UTF-8 sequence length starting at p, validating continuation bytes and the * lead-byte ranges (RFC 3629: no overlongs, no surrogates, max U+10FFFF). * `remaining` makes truncated multibyte input safe to inspect. */
| 102 | * lead-byte ranges (RFC 3629: no overlongs, no surrogates, max U+10FFFF). |
| 103 | * `remaining` makes truncated multibyte input safe to inspect. */ |
| 104 | static size_t utf8_sequence_length(const unsigned char *p, size_t remaining) { |
| 105 | if (!p || remaining == 0) { |
| 106 | return 0; |
| 107 | } |
| 108 | unsigned char c = p[0]; |
| 109 | if (c < 0x80) { |
| 110 | return 1; |
| 111 | } |
| 112 | if (c < 0xC2) { |
| 113 | return 0; /* bare continuation byte or overlong lead */ |
| 114 | } |
| 115 | if (c < 0xE0) { |
| 116 | return remaining >= 2 && (p[1] & 0xC0) == 0x80 ? 2 : 0; |
| 117 | } |
| 118 | if (c < 0xF0) { |
| 119 | if (remaining < 3) { |
| 120 | return 0; |
| 121 | } |
| 122 | unsigned char lo = c == 0xE0 ? 0xA0 : 0x80; |
| 123 | unsigned char hi = c == 0xED ? 0x9F : 0xBF; |
| 124 | if (p[1] < lo || p[1] > hi || (p[2] & 0xC0) != 0x80) { |
| 125 | return 0; |
| 126 | } |
| 127 | return 3; |
| 128 | } |
| 129 | if (c < 0xF5) { |
| 130 | if (remaining < 4) { |
| 131 | return 0; |
| 132 | } |
| 133 | unsigned char lo = c == 0xF0 ? 0x90 : 0x80; |
| 134 | unsigned char hi = c == 0xF4 ? 0x8F : 0xBF; |
| 135 | if (p[1] < lo || p[1] > hi || (p[2] & 0xC0) != 0x80 || (p[3] & 0xC0) != 0x80) { |
| 136 | return 0; |
| 137 | } |
| 138 | return 4; |
| 139 | } |
| 140 | return 0; |
| 141 | } |
| 142 | |
| 143 | static bool output_text_requires_encoding(const char *data, size_t len) { |
| 144 | static const char bytes_prefix[] = "@bytes:"; |
no outgoing calls
no test coverage detected