[MIPS GlobalISel] Select MSA vector generic and builtin add
[llvm-complete.git] / lib / Support / StringRef.cpp
blob4bafc4ec7181920e6df1aa0ed59bec15caa7e3ff
1 //===-- StringRef.cpp - Lightweight String References ---------------------===//
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 //===----------------------------------------------------------------------===//
9 #include "llvm/ADT/StringRef.h"
10 #include "llvm/ADT/APFloat.h"
11 #include "llvm/ADT/APInt.h"
12 #include "llvm/ADT/Hashing.h"
13 #include "llvm/ADT/StringExtras.h"
14 #include "llvm/ADT/edit_distance.h"
15 #include <bitset>
17 using namespace llvm;
19 // MSVC emits references to this into the translation units which reference it.
20 #ifndef _MSC_VER
21 const size_t StringRef::npos;
22 #endif
24 // strncasecmp() is not available on non-POSIX systems, so define an
25 // alternative function here.
26 static int ascii_strncasecmp(const char *LHS, const char *RHS, size_t Length) {
27 for (size_t I = 0; I < Length; ++I) {
28 unsigned char LHC = toLower(LHS[I]);
29 unsigned char RHC = toLower(RHS[I]);
30 if (LHC != RHC)
31 return LHC < RHC ? -1 : 1;
33 return 0;
36 /// compare_lower - Compare strings, ignoring case.
37 int StringRef::compare_lower(StringRef RHS) const {
38 if (int Res = ascii_strncasecmp(Data, RHS.Data, std::min(Length, RHS.Length)))
39 return Res;
40 if (Length == RHS.Length)
41 return 0;
42 return Length < RHS.Length ? -1 : 1;
45 /// Check if this string starts with the given \p Prefix, ignoring case.
46 bool StringRef::startswith_lower(StringRef Prefix) const {
47 return Length >= Prefix.Length &&
48 ascii_strncasecmp(Data, Prefix.Data, Prefix.Length) == 0;
51 /// Check if this string ends with the given \p Suffix, ignoring case.
52 bool StringRef::endswith_lower(StringRef Suffix) const {
53 return Length >= Suffix.Length &&
54 ascii_strncasecmp(end() - Suffix.Length, Suffix.Data, Suffix.Length) == 0;
57 size_t StringRef::find_lower(char C, size_t From) const {
58 char L = toLower(C);
59 return find_if([L](char D) { return toLower(D) == L; }, From);
62 /// compare_numeric - Compare strings, handle embedded numbers.
63 int StringRef::compare_numeric(StringRef RHS) const {
64 for (size_t I = 0, E = std::min(Length, RHS.Length); I != E; ++I) {
65 // Check for sequences of digits.
66 if (isDigit(Data[I]) && isDigit(RHS.Data[I])) {
67 // The longer sequence of numbers is considered larger.
68 // This doesn't really handle prefixed zeros well.
69 size_t J;
70 for (J = I + 1; J != E + 1; ++J) {
71 bool ld = J < Length && isDigit(Data[J]);
72 bool rd = J < RHS.Length && isDigit(RHS.Data[J]);
73 if (ld != rd)
74 return rd ? -1 : 1;
75 if (!rd)
76 break;
78 // The two number sequences have the same length (J-I), just memcmp them.
79 if (int Res = compareMemory(Data + I, RHS.Data + I, J - I))
80 return Res < 0 ? -1 : 1;
81 // Identical number sequences, continue search after the numbers.
82 I = J - 1;
83 continue;
85 if (Data[I] != RHS.Data[I])
86 return (unsigned char)Data[I] < (unsigned char)RHS.Data[I] ? -1 : 1;
88 if (Length == RHS.Length)
89 return 0;
90 return Length < RHS.Length ? -1 : 1;
93 // Compute the edit distance between the two given strings.
