对于c++ 17 c++ 20，您将对反思研究小组(SG7)的工作感兴趣。还有一系列平行的论文，包括措辞(P0194)和基本原理、设计和进化(P0385)。(链接解析为每个系列的最新论文。)

从P0194r2(2016-10-15)开始，该语法将使用建议的reflexpr关键字:

meta::get_base_name_v<
  meta::get_element_m<
    meta::get_enumerators_m<reflexpr(MyEnum)>,
    0>
  >

例如(改编自Matus Choclik的reflexpr clang分支):

#include <reflexpr>
#include <iostream>

enum MyEnum { AAA = 1, BBB, CCC = 99 };

int main()
{
  auto name_of_MyEnum_0 = 
    std::meta::get_base_name_v<
      std::meta::get_element_m<
        std::meta::get_enumerators_m<reflexpr(MyEnum)>,
        0>
    >;

  // prints "AAA"
  std::cout << name_of_MyEnum_0 << std::endl;
}

静态反射未能进入c++ 17(更确切地说，进入了2016年11月在Issaquah举行的标准会议上提出的可能是最终草案)，但有信心它将进入c++ 20;摘自赫布·萨特的旅行报告:

特别是，反射研究小组审查了最新合并的静态反射提案，并发现它准备在我们的下一次会议上进入主要的进化小组，开始考虑TS或下一个标准的统一静态反射提案。

下面的解决方案是基于给定enum的std::array<std::string,N>。

对于将enum转换为std::string，我们只需将enum转换为size_t，然后从数组中查找字符串。该操作是O(1)，不需要堆分配。

#include <boost/preprocessor/seq/transform.hpp>
#include <boost/preprocessor/seq/enum.hpp>
#include <boost/preprocessor/stringize.hpp>

#include <string>
#include <array>
#include <iostream>

#define STRINGIZE(s, data, elem) BOOST_PP_STRINGIZE(elem)

// ENUM
// ============================================================================
#define ENUM(X, SEQ) \
struct X {   \
    enum Enum {BOOST_PP_SEQ_ENUM(SEQ)}; \
    static const std::array<std::string,BOOST_PP_SEQ_SIZE(SEQ)> array_of_strings() { \
        return {{BOOST_PP_SEQ_ENUM(BOOST_PP_SEQ_TRANSFORM(STRINGIZE, 0, SEQ))}}; \
    } \
    static std::string to_string(Enum e) { \
        auto a = array_of_strings(); \
        return a[static_cast<size_t>(e)]; \
    } \
}

对于std::string到enum的转换，我们必须对数组进行线性搜索，并将数组索引强制转换为enum。

这里有一些用法示例:http://coliru.stacked-crooked.com/a/e4212f93bee65076

编辑:重做我的解决方案，以便自定义Enum可以在类中使用。

我写了一个库来解决这个问题，所有的事情都发生在编译时，除了获取消息。

用法:

使用宏DEF_MSG定义宏和消息对:

DEF_MSG(CODE_OK,   "OK!")
DEF_MSG(CODE_FAIL, "Fail!")

CODE_OK是要使用的宏，“OK!”是相应的消息。

使用get_message()或gm()来获取消息:

get_message(CODE_FAIL);  // will return "Fail!"
gm(CODE_FAIL);           // works exactly the same as above

使用MSG_NUM查找已经定义了多少个宏。它会自动增加，你不需要做任何事情。

预定义的消息:

MSG_OK:     OK
MSG_BOTTOM: Message bottom

项目:libcodemsg

标准库不会创建额外的数据。一切都发生在编译时。在message_def.h中，它生成一个名为MSG_CODE的enum;在message_def.c中，它生成一个变量，保存静态const char* _g_messages[]中的所有字符串。

在这种情况下，库只能创建一个枚举。这对于返回值非常理想，例如:

MSG_CODE foo(void) {
    return MSG_OK; // or something else
}

MSG_CODE ret = foo();

if (MSG_OK != ret) {
    printf("%s\n", gm(ret););
}

我喜欢这种设计的另一个原因是，您可以在不同的文件中管理消息定义。

我发现这个问题的解看起来好多了。

只需要生成枚举。为此目的编写一个生成器大约需要5分钟的工作。

生成器代码在java和python，超级容易移植到任何你喜欢的语言，包括c++。

而且非常容易扩展任何你想要的功能。

示例输入:

First = 5
Second
Third = 7
Fourth
Fifth=11

生成头文件:

#include <iosfwd>

enum class Hallo
{
    First = 5,
    Second = 6,
    Third = 7,
    Fourth = 8,
    Fifth = 11
};

std::ostream & operator << (std::ostream &, const Hallo&);

生成的CPP文件

#include <ostream>

#include "Hallo.h"

std::ostream & operator << (std::ostream &out, const Hallo&value)
{
    switch(value)
    {
    case Hallo::First:
        out << "First";
        break;
    case Hallo::Second:
        out << "Second";
        break;
    case Hallo::Third:
        out << "Third";
        break;
    case Hallo::Fourth:
        out << "Fourth";
        break;
    case Hallo::Fifth:
        out << "Fifth";
        break;
    default:
        out << "<unknown>";
    }

    return out;
}

生成器以非常简洁的形式作为移植和扩展的模板。这个示例代码确实试图避免覆盖任何文件，但使用它仍然要自担风险。

package cppgen;

import java.io.BufferedReader;
import java.io.File;
import java.io.FileInputStream;
import java.io.FileOutputStream;
import java.io.InputStreamReader;
import java.io.OutputStreamWriter;
import java.io.PrintWriter;
import java.nio.charset.Charset;
import java.util.LinkedHashMap;
import java.util.Map;
import java.util.Map.Entry;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

public class EnumGenerator
{
    static void fail(String message)
    {
        System.err.println(message);
        System.exit(1);
    }

    static void run(String[] args)
    throws Exception
    {
        Pattern pattern = Pattern.compile("\\s*(\\w+)\\s*(?:=\\s*(\\d+))?\\s*", Pattern.UNICODE_CHARACTER_CLASS);
        Charset charset = Charset.forName("UTF8");
        String tab = "    ";

        if (args.length != 3)
        {
            fail("Required arguments: <enum name> <input file> <output dir>");
        }

        String enumName = args[0];

        File inputFile = new File(args[1]);

        if (inputFile.isFile() == false)
        {
            fail("Not a file: [" + inputFile.getCanonicalPath() + "]");
        }

        File outputDir = new File(args[2]);

        if (outputDir.isDirectory() == false)
        {
            fail("Not a directory: [" + outputDir.getCanonicalPath() + "]");
        }

        File headerFile = new File(outputDir, enumName + ".h");
        File codeFile = new File(outputDir, enumName + ".cpp");

        for (File file : new File[] { headerFile, codeFile })
        {
            if (file.exists())
            {
                fail("Will not overwrite file [" + file.getCanonicalPath() + "]");
            }
        }

        int nextValue = 0;

        Map<String, Integer> fields = new LinkedHashMap<>();

        try
        (
            BufferedReader reader = new BufferedReader(new InputStreamReader(new FileInputStream(inputFile), charset));
        )
        {
            while (true)
            {
                String line = reader.readLine();

                if (line == null)
                {
                    break;
                }

                if (line.trim().length() == 0)
                {
                    continue;
                }

                Matcher matcher = pattern.matcher(line);

                if (matcher.matches() == false)
                {
                    fail("Syntax error: [" + line + "]");
                }

                String fieldName = matcher.group(1);

                if (fields.containsKey(fieldName))
                {
                    fail("Double fiend name: " + fieldName);
                }

                String valueString = matcher.group(2);

                if (valueString != null)
                {
                    int value = Integer.parseInt(valueString);

                    if (value < nextValue)
                    {
                        fail("Not a monotonous progression from " + nextValue + " to " + value + " for enum field " + fieldName);
                    }

                    nextValue = value;
                }

                fields.put(fieldName, nextValue);

                ++nextValue;
            }
        }

        try
        (
            PrintWriter headerWriter = new PrintWriter(new OutputStreamWriter(new FileOutputStream(headerFile), charset));
            PrintWriter codeWriter = new PrintWriter(new OutputStreamWriter(new FileOutputStream(codeFile), charset));
        )
        {
            headerWriter.println();
            headerWriter.println("#include <iosfwd>");
            headerWriter.println();
            headerWriter.println("enum class " + enumName);
            headerWriter.println('{');
            boolean first = true;
            for (Entry<String, Integer> entry : fields.entrySet())
            {
                if (first == false)
                {
                    headerWriter.println(",");
                }

                headerWriter.print(tab + entry.getKey() + " = " + entry.getValue());

                first = false;
            }
            if (first == false)
            {
                headerWriter.println();
            }
            headerWriter.println("};");
            headerWriter.println();
            headerWriter.println("std::ostream & operator << (std::ostream &, const " + enumName + "&);");
            headerWriter.println();

