Boost 将Spirit标记化表达式增强为向量

我试图解析一个表达式，该表达式也可以包含标识符，并将每个元素推送到

std:：vector

，我提出了以下语法：

#include <boost/spirit/include/qi.hpp>
#include <boost/spirit/include/phoenix_core.hpp>
#include <boost/spirit/include/phoenix_operator.hpp>
#include <vector>

namespace qi = boost::spirit::qi;

struct Tokeniser
    : boost::spirit::qi::grammar <std::string::const_iterator, std::vector <std::string> (), boost::spirit::ascii::space_type>
{
    Tokeniser() : Tokeniser::base_type(expression)
    {
        namespace qi = boost::spirit::qi;
        expression = 
            term >>
            *( (qi::string("+")[qi::_val.push_back(qi::_1)] >> term) |
               (qi::string("-")[qi::_val.push_back(qi::_1)] >> term) );

        term = 
            factor >>
            *( (qi::string("*")[qi::_val.push_back(qi::_1)] >> factor) |
               (qi::string("/")[qi::_val.push_back(qi::_1)] >> factor) );

        factor =
            (identifier | myDouble_)[qi::_val.push_back(qi::_1)] |
            qi::string("(")[qi::_val.push_back(qi::_1)] >> expression >> qi::string(")")[qi::_val.push_back(qi::_1)];

        identifier = qi::raw [ qi::lexeme[ (qi::alpha | '_') >> *(qi::alnum | '_') ] ]; 

        myDouble_ = qi::raw [ qi::double_ ];
    }

    boost::spirit::qi::rule<std::string::const_iterator, std::vector <std::string> (), boost::spirit::ascii::space_type> expression;
    boost::spirit::qi::rule<std::string::const_iterator, boost::spirit::ascii::space_type>                               factor;
    boost::spirit::qi::rule<std::string::const_iterator, boost::spirit::ascii::space_type>                               term;        

    boost::spirit::qi::rule<std::string::const_iterator, std::string(), boost::spirit::ascii::space_type> identifier;
    boost::spirit::qi::rule<std::string::const_iterator, std::string(), boost::spirit::ascii::space_type> myDouble_;
};

#包括
#包括
#包括
#包括
名称空间qi=boost:：spirit:：qi；
结构标记器
：boost：：spirit：：qi：：语法
{
Tokeniser（）：Tokeniser:：base_类型（表达式）
{
名称空间qi=boost:：spirit:：qi；
表达式=
术语>>
*（（qi：：字符串（“+”[qi:：_val.push_back（qi:：_1）]>>术语）|
（qi：：字符串（“-”[qi:：_val.push_back（qi:_1）]>>术语））；
术语=
因子>>
*（（qi：：字符串（“*”[qi:：_val.push_back（qi:：_1）]>>因子）|
（qi：：字符串（“/”[qi:：_val.push_back（qi:：_1）]>>因子）；
因素=
（标识符| myDouble）[qi:：_val.push_back（qi:：_1）]|
qi:：string（“”[qi:：_val.push_back（qi:：_1）]>>表达式>>qi：：string（“”）[qi:_val.push_back（qi:_1）]；
identifier=qi:：raw[qi:：lexeme[（qi:：alpha |''）>*（qi:：alnum |''）]；
myDouble=qi:：raw[qi:：double]；
}
boost:：spirit:：qi:：规则表达式；
提振：：精神：：气：：规则因素；
提振：：精神：：气：：规则术语；
boost:：spirit:：qi:：规则标识符；
提振：：精神：：气：：规则myDouble；
};

但是，我得到以下错误

“const struct boost:：phoenix:：actor”没有名为“push_back”的成员

有没有直接的方法来执行我想做的事情？
是的，占位符类型（显然）没有
push_back
成员

C++是强类型的。任何延迟的操作都是一种“幻觉”：参与者通过组合特殊用途的类型在表达式模板中表示，这些类型可以在以后进行“评估”