94 unsigned StringRef::edit_distance(llvm::StringRef Other,
95 bool AllowReplacements,
96 unsigned MaxEditDistance) const {
97 return llvm::ComputeEditDistance(
98 makeArrayRef(data(), size()),
99 makeArrayRef(Other.data(), Other.size()),
100 AllowReplacements, MaxEditDistance);
103 //===----------------------------------------------------------------------===//
104 // String Operations
105 //===----------------------------------------------------------------------===//
107 std::string StringRef::lower() const {
108 std::string Result(size(), char());
109 for (size_type i = 0, e = size(); i != e; ++i) {
110 Result[i] = toLower(Data[i]);
112 return Result;
115 std::string StringRef::upper() const {
116 std::string Result(size(), char());
117 for (size_type i = 0, e = size(); i != e; ++i) {
118 Result[i] = toUpper(Data[i]);
120 return Result;
123 //===----------------------------------------------------------------------===//
124 // String Searching
125 //===----------------------------------------------------------------------===//
128 /// find - Search for the first string \arg Str in the string.
130 /// \return - The index of the first occurrence of \arg Str, or npos if not
131 /// found.
132 size_t StringRef::find(StringRef Str, size_t From) const {
133 if (From > Length)
134 return npos;
136 const char *Start = Data + From;
137 size_t Size = Length - From;
139 const char *Needle = Str.data();
140 size_t N = Str.size();
141 if (N == 0)
142 return From;
143 if (Size < N)
144 return npos;
145 if (N == 1) {
146 const char *Ptr = (const char *)::memchr(Start, Needle[0], Size);
147 return Ptr == nullptr ? npos : Ptr - Data;
150 const char *Stop = Start + (Size - N + 1);
152 // For short haystacks or unsupported needles fall back to the naive algorithm
153 if (Size < 16 || N > 255) {
154 do {
155 if (std::memcmp(Start, Needle, N) == 0)
156 return Start - Data;
157 ++Start;
158 } while (Start < Stop);
159 return npos;
162 // Build the bad char heuristic table, with uint8_t to reduce cache thrashing.
163 uint8_t BadCharSkip[256];
164 std::memset(BadCharSkip, N, 256);
165 for (unsigned i = 0; i != N-1; ++i)
166 BadCharSkip[(uint8_t)Str[i]] = N-1-i;
168 do {
169 uint8_t Last = Start[N - 1];
170 if (LLVM_UNLIKELY(Last == (uint8_t)Needle[N - 1]))
171 if (std::memcmp(Start, Needle, N - 1) == 0)
172 return Start - Data;
174 // Otherwise skip the appropriate number of bytes.
175 Start += BadCharSkip[Last];
176 } while (Start < Stop);
178 return npos;
181 size_t StringRef::find_lower(StringRef Str, size_t From) const {
182 StringRef This = substr(From);
183 while (This.size() >= Str.size()) {
184 if (This.startswith_lower(Str))
185 return From;
186 This = This.drop_front();
187 ++From;
189 return npos;
192 size_t StringRef::rfind_lower(char C, size_t From) const {
193 From = std::min(From, Length);
194 size_t i = From;
195 while (i != 0) {
196 --i;
197 if (toLower(Data[i]) == toLower(C))
198 return i;
200 return npos;
203 /// rfind - Search for the last string \arg Str in the string.
205 /// \return - The index of the last occurrence of \arg Str, or npos if not
206 /// found.
207 size_t StringRef::rfind(StringRef Str) const {
208 size_t N = Str.size();
209 if (N > Length)
210 return npos;
211 for (size_t i = Length - N + 1, e = 0; i != e;) {
212 --i;
213 if (substr(i, N).equals(Str))
214 return i;
216 return npos;
219 size_t StringRef::rfind_lower(StringRef Str) const {
220 size_t N = Str.size();
221 if (N > Length)
222 return npos;
223 for (size_t i = Length - N + 1, e = 0; i != e;) {
224 --i;
225 if (substr(i, N).equals_lower(Str))
226 return i;
228 return npos;
231 /// find_first_of - Find the first character in the string that is in \arg
232 /// Chars, or npos if not found.
