使用boost::spirit::x3解析自定义字符串组的通用规则

huangapple go评论74阅读模式
英文:

generic rule for parsing custom group of strings using boost::spirit::x3

问题

你的理解基本正确,但是在代码中有一些语法和符号错误。这里是你的代码的修正版本:

//#define BOOST_SPIRIT_X3_DEBUG
//#define DEBUG_SYMBOLS
#include <iostream>
#include <chrono>
#include <boost/spirit/home/x3.hpp>

using namespace boost::spirit;

auto myRule = +x3::lexeme[ +x3::alnum >> *(x3::char_('.') >> +x3::alnum)] % ',';

void print(std::string test, std::vector<std::string>& allHeroes)
{
    std::cout << "----" << test << std::endl;
    for (auto& hero: allHeroes)
    {
        std::cout << hero << std::endl;    
    }
    std::cout << "-----------------------------" << std::endl;
}

void test_1()
{
    std::string heroSelections = "mirana,slark";
    std::vector<std::string> allHeroes;
    boost::spirit::x3::parse(heroSelections.begin(), heroSelections.end(), myRule, allHeroes);
    print("test_1", allHeroes);
    if (allHeroes.size() == 2)
    {
        assert(allHeroes[0] == "mirana");
        assert(allHeroes[1] == "slark"); 
        std::cout << "PASS" << std::endl;
    }
    else
    {
        std::cout << "FAIL" << std::endl;
    }
}

void test_2()
{
    std::string heroSelections = "hero.0.range.1.mirana,slark";
    std::vector<std::string> allHeroes;
    boost::spirit::x3::parse(heroSelections.begin(), heroSelections.end(), myRule, allHeroes);
    print("test_2", allHeroes);
    if (allHeroes.size() == 2)
    {
        assert(allHeroes[0] == "hero.0.range.1.mirana");
        assert(allHeroes[1] == "slark");  
        std::cout << "PASS" << std::endl;
    }
    else
    {
        std::cout << "FAIL" << std::endl;
    }
}

void test_3()
{
    std::string heroSelections = "hero.0.mirana,slark";
    std::vector<std::string> allHeroes;
    boost::spirit::x3::parse(heroSelections.begin(), heroSelections.end(), myRule, allHeroes);
    print("test_3", allHeroes);
    if (allHeroes.size() == 2)
    {
        assert(allHeroes[0] == "hero.0.mirana");
        assert(allHeroes[1] == "slark");   
        std::cout << "PASS" << std::endl;
    }
    else
    {
        std::cout << "FAIL" << std::endl;
    }
}

int main()
{
    test_1();
    test_2();
    test_3();
    return 0;
}

修正后的代码中,我更正了myRule的定义,以正确地解析字符串和点分隔的子字符串。我还更正了注释中的特殊字符,使其在代码中正确显示。这应该可以正确地解析你的用例。

英文:

I am trying make a generic rule for my use case with using boost::spirit::x3.

use cases are

  1. input "string,string" output [string, string]
  2. input "string.string,string" output [string.string, string]
  3. input "string.0.string.1.string,string" output [string.0.string.1.string, string]

coliru

//#define BOOST_SPIRIT_X3_DEBUG
//#define DEBUG_SYMBOLS
#include &lt;iostream&gt;
#include &lt;chrono&gt;
#include &lt;boost/spirit/home/x3.hpp&gt;
using namespace boost::spirit;
auto myRule = +x3::lexeme[ +x3::alnum &gt;&gt; -x3::char_(&#39;.&#39;)] % +x3::char_(&#39;,&#39;);
void print(std::string test, std::vector&lt;std::string&gt;&amp; allHeroes)
{
std::cout &lt;&lt; &quot;----&quot; &lt;&lt; test &lt;&lt; std::endl;
for (auto&amp; hero: allHeroes)
{
std::cout &lt;&lt; hero &lt;&lt; std::endl;    
}
std::cout &lt;&lt; &quot;-----------------------------&quot; &lt;&lt; std::endl;
}
void test_1()
{
std::string heroSelections = &quot;mirana,slark&quot;;
std::vector&lt;std::string&gt; allHeroes;
boost::spirit::x3::parse(heroSelections.begin(), heroSelections.end(), myRule, allHeroes);
print(&quot;test_1&quot;, allHeroes);
if (allHeroes.size() == 2)
{
assert(allHeroes[0] == &quot;mirana&quot;);
assert(allHeroes[1] == &quot;slark&quot;); 
std::cout &lt;&lt; &quot;PASS&quot; &lt;&lt; std::endl;
}
else
{
std::cout &lt;&lt; &quot;FAIL&quot; &lt;&lt; std::endl;
}
}
void test_2()
{
std::string heroSelections = &quot;hero.0.range.1.mirana,slark&quot;;
std::vector&lt;std::string&gt; allHeroes;
boost::spirit::x3::parse(heroSelections.begin(), heroSelections.end(), myRule, allHeroes);
print(&quot;test_2&quot;, allHeroes);
if (allHeroes.size() == 2)
{
assert(allHeroes[0] == &quot;hero.0.range.1.mirana&quot;);
assert(allHeroes[1] == &quot;slark&quot;);  
std::cout &lt;&lt; &quot;PASS&quot; &lt;&lt; std::endl;
}
else
{
std::cout &lt;&lt; &quot;FAIL&quot; &lt;&lt; std::endl;
}
}
void test_3()
{
std::string heroSelections = &quot;hero.0.mirana,slark&quot;;
std::vector&lt;std::string&gt; allHeroes;
boost::spirit::x3::parse(heroSelections.begin(), heroSelections.end(), myRule, allHeroes);
print(&quot;test_3&quot;, allHeroes);
if (allHeroes.size() == 2)
{
assert(allHeroes[0] == &quot;hero.0.mirana&quot;);
assert(allHeroes[1] == &quot;slark&quot;);   
std::cout &lt;&lt; &quot;PASS&quot; &lt;&lt; std::endl;
}
else
{
std::cout &lt;&lt; &quot;FAIL&quot; &lt;&lt; std::endl;
}
}
int main()
{
test_1();
test_2();
test_3();
return 0;
}

