1 //===--- StringRef.h - Constant String Reference Wrapper --------*- C++ -*-===//
3 // The LLVM Compiler Infrastructure
5 // This file is distributed under the University of Illinois Open Source
6 // License. See LICENSE.TXT for details.
8 //===----------------------------------------------------------------------===//
10 #ifndef LLVM_ADT_STRINGREF_H
11 #define LLVM_ADT_STRINGREF_H
22 class SmallVectorImpl;
27 /// Helper functions for StringRef::getAsInteger.
28 bool getAsUnsignedInteger(StringRef Str, unsigned Radix,
29 unsigned long long &Result);
31 bool getAsSignedInteger(StringRef Str, unsigned Radix, long long &Result);
33 /// StringRef - Represent a constant reference to a string, i.e. a character
34 /// array and a length, which need not be null terminated.
36 /// This class does not own the string data, it is expected to be used in
37 /// situations where the character data resides in some other buffer, whose
38 /// lifetime extends past that of the StringRef. For this reason, it is not in
39 /// general safe to store a StringRef.
42 typedef const char *iterator;
43 typedef const char *const_iterator;
44 static const size_t npos = ~size_t(0);
45 typedef size_t size_type;
48 /// The start of the string, in an external buffer.
51 /// The length of the string.
54 // Workaround memcmp issue with null pointers (undefined behavior)
55 // by providing a specialized version
56 static int compareMemory(const char *Lhs, const char *Rhs, size_t Length) {
57 if (Length == 0) { return 0; }
58 return ::memcmp(Lhs,Rhs,Length);
62 /// @name Constructors
65 /// Construct an empty string ref.
66 /*implicit*/ StringRef() : Data(nullptr), Length(0) {}
68 /// Construct a string ref from a cstring.
69 /*implicit*/ StringRef(const char *Str)
71 assert(Str && "StringRef cannot be built from a NULL argument");
72 Length = ::strlen(Str); // invoking strlen(NULL) is undefined behavior
75 /// Construct a string ref from a pointer and length.
76 /*implicit*/ StringRef(const char *data, size_t length)
77 : Data(data), Length(length) {
78 assert((data || length == 0) &&
79 "StringRef cannot be built from a NULL argument with non-null length");
82 /// Construct a string ref from an std::string.
83 /*implicit*/ StringRef(const std::string &Str)
84 : Data(Str.data()), Length(Str.length()) {}
90 iterator begin() const { return Data; }
92 iterator end() const { return Data + Length; }
95 /// @name String Operations
98 /// data - Get a pointer to the start of the string (which may not be null
100 const char *data() const { return Data; }
102 /// empty - Check if the string is empty.
103 bool empty() const { return Length == 0; }
105 /// size - Get the string size.
106 size_t size() const { return Length; }
108 /// front - Get the first character in the string.
114 /// back - Get the last character in the string.
117 return Data[Length-1];
120 // copy - Allocate copy in Allocator and return StringRef to it.
121 template <typename Allocator> StringRef copy(Allocator &A) const {
122 char *S = A.template Allocate<char>(Length);
123 std::copy(begin(), end(), S);
124 return StringRef(S, Length);
127 /// equals - Check for string equality, this is more efficient than
128 /// compare() when the relative ordering of inequal strings isn't needed.
129 bool equals(StringRef RHS) const {
130 return (Length == RHS.Length &&
131 compareMemory(Data, RHS.Data, RHS.Length) == 0);
134 /// equals_lower - Check for string equality, ignoring case.
135 bool equals_lower(StringRef RHS) const {
136 return Length == RHS.Length && compare_lower(RHS) == 0;
139 /// compare - Compare two strings; the result is -1, 0, or 1 if this string
140 /// is lexicographically less than, equal to, or greater than the \p RHS.
141 int compare(StringRef RHS) const {
142 // Check the prefix for a mismatch.
143 if (int Res = compareMemory(Data, RHS.Data, std::min(Length, RHS.Length)))
144 return Res < 0 ? -1 : 1;
146 // Otherwise the prefixes match, so we only need to check the lengths.
147 if (Length == RHS.Length)
149 return Length < RHS.Length ? -1 : 1;
152 /// compare_lower - Compare two strings, ignoring case.
