mini_lambda.cpp
来自「Boost provides free peer-reviewed portab」· C++ 代码 · 共 233 行
CPP
233 行
//[ Lambda///////////////////////////////////////////////////////////////////////////////// Copyright 2008 Eric Niebler. Distributed under the Boost// Software License, Version 1.0. (See accompanying file// LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)//// This example builds a simple but functional lambda library using Proto.#include <iostream>#include <algorithm>#include <boost/mpl/int.hpp>#include <boost/mpl/min_max.hpp>#include <boost/mpl/eval_if.hpp>#include <boost/mpl/identity.hpp>#include <boost/mpl/next_prior.hpp>#include <boost/fusion/tuple.hpp>#include <boost/typeof/typeof.hpp>#include <boost/typeof/std/ostream.hpp>#include <boost/typeof/std/iostream.hpp>#include <boost/proto/core.hpp>#include <boost/proto/context.hpp>#include <boost/proto/transform.hpp>namespace mpl = boost::mpl;namespace proto = boost::proto;namespace fusion = boost::fusion;using proto::_;// Forward declaration of the lambda expression wrappertemplate<typename T>struct lambda;struct lambda_domain : proto::domain<proto::pod_generator<lambda> >{};template<typename I>struct placeholder{ typedef I arity;};template<typename T>struct placeholder_arity{ typedef typename T::arity type;};// The lambda grammar, with the transforms for calculating the max aritystruct Lambda : proto::or_< proto::when< proto::terminal< placeholder<_> > , mpl::next<placeholder_arity<proto::_value> >() > , proto::when< proto::terminal<_> , mpl::int_<0>() > , proto::when< proto::nary_expr<_, proto::vararg<_> > , proto::fold<_, mpl::int_<0>(), mpl::max<Lambda,proto::_state>()> > >{};// simple wrapper for calculating a lambda expression's arity.template<typename Expr>struct lambda_arity : boost::result_of<Lambda(Expr, mpl::void_, mpl::void_)>{};// The lambda context is the same as the default context// with the addition of special handling for lambda placeholderstemplate<typename Tuple>struct lambda_context : proto::callable_context<lambda_context<Tuple> const>{ lambda_context(Tuple const &args) : args_(args) {} template<typename Sig> struct result; template<typename This, typename I> struct result<This(proto::tag::terminal, placeholder<I> const &)> : fusion::result_of::at<Tuple, I> {}; template<typename I> typename fusion::result_of::at<Tuple, I>::type operator ()(proto::tag::terminal, placeholder<I> const &) const { return fusion::at<I>(this->args_); } Tuple args_;};// The lambda<> expression wrapper makes expressions polymorphic// function objectstemplate<typename T>struct lambda{ BOOST_PROTO_BASIC_EXTENDS(T, lambda<T>, lambda_domain) BOOST_PROTO_EXTENDS_ASSIGN() BOOST_PROTO_EXTENDS_SUBSCRIPT() // Careful not to evaluate the return type of the nullary function // unless we have a nullary lambda! typedef typename mpl::eval_if< typename lambda_arity<T>::type , mpl::identity<void> , proto::result_of::eval<T const, lambda_context<fusion::tuple<> > > >::type nullary_type; // Define our operator () that evaluates the lambda expression. nullary_type operator ()() const { fusion::tuple<> args; lambda_context<fusion::tuple<> > ctx(args); return proto::eval(*this, ctx); } template<typename A0> typename proto::result_of::eval<T const, lambda_context<fusion::tuple<A0 const &> > >::type operator ()(A0 const &a0) const { fusion::tuple<A0 const &> args(a0); lambda_context<fusion::tuple<A0 const &> > ctx(args); return proto::eval(*this, ctx); } template<typename A0, typename A1> typename proto::result_of::eval<T const, lambda_context<fusion::tuple<A0 const &, A1 const &> > >::type operator ()(A0 const &a0, A1 const &a1) const { fusion::tuple<A0 const &, A1 const &> args(a0, a1); lambda_context<fusion::tuple<A0 const &, A1 const &> > ctx(args); return proto::eval(*this, ctx); }};// Define some lambda placeholderslambda<proto::terminal<placeholder<mpl::int_<0> > >::type> const _1 = {{}};lambda<proto::terminal<placeholder<mpl::int_<1> > >::type> const _2 = {{}};template<typename T>lambda<typename proto::terminal<T>::type> const val(T const &t){ lambda<typename proto::terminal<T>::type> that = {{t}}; return that;}template<typename T>lambda<typename proto::terminal<T &>::type> const var(T &t){ lambda<typename proto::terminal<T &>::type> that = {{t}}; return that;}template<typename T>struct construct_helper{ typedef T result_type; // for TR1 result_of T operator()() const { return T(); } template<typename A0> T operator()(A0 const &a0) const { return T(a0); } template<typename A0, typename A1> T operator()(A0 const &a0, A1 const &a1) const { return T(a0, a1); }};// Generate BOOST_PROTO_MAX_ARITY-1 overloads of the// construct function template like the one defined above.BOOST_PROTO_DEFINE_VARARG_FUNCTION_TEMPLATE( \ construct \ , lambda_domain \ , (proto::tag::function) \ , ((construct_helper)(typename)) \)struct S{ S() {} S(int i, char c) { std::cout << "S(" << i << "," << c << ")\n"; }};int main(){ // Create some lambda objects and immediately // invoke them by applying their operator(): int i = ( (_1 + 2) / 4 )(42); std::cout << i << std::endl; // prints 11 int j = ( (-(_1 + 2)) / 4 )(42); std::cout << j << std::endl; // prints -11 double d = ( (4 - _2) * 3 )(42, 3.14); std::cout << d << std::endl; // prints 2.58 // check non-const ref terminals (std::cout << _1 << " -- " << _2 << '\n')(42, "Life, the Universe and Everything!"); // prints "42 -- Life, the Universe and Everything!" // "Nullary" lambdas work too int k = (val(1) + val(2))(); std::cout << k << std::endl; // prints 3 // check array indexing for kicks int integers[5] = {0}; (var(integers)[2] = 2)(); (var(integers)[_1] = _1)(3); std::cout << integers[2] << std::endl; // prints 2 std::cout << integers[3] << std::endl; // prints 3 // Now use a lambda with an STL algorithm! int rgi[4] = {1,2,3,4}; char rgc[4] = {'a','b','c','d'}; S rgs[4]; std::transform(rgi, rgi+4, rgc, rgs, construct<S>(_1, _2)); return 0;}//]
⌨️ 快捷键说明
复制代码Ctrl + C
搜索代码Ctrl + F
全屏模式F11
增大字号Ctrl + =
减小字号Ctrl + -
显示快捷键?