一、移除性算法 (remove)
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 | // TEMPLATE FUNCTION remove_copy template < class _InIt, class _OutIt, class _Ty > inline _OutIt _Remove_copy(_InIt _First, _InIt _Last, _OutIt _Dest, const _Ty &_Val, _Range_checked_iterator_tag) { // copy omitting each matching _Val _DEBUG_RANGE(_First, _Last); _DEBUG_POINTER(_Dest); for (; _First != _Last; ++_First) if (!(*_First == _Val)) *_Dest++ = *_First; return (_Dest); } template < class _InIt, class _OutIt, class _Ty > inline _OutIt unchecked_remove_copy(_InIt _First, _InIt _Last, _OutIt _Dest, const _Ty &_Val) { // copy omitting each matching _Val return _STD _Remove_copy(_CHECKED_BASE(_First), _CHECKED_BASE(_Last), _Dest, _Val, _STD _Range_checked_iterator_tag()); } // TEMPLATE FUNCTION remove template < class _FwdIt, class _Ty > inline _FwdIt remove(_FwdIt _First, _FwdIt _Last, const _Ty &_Val) { // remove each matching _Val _First = find(_First, _Last, _Val); if (_First == _Last) return (_First); // empty sequence, all done else { // nonempty sequence, worth doing _FwdIt _First1 = _First; return (_STDEXT unchecked_remove_copy(++_First1, _Last, _First, _Val)); } } |
如下图所示:
假设现在想要remove 的元素是3,则传入到 _Remove_copy 函数的3个参数如上图第一行所示,Val 即3。
接着遍历First ~ Last 区间的元素,将非移除元素拷贝到前面,覆盖前面的元素,最后的指向如图,函数返回的是Dest 位置,如下代
码所示:
for (; _First != _Last; ++_First)
if (!(*_First == _Val))
*_Dest++ = *_First;
由上图可看出移除性算法并没有改变元素的个数,如果要真正删除,可以将remove 的返回值传入erase 进行删除,如:
v.erase(remove(v.begin(), v.end(), 3), v.end()); 即会将后面两个元素4 和 5 删除掉。
示例代码1:
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 | #include <iostream> #include <vector> #include <list> #include <algorithm> using namespace std; void print_element( int n) { cout << n << ' '; } int main( void) { int a[] = { 1, 3, 2, 3, 4, 5 }; vector< int> v(a, a + 6); for_each(v.begin(), v.end(), print_element); cout << endl; /*remove(v.begin(), v.end(), 3); for_each(v.begin(), v.end(), print_element); cout<<endl;*/ v.erase(remove(v.begin(), v.end(), 3), v.end()); for_each(v.begin(), v.end(), print_element); cout << endl; return 0; } |
二、变序性算法( rotate)
1 2 3 4 5 6 7 | template< class _FwdIt> inline void rotate(_FwdIt _First, _FwdIt _Mid, _FwdIt _Last) { // rotate [_First, _Last) if (_First != _Mid && _Mid != _Last) _Rotate(_CHECKED_BASE(_First), _CHECKED_BASE(_Mid), _CHECKED_BASE(_Last), _Iter_cat(_First)); } |
rotate 调用了_Rotate,实际上_Rotate 继续调用了某个函数,内部实现代码比较长,而且不容易看懂,这边可以看一下简易的等价
版本实现,来自这里,如下:
1 2 3 4 5 6 7 8 9 10 11 12 | template < class ForwardIterator> void rotate (ForwardIterator first, ForwardIterator middle, ForwardIterator last) { ForwardIterator next = middle; while (first != next) { swap (*first++, *next++); if (next == last) next = middle; else if (first == middle) middle = next; } } |
假设一个容器有 1 2 3 4 5 6 六个元素,现在想把 1 2 放到后面去,可以这样写 rotate(v.begin(), v.begin()+2, v.end()); 如下图所示:
即将first 与 next 对应的元素互换且不断向前推进,直到first == next 为止。
三、排序算法 (sort)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 | template< class _RanIt> inline void sort(_RanIt _First, _RanIt _Last) { // order [_First, _Last), using operator< _DEBUG_RANGE(_First, _Last); std::_Sort(_CHECKED_BASE(_First), _CHECKED_BASE(_Last), _Last - _First); } template < class _RanIt, class _Pr > inline void sort(_RanIt _First, _RanIt _Last, _Pr _Pred) { // order [_First, _Last), using _Pred _DEBUG_RANGE(_First, _Last); _DEBUG_POINTER(_Pred); std::_Sort(_CHECKED_BASE(_First), _CHECKED_BASE(_Last), _Last - _First, _Pred); } |
sort 重载了两个版本,第一个版本只有2个参数,默认按从小到大排序(using operator<)
;第二个版本有三个参数,即可以自定义比较逻辑
(_Pred)。它们都调用了标准库的std::_Sort, 跟踪进去发现比较复杂,在_Sort 内会根据一些条件选择不同的排序方式,如标准库的堆排序,合并
排序,插入排序等等。站在使用的角度看,没必要去深究,但如果是想学习相关的排序,那是很好的资源。
示例代码2:
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 | #include <iostream> #include <vector> #include <list> #include <algorithm> using namespace std; void print_element( int n) { cout << n << ' '; } bool my_greater( int a, int b) { return a > b; } int main( void) { int a[] = { 1, 2, 3, 4, 5, 6 }; vector< int> v(a, a + 6); for_each(v.begin(), v.end(), print_element); cout << endl; rotate(v.begin(), v.begin() + 2, v.end()); for_each(v.begin(), v.end(), print_element); cout << endl; sort(v.begin(), v.end()); for_each(v.begin(), v.end(), print_element); cout << endl; sort(v.begin(), v.end(), my_greater); for_each(v.begin(), v.end(), print_element); cout << endl; return 0; } |
