mg_string.c

#define REPLACE_SKIP
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <stdarg.h>
#include <ctype.h>
#include <time.h>
#include "mg_string.h"
#undef REPLACE_SKIP
// Skip the characters until one of the delimiters characters found.
// 0-terminate resulting word. Skip the delimiter and following whitespaces.
// Advance pointer to buffer to the next word. Return found 0-terminated word.
// Delimiters can be quoted with quotechar.
REPLACE_STATIC char *skip_quoted(char **buf, const char *delimiters,
                         const char *whitespace, char quotechar) {
  char *p, *begin_word, *end_word, *end_whitespace;

  begin_word = *buf;
  end_word = begin_word + strcspn(begin_word, delimiters);

  // Check for quotechar
  if (end_word > begin_word) {
    p = end_word - 1;
    while (*p == quotechar) {
      // If there is anything beyond end_word, copy it
      if (*end_word == '\0') {
        *p = '\0';
        break;
      } else {
        size_t end_off = strcspn(end_word + 1, delimiters);
        memmove (p, end_word, end_off + 1);
        p += end_off; // p must correspond to end_word - 1
        end_word += end_off + 1;
      }
    }
    for (p++; p < end_word; p++) {
      *p = '\0';
    }
  }

  if (*end_word == '\0') {
    *buf = end_word;
  } else {
    end_whitespace = end_word + 1 + strspn(end_word + 1, whitespace);

    for (p = end_word; p < end_whitespace; p++) {
      *p = '\0';
    }

    *buf = end_whitespace;
  }

  return begin_word;
}

// Simplified version of skip_quoted without quote char
// and whitespace == delimiters
REPLACE_STATIC char *skip(char **buf, const char *delimiters) {
  return skip_quoted(buf, delimiters, delimiters, 0);
}

// A helper function for traversing a comma separated list of values.
// It returns a list pointer shifted to the next value, or NULL if the end
// of the list found.
// Value is stored in val vector. If value has form "x=y", then eq_val
// vector is initialized to point to the "y" part, and val vector length
// is adjusted to point only to "x".
REPLACE_STATIC const char *next_option(const char *list, struct vec *val,
                               struct vec *eq_val) {
  if (list == NULL || *list == '\0') {
    // End of the list
    list = NULL;
  } else {
    val->ptr = list;
    if ((list = strchr(val->ptr, ',')) != NULL) {
      // Comma found. Store length and shift the list ptr
      val->len = list - val->ptr;
      list++;
    } else {
      // This value is the last one
      list = val->ptr + strlen(val->ptr);
      val->len = list - val->ptr;
    }

    if (eq_val != NULL) {
      // Value has form "x=y", adjust pointers and lengths
      // so that val points to "x", and eq_val points to "y".
      eq_val->len = 0;
      eq_val->ptr = (const char *) memchr(val->ptr, '=', val->len);
      if (eq_val->ptr != NULL) {
        eq_val->ptr++;  // Skip over '=' character
        eq_val->len = val->ptr + val->len - eq_val->ptr;
        val->len = (eq_val->ptr - val->ptr) - 1;
      }
    }
  }

  return list;
}

REPLACE_STATIC int match_prefix(const char *pattern, int pattern_len, const char *str) {
  const char *or_str;
  int i, j, len, res;

  if ((or_str = (const char *) memchr(pattern, '|', pattern_len)) != NULL) {
    res = match_prefix(pattern, or_str - pattern, str);
    return res > 0 ? res :
        match_prefix(or_str + 1, (pattern + pattern_len) - (or_str + 1), str);
  }

  i = j = 0;
  res = -1;
  for (; i < pattern_len; i++, j++) {
    if (pattern[i] == '?' && str[j] != '\0') {
      continue;
    } else if (pattern[i] == '$') {
      return str[j] == '\0' ? j : -1;
    } else if (pattern[i] == '*') {
      i++;
      if (pattern[i] == '*') {
        i++;
        len = (int) strlen(str + j);
      } else {
        len = (int) strcspn(str + j, "/");
      }
      if (i == pattern_len) {
        return j + len;
      }
      do {
        res = match_prefix(pattern + i, pattern_len - i, str + j + len);
      } while (res == -1 && len-- > 0);
      return res == -1 ? -1 : j + res + len;
    } else if (pattern[i] != str[j]) {
      return -1;
    }
  }
  return j;
}

