From Jason Turner

[template.valarray]

C++20 → C++23 2.9 KB 22 lines
Current Draft | LaTeX Source
Archived: C++20 | C++23
Markdown: C++20 | C++23
Explore in Adventure Game
C++ Weekly: Ep 367

Diff to HTML by rtfpessoa

Files changed (1) hide show
  1. tmp/tmpc4n8h5hm/{from.md → to.md} +8 -8
tmp/tmpc4n8h5hm/{from.md → to.md} RENAMED
@@ -100,19 +100,19 @@ elements numbered sequentially from zero. It is a representation of the
100
  mathematical concept of an ordered set of values. For convenience, an
101
  object of type `valarray<T>` is referred to as an “array” throughout the
102
  remainder of  [[numarray]]. The illusion of higher dimensionality may be
103
  produced by the familiar idiom of computed indices, together with the
104
  powerful subsetting capabilities provided by the generalized subscript
105
- operators.[^8]
106
 
107
  #### Constructors <a id="valarray.cons">[[valarray.cons]]</a>
108
 
109
  ``` cpp
110
  valarray();
111
  ```
112
 
113
- *Effects:* Constructs a `valarray` that has zero length.[^9]
114
 
115
  ``` cpp
116
  explicit valarray(size_t n);
117
  ```
118
 
@@ -132,19 +132,19 @@ valarray(const T* p, size_t n);
132
 
133
  *Preconditions:* \[`p`, `p + n`) is a valid range.
134
 
135
  *Effects:* Constructs a `valarray` that has length `n`. The values of
136
  the elements of the array are initialized with the first `n` values
137
- pointed to by the first argument.[^10]
138
 
139
  ``` cpp
140
  valarray(const valarray& v);
141
  ```
142
 
143
  *Effects:* Constructs a `valarray` that has the same length as `v`. The
144
  elements are initialized with the values of the corresponding elements
145
- of `v`.[^11]
146
 
147
  ``` cpp
148
  valarray(valarray&& v) noexcept;
149
  ```
150
 
@@ -255,12 +255,12 @@ evaluates to `true` for all `size_t i` and `size_t j` such that
255
 
256
  The expression `addressof(a[i]) != addressof(b[j])` evaluates to `true`
257
  for any two arrays `a` and `b` and for any `size_t i` and `size_t j`
258
  such that `i < a.size()` and `j < b.size()`.
259
 
260
- [*Note 2*: This property indicates an absence of aliasing and may be
261
- used to advantage by optimizing compilers. Compilers may take advantage
262
  of inlining, constant propagation, loop fusion, tracking of pointers
263
  obtained from `operator new`, and other techniques to generate efficient
264
  `valarray`s. — *end note*]
265
 
266
  The reference returned by the subscript operator for an array shall be
@@ -568,12 +568,12 @@ is `(*this)[`*`I`*` + n]` if *`I`*` + n` is non-negative and less than
568
  value of `n` shifts the elements left `n` places, with zero
569
  fill. — *end note*]
570
 
571
  [*Example 1*: If the argument has the value -2, the first two elements
572
  of the result will be value-initialized [[dcl.init]]; the third element
573
- of the result will be assigned the value of the first element of the
574
- argument; etc. — *end example*]
575
 
576
  ``` cpp
577
  valarray cshift(int n) const;
578
  ```
579
 
 
100
  mathematical concept of an ordered set of values. For convenience, an
101
  object of type `valarray<T>` is referred to as an “array” throughout the
102
  remainder of  [[numarray]]. The illusion of higher dimensionality may be
103
  produced by the familiar idiom of computed indices, together with the
104
  powerful subsetting capabilities provided by the generalized subscript
105
+ operators.[^9]
106
 
107
  #### Constructors <a id="valarray.cons">[[valarray.cons]]</a>
108
 
109
  ``` cpp
110
  valarray();
111
  ```
112
 
113
+ *Effects:* Constructs a `valarray` that has zero length.[^10]
114
 
115
  ``` cpp
116
  explicit valarray(size_t n);
117
  ```
118
 
 
132
 
133
  *Preconditions:* \[`p`, `p + n`) is a valid range.
134
 
135
  *Effects:* Constructs a `valarray` that has length `n`. The values of
136
  the elements of the array are initialized with the first `n` values
137
+ pointed to by the first argument.[^11]
138
 
139
  ``` cpp
140
  valarray(const valarray& v);
141
  ```
142
 
143
  *Effects:* Constructs a `valarray` that has the same length as `v`. The
144
  elements are initialized with the values of the corresponding elements
145
+ of `v`.[^12]
146
 
147
  ``` cpp
148
  valarray(valarray&& v) noexcept;
149
  ```
150
 
 
255
 
256
  The expression `addressof(a[i]) != addressof(b[j])` evaluates to `true`
257
  for any two arrays `a` and `b` and for any `size_t i` and `size_t j`
258
  such that `i < a.size()` and `j < b.size()`.
259
 
260
+ [*Note 2*: This property indicates an absence of aliasing and can be
261
+ used to advantage by optimizing compilers. Compilers can take advantage
262
  of inlining, constant propagation, loop fusion, tracking of pointers
263
  obtained from `operator new`, and other techniques to generate efficient
264
  `valarray`s. — *end note*]
265
 
266
  The reference returned by the subscript operator for an array shall be
 
568
  value of `n` shifts the elements left `n` places, with zero
569
  fill. — *end note*]
570
 
571
  [*Example 1*: If the argument has the value -2, the first two elements
572
  of the result will be value-initialized [[dcl.init]]; the third element
573
+ of the result will be assigned the value of the first element of
574
+ `*this`; etc. — *end example*]
575
 
576
  ``` cpp
577
  valarray cshift(int n) const;
578
  ```
579