234 /// Note: O(size() + Chars.size())
235 StringRef::size_type StringRef::find_first_of(StringRef Chars,
236 size_t From) const {
237 std::bitset<1 << CHAR_BIT> CharBits;
238 for (size_type i = 0; i != Chars.size(); ++i)
239 CharBits.set((unsigned char)Chars[i]);
241 for (size_type i = std::min(From, Length), e = Length; i != e; ++i)
242 if (CharBits.test((unsigned char)Data[i]))
243 return i;
244 return npos;
247 /// find_first_not_of - Find the first character in the string that is not
248 /// \arg C or npos if not found.
249 StringRef::size_type StringRef::find_first_not_of(char C, size_t From) const {
250 for (size_type i = std::min(From, Length), e = Length; i != e; ++i)
251 if (Data[i] != C)
252 return i;
253 return npos;
256 /// find_first_not_of - Find the first character in the string that is not
257 /// in the string \arg Chars, or npos if not found.
259 /// Note: O(size() + Chars.size())
260 StringRef::size_type StringRef::find_first_not_of(StringRef Chars,
261 size_t From) const {
262 std::bitset<1 << CHAR_BIT> CharBits;
263 for (size_type i = 0; i != Chars.size(); ++i)
264 CharBits.set((unsigned char)Chars[i]);
266 for (size_type i = std::min(From, Length), e = Length; i != e; ++i)
267 if (!CharBits.test((unsigned char)Data[i]))
268 return i;
269 return npos;
272 /// find_last_of - Find the last character in the string that is in \arg C,
273 /// or npos if not found.
275 /// Note: O(size() + Chars.size())
276 StringRef::size_type StringRef::find_last_of(StringRef Chars,
277 size_t From) const {
278 std::bitset<1 << CHAR_BIT> CharBits;
279 for (size_type i = 0; i != Chars.size(); ++i)
280 CharBits.set((unsigned char)Chars[i]);
282 for (size_type i = std::min(From, Length) - 1, e = -1; i != e; --i)
283 if (CharBits.test((unsigned char)Data[i]))
284 return i;
285 return npos;
288 /// find_last_not_of - Find the last character in the string that is not
289 /// \arg C, or npos if not found.
290 StringRef::size_type StringRef::find_last_not_of(char C, size_t From) const {
291 for (size_type i = std::min(From, Length) - 1, e = -1; i != e; --i)
292 if (Data[i] != C)
293 return i;
294 return npos;
297 /// find_last_not_of - Find the last character in the string that is not in
298 /// \arg Chars, or npos if not found.
300 /// Note: O(size() + Chars.size())
301 StringRef::size_type StringRef::find_last_not_of(StringRef Chars,
302 size_t From) const {
303 std::bitset<1 << CHAR_BIT> CharBits;
304 for (size_type i = 0, e = Chars.size(); i != e; ++i)
305 CharBits.set((unsigned char)Chars[i]);
307 for (size_type i = std::min(From, Length) - 1, e = -1; i != e; --i)
308 if (!CharBits.test((unsigned char)Data[i]))
309 return i;
310 return npos;
313 void StringRef::split(SmallVectorImpl<StringRef> &A,
314 StringRef Separator, int MaxSplit,
315 bool KeepEmpty) const {
316 StringRef S = *this;
318 // Count down from MaxSplit. When MaxSplit is -1, this will just split
319 // "forever". This doesn't support splitting more than 2^31 times
320 // intentionally; if we ever want that we can make MaxSplit a 64-bit integer
321 // but that seems unlikely to be useful.
322 while (MaxSplit-- != 0) {
323 size_t Idx = S.find(Separator);
324 if (Idx == npos)
325 break;
327 // Push this split.
328 if (KeepEmpty || Idx > 0)
329 A.push_back(S.slice(0, Idx));
331 // Jump forward.
332 S = S.slice(Idx + Separator.size(), npos);
335 // Push the tail.
336 if (KeepEmpty || !S.empty())
337 A.push_back(S);
340 void StringRef::split(SmallVectorImpl<StringRef> &A, char Separator,
341 int MaxSplit, bool KeepEmpty) const {
342 StringRef S = *this;
344 // Count down from MaxSplit. When MaxSplit is -1, this will just split
345 // "forever". This doesn't support splitting more than 2^31 times
346 // intentionally; if we ever want that we can make MaxSplit a 64-bit integer
347 // but that seems unlikely to be useful.
