Details | Last modification | View Log | RSS feed
Rev | Author | Line No. | Line |
---|---|---|---|
14 | pmbaty | 1 | //===- StringRef.h - Constant String Reference Wrapper ----------*- 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 | #ifndef LLVM_ADT_STRINGREF_H |
||
10 | #define LLVM_ADT_STRINGREF_H |
||
11 | |||
12 | #include "llvm/ADT/DenseMapInfo.h" |
||
13 | #include "llvm/ADT/STLFunctionalExtras.h" |
||
14 | #include "llvm/ADT/iterator_range.h" |
||
15 | #include "llvm/Support/Compiler.h" |
||
16 | #include <algorithm> |
||
17 | #include <cassert> |
||
18 | #include <cstddef> |
||
19 | #include <cstring> |
||
20 | #include <limits> |
||
21 | #include <string> |
||
22 | #include <string_view> |
||
23 | #include <type_traits> |
||
24 | #include <utility> |
||
25 | |||
26 | namespace llvm { |
||
27 | |||
28 | class APInt; |
||
29 | class hash_code; |
||
30 | template <typename T> class SmallVectorImpl; |
||
31 | class StringRef; |
||
32 | |||
33 | /// Helper functions for StringRef::getAsInteger. |
||
34 | bool getAsUnsignedInteger(StringRef Str, unsigned Radix, |
||
35 | unsigned long long &Result); |
||
36 | |||
37 | bool getAsSignedInteger(StringRef Str, unsigned Radix, long long &Result); |
||
38 | |||
39 | bool consumeUnsignedInteger(StringRef &Str, unsigned Radix, |
||
40 | unsigned long long &Result); |
||
41 | bool consumeSignedInteger(StringRef &Str, unsigned Radix, long long &Result); |
||
42 | |||
43 | /// StringRef - Represent a constant reference to a string, i.e. a character |
||
44 | /// array and a length, which need not be null terminated. |
||
45 | /// |
||
46 | /// This class does not own the string data, it is expected to be used in |
||
47 | /// situations where the character data resides in some other buffer, whose |
||
48 | /// lifetime extends past that of the StringRef. For this reason, it is not in |
||
49 | /// general safe to store a StringRef. |
||
50 | class LLVM_GSL_POINTER StringRef { |
||
51 | public: |
||
52 | static constexpr size_t npos = ~size_t(0); |
||
53 | |||
54 | using iterator = const char *; |
||
55 | using const_iterator = const char *; |
||
56 | using size_type = size_t; |
||
57 | |||
58 | private: |
||
59 | /// The start of the string, in an external buffer. |
||
60 | const char *Data = nullptr; |
||
61 | |||
62 | /// The length of the string. |
||
63 | size_t Length = 0; |
||
64 | |||
65 | // Workaround memcmp issue with null pointers (undefined behavior) |
||
66 | // by providing a specialized version |
||
67 | static int compareMemory(const char *Lhs, const char *Rhs, size_t Length) { |
||
68 | if (Length == 0) { return 0; } |
||
69 | return ::memcmp(Lhs,Rhs,Length); |
||
70 | } |
||
71 | |||
72 | public: |
||
73 | /// @name Constructors |
||
74 | /// @{ |
||
75 | |||
76 | /// Construct an empty string ref. |
||
77 | /*implicit*/ StringRef() = default; |
||
78 | |||
79 | /// Disable conversion from nullptr. This prevents things like |
||
80 | /// if (S == nullptr) |
||
81 | StringRef(std::nullptr_t) = delete; |
||
82 | |||
83 | /// Construct a string ref from a cstring. |
||
84 | /*implicit*/ constexpr StringRef(const char *Str) |
||
85 | : Data(Str), Length(Str ? |
||
86 | // GCC 7 doesn't have constexpr char_traits. Fall back to __builtin_strlen. |
||
87 | #if defined(_GLIBCXX_RELEASE) && _GLIBCXX_RELEASE < 8 |
||
88 | __builtin_strlen(Str) |
||
89 | #else |
||
90 | std::char_traits<char>::length(Str) |
||
91 | #endif |
||
92 | : 0) { |
||
93 | } |
||
94 | |||
95 | /// Construct a string ref from a pointer and length. |
||
96 | /*implicit*/ constexpr StringRef(const char *data, size_t length) |
||
97 | : Data(data), Length(length) {} |
||
98 | |||
99 | /// Construct a string ref from an std::string. |
||
100 | /*implicit*/ StringRef(const std::string &Str) |
||
101 | : Data(Str.data()), Length(Str.length()) {} |
||
102 | |||
103 | /// Construct a string ref from an std::string_view. |
||
104 | /*implicit*/ constexpr StringRef(std::string_view Str) |
||
105 | : Data(Str.data()), Length(Str.size()) {} |
||
106 | |||
107 | /// @} |
||
108 | /// @name Iterators |
||
109 | /// @{ |
||
110 | |||
111 | iterator begin() const { return Data; } |
||
112 | |||
113 | iterator end() const { return Data + Length; } |
||
114 | |||
115 | const unsigned char *bytes_begin() const { |
||
116 | return reinterpret_cast<const unsigned char *>(begin()); |
||
117 | } |
||
118 | const unsigned char *bytes_end() const { |
||
119 | return reinterpret_cast<const unsigned char *>(end()); |
||
120 | } |
||
121 | iterator_range<const unsigned char *> bytes() const { |
||
122 | return make_range(bytes_begin(), bytes_end()); |
||
123 | } |
||
124 | |||
125 | /// @} |
||
126 | /// @name String Operations |
||
127 | /// @{ |
||
128 | |||
129 | /// data - Get a pointer to the start of the string (which may not be null |
||
130 | /// terminated). |
||
131 | [[nodiscard]] const char *data() const { return Data; } |
||
132 | |||