153 int compare_lower(StringRef RHS) const;
155 /// compare_numeric - Compare two strings, treating sequences of digits as
157 int compare_numeric(StringRef RHS) const;
159 /// \brief Determine the edit distance between this string and another
162 /// \param Other the string to compare this string against.
164 /// \param AllowReplacements whether to allow character
165 /// replacements (change one character into another) as a single
166 /// operation, rather than as two operations (an insertion and a
169 /// \param MaxEditDistance If non-zero, the maximum edit distance that
170 /// this routine is allowed to compute. If the edit distance will exceed
171 /// that maximum, returns \c MaxEditDistance+1.
173 /// \returns the minimum number of character insertions, removals,
174 /// or (if \p AllowReplacements is \c true) replacements needed to
175 /// transform one of the given strings into the other. If zero,
176 /// the strings are identical.
177 unsigned edit_distance(StringRef Other, bool AllowReplacements = true,
178 unsigned MaxEditDistance = 0) const;
180 /// str - Get the contents as an std::string.
181 std::string str() const {
182 if (!Data) return std::string();
183 return std::string(Data, Length);
187 /// @name Operator Overloads
190 char operator[](size_t Index) const {
191 assert(Index < Length && "Invalid index!");
196 /// @name Type Conversions
199 operator std::string() const {
204 /// @name String Predicates
207 /// Check if this string starts with the given \p Prefix.
208 bool startswith(StringRef Prefix) const {
209 return Length >= Prefix.Length &&
210 compareMemory(Data, Prefix.Data, Prefix.Length) == 0;
213 /// Check if this string starts with the given \p Prefix, ignoring case.
214 bool startswith_lower(StringRef Prefix) const;
216 /// Check if this string ends with the given \p Suffix.
217 bool endswith(StringRef Suffix) const {
218 return Length >= Suffix.Length &&
219 compareMemory(end() - Suffix.Length, Suffix.Data, Suffix.Length) == 0;
222 /// Check if this string ends with the given \p Suffix, ignoring case.
223 bool endswith_lower(StringRef Suffix) const;
226 /// @name String Searching
229 /// Search for the first character \p C in the string.
231 /// \returns The index of the first occurrence of \p C, or npos if not
233 size_t find(char C, size_t From = 0) const {
234 for (size_t i = std::min(From, Length), e = Length; i != e; ++i)
240 /// Search for the first string \p Str in the string.
242 /// \returns The index of the first occurrence of \p Str, or npos if not
244 size_t find(StringRef Str, size_t From = 0) const;
246 /// Search for the last character \p C in the string.
248 /// \returns The index of the last occurrence of \p C, or npos if not
250 size_t rfind(char C, size_t From = npos) const {
251 From = std::min(From, Length);
261 /// Search for the last string \p Str in the string.
263 /// \returns The index of the last occurrence of \p Str, or npos if not
265 size_t rfind(StringRef Str) const;
267 /// Find the first character in the string that is \p C, or npos if not
268 /// found. Same as find.
269 size_t find_first_of(char C, size_t From = 0) const {
270 return find(C, From);
273 /// Find the first character in the string that is in \p Chars, or npos if
276 /// Complexity: O(size() + Chars.size())
277 size_t find_first_of(StringRef Chars, size_t From = 0) const;
279 /// Find the first character in the string that is not \p C or npos if not
281 size_t find_first_not_of(char C, size_t From = 0) const;
283 /// Find the first character in the string that is not in the string
284 /// \p Chars, or npos if not found.
286 /// Complexity: O(size() + Chars.size())
287 size_t find_first_not_of(StringRef Chars, size_t From = 0) const;
289 /// Find the last character in the string that is \p C, or npos if not
291 size_t find_last_of(char C, size_t From = npos) const {
292 return rfind(C, From);
295 /// Find the last character in the string that is in \p C, or npos if not
298 /// Complexity: O(size() + Chars.size())
299 size_t find_last_of(StringRef Chars, size_t From = npos) const;
301 /// Find the last character in the string that is not \p C, or npos if not
303 size_t find_last_not_of(char C, size_t From = npos) const;
305 /// Find the last character in the string that is not in \p Chars, or
306 /// npos if not found.
308 /// Complexity: O(size() + Chars.size())
309 size_t find_last_not_of(StringRef Chars, size_t From = npos) const;
312 /// @name Helpful Algorithms
315 /// Return the number of occurrences of \p C in the string.
