Subversion Repositories HelenOS

 *

 * Strings and characters use the Universal Character Set (UCS). The standard

 * strings, called just strings are encoded in UTF-8. Wide strings (encoded

 * in UTF-32) are supported to a limited degree. A single character is

 * represented as wchar_t.@n

 *

 * Overview of the terminology:@n

 *

 *  Term                  Meaning

 *  --------------------  ----------------------------------------------------

 *  byte                  8 bits stored in uint8_t (unsigned 8 bit integer)

 *

 *  character             UTF-32 encoded Unicode character, stored in wchar_t

 *                        (signed 32 bit integer), code points 0 .. 1114111

 *                        are valid

 *

 *  ASCII character       7 bit encoded ASCII character, stored in char

 *                        (usually signed 8 bit integer), code points 0 .. 127

 *                        are valid

 *

 *  string                UTF-8 encoded NULL-terminated Unicode string, char *

 *

 *  wide string           UTF-32 encoded NULL-terminated Unicode string,

 *                        wchar_t *

 *

 *  [wide] string size    number of BYTES in a [wide] string (excluding

 *                        the NULL-terminator), size_t

 *

 *  [wide] string length  number of CHARACTERS in a [wide] string (excluding

 *                        the NULL-terminator), count_t

 *

 *  [wide] string width   number of display cells on a monospace display taken

 *                        by a [wide] string, count_t

 *

 *

 * Overview of string metrics:@n

 *

 *  Metric  Abbrev.  Type     Meaning

 *  ------  ------   ------   -------------------------------------------------

 *  size    n        size_t   number of BYTES in a string (excluding the

 *                            NULL-terminator)

 *

 *  length  l        count_t  number of CHARACTERS in a string (excluding the

 *                            null terminator)

 *

 *  width  w         count_t  number of display cells on a monospace display

 *                            taken by a string

 *

 *

 * Function naming prefixes:@n

 *

 *  chr_    operate on characters

 *  ascii_  operate on ASCII characters

 *  str_    operate on strings

 *  wstr_   operate on wide strings

 *

 *  [w]str_[n|l|w]  operate on a prefix limited by size, length

 *                  or width

 *

 *

 * A specific character inside a [wide] string can be referred to by:@n

 *

 *  pointer (char *, wchar_t *)

 *  byte offset (size_t)

 *  character index (count_t)

 *

#include <align.h>

/** Byte mask consisting of lowest @n bits (out of 8) */

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

/** Byte mask consisting of lowest @n bits (out of 32) */

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

/** Byte mask consisting of highest @n bits (out of 8) */

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

/** Number of data bits in a UTF-8 continuation byte */

#define CONT_BITS  6

 * Decode a single character from a string of size @a size. Decoding starts

 * at @a offset and this offset is moved to the beginning of the next

 * character. In case of decoding error, offset generally advances at least

 * @param str    String (not necessarily NULL-terminated).

 * @param offset Byte offset in string where to start decoding.

 *

 * @return Value of decoded character, U_SPECIAL on decoding error or

 *         NULL if attempt to decode beyond @a size.

    if (*offset + 1 > size)

    /* First byte read from string */

    /* 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--;

    }

 * it into a buffer at @a offset. Encoding starts at @a offset and this offset

 *

 * @param ch     Input character.

 * @param str    Output buffer.

 * @param offset Byte offset where to start writing.

 *

 * @return EOK if the character was encoded successfully, EOVERFLOW if there

 *     was not enough space in the output buffer or EINVAL if the character

 *     code was invalid.

 */

int chr_encode(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;

}

/** Get size of string.

 *

 * Get the number of bytes which are used by the string @a str (excluding the

 * NULL-terminator).

        size++;

}

/** Get size of wide string.

 *

 * Get the number of bytes which are used by the wide string @a str (excluding the

 * NULL-terminator).

 *

 * @param str Wide string to consider.

 *

 * @return Number of bytes used by the wide string

 *

 */

size_t wstr_size(const wchar_t *str)

{

    return (wstr_length(str) * sizeof(wchar_t));

}

/** Get size of string with length limit.

 *

 * Get the number of bytes which are used by up to @a max_len first

 * characters in the string @a str. If @a max_len is greater than

 * the length of @a str, the entire string is measured (excluding the

 * NULL-terminator).

 *

 * @param str     String to consider.

 * @param max_len Maximum number of characters to measure.

 *

 * @return Number of bytes used by the characters.

 *

 */

size_t str_lsize(const char *str, count_t max_len)

{

    count_t len = 0;

    while (len < max_len) {

        if (str_decode(str, &offset, STR_NO_LIMIT) == 0)

        len++;

}

/** Get size of wide string with length limit.

 *

 * Get the number of bytes which are used by up to @a max_len first

 * wide characters in the wide string @a str. If @a max_len is greater than

 * the length of @a str, the entire wide string is measured (excluding the

 * NULL-terminator).

 *

 * @param str     Wide string to consider.

 * @param max_len Maximum number of wide characters to measure.

 *

 * @return Number of bytes used by the wide characters.

 *

 */

size_t wstr_lsize(const wchar_t *str, count_t max_len)

{

    return (wstr_nlength(str, max_len * sizeof(wchar_t)) * sizeof(wchar_t));

}

/** Get number of characters in a string.

 *

 * @param str NULL-terminated string.

 *

 * @return Number of characters in string.

 *

 */

count_t str_length(const char *str)

{

    count_t len = 0;

    size_t offset = 0;

    while (str_decode(str, &offset, STR_NO_LIMIT) != 0)

/** Get number of characters in a wide string.

 *

 * @param str NULL-terminated wide string.

