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
20 class SmallVectorImpl;
22 /// StringRef - Represent a constant reference to a string, i.e. a character
23 /// array and a length, which need not be null terminated.
25 /// This class does not own the string data, it is expected to be used in
26 /// situations where the character data resides in some other buffer, whose
27 /// lifetime extends past that of the StringRef. For this reason, it is not in
28 /// general safe to store a StringRef.
31 typedef const char *iterator;
32 static const size_t npos = ~size_t(0);
33 typedef size_t size_type;
36 /// The start of the string, in an external buffer.
39 /// The length of the string.
42 // Workaround PR5482: nearly all gcc 4.x miscompile StringRef and std::min()
43 // Changing the arg of min to be an integer, instead of a reference to an
44 // integer works around this bug.
45 size_t min(size_t a, size_t b) const
50 size_t max(size_t a, size_t b) const
56 /// @name Constructors
59 /// Construct an empty string ref.
60 /*implicit*/ StringRef() : Data(0), Length(0) {}
62 /// Construct a string ref from a cstring.
63 /*implicit*/ StringRef(const char *Str)
64 : Data(Str), Length(::strlen(Str)) {}
66 /// Construct a string ref from a pointer and length.
67 /*implicit*/ StringRef(const char *data, size_t length)
68 : Data(data), Length(length) {}
70 /// Construct a string ref from an std::string.
71 /*implicit*/ StringRef(const std::string &Str)
72 : Data(Str.data()), Length(Str.length()) {}
78 iterator begin() const { return Data; }
80 iterator end() const { return Data + Length; }
83 /// @name String Operations
86 /// data - Get a pointer to the start of the string (which may not be null
88 const char *data() const { return Data; }
90 /// empty - Check if the string is empty.
91 bool empty() const { return Length == 0; }
93 /// size - Get the string size.
94 size_t size() const { return Length; }
96 /// front - Get the first character in the string.
102 /// back - Get the last character in the string.
105 return Data[Length-1];
108 /// equals - Check for string equality, this is more efficient than
109 /// compare() when the relative ordering of inequal strings isn't needed.
110 bool equals(StringRef RHS) const {
111 return (Length == RHS.Length &&
112 memcmp(Data, RHS.Data, RHS.Length) == 0);
115 /// equals_lower - Check for string equality, ignoring case.
116 bool equals_lower(StringRef RHS) const {
117 return Length == RHS.Length && compare_lower(RHS) == 0;
120 /// compare - Compare two strings; the result is -1, 0, or 1 if this string
121 /// is lexicographically less than, equal to, or greater than the \arg RHS.
122 int compare(StringRef RHS) const {
123 // Check the prefix for a mismatch.
124 if (int Res = memcmp(Data, RHS.Data, min(Length, RHS.Length)))
125 return Res < 0 ? -1 : 1;
127 // Otherwise the prefixes match, so we only need to check the lengths.
128 if (Length == RHS.Length)
130 return Length < RHS.Length ? -1 : 1;
133 /// compare_lower - Compare two strings, ignoring case.
134 int compare_lower(StringRef RHS) const;
136 /// \brief Determine the edit distance between this string and another
139 /// \param Other the string to compare this string against.
141 /// \param AllowReplacements whether to allow character
142 /// replacements (change one character into another) as a single
143 /// operation, rather than as two operations (an insertion and a
146 /// \returns the minimum number of character insertions, removals,
147 /// or (if \p AllowReplacements is \c true) replacements needed to
148 /// transform one of the given strings into the other. If zero,
149 /// the strings are identical.
150 unsigned edit_distance(StringRef Other, bool AllowReplacements = true);
152 /// str - Get the contents as an std::string.
153 std::string str() const { return std::string(Data, Length); }
156 /// @name Operator Overloads
159 char operator[](size_t Index) const {
160 assert(Index < Length && "Invalid index!");
165 /// @name Type Conversions
168 operator std::string() const {
173 /// @name String Predicates
176 /// startswith - Check if this string starts with the given \arg Prefix.