133 | /// empty - Check if the string is empty. |
||
134 | [[nodiscard]] constexpr bool empty() const { return Length == 0; } |
||
135 | |||
136 | /// size - Get the string size. |
||
137 | [[nodiscard]] constexpr size_t size() const { return Length; } |
||
138 | |||
139 | /// front - Get the first character in the string. |
||
140 | [[nodiscard]] char front() const { |
||
141 | assert(!empty()); |
||
142 | return Data[0]; |
||
143 | } |
||
144 | |||
145 | /// back - Get the last character in the string. |
||
146 | [[nodiscard]] char back() const { |
||
147 | assert(!empty()); |
||
148 | return Data[Length-1]; |
||
149 | } |
||
150 | |||
151 | // copy - Allocate copy in Allocator and return StringRef to it. |
||
152 | template <typename Allocator> |
||
153 | [[nodiscard]] StringRef copy(Allocator &A) const { |
||
154 | // Don't request a length 0 copy from the allocator. |
||
155 | if (empty()) |
||
156 | return StringRef(); |
||
157 | char *S = A.template Allocate<char>(Length); |
||
158 | std::copy(begin(), end(), S); |
||
159 | return StringRef(S, Length); |
||
160 | } |
||
161 | |||
162 | /// equals - Check for string equality, this is more efficient than |
||
163 | /// compare() when the relative ordering of inequal strings isn't needed. |
||
164 | [[nodiscard]] bool equals(StringRef RHS) const { |
||
165 | return (Length == RHS.Length && |
||
166 | compareMemory(Data, RHS.Data, RHS.Length) == 0); |
||
167 | } |
||
168 | |||
169 | /// Check for string equality, ignoring case. |
||
170 | [[nodiscard]] bool equals_insensitive(StringRef RHS) const { |
||
171 | return Length == RHS.Length && compare_insensitive(RHS) == 0; |
||
172 | } |
||
173 | |||
174 | /// compare - Compare two strings; the result is negative, zero, or positive |
||
175 | /// if this string is lexicographically less than, equal to, or greater than |
||
176 | /// the \p RHS. |
||
177 | [[nodiscard]] int compare(StringRef RHS) const { |
||
178 | // Check the prefix for a mismatch. |
||
179 | if (int Res = compareMemory(Data, RHS.Data, std::min(Length, RHS.Length))) |
||
180 | return Res < 0 ? -1 : 1; |
||
181 | |||
182 | // Otherwise the prefixes match, so we only need to check the lengths. |
||
183 | if (Length == RHS.Length) |
||
184 | return 0; |
||
185 | return Length < RHS.Length ? -1 : 1; |
||
186 | } |
||
187 | |||
188 | /// Compare two strings, ignoring case. |
||
189 | [[nodiscard]] int compare_insensitive(StringRef RHS) const; |
||
190 | |||
191 | /// compare_numeric - Compare two strings, treating sequences of digits as |
||
192 | /// numbers. |
||
193 | [[nodiscard]] int compare_numeric(StringRef RHS) const; |
||
194 | |||
195 | /// Determine the edit distance between this string and another |
||
196 | /// string. |
||
197 | /// |
||
198 | /// \param Other the string to compare this string against. |
||
199 | /// |
||
200 | /// \param AllowReplacements whether to allow character |
||
201 | /// replacements (change one character into another) as a single |
||
202 | /// operation, rather than as two operations (an insertion and a |
||
203 | /// removal). |
||
204 | /// |
||
205 | /// \param MaxEditDistance If non-zero, the maximum edit distance that |
||
206 | /// this routine is allowed to compute. If the edit distance will exceed |
||
207 | /// that maximum, returns \c MaxEditDistance+1. |
||
208 | /// |
||
209 | /// \returns the minimum number of character insertions, removals, |
||
210 | /// or (if \p AllowReplacements is \c true) replacements needed to |
||
211 | /// transform one of the given strings into the other. If zero, |
||
212 | /// the strings are identical. |
||
213 | [[nodiscard]] unsigned edit_distance(StringRef Other, |
||
214 | bool AllowReplacements = true, |
||
215 | unsigned MaxEditDistance = 0) const; |
||
216 | |||
217 | [[nodiscard]] unsigned |
||
218 | edit_distance_insensitive(StringRef Other, bool AllowReplacements = true, |
||
219 | unsigned MaxEditDistance = 0) const; |
||
220 | |||
221 | /// str - Get the contents as an std::string. |
||
222 | [[nodiscard]] std::string str() const { |
||
223 | if (!Data) return std::string(); |
||
224 | return std::string(Data, Length); |
||
225 | } |
||
226 | |||
227 | /// @} |
||
228 | /// @name Operator Overloads |
||
229 | /// @{ |
||
230 | |||
231 | [[nodiscard]] char operator[](size_t Index) const { |
||
232 | assert(Index < Length && "Invalid index!"); |
||
233 | return Data[Index]; |
||
234 | } |
||
235 | |||
236 | /// Disallow accidental assignment from a temporary std::string. |
||
237 | /// |
||
238 | /// The declaration here is extra complicated so that `stringRef = {}` |
||
239 | /// and `stringRef = "abc"` continue to select the move assignment operator. |
||
240 | template <typename T> |
||
241 | std::enable_if_t<std::is_same<T, std::string>::value, StringRef> & |
||
242 | operator=(T &&Str) = delete; |
||
243 | |||
244 | /// @} |
||
245 | /// @name Type Conversions |
||
246 | /// @{ |
||
247 | |||
248 | operator std::string_view() const { |
