1 | //===-- StringExtractor.cpp -----------------------------------------------===// |
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 "lldb/Utility/StringExtractor.h" |
10 | #include "llvm/ADT/StringExtras.h" |
11 | |
12 | #include <tuple> |
13 | |
14 | #include <cctype> |
15 | #include <cstdlib> |
16 | #include <cstring> |
17 | |
18 | static inline int xdigit_to_sint(char ch) { |
19 | if (ch >= 'a' && ch <= 'f') |
20 | return 10 + ch - 'a'; |
21 | if (ch >= 'A' && ch <= 'F') |
22 | return 10 + ch - 'A'; |
23 | if (ch >= '0' && ch <= '9') |
24 | return ch - '0'; |
25 | return -1; |
26 | } |
27 | |
28 | // StringExtractor constructor |
29 | StringExtractor::() : m_packet() {} |
30 | |
31 | StringExtractor::(llvm::StringRef packet_str) : m_packet() { |
32 | m_packet.assign(first: packet_str.begin(), last: packet_str.end()); |
33 | } |
34 | |
35 | StringExtractor::(const char *packet_cstr) : m_packet() { |
36 | if (packet_cstr) |
37 | m_packet.assign(s: packet_cstr); |
38 | } |
39 | |
40 | // Destructor |
41 | StringExtractor::() = default; |
42 | |
43 | char StringExtractor::(char fail_value) { |
44 | if (m_index < m_packet.size()) { |
45 | char ch = m_packet[m_index]; |
46 | ++m_index; |
47 | return ch; |
48 | } |
49 | m_index = UINT64_MAX; |
50 | return fail_value; |
51 | } |
52 | |
53 | // If a pair of valid hex digits exist at the head of the StringExtractor they |
54 | // are decoded into an unsigned byte and returned by this function |
55 | // |
56 | // If there is not a pair of valid hex digits at the head of the |
57 | // StringExtractor, it is left unchanged and -1 is returned |
58 | int StringExtractor::() { |
59 | SkipSpaces(); |
60 | if (GetBytesLeft() < 2) { |
61 | return -1; |
62 | } |
63 | const int hi_nibble = xdigit_to_sint(ch: m_packet[m_index]); |
64 | const int lo_nibble = xdigit_to_sint(ch: m_packet[m_index + 1]); |
65 | if (hi_nibble == -1 || lo_nibble == -1) { |
66 | return -1; |
67 | } |
68 | m_index += 2; |
69 | return static_cast<uint8_t>((hi_nibble << 4) + lo_nibble); |
70 | } |
71 | |
72 | // Extract an unsigned character from two hex ASCII chars in the packet string, |
73 | // or return fail_value on failure |
74 | uint8_t StringExtractor::(uint8_t fail_value, bool set_eof_on_fail) { |
75 | // On success, fail_value will be overwritten with the next character in the |
76 | // stream |
77 | GetHexU8Ex(ch&: fail_value, set_eof_on_fail); |
78 | return fail_value; |
79 | } |
80 | |
81 | bool StringExtractor::(uint8_t &ch, bool set_eof_on_fail) { |
82 | int byte = DecodeHexU8(); |
83 | if (byte == -1) { |
84 | if (set_eof_on_fail || m_index >= m_packet.size()) |
85 | m_index = UINT64_MAX; |
86 | // ch should not be changed in case of failure |
87 | return false; |
88 | } |
89 | ch = static_cast<uint8_t>(byte); |
90 | return true; |
91 | } |
92 | |
93 | uint32_t StringExtractor::(uint32_t fail_value, int base) { |
94 | if (m_index < m_packet.size()) { |
95 | char *end = nullptr; |
96 | const char *start = m_packet.c_str(); |
97 | const char *cstr = start + m_index; |
98 | uint32_t result = static_cast<uint32_t>(::strtoul(nptr: cstr, endptr: &end, base: base)); |
99 | |
100 | if (end && end != cstr) { |
101 | m_index = end - start; |
102 | return result; |
103 | } |
104 | } |
105 | return fail_value; |
106 | } |
107 | |
108 | int32_t StringExtractor::(int32_t fail_value, int base) { |
109 | if (m_index < m_packet.size()) { |
110 | char *end = nullptr; |
111 | const char *start = m_packet.c_str(); |
112 | const char *cstr = start + m_index; |
113 | int32_t result = static_cast<int32_t>(::strtol(nptr: cstr, endptr: &end, base: base)); |
114 | |
115 | if (end && end != cstr) { |
116 | m_index = end - start; |
117 | return result; |
118 | } |
119 | } |
120 | return fail_value; |
121 | } |
122 | |
123 | uint64_t StringExtractor::(uint64_t fail_value, int base) { |
124 | if (m_index < m_packet.size()) { |
