From f951c2357fc4c1f25c7af34958e37cd65a6979d2 Mon Sep 17 00:00:00 2001 From: emersion Date: Thu, 17 Jan 2019 12:45:41 +0100 Subject: [PATCH] Remove unicode.c --- common/meson.build | 1 - common/unicode.c | 101 --------------------------------------------- include/unicode.h | 33 --------------- 3 files changed, 135 deletions(-) delete mode 100644 common/unicode.c delete mode 100644 include/unicode.h diff --git a/common/meson.build b/common/meson.build index 4ad872d1d..3af1f1d56 100644 --- a/common/meson.build +++ b/common/meson.build @@ -9,7 +9,6 @@ lib_sway_common = static_library( 'list.c', 'pango.c', 'stringop.c', - 'unicode.c', 'util.c' ), dependencies: [ diff --git a/common/unicode.c b/common/unicode.c deleted file mode 100644 index 5070e0835..000000000 --- a/common/unicode.c +++ /dev/null @@ -1,101 +0,0 @@ -#include -#include -#include "unicode.h" - -size_t utf8_chsize(uint32_t ch) { - if (ch < 0x80) { - return 1; - } else if (ch < 0x800) { - return 2; - } else if (ch < 0x10000) { - return 3; - } - return 4; -} - -static const uint8_t masks[] = { - 0x7F, - 0x1F, - 0x0F, - 0x07, - 0x03, - 0x01 -}; - -uint32_t utf8_decode(const char **char_str) { - uint8_t **s = (uint8_t **)char_str; - - uint32_t cp = 0; - if (**s < 128) { - // shortcut - cp = **s; - ++*s; - return cp; - } - int size = utf8_size((char *)*s); - if (size == -1) { - ++*s; - return UTF8_INVALID; - } - uint8_t mask = masks[size - 1]; - cp = **s & mask; - ++*s; - while (--size) { - cp <<= 6; - cp |= **s & 0x3f; - ++*s; - } - return cp; -} - -size_t utf8_encode(char *str, uint32_t ch) { - size_t len = 0; - uint8_t first; - - if (ch < 0x80) { - first = 0; - len = 1; - } else if (ch < 0x800) { - first = 0xc0; - len = 2; - } else if (ch < 0x10000) { - first = 0xe0; - len = 3; - } else { - first = 0xf0; - len = 4; - } - - for (size_t i = len - 1; i > 0; --i) { - str[i] = (ch & 0x3f) | 0x80; - ch >>= 6; - } - - str[0] = ch | first; - return len; -} - - -static const struct { - uint8_t mask; - uint8_t result; - int octets; -} sizes[] = { - { 0x80, 0x00, 1 }, - { 0xE0, 0xC0, 2 }, - { 0xF0, 0xE0, 3 }, - { 0xF8, 0xF0, 4 }, - { 0xFC, 0xF8, 5 }, - { 0xFE, 0xF8, 6 }, - { 0x80, 0x80, -1 }, -}; - -int utf8_size(const char *s) { - uint8_t c = (uint8_t)*s; - for (size_t i = 0; i < sizeof(sizes) / sizeof(*sizes); ++i) { - if ((c & sizes[i].mask) == sizes[i].result) { - return sizes[i].octets; - } - } - return -1; -} diff --git a/include/unicode.h b/include/unicode.h deleted file mode 100644 index e2ee9588c..000000000 --- a/include/unicode.h +++ /dev/null @@ -1,33 +0,0 @@ -#ifndef _SWAY_UNICODE_H -#define _SWAY_UNICODE_H -#include -#include - -// Technically UTF-8 supports up to 6 byte codepoints, but Unicode itself -// doesn't really bother with more than 4. -#define UTF8_MAX_SIZE 4 - -#define UTF8_INVALID 0x80 - -/** - * Grabs the next UTF-8 character and advances the string pointer - */ -uint32_t utf8_decode(const char **str); - -/** - * Encodes a character as UTF-8 and returns the length of that character. - */ -size_t utf8_encode(char *str, uint32_t ch); - -/** - * Returns the size of the next UTF-8 character - */ -int utf8_size(const char *str); - -/** - * Returns the size of a UTF-8 character - */ -size_t utf8_chsize(uint32_t ch); - -#endif -