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 /// str - Get the contents as an std::string.
137 std::string str() const { return std::string(Data, Length); }
140 /// @name Operator Overloads
143 char operator[](size_t Index) const {
144 assert(Index < Length && "Invalid index!");
149 /// @name Type Conversions
152 operator std::string() const {
157 /// @name String Predicates
160 /// startswith - Check if this string starts with the given \arg Prefix.
161 bool startswith(StringRef Prefix) const {
162 return substr(0, Prefix.Length).equals(Prefix);
165 /// endswith - Check if this string ends with the given \arg Suffix.
166 bool endswith(StringRef Suffix) const {
167 return slice(size() - Suffix.Length, size()).equals(Suffix);
171 /// @name String Searching
174 /// find - Search for the first character \arg C in the string.
176 /// \return - The index of the first occurence of \arg C, or npos if not
178 size_t find(char C, size_t From = 0) const {
179 for (size_t i = min(From, Length), e = Length; i != e; ++i)
185 /// find - Search for the first string \arg Str in the string.
187 /// \return - The index of the first occurence of \arg Str, or npos if not
189 size_t find(StringRef Str, size_t From = 0) const;
191 /// rfind - Search for the last character \arg C in the string.
193 /// \return - The index of the last occurence of \arg C, or npos if not
195 size_t rfind(char C, size_t From = npos) const {
196 From = min(From, Length);
206 /// rfind - Search for the last string \arg Str in the string.
208 /// \return - The index of the last occurence of \arg Str, or npos if not
210 size_t rfind(StringRef Str) const;
212 /// find_first_of - Find the first character in the string that is \arg C,
213 /// or npos if not found. Same as find.
214 size_type find_first_of(char C, size_t = 0) const { return find(C); }
216 /// find_first_of - Find the first character in the string that is in \arg
217 /// Chars, or npos if not found.
219 /// Note: O(size() * Chars.size())
220 size_type find_first_of(StringRef Chars, size_t From = 0) const;
222 /// find_first_not_of - Find the first character in the string that is not
223 /// \arg C or npos if not found.
224 size_type find_first_not_of(char C, size_t From = 0) const;
226 /// find_first_not_of - Find the first character in the string that is not
227 /// in the string \arg Chars, or npos if not found.
229 /// Note: O(size() * Chars.size())
230 size_type find_first_not_of(StringRef Chars, size_t From = 0) const;
233 /// @name Helpful Algorithms
236 /// count - Return the number of occurrences of \arg C in the string.
237 size_t count(char C) const {
239 for (size_t i = 0, e = Length; i != e; ++i)
245 /// count - Return the number of non-overlapped occurrences of \arg Str in
247 size_t count(StringRef Str) const;
249 /// getAsInteger - Parse the current string as an integer of the specified
250 /// radix. If Radix is specified as zero, this does radix autosensing using
251 /// extended C rules: 0 is octal, 0x is hex, 0b is binary.
253 /// If the string is invalid or if only a subset of the string is valid,
254 /// this returns true to signify the error. The string is considered
255 /// erroneous if empty.
257 bool getAsInteger(unsigned Radix, long long &Result) const;
258 bool getAsInteger(unsigned Radix, unsigned long long &Result) const;
259 bool getAsInteger(unsigned Radix, int &Result) const;
260 bool getAsInteger(unsigned Radix, unsigned &Result) const;
262 // TODO: Provide overloads for int/unsigned that check for overflow.
265 /// @name Substring Operations
268 /// substr - Return a reference to the substring from [Start, Start + N).
270 /// \param Start - The index of the starting character in the substring; if
271 /// the index is npos or greater than the length of the string then the
272 /// empty substring will be returned.
274 /// \param N - The number of characters to included in the substring. If N
275 /// exceeds the number of characters remaining in the string, the string
276 /// suffix (starting with \arg Start) will be returned.
277 StringRef substr(size_t Start, size_t N = npos) const {
278 Start = min(Start, Length);
279 return StringRef(Data + Start, min(N, Length - Start));
282 /// slice - Return a reference to the substring from [Start, End).
284 /// \param Start - The index of the starting character in the substring; if
285 /// the index is npos or greater than the length of the string then the
286 /// empty substring will be returned.
