str.c

Go to the documentation of this file.

/*
 * Copyright (c) 2005 Martin Decky
 * Copyright (c) 2008 Jiri Svoboda
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 *
 * - Redistributions of source code must retain the above copyright
 *   notice, this list of conditions and the following disclaimer.
 * - Redistributions in binary form must reproduce the above copyright
 *   notice, this list of conditions and the following disclaimer in the
 *   documentation and/or other materials provided with the distribution.
 * - The name of the author may not be used to endorse or promote products
 *   derived from this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 */

#include <str.h>
#include <stdlib.h>
#include <assert.h>
#include <stdint.h>
#include <ctype.h>
#include <malloc.h>
#include <errno.h>
#include <align.h>
#include <mem.h>
#include <str.h>

#define LO_MASK_8(n)  ((uint8_t) ((1 << (n)) - 1))

#define LO_MASK_32(n)  ((uint32_t) ((1 << (n)) - 1))

#define HI_MASK_8(n)  (~LO_MASK_8(8 - (n)))

#define CONT_BITS  6

wchar_t str_decode(const char *str, size_t *offset, size_t size)
{
        if (*offset + 1 > size)
                return 0;
        
        /* First byte read from string */
        uint8_t b0 = (uint8_t) str[(*offset)++];
        
        /* Determine code length */
        
        unsigned int b0_bits;  /* Data bits in first byte */
        unsigned int cbytes;   /* Number of continuation bytes */
        
        if ((b0 & 0x80) == 0) {
                /* 0xxxxxxx (Plain ASCII) */
                b0_bits = 7;
                cbytes = 0;
        } else if ((b0 & 0xe0) == 0xc0) {
                /* 110xxxxx 10xxxxxx */
                b0_bits = 5;
                cbytes = 1;
        } else if ((b0 & 0xf0) == 0xe0) {
                /* 1110xxxx 10xxxxxx 10xxxxxx */
                b0_bits = 4;
                cbytes = 2;
        } else if ((b0 & 0xf8) == 0xf0) {
                /* 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */
                b0_bits = 3;
                cbytes = 3;
        } else {
                /* 10xxxxxx -- unexpected continuation byte */
                return U_SPECIAL;
        }
        
        if (*offset + cbytes > size)
                return U_SPECIAL;
        
        wchar_t ch = b0 & LO_MASK_8(b0_bits);
        
        /* Decode continuation bytes */
        while (cbytes > 0) {
                uint8_t b = (uint8_t) str[(*offset)++];
                
                /* Must be 10xxxxxx */
                if ((b & 0xc0) != 0x80)
                        return U_SPECIAL;
                
                /* Shift data bits to ch */
                ch = (ch << CONT_BITS) | (wchar_t) (b & LO_MASK_8(CONT_BITS));
                cbytes--;
        }
        
        return ch;
}

int chr_encode(const wchar_t ch, char *str, size_t *offset, size_t size)
{
        if (*offset >= size)
                return EOVERFLOW;
        
        if (!chr_check(ch))
                return EINVAL;
        
        /* Unsigned version of ch (bit operations should only be done
           on unsigned types). */
        uint32_t cc = (uint32_t) ch;
        
        /* Determine how many continuation bytes are needed */
        
        unsigned int b0_bits;  /* Data bits in first byte */
        unsigned int cbytes;   /* Number of continuation bytes */
        
        if ((cc & ~LO_MASK_32(7)) == 0) {
                b0_bits = 7;
                cbytes = 0;
        } else if ((cc & ~LO_MASK_32(11)) == 0) {
                b0_bits = 5;
                cbytes = 1;
        } else if ((cc & ~LO_MASK_32(16)) == 0) {
                b0_bits = 4;
                cbytes = 2;
        } else if ((cc & ~LO_MASK_32(21)) == 0) {
                b0_bits = 3;
                cbytes = 3;
        } else {
                /* Codes longer than 21 bits are not supported */
                return EINVAL;
        }
        
        /* Check for available space in buffer */
        if (*offset + cbytes >= size)
                return EOVERFLOW;
        
        /* Encode continuation bytes */
        unsigned int i;
        for (i = cbytes; i > 0; i--) {
                str[*offset + i] = 0x80 | (cc & LO_MASK_32(CONT_BITS));
                cc = cc >> CONT_BITS;
        }
        
