| 1 | //===-- StdStringExtractor.cpp ----------------------------------*- C++ -*-===// |
| 2 | // |
| 3 | // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. |
| 4 | // See https://llvm.org/LICENSE.txt for license information. |
| 5 | // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception |
| 6 | // |
| 7 | //===----------------------------------------------------------------------===// |
| 8 | |
| 9 | #include "StdStringExtractor.h" |
| 10 | |
| 11 | #include <cstdlib> |
| 12 | |
| 13 | static inline int xdigit_to_sint(char ch) { |
| 14 | if (ch >= 'a' && ch <= 'f') |
| 15 | return 10 + ch - 'a'; |
| 16 | if (ch >= 'A' && ch <= 'F') |
| 17 | return 10 + ch - 'A'; |
| 18 | if (ch >= '0' && ch <= '9') |
| 19 | return ch - '0'; |
| 20 | return -1; |
| 21 | } |
| 22 | |
| 23 | // StdStringExtractor constructor |
| 24 | StdStringExtractor::() : m_packet(), m_index(0) {} |
| 25 | |
| 26 | StdStringExtractor::(const char *packet_cstr) |
| 27 | : m_packet(), m_index(0) { |
| 28 | if (packet_cstr) |
| 29 | m_packet.assign(s: packet_cstr); |
| 30 | } |
| 31 | |
| 32 | // Destructor |
| 33 | StdStringExtractor::() = default; |
| 34 | |
| 35 | char StdStringExtractor::(char fail_value) { |
| 36 | if (m_index < m_packet.size()) { |
| 37 | char ch = m_packet[m_index]; |
| 38 | ++m_index; |
| 39 | return ch; |
| 40 | } |
| 41 | m_index = UINT64_MAX; |
| 42 | return fail_value; |
| 43 | } |
| 44 | |
| 45 | // If a pair of valid hex digits exist at the head of the |
| 46 | // StdStringExtractor they are decoded into an unsigned byte and returned |
| 47 | // by this function |
| 48 | // |
| 49 | // If there is not a pair of valid hex digits at the head of the |
| 50 | // StdStringExtractor, it is left unchanged and -1 is returned |
| 51 | int StdStringExtractor::() { |
| 52 | SkipSpaces(); |
| 53 | if (GetBytesLeft() < 2) { |
| 54 | return -1; |
| 55 | } |
| 56 | const int hi_nibble = xdigit_to_sint(ch: m_packet[m_index]); |
| 57 | const int lo_nibble = xdigit_to_sint(ch: m_packet[m_index + 1]); |
| 58 | if (hi_nibble == -1 || lo_nibble == -1) { |
| 59 | return -1; |
| 60 | } |
| 61 | m_index += 2; |
| 62 | return (uint8_t)((hi_nibble << 4) + lo_nibble); |
| 63 | } |
| 64 | |
| 65 | // Extract an unsigned character from two hex ASCII chars in the packet |
| 66 | // string, or return fail_value on failure |
| 67 | uint8_t StdStringExtractor::(uint8_t fail_value, bool set_eof_on_fail) { |
| 68 | // On success, fail_value will be overwritten with the next |
| 69 | // character in the stream |
| 70 | GetHexU8Ex(ch&: fail_value, set_eof_on_fail); |
| 71 | return fail_value; |
| 72 | } |
| 73 | |
| 74 | bool StdStringExtractor::(uint8_t &ch, bool set_eof_on_fail) { |
| 75 | int byte = DecodeHexU8(); |
| 76 | if (byte == -1) { |
| 77 | if (set_eof_on_fail || m_index >= m_packet.size()) |
| 78 | m_index = UINT64_MAX; |
| 79 | // ch should not be changed in case of failure |
| 80 | return false; |
| 81 | } |
| 82 | ch = (uint8_t)byte; |