||
249 | return std::string_view(data(), size()); |
||
250 | } |
||
251 | |||
252 | /// @} |
||
253 | /// @name String Predicates |
||
254 | /// @{ |
||
255 | |||
256 | /// Check if this string starts with the given \p Prefix. |
||
257 | [[nodiscard]] bool starts_with(StringRef Prefix) const { |
||
258 | return Length >= Prefix.Length && |
||
259 | compareMemory(Data, Prefix.Data, Prefix.Length) == 0; |
||
260 | } |
||
261 | [[nodiscard]] bool startswith(StringRef Prefix) const { |
||
262 | return starts_with(Prefix); |
||
263 | } |
||
264 | |||
265 | /// Check if this string starts with the given \p Prefix, ignoring case. |
||
266 | [[nodiscard]] bool starts_with_insensitive(StringRef Prefix) const; |
||
267 | [[nodiscard]] bool startswith_insensitive(StringRef Prefix) const { |
||
268 | return starts_with_insensitive(Prefix); |
||
269 | } |
||
270 | |||
271 | /// Check if this string ends with the given \p Suffix. |
||
272 | [[nodiscard]] bool ends_with(StringRef Suffix) const { |
||
273 | return Length >= Suffix.Length && |
||
274 | compareMemory(end() - Suffix.Length, Suffix.Data, Suffix.Length) == |
||
275 | 0; |
||
276 | } |
||
277 | [[nodiscard]] bool endswith(StringRef Suffix) const { |
||
278 | return ends_with(Suffix); |
||
279 | } |
||
280 | |||
281 | /// Check if this string ends with the given \p Suffix, ignoring case. |
||
282 | [[nodiscard]] bool ends_with_insensitive(StringRef Suffix) const; |
||
283 | [[nodiscard]] bool endswith_insensitive(StringRef Suffix) const { |
||
284 | return ends_with_insensitive(Suffix); |
||
285 | } |
||
286 | |||
287 | /// @} |
||
288 | /// @name String Searching |
||
289 | /// @{ |
||
290 | |||
291 | /// Search for the first character \p C in the string. |
||
292 | /// |
||
293 | /// \returns The index of the first occurrence of \p C, or npos if not |
||
294 | /// found. |
||
295 | [[nodiscard]] size_t find(char C, size_t From = 0) const { |
||
296 | return std::string_view(*this).find(C, From); |
||
297 | } |
||
298 | |||
299 | /// Search for the first character \p C in the string, ignoring case. |
||
300 | /// |
||
301 | /// \returns The index of the first occurrence of \p C, or npos if not |
||
302 | /// found. |
||
303 | [[nodiscard]] size_t find_insensitive(char C, size_t From = 0) const; |
||
304 | |||
305 | /// Search for the first character satisfying the predicate \p F |
||
306 | /// |
||
307 | /// \returns The index of the first character satisfying \p F starting from |
||
308 | /// \p From, or npos if not found. |
||
309 | [[nodiscard]] size_t find_if(function_ref<bool(char)> F, |
||
310 | size_t From = 0) const { |
||
311 | StringRef S = drop_front(From); |
||
312 | while (!S.empty()) { |
||
313 | if (F(S.front())) |
||
314 | return size() - S.size(); |
||
315 | S = S.drop_front(); |
||
316 | } |
||
317 | return npos; |
||
318 | } |
||
319 | |||
320 | /// Search for the first character not satisfying the predicate \p F |
||
321 | /// |
||
322 | /// \returns The index of the first character not satisfying \p F starting |
||
323 | /// from \p From, or npos if not found. |
||
324 | [[nodiscard]] size_t find_if_not(function_ref<bool(char)> F, |
||
325 | size_t From = 0) const { |
||
326 | return find_if([F](char c) { return !F(c); }, From); |
||
327 | } |
||
328 | |||
329 | /// Search for the first string \p Str in the string. |
||
330 | /// |
||
331 | /// \returns The index of the first occurrence of \p Str, or npos if not |
||
332 | /// found. |
||
333 | [[nodiscard]] size_t find(StringRef Str, size_t From = 0) const; |
||
334 | |||
335 | /// Search for the first string \p Str in the string, ignoring case. |
||
336 | /// |
||
337 | /// \returns The index of the first occurrence of \p Str, or npos if not |
||
338 | /// found. |
||
339 | [[nodiscard]] size_t find_insensitive(StringRef Str, size_t From = 0) const; |
||
340 | |||
341 | /// Search for the last character \p C in the string. |
||
342 | /// |
||
343 | /// \returns The index of the last occurrence of \p C, or npos if not |
||
344 | /// found. |
||
345 | [[nodiscard]] size_t rfind(char C, size_t From = npos) const { |
||
346 | From = std::min(From, Length); |
||
347 | size_t i = From; |
||
348 | while (i != 0) { |
||
349 | --i; |
||
350 | if (Data[i] == C) |
||
351 | return i; |
||
352 | } |
||
353 | return npos; |
||
354 | } |
||
355 | |||
356 | /// Search for the last character \p C in the string, ignoring case. |
||
357 | /// |
||
358 | /// \returns The index of the last occurrence of \p C, or npos if not |
||
359 | /// found. |
||
360 | [[nodiscard]] size_t rfind_insensitive(char C, size_t From = npos) const; |
||
361 | |||
362 | /// Search for the last string \p Str in the string. |
||
363 | /// |
||
364 | /// \returns The index of the last occurrence of \p Str, or npos if not |
||
365 | /// found. |
||
366 | [[nodiscard]] size_t rfind(StringRef Str) const; |
||
367 | |||
368 | /// Search for the last string \p Str in the string, ignoring case. |
||
