diff options
Diffstat (limited to 'src/utf8.c')
| -rw-r--r-- | src/utf8.c | 67 |
1 files changed, 0 insertions, 67 deletions
diff --git a/src/utf8.c b/src/utf8.c deleted file mode 100644 index abf5ef7..0000000 --- a/src/utf8.c +++ /dev/null @@ -1,67 +0,0 @@ -#include "utf8.h" - -#include <stdio.h> - -bool utf8_byte_is_unicode_start(uint8_t byte) { return (byte & 0xc0) == 0xc0; } -bool utf8_byte_is_unicode_continuation(uint8_t byte) { - return utf8_byte_is_unicode(byte) && !utf8_byte_is_unicode_start(byte); -} -bool utf8_byte_is_unicode(uint8_t byte) { return (byte & 0x80) != 0x0; } -bool utf8_byte_is_ascii(uint8_t byte) { return !utf8_byte_is_unicode(byte); } - -uint32_t utf8_nbytes_in_char(uint8_t byte) { - // length of char is the number of leading ones - // flip it and count number of leading zeros - uint8_t invb = ~byte; - return __builtin_clz((uint32_t)invb) - 24; -} - -// TODO: grapheme clusters, this returns the number of unicode code points -uint32_t utf8_nchars(uint8_t *bytes, uint32_t nbytes) { - uint32_t nchars = 0; - uint32_t expected = 0; - for (uint32_t bi = 0; bi < nbytes; ++bi) { - uint8_t byte = bytes[bi]; - if (utf8_byte_is_unicode(byte)) { - if (utf8_byte_is_unicode_start(byte)) { - expected = utf8_nbytes_in_char(byte) - 1; - } else { // continuation byte - --expected; - if (expected == 0) { - ++nchars; - } - } - } else { // ascii - ++nchars; - } - } - return nchars; -} - -// TODO: grapheme clusters, this uses the number of unicode code points -uint32_t utf8_nbytes(uint8_t *bytes, uint32_t nbytes, uint32_t nchars) { - - uint32_t bi = 0; - uint32_t chars = 0; - uint32_t expected = 0; - - while (chars < nchars && bi < nbytes) { - uint8_t byte = bytes[bi]; - if (utf8_byte_is_unicode(byte)) { - if (utf8_byte_is_unicode_start(byte)) { - expected = utf8_nbytes_in_char(byte) - 1; - } else { // continuation char - --expected; - if (expected == 0) { - ++chars; - } - } - } else { // ascii - ++chars; - } - - ++bi; - } - - return bi; -} |