| 83 | return true; |
| 84 | } |
| 85 | |
| 86 | uint32_t StdStringExtractor::(uint32_t fail_value, int base) { |
| 87 | if (m_index < m_packet.size()) { |
| 88 | char *end = nullptr; |
| 89 | const char *start = m_packet.c_str(); |
| 90 | const char *cstr = start + m_index; |
| 91 | uint32_t result = static_cast<uint32_t>(::strtoul(nptr: cstr, endptr: &end, base: base)); |
| 92 | |
| 93 | if (end && end != cstr) { |
| 94 | m_index = end - start; |
| 95 | return result; |
| 96 | } |
| 97 | } |
| 98 | return fail_value; |
| 99 | } |
| 100 | |
| 101 | int32_t StdStringExtractor::(int32_t fail_value, int base) { |
| 102 | if (m_index < m_packet.size()) { |
| 103 | char *end = nullptr; |
| 104 | const char *start = m_packet.c_str(); |
| 105 | const char *cstr = start + m_index; |
| 106 | int32_t result = static_cast<int32_t>(::strtol(nptr: cstr, endptr: &end, base: base)); |
| 107 | |
| 108 | if (end && end != cstr) { |
| 109 | m_index = end - start; |
| 110 | return result; |
| 111 | } |
| 112 | } |
| 113 | return fail_value; |
| 114 | } |
| 115 | |
| 116 | uint64_t StdStringExtractor::(uint64_t fail_value, int base) { |
| 117 | if (m_index < m_packet.size()) { |
| 118 | char *end = nullptr; |
| 119 | const char *start = m_packet.c_str(); |
| 120 | const char *cstr = start + m_index; |
| 121 | uint64_t result = ::strtoull(nptr: cstr, endptr: &end, base: base); |
| 122 | |
| 123 | if (end && end != cstr) { |
| 124 | m_index = end - start; |
| 125 | return result; |
| 126 | } |
| 127 | } |
| 128 | return fail_value; |
| 129 | } |
| 130 | |
| 131 | int64_t StdStringExtractor::(int64_t fail_value, int base) { |
| 132 | if (m_index < m_packet.size()) { |
| 133 | char *end = nullptr; |
| 134 | const char *start = m_packet.c_str(); |
| 135 | const char *cstr = start + m_index; |
| 136 | int64_t result = ::strtoll(nptr: cstr, endptr: &end, base: base); |
| 137 | |
| 138 | if (end && end != cstr) { |
| 139 | m_index = end - start; |
| 140 | return result; |
| 141 | } |
| 142 | } |
| 143 | return fail_value; |
| 144 | } |
| 145 | |
| 146 | uint32_t StdStringExtractor::(bool little_endian, |
| 147 | uint32_t fail_value) { |
| 148 | uint32_t result = 0; |
| 149 | uint32_t nibble_count = 0; |
| 150 | |
| 151 | SkipSpaces(); |
| 152 | if (little_endian) { |
| 153 | uint32_t shift_amount = 0; |
| 154 | while (m_index < m_packet.size() && ::isxdigit(m_packet[m_index])) { |
| 155 | // Make sure we don't exceed the size of a uint32_t... |
| 156 | if (nibble_count >= (sizeof(uint32_t) * 2)) { |
| 157 | m_index = UINT64_MAX; |
| 158 | return fail_value; |
| 159 | } |
| 160 | |
| 161 | uint8_t nibble_lo; |
| 162 | uint8_t nibble_hi = xdigit_to_sint(ch: m_packet[m_index]); |
| 163 | ++m_index; |
| 164 | if (m_index < m_packet.size() && ::isxdigit(m_packet[m_index])) { |
| 165 | nibble_lo = xdigit_to_sint(ch: m_packet[m_index]); |
| 166 | ++m_index; |
| 167 | result |= ((uint32_t)nibble_hi << (shift_amount + 4)); |
| 168 | result |= ((uint32_t)nibble_lo << shift_amount); |