369 | /// |
||
370 | /// \returns The index of the last occurrence of \p Str, or npos if not |
||
371 | /// found. |
||
372 | [[nodiscard]] size_t rfind_insensitive(StringRef Str) const; |
||
373 | |||
374 | /// Find the first character in the string that is \p C, or npos if not |
||
375 | /// found. Same as find. |
||
376 | [[nodiscard]] size_t find_first_of(char C, size_t From = 0) const { |
||
377 | return find(C, From); |
||
378 | } |
||
379 | |||
380 | /// Find the first character in the string that is in \p Chars, or npos if |
||
381 | /// not found. |
||
382 | /// |
||
383 | /// Complexity: O(size() + Chars.size()) |
||
384 | [[nodiscard]] size_t find_first_of(StringRef Chars, size_t From = 0) const; |
||
385 | |||
386 | /// Find the first character in the string that is not \p C or npos if not |
||
387 | /// found. |
||
388 | [[nodiscard]] size_t find_first_not_of(char C, size_t From = 0) const; |
||
389 | |||
390 | /// Find the first character in the string that is not in the string |
||
391 | /// \p Chars, or npos if not found. |
||
392 | /// |
||
393 | /// Complexity: O(size() + Chars.size()) |
||
394 | [[nodiscard]] size_t find_first_not_of(StringRef Chars, |
||
395 | size_t From = 0) const; |
||
396 | |||
397 | /// Find the last character in the string that is \p C, or npos if not |
||
398 | /// found. |
||
399 | [[nodiscard]] size_t find_last_of(char C, size_t From = npos) const { |
||
400 | return rfind(C, From); |
||
401 | } |
||
402 | |||
403 | /// Find the last character in the string that is in \p C, or npos if not |
||
404 | /// found. |
||
405 | /// |
||
406 | /// Complexity: O(size() + Chars.size()) |
||
407 | [[nodiscard]] size_t find_last_of(StringRef Chars, |
||
408 | size_t From = npos) const; |
||
409 | |||
410 | /// Find the last character in the string that is not \p C, or npos if not |
||
411 | /// found. |
||
412 | [[nodiscard]] size_t find_last_not_of(char C, size_t From = npos) const; |
||
413 | |||
414 | /// Find the last character in the string that is not in \p Chars, or |
||
415 | /// npos if not found. |
||
416 | /// |
||
417 | /// Complexity: O(size() + Chars.size()) |
||
418 | [[nodiscard]] size_t find_last_not_of(StringRef Chars, |
||
419 | size_t From = npos) const; |
||
420 | |||
421 | /// Return true if the given string is a substring of *this, and false |
||
422 | /// otherwise. |
||
423 | [[nodiscard]] bool contains(StringRef Other) const { |
||
424 | return find(Other) != npos; |
||
425 | } |
||
426 | |||
427 | /// Return true if the given character is contained in *this, and false |
||
428 | /// otherwise. |
||
429 | [[nodiscard]] bool contains(char C) const { |
||
430 | return find_first_of(C) != npos; |
||
431 | } |
||
432 | |||
433 | /// Return true if the given string is a substring of *this, and false |
||
434 | /// otherwise. |
||
435 | [[nodiscard]] bool contains_insensitive(StringRef Other) const { |
||
436 | return find_insensitive(Other) != npos; |
||
437 | } |
||
438 | |||
439 | /// Return true if the given character is contained in *this, and false |
||
440 | /// otherwise. |
||
441 | [[nodiscard]] bool contains_insensitive(char C) const { |
||
442 | return find_insensitive(C) != npos; |
||
443 | } |
||
444 | |||
445 | /// @} |
||
446 | /// @name Helpful Algorithms |
||
447 | /// @{ |
||
448 | |||
449 | /// Return the number of occurrences of \p C in the string. |
||
450 | [[nodiscard]] size_t count(char C) const { |
||
451 | size_t Count = 0; |
||
452 | for (size_t i = 0, e = Length; i != e; ++i) |
||
453 | if (Data[i] == C) |
||
454 | ++Count; |
||
455 | return Count; |
||
456 | } |
||
457 | |||
458 | /// Return the number of non-overlapped occurrences of \p Str in |
||
459 | /// the string. |
||
460 | size_t count(StringRef Str) const; |
||
461 | |||
462 | /// Parse the current string as an integer of the specified radix. If |
||
463 | /// \p Radix is specified as zero, this does radix autosensing using |
||
464 | /// extended C rules: 0 is octal, 0x is hex, 0b is binary. |
||
465 | /// |
||
466 | /// If the string is invalid or if only a subset of the string is valid, |
||
467 | /// this returns true to signify the error. The string is considered |
||
468 | /// erroneous if empty or if it overflows T. |
||
469 | template <typename T> bool getAsInteger(unsigned Radix, T &Result) const { |
||
470 | if constexpr (std::numeric_limits<T>::is_signed) { |
||
471 | long long LLVal; |
||
472 | if (getAsSignedInteger(*this, Radix, LLVal) || |
||
473 | static_cast<T>(LLVal) != LLVal) |
||
474 | return true; |
||
475 | Result = LLVal; |
||
476 | } else { |
||
477 | unsigned long long ULLVal; |
||
478 | // The additional cast to unsigned long long is required to avoid the |
||
479 | // Visual C++ warning C4805: '!=' : unsafe mix of type 'bool' and type |
||
480 | // 'unsigned __int64' when instantiating getAsInteger with T = bool. |
||
481 | if (getAsUnsignedInteger(*this, Radix, ULLVal) || |
||
