util.c

关于tor匿名通信的源代码

    return strncasecmp(s1+(n1-n2), s2, n2);
}

/** Return a pointer to the first char of s that is not whitespace and
 * not a comment, or to the terminating NUL if no such character exists.
 */
const char *
eat_whitespace(const char *s)
{
  tor_assert(s);

  while (1) {
    switch (*s) {
    case '\0':
    default:
      return s;
    case ' ':
    case '\t':
    case '\n':
    case '\r':
      ++s;
      break;
    case '#':
      ++s;
      while (*s && *s != '\n')
        ++s;
    }
  }
}

/** Return a pointer to the first char of s that is not whitespace and
 * not a comment, or to the terminating NUL if no such character exists.
 */
const char *
eat_whitespace_eos(const char *s, const char *eos)
{
  tor_assert(s);
  tor_assert(eos && s <= eos);

  while (s < eos) {
    switch (*s) {
    case '\0':
    default:
      return s;
    case ' ':
    case '\t':
    case '\n':
    case '\r':
      ++s;
      break;
    case '#':
      ++s;
      while (s < eos && *s && *s != '\n')
        ++s;
    }
  }
  return s;
}

/** Return a pointer to the first char of s that is not a space or a tab
 * or a \\r, or to the terminating NUL if no such character exists. */
const char *
eat_whitespace_no_nl(const char *s)
{
  while (*s == ' ' || *s == '\t' || *s == '\r')
    ++s;
  return s;
}

/** As eat_whitespace_no_nl, but stop at <b>eos</b> whether we have
 * found a non-whitespace character or not. */
const char *
eat_whitespace_eos_no_nl(const char *s, const char *eos)
{
  while (s < eos && (*s == ' ' || *s == '\t' || *s == '\r'))
    ++s;
  return s;
}

/** Return a pointer to the first char of s that is whitespace or <b>#</b>,
 * or to the terminating NUL if no such character exists.
 */
const char *
find_whitespace(const char *s)
{
  /* tor_assert(s); */
  while (1) {
    switch (*s)
    {
    case '\0':
    case '#':
    case ' ':
    case '\r':
    case '\n':
    case '\t':
      return s;
    default:
      ++s;
    }
  }
}

/** As find_whitespace, but stop at <b>eos</b> whether we have found a
 * whitespace or not. */
const char *
find_whitespace_eos(const char *s, const char *eos)
{
  /* tor_assert(s); */
  while (s < eos) {
    switch (*s)
    {
    case '\0':
    case '#':
    case ' ':
    case '\r':
    case '\n':
    case '\t':
      return s;
    default:
      ++s;
    }
  }
  return s;
}

/** Return true iff the 'len' bytes at 'mem' are all zero. */
int
tor_mem_is_zero(const char *mem, size_t len)
{
  static const char ZERO[] = {
    0,0,0,0, 0,0,0,0, 0,0,0,0, 0,0,0,0, 0,0,0,0, 0,0,0,0, 0,0,0,0, 0,0,0,0,
  };
  while (len >= sizeof(ZERO)) {
    if (memcmp(mem, ZERO, sizeof(ZERO)))
      return 0;
    len -= sizeof(ZERO);
    mem += sizeof(ZERO);
  }
  /* Deal with leftover bytes. */
  if (len)
    return ! memcmp(mem, ZERO, len);

  return 1;
}

/** Return true iff the DIGEST_LEN bytes in digest are all zero. */
int
tor_digest_is_zero(const char *digest)
{
  return tor_mem_is_zero(digest, DIGEST_LEN);
}

/* Helper: common code to check whether the result of a strtol or strtoul or
 * strtoll is correct. */
#define CHECK_STRTOX_RESULT()                           \
  /* Was at least one character converted? */           \
  if (endptr == s)                                      \
    goto err;                                           \
  /* Were there unexpected unconverted characters? */   \
  if (!next && *endptr)                                 \
    goto err;                                           \
  /* Is r within limits? */                             \
  if (r < min || r > max)                               \
    goto err;                                           \
  if (ok) *ok = 1;                                      \
  if (next) *next = endptr;                             \
  return r;                                             \
 err:                                                   \
  if (ok) *ok = 0;                                      \
  if (next) *next = endptr;                             \
  return 0