表达式模板简介

#define BOOST_SPIRIT_DEBUG
#include <boost/spirit/include/qi.hpp>
#include <boost/spirit/include/phoenix.hpp>
#include <boost/phoenix/stl.hpp>
#include <vector>

namespace qi = boost::spirit::qi;
namespace px = boost::phoenix;

template <typename It = std::string::const_iterator>
struct Tokeniser
    : qi::grammar <It, std::vector <std::string> (), boost::spirit::ascii::space_type>
{
    Tokeniser() : Tokeniser::base_type(expression)
    {
        auto push      = px::push_back(qi::_val, qi::_1);
        auto propagate = px::insert(qi::_val, px::end(qi::_val), px::begin(qi::_1), px::end(qi::_1));

        expression = 
            term[propagate] >>
            *( (qi::string("+")[push] >> term[propagate]) |
               (qi::string("-")[push] >> term[propagate]) );

        term = 
            factor[propagate] >>
            *( (qi::string("*")[push] >> factor[propagate]) |
               (qi::string("/")[push] >> factor[propagate]) );

        factor =
            (identifier | myDouble_)[push] |
            qi::string("(")[push] >> expression[propagate] >> qi::string(")")[push];

        identifier = qi::raw [ qi::lexeme[ (qi::alpha | '_') >> *(qi::alnum | '_') ] ]; 

        myDouble_ = qi::raw [ qi::double_ ];

        BOOST_SPIRIT_DEBUG_NODES((expression)(term)(factor)(identifier)(myDouble_))
    }

    qi::rule<It, std::vector<std::string>(), boost::spirit::ascii::space_type> expression;
    qi::rule<It, std::vector<std::string>(), boost::spirit::ascii::space_type> factor;
    qi::rule<It, std::vector<std::string>(), boost::spirit::ascii::space_type> term;
    qi::rule<It, std::string(), boost::spirit::ascii::space_type> identifier;
    qi::rule<It, std::string(), boost::spirit::ascii::space_type> myDouble_;
};

int main() {
    Tokeniser<> tok;

    std::string const input = "x + 89/(y*y)";

    auto f = input.begin(), l = input.end();
    std::vector<std::string> tokens;
    if (phrase_parse(f, l, tok, boost::spirit::ascii::space, tokens)) {
        std::cout << "Parsed " << tokens.size() << " tokens:\n";
        for (auto& token : tokens)
            std::cout << " - '" << token << "'\n";
    } else {
        std::cout << "Parse failed\n";
    }

    if (f != l)
        std::cout << "Remaining unparsed input: '" << std::string(f,l) << "'\n";
}

#define BOOST_SPIRIT_DEBUG
#include <boost/spirit/include/qi.hpp>
#include <vector>

namespace qi = boost::spirit::qi;

template <typename It = std::string::const_iterator>
struct Tokeniser : qi::grammar <It, std::vector <std::string>()> {
    Tokeniser() : Tokeniser::base_type(start)
    {
        start = qi::skip(boost::spirit::ascii::space) [expression];
        expression =
            term >>
            *( (qi::string("+") >> term) |
               (qi::string("-") >> term) );

        term =
            factor >>
            *( (qi::string("*") >> factor) |
               (qi::string("/") >> factor) );

        factor =
            (identifier | myDouble_) |
            qi::string("(") >> expression >> qi::string(")");

        identifier = qi::raw [ (qi::alpha | '_') >> *(qi::alnum | '_') ]; 
        myDouble_  = qi::raw [ qi::double_ ];

        BOOST_SPIRIT_DEBUG_NODES((expression)(term)(factor)(identifier)(myDouble_))
    }

    qi::rule<It, std::vector<std::string>()> start;
    qi::rule<It, std::vector<std::string>(), boost::spirit::ascii::space_type> expression, factor, term;
    qi::rule<It, std::string()> identifier, myDouble_;
};

int main() {
    Tokeniser<> tok;

    std::string const input = "x + 89/(y*y)";

    auto f = input.begin(), l = input.end();
    std::vector<std::string> tokens;
    if (parse(f, l, tok, tokens)) {
        std::cout << "Parsed " << tokens.size() << " tokens:\n";
        for (auto& token : tokens)
            std::cout << " - '" << token << "'\n";
    } else {
        std::cout << "Parse failed\n";
    }

    if (f != l)
        std::cout << "Remaining unparsed input: '" << std::string(f,l) << "'\n";
}