482 | static_cast<unsigned long long>(static_cast<T>(ULLVal)) != ULLVal) |
||
483 | return true; |
||
484 | Result = ULLVal; |
||
485 | } |
||
486 | return false; |
||
487 | } |
||
488 | |||
489 | /// Parse the current string as an integer of the specified radix. If |
||
490 | /// \p Radix is specified as zero, this does radix autosensing using |
||
491 | /// extended C rules: 0 is octal, 0x is hex, 0b is binary. |
||
492 | /// |
||
493 | /// If the string does not begin with a number of the specified radix, |
||
494 | /// this returns true to signify the error. The string is considered |
||
495 | /// erroneous if empty or if it overflows T. |
||
496 | /// The portion of the string representing the discovered numeric value |
||
497 | /// is removed from the beginning of the string. |
||
498 | template <typename T> bool consumeInteger(unsigned Radix, T &Result) { |
||
499 | if constexpr (std::numeric_limits<T>::is_signed) { |
||
500 | long long LLVal; |
||
501 | if (consumeSignedInteger(*this, Radix, LLVal) || |
||
502 | static_cast<long long>(static_cast<T>(LLVal)) != LLVal) |
||
503 | return true; |
||
504 | Result = LLVal; |
||
505 | } else { |
||
506 | unsigned long long ULLVal; |
||
507 | if (consumeUnsignedInteger(*this, Radix, ULLVal) || |
||
508 | static_cast<unsigned long long>(static_cast<T>(ULLVal)) != ULLVal) |
||
509 | return true; |
||
510 | Result = ULLVal; |
||
511 | } |
||
512 | return false; |
||
513 | } |
||
514 | |||
515 | /// Parse the current string as an integer of the specified \p Radix, or of |
||
516 | /// an autosensed radix if the \p Radix given is 0. The current value in |
||
517 | /// \p Result is discarded, and the storage is changed to be wide enough to |
||
518 | /// store the parsed integer. |
||
519 | /// |
||
520 | /// \returns true if the string does not solely consist of a valid |
||
521 | /// non-empty number in the appropriate base. |
||
522 | /// |
||
523 | /// APInt::fromString is superficially similar but assumes the |
||
524 | /// string is well-formed in the given radix. |
||
525 | bool getAsInteger(unsigned Radix, APInt &Result) const; |
||
526 | |||
527 | /// Parse the current string as an IEEE double-precision floating |
||
528 | /// point value. The string must be a well-formed double. |
||
529 | /// |
||
530 | /// If \p AllowInexact is false, the function will fail if the string |
||
531 | /// cannot be represented exactly. Otherwise, the function only fails |
||
532 | /// in case of an overflow or underflow, or an invalid floating point |
||
533 | /// representation. |
||
534 | bool getAsDouble(double &Result, bool AllowInexact = true) const; |
||
535 | |||
536 | /// @} |
||
537 | /// @name String Operations |
||
538 | /// @{ |
||
539 | |||
540 | // Convert the given ASCII string to lowercase. |
||
541 | [[nodiscard]] std::string lower() const; |
||
542 | |||
543 | /// Convert the given ASCII string to uppercase. |
||
544 | [[nodiscard]] std::string upper() const; |
||
545 | |||
546 | /// @} |
||
547 | /// @name Substring Operations |
||
548 | /// @{ |
||
549 | |||
550 | /// Return a reference to the substring from [Start, Start + N). |
||
551 | /// |
||
552 | /// \param Start The index of the starting character in the substring; if |
||
553 | /// the index is npos or greater than the length of the string then the |
||
554 | /// empty substring will be returned. |
||
555 | /// |
||
556 | /// \param N The number of characters to included in the substring. If N |
||
557 | /// exceeds the number of characters remaining in the string, the string |
||
558 | /// suffix (starting with \p Start) will be returned. |
||
559 | [[nodiscard]] constexpr StringRef substr(size_t Start, |
||
560 | size_t N = npos) const { |
||
561 | Start = std::min(Start, Length); |
||
562 | return StringRef(Data + Start, std::min(N, Length - Start)); |
||
563 | } |
||
564 | |||
565 | /// Return a StringRef equal to 'this' but with only the first \p N |
||
566 | /// elements remaining. If \p N is greater than the length of the |
||
567 | /// string, the entire string is returned. |
||
568 | [[nodiscard]] StringRef take_front(size_t N = 1) const { |
||
569 | if (N >= size()) |
||
570 | return *this; |
||
571 | return drop_back(size() - N); |
||
572 | } |
||
573 | |||
574 | /// Return a StringRef equal to 'this' but with only the last \p N |
||
575 | /// elements remaining. If \p N is greater than the length of the |
||
576 | /// string, the entire string is returned. |
||
577 | [[nodiscard]] StringRef take_back(size_t N = 1) const { |
||
578 | if (N >= size()) |
||
579 | return *this; |
||
580 | return drop_front(size() - N); |
||
581 | } |
||
582 | |||
583 | /// Return the longest prefix of 'this' such that every character |
||
584 | /// in the prefix satisfies the given predicate. |
||
585 | [[nodiscard]] StringRef take_while(function_ref<bool(char)> F) const { |
||
586 | return substr(0, find_if_not(F)); |
||
587 | } |
||
588 | |||
589 | /// Return the longest prefix of 'this' such that no character in |