288 /// \param End - The index following the last character to include in the
289 /// substring. If this is npos, or less than \arg Start, or exceeds the
290 /// number of characters remaining in the string, the string suffix
291 /// (starting with \arg Start) will be returned.
292 StringRef slice(size_t Start, size_t End) const {
293 Start = min(Start, Length);
294 End = min(max(Start, End), Length);
295 return StringRef(Data + Start, End - Start);
298 /// split - Split into two substrings around the first occurence of a
299 /// separator character.
301 /// If \arg Separator is in the string, then the result is a pair (LHS, RHS)
302 /// such that (*this == LHS + Separator + RHS) is true and RHS is
303 /// maximal. If \arg Separator is not in the string, then the result is a
304 /// pair (LHS, RHS) where (*this == LHS) and (RHS == "").
306 /// \param Separator - The character to split on.
307 /// \return - The split substrings.
308 std::pair<StringRef, StringRef> split(char Separator) const {
309 size_t Idx = find(Separator);
311 return std::make_pair(*this, StringRef());
312 return std::make_pair(slice(0, Idx), slice(Idx+1, npos));
315 /// split - Split into two substrings around the first occurence of a
316 /// separator string.
318 /// If \arg Separator is in the string, then the result is a pair (LHS, RHS)
319 /// such that (*this == LHS + Separator + RHS) is true and RHS is
320 /// maximal. If \arg Separator is not in the string, then the result is a
321 /// pair (LHS, RHS) where (*this == LHS) and (RHS == "").
323 /// \param Separator - The string to split on.
324 /// \return - The split substrings.
325 std::pair<StringRef, StringRef> split(StringRef Separator) const {
326 size_t Idx = find(Separator);
328 return std::make_pair(*this, StringRef());
329 return std::make_pair(slice(0, Idx), slice(Idx + Separator.size(), npos));
332 /// split - Split into substrings around the occurences of a separator
335 /// Each substring is stored in \arg A. If \arg MaxSplit is >= 0, at most
336 /// \arg MaxSplit splits are done and consequently <= \arg MaxSplit
337 /// elements are added to A.
338 /// If \arg KeepEmpty is false, empty strings are not added to \arg A. They
339 /// still count when considering \arg MaxSplit
340 /// An useful invariant is that
341 /// Separator.join(A) == *this if MaxSplit == -1 and KeepEmpty == true
343 /// \param A - Where to put the substrings.
344 /// \param Separator - The string to split on.
345 /// \param MaxSplit - The maximum number of times the string is split.
346 /// \parm KeepEmpty - True if empty substring should be added.
347 void split(SmallVectorImpl<StringRef> &A,
348 StringRef Separator, int MaxSplit = -1,
349 bool KeepEmpty = true) const;
351 /// rsplit - Split into two substrings around the last occurence of a
352 /// separator character.
354 /// If \arg Separator is in the string, then the result is a pair (LHS, RHS)
355 /// such that (*this == LHS + Separator + RHS) is true and RHS is
356 /// minimal. If \arg Separator is not in the string, then the result is a
357 /// pair (LHS, RHS) where (*this == LHS) and (RHS == "").
359 /// \param Separator - The character to split on.
360 /// \return - The split substrings.
361 std::pair<StringRef, StringRef> rsplit(char Separator) const {
362 size_t Idx = rfind(Separator);
364 return std::make_pair(*this, StringRef());
365 return std::make_pair(slice(0, Idx), slice(Idx+1, npos));
371 /// @name StringRef Comparison Operators
374 inline bool operator==(StringRef LHS, StringRef RHS) {
375 return LHS.equals(RHS);
378 inline bool operator!=(StringRef LHS, StringRef RHS) {
379 return !(LHS == RHS);
382 inline bool operator<(StringRef LHS, StringRef RHS) {
383 return LHS.compare(RHS) == -1;
386 inline bool operator<=(StringRef LHS, StringRef RHS) {
387 return LHS.compare(RHS) != 1;
390 inline bool operator>(StringRef LHS, StringRef RHS) {
391 return LHS.compare(RHS) == 1;
394 inline bool operator>=(StringRef LHS, StringRef RHS) {
395 return LHS.compare(RHS) != -1;