        /* Encode first byte */
        str[*offset] = (cc & LO_MASK_32(b0_bits)) | HI_MASK_8(8 - b0_bits - 1);
        
        /* Advance offset */
        *offset += cbytes + 1;
        
        return EOK;
}

size_t str_size(const char *str)
{
        size_t size = 0;
        
        while (*str++ != 0)
                size++;
        
        return size;
}

size_t wstr_size(const wchar_t *str)
{
        return (wstr_length(str) * sizeof(wchar_t));
}

size_t str_lsize(const char *str, size_t max_len)
{
        size_t len = 0;
        size_t offset = 0;
        
        while (len < max_len) {
                if (str_decode(str, &offset, STR_NO_LIMIT) == 0)
                        break;
                
                len++;
        }
        
        return offset;
}

size_t wstr_lsize(const wchar_t *str, size_t max_len)
{
        return (wstr_nlength(str, max_len * sizeof(wchar_t)) * sizeof(wchar_t));
}

size_t str_length(const char *str)
{
        size_t len = 0;
        size_t offset = 0;
        
        while (str_decode(str, &offset, STR_NO_LIMIT) != 0)
                len++;
        
        return len;
}

size_t wstr_length(const wchar_t *wstr)
{
        size_t len = 0;
        
        while (*wstr++ != 0)
                len++;
        
        return len;
}

size_t str_nlength(const char *str, size_t size)
{
        size_t len = 0;
        size_t offset = 0;
        
        while (str_decode(str, &offset, size) != 0)
                len++;
        
        return len;
}

size_t wstr_nlength(const wchar_t *str, size_t size)
{
        size_t len = 0;
        size_t limit = ALIGN_DOWN(size, sizeof(wchar_t));
        size_t offset = 0;
        
        while ((offset < limit) && (*str++ != 0)) {
                len++;
                offset += sizeof(wchar_t);
        }
        
        return len;
}

bool ascii_check(wchar_t ch)
{
        if ((ch >= 0) && (ch <= 127))
                return true;
        
        return false;
}

bool chr_check(wchar_t ch)
{
        if ((ch >= 0) && (ch <= 1114111))
                return true;
        
        return false;
}

int str_cmp(const char *s1, const char *s2)
{
        wchar_t c1 = 0;
        wchar_t c2 = 0;
        
        size_t off1 = 0;
        size_t off2 = 0;

        while (true) {
                c1 = str_decode(s1, &off1, STR_NO_LIMIT);
                c2 = str_decode(s2, &off2, STR_NO_LIMIT);

                if (c1 < c2)
                        return -1;
                
                if (c1 > c2)
                        return 1;

                if (c1 == 0 || c2 == 0)
                        break;          
        }

        return 0;
}

int str_lcmp(const char *s1, const char *s2, size_t max_len)
{
        wchar_t c1 = 0;
        wchar_t c2 = 0;
        
        size_t off1 = 0;
        size_t off2 = 0;
        
        size_t len = 0;

        while (true) {
                if (len >= max_len)
                        break;

                c1 = str_decode(s1, &off1, STR_NO_LIMIT);
                c2 = str_decode(s2, &off2, STR_NO_LIMIT);

                if (c1 < c2)
                        return -1;

                if (c1 > c2)
                        return 1;

                if (c1 == 0 || c2 == 0)
                        break;

                ++len;  
        }

        return 0;

}

void str_cpy(char *dest, size_t size, const char *src)
{
        /* There must be space for a null terminator in the buffer. */
        assert(size > 0);
        
        size_t src_off = 0;
        size_t dest_off = 0;
        
        wchar_t ch;
        while ((ch = str_decode(src, &src_off, STR_NO_LIMIT)) != 0) {
                if (chr_encode(ch, dest, &dest_off, size - 1) != EOK)
                        break;
        }
        
        dest[dest_off] = '\0';
}

void str_ncpy(char *dest, size_t size, const char *src, size_t n)
{
        /* There must be space for a null terminator in the buffer. */
        assert(size > 0);
        
        size_t src_off = 0;
        size_t dest_off = 0;
        
        wchar_t ch;
        while ((ch = str_decode(src, &src_off, n)) != 0) {
                if (chr_encode(ch, dest, &dest_off, size - 1) != EOK)
                        break;
        }
        
        dest[dest_off] = '\0';
}

void str_append(char *dest, size_t size, const char *src)
{
        size_t dstr_size;

        dstr_size = str_size(dest);
        str_cpy(dest + dstr_size, size - dstr_size, src);
}

void wstr_to_str(char *dest, size_t size, const wchar_t *src)
{
        wchar_t ch;
        size_t src_idx;
        size_t dest_off;

