diff options
Diffstat (limited to 'src/mds-kbdc/string.c')
| -rw-r--r-- | src/mds-kbdc/string.c | 166 |
1 files changed, 83 insertions, 83 deletions
diff --git a/src/mds-kbdc/string.c b/src/mds-kbdc/string.c index a334611..c0e9852 100644 --- a/src/mds-kbdc/string.c +++ b/src/mds-kbdc/string.c @@ -30,12 +30,13 @@ * @param string The string * @return The length of the string */ -size_t string_length(const char32_t* restrict string) +size_t +string_length(const char32_t *restrict string) { - size_t i = 0; - while (string[i] != -1) - i++; - return i; + size_t i = 0; + while (string[i] != -1) + i++; + return i; } @@ -45,43 +46,41 @@ size_t string_length(const char32_t* restrict string) * @param string The UTF-8 string * @return The string in UTF-32, `NULL` on error */ -char32_t* string_decode(const char* restrict string) +char32_t * +string_decode(const char *restrict string) { - size_t i, j, n, length = 0; - char32_t* rc; - - /* Get the length of the UTF-32 string, excluding termination. */ - for (i = 0; string[i]; i++) - if ((string[i] & 0xC0) != 0x80) - length++; - - /* Allocated UTF-32 string. */ - fail_if (xmalloc(rc, length + 1, char32_t)); - - /* Convert to UTF-32. */ - for (i = j = n = 0; string[i]; i++) - { - char c = string[i]; - if (n) - { - rc[j] <<= 6, rc[j] |= c & 0x3F; - if (--n == 0) - j++; - } - else if ((c & 0xC0) == 0xC0) - { - while (c & 0x80) - n++, c = (char)(c << 1); - rc[j] = c >> n--; + size_t i, j, n, length = 0; + char32_t *rc; + char c; + + /* Get the length of the UTF-32 string, excluding termination. */ + for (i = 0; string[i]; i++) + if ((string[i] & 0xC0) != 0x80) + length++; + + /* Allocated UTF-32 string. */ + fail_if (xmalloc(rc, length + 1, char32_t)); + + /* Convert to UTF-32. */ + for (i = j = n = 0; string[i]; i++) { + c = string[i]; + if (n) { + rc[j] <<= 6, rc[j] |= c & 0x3F; + if (!--n) + j++; + } else if ((c & 0xC0) == 0xC0) { + while (c & 0x80) + n++, c = (char)(c << 1); + rc[j] = c >> n--; + } else { + rc[j++] = c & 255; + } } - else - rc[j++] = c & 255; - } - /* -1-terminate and return. */ - return rc[length] = -1, rc; - fail: - return NULL; + /* -1-terminate and return. */ + return rc[length] = -1, rc; +fail: + return NULL; } @@ -93,42 +92,43 @@ char32_t* string_decode(const char* restrict string) * @param string The UTF-32 string * @return The string in UTF-8, `NULL` on error */ -char* string_encode(const char32_t* restrict string) +char * +string_encode(const char32_t *restrict string) { - size_t i, j, n = string_length(string); - char* restrict rc; - - /* Allocated Modified UTF-8 string. */ - fail_if (xmalloc(rc, 7 * n + 1, char)); - - /* Convert to Modified UTF-8. */ - for (i = j = 0; i < n; i++) - { -#define _c(s) rc[j++] = (char)(((c >> (s)) & 0x3F) | 0x80) -#define _t(s) (0 < c) && ((uint32_t)c < (uint32_t)(1ULL << s)) - char32_t c = string[i]; - if (c == 0) rc[j++] = (char)0xC0, rc[j++] = (char)0x80; - else if (_t( 7)) rc[j++] = (char)c; - else if (_t(11)) rc[j++] = (char)((c >> 6) | 0xC0), _c( 0); - else if (_t(16)) rc[j++] = (char)((c >> 12) | 0xE0), _c( 6), _c( 0); - else if (_t(21)) rc[j++] = (char)((c >> 18) | 0xF0), _c(12), _c( 6), _c( 0); - /* UTF-8 actually ends here, fits 32 planes. */ - else if (_t(26)) rc[j++] = (char)((c >> 24) | 0xF8), _c(18), _c(12), _c( 6), _c(0); - else if (_t(31)) rc[j++] = (char)((c >> 30) | 0xFC), _c(24), _c(18), _c(12), _c(6), _c(0); - /* The orginal UTF-8 specification ended here, fits 31 bits. - * However, we added another byte so we can fit 32 bits - * (actually we ca now fit 36 bits.) - * However, we only needed this in `string_decode` which would - * not require any changed, but we added it here for symmetry. */ - else rc[j++] = (char)0xFE, _c(30), _c(24), _c(18), _c(12), _c(6), _c(0); + size_t i, j, n = string_length(string); + char *restrict rc; + char32_t c; + + /* Allocated Modified UTF-8 string. */ + fail_if (xmalloc(rc, 7 * n + 1, char)); + + /* Convert to Modified UTF-8. */ + for (i = j = 0; i < n; i++) { +#define _c(s) rc[j++] = (char)(((c >> (s)) & 0x3F) | 0x80) +#define _t(s) (0 < c) && ((uint32_t)c < (uint32_t)(1ULL << (s))) + c = string[i]; + if (!c) rc[j++] = (char)0xC0, rc[j++] = (char)0x80; + else if (_t( 7)) rc[j++] = (char)c; + else if (_t(11)) rc[j++] = (char)((c >> 6) | 0xC0), _c( 0); + else if (_t(16)) rc[j++] = (char)((c >> 12) | 0xE0), _c( 6), _c( 0); + else if (_t(21)) rc[j++] = (char)((c >> 18) | 0xF0), _c(12), _c( 6), _c( 0); + /* UTF-8 actually ends here, fits 32 planes. */ + else if (_t(26)) rc[j++] = (char)((c >> 24) | 0xF8), _c(18), _c(12), _c( 6), _c(0); + else if (_t(31)) rc[j++] = (char)((c >> 30) | 0xFC), _c(24), _c(18), _c(12), _c(6), _c(0); + /* The orginal UTF-8 specification ended here, fits 31 bits. + * However, we added another byte so we can fit 32 bits + * (actually we ca now fit 36 bits.) + * However, we only needed this in `string_decode` which would + * not require any changed, but we added it here for symmetry. */ + else rc[j++] = (char)0xFE, _c(30), _c(24), _c(18), _c(12), _c(6), _c(0); #undef _t #undef _c - } - - /* NUL-terminate and return. */ - return rc[j] = '\0', rc; - fail: - return NULL; + } + + /* NUL-terminate and return. */ + return rc[j] = '\0', rc; +fail: + return NULL; } @@ -138,16 +138,16 @@ char* string_encode(const char32_t* restrict string) * @param string The string * @return A duplicate of the string, `NULL` on error or if `string` is `NULL` */ -char32_t* string_dup(const char32_t* restrict string) +char32_t * +string_dup(const char32_t *restrict string) { - size_t n; - char32_t* rc; - if (string == NULL) - return NULL; - n = string_length(string) + 1; - fail_if (xmemdup(rc, string, n, char32_t)); - return rc; - fail: - return NULL; + size_t n; + char32_t *rc; + if (!string) + return NULL; + n = string_length(string) + 1; + fail_if (xmemdup(rc, string, n, char32_t)); + return rc; +fail: + return NULL; } - |