i have tried rules

  1. auto myRule = +x3::lexeme[ +x3::alnum &gt;&gt; -x3::char_(&#39;.&#39;)] % +x3::char_(&#39;,&#39;);

    my understanding:

    +x3::alnum &gt;&gt; -x3::char_(&#39;.&#39;) allows format string and string.string and string.string.string until % +x3::char_(&#39;,&#39;) is met

  2. auto myRule = +x3::alnum &gt;&gt; +x3::lexeme[ +x3::alnum &gt;&gt; -x3::char_(&#39;.&#39;)] % +x3::char_(&#39;,&#39;);

+x3::alnum allows format string and then +x3::lexeme[ +x3::alnum &gt;&gt; -x3::char_(&#39;.&#39;)] allows all string.string until % +x3::char_(&#39;,&#39;).

But clearly my understanding is incorrect here.

what am i doing wrong ?

答案1

得分: 2

以下是翻译好的部分:

It&#39;s [documented][1], just _check_ your understanding:
&gt; The list operator, `a % b`, is a binary operator that matches a list of one or more repetitions of `a` separated by occurrences of `b`. This is equivalent to `a &gt;&gt; *(b &gt;&gt; a)`.
Simplifying the test program making it readable (without all the repetition):
**[Live On Coliru](http://coliru.stacked-crooked.com/a/779268f5d0211436)**
#include &lt;boost/spirit/home/x3.hpp&gt;
#include &lt;fmt/ranges.h&gt;
using Heroes = std::vector&lt;std::string&gt;;
struct {
std::string txt;
Heroes      expected;
} testcases[]{
{
&quot;mirana,slark&quot;,
{&quot;mirana&quot;, &quot;slark&quot;},
},
{
&quot;hero.0.range.1.mirana,slark&quot;,
{&quot;hero.0.range.1.mirana&quot;, &quot;slark&quot;},
},
{
&quot;hero.0.mirana,slark&quot;,
{&quot;hero.0.mirana&quot;, &quot;slark&quot;},
},
};
int main() {
namespace x3 = boost::spirit::x3;
auto const myRule = +x3::lexeme[+x3::alnum &gt;&gt; -x3::char_(&#39;.&#39;)] % +x3::lit(&#39;,&#39;);
for (auto [test, expected] : testcases) {
Heroes actual;
parse(test.begin(), test.end(), myRule, actual);
fmt::print(&quot;{}\t&#39;{}&#39; -&gt; {}\n&quot;, (actual == expected ? &quot;PASS&quot; : &quot;FAIL&quot;), test, actual);
}
}
Prints
PASS	&#39;mirana,slark&#39; -&gt; [&quot;mirana&quot;, &quot;slark&quot;]
FAIL	&#39;hero.0.range.1.mirana,slark&#39; -&gt; [&quot;hero.&quot;, &quot;0.&quot;, &quot;range.&quot;, &quot;1.&quot;, &quot;mirana&quot;, &quot;slark&quot;]
FAIL	&#39;hero.0.mirana,slark&#39; -&gt; [&quot;hero.&quot;, &quot;0.&quot;, &quot;mirana&quot;, &quot;slark&quot;]
I&#39;d suggest simplification like
auto const hero   = +x3::char_(&quot;a-zA-Z0-9.&quot;);
auto const myRule = hero % &#39;,&#39;;
[Printing][2]:
PASS	&#39;mirana,slark&#39; -&gt; [&quot;mirana&quot;, &quot;slark&quot;]
PASS	&#39;hero.0.range.1.mirana,slark&#39; -&gt; [&quot;hero.0.range.1.mirana&quot;, &quot;slark&quot;]
PASS	&#39;hero.0.mirana,slark&#39; -&gt; [&quot;hero.0.mirana&quot;, &quot;slark&quot;]
However, if you actually want to be stricter than the tests suggest, consider:
auto const hero   = x3::raw[ +x3::alnum % &#39;.&#39; ];
auto const myRule = hero % &#39;,&#39;; // or % +x3::lit(&#39;,&#39;);
Also [printing](http://coliru.stacked-crooked.com/a/8296c7219d460a0a) the same
PASS	&#39;mirana,slark&#39; -&gt; [&quot;mirana&quot;, &quot;slark&quot;]
PASS	&#39;hero.0.range.1.mirana,slark&#39; -&gt; [&quot;hero.0.range.1.mirana&quot;, &quot;slark&quot;]
PASS	&#39;hero.0.mirana,slark&#39; -&gt; [&quot;hero.0.mirana&quot;, &quot;slark&quot;]
[1]: https://www.boost.org/doc/libs/1_82_0/libs/spirit/doc/html/spirit/qi/reference/operator.html
[2]: http://coliru.stacked-crooked.com/a/be68c11d7b738f3a
英文:

It's documented, just check your understanding:

> The list operator, a % b, is a binary operator that matches a list of one or more repetitions of a separated by occurrences of b. This is equivalent to a &gt;&gt; *(b &gt;&gt; a).

Simplifying the test program making it readable (without all the repetition):

Live On Coliru

#include &lt;boost/spirit/home/x3.hpp&gt;
#include &lt;fmt/ranges.h&gt;
using Heroes = std::vector&lt;std::string&gt;;
struct {
std::string txt;
Heroes      expected;
} testcases[]{
{
&quot;mirana,slark&quot;,
{&quot;mirana&quot;, &quot;slark&quot;},
},
{
&quot;hero.0.range.1.mirana,slark&quot;,
{&quot;hero.0.range.1.mirana&quot;, &quot;slark&quot;},
},
{
&quot;hero.0.mirana,slark&quot;,
{&quot;hero.0.mirana&quot;, &quot;slark&quot;},
},
};
int main() {
namespace x3 = boost::spirit::x3;
auto const myRule = +x3::lexeme[+x3::alnum &gt;&gt; -x3::char_(&#39;.&#39;)] % +x3::lit(&#39;,&#39;);
for (auto [test, expected] : testcases) {
Heroes actual;
parse(test.begin(), test.end(), myRule, actual);
fmt::print(&quot;{}\t&#39;{}&#39; -&gt; {}\n&quot;, (actual == expected ? &quot;PASS&quot; : &quot;FAIL&quot;), test, actual);
}
}

Prints

PASS	&#39;mirana,slark&#39; -&gt; [&quot;mirana&quot;, &quot;slark&quot;]
FAIL	&#39;hero.0.range.1.mirana,slark&#39; -&gt; [&quot;hero.&quot;, &quot;0.&quot;, &quot;range.&quot;, &quot;1.&quot;, &quot;mirana&quot;, &quot;slark&quot;]
FAIL	&#39;hero.0.mirana,slark&#39; -&gt; [&quot;hero.&quot;, &quot;0.&quot;, &quot;mirana&quot;, &quot;slark&quot;]

I'd suggest simplification like

auto const hero   = +x3::char_(&quot;a-zA-Z0-9.&quot;);
auto const myRule = hero % &#39;,&#39;;

Printing:

PASS	&#39;mirana,slark&#39; -&gt; [&quot;mirana&quot;, &quot;slark&quot;]
PASS	&#39;hero.0.range.1.mirana,slark&#39; -&gt; [&quot;hero.0.range.1.mirana&quot;, &quot;slark&quot;]
PASS	&#39;hero.0.mirana,slark&#39; -&gt; [&quot;hero.0.mirana&quot;, &quot;slark&quot;]

However, if you actually want to be stricter than the tests suggest, consider:

auto const hero   = x3::raw[ +x3::alnum % &#39;.&#39; ];
auto const myRule = hero % &#39;,&#39;; // or % +x3::lit(&#39;,&#39;);

Also printing the same

PASS	&#39;mirana,slark&#39; -&gt; [&quot;mirana&quot;, &quot;slark&quot;]
PASS	&#39;hero.0.range.1.mirana,slark&#39; -&gt; [&quot;hero.0.range.1.mirana&quot;, &quot;slark&quot;]
PASS	&#39;hero.0.mirana,slark&#39; -&gt; [&quot;hero.0.mirana&quot;, &quot;slark&quot;]

huangapple
  • 本文由 发表于 2023年6月12日 01:47:43
  • 转载请务必保留本文链接:https://go.coder-hub.com/76451742.html
匿名

发表评论

匿名网友

:?: :razz: :sad: :evil: :!: :smile: :oops: :grin: :eek: :shock: :???: :cool: :lol: :mad: :twisted: :roll: :wink: :idea: :arrow: :neutral: :cry: :mrgreen:

确定