||
590 | /// the prefix satisfies the given predicate. |
||
591 | [[nodiscard]] StringRef take_until(function_ref<bool(char)> F) const { |
||
592 | return substr(0, find_if(F)); |
||
593 | } |
||
594 | |||
595 | /// Return a StringRef equal to 'this' but with the first \p N elements |
||
596 | /// dropped. |
||
597 | [[nodiscard]] StringRef drop_front(size_t N = 1) const { |
||
598 | assert(size() >= N && "Dropping more elements than exist"); |
||
599 | return substr(N); |
||
600 | } |
||
601 | |||
602 | /// Return a StringRef equal to 'this' but with the last \p N elements |
||
603 | /// dropped. |
||
604 | [[nodiscard]] StringRef drop_back(size_t N = 1) const { |
||
605 | assert(size() >= N && "Dropping more elements than exist"); |
||
606 | return substr(0, size()-N); |
||
607 | } |
||
608 | |||
609 | /// Return a StringRef equal to 'this', but with all characters satisfying |
||
610 | /// the given predicate dropped from the beginning of the string. |
||
611 | [[nodiscard]] StringRef drop_while(function_ref<bool(char)> F) const { |
||
612 | return substr(find_if_not(F)); |
||
613 | } |
||
614 | |||
615 | /// Return a StringRef equal to 'this', but with all characters not |
||
616 | /// satisfying the given predicate dropped from the beginning of the string. |
||
617 | [[nodiscard]] StringRef drop_until(function_ref<bool(char)> F) const { |
||
618 | return substr(find_if(F)); |
||
619 | } |
||
620 | |||
621 | /// Returns true if this StringRef has the given prefix and removes that |
||
622 | /// prefix. |
||
623 | bool consume_front(StringRef Prefix) { |
||
624 | if (!starts_with(Prefix)) |
||
625 | return false; |
||
626 | |||
627 | *this = drop_front(Prefix.size()); |
||
628 | return true; |
||
629 | } |
||
630 | |||
631 | /// Returns true if this StringRef has the given prefix, ignoring case, |
||
632 | /// and removes that prefix. |
||
633 | bool consume_front_insensitive(StringRef Prefix) { |
||
634 | if (!startswith_insensitive(Prefix)) |
||
635 | return false; |
||
636 | |||
637 | *this = drop_front(Prefix.size()); |
||
638 | return true; |
||
639 | } |
||
640 | |||
641 | /// Returns true if this StringRef has the given suffix and removes that |
||
642 | /// suffix. |
||
643 | bool consume_back(StringRef Suffix) { |
||
644 | if (!ends_with(Suffix)) |
||
645 | return false; |
||
646 | |||
647 | *this = drop_back(Suffix.size()); |
||
648 | return true; |
||
649 | } |
||
650 | |||
651 | /// Returns true if this StringRef has the given suffix, ignoring case, |
||
652 | /// and removes that suffix. |
||
653 | bool consume_back_insensitive(StringRef Suffix) { |
||
654 | if (!endswith_insensitive(Suffix)) |
||
655 | return false; |
||
656 | |||
657 | *this = drop_back(Suffix.size()); |
||
658 | return true; |
||
659 | } |
||
660 | |||
661 | /// Return a reference to the substring from [Start, End). |
||
662 | /// |
||
663 | /// \param Start The index of the starting character in the substring; if |
||
664 | /// the index is npos or greater than the length of the string then the |
||
665 | /// empty substring will be returned. |
||
666 | /// |
||
667 | /// \param End The index following the last character to include in the |
||
668 | /// substring. If this is npos or exceeds the number of characters |
||
669 | /// remaining in the string, the string suffix (starting with \p Start) |
||
670 | /// will be returned. If this is less than \p Start, an empty string will |
||
671 | /// be returned. |
||
672 | [[nodiscard]] StringRef slice(size_t Start, size_t End) const { |
||
673 | Start = std::min(Start, Length); |
||
674 | End = std::min(std::max(Start, End), Length); |
||
675 | return StringRef(Data + Start, End - Start); |
||
676 | } |
||
677 | |||
678 | /// Split into two substrings around the first occurrence of a separator |
||
679 | /// character. |
||
680 | /// |
||
681 | /// If \p Separator is in the string, then the result is a pair (LHS, RHS) |
||
682 | /// such that (*this == LHS + Separator + RHS) is true and RHS is |
||
683 | /// maximal. If \p Separator is not in the string, then the result is a |
||
684 | /// pair (LHS, RHS) where (*this == LHS) and (RHS == ""). |
||
685 | /// |
||
686 | /// \param Separator The character to split on. |
||
687 | /// \returns The split substrings. |
||
688 | [[nodiscard]] std::pair<StringRef, StringRef> split(char Separator) const { |
||
689 | return split(StringRef(&Separator, 1)); |
||
690 | } |
||
691 | |||
692 | /// Split into two substrings around the first occurrence of a separator |
||
693 | /// string. |
||
694 | /// |
||
695 | /// If \p Separator is in the string, then the result is a pair (LHS, RHS) |
||
696 | /// such that (*this == LHS + Separator + RHS) is true and RHS is |
||
697 | /// maximal. If \p Separator is not in the string, then the result is a |
||
698 | /// pair (LHS, RHS) where (*this == LHS) and (RHS == ""). |
||
699 | /// |
||
700 | /// \param Separator - The string to split on. |
||