/** Extract a long from the start of s, in the given numeric base.  If
 * there is unconverted data and next is provided, set *next to the
 * first unconverted character.  An error has occurred if no characters
 * are converted; or if there are unconverted characters and next is NULL; or
 * if the parsed value is not between min and max.  When no error occurs,
 * return the parsed value and set *ok (if provided) to 1.  When an error
 * occurs, return 0 and set *ok (if provided) to 0.
 */
long
tor_parse_long(const char *s, int base, long min, long max,
               int *ok, char **next)
{
  char *endptr;
  long r;

  r = strtol(s, &endptr, base);
  CHECK_STRTOX_RESULT();
}

/** As tor_parse_long(), but return an unsigned long. */
unsigned long
tor_parse_ulong(const char *s, int base, unsigned long min,
                unsigned long max, int *ok, char **next)
{
  char *endptr;
  unsigned long r;

  r = strtoul(s, &endptr, base);
  CHECK_STRTOX_RESULT();
}

/** As tor_parse_log, but return a unit64_t.  Only base 10 is guaranteed to
 * work for now. */
uint64_t
tor_parse_uint64(const char *s, int base, uint64_t min,
                 uint64_t max, int *ok, char **next)
{
  char *endptr;
  uint64_t r;

#ifdef HAVE_STRTOULL
  r = (uint64_t)strtoull(s, &endptr, base);
#elif defined(MS_WINDOWS)
#if defined(_MSC_VER) && _MSC_VER < 1300
  tor_assert(base <= 10);
  r = (uint64_t)_atoi64(s);
  endptr = (char*)s;
  while (TOR_ISSPACE(*endptr)) endptr++;
  while (TOR_ISDIGIT(*endptr)) endptr++;
#else
  r = (uint64_t)_strtoui64(s, &endptr, base);
#endif
#elif SIZEOF_LONG == 8
  r = (uint64_t)strtoul(s, &endptr, base);
#else
#error "I don't know how to parse 64-bit numbers."
#endif

  CHECK_STRTOX_RESULT();
}

/** Encode the <b>srclen</b> bytes at <b>src</b> in a NUL-terminated,
 * uppercase hexadecimal string; store it in the <b>destlen</b>-byte buffer
 * <b>dest</b>.
 */
void
base16_encode(char *dest, size_t destlen, const char *src, size_t srclen)
{
  const char *end;
  char *cp;

  tor_assert(destlen >= srclen*2+1);
  tor_assert(destlen < SIZE_T_CEILING);

  cp = dest;
  end = src+srclen;
  while (src<end) {
    *cp++ = "0123456789ABCDEF"[ (*(const uint8_t*)src) >> 4 ];
    *cp++ = "0123456789ABCDEF"[ (*(const uint8_t*)src) & 0xf ];
    ++src;
  }
  *cp = '\0';
}

/** Helper: given a hex digit, return its value, or -1 if it isn't hex. */
static INLINE int
hex_decode_digit(char c)
{
  switch (c) {
    case '0': return 0;
    case '1': return 1;
    case '2': return 2;
    case '3': return 3;
    case '4': return 4;
    case '5': return 5;
    case '6': return 6;
    case '7': return 7;
    case '8': return 8;
    case '9': return 9;
    case 'A': case 'a': return 10;
    case 'B': case 'b': return 11;
    case 'C': case 'c': return 12;
    case 'D': case 'd': return 13;
    case 'E': case 'e': return 14;
    case 'F': case 'f': return 15;
    default:
      return -1;
  }
}

/** Given a hexadecimal string of <b>srclen</b> bytes in <b>src</b>, decode it
 * and store the result in the <b>destlen</b>-byte buffer at <b>dest</b>.
 * Return 0 on success, -1 on failure. */
int
base16_decode(char *dest, size_t destlen, const char *src, size_t srclen)
{
  const char *end;

  int v1,v2;
  if ((srclen % 2) != 0)
    return -1;
  if (destlen < srclen/2 || destlen > SIZE_T_CEILING)
    return -1;
  end = src+srclen;
  while (src<end) {
    v1 = hex_decode_digit(*src);
    v2 = hex_decode_digit(*(src+1));
    if (v1<0||v2<0)
      return -1;
    *(uint8_t*)dest = (v1<<4)|v2;
    ++dest;
    src+=2;
  }
  return 0;
}

