1// © 2016 and later: Unicode, Inc. and others.
2// License & terms of use: http://www.unicode.org/copyright.html
3// Copyright (C) 2009-2013, International Business Machines
4// Corporation and others. All Rights Reserved.
5//
6// Copyright 2001 and onwards Google Inc.
7// Author: Sanjay Ghemawat
8
9// This code is a contribution of Google code, and the style used here is
10// a compromise between the original Google code and the ICU coding guidelines.
11// For example, data types are ICU-ified (size_t,int->int32_t),
12// and API comments doxygen-ified, but function names and behavior are
13// as in the original, if possible.
14// Assertion-style error handling, not available in ICU, was changed to
15// parameter "pinning" similar to UnicodeString.
16//
17// In addition, this is only a partial port of the original Google code,
18// limited to what was needed so far. The (nearly) complete original code
19// is in the ICU svn repository at icuhtml/trunk/design/strings/contrib
20// (see ICU ticket 6765, r25517).
21
22#ifndef __STRINGPIECE_H__
23#define __STRINGPIECE_H__
24
25/**
26 * \file
27 * \brief C++ API: StringPiece: Read-only byte string wrapper class.
28 */
29
30#include "unicode/utypes.h"
31
32#if U_SHOW_CPLUSPLUS_API
33
34#include <cstddef>
35#include <type_traits>
36
37#include "unicode/uobject.h"
38#include "unicode/std_string.h"
39
40// Arghh! I wish C++ literals were "string".
41
42U_NAMESPACE_BEGIN
43
44/**
45 * A string-like object that points to a sized piece of memory.
46 *
47 * We provide non-explicit singleton constructors so users can pass
48 * in a "const char*" or a "string" wherever a "StringPiece" is
49 * expected.
50 *
51 * Functions or methods may use StringPiece parameters to accept either a
52 * "const char*" or a "string" value that will be implicitly converted to a
53 * StringPiece.
54 *
55 * Systematic usage of StringPiece is encouraged as it will reduce unnecessary
56 * conversions from "const char*" to "string" and back again.
57 *
58 * @stable ICU 4.2
59 */
60class U_COMMON_API StringPiece : public UMemory {
61 private:
62 const char* ptr_;
63 int32_t length_;
64
65 public:
66 /**
67 * Default constructor, creates an empty StringPiece.
68 * @stable ICU 4.2
69 */
70 StringPiece() : ptr_(NULL), length_(0) { }
71 /**
72 * Constructs from a NUL-terminated const char * pointer.
73 * @param str a NUL-terminated const char * pointer
74 * @stable ICU 4.2
75 */
76 StringPiece(const char* str);
77 /**
78 * Constructs from a std::string.
79 * @stable ICU 4.2
80 */
81 StringPiece(const std::string& str)
82 : ptr_(str.data()), length_(static_cast<int32_t>(str.size())) { }
83#ifndef U_HIDE_DRAFT_API
84 /**
85 * Constructs from some other implementation of a string piece class, from any
86 * C++ record type that has these two methods:
87 *
88 * \code{.cpp}
89 *
90 * struct OtherStringPieceClass {
91 * const char* data();
92 * size_t size();
93 * };
94 *
95 * \endcode
96 *
97 * The other string piece class will typically be std::string_view from C++17
98 * or absl::string_view from Abseil.
99 *
100 * @param str the other string piece
101 * @draft ICU 65
102 */
103 template <typename T,
104 typename = typename std::enable_if<
105 std::is_same<decltype(T().data()), const char*>::value &&
106 std::is_same<decltype(T().size()), size_t>::value>::type>
107 StringPiece(T str)
108 : ptr_(str.data()), length_(static_cast<int32_t>(str.size())) {}
109#endif // U_HIDE_DRAFT_API
110 /**
111 * Constructs from a const char * pointer and a specified length.
112 * @param offset a const char * pointer (need not be terminated)
113 * @param len the length of the string; must be non-negative
114 * @stable ICU 4.2
115 */
116 StringPiece(const char* offset, int32_t len) : ptr_(offset), length_(len) { }
117 /**
118 * Substring of another StringPiece.
119 * @param x the other StringPiece
120 * @param pos start position in x; must be non-negative and <= x.length().
