case 'C':
        if (isdigit(*p))    /* Set copy string */
          {
          while(isdigit(*p)) n = n * 10 + *p++ - '0';
          copystrings |= 1 << n;
          }
        else if (isalnum(*p))
          {
          uschar *npp = copynamesptr;
          while (isalnum(*p)) *npp++ = *p++;
          *npp++ = 0;
          *npp = 0;
          n = pcre_get_stringnumber(re, (char *)copynamesptr);
          if (n < 0)
            fprintf(outfile, "no parentheses with name \"%s\"\n", copynamesptr);
          copynamesptr = npp;
          }
        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;
          }