125 | char *end = nullptr; |
126 | const char *start = m_packet.c_str(); |
127 | const char *cstr = start + m_index; |
128 | uint64_t result = ::strtoull(nptr: cstr, endptr: &end, base: base); |
129 | |
130 | if (end && end != cstr) { |
131 | m_index = end - start; |
132 | return result; |
133 | } |
134 | } |
135 | return fail_value; |
136 | } |
137 | |
138 | int64_t StringExtractor::(int64_t fail_value, int base) { |
139 | if (m_index < m_packet.size()) { |
140 | char *end = nullptr; |
141 | const char *start = m_packet.c_str(); |
142 | const char *cstr = start + m_index; |
143 | int64_t result = ::strtoll(nptr: cstr, endptr: &end, base: base); |
144 | |
145 | if (end && end != cstr) { |
146 | m_index = end - start; |
147 | return result; |
148 | } |
149 | } |
150 | return fail_value; |
151 | } |
152 | |
153 | uint32_t StringExtractor::(bool little_endian, |
154 | uint32_t fail_value) { |
155 | uint32_t result = 0; |
156 | uint32_t nibble_count = 0; |
157 | |
158 | SkipSpaces(); |
159 | if (little_endian) { |
160 | uint32_t shift_amount = 0; |
161 | while (m_index < m_packet.size() && ::isxdigit(m_packet[m_index])) { |
162 | // Make sure we don't exceed the size of a uint32_t... |
163 | if (nibble_count >= (sizeof(uint32_t) * 2)) { |
164 | m_index = UINT64_MAX; |
165 | return fail_value; |
166 | } |
167 | |
168 | uint8_t nibble_lo; |
169 | uint8_t nibble_hi = xdigit_to_sint(ch: m_packet[m_index]); |
170 | ++m_index; |
171 | if (m_index < m_packet.size() && ::isxdigit(m_packet[m_index])) { |
172 | nibble_lo = xdigit_to_sint(ch: m_packet[m_index]); |
173 | ++m_index; |
174 | result |= (static_cast<uint32_t>(nibble_hi) << (shift_amount + 4)); |
175 | result |= (static_cast<uint32_t>(nibble_lo) << shift_amount); |
176 | nibble_count += 2; |
177 | shift_amount += 8; |
178 | } else { |
179 | result |= (static_cast<uint32_t>(nibble_hi) << shift_amount); |
180 | nibble_count += 1; |
181 | shift_amount += 4; |
182 | } |
183 | } |
184 | } else { |
185 | while (m_index < m_packet.size() && ::isxdigit(m_packet[m_index])) { |
186 | // Make sure we don't exceed the size of a uint32_t... |
187 | if (nibble_count >= (sizeof(uint32_t) * 2)) { |
188 | m_index = UINT64_MAX; |
189 | return fail_value; |
190 | } |
191 | |
192 | uint8_t nibble = xdigit_to_sint(ch: m_packet[m_index]); |
193 | // Big Endian |
194 | result <<= 4; |
195 | result |= nibble; |
196 | |
197 | ++m_index; |
198 | ++nibble_count; |
199 | } |
200 | } |
201 | return result; |
202 | } |
203 | |
204 | uint64_t StringExtractor::(bool little_endian, |
205 | uint64_t fail_value) { |
206 | uint64_t result = 0; |
207 | uint32_t nibble_count = 0; |
208 | |
209 | SkipSpaces(); |
210 | if (little_endian) { |
211 | uint32_t shift_amount = 0; |
212 | while (m_index < m_packet.size() && ::isxdigit(m_packet[m_index])) { |
213 | // Make sure we don't exceed the size of a uint64_t... |
214 | if (nibble_count >= (sizeof(uint64_t) * 2)) { |
215 | m_index = UINT64_MAX; |
216 | return fail_value; |
217 | } |
218 | |
219 | uint8_t nibble_lo; |
220 | uint8_t nibble_hi = xdigit_to_sint(ch: m_packet[m_index]); |
221 | ++m_index; |
222 | if (m_index < m_packet.size() && ::isxdigit(m_packet[m_index])) { |
223 | nibble_lo = xdigit_to_sint(ch: m_packet[m_index]); |
224 | ++m_index; |
225 | result |= (static_cast<uint64_t>(nibble_hi) << (shift_amount + 4)); |
226 | result |= (static_cast<uint64_t>(nibble_lo) << shift_amount); |
227 | nibble_count += 2; |
228 | shift_amount += 8; |
229 | } else { |
230 | result |= (static_cast<uint64_t>(nibble_hi) << shift_amount); |
231 | nibble_count += 1; |
232 | shift_amount += 4; |
233 | } |
234 | } |
235 | } else { |
236 | while (m_index < m_packet.size() && ::isxdigit(m_packet[m_index])) { |
237 | // Make sure we don't exceed the size of a uint64_t... |
238 | if (nibble_count >= (sizeof(uint64_t) * 2)) { |
239 | m_index = UINT64_MAX; |
