diff --git a/BAN/include/BAN/UTF8.h b/BAN/include/BAN/UTF8.h index 918d1be18f..05066eef45 100644 --- a/BAN/include/BAN/UTF8.h +++ b/BAN/include/BAN/UTF8.h @@ -21,20 +21,21 @@ namespace BAN::UTF8 return 0; } - constexpr uint32_t to_codepoint(uint8_t* bytes) + template requires (sizeof(T) == 1) + constexpr uint32_t to_codepoint(const T* bytes) { uint32_t length = byte_length(bytes[0]); for (uint32_t i = 1; i < length; i++) - if ((bytes[i] & 0xC0) != 0x80) + if (((uint8_t)bytes[i] & 0xC0) != 0x80) return UTF8::invalid; switch (length) { - case 1: return ((bytes[0] & 0x80) != 0x00) ? UTF8::invalid : bytes[0]; - case 2: return ((bytes[0] & 0xE0) != 0xC0) ? UTF8::invalid : ((bytes[0] & 0x1F) << 6) | (bytes[1] & 0x3F); - case 3: return ((bytes[0] & 0xF0) != 0xE0) ? UTF8::invalid : ((bytes[0] & 0x0F) << 12) | ((bytes[1] & 0x3F) << 6) | (bytes[2] & 0x3F); - case 4: return ((bytes[0] & 0xF8) != 0xF0) ? UTF8::invalid : ((bytes[0] & 0x07) << 18) | ((bytes[1] & 0x3F) << 12) | ((bytes[2] & 0x3F) << 6) | (bytes[3] & 0x3F); + case 1: return (((uint8_t)bytes[0] & 0x80) != 0x00) ? UTF8::invalid : (uint8_t)bytes[0]; + case 2: return (((uint8_t)bytes[0] & 0xE0) != 0xC0) ? UTF8::invalid : (((uint8_t)bytes[0] & 0x1F) << 6) | ((uint8_t)bytes[1] & 0x3F); + case 3: return (((uint8_t)bytes[0] & 0xF0) != 0xE0) ? UTF8::invalid : (((uint8_t)bytes[0] & 0x0F) << 12) | (((uint8_t)bytes[1] & 0x3F) << 6) | ((uint8_t)bytes[2] & 0x3F); + case 4: return (((uint8_t)bytes[0] & 0xF8) != 0xF0) ? UTF8::invalid : (((uint8_t)bytes[0] & 0x07) << 18) | (((uint8_t)bytes[1] & 0x3F) << 12) | (((uint8_t)bytes[2] & 0x3F) << 6) | ((uint8_t)bytes[3] & 0x3F); } return UTF8::invalid;