348 while (MaxSplit-- != 0) {
349 size_t Idx = S.find(Separator);
350 if (Idx == npos)
351 break;
353 // Push this split.
354 if (KeepEmpty || Idx > 0)
355 A.push_back(S.slice(0, Idx));
357 // Jump forward.
358 S = S.slice(Idx + 1, npos);
361 // Push the tail.
362 if (KeepEmpty || !S.empty())
363 A.push_back(S);
366 //===----------------------------------------------------------------------===//
367 // Helpful Algorithms
368 //===----------------------------------------------------------------------===//
370 /// count - Return the number of non-overlapped occurrences of \arg Str in
371 /// the string.
372 size_t StringRef::count(StringRef Str) const {
373 size_t Count = 0;
374 size_t N = Str.size();
375 if (N > Length)
376 return 0;
377 for (size_t i = 0, e = Length - N + 1; i != e; ++i)
378 if (substr(i, N).equals(Str))
379 ++Count;
380 return Count;
383 static unsigned GetAutoSenseRadix(StringRef &Str) {
384 if (Str.empty())
385 return 10;
387 if (Str.startswith("0x") || Str.startswith("0X")) {
388 Str = Str.substr(2);
389 return 16;
392 if (Str.startswith("0b") || Str.startswith("0B")) {
393 Str = Str.substr(2);
394 return 2;
397 if (Str.startswith("0o")) {
398 Str = Str.substr(2);
399 return 8;
402 if (Str[0] == '0' && Str.size() > 1 && isDigit(Str[1])) {
403 Str = Str.substr(1);
404 return 8;
407 return 10;
410 bool llvm::consumeUnsignedInteger(StringRef &Str, unsigned Radix,
411 unsigned long long &Result) {
412 // Autosense radix if not specified.
413 if (Radix == 0)
414 Radix = GetAutoSenseRadix(Str);
416 // Empty strings (after the radix autosense) are invalid.
417 if (Str.empty()) return true;
419 // Parse all the bytes of the string given this radix. Watch for overflow.
420 StringRef Str2 = Str;
421 Result = 0;
422 while (!Str2.empty()) {
423 unsigned CharVal;
424 if (Str2[0] >= '0' && Str2[0] <= '9')
425 CharVal = Str2[0] - '0';
426 else if (Str2[0] >= 'a' && Str2[0] <= 'z')
427 CharVal = Str2[0] - 'a' + 10;
428 else if (Str2[0] >= 'A' && Str2[0] <= 'Z')
429 CharVal = Str2[0] - 'A' + 10;
430 else
431 break;
433 // If the parsed value is larger than the integer radix, we cannot
434 // consume any more characters.
435 if (CharVal >= Radix)
436 break;
438 // Add in this character.
439 unsigned long long PrevResult = Result;
440 Result = Result * Radix + CharVal;
442 // Check for overflow by shifting back and seeing if bits were lost.
443 if (Result / Radix < PrevResult)
444 return true;
446 Str2 = Str2.substr(1);
449 // We consider the operation a failure if no characters were consumed
450 // successfully.
451 if (Str.size() == Str2.size())
452 return true;
454 Str = Str2;
455 return false;
458 bool llvm::consumeSignedInteger(StringRef &Str, unsigned Radix,
459 long long &Result) {
460 unsigned long long ULLVal;
462 // Handle positive strings first.
463 if (Str.empty() || Str.front() != '-') {
464 if (consumeUnsignedInteger(Str, Radix, ULLVal) ||
465 // Check for value so large it overflows a signed value.
466 (long long)ULLVal < 0)
467 return true;
468 Result = ULLVal;
469 return false;
472 // Get the positive part of the value.
473 StringRef Str2 = Str.drop_front(1);
474 if (consumeUnsignedInteger(Str2, Radix, ULLVal) ||
475 // Reject values so large they'd overflow as negative signed, but allow
476 // "-0". This negates the unsigned so that the negative isn't undefined
477 // on signed overflow.