            codeWriter.println();
            codeWriter.println("#include <ostream>");
            codeWriter.println();
            codeWriter.println("#include \"" + enumName + ".h\"");
            codeWriter.println();
            codeWriter.println("std::ostream & operator << (std::ostream &out, const " + enumName + "&value)");
            codeWriter.println('{');
            codeWriter.println(tab + "switch(value)");
            codeWriter.println(tab + '{');
            first = true;
            for (Entry<String, Integer> entry : fields.entrySet())
            {
                codeWriter.println(tab + "case " + enumName + "::" + entry.getKey() + ':');
                codeWriter.println(tab + tab + "out << \"" + entry.getKey() + "\";");
                codeWriter.println(tab + tab + "break;");

                first = false;
            }
            codeWriter.println(tab + "default:");
            codeWriter.println(tab + tab + "out << \"<unknown>\";");
            codeWriter.println(tab + '}');
            codeWriter.println();
            codeWriter.println(tab + "return out;");
            codeWriter.println('}');
            codeWriter.println();
        }
    }

    public static void main(String[] args)
    {
        try
        {
            run(args);
        }
        catch(Exception exc)
        {
            exc.printStackTrace();
            System.exit(1);
        }
    }
}

并将其移植到Python 3.5，因为它的不同可能会有所帮助

import re
import collections
import sys
import io
import os

def fail(*args):
    print(*args)
    exit(1)

pattern = re.compile(r'\s*(\w+)\s*(?:=\s*(\d+))?\s*')
tab = "    "

if len(sys.argv) != 4:
    n=0
    for arg in sys.argv:
        print("arg", n, ":", arg, " / ", sys.argv[n])
        n += 1
    fail("Required arguments: <enum name> <input file> <output dir>")

enumName = sys.argv[1]

inputFile = sys.argv[2]

if not os.path.isfile(inputFile):
    fail("Not a file: [" + os.path.abspath(inputFile) + "]")

outputDir = sys.argv[3]

if not os.path.isdir(outputDir):
    fail("Not a directory: [" + os.path.abspath(outputDir) + "]")

headerFile = os.path.join(outputDir, enumName + ".h")
codeFile = os.path.join(outputDir, enumName + ".cpp")

for file in [ headerFile, codeFile ]:
    if os.path.exists(file):
        fail("Will not overwrite file [" + os.path.abspath(file) + "]")

nextValue = 0

fields = collections.OrderedDict()

for line in open(inputFile, 'r'):
    line = line.strip()

    if len(line) == 0:
        continue

    match = pattern.match(line)

    if match == None:
        fail("Syntax error: [" + line + "]")

    fieldName = match.group(1)

    if fieldName in fields:
        fail("Double field name: " + fieldName)

    valueString = match.group(2)

    if valueString != None:
        value = int(valueString)

        if value < nextValue:
            fail("Not a monotonous progression from " + nextValue + " to " + value + " for enum field " + fieldName)

        nextValue = value

    fields[fieldName] = nextValue

    nextValue += 1

headerWriter = open(headerFile, 'w')
codeWriter = open(codeFile, 'w')

try:
    headerWriter.write("\n")
    headerWriter.write("#include <iosfwd>\n")
    headerWriter.write("\n")
    headerWriter.write("enum class " + enumName + "\n")
    headerWriter.write("{\n")
    first = True
    for fieldName, fieldValue in fields.items():
        if not first:
            headerWriter.write(",\n")

        headerWriter.write(tab + fieldName + " = " + str(fieldValue))

        first = False
    if not first:
        headerWriter.write("\n")
    headerWriter.write("};\n")
    headerWriter.write("\n")
    headerWriter.write("std::ostream & operator << (std::ostream &, const " + enumName + "&);\n")
    headerWriter.write("\n")

    codeWriter.write("\n")
    codeWriter.write("#include <ostream>\n")
    codeWriter.write("\n")
    codeWriter.write("#include \"" + enumName + ".h\"\n")
    codeWriter.write("\n")
    codeWriter.write("std::ostream & operator << (std::ostream &out, const " + enumName + "&value)\n")
    codeWriter.write("{\n")
    codeWriter.write(tab + "switch(value)\n")
    codeWriter.write(tab + "{\n")
    for fieldName in fields.keys():
        codeWriter.write(tab + "case " + enumName + "::" + fieldName + ":\n")
        codeWriter.write(tab + tab + "out << \"" + fieldName + "\";\n")
        codeWriter.write(tab + tab + "break;\n")
    codeWriter.write(tab + "default:\n")
    codeWriter.write(tab + tab + "out << \"<unknown>\";\n")
    codeWriter.write(tab + "}\n")
    codeWriter.write("\n")
    codeWriter.write(tab + "return out;\n")
    codeWriter.write("}\n")
    codeWriter.write("\n")
finally:
    headerWriter.close()
    codeWriter.close()