        /* There must be space for a null terminator in the buffer. */
        assert(size > 0);
        
        src_idx = 0;
        dest_off = 0;

        while ((ch = src[src_idx++]) != 0) {
                if (chr_encode(ch, dest, &dest_off, size - 1) != EOK)
                        break;
        }

        dest[dest_off] = '\0';
}

char *wstr_to_astr(const wchar_t *src)
{
        char dbuf[STR_BOUNDS(1)];
        char *str;
        wchar_t ch;

        size_t src_idx;
        size_t dest_off;
        size_t dest_size;

        /* Compute size of encoded string. */

        src_idx = 0;
        dest_size = 0;

        while ((ch = src[src_idx++]) != 0) {
                dest_off = 0;
                if (chr_encode(ch, dbuf, &dest_off, STR_BOUNDS(1)) != EOK)
                        break;
                dest_size += dest_off;
        }

        str = malloc(dest_size + 1);
        if (str == NULL)
                return NULL;

        /* Encode string. */

        src_idx = 0;
        dest_off = 0;

        while ((ch = src[src_idx++]) != 0) {
                if (chr_encode(ch, str, &dest_off, dest_size) != EOK)
                        break;
        }

        str[dest_size] = '\0';
        return str;
}


void str_to_wstr(wchar_t *dest, size_t dlen, const char *src)
{
        size_t offset;
        size_t di;
        wchar_t c;

        assert(dlen > 0);

        offset = 0;
        di = 0;

        do {
                if (di >= dlen - 1)
                        break;

                c = str_decode(src, &offset, STR_NO_LIMIT);
                dest[di++] = c;
        } while (c != '\0');

        dest[dlen - 1] = '\0';
}

char *str_chr(const char *str, wchar_t ch)
{
        wchar_t acc;
        size_t off = 0;
        size_t last = 0;
        
        while ((acc = str_decode(str, &off, STR_NO_LIMIT)) != 0) {
                if (acc == ch)
                        return (char *) (str + last);
                last = off;
        }
        
        return NULL;
}

char *str_rchr(const char *str, wchar_t ch)
{
        wchar_t acc;
        size_t off = 0;
        size_t last = 0;
        const char *res = NULL;
        
        while ((acc = str_decode(str, &off, STR_NO_LIMIT)) != 0) {
                if (acc == ch)
                        res = (str + last);
                last = off;
        }
        
        return (char *) res;
}

bool wstr_linsert(wchar_t *str, wchar_t ch, size_t pos, size_t max_pos)
{
        size_t len = wstr_length(str);
        
        if ((pos > len) || (pos + 1 > max_pos))
                return false;
        
        size_t i;
        for (i = len; i + 1 > pos; i--)
                str[i + 1] = str[i];
        
        str[pos] = ch;
        
        return true;
}

bool wstr_remove(wchar_t *str, size_t pos)
{
        size_t len = wstr_length(str);
        
        if (pos >= len)
                return false;
        
        size_t i;
        for (i = pos + 1; i <= len; i++)
                str[i - 1] = str[i];
        
        return true;
}

int stricmp(const char *a, const char *b)
{
        int c = 0;
        
        while (a[c] && b[c] && (!(tolower(a[c]) - tolower(b[c]))))
                c++;
        
        return (tolower(a[c]) - tolower(b[c]));
}

static unsigned long
_strtoul(const char *nptr, char **endptr, int base, char *sgn)
{
        unsigned char c;
        unsigned long result = 0;
        unsigned long a, b;
        const char *str = nptr;
        const char *tmpptr;
        
        while (isspace(*str))
                str++;
        
        if (*str == '-') {
                *sgn = 1;
                ++str;
        } else if (*str == '+')
                ++str;
        
        if (base) {
                if ((base == 1) || (base > 36)) {
                        /* FIXME: set errno to EINVAL */
                        return 0;
                }
                if ((base == 16) && (*str == '0') && ((str[1] == 'x') ||
                    (str[1] == 'X'))) {
                        str += 2;
                }
        } else {
                base = 10;
                