177 bool startswith(StringRef Prefix) const {
178 return Length >= Prefix.Length &&
179 memcmp(Data, Prefix.Data, Prefix.Length) == 0;
182 /// endswith - Check if this string ends with the given \arg Suffix.
183 bool endswith(StringRef Suffix) const {
184 return Length >= Suffix.Length &&
185 memcmp(end() - Suffix.Length, Suffix.Data, Suffix.Length) == 0;
189 /// @name String Searching
192 /// find - Search for the first character \arg C in the string.
194 /// \return - The index of the first occurence of \arg C, or npos if not
196 size_t find(char C, size_t From = 0) const {
197 for (size_t i = min(From, Length), e = Length; i != e; ++i)
203 /// find - Search for the first string \arg Str in the string.
205 /// \return - The index of the first occurence of \arg Str, or npos if not
207 size_t find(StringRef Str, size_t From = 0) const;
209 /// rfind - Search for the last character \arg C in the string.
211 /// \return - The index of the last occurence of \arg C, or npos if not
213 size_t rfind(char C, size_t From = npos) const {
214 From = min(From, Length);
224 /// rfind - Search for the last string \arg Str in the string.
226 /// \return - The index of the last occurence of \arg Str, or npos if not
228 size_t rfind(StringRef Str) const;
230 /// find_first_of - Find the first character in the string that is \arg C,
231 /// or npos if not found. Same as find.
232 size_type find_first_of(char C, size_t = 0) const { return find(C); }
234 /// find_first_of - Find the first character in the string that is in \arg
235 /// Chars, or npos if not found.
237 /// Note: O(size() * Chars.size())
238 size_type find_first_of(StringRef Chars, size_t From = 0) const;
240 /// find_first_not_of - Find the first character in the string that is not
241 /// \arg C or npos if not found.
242 size_type find_first_not_of(char C, size_t From = 0) const;
244 /// find_first_not_of - Find the first character in the string that is not
245 /// in the string \arg Chars, or npos if not found.
247 /// Note: O(size() * Chars.size())
248 size_type find_first_not_of(StringRef Chars, size_t From = 0) const;
251 /// @name Helpful Algorithms
254 /// count - Return the number of occurrences of \arg C in the string.
255 size_t count(char C) const {
257 for (size_t i = 0, e = Length; i != e; ++i)
263 /// count - Return the number of non-overlapped occurrences of \arg Str in
265 size_t count(StringRef Str) const;
267 /// getAsInteger - Parse the current string as an integer of the specified
268 /// radix. If Radix is specified as zero, this does radix autosensing using
269 /// extended C rules: 0 is octal, 0x is hex, 0b is binary.
271 /// If the string is invalid or if only a subset of the string is valid,
272 /// this returns true to signify the error. The string is considered
273 /// erroneous if empty.
275 bool getAsInteger(unsigned Radix, long long &Result) const;
276 bool getAsInteger(unsigned Radix, unsigned long long &Result) const;
277 bool getAsInteger(unsigned Radix, int &Result) const;
278 bool getAsInteger(unsigned Radix, unsigned &Result) const;
280 // TODO: Provide overloads for int/unsigned that check for overflow.
283 /// @name Substring Operations
286 /// substr - Return a reference to the substring from [Start, Start + N).
288 /// \param Start - The index of the starting character in the substring; if
289 /// the index is npos or greater than the length of the string then the
290 /// empty substring will be returned.
292 /// \param N - The number of characters to included in the substring. If N
293 /// exceeds the number of characters remaining in the string, the string
294 /// suffix (starting with \arg Start) will be returned.
295 StringRef substr(size_t Start, size_t N = npos) const {
296 Start = min(Start, Length);
297 return StringRef(Data + Start, min(N, Length - Start));
300 /// slice - Return a reference to the substring from [Start, End).
302 /// \param Start - The index of the starting character in the substring; if
303 /// the index is npos or greater than the length of the string then the
304 /// empty substring will be returned.
306 /// \param End - The index following the last character to include in the
307 /// substring. If this is npos, or less than \arg Start, or exceeds the
308 /// number of characters remaining in the string, the string suffix
309 /// (starting with \arg Start) will be returned.