478 (long long)-ULLVal > 0)
479 return true;
481 Str = Str2;
482 Result = -ULLVal;
483 return false;
486 /// GetAsUnsignedInteger - Workhorse method that converts a integer character
487 /// sequence of radix up to 36 to an unsigned long long value.
488 bool llvm::getAsUnsignedInteger(StringRef Str, unsigned Radix,
489 unsigned long long &Result) {
490 if (consumeUnsignedInteger(Str, Radix, Result))
491 return true;
493 // For getAsUnsignedInteger, we require the whole string to be consumed or
494 // else we consider it a failure.
495 return !Str.empty();
498 bool llvm::getAsSignedInteger(StringRef Str, unsigned Radix,
499 long long &Result) {
500 if (consumeSignedInteger(Str, Radix, Result))
501 return true;
503 // For getAsSignedInteger, we require the whole string to be consumed or else
504 // we consider it a failure.
505 return !Str.empty();
508 bool StringRef::getAsInteger(unsigned Radix, APInt &Result) const {
509 StringRef Str = *this;
511 // Autosense radix if not specified.
512 if (Radix == 0)
513 Radix = GetAutoSenseRadix(Str);
515 assert(Radix > 1 && Radix <= 36);
517 // Empty strings (after the radix autosense) are invalid.
518 if (Str.empty()) return true;
520 // Skip leading zeroes. This can be a significant improvement if
521 // it means we don't need > 64 bits.
522 while (!Str.empty() && Str.front() == '0')
523 Str = Str.substr(1);
525 // If it was nothing but zeroes....
526 if (Str.empty()) {
527 Result = APInt(64, 0);
528 return false;
531 // (Over-)estimate the required number of bits.
532 unsigned Log2Radix = 0;
533 while ((1U << Log2Radix) < Radix) Log2Radix++;
534 bool IsPowerOf2Radix = ((1U << Log2Radix) == Radix);
536 unsigned BitWidth = Log2Radix * Str.size();
537 if (BitWidth < Result.getBitWidth())
538 BitWidth = Result.getBitWidth(); // don't shrink the result
539 else if (BitWidth > Result.getBitWidth())
540 Result = Result.zext(BitWidth);
542 APInt RadixAP, CharAP; // unused unless !IsPowerOf2Radix
543 if (!IsPowerOf2Radix) {
544 // These must have the same bit-width as Result.
545 RadixAP = APInt(BitWidth, Radix);
546 CharAP = APInt(BitWidth, 0);
549 // Parse all the bytes of the string given this radix.
550 Result = 0;
551 while (!Str.empty()) {
552 unsigned CharVal;
553 if (Str[0] >= '0' && Str[0] <= '9')
554 CharVal = Str[0]-'0';
555 else if (Str[0] >= 'a' && Str[0] <= 'z')
556 CharVal = Str[0]-'a'+10;
557 else if (Str[0] >= 'A' && Str[0] <= 'Z')
558 CharVal = Str[0]-'A'+10;
559 else
560 return true;
562 // If the parsed value is larger than the integer radix, the string is
563 // invalid.
564 if (CharVal >= Radix)
565 return true;
567 // Add in this character.
568 if (IsPowerOf2Radix) {
569 Result <<= Log2Radix;
570 Result |= CharVal;
571 } else {
572 Result *= RadixAP;
573 CharAP = CharVal;
574 Result += CharAP;
577 Str = Str.substr(1);
580 return false;
583 bool StringRef::getAsDouble(double &Result, bool AllowInexact) const {
584 APFloat F(0.0);
585 APFloat::opStatus Status =
586 F.convertFromString(*this, APFloat::rmNearestTiesToEven);
587 if (Status != APFloat::opOK) {
588 if (!AllowInexact || !(Status & APFloat::opInexact))
589 return true;
592 Result = F.convertToDouble();
593 return false;
596 // Implementation of StringRef hashing.
597 hash_code llvm::hash_value(StringRef S) {
598 return hash_combine_range(S.begin(), S.end());