| 169 | nibble_count += 2; |
| 170 | shift_amount += 8; |
| 171 | } else { |
| 172 | result |= ((uint32_t)nibble_hi << shift_amount); |
| 173 | nibble_count += 1; |
| 174 | shift_amount += 4; |
| 175 | } |
| 176 | } |
| 177 | } else { |
| 178 | while (m_index < m_packet.size() && ::isxdigit(m_packet[m_index])) { |
| 179 | // Make sure we don't exceed the size of a uint32_t... |
| 180 | if (nibble_count >= (sizeof(uint32_t) * 2)) { |
| 181 | m_index = UINT64_MAX; |
| 182 | return fail_value; |
| 183 | } |
| 184 | |
| 185 | uint8_t nibble = xdigit_to_sint(ch: m_packet[m_index]); |
| 186 | // Big Endian |
| 187 | result <<= 4; |
| 188 | result |= nibble; |
| 189 | |
| 190 | ++m_index; |
| 191 | ++nibble_count; |
| 192 | } |
| 193 | } |
| 194 | return result; |
| 195 | } |
| 196 | |
| 197 | uint64_t StdStringExtractor::(bool little_endian, |
| 198 | uint64_t fail_value) { |
| 199 | uint64_t result = 0; |
| 200 | uint32_t nibble_count = 0; |
| 201 | |
| 202 | SkipSpaces(); |
| 203 | if (little_endian) { |
| 204 | uint32_t shift_amount = 0; |
| 205 | while (m_index < m_packet.size() && ::isxdigit(m_packet[m_index])) { |
| 206 | // Make sure we don't exceed the size of a uint64_t... |
| 207 | if (nibble_count >= (sizeof(uint64_t) * 2)) { |
| 208 | m_index = UINT64_MAX; |
| 209 | return fail_value; |
| 210 | } |
| 211 | |
| 212 | uint8_t nibble_lo; |
| 213 | uint8_t nibble_hi = xdigit_to_sint(ch: m_packet[m_index]); |
| 214 | ++m_index; |
| 215 | if (m_index < m_packet.size() && ::isxdigit(m_packet[m_index])) { |
| 216 | nibble_lo = xdigit_to_sint(ch: m_packet[m_index]); |
| 217 | ++m_index; |
| 218 | result |= ((uint64_t)nibble_hi << (shift_amount + 4)); |
| 219 | result |= ((uint64_t)nibble_lo << shift_amount); |
| 220 | nibble_count += 2; |
| 221 | shift_amount += 8; |
| 222 | } else { |
| 223 | result |= ((uint64_t)nibble_hi << shift_amount); |
| 224 | nibble_count += 1; |
| 225 | shift_amount += 4; |
| 226 | } |
| 227 | } |
| 228 | } else { |
| 229 | while (m_index < m_packet.size() && ::isxdigit(m_packet[m_index])) { |
| 230 | // Make sure we don't exceed the size of a uint64_t... |
| 231 | if (nibble_count >= (sizeof(uint64_t) * 2)) { |
| 232 | m_index = UINT64_MAX; |
| 233 | return fail_value; |
| 234 | } |
| 235 | |
| 236 | uint8_t nibble = xdigit_to_sint(ch: m_packet[m_index]); |
| 237 | // Big Endian |
| 238 | result <<= 4; |
| 239 | result |= nibble; |
| 240 | |
| 241 | ++m_index; |
| 242 | ++nibble_count; |
| 243 | } |
| 244 | } |
| 245 | return result; |
| 246 | } |
| 247 | |
| 248 | size_t StdStringExtractor::(void *dst_void, size_t dst_len, |
| 249 | uint8_t fail_fill_value) { |
| 250 | uint8_t *dst = (uint8_t *)dst_void; |
| 251 | size_t = 0; |
| 252 | while (bytes_extracted < dst_len && GetBytesLeft()) { |
| 253 | dst[bytes_extracted] = GetHexU8(fail_value: fail_fill_value); |
| 254 | if (IsGood()) |
| 255 | ++bytes_extracted; |
| 256 | else |
| 257 | break; |
| 258 | } |
| 259 | |
| 260 | for (size_t i = bytes_extracted; i < dst_len; ++i) |