 *

 * @return Number of characters in @a str.

 *

 */

count_t wstr_length(const wchar_t *wstr)

{

    count_t len = 0;

    while (*wstr++ != 0)

        len++;

    return len;

}

/** Get number of characters in a string with size limit.

 *

 * @param str  NULL-terminated string.

 * @param size Maximum number of bytes to consider.

 *

 * @return Number of characters in string.

 *

 */

count_t str_nlength(const char *str, size_t size)

{

    count_t len = 0;

    size_t offset = 0;

    while (str_decode(str, &offset, size) != 0)

        len++;

    return len;

}

 * @param str  NULL-terminated string.

 * @param size Maximum number of bytes to consider.

 *

 * @return Number of characters in string.

 *

 */

count_t wstr_nlength(const wchar_t *str, size_t size)

{

    count_t len = 0;

    count_t limit = ALIGN_DOWN(size, sizeof(wchar_t));

    count_t offset = 0;

    while ((offset < limit) && (*str++ != 0)) {

        len++;

        offset += sizeof(wchar_t);

    }

    return len;

}

/** Check whether character is plain ASCII.

 *

 * @return True if character is plain ASCII.

 *

 */

bool ascii_check(wchar_t ch)

{

    if ((ch >= 0) && (ch <= 127))

        return true;

    return false;

}

/** Check whether character is valid

 *

 * @return True if character is a valid Unicode code point.

 *

 */

bool chr_check(wchar_t ch)

{

    if ((ch >= 0) && (ch <= 1114111))

        return true;

    return false;

}

/** Compare two NULL terminated strings.

 *

 *         1 if second smaller.

    wchar_t c1 = 0;

    size_t off1 = 0;

    size_t off2 = 0;

    while (true) {

        c1 = str_decode(s1, &off1, STR_NO_LIMIT);

        if (c1 == 0 || c2 == 0)

            break;     

    return 0;

}

/** Compare two NULL terminated strings with length limit.

 *

 * Do a char-by-char comparison of two NULL-terminated strings.

 * The strings are considered equal iff they consist of the same

 * characters on the minimum of their lengths and the length limit.

 *

 * @param s1      First string to compare.

 * @param s2      Second string to compare.

 * @param max_len Maximum number of characters to consider.

 *

 * @return 0 if the strings are equal, -1 if first is smaller,

 *         1 if second smaller.

 *

 */

int str_lcmp(const char *s1, const char *s2, count_t max_len)

{

    wchar_t c1 = 0;

    wchar_t c2 = 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;

            break;

        ++len; 

    }

    return 0;

}

 * Copy source string @a src to destination buffer @a dst.

 * No more than @a size bytes are written. NULL-terminator is always

 * written after the last succesfully copied character (i.e. if the

 * destination buffer is has at least 1 byte, it will be always

 * NULL-terminated).

 *

    /* No space for the NULL-terminator in the buffer */

        return;

    wchar_t ch;

    size_t str_off = 0;

    size_t dst_off = 0;

    while ((ch = str_decode(src, &str_off, STR_NO_LIMIT)) != 0) {

        if (chr_encode(ch, dst, &dst_off, size) != EOK)

            break;

    }

        dst[size - 1] = 0;

    else

        dst[dst_off] = 0;

}

/** Copy NULL-terminated wide string to string

 *

 * Copy source wide string @a src to destination buffer @a dst.

 * No more than @a size bytes are written. NULL-terminator is always

 * written after the last succesfully copied character (i.e. if the

 * destination buffer is has at least 1 byte, it will be always

 * NULL-terminated).

 *

 * @param src   Source wide string.

 * @param dst   Destination buffer.

 * @param count Size of the destination buffer.

 *

 */

void wstr_nstr(char *dst, const wchar_t *src, size_t size)

{

    /* No space for the NULL-terminator in the buffer */

    wchar_t ch;

    count_t src_idx = 0;

    size_t dst_off = 0;

    while ((ch = src[src_idx++]) != 0) {

        if (chr_encode(ch, dst, &dst_off, size) != EOK)

            break;

    if (dst_off >= size)

    else

        dst[dst_off] = 0;

 *

 * @return Pointer to character in @a str or NULL if not found.

    wchar_t acc;

    while ((acc = str_decode(str, &off, STR_NO_LIMIT)) != 0) {

/** Insert a wide character into a wide string.

 *

 * Insert a wide character into a wide string at position

 * @a pos. The characters after the position are shifted.

 *

 * @param str     String to insert to.

 * @param ch      Character to insert to.

 * @param pos     Character index where to insert.

 @ @param max_pos Characters in the buffer.

 *

 * @return True if the insertion was sucessful, false if the position

 *         is out of bounds.

 *

 */

bool wstr_linsert(wchar_t *str, wchar_t ch, count_t pos, count_t max_pos)

{

    count_t len = wstr_length(str);

    if ((pos > len) || (pos + 1 > max_pos))

        return false;

    count_t i;

    for (i = len; i + 1 > pos; i--)

        str[i + 1] = str[i];

    str[pos] = ch;

    return true;

}

/** Remove a wide character from a wide string.

 *

 * Remove a wide character from a wide string at position

 * @a pos. The characters after the position are shifted.

 *

 * @param str String to remove from.

 * @param pos Character index to remove.

 *

 * @return True if the removal was sucessful, false if the position

 *         is out of bounds.

 *

 */

bool wstr_remove(wchar_t *str, count_t pos)

{

    count_t len = wstr_length(str);

    if (pos >= len)

        return false;

    count_t i;

    for (i = pos + 1; i <= len; i++)

        str[i - 1] = str[i];

    return true;

}