C++ - 如何提取字符串中的有效字符串？

Question

问题:我正在尝试使用 C++ 从游戏名称中提取古代防御 (DotA) 的有效游戏模式。

详细信息:

• 游戏名称最长不能超过 31 个字符
• 共有三种游戏模式类别:主要、次要和其他
  • 只能选择一种主要游戏模式
  • 某些主要游戏模式与某些次要游戏模式不兼容
  • 某些次要游戏模式与其他次要游戏模式不兼容
  • 杂项游戏模式可以与所有其他游戏模式相结合

这里是各种游戏模式的列表，图表显示了每种模式与哪些次要模式兼容(X 表示不兼容):

// Only 1 primary allowed
static char *Primary[] = {
          // Compatible with > | dm | rv | mm | du | sh | aa | ai | as | id | em | np | sc | om | nt | nm | nb | ro | mo | sp | 
    "ap", // All Pick          |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |
    "ar", // All Random        |    | X  |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |
    "tr", // Team Random       | X  | X  |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |
    "mr", // Mode Random       | X  | X  |    |    | X  | X  | X  | X  |    |    |    |    |    |    |    |    | X  | X  |    |
    "lm", // League Mode       | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  |    |
    "rd", // Random Draft      | X  | X  | X  |    | X  | X  | X  | X  |    |    |    |    |    |    |    |    | X  | X  |    |
    "vr", // Vote Random       | X  | X  | X  |    | X  | X  | X  | X  |    |    |    |    |    |    |    |    | X  | X  |    |
    "el", // Extended League   | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  |    |
    "sd", // Single Draft      | X  | X  | X  |    | X  | X  | X  | X  |    |    |    |    |    |    |    |    | X  | X  |    |
    "cm", // Captains Mode     | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  | X  |
    "cd"  // Captains Draft    | X  | X  | X  |    | X  | X  | X  | X  |    |    |    |    |    |    |    |    | X  | X  |    |
};

static char *Secondary[] = {
          // Compatible with > | dm | rv | mm | du | sh | aa | ai | as | id | em | np | sc | om | nt | nm | nb | ro | mo | sp | 
    "dm", // Death Match       |    | X  | X  |    | X  | X  | X  | X  |    |    |    |    |    |    |    |    | X  | X  |    |
    "rv", // Reverse Mode      | X  |    |    |    | X  |    |    |    |    |    |    |    |    |    |    |    |    |    |    |
    "mm", // Mirror Match      | X  |    |    |    | X  |    |    |    |    |    |    |    |    |    |    |    |    |    |    |
    "du", // Duplicate Mode    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |
    "sh", // Same Hero         | X  | X  | X  |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |
    "aa", // All Agility       | X  |    |    |    |    |    | X  | X  |    |    |    |    |    |    |    |    |    |    |    |
    "ai", // All Intelligence  | X  |    |    |    |    | X  |    | X  |    |    |    |    |    |    |    |    |    |    |    |
    "as", // All Strength      | X  |    |    |    |    | X  | X  |    |    |    |    |    |    |    |    |    |    |    |    |
    "id", // Item Drop         |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |
    "em", // Easy Mode         |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |
    "np", // No Powerups       |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |
    "sc", // Super Creeps      |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    | 
    "om", // Only Mid          |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |
    "nt", // No Top            |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |
    "nm", // No Middle         |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |
    "nb", // No Bottom         |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    | 
    "ro", // Range Only        | X  |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    | X  |    | 
    "mo", // Melee Only        | X  |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    | X  |    |    | 
    "sp"  // Shuffle Players   |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |    |
};

// These options are always available
static char *Misc[] = {
    "ns", // No Swap
    "nr", // No Repick
    "ts", // Terrain Snow
    "pm", // Pooling Mode
    "oi", // Observer Info
    "mi", // Mini Heroes
    "fr", // Fast Respawn
    "so"  // Switch On
};

示例:这是一些示例输入，具有所需的输出:

“DotA v6.60 -RDSOSP USA/CA LC!” -> “rdsosp”

“DOTA AREMDM 美国 LC”->“aremdm”

“DotA v6.60 -ApEmDuSpId USA BL”->“apemduspid”

注意:解决方案不一定必须提供实际代码、伪代码，甚至只是对您将如何处理它的解释也是可以接受和首选的。此外，该解决方案需要足够灵活，以便我可以相当轻松地添加另一种游戏模式。还可以安全地假设在游戏名称中，游戏模式将始终以主要游戏模式开头。

---

结果:

#include <cstdarg>
#include <algorithm>
#include <iostream>
#include <string>
#include <sstream>
#include <map>
#include <vector>

std::map<std::string, std::vector<std::string> > ModeCompatibilityMap;

static const unsigned int PrimaryModesCount = 11;
static char *PrimaryModes[] = { 
    "ap", "ar", "tr", "mr", "lm", "rd", "vr", "el", "sd", "cm", "cd"
};

static const unsigned int SecondaryModesCounts = 19;
static char *SecondaryModes[] = {
    "dm", "rv", "mm", "du", "sh", "aa", "ai", "as", "id", "em", "np",
    "sc", "om", "nt", "nm", "nb", "ro", "mo", "sp"
};

static const unsigned int MiscModesCount = 8;
static char *MiscModes[] = {
    "ns", "nr", "ts", "pm", "oi", "mi", "fr", "so" 
};

std::vector<std::string> Vectorize( int count, ... ) {
    std::vector<std::string> result;

    va_list vl;
    va_start( vl, count );

    for ( int i = 0; i < count; ++i ) {
        char *buffer = va_arg( vl, char * );
        result.push_back( buffer );
    }

    va_end( vl );

    return result;
}

void InitializeModeCompatibilityMap() {
    // Primary
    ModeCompatibilityMap["ar"] = Vectorize( 1, "rv" );
    ModeCompatibilityMap["tr"] = Vectorize( 2, "dm", "rv" );
    ModeCompatibilityMap["mr"] = Vectorize( 8, "dm", "rv", "sh", "aa", "ai", "as", "ro", "mo" );
    ModeCompatibilityMap["lm"] = Vectorize( 18, "dm", "rv", "mm", "du", "sh", "aa", "ai", "as", "id", "em", "np", "sc", "om", "nt", "nm", "nb", "ro", "mo" );
    ModeCompatibilityMap["rd"] = Vectorize( 9, "dm", "rv", "mm", "sh", "aa", "ai", "as", "ro", "mo" );
    ModeCompatibilityMap["vr"] = Vectorize( 9, "dm", "rv", "mm", "sh", "aa", "ai", "as", "ro", "mo" );
    ModeCompatibilityMap["el"] = Vectorize( 18, "dm", "rv", "mm", "du", "sh", "aa", "ai", "as", "id", "em", "np", "sc", "om", "nt", "nm", "nb", "ro", "mo" );
    ModeCompatibilityMap["sd"] = Vectorize( 9, "dm", "rv", "mm", "sh", "aa", "ai", "as", "ro", "mo" );
    ModeCompatibilityMap["cm"] = Vectorize( 19, "dm", "rv", "mm", "du", "sh", "aa", "ai", "as", "id", "em", "np", "sc", "om", "nt", "nm", "nb", "ro", "mo", "sp" );
    ModeCompatibilityMap["cd"] = Vectorize( 9, "dm", "rv", "mm", "sh", "aa", "ai", "as", "ro", "mo" );   
    // Secondary
    ModeCompatibilityMap["dm"] = Vectorize( 8, "rv", "mm", "sh", "aa", "ai", "as", "ro", "mo" );
    ModeCompatibilityMap["rv"] = Vectorize( 2, "dm", "sh" );
    ModeCompatibilityMap["mm"] = Vectorize( 2, "dm", "sh" );
    ModeCompatibilityMap["sh"] = Vectorize( 3, "dm", "rv", "mm" );
    ModeCompatibilityMap["aa"] = Vectorize( 3, "dm", "ai", "as" );
    ModeCompatibilityMap["ai"] = Vectorize( 3, "dm", "aa", "as" );
    ModeCompatibilityMap["as"] = Vectorize( 3, "dm", "aa", "ai" );
    ModeCompatibilityMap["ro"] = Vectorize( 2, "dm", "mo" );
    ModeCompatibilityMap["mo"] = Vectorize( 2, "dm", "ro" );
}

std::vector<std::string> Tokenize( const std::string &string ) {
    std::vector<std::string> tokens;
    std::string token;
    std::stringstream ss( string );

    while ( ss >> token ) {
        tokens.push_back( token );
    }

    return tokens;
}

void SanitizeString( std::string &in ) {
    std::transform( in.begin(), in.end(), in.begin(), tolower );

    for ( size_t i = 0; i < in.size(); ++i ) {
        if ( in[i] < 'a' || in[i] > 'z' ) {
            in.erase( i--, 1 );
        }
    }
}

