* Decode a character from UTF-8. * @param buf Binary data. * @return Length and character. Length is zero, if the input data is invalid. */
| 46 | * @return Length and character. Length is zero, if the input data is invalid. |
| 47 | */ |
| 48 | [[nodiscard]] std::pair<size_t, char32_t> DecodeUtf8(std::string_view buf) |
| 49 | { |
| 50 | if (buf.size() >= 1 && !HasBit(buf[0], 7)) { |
| 51 | /* Single byte character: 0xxxxxxx */ |
| 52 | char32_t c = buf[0]; |
| 53 | return {1, c}; |
| 54 | } else if (buf.size() >= 2 && GB(buf[0], 5, 3) == 6) { |
| 55 | if (IsUtf8Part(buf[1])) { |
| 56 | /* Double byte character: 110xxxxx 10xxxxxx */ |
| 57 | char32_t c = GB(buf[0], 0, 5) << 6 | GB(buf[1], 0, 6); |
| 58 | if (c >= 0x80) return {2, c}; |
| 59 | } |
| 60 | } else if (buf.size() >= 3 && GB(buf[0], 4, 4) == 14) { |
| 61 | if (IsUtf8Part(buf[1]) && IsUtf8Part(buf[2])) { |
| 62 | /* Triple byte character: 1110xxxx 10xxxxxx 10xxxxxx */ |
| 63 | char32_t c = GB(buf[0], 0, 4) << 12 | GB(buf[1], 0, 6) << 6 | GB(buf[2], 0, 6); |
| 64 | if (c >= 0x800) return {3, c}; |
| 65 | } |
| 66 | } else if (buf.size() >= 4 && GB(buf[0], 3, 5) == 30) { |
| 67 | if (IsUtf8Part(buf[1]) && IsUtf8Part(buf[2]) && IsUtf8Part(buf[3])) { |
| 68 | /* 4 byte character: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */ |
| 69 | char32_t c = GB(buf[0], 0, 3) << 18 | GB(buf[1], 0, 6) << 12 | GB(buf[2], 0, 6) << 6 | GB(buf[3], 0, 6); |
| 70 | if (c >= 0x10000 && c <= 0x10FFFF) return {4, c}; |
| 71 | } |
| 72 | } |
| 73 | return {}; |
| 74 | } |
| 75 | |
| 76 | /** |
| 77 | * Create iterator pointing at codepoint, which occupies the byte position "offset". |
no test coverage detected