以防你想了解它的实际工作原理，这是一个从头开始的简单例子。注释描述了代码的各个部分的功能：

// we have lazy placeholder types:
template <int N> struct placeholder {};
placeholder<1> _1;
placeholder<2> _2;
placeholder<3> _3;

// note that every type here is stateless, and acts just like a more
// complicated placeholder.
// We can have expressions, like binary addition:
template <typename L, typename R> struct addition { };
template <typename L, typename R> struct multiplication { };

// here is the "factory" for our expression template:
template <typename L, typename R> addition<L,R> operator+(L const&, R const&) { return {}; }
template <typename L, typename R> multiplication<L,R> operator*(L const&, R const&) { return {}; }

///////////////////////////////////////////////
// To evaluate/interpret the expressions, we have to define "evaluation" for each type of placeholder:
template <typename Ctx, int N> 
    auto eval(Ctx& ctx, placeholder<N>) { return ctx.arg(N); }
template <typename Ctx, typename L, typename R>
    auto eval(Ctx& ctx, addition<L, R>) { return eval(ctx, L{}) + eval(ctx, R{}); }
template <typename Ctx, typename L, typename R>
    auto eval(Ctx& ctx, multiplication<L, R>) { return eval(ctx, L{}) * eval(ctx, R{}); }

///////////////////////////////////////////////
// A simple real-life context would contain the arguments:
#include <vector>
struct Context {
    std::vector<double> _args;

    // define the operation to get an argument from this context:
    double arg(int i) const { return _args.at(i-1); }
};

#include <iostream>
int main() {
    auto foo = _1 + _2 + _3;

    Context ctx { { 3, 10, -4 } };
    std::cout << "foo: " << eval(ctx, foo) << "\n";
    std::cout << "_1 + _2 * _3: " << eval(ctx, _1 + _2 * _3) << "\n";
}

确定行动：
您必须“描述”push_back
操作，而不是试图在占位符上找到这样的操作。凤凰城有你的背：

#include <boost/phoenix/stl.hpp>

但是，这还有一个额外的问题，
\u val
被解析为规则的属性类型。但是您的一些规则没有声明属性类型，因此它默认为
qi:：unused_type
。显然，为该属性生成“push\u back”求值代码对于
未使用的\u类型
不起作用

让我们修复这些声明：

qi::rule<It, std::vector<std::string>(), boost::spirit::ascii::space_type> expression;
qi::rule<It, std::vector<std::string>(), boost::spirit::ascii::space_type> factor;
qi::rule<It, std::vector<std::string>(), boost::spirit::ascii::space_type> term;

现在，使用

#define BOOST_SPIRIT_DEBUG
#include <boost/spirit/include/qi.hpp>
#include <boost/spirit/include/phoenix.hpp>
#include <boost/phoenix/stl.hpp>
#include <vector>

namespace qi = boost::spirit::qi;
namespace px = boost::phoenix;

template <typename It = std::string::const_iterator>
struct Tokeniser
    : qi::grammar <It, std::vector <std::string> (), boost::spirit::ascii::space_type>
{
    Tokeniser() : Tokeniser::base_type(expression)
    {
        auto push      = px::push_back(qi::_val, qi::_1);
        auto propagate = px::insert(qi::_val, px::end(qi::_val), px::begin(qi::_1), px::end(qi::_1));

        expression = 
            term[propagate] >>
            *( (qi::string("+")[push] >> term[propagate]) |
               (qi::string("-")[push] >> term[propagate]) );

        term = 
            factor[propagate] >>
            *( (qi::string("*")[push] >> factor[propagate]) |
               (qi::string("/")[push] >> factor[propagate]) );

        factor =
            (identifier | myDouble_)[push] |
            qi::string("(")[push] >> expression[propagate] >> qi::string(")")[push];

        identifier = qi::raw [ qi::lexeme[ (qi::alpha | '_') >> *(qi::alnum | '_') ] ]; 

        myDouble_ = qi::raw [ qi::double_ ];