四、已序区间算法 (lower_bound 、upper_bound)
使用这些算法的前提是区间已经是有序的。
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 | // TEMPLATE FUNCTION lower_bound template < class _FwdIt, class _Ty, class _Diff > inline _FwdIt _Lower_bound(_FwdIt _First, _FwdIt _Last, const _Ty &_Val, _Diff *) { // find first element not before _Val, using operator< _DEBUG_ORDER_SINGLE(_First, _Last, true); _Diff _Count = 0; _Distance(_First, _Last, _Count); for (; 0 < _Count; ) { // divide and conquer, find half that contains answer _Diff _Count2 = _Count / 2; _FwdIt _Mid = _First; std::advance(_Mid, _Count2); _DEBUG_ORDER_SINGLE(_Mid, _Last, false); if (_DEBUG_LT(*_Mid, _Val)) _First = ++_Mid, _Count -= _Count2 + 1; else _Count = _Count2; } return (_First); } template < class _FwdIt, class _Ty > inline _FwdIt lower_bound(_FwdIt _First, _FwdIt _Last, const _Ty &_Val) { // find first element not before _Val, using operator< _ASSIGN_FROM_BASE(_First, _Lower_bound(_CHECKED_BASE(_First), _CHECKED_BASE(_Last), _Val, _Dist_type(_First))); return _First; } |
lower_bound() 返回第一个“大于等于给定值" 的元素位置,其实还重载了另一个low_bound 版本,如下:
1 2 3 4 5 6 7 | // TEMPLATE FUNCTION lower_bound WITH PRED template < class _FwdIt, class _Ty, class _Diff, class _Pr > inline _FwdIt _Lower_bound(_FwdIt _First, _FwdIt _Last, const _Ty &_Val, _Pr _Pred, _Diff *) |
也就是可以自定义比较逻辑,需要注意的是如果使用这个版本,那么区间应该本来就是按comp 方法排序的,如下面这句话:
The elements are compared using operator<
for the first version, and comp for the second. The elements in the range shall already
be sorted according to this same criterion (operator<
or comp), or at least partitioned with respect to val.
由于是已序区间,所以函数内用的是二分查找,而两个版本主要的代码不同在于:
_DEBUG_LT(*_Mid, _Val)
_DEBUG_LT_PRED(_Pred, *_Mid, _Val)
upper_bound 与 lower_bound 类似,不过返回的是第一个”大于给定值“ 的元素位置。
示例代码3:
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 | #include <iostream> #include <vector> #include <list> #include <algorithm> using namespace std; void print_element( int n) { cout << n << ' '; } int main( void) { int a[] = { 1, 10, 10, 14, 15, 16 }; vector< int> v(a, a + 6); for_each(v.begin(), v.end(), print_element); cout << endl; vector< int>::iterator it; it = lower_bound(v.begin(), v.end(), 10); if (it != v.end()) { cout << it - v.begin() << endl; } it = upper_bound(v.begin(), v.end(), 10); if (it != v.end()) { cout << it - v.begin() << endl; } return 0; } |
五、数值算法(accumulate)
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 | // TEMPLATE FUNCTION accumulate template < class _InIt, class _Ty > inline _Ty _Accumulate(_InIt _First, _InIt _Last, _Ty _Val) { // return sum of _Val and all in [_First, _Last) _DEBUG_RANGE(_First, _Last); for (; _First != _Last; ++_First) _Val = _Val + *_First; return (_Val); } template < class _InIt, class _Ty > inline _Ty accumulate(_InIt _First, _InIt _Last, _Ty _Val) { // return sum of _Val and all in [_First, _Last) return _Accumulate(_CHECKED_BASE(_First), _CHECKED_BASE(_Last), _Val); } // TEMPLATE FUNCTION accumulate WITH BINOP template < class _InIt, class _Ty, class _Fn2 > inline _Ty _Accumulate(_InIt _First, _InIt _Last, _Ty _Val, _Fn2 _Func) { // return sum of _Val and all in [_First, _Last), using _Func _DEBUG_RANGE(_First, _Last); _DEBUG_POINTER(_Func); for (; _First != _Last; ++_First) _Val = _Func(_Val, *_First); return (_Val); } template < class _InIt, class _Ty, class _Fn2 > inline _Ty accumulate(_InIt _First, _InIt _Last, _Ty _Val, _Fn2 _Func) { // return sum of _Val and all in [_First, _Last), using _Func return _Accumulate(_CHECKED_BASE(_First), _CHECKED_BASE(_Last), _Val, _Func); } |
accumulate 重载了两个版本,第一个版本实现的是累加,第二个版本带_Func 参数,可以自定义计算,比如累乘等。代码都比较好理解,就不赘述
了。看下面的示例代码4:
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 | #include <iostream> #include <vector> #include <list> #include <algorithm> #include <numeric> using namespace std; void print_element( int n) { cout << n << ' '; } int mult( int a, int b) { return a * b; } int main( void) { int a[] = { 1, 2, 3, 4, 5 }; vector< int> v(a, a + 5); for_each(v.begin(), v.end(), print_element); cout << endl; // 累加 cout << accumulate(v.begin(), v.end(), 0) << endl; // 累乘 cout << accumulate(v.begin(), v.end(), 1, mult) << endl; return 0; } |
参考:
C++ primer 第四版
Effective C++ 3rd
C++编程规范