310 StringRef slice(size_t Start, size_t End) const {
311 Start = min(Start, Length);
312 End = min(max(Start, End), Length);
313 return StringRef(Data + Start, End - Start);
316 /// split - Split into two substrings around the first occurence of a
317 /// separator character.
319 /// If \arg Separator is in the string, then the result is a pair (LHS, RHS)
320 /// such that (*this == LHS + Separator + RHS) is true and RHS is
321 /// maximal. If \arg Separator is not in the string, then the result is a
322 /// pair (LHS, RHS) where (*this == LHS) and (RHS == "").
324 /// \param Separator - The character to split on.
325 /// \return - The split substrings.
326 std::pair<StringRef, StringRef> split(char Separator) const {
327 size_t Idx = find(Separator);
329 return std::make_pair(*this, StringRef());
330 return std::make_pair(slice(0, Idx), slice(Idx+1, npos));
333 /// split - Split into two substrings around the first occurence of a
334 /// separator string.
336 /// If \arg Separator is in the string, then the result is a pair (LHS, RHS)
337 /// such that (*this == LHS + Separator + RHS) is true and RHS is
338 /// maximal. If \arg Separator is not in the string, then the result is a
339 /// pair (LHS, RHS) where (*this == LHS) and (RHS == "").
341 /// \param Separator - The string to split on.
342 /// \return - The split substrings.
343 std::pair<StringRef, StringRef> split(StringRef Separator) const {
344 size_t Idx = find(Separator);
346 return std::make_pair(*this, StringRef());
347 return std::make_pair(slice(0, Idx), slice(Idx + Separator.size(), npos));
350 /// split - Split into substrings around the occurences of a separator
353 /// Each substring is stored in \arg A. If \arg MaxSplit is >= 0, at most
354 /// \arg MaxSplit splits are done and consequently <= \arg MaxSplit
355 /// elements are added to A.
356 /// If \arg KeepEmpty is false, empty strings are not added to \arg A. They
357 /// still count when considering \arg MaxSplit
358 /// An useful invariant is that
359 /// Separator.join(A) == *this if MaxSplit == -1 and KeepEmpty == true
361 /// \param A - Where to put the substrings.
362 /// \param Separator - The string to split on.
363 /// \param MaxSplit - The maximum number of times the string is split.
364 /// \parm KeepEmpty - True if empty substring should be added.
365 void split(SmallVectorImpl<StringRef> &A,
366 StringRef Separator, int MaxSplit = -1,
367 bool KeepEmpty = true) const;
369 /// rsplit - Split into two substrings around the last occurence of a
370 /// separator character.
372 /// If \arg Separator is in the string, then the result is a pair (LHS, RHS)
373 /// such that (*this == LHS + Separator + RHS) is true and RHS is
374 /// minimal. If \arg Separator is not in the string, then the result is a
375 /// pair (LHS, RHS) where (*this == LHS) and (RHS == "").
377 /// \param Separator - The character to split on.
378 /// \return - The split substrings.
379 std::pair<StringRef, StringRef> rsplit(char Separator) const {
380 size_t Idx = rfind(Separator);
382 return std::make_pair(*this, StringRef());
383 return std::make_pair(slice(0, Idx), slice(Idx+1, npos));
389 /// @name StringRef Comparison Operators
392 inline bool operator==(StringRef LHS, StringRef RHS) {
393 return LHS.equals(RHS);
396 inline bool operator!=(StringRef LHS, StringRef RHS) {
397 return !(LHS == RHS);
400 inline bool operator<(StringRef LHS, StringRef RHS) {
401 return LHS.compare(RHS) == -1;
404 inline bool operator<=(StringRef LHS, StringRef RHS) {
405 return LHS.compare(RHS) != 1;
408 inline bool operator>(StringRef LHS, StringRef RHS) {
409 return LHS.compare(RHS) == 1;
412 inline bool operator>=(StringRef LHS, StringRef RHS) {
413 return LHS.compare(RHS) != -1;