// URL-decode input buffer into destination buffer.
// 0-terminate the destination buffer. Return the length of decoded data.
// form-url-encoded data differs from URI encoding in a way that it
// uses '+' as character for space, see RFC 1866 section 8.2.1
// http://ftp.ics.uci.edu/pub/ietf/html/rfc1866.txt
REPLACE_STATIC int url_decode(const char *src, int src_len, char *dst,
                      int dst_len, int is_form_url_encoded) {
  int i, j, a, b;
#define HEXTOI(x) (isdigit(x) ? x - '0' : x - 'W')

  for (i = j = 0; i < src_len && j < dst_len - 1; i++, j++) {
    if (src[i] == '%' &&
        isxdigit(* (const unsigned char *) (src + i + 1)) &&
        isxdigit(* (const unsigned char *) (src + i + 2))) {
      a = tolower(* (const unsigned char *) (src + i + 1));
      b = tolower(* (const unsigned char *) (src + i + 2));
      dst[j] = (char) ((HEXTOI(a) << 4) | HEXTOI(b));
      i += 2;
    } else if (is_form_url_encoded && src[i] == '+') {
      dst[j] = ' ';
    } else {
      dst[j] = src[i];
    }
  }

  dst[j] = '\0'; // Null-terminate the destination

  return i >= src_len ? j : -1;
}

// Check whether full request is buffered. Return:
//   -1  if request is malformed
//    0  if request is not yet fully buffered
//   >0  actual request length, including last \r\n\r\n
REPLACE_STATIC int get_request_len(const char *buf, int buflen) {
  const char *s, *e;
  int len = 0;

  for (s = buf, e = s + buflen - 1; len <= 0 && s < e; s++)
    // Control characters are not allowed but >=128 is.
    if (!isprint(* (const unsigned char *) s) && *s != '\r' &&
        *s != '\n' && * (const unsigned char *) s < 128) {
      len = -1;
      break; // [i_a] abort scan as soon as one malformed character is found; don't let subsequent \r\n\r\n win us over anyhow
    } else if (s[0] == '\n' && s[1] == '\n') {
      len = (int) (s - buf) + 2;
    } else if (s[0] == '\n' && &s[1] < e &&
        s[1] == '\r' && s[2] == '\n') {
      len = (int) (s - buf) + 3;
    }

  return len;
}

// Protect against directory disclosure attack by removing '..',
// excessive '/' and '\' characters
REPLACE_STATIC void remove_double_dots_and_double_slashes(char *s) {
  char *p = s;

  while (*s != '\0') {
    *p++ = *s++;
    if (s[-1] == '/' || s[-1] == '\\') {
      // Skip all following slashes, backslashes and double-dots
      while (s[0] != '\0') {
        if (s[0] == '/' || s[0] == '\\') {
          s++;
        } else if (s[0] == '.' && s[1] == '.') {
          s += 2;
        } else {
          break;
        }
      }
    }
  }
  *p = '\0';
}

// Stringify binary data. Output buffer must be twice as big as input,
// because each byte takes 2 bytes in string representation
REPLACE_STATIC void bin2str(char *to, const unsigned char *p, size_t len) {
  static const char *hex = "0123456789abcdef";

  for (; len--; p++) {
    *to++ = hex[p[0] >> 4];
    *to++ = hex[p[0] & 0x0f];
  }
  *to = '\0';
}

REPLACE_STATIC void url_encode(const char *src, char *dst, size_t dst_len) {
  static const char *dont_escape = "._-$,;~()";
  static const char *hex = "0123456789abcdef";
  const char *end = dst + dst_len - 1;

  for (; *src != '\0' && dst < end; src++, dst++) {
    if (isalnum(*(const unsigned char *) src) ||
        strchr(dont_escape, * (const unsigned char *) src) != NULL) {
      *dst = *src;
    } else if (dst + 2 < end) {
      dst[0] = '%';
      dst[1] = hex[(* (const unsigned char *) src) >> 4];
      dst[2] = hex[(* (const unsigned char *) src) & 0xf];
      dst += 2;
    }
  }