std::vector<std::string> SplitString( const std::string &in, int count ) {
    std::vector<std::string> result;

    if ( in.length() % count != 0 ) {
        return result;
    }

    for ( std::string::const_iterator i = in.begin(); i != in.end(); i += count ) {
        result.push_back( std::string( i, i + count ) );
    }

    return result;
}

bool IsPrimaryGameMode( const std::string &in ) {
    for ( int i = 0; i < PrimaryModesCount; ++i ) {
        if ( strcmp( PrimaryModes[i], in.c_str() ) == 0 ) {
            return true;
        }
    }

    return false;
}

bool IsSecondaryGameMode( const std::string &in ) {
    for ( int i = 0; i < SecondaryModesCounts; ++i ) {
        if ( strcmp( SecondaryModes[i], in.c_str() ) == 0 ) {
            return true;
        }
    }

    return false;
}

bool IsMiscGameMode( const std::string &in ) {
    for ( int i = 0; i < MiscModesCount; ++i ) {
        if ( strcmp( MiscModes[i], in.c_str() ) == 0 ) {
            return true;
        }
    }

    return false;
}

bool IsValidGameMode( std::string in, std::string &out ) {
    // 1. Strip all non-letters from the string and convert it to lower-case
    SanitizeString( in );

    // 2. Confirm that it is a multiple of 2
    if ( in.length() == 0 || in.length() % 2 != 0 ) {
        return false;
    }

    // 3. Split the string further into strings of 2 characters
    std::vector<std::string> modes = SplitString( in, 2 );

    // 4. Verify that each game mode is a valid game mode and is compatible with the others
    bool primaryModeSet = false;

    for ( size_t i = 0; i < modes.size(); ++i ) {
        if ( IsPrimaryGameMode( modes[i] ) || IsSecondaryGameMode( modes[i] ) ) {
            if ( IsPrimaryGameMode( modes[i] ) ) {
                if ( primaryModeSet ) {
                    return false;
                }

                primaryModeSet = true;
            }

            if ( ModeCompatibilityMap.count( modes[i] ) > 0 ) {
                std::vector<std::string> badModes = ModeCompatibilityMap[modes[i]];

                for ( size_t j = 0; j < badModes.size(); ++j ) {
                    for ( size_t k = 0; k < modes.size(); ++k ) {
                        if ( badModes[j] == modes[k] ) {
                            return false;
                        }
                    }
                }
            }
        } else if ( !IsMiscGameMode( modes[i] ) ) {
            return false;
        }
    }

    // 5. Assign the output variable with the game mode and return true
    out = in;

    return true;
}

std::string ExtractGameMode( const std::string &gameName ) {
    std::vector<std::string> tokens = Tokenize( gameName );

    std::string gameMode;

    for ( size_t i = 0; i < tokens.size(); ++i ) {
        if ( IsValidGameMode( tokens[i], gameMode ) ) {
            return gameMode;
        }
    }

    return "";
}

int main( int argc, char *argv[] ) {
    InitializeModeCompatibilityMap();

    std::string gameName = "DotA v6.60 -RDEM USA/CA LC";
    std::string gameMode = ExtractGameMode( gameName );

    std::cout << "Name: " << gameName << std::endl;
    std::cout << "Mode: " << gameMode << std::endl;

    return 0;
}

输出:

Name: DotA v6.60 -RDEM USA/CA LC

Mode: rdem

---

如果有人想查看此代码并让我知道他们会更改什么，我们将不胜感激。

谢谢。

Answer 1

创建 bool 数组来复制您放入注释中的表格。除了用“true”或“false”代替“X”或空白(所以“true”表示模式组合有效，“false”表示无效)。

使用此表查找组合是否有效:

 bool IsSecondaryValidWithPrimary(unsigned int primaryIndex, unsigned int secondaryIndex)
 {
     static bool secondaryValidWithPrimary[numPrimaryModes][numSecondaryModes] = {...}

     if (primaryIndex < numPrimaryModes && secondaryIndex < numSecondaryModes)
     {
         return secondaryValidWithPrimary[primaryIndex][secondaryIndex]
     }
     else
     {
         //... this should never happen, throw your favorite exception
     }
 }

当然，这需要您将每个 2 个字符的字符串转换为要测试的正确数组索引。遍历所有可能的组合并检查其是否有效。我怀疑您是否真的关心此设置中的性能，所以这应该能很好地工作。

对其他有效性检查(次要与另一个次要)或您具有兼容性规则的任何其他模式组合执行相同的操作。