240 | return fail_value; |
241 | } |
242 | |
243 | uint8_t nibble = xdigit_to_sint(ch: m_packet[m_index]); |
244 | // Big Endian |
245 | result <<= 4; |
246 | result |= nibble; |
247 | |
248 | ++m_index; |
249 | ++nibble_count; |
250 | } |
251 | } |
252 | return result; |
253 | } |
254 | |
255 | bool StringExtractor::(const llvm::StringRef &str) { |
256 | llvm::StringRef S = GetStringRef(); |
257 | if (!S.starts_with(Prefix: str)) |
258 | return false; |
259 | else |
260 | m_index += str.size(); |
261 | return true; |
262 | } |
263 | |
264 | size_t StringExtractor::(llvm::MutableArrayRef<uint8_t> dest, |
265 | uint8_t fail_fill_value) { |
266 | size_t = 0; |
267 | while (!dest.empty() && GetBytesLeft() > 0) { |
268 | dest[0] = GetHexU8(fail_value: fail_fill_value); |
269 | if (!IsGood()) |
270 | break; |
271 | ++bytes_extracted; |
272 | dest = dest.drop_front(); |
273 | } |
274 | |
275 | if (!dest.empty()) |
276 | ::memset(s: dest.data(), c: fail_fill_value, n: dest.size()); |
277 | |
278 | return bytes_extracted; |
279 | } |
280 | |
281 | // Decodes all valid hex encoded bytes at the head of the StringExtractor, |
282 | // limited by dst_len. |
283 | // |
284 | // Returns the number of bytes successfully decoded |
285 | size_t StringExtractor::(llvm::MutableArrayRef<uint8_t> dest) { |
286 | size_t = 0; |
287 | while (!dest.empty()) { |
288 | int decode = DecodeHexU8(); |
289 | if (decode == -1) |
290 | break; |
291 | dest[0] = static_cast<uint8_t>(decode); |
292 | dest = dest.drop_front(); |
293 | ++bytes_extracted; |
294 | } |
295 | return bytes_extracted; |
296 | } |
297 | |
298 | size_t StringExtractor::(std::string &str) { |
299 | str.clear(); |
300 | str.reserve(res: GetBytesLeft() / 2); |
301 | char ch; |
302 | while ((ch = GetHexU8()) != '\0') |
303 | str.append(n: 1, c: ch); |
304 | return str.size(); |
305 | } |
306 | |
307 | size_t StringExtractor::(std::string &str, |
308 | uint32_t nibble_length) { |
309 | str.clear(); |
310 | |
311 | uint32_t nibble_count = 0; |
312 | for (const char *pch = Peek(); |
313 | (nibble_count < nibble_length) && (pch != nullptr); |
314 | str.append(n: 1, c: GetHexU8(fail_value: 0, set_eof_on_fail: false)), pch = Peek(), nibble_count += 2) { |
315 | } |
316 | |
317 | return str.size(); |
318 | } |
319 | |
320 | size_t StringExtractor::(std::string &str, |
321 | char terminator) { |
322 | str.clear(); |
323 | char ch; |
324 | while ((ch = GetHexU8(fail_value: 0, set_eof_on_fail: false)) != '\0') |
325 | str.append(n: 1, c: ch); |
326 | if (Peek() && *Peek() == terminator) |
327 | return str.size(); |
328 | |
329 | str.clear(); |
330 | return str.size(); |
331 | } |
332 | |
333 | bool StringExtractor::(llvm::StringRef &name, |
334 | llvm::StringRef &value) { |
335 | // Read something in the form of NNNN:VVVV; where NNNN is any character that |
336 | // is not a colon, followed by a ':' character, then a value (one or more ';' |
337 | // chars), followed by a ';' |
338 | if (m_index >= m_packet.size()) |
339 | return fail(); |
340 | |
341 | llvm::StringRef view(m_packet); |
342 | if (view.empty()) |
343 | return fail(); |
344 | |
345 | llvm::StringRef a, b, c, d; |
346 | view = view.substr(Start: m_index); |
347 | std::tie(args&: a, args&: b) = view.split(Separator: ':'); |
348 | if (a.empty() || b.empty()) |
349 | return fail(); |
350 | std::tie(args&: c, args&: d) = b.split(Separator: ';'); |
351 | if (b == c && d.empty()) |
352 | return fail(); |
353 | |
354 | name = a; |
355 | value = c; |
356 | if (d.empty()) |
357 | m_index = m_packet.size(); |
358 | else { |
359 | size_t bytes_consumed = d.data() - view.data(); |
360 | m_index += bytes_consumed; |
361 | } |
362 | return true; |
363 | } |
364 | |
365 | void StringExtractor::() { |
366 | const size_t n = m_packet.size(); |
367 | while (m_index < n && llvm::isSpace(C: m_packet[m_index])) |
368 | ++m_index; |
369 | } |
370 | |