| 261 | dst[i] = fail_fill_value; |
| 262 | |
| 263 | return bytes_extracted; |
| 264 | } |
| 265 | |
| 266 | // Decodes all valid hex encoded bytes at the head of the |
| 267 | // StdStringExtractor, limited by dst_len. |
| 268 | // |
| 269 | // Returns the number of bytes successfully decoded |
| 270 | size_t StdStringExtractor::(void *dst_void, size_t dst_len) { |
| 271 | uint8_t *dst = (uint8_t *)dst_void; |
| 272 | size_t = 0; |
| 273 | while (bytes_extracted < dst_len) { |
| 274 | int decode = DecodeHexU8(); |
| 275 | if (decode == -1) { |
| 276 | break; |
| 277 | } |
| 278 | dst[bytes_extracted++] = (uint8_t)decode; |
| 279 | } |
| 280 | return bytes_extracted; |
| 281 | } |
| 282 | |
| 283 | size_t StdStringExtractor::(std::string &str) { |
| 284 | str.clear(); |
| 285 | str.reserve(res: GetBytesLeft() / 2); |
| 286 | char ch; |
| 287 | while ((ch = GetHexU8()) != '\0') |
| 288 | str.append(n: 1, c: ch); |
| 289 | return str.size(); |
| 290 | } |
| 291 | |
| 292 | size_t StdStringExtractor::(std::string &str, |
| 293 | uint32_t nibble_length) { |
| 294 | str.clear(); |
| 295 | |
| 296 | uint32_t nibble_count = 0; |
| 297 | for (const char *pch = Peek(); |
| 298 | (nibble_count < nibble_length) && (pch != nullptr); |
| 299 | str.append(n: 1, c: GetHexU8(fail_value: 0, set_eof_on_fail: false)), pch = Peek(), nibble_count += 2) { |
| 300 | } |
| 301 | |
| 302 | return str.size(); |
| 303 | } |
| 304 | |
| 305 | size_t StdStringExtractor::(std::string &str, |
| 306 | char terminator) { |
| 307 | str.clear(); |
| 308 | char ch; |
| 309 | while ((ch = GetHexU8(fail_value: 0, set_eof_on_fail: false)) != '\0') |
| 310 | str.append(n: 1, c: ch); |
| 311 | if (Peek() && *Peek() == terminator) |
| 312 | return str.size(); |
| 313 | |
| 314 | str.clear(); |
| 315 | return str.size(); |
| 316 | } |
| 317 | |
| 318 | bool StdStringExtractor::(std::string &name, |
| 319 | std::string &value) { |
| 320 | // Read something in the form of NNNN:VVVV; where NNNN is any character |
| 321 | // that is not a colon, followed by a ':' character, then a value (one or |
| 322 | // more ';' chars), followed by a ';' |
| 323 | if (m_index < m_packet.size()) { |
| 324 | const size_t colon_idx = m_packet.find(c: ':', pos: m_index); |
| 325 | if (colon_idx != std::string::npos) { |
| 326 | const size_t semicolon_idx = m_packet.find(c: ';', pos: colon_idx); |
| 327 | if (semicolon_idx != std::string::npos) { |
| 328 | name.assign(str: m_packet, pos: m_index, n: colon_idx - m_index); |
| 329 | value.assign(str: m_packet, pos: colon_idx + 1, n: semicolon_idx - (colon_idx + 1)); |
| 330 | m_index = semicolon_idx + 1; |
| 331 | return true; |
| 332 | } |
| 333 | } |
| 334 | } |
| 335 | m_index = UINT64_MAX; |
| 336 | return false; |
| 337 | } |
| 338 | |
| 339 | void StdStringExtractor::() { |
| 340 | const size_t n = m_packet.size(); |
| 341 | while (m_index < n && isspace(m_packet[m_index])) |
| 342 | ++m_index; |
| 343 | } |
| 344 | |