我的解决方案，使用预处理器定义。

您可以在https://repl.it/@JomaCorpFX/nameof#main.cpp上查看此代码

#include <iostream>
#include <stdexcept>
#include <regex>

typedef std::string String;
using namespace std::literals::string_literals;

class Strings
{
public:
    static String TrimStart(const std::string& data)
    {
        String s = data;
        s.erase(s.begin(), std::find_if(s.begin(), s.end(), [](unsigned char ch) {
            return !std::isspace(ch);
        }));
        return s;
    }

    static String TrimEnd(const std::string& data)
    {
        String s = data;
        s.erase(std::find_if(s.rbegin(), s.rend(), [](unsigned char ch) {
            return !std::isspace(ch);
        }).base(),
            s.end());
        return s;
    }

    static String Trim(const std::string& data)
    {
        return TrimEnd(TrimStart(data));
    }

    static String Replace(const String& data, const String& toFind, const String& toReplace)
    {
        String result = data;
        size_t pos = 0;
        while ((pos = result.find(toFind, pos)) != String::npos)
        {
            result.replace(pos, toFind.length(), toReplace);
            pos += toReplace.length();
            pos = result.find(toFind, pos);
        }
        return result;
    }

};

static String Nameof(const String& name)
{
    std::smatch groups;
    String str = Strings::Trim(name);
    if (std::regex_match(str, groups, std::regex(u8R"(^&?([_a-zA-Z]\w*(->|\.|::))*([_a-zA-Z]\w*)$)")))
    {
        if (groups.size() == 4)
        {
            return groups[3];
        }
    }
    throw std::invalid_argument(Strings::Replace(u8R"(nameof(#). Invalid identifier "#".)", u8"#", name));
}

#define nameof(name) Nameof(u8## #name ## s)
#define cnameof(name) Nameof(u8## #name ## s).c_str()

enum TokenType {
    COMMA,
    PERIOD,
    Q_MARK
};

struct MyClass
{
    enum class MyEnum : char {
        AAA = -8,
        BBB = '8',
        CCC = AAA + BBB
    };
};

int main() {
    String greetings = u8"Hello"s;
    std::cout << nameof(COMMA) << std::endl;
    std::cout << nameof(TokenType::PERIOD) << std::endl;
    std::cout << nameof(TokenType::Q_MARK) << std::endl;
    std::cout << nameof(int) << std::endl;
    std::cout << nameof(std::string) << std::endl;
    std::cout << nameof(Strings) << std::endl;
    std::cout << nameof(String) << std::endl;
    std::cout << nameof(greetings) << std::endl;
    std::cout << nameof(&greetings) << std::endl;
    std::cout << nameof(greetings.c_str) << std::endl;
    std::cout << nameof(std::string::npos) << std::endl;
    std::cout << nameof(MyClass::MyEnum::AAA) << std::endl;
    std::cout << nameof(MyClass::MyEnum::BBB) << std::endl;
    std::cout << nameof(MyClass::MyEnum::CCC) << std::endl;


    std::cin.get();
    return 0;
}

输出

COMMA
PERIOD
Q_MARK
int
string
Strings
String
greetings
greetings
c_str
npos
AAA
BBB
CCC

铛

视觉C + +。

我的答案在这里。

你可以同时获得枚举值名称和这些索引，如deque of string。

这种方法只需要少量的复制粘贴和编辑。

当需要枚举类类型值时，需要将获得的结果从size_t类型转换为枚举类类型，但我认为这是一种非常可移植和强大的处理枚举类的方法。

enum class myenum
{
  one = 0,
  two,
  three,
};

deque<string> ssplit(const string &_src, boost::regex &_re)
{
  boost::sregex_token_iterator it(_src.begin(), _src.end(), _re, -1);
  boost::sregex_token_iterator e;
  deque<string> tokens;
  while (it != e)
    tokens.push_back(*it++);
  return std::move(tokens);
}

int main()
{
  regex re(",");
  deque<string> tokens = ssplit("one,two,three", re);
  for (auto &t : tokens) cout << t << endl;
    getchar();
  return 0;
}