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#include "unicode/uobject.h"
32#include "unicode/std_string.h"
33
34// Arghh!  I wish C++ literals were "string".
35
36U_NAMESPACE_BEGIN
37
38/**
39 * A string-like object that points to a sized piece of memory.
40 *
41 * We provide non-explicit singleton constructors so users can pass
42 * in a "const char*" or a "string" wherever a "StringPiece" is
43 * expected.
44 *
45 * Functions or methods may use StringPiece parameters to accept either a
46 * "const char*" or a "string" value that will be implicitly converted to a
47 * StringPiece.
48 *
49 * Systematic usage of StringPiece is encouraged as it will reduce unnecessary
50 * conversions from "const char*" to "string" and back again.
51 *
52 * @stable ICU 4.2
53 */
54class U_COMMON_API StringPiece : public UMemory {
55 private:
56  const char*   ptr_;
57  int32_t       length_;
58
59 public:
60  /**
61   * Default constructor, creates an empty StringPiece.
62   * @stable ICU 4.2
63   */
64  StringPiece() : ptr_(NULL), length_(0) { }
65  /**
66   * Constructs from a NUL-terminated const char * pointer.
67   * @param str a NUL-terminated const char * pointer
68   * @stable ICU 4.2
69   */
70  StringPiece(const char* str);
71  /**
72   * Constructs from a std::string.
73   * @stable ICU 4.2
74   */
75  StringPiece(const std::string& str)
76    : ptr_(str.data()), length_(static_cast<int32_t>(str.size())) { }
77  /**
78   * Constructs from a const char * pointer and a specified length.
79   * @param offset a const char * pointer (need not be terminated)
80   * @param len the length of the string; must be non-negative
81   * @stable ICU 4.2
82   */
83  StringPiece(const char* offset, int32_t len) : ptr_(offset), length_(len) { }
84  /**
85   * Substring of another StringPiece.
86   * @param x the other StringPiece
87   * @param pos start position in x; must be non-negative and <= x.length().
88   * @stable ICU 4.2
89   */
90  StringPiece(const StringPiece& x, int32_t pos);
91  /**
92   * Substring of another StringPiece.
93   * @param x the other StringPiece
94   * @param pos start position in x; must be non-negative and <= x.length().
95   * @param len length of the substring;
96   *            must be non-negative and will be pinned to at most x.length() - pos.
97   * @stable ICU 4.2
98   */
99  StringPiece(const StringPiece& x, int32_t pos, int32_t len);
100
101  /**
102   * Returns the string pointer. May be NULL if it is empty.
103   *
104   * data() may return a pointer to a buffer with embedded NULs, and the
105   * returned buffer may or may not be null terminated.  Therefore it is
106   * typically a mistake to pass data() to a routine that expects a NUL
107   * terminated string.
108   * @return the string pointer
109   * @stable ICU 4.2
110   */
111  const char* data() const { return ptr_; }
112  /**
113   * Returns the string length. Same as length().
114   * @return the string length
115   * @stable ICU 4.2
116   */
117  int32_t size() const { return length_; }
118  /**
119   * Returns the string length. Same as size().
120   * @return the string length
121   * @stable ICU 4.2
122   */
123  int32_t length() const { return length_; }
124  /**
125   * Returns whether the string is empty.
126   * @return TRUE if the string is empty
127   * @stable ICU 4.2
128   */
129  UBool empty() const { return length_ == 0; }
130
131  /**
132   * Sets to an empty string.
133   * @stable ICU 4.2
134   */
135  void clear() { ptr_ = NULL; length_ = 0; }
136
137  /**
138   * Reset the stringpiece to refer to new data.
139   * @param xdata pointer the new string data.  Need not be nul terminated.
140   * @param len the length of the new data
141   * @stable ICU 4.8
142   */
143  void set(const char* xdata, int32_t len) { ptr_ = xdata; length_ = len; }
144
145  /**
146   * Reset the stringpiece to refer to new data.
147   * @param str a pointer to a NUL-terminated string.
148   * @stable ICU 4.8
149   */
150  void set(const char* str);
151
152  /**
153   * Removes the first n string units.
154   * @param n prefix length, must be non-negative and <=length()
155   * @stable ICU 4.2
156   */
157  void remove_prefix(int32_t n) {
158    if (n >= 0) {
159      if (n > length_) {
160        n = length_;
161      }
162      ptr_ += n;
163      length_ -= n;
164    }
165  }
166
167  /**
168   * Removes the last n string units.
169   * @param n suffix length, must be non-negative and <=length()
170   * @stable ICU 4.2
171   */
172  void remove_suffix(int32_t n) {
173    if (n >= 0) {
174      if (n <= length_) {
175        length_ -= n;
176      } else {
177        length_ = 0;
178      }
179    }
180  }
181
182  /**
183   * Maximum integer, used as a default value for substring methods.
184   * @stable ICU 4.2
185   */
186  static const int32_t npos; // = 0x7fffffff;
187
188  /**
189   * Returns a substring of this StringPiece.
190   * @param pos start position; must be non-negative and <= length().
191   * @param len length of the substring;
192   *            must be non-negative and will be pinned to at most length() - pos.
193   * @return the substring StringPiece
194   * @stable ICU 4.2
195   */
196  StringPiece substr(int32_t pos, int32_t len = npos) const {
197    return StringPiece(*this, pos, len);
198  }
199};
200
201/**
202 * Global operator == for StringPiece
203 * @param x The first StringPiece to compare.
204 * @param y The second StringPiece to compare.
205 * @return TRUE if the string data is equal
206 * @stable ICU 4.8
207 */
208U_EXPORT UBool U_EXPORT2
209operator==(const StringPiece& x, const StringPiece& y);
210
211/**
212 * Global operator != for StringPiece
213 * @param x The first StringPiece to compare.
214 * @param y The second StringPiece to compare.
215 * @return TRUE if the string data is not equal
216 * @stable ICU 4.8
217 */
218inline UBool operator!=(const StringPiece& x, const StringPiece& y) {
219  return !(x == y);
220}
221
222U_NAMESPACE_END
223
224#endif  // __STRINGPIECE_H__
225