aboutsummaryrefslogtreecommitdiff
path: root/quantum/unicode/utf8.c
diff options
context:
space:
mode:
authorNick Brassel <[email protected]>2022-11-28 07:54:00 +1100
committerNick Brassel <[email protected]>2022-11-28 07:54:00 +1100
commit4020674163fc80914059c4c9c3be5c0ae00bd150 (patch)
tree6f4187d72b04d03572adf507502afbda9726d696 /quantum/unicode/utf8.c
parent8f9b49dc5b05fd3421e47aa76822a5b2199dfca6 (diff)
parent9e78e65a566487b2f4fe7b663971a01deb6ddad2 (diff)
downloadqmk_firmware-0.19.0.tar.gz
qmk_firmware-0.19.0.zip
Merge remote-tracking branch 'upstream/develop'0.19.0
Diffstat (limited to 'quantum/unicode/utf8.c')
-rw-r--r--quantum/unicode/utf8.c46
1 files changed, 46 insertions, 0 deletions
diff --git a/quantum/unicode/utf8.c b/quantum/unicode/utf8.c
new file mode 100644
index 0000000000..4b2cd4d8d4
--- /dev/null
+++ b/quantum/unicode/utf8.c
@@ -0,0 +1,46 @@
+/* Copyright 2021 QMK
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "utf8.h"
+
+// Borrowed from https://nullprogram.com/blog/2017/10/06/
+const char *decode_utf8(const char *str, int32_t *code_point) {
+ const char *next;
+
+ if (str[0] < 0x80) { // U+0000-007F
+ *code_point = str[0];
+ next = str + 1;
+ } else if ((str[0] & 0xE0) == 0xC0) { // U+0080-07FF
+ *code_point = ((int32_t)(str[0] & 0x1F) << 6) | ((int32_t)(str[1] & 0x3F) << 0);
+ next = str + 2;
+ } else if ((str[0] & 0xF0) == 0xE0) { // U+0800-FFFF
+ *code_point = ((int32_t)(str[0] & 0x0F) << 12) | ((int32_t)(str[1] & 0x3F) << 6) | ((int32_t)(str[2] & 0x3F) << 0);
+ next = str + 3;
+ } else if ((str[0] & 0xF8) == 0xF0 && (str[0] <= 0xF4)) { // U+10000-10FFFF
+ *code_point = ((int32_t)(str[0] & 0x07) << 18) | ((int32_t)(str[1] & 0x3F) << 12) | ((int32_t)(str[2] & 0x3F) << 6) | ((int32_t)(str[3] & 0x3F) << 0);
+ next = str + 4;
+ } else {
+ *code_point = -1;
+ next = str + 1;
+ }
+
+ // part of a UTF-16 surrogate pair - invalid
+ if (*code_point >= 0xD800 && *code_point <= 0xDFFF) {
+ *code_point = -1;
+ }
+
+ return next;
+}