/** Allocate and return a new string representing the contents of <b>s</b>,
 * surrounded by quotes and using standard C escapes.
 *
 * Generally, we use this for logging values that come in over the network to
 * keep them from tricking users, and for sending certain values to the
 * controller.
 *
 * We trust values from the resolver, OS, configuration file, and command line
 * to not be maliciously ill-formed.  We validate incoming routerdescs and
 * SOCKS requests and addresses from BEGIN cells as they're parsed;
 * afterwards, we trust them as non-malicious.
 */
char *
esc_for_log(const char *s)
{
  const char *cp;
  char *result, *outp;
  size_t len = 3;
  if (!s) {
    return tor_strdup("");
  }

  for (cp = s; *cp; ++cp) {
    switch (*cp) {
      case '\\':
      case '\"':
      case '\'':
        len += 2;
        break;
      default:
        if (TOR_ISPRINT(*cp) && ((uint8_t)*cp)<127)
          ++len;
        else
          len += 4;
        break;
    }
  }

  result = outp = tor_malloc(len);
  *outp++ = '\"';
  for (cp = s; *cp; ++cp) {
    switch (*cp) {
      case '\\':
      case '\"':
      case '\'':
        *outp++ = '\\';
        *outp++ = *cp;
        break;
      case '\n':
        *outp++ = '\\';
        *outp++ = 'n';
        break;
      case '\t':
        *outp++ = '\\';
        *outp++ = 't';
        break;
      case '\r':
        *outp++ = '\\';
        *outp++ = 'r';
        break;
      default:
        if (TOR_ISPRINT(*cp) && ((uint8_t)*cp)<127) {
          *outp++ = *cp;
        } else {
          tor_snprintf(outp, 5, "\\%03o", (int)(uint8_t) *cp);
          outp += 4;
        }
        break;
    }
  }

  *outp++ = '\"';
  *outp++ = 0;

  return result;
}

/** Allocate and return a new string representing the contents of <b>s</b>,
 * surrounded by quotes and using standard C escapes.
 *
 * THIS FUNCTION IS NOT REENTRANT.  Don't call it from outside the main
 * thread.  Also, each call invalidates the last-returned value, so don't
 * try log_warn(LD_GENERAL, "%s %s", escaped(a), escaped(b));
 */
const char *
escaped(const char *s)
{
  static char *_escaped_val = NULL;
  if (_escaped_val)
    tor_free(_escaped_val);

  if (s)
    _escaped_val = esc_for_log(s);
  else
    _escaped_val = NULL;

  return _escaped_val;
}

/** Rudimentary string wrapping code: given a un-wrapped <b>string</b> (no
 * newlines!), break the string into newline-terminated lines of no more than
 * <b>width</b> characters long (not counting newline) and insert them into
 * <b>out</b> in order.  Precede the first line with prefix0, and subsequent
 * lines with prefixRest.
 */
/* This uses a stupid greedy wrapping algorithm right now:
 *  - For each line:
 *    - Try to fit as much stuff as possible, but break on a space.
 *    - If the first "word" of the line will extend beyond the allowable
 *      width, break the word at the end of the width.
 */
void
wrap_string(smartlist_t *out, const char *string, size_t width,
            const char *prefix0, const char *prefixRest)
{
  size_t p0Len, pRestLen, pCurLen;
  const char *eos, *prefixCur;
  tor_assert(out);
  tor_assert(string);
  tor_assert(width);
  if (!prefix0)
    prefix0 = "";
  if (!prefixRest)
    prefixRest = "";

  p0Len = strlen(prefix0);
  pRestLen = strlen(prefixRest);
  tor_assert(width > p0Len && width > pRestLen);
  eos = strchr(string, '\0');
  tor_assert(eos);
  pCurLen = p0Len;
  prefixCur = prefix0;

  while ((eos-string)+pCurLen > width) {
    const char *eol = string + width - pCurLen;
    while (eol > string && *eol != ' ')
      --eol;
    /* eol is now the last space that can fit, or the start of the string. */
    if (eol > string) {
      size_t line_len = (eol-string) + pCurLen + 2;
      char *line = tor_malloc(line_len);
      memcpy(line, prefixCur, pCurLen);
      memcpy(line+pCurLen, string, eol-string);
      line[line_len-2] = '\n';
      line[line_len-1] = '\0';
      smartlist_add(out, line);
      string = eol + 1;
    } else {
      size_t line_len = width + 2;
      char *line = tor_malloc(line_len);
      memcpy(line, prefixCur, pCurLen);
      memcpy(line+pCurLen, string, width - pCurLen);
      line[line_len-2] = '\n';
      line[line_len-1] = '\0';
      smartlist_add(out, line);
      string += width-pCurLen;
    }
    prefixCur = prefixRest;
    pCurLen = pRestLen;
  }