                if (*str == '0') {
                        base = 8;
                        if ((str[1] == 'X') || (str[1] == 'x'))  {
                                base = 16;
                                str += 2;
                        }
                } 
        }
        
        tmpptr = str;

        while (*str) {
                c = *str;
                c = (c >= 'a' ? c - 'a' + 10 : (c >= 'A' ? c - 'A' + 10 :
                    (c <= '9' ? c - '0' : 0xff)));
                if (c > base) {
                        break;
                }
                
                a = (result & 0xff) * base + c;
                b = (result >> 8) * base + (a >> 8);
                
                if (b > (ULONG_MAX >> 8)) {
                        /* overflow */
                        /* FIXME: errno = ERANGE*/
                        return ULONG_MAX;
                }
        
                result = (b << 8) + (a & 0xff);
                ++str;
        }
        
        if (str == tmpptr) {
                /*
                 * No number was found => first invalid character is the first
                 * character of the string.
                 */
                /* FIXME: set errno to EINVAL */
                str = nptr;
                result = 0;
        }
        
        if (endptr)
                *endptr = (char *) str;

        if (nptr == str) { 
                /*FIXME: errno = EINVAL*/
                return 0;
        }

        return result;
}

long int strtol(const char *nptr, char **endptr, int base)
{
        char sgn = 0;
        unsigned long number = 0;
        
        number = _strtoul(nptr, endptr, base, &sgn);

        if (number > LONG_MAX) {
                if ((sgn) && (number == (unsigned long) (LONG_MAX) + 1)) {
                        /* FIXME: set 0 to errno */
                        return number;          
                }
                /* FIXME: set ERANGE to errno */
                return (sgn ? LONG_MIN : LONG_MAX);     
        }
        
        return (sgn ? -number : number);
}

char *str_dup(const char *src)
{
        size_t size = str_size(src) + 1;
        char *dest = (char *) malloc(size);
        if (dest == NULL)
                return (char *) NULL;
        
        str_cpy(dest, size, src);
        return dest;
}

char *str_ndup(const char *src, size_t n)
{
        size_t size = str_size(src);
        if (size > n)
                size = n;
        
        char *dest = (char *) malloc(size + 1);
        if (dest == NULL)
                return (char *) NULL;
        
        str_ncpy(dest, size + 1, src, size);
        return dest;
}


unsigned long strtoul(const char *nptr, char **endptr, int base)
{
        char sgn = 0;
        unsigned long number = 0;
        
        number = _strtoul(nptr, endptr, base, &sgn);

        return (sgn ? -number : number);
}

char *strtok(char *s, const char *delim)
{
        static char *next;

        return strtok_r(s, delim, &next);
}

char *strtok_r(char *s, const char *delim, char **next)
{
        char *start, *end;

        if (s == NULL)
                s = *next;

        /* Skip over leading delimiters. */
        while (*s && (str_chr(delim, *s) != NULL)) ++s;
        start = s;

        /* Skip over token characters. */
        while (*s && (str_chr(delim, *s) == NULL)) ++s;
        end = s;
        *next = (*s ? s + 1 : s);

        if (start == end) {
                return NULL;    /* No more tokens. */
        }

        /* Overwrite delimiter with NULL terminator. */
        *end = '\0';
        return start;
}

static int str_uint(const char *nptr, char **endptr, unsigned int base,
    bool *neg, uint64_t *result)
{
        assert(endptr != NULL);
        assert(neg != NULL);
        assert(result != NULL);
        
        *neg = false;
        const char *str = nptr;
        
        /* Ignore leading whitespace */
        while (isspace(*str))
                str++;
        
        if (*str == '-') {
                *neg = true;
                str++;
        } else if (*str == '+')
                str++;
        
        if (base == 0) {
                /* Decode base if not specified */
                base = 10;
                
                if (*str == '0') {
                        base = 8;
                        str++;
                        
                        switch (*str) {
                        case 'b':
                        case 'B':
                                base = 2;
                                str++;
                                break;
                        case 'o':
                        case 'O':
                                base = 8;
                                str++;
                                break;
                        case 'd':
                        case 'D':
                        case 't':
                        case 'T':
                                base = 10;
                                str++;
                                break;
                        case 'x':
                        case 'X':
                                base = 16;
                                str++;
                                break;
                        default:
                                str--;
                        }
                }
        } else {
                /* Check base range */
                if ((base < 2) || (base > 36)) {
                        *endptr = (char *) str;
                        return EINVAL;
                }
        }
        