701 | /// \return - The split substrings. |
||
702 | [[nodiscard]] std::pair<StringRef, StringRef> |
||
703 | split(StringRef Separator) const { |
||
704 | size_t Idx = find(Separator); |
||
705 | if (Idx == npos) |
||
706 | return std::make_pair(*this, StringRef()); |
||
707 | return std::make_pair(slice(0, Idx), slice(Idx + Separator.size(), npos)); |
||
708 | } |
||
709 | |||
710 | /// Split into two substrings around the last occurrence of a separator |
||
711 | /// string. |
||
712 | /// |
||
713 | /// If \p Separator is in the string, then the result is a pair (LHS, RHS) |
||
714 | /// such that (*this == LHS + Separator + RHS) is true and RHS is |
||
715 | /// minimal. If \p Separator is not in the string, then the result is a |
||
716 | /// pair (LHS, RHS) where (*this == LHS) and (RHS == ""). |
||
717 | /// |
||
718 | /// \param Separator - The string to split on. |
||
719 | /// \return - The split substrings. |
||
720 | [[nodiscard]] std::pair<StringRef, StringRef> |
||
721 | rsplit(StringRef Separator) const { |
||
722 | size_t Idx = rfind(Separator); |
||
723 | if (Idx == npos) |
||
724 | return std::make_pair(*this, StringRef()); |
||
725 | return std::make_pair(slice(0, Idx), slice(Idx + Separator.size(), npos)); |
||
726 | } |
||
727 | |||
728 | /// Split into substrings around the occurrences of a separator string. |
||
729 | /// |
||
730 | /// Each substring is stored in \p A. If \p MaxSplit is >= 0, at most |
||
731 | /// \p MaxSplit splits are done and consequently <= \p MaxSplit + 1 |
||
732 | /// elements are added to A. |
||
733 | /// If \p KeepEmpty is false, empty strings are not added to \p A. They |
||
734 | /// still count when considering \p MaxSplit |
||
735 | /// An useful invariant is that |
||
736 | /// Separator.join(A) == *this if MaxSplit == -1 and KeepEmpty == true |
||
737 | /// |
||
738 | /// \param A - Where to put the substrings. |
||
739 | /// \param Separator - The string to split on. |
||
740 | /// \param MaxSplit - The maximum number of times the string is split. |
||
741 | /// \param KeepEmpty - True if empty substring should be added. |
||
742 | void split(SmallVectorImpl<StringRef> &A, |
||
743 | StringRef Separator, int MaxSplit = -1, |
||
744 | bool KeepEmpty = true) const; |
||
745 | |||
746 | /// Split into substrings around the occurrences of a separator character. |
||
747 | /// |
||
748 | /// Each substring is stored in \p A. If \p MaxSplit is >= 0, at most |
||
749 | /// \p MaxSplit splits are done and consequently <= \p MaxSplit + 1 |
||
750 | /// elements are added to A. |
||
751 | /// If \p KeepEmpty is false, empty strings are not added to \p A. They |
||
752 | /// still count when considering \p MaxSplit |
||
753 | /// An useful invariant is that |
||
754 | /// Separator.join(A) == *this if MaxSplit == -1 and KeepEmpty == true |
||
755 | /// |
||
756 | /// \param A - Where to put the substrings. |
||
757 | /// \param Separator - The string to split on. |
||
758 | /// \param MaxSplit - The maximum number of times the string is split. |
||
759 | /// \param KeepEmpty - True if empty substring should be added. |
||
760 | void split(SmallVectorImpl<StringRef> &A, char Separator, int MaxSplit = -1, |
||
761 | bool KeepEmpty = true) const; |
||
762 | |||
763 | /// Split into two substrings around the last occurrence of a separator |
||
764 | /// character. |
||
765 | /// |
||
766 | /// If \p Separator is in the string, then the result is a pair (LHS, RHS) |
||
767 | /// such that (*this == LHS + Separator + RHS) is true and RHS is |
||
768 | /// minimal. If \p Separator is not in the string, then the result is a |
||
769 | /// pair (LHS, RHS) where (*this == LHS) and (RHS == ""). |
||
770 | /// |
||
771 | /// \param Separator - The character to split on. |
||
772 | /// \return - The split substrings. |
||
773 | [[nodiscard]] std::pair<StringRef, StringRef> rsplit(char Separator) const { |
||
774 | return rsplit(StringRef(&Separator, 1)); |
||
775 | } |
||
776 | |||
777 | /// Return string with consecutive \p Char characters starting from the |
||
778 | /// the left removed. |
||
779 | [[nodiscard]] StringRef ltrim(char Char) const { |
||
780 | return drop_front(std::min(Length, find_first_not_of(Char))); |
||
781 | } |
||
782 | |||
783 | /// Return string with consecutive characters in \p Chars starting from |
||
784 | /// the left removed. |
||
785 | [[nodiscard]] StringRef ltrim(StringRef Chars = " \t\n\v\f\r") const { |
||
786 | return drop_front(std::min(Length, find_first_not_of(Chars))); |
||
787 | } |
||
788 | |||
789 | /// Return string with consecutive \p Char characters starting from the |
||
790 | /// right removed. |
||
791 | [[nodiscard]] StringRef rtrim(char Char) const { |
||
792 | return drop_back(Length - std::min(Length, find_last_not_of(Char) + 1)); |
||
793 | } |
||
794 | |||
795 | /// Return string with consecutive characters in \p Chars starting from |
||
796 | /// the right removed. |
||