        BOOST_SPIRIT_DEBUG_NODES((expression)(term)(factor)(identifier)(myDouble_))
    }

    qi::rule<It, std::vector<std::string>(), boost::spirit::ascii::space_type> expression;
    qi::rule<It, std::vector<std::string>(), boost::spirit::ascii::space_type> factor;
    qi::rule<It, std::vector<std::string>(), boost::spirit::ascii::space_type> term;
    qi::rule<It, std::string(), boost::spirit::ascii::space_type> identifier;
    qi::rule<It, std::string(), boost::spirit::ascii::space_type> myDouble_;
};

int main() {
    Tokeniser<> tok;

    std::string const input = "x + 89/(y*y)";

    auto f = input.begin(), l = input.end();
    std::vector<std::string> tokens;
    if (phrase_parse(f, l, tok, boost::spirit::ascii::space, tokens)) {
        std::cout << "Parsed " << tokens.size() << " tokens:\n";
        for (auto& token : tokens)
            std::cout << " - '" << token << "'\n";
    } else {
        std::cout << "Parse failed\n";
    }

    if (f != l)
        std::cout << "Remaining unparsed input: '" << std::string(f,l) << "'\n";
}

#define BOOST_SPIRIT_DEBUG
#include <boost/spirit/include/qi.hpp>
#include <vector>

namespace qi = boost::spirit::qi;

template <typename It = std::string::const_iterator>
struct Tokeniser : qi::grammar <It, std::vector <std::string>()> {
    Tokeniser() : Tokeniser::base_type(start)
    {
        start = qi::skip(boost::spirit::ascii::space) [expression];
        expression =
            term >>
            *( (qi::string("+") >> term) |
               (qi::string("-") >> term) );

        term =
            factor >>
            *( (qi::string("*") >> factor) |
               (qi::string("/") >> factor) );

        factor =
            (identifier | myDouble_) |
            qi::string("(") >> expression >> qi::string(")");

        identifier = qi::raw [ (qi::alpha | '_') >> *(qi::alnum | '_') ]; 
        myDouble_  = qi::raw [ qi::double_ ];

        BOOST_SPIRIT_DEBUG_NODES((expression)(term)(factor)(identifier)(myDouble_))
    }

    qi::rule<It, std::vector<std::string>()> start;
    qi::rule<It, std::vector<std::string>(), boost::spirit::ascii::space_type> expression, factor, term;
    qi::rule<It, std::string()> identifier, myDouble_;
};

int main() {
    Tokeniser<> tok;

    std::string const input = "x + 89/(y*y)";

    auto f = input.begin(), l = input.end();
    std::vector<std::string> tokens;
    if (parse(f, l, tok, tokens)) {
        std::cout << "Parsed " << tokens.size() << " tokens:\n";
        for (auto& token : tokens)
            std::cout << " - '" << token << "'\n";
    } else {
        std::cout << "Parse failed\n";
    }

    if (f != l)
        std::cout << "Remaining unparsed input: '" << std::string(f,l) << "'\n";
}

大扫除
一般来说，避免语义动作（参见我的答案——特别是关于副作用的子弹）。大多数情况下，您可以不受自动属性传播的影响。我想说这是Boost Spirit的关键卖点

进一步简化跳过/lexemes（）的操作确实可以显著减少代码和编译时间：


#define BOOST_SPIRIT_DEBUG
#include <boost/spirit/include/qi.hpp>
#include <boost/spirit/include/phoenix.hpp>
#include <boost/phoenix/stl.hpp>
#include <vector>

namespace qi = boost::spirit::qi;
namespace px = boost::phoenix;

template <typename It = std::string::const_iterator>
struct Tokeniser
    : qi::grammar <It, std::vector <std::string> (), boost::spirit::ascii::space_type>
{
    Tokeniser() : Tokeniser::base_type(expression)
    {
        auto push      = px::push_back(qi::_val, qi::_1);
        auto propagate = px::insert(qi::_val, px::end(qi::_val), px::begin(qi::_1), px::end(qi::_1));

        expression = 
            term[propagate] >>
            *( (qi::string("+")[push] >> term[propagate]) |
               (qi::string("-")[push] >> term[propagate]) );

        term = 
            factor[propagate] >>
            *( (qi::string("*")[push] >> factor[propagate]) |
               (qi::string("/")[push] >> factor[propagate]) );

        factor =
            (identifier | myDouble_)[push] |
            qi::string("(")[push] >> expression[propagate] >> qi::string(")")[push];

        identifier = qi::raw [ qi::lexeme[ (qi::alpha | '_') >> *(qi::alnum | '_') ] ]; 

        myDouble_ = qi::raw [ qi::double_ ];