        *result = 0;
        const char *startstr = str;
        
        while (*str != 0) {
                unsigned int digit;
                
                if ((*str >= 'a') && (*str <= 'z'))
                        digit = *str - 'a' + 10;
                else if ((*str >= 'A') && (*str <= 'Z'))
                        digit = *str - 'A' + 10;
                else if ((*str >= '0') && (*str <= '9'))
                        digit = *str - '0';
                else
                        break;
                
                if (digit >= base)
                        break;
                
                uint64_t prev = *result;
                *result = (*result) * base + digit;
                
                if (*result < prev) {
                        /* Overflow */
                        *endptr = (char *) str;
                        return EOVERFLOW;
                }
                
                str++;
        }
        
        if (str == startstr) {
                /*
                 * No digits were decoded => first invalid character is
                 * the first character of the string.
                 */
                str = nptr;
        }
        
        *endptr = (char *) str;
        
        if (str == nptr)
                return EINVAL;
        
        return EOK;
}

int str_uint64(const char *nptr, char **endptr, unsigned int base,
    bool strict, uint64_t *result)
{
        assert(result != NULL);
        
        bool neg;
        char *lendptr;
        int ret = str_uint(nptr, &lendptr, base, &neg, result);
        
        if (endptr != NULL)
                *endptr = (char *) lendptr;
        
        if (ret != EOK)
                return ret;
        
        /* Do not allow negative values */
        if (neg)
                return EINVAL;
        
        /* Check whether we are at the end of
           the string in strict mode */
        if ((strict) && (*lendptr != 0))
                return EINVAL;
        
        return EOK;
}

int str_size_t(const char *nptr, char **endptr, unsigned int base,
    bool strict, size_t *result)
{
        assert(result != NULL);
        
        bool neg;
        char *lendptr;
        uint64_t res;
        int ret = str_uint(nptr, &lendptr, base, &neg, &res);
        
        if (endptr != NULL)
                *endptr = (char *) lendptr;
        
        if (ret != EOK)
                return ret;
        
        /* Do not allow negative values */
        if (neg)
                return EINVAL;
        
        /* Check whether we are at the end of
           the string in strict mode */
        if ((strict) && (*lendptr != 0))
                return EINVAL;
        
        /* Check for overflow */
        size_t _res = (size_t) res;
        if (_res != res)
                return EOVERFLOW;
        
        *result = _res;
        
        return EOK;
}

void order_suffix(const uint64_t val, uint64_t *rv, char *suffix)
{
        if (val > UINT64_C(10000000000000000000)) {
                *rv = val / UINT64_C(1000000000000000000);
                *suffix = 'Z';
        } else if (val > UINT64_C(1000000000000000000)) {
                *rv = val / UINT64_C(1000000000000000);
                *suffix = 'E';
        } else if (val > UINT64_C(1000000000000000)) {
                *rv = val / UINT64_C(1000000000000);
                *suffix = 'T';
        } else if (val > UINT64_C(1000000000000)) {
                *rv = val / UINT64_C(1000000000);
                *suffix = 'G';
        } else if (val > UINT64_C(1000000000)) {
                *rv = val / UINT64_C(1000000);
                *suffix = 'M';
        } else if (val > UINT64_C(1000000)) {
                *rv = val / UINT64_C(1000);
                *suffix = 'k';
        } else {
                *rv = val;
                *suffix = ' ';
        }
}

void bin_order_suffix(const uint64_t val, uint64_t *rv, const char **suffix,
    bool fixed)
{
        if (val > UINT64_C(1152921504606846976)) {
                *rv = val / UINT64_C(1125899906842624);
                *suffix = "EiB";
        } else if (val > UINT64_C(1125899906842624)) {
                *rv = val / UINT64_C(1099511627776);
                *suffix = "TiB";
        } else if (val > UINT64_C(1099511627776)) {
                *rv = val / UINT64_C(1073741824);
                *suffix = "GiB";
        } else if (val > UINT64_C(1073741824)) {
                *rv = val / UINT64_C(1048576);
                *suffix = "MiB";
        } else if (val > UINT64_C(1048576)) {
                *rv = val / UINT64_C(1024);
                *suffix = "KiB";
        } else {
                *rv = val;
                if (fixed)
                        *suffix = "B  ";
                else
                        *suffix = "B";
        }
}

---