121 * @stable ICU 4.2
122 */
123 StringPiece(const StringPiece& x, int32_t pos);
124 /**
125 * Substring of another StringPiece.
126 * @param x the other StringPiece
127 * @param pos start position in x; must be non-negative and <= x.length().
128 * @param len length of the substring;
129 * must be non-negative and will be pinned to at most x.length() - pos.
130 * @stable ICU 4.2
131 */
132 StringPiece(const StringPiece& x, int32_t pos, int32_t len);
133
134 /**
135 * Returns the string pointer. May be NULL if it is empty.
136 *
137 * data() may return a pointer to a buffer with embedded NULs, and the
138 * returned buffer may or may not be null terminated. Therefore it is
139 * typically a mistake to pass data() to a routine that expects a NUL
140 * terminated string.
141 * @return the string pointer
142 * @stable ICU 4.2
143 */
144 const char* data() const { return ptr_; }
145 /**
146 * Returns the string length. Same as length().
147 * @return the string length
148 * @stable ICU 4.2
149 */
150 int32_t size() const { return length_; }
151 /**
152 * Returns the string length. Same as size().
153 * @return the string length
154 * @stable ICU 4.2
155 */
156 int32_t length() const { return length_; }
157 /**
158 * Returns whether the string is empty.
159 * @return TRUE if the string is empty
160 * @stable ICU 4.2
161 */
162 UBool empty() const { return length_ == 0; }
163
164 /**
165 * Sets to an empty string.
166 * @stable ICU 4.2
167 */
168 void clear() { ptr_ = NULL; length_ = 0; }
169
170 /**
171 * Reset the stringpiece to refer to new data.
172 * @param xdata pointer the new string data. Need not be nul terminated.
173 * @param len the length of the new data
174 * @stable ICU 4.8
175 */
176 void set(const char* xdata, int32_t len) { ptr_ = xdata; length_ = len; }
177
178 /**
179 * Reset the stringpiece to refer to new data.
180 * @param str a pointer to a NUL-terminated string.
181 * @stable ICU 4.8
182 */
183 void set(const char* str);
184
185 /**
186 * Removes the first n string units.
187 * @param n prefix length, must be non-negative and <=length()
188 * @stable ICU 4.2
189 */
190 void remove_prefix(int32_t n) {
191 if (n >= 0) {
192 if (n > length_) {
193 n = length_;
194 }
195 ptr_ += n;
196 length_ -= n;
197 }
198 }
199
200 /**
201 * Removes the last n string units.
202 * @param n suffix length, must be non-negative and <=length()
203 * @stable ICU 4.2
204 */
205 void remove_suffix(int32_t n) {
206 if (n >= 0) {
207 if (n <= length_) {
208 length_ -= n;
209 } else {
210 length_ = 0;
211 }
212 }
213 }
214
215 /**
216 * Maximum integer, used as a default value for substring methods.
217 * @stable ICU 4.2
218 */
219 static const int32_t npos; // = 0x7fffffff;
220
221 /**
222 * Returns a substring of this StringPiece.
223 * @param pos start position; must be non-negative and <= length().
224 * @param len length of the substring;
225 * must be non-negative and will be pinned to at most length() - pos.
226 * @return the substring StringPiece
227 * @stable ICU 4.2
228 */
229 StringPiece substr(int32_t pos, int32_t len = npos) const {
230 return StringPiece(*this, pos, len);
231 }
232};
233
234/**
235 * Global operator == for StringPiece
236 * @param x The first StringPiece to compare.
237 * @param y The second StringPiece to compare.
238 * @return TRUE if the string data is equal
239 * @stable ICU 4.8
240 */
241U_EXPORT UBool U_EXPORT2
242operator==(const StringPiece& x, const StringPiece& y);
243
244/**
245 * Global operator != for StringPiece
246 * @param x The first StringPiece to compare.
247 * @param y The second StringPiece to compare.
248 * @return TRUE if the string data is not equal
249 * @stable ICU 4.8
250 */
251inline UBool operator!=(const StringPiece& x, const StringPiece& y) {
252 return !(x == y);
253}
254
255U_NAMESPACE_END
256
257#endif /* U_SHOW_CPLUSPLUS_API */
258
259#endif // __STRINGPIECE_H__
260