        BOOST_SPIRIT_DEBUG_NODES((expression)(term)(factor)(identifier)(myDouble_))
    }

    qi::rule<It, std::vector<std::string>(), boost::spirit::ascii::space_type> expression;
    qi::rule<It, std::vector<std::string>(), boost::spirit::ascii::space_type> factor;
    qi::rule<It, std::vector<std::string>(), boost::spirit::ascii::space_type> term;
    qi::rule<It, std::string(), boost::spirit::ascii::space_type> identifier;
    qi::rule<It, std::string(), boost::spirit::ascii::space_type> myDouble_;
};

int main() {
    Tokeniser<> tok;

    std::string const input = "x + 89/(y*y)";

    auto f = input.begin(), l = input.end();
    std::vector<std::string> tokens;
    if (phrase_parse(f, l, tok, boost::spirit::ascii::space, tokens)) {
        std::cout << "Parsed " << tokens.size() << " tokens:\n";
        for (auto& token : tokens)
            std::cout << " - '" << token << "'\n";
    } else {
        std::cout << "Parse failed\n";
    }

    if (f != l)
        std::cout << "Remaining unparsed input: '" << std::string(f,l) << "'\n";
}

#define BOOST_SPIRIT_DEBUG
#include <boost/spirit/include/qi.hpp>
#include <vector>

namespace qi = boost::spirit::qi;

template <typename It = std::string::const_iterator>
struct Tokeniser : qi::grammar <It, std::vector <std::string>()> {
    Tokeniser() : Tokeniser::base_type(start)
    {
        start = qi::skip(boost::spirit::ascii::space) [expression];
        expression =
            term >>
            *( (qi::string("+") >> term) |
               (qi::string("-") >> term) );

        term =
            factor >>
            *( (qi::string("*") >> factor) |
               (qi::string("/") >> factor) );

        factor =
            (identifier | myDouble_) |
            qi::string("(") >> expression >> qi::string(")");

        identifier = qi::raw [ (qi::alpha | '_') >> *(qi::alnum | '_') ]; 
        myDouble_  = qi::raw [ qi::double_ ];

        BOOST_SPIRIT_DEBUG_NODES((expression)(term)(factor)(identifier)(myDouble_))
    }

    qi::rule<It, std::vector<std::string>()> start;
    qi::rule<It, std::vector<std::string>(), boost::spirit::ascii::space_type> expression, factor, term;
    qi::rule<It, std::string()> identifier, myDouble_;
};

int main() {
    Tokeniser<> tok;

    std::string const input = "x + 89/(y*y)";

    auto f = input.begin(), l = input.end();
    std::vector<std::string> tokens;
    if (parse(f, l, tok, tokens)) {
        std::cout << "Parsed " << tokens.size() << " tokens:\n";
        for (auto& token : tokens)
            std::cout << " - '" << token << "'\n";
    } else {
        std::cout << "Parse failed\n";
    }

    if (f != l)
        std::cout << "Remaining unparsed input: '" << std::string(f,l) << "'\n";
}

遗留问题
你有没有想过回溯行为？我认为您需要在您的规则中明智地放置一些
qi:：hold[]
指令，请参见例如

重大问题：
当你想要的只是令牌时，你为什么要“解析”呢
你将如何处理代币？解析器构建了大量信息，但又一次丢弃了这些信息
简短回答：是的。不，你不应该。我的回答对这里发生的事情有一个更长的解释。
Parsed 9 tokens:
 - 'x'
 - '+'
 - '89'
 - '/'
 - '('
 - 'y'
 - '*'
 - 'y'
 - ')'