  *dst = '\0';
}


REPLACE_STATIC void base64_encode(const unsigned char *src, int src_len, char *dst) {
  static const char *b64 =
    "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
  int i, j, a, b, c;

  for (i = j = 0; i < src_len; i += 3) {
    a = src[i];
    b = i + 1 >= src_len ? 0 : src[i + 1];
    c = i + 2 >= src_len ? 0 : src[i + 2];

    dst[j++] = b64[a >> 2];
    dst[j++] = b64[((a & 3) << 4) | (b >> 4)];
    if (i + 1 < src_len) {
      dst[j++] = b64[(b & 15) << 2 | (c >> 6)];
    }
    if (i + 2 < src_len) {
      dst[j++] = b64[c & 63];
    }
  }
  while (j % 4 != 0) {
    dst[j++] = '=';
  }
  dst[j++] = '\0';
}

REPLACE_STATIC void mg_strlcpy(register char *dst, register const char *src, size_t n) {
  for (; *src != '\0' && n > 1; n--) {
    *dst++ = *src++;
  }
  *dst = '\0';
}

REPLACE_STATIC int lowercase(const char *s) {
  return tolower(* (const unsigned char *) s);
}

REPLACE_STATIC int mg_strncasecmp(const char *s1, const char *s2, size_t len) {
  int diff = 0;

  if (len > 0)
    do {
      diff = lowercase(s1++) - lowercase(s2++);
    } while (diff == 0 && s1[-1] != '\0' && --len > 0);

  return diff;
}

REPLACE_STATIC int mg_strcasecmp(const char *s1, const char *s2) {
  int diff;

  do {
    diff = lowercase(s1++) - lowercase(s2++);
  } while (diff == 0 && s1[-1] != '\0');

  return diff;
}

REPLACE_STATIC char * mg_strndup(const char *ptr, size_t len) {
  char *p;

  if ((p = (char *) malloc(len + 1)) != NULL) {
    mg_strlcpy(p, ptr, len + 1);
  }

  return p;
}

REPLACE_STATIC char * mg_strdup(const char *str) {
  return mg_strndup(str, strlen(str));
}

static const char *month_names[] = {
  "Jan", "Feb", "Mar", "Apr", "May", "Jun",
  "Jul", "Aug", "Sep", "Oct", "Nov", "Dec"
};

// Convert month to the month number. Return -1 on error, or month number
static int get_month_index(const char *s) {
  size_t i;

  for (i = 0; i < ARRAY_SIZE(month_names); i++)
    if (!strcmp(s, month_names[i]))
      return (int) i;

  return -1;
}

static int num_leap_years(int year) {
  return year / 4 - year / 100 + year / 400;
}

// Parse UTC date-time string, and return the corresponding time_t value.
REPLACE_STATIC time_t parse_date_string(const char *datetime) {
  static const unsigned short days_before_month[] = {
    0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334
  };
  char month_str[32];
  int second, minute, hour, day, month, year, leap_days, days;
  time_t result = (time_t) 0;

  if (((sscanf(datetime, "%d/%3s/%d %d:%d:%d",
               &day, month_str, &year, &hour, &minute, &second) == 6) ||
       (sscanf(datetime, "%d %3s %d %d:%d:%d",
               &day, month_str, &year, &hour, &minute, &second) == 6) ||
       (sscanf(datetime, "%*3s, %d %3s %d %d:%d:%d",
               &day, month_str, &year, &hour, &minute, &second) == 6) ||
       (sscanf(datetime, "%d-%3s-%d %d:%d:%d",
               &day, month_str, &year, &hour, &minute, &second) == 6)) &&
      year > 1970 &&
      (month = get_month_index(month_str)) != -1) {
    leap_days = num_leap_years(year) - num_leap_years(1970);
    year -= 1970;
    days = year * 365 + days_before_month[month] + (day - 1) + leap_days;
    result = days * 24 * 3600 + hour * 3600 + minute * 60 + second;
  }

  return result;
}