316 size_t count(char C) const {
318 for (size_t i = 0, e = Length; i != e; ++i)
324 /// Return the number of non-overlapped occurrences of \p Str in
326 size_t count(StringRef Str) const;
328 /// Parse the current string as an integer of the specified radix. If
329 /// \p Radix is specified as zero, this does radix autosensing using
330 /// extended C rules: 0 is octal, 0x is hex, 0b is binary.
332 /// If the string is invalid or if only a subset of the string is valid,
333 /// this returns true to signify the error. The string is considered
334 /// erroneous if empty or if it overflows T.
335 template <typename T>
336 typename std::enable_if<std::numeric_limits<T>::is_signed, bool>::type
337 getAsInteger(unsigned Radix, T &Result) const {
339 if (getAsSignedInteger(*this, Radix, LLVal) ||
340 static_cast<T>(LLVal) != LLVal)
346 template <typename T>
347 typename std::enable_if<!std::numeric_limits<T>::is_signed, bool>::type
348 getAsInteger(unsigned Radix, T &Result) const {
349 unsigned long long ULLVal;
350 // The additional cast to unsigned long long is required to avoid the
351 // Visual C++ warning C4805: '!=' : unsafe mix of type 'bool' and type
352 // 'unsigned __int64' when instantiating getAsInteger with T = bool.
353 if (getAsUnsignedInteger(*this, Radix, ULLVal) ||
354 static_cast<unsigned long long>(static_cast<T>(ULLVal)) != ULLVal)
360 /// Parse the current string as an integer of the specified \p Radix, or of
361 /// an autosensed radix if the \p Radix given is 0. The current value in
362 /// \p Result is discarded, and the storage is changed to be wide enough to
363 /// store the parsed integer.
365 /// \returns true if the string does not solely consist of a valid
366 /// non-empty number in the appropriate base.
368 /// APInt::fromString is superficially similar but assumes the
369 /// string is well-formed in the given radix.
370 bool getAsInteger(unsigned Radix, APInt &Result) const;
373 /// @name String Operations
376 // Convert the given ASCII string to lowercase.
377 std::string lower() const;
379 /// Convert the given ASCII string to uppercase.
380 std::string upper() const;
383 /// @name Substring Operations
386 /// Return a reference to the substring from [Start, Start + N).
388 /// \param Start The index of the starting character in the substring; if
389 /// the index is npos or greater than the length of the string then the
390 /// empty substring will be returned.
392 /// \param N The number of characters to included in the substring. If N
393 /// exceeds the number of characters remaining in the string, the string
394 /// suffix (starting with \p Start) will be returned.
395 StringRef substr(size_t Start, size_t N = npos) const {
396 Start = std::min(Start, Length);
397 return StringRef(Data + Start, std::min(N, Length - Start));
400 /// Return a StringRef equal to 'this' but with the first \p N elements
402 StringRef drop_front(size_t N = 1) const {
403 assert(size() >= N && "Dropping more elements than exist");
407 /// Return a StringRef equal to 'this' but with the last \p N elements
409 StringRef drop_back(size_t N = 1) const {
410 assert(size() >= N && "Dropping more elements than exist");
411 return substr(0, size()-N);
414 /// Return a reference to the substring from [Start, End).
416 /// \param Start The index of the starting character in the substring; if
417 /// the index is npos or greater than the length of the string then the
418 /// empty substring will be returned.
420 /// \param End The index following the last character to include in the
421 /// substring. If this is npos, or less than \p Start, or exceeds the
422 /// number of characters remaining in the string, the string suffix
423 /// (starting with \p Start) will be returned.
424 StringRef slice(size_t Start, size_t End) const {
425 Start = std::min(Start, Length);
426 End = std::min(std::max(Start, End), Length);
427 return StringRef(Data + Start, End - Start);
430 /// Split into two substrings around the first occurrence of a separator
433 /// If \p Separator is in the string, then the result is a pair (LHS, RHS)
434 /// such that (*this == LHS + Separator + RHS) is true and RHS is
435 /// maximal. If \p Separator is not in the string, then the result is a
436 /// pair (LHS, RHS) where (*this == LHS) and (RHS == "").
438 /// \param Separator The character to split on.
439 /// \returns The split substrings.