797 | [[nodiscard]] StringRef rtrim(StringRef Chars = " \t\n\v\f\r") const { |
||
798 | return drop_back(Length - std::min(Length, find_last_not_of(Chars) + 1)); |
||
799 | } |
||
800 | |||
801 | /// Return string with consecutive \p Char characters starting from the |
||
802 | /// left and right removed. |
||
803 | [[nodiscard]] StringRef trim(char Char) const { |
||
804 | return ltrim(Char).rtrim(Char); |
||
805 | } |
||
806 | |||
807 | /// Return string with consecutive characters in \p Chars starting from |
||
808 | /// the left and right removed. |
||
809 | [[nodiscard]] StringRef trim(StringRef Chars = " \t\n\v\f\r") const { |
||
810 | return ltrim(Chars).rtrim(Chars); |
||
811 | } |
||
812 | |||
813 | /// Detect the line ending style of the string. |
||
814 | /// |
||
815 | /// If the string contains a line ending, return the line ending character |
||
816 | /// sequence that is detected. Otherwise return '\n' for unix line endings. |
||
817 | /// |
||
818 | /// \return - The line ending character sequence. |
||
819 | [[nodiscard]] StringRef detectEOL() const { |
||
820 | size_t Pos = find('\r'); |
||
821 | if (Pos == npos) { |
||
822 | // If there is no carriage return, assume unix |
||
823 | return "\n"; |
||
824 | } |
||
825 | if (Pos + 1 < Length && Data[Pos + 1] == '\n') |
||
826 | return "\r\n"; // Windows |
||
827 | if (Pos > 0 && Data[Pos - 1] == '\n') |
||
828 | return "\n\r"; // You monster! |
||
829 | return "\r"; // Classic Mac |
||
830 | } |
||
831 | /// @} |
||
832 | }; |
||
833 | |||
834 | /// A wrapper around a string literal that serves as a proxy for constructing |
||
835 | /// global tables of StringRefs with the length computed at compile time. |
||
836 | /// In order to avoid the invocation of a global constructor, StringLiteral |
||
837 | /// should *only* be used in a constexpr context, as such: |
||
838 | /// |
||
839 | /// constexpr StringLiteral S("test"); |
||
840 | /// |
||
841 | class StringLiteral : public StringRef { |
||
842 | private: |
||
843 | constexpr StringLiteral(const char *Str, size_t N) : StringRef(Str, N) { |
||
844 | } |
||
845 | |||
846 | public: |
||
847 | template <size_t N> |
||
848 | constexpr StringLiteral(const char (&Str)[N]) |
||
849 | #if defined(__clang__) && __has_attribute(enable_if) |
||
850 | #pragma clang diagnostic push |
||
851 | #pragma clang diagnostic ignored "-Wgcc-compat" |
||
852 | __attribute((enable_if(__builtin_strlen(Str) == N - 1, |
||
853 | "invalid string literal"))) |
||
854 | #pragma clang diagnostic pop |
||
855 | #endif |
||
856 | : StringRef(Str, N - 1) { |
||
857 | } |
||
858 | |||
859 | // Explicit construction for strings like "foo\0bar". |
||
860 | template <size_t N> |
||
861 | static constexpr StringLiteral withInnerNUL(const char (&Str)[N]) { |
||
862 | return StringLiteral(Str, N - 1); |
||
863 | } |
||
864 | }; |
||
865 | |||
866 | /// @name StringRef Comparison Operators |
||
867 | /// @{ |
||
868 | |||
869 | inline bool operator==(StringRef LHS, StringRef RHS) { |
||
870 | return LHS.equals(RHS); |
||
871 | } |
||
872 | |||
873 | inline bool operator!=(StringRef LHS, StringRef RHS) { return !(LHS == RHS); } |
||
874 | |||
875 | inline bool operator<(StringRef LHS, StringRef RHS) { |
||
876 | return LHS.compare(RHS) < 0; |
||
877 | } |
||
878 | |||
879 | inline bool operator<=(StringRef LHS, StringRef RHS) { |
||
880 | return LHS.compare(RHS) <= 0; |
||
881 | } |
||
882 | |||
883 | inline bool operator>(StringRef LHS, StringRef RHS) { |
||
884 | return LHS.compare(RHS) > 0; |
||
885 | } |
||
886 | |||
887 | inline bool operator>=(StringRef LHS, StringRef RHS) { |
||
888 | return LHS.compare(RHS) >= 0; |
||
889 | } |
||
890 | |||
891 | inline std::string &operator+=(std::string &buffer, StringRef string) { |
||
892 | return buffer.append(string.data(), string.size()); |
||
893 | } |
||
894 | |||
895 | /// @} |
||
896 | |||
897 | /// Compute a hash_code for a StringRef. |
||
898 | [[nodiscard]] hash_code hash_value(StringRef S); |
||
899 | |||
900 | // Provide DenseMapInfo for StringRefs. |
||
901 | template <> struct DenseMapInfo<StringRef, void> { |
||
902 | static inline StringRef getEmptyKey() { |
||
903 | return StringRef( |
||
904 | reinterpret_cast<const char *>(~static_cast<uintptr_t>(0)), 0); |
||
905 | } |
||
906 | |||
907 | static inline StringRef getTombstoneKey() { |
||
908 | return StringRef( |
||
909 | reinterpret_cast<const char *>(~static_cast<uintptr_t>(1)), 0); |
||
910 | } |
||
911 | |||
912 | static unsigned getHashValue(StringRef Val); |
||
913 | |||
914 | static bool isEqual(StringRef LHS, StringRef RHS) { |
||
915 | if (RHS.data() == getEmptyKey().data()) |
||
916 | return LHS.data() == getEmptyKey().data(); |
||
917 | if (RHS.data() == getTombstoneKey().data()) |
||
918 | return LHS.data() == getTombstoneKey().data(); |
||
919 | return LHS == RHS; |
||
920 | } |
||
921 | }; |
||
922 | |||
923 | } // end namespace llvm |
||
924 | |||
925 | #endif // LLVM_ADT_STRINGREF_H |