summaryrefslogtreecommitdiffstats
path: root/xpcom/string/nsTPromiseFlatString.h
blob: 0033d074ce5c5dad30297883e99d49e2d96a503e (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
/* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* vim: set ts=8 sts=2 et sw=2 tw=80: */
/* This Source Code Form is subject to the terms of the Mozilla Public
 * License, v. 2.0. If a copy of the MPL was not distributed with this
 * file, You can obtain one at http://mozilla.org/MPL/2.0/. */

#ifndef nsTPromiseFlatString_h
#define nsTPromiseFlatString_h

#include "mozilla/Attributes.h"
#include "nsTString.h"

/**
 * NOTE:
 *
 * Try to avoid flat strings.  |PromiseFlat[C]String| will help you as a last
 * resort, and this may be necessary when dealing with legacy or OS calls,
 * but in general, requiring a null-terminated array of characters kills many
 * of the performance wins the string classes offer.  Write your own code to
 * use |nsA[C]String&|s for parameters.  Write your string proccessing
 * algorithms to exploit iterators.  If you do this, you will benefit from
 * being able to chain operations without copying or allocating and your code
 * will be significantly more efficient.  Remember, a function that takes an
 * |const nsA[C]String&| can always be passed a raw character pointer by
 * wrapping it (for free) in a |nsDependent[C]String|.  But a function that
 * takes a character pointer always has the potential to force allocation and
 * copying.
 *
 *
 * How to use it:
 *
 * A |nsPromiseFlat[C]String| doesn't necessarily own the characters it
 * promises.  You must never use it to promise characters out of a string
 * with a shorter lifespan.  The typical use will be something like this:
 *
 *   SomeOSFunction( PromiseFlatCString(aCSubstring).get() ); // GOOD
 *
 * Here's a BAD use:
 *
 *  const char* buffer = PromiseFlatCString(aCSubstring).get();
 *  SomeOSFunction(buffer); // BAD!! |buffer| is a dangling pointer
 *
 * The only way to make one is with the function |PromiseFlat[C]String|,
 * which produce a |const| instance.  ``What if I need to keep a promise
 * around for a little while?'' you might ask.  In that case, you can keep a
 * reference, like so:
 *
 *   const nsCString& flat = PromiseFlatString(aCSubstring);
 *     // Temporaries usually die after the full expression containing the
 *     // expression that created the temporary is evaluated.  But when a
 *     // temporary is assigned to a local reference, the temporary's lifetime
 *     // is extended to the reference's lifetime (C++11 [class.temporary]p5).
 *     //
 *     // This reference holds the anonymous temporary alive.  But remember: it
 *     // must _still_ have a lifetime shorter than that of |aCSubstring|, and
 *     // |aCSubstring| must not be changed while the PromiseFlatString lives.
 *
 *  SomeOSFunction(flat.get());
 *  SomeOtherOSFunction(flat.get());
 *
 *
 * How does it work?
 *
 * A |nsPromiseFlat[C]String| is just a wrapper for another string.  If you
 * apply it to a string that happens to be flat, your promise is just a
 * dependent reference to the string's data.  If you apply it to a non-flat
 * string, then a temporary flat string is created for you, by allocating and
 * copying.  In the event that you end up assigning the result into a sharing
 * string (e.g., |nsTString|), the right thing happens.
 */

template <typename T>
class MOZ_STACK_CLASS nsTPromiseFlatString : public nsTString<T> {
 public:
  typedef nsTPromiseFlatString<T> self_type;
  typedef nsTString<T> base_string_type;
  typedef typename base_string_type::substring_type substring_type;
  typedef typename base_string_type::string_type string_type;
  typedef typename base_string_type::substring_tuple_type substring_tuple_type;
  typedef typename base_string_type::char_type char_type;
  typedef typename base_string_type::size_type size_type;

  // These are only for internal use within the string classes:
  typedef typename base_string_type::DataFlags DataFlags;
  typedef typename base_string_type::ClassFlags ClassFlags;

 private:
  void Init(const substring_type&);

  // NOT TO BE IMPLEMENTED
  void operator=(const self_type&) = delete;

  // NOT TO BE IMPLEMENTED
  nsTPromiseFlatString(const self_type&) = delete;

  // NOT TO BE IMPLEMENTED
  nsTPromiseFlatString() = delete;

  // NOT TO BE IMPLEMENTED
  nsTPromiseFlatString(const string_type& aStr) = delete;

 public:
  explicit nsTPromiseFlatString(const substring_type& aStr) : string_type() {
    Init(aStr);
  }

  explicit nsTPromiseFlatString(const substring_tuple_type& aTuple)
      : string_type() {
    // nothing else to do here except assign the value of the tuple
    // into ourselves.
    this->Assign(aTuple);
  }
};

extern template class nsTPromiseFlatString<char>;
extern template class nsTPromiseFlatString<char16_t>;

// We template this so that the constructor is chosen based on the type of the
// parameter. This allows us to reject attempts to promise a flat flat string.
template <class T>
const nsTPromiseFlatString<T> TPromiseFlatString(
    const typename nsTPromiseFlatString<T>::substring_type& aString) {
  return nsTPromiseFlatString<T>(aString);
}

template <class T>
const nsTPromiseFlatString<T> TPromiseFlatString(
    const typename nsTPromiseFlatString<T>::substring_tuple_type& aString) {
  return nsTPromiseFlatString<T>(aString);
}

#ifndef PromiseFlatCString
#  define PromiseFlatCString TPromiseFlatString<char>
#endif

#ifndef PromiseFlatString
#  define PromiseFlatString TPromiseFlatString<char16_t>
#endif

#endif