440 std::pair<StringRef, StringRef> split(char Separator) const {
441 size_t Idx = find(Separator);
443 return std::make_pair(*this, StringRef());
444 return std::make_pair(slice(0, Idx), slice(Idx+1, npos));
447 /// Split into two substrings around the first occurrence of a separator
450 /// If \p Separator is in the string, then the result is a pair (LHS, RHS)
451 /// such that (*this == LHS + Separator + RHS) is true and RHS is
452 /// maximal. If \p Separator is not in the string, then the result is a
453 /// pair (LHS, RHS) where (*this == LHS) and (RHS == "").
455 /// \param Separator - The string to split on.
456 /// \return - The split substrings.
457 std::pair<StringRef, StringRef> split(StringRef Separator) const {
458 size_t Idx = find(Separator);
460 return std::make_pair(*this, StringRef());
461 return std::make_pair(slice(0, Idx), slice(Idx + Separator.size(), npos));
464 /// Split into substrings around the occurrences of a separator string.
466 /// Each substring is stored in \p A. If \p MaxSplit is >= 0, at most
467 /// \p MaxSplit splits are done and consequently <= \p MaxSplit
468 /// elements are added to A.
469 /// If \p KeepEmpty is false, empty strings are not added to \p A. They
470 /// still count when considering \p MaxSplit
471 /// An useful invariant is that
472 /// Separator.join(A) == *this if MaxSplit == -1 and KeepEmpty == true
474 /// \param A - Where to put the substrings.
475 /// \param Separator - The string to split on.
476 /// \param MaxSplit - The maximum number of times the string is split.
477 /// \param KeepEmpty - True if empty substring should be added.
478 void split(SmallVectorImpl<StringRef> &A,
479 StringRef Separator, int MaxSplit = -1,
480 bool KeepEmpty = true) const;
482 /// Split into two substrings around the last occurrence of a separator
485 /// If \p Separator is in the string, then the result is a pair (LHS, RHS)
486 /// such that (*this == LHS + Separator + RHS) is true and RHS is
487 /// minimal. If \p Separator is not in the string, then the result is a
488 /// pair (LHS, RHS) where (*this == LHS) and (RHS == "").
490 /// \param Separator - The character to split on.
491 /// \return - The split substrings.
492 std::pair<StringRef, StringRef> rsplit(char Separator) const {
493 size_t Idx = rfind(Separator);
495 return std::make_pair(*this, StringRef());
496 return std::make_pair(slice(0, Idx), slice(Idx+1, npos));
499 /// Return string with consecutive characters in \p Chars starting from
500 /// the left removed.
501 StringRef ltrim(StringRef Chars = " \t\n\v\f\r") const {
502 return drop_front(std::min(Length, find_first_not_of(Chars)));
505 /// Return string with consecutive characters in \p Chars starting from
506 /// the right removed.
507 StringRef rtrim(StringRef Chars = " \t\n\v\f\r") const {
508 return drop_back(Length - std::min(Length, find_last_not_of(Chars) + 1));
511 /// Return string with consecutive characters in \p Chars starting from
512 /// the left and right removed.
513 StringRef trim(StringRef Chars = " \t\n\v\f\r") const {
514 return ltrim(Chars).rtrim(Chars);
520 /// @name StringRef Comparison Operators
523 inline bool operator==(StringRef LHS, StringRef RHS) {
524 return LHS.equals(RHS);
527 inline bool operator!=(StringRef LHS, StringRef RHS) {
528 return !(LHS == RHS);
531 inline bool operator<(StringRef LHS, StringRef RHS) {
532 return LHS.compare(RHS) == -1;
535 inline bool operator<=(StringRef LHS, StringRef RHS) {
536 return LHS.compare(RHS) != 1;
539 inline bool operator>(StringRef LHS, StringRef RHS) {
540 return LHS.compare(RHS) == 1;
543 inline bool operator>=(StringRef LHS, StringRef RHS) {
544 return LHS.compare(RHS) != -1;
547 inline std::string &operator+=(std::string &buffer, StringRef string) {
548 return buffer.append(string.data(), string.size());
553 /// \brief Compute a hash_code for a StringRef.
554 hash_code hash_value(StringRef S);
556 // StringRefs can be treated like a POD type.
557 template <typename T> struct isPodLike;
558 template <> struct isPodLike<StringRef> { static const bool value = true; };