使用C++实现简单的文章生成器

class wordpair
{
private:
    string preword;            // 前缀
    map<string, int> sufwords; // 后缀，次数
    int count;                 // 总次数

public:
    wordpair(string pre);
    wordpair(string pre, string suf);
    wordpair(string pre, map<string, int> suf);
    ~wordpair();
    string getPreword() const;
    map<string, int> getSufwords() const;
    void setPreword(string pre);
    void setSufwords(map<string, int> suf);
    string toJson() const;

    void addSufword(string suf);
    string chooseSufword() const;
};

class createArticle
{
private:
    vector<wordpair> wordpairlist;
    string article;

public:
    createArticle();
    ~createArticle();

    void importWords(string filename, int len_pre = 1, int len_suf = 1);
    void exportWords(string filename);
    void addWordPair(string pre, string suf);
    void generateArticle(string startword, int lenout = 10000);
    void printArticle(string filename);
};

string wordpair::toJson() const
{
    string str = "\"";
    str += this->preword + "\" : {";

    for (auto &it : this->sufwords)
    {
        str += "\"" + it.first + "\"" + ":" + to_string(it.second) + ",";
    }
    str += "}";
    return str;
}

void wordpair::addSufword(string suf)
{
    for (auto &it : this->sufwords)
    {
        if (it.first == suf)
        {
            it.second++;
            return;
        }
    }
    this->sufwords[suf] = 1; // if the word is not in the map, add it with a count of 1
}

string wordpair::chooseSufword() const
{
    if (this->sufwords.size() == 1)//如果只有一个后缀词就直接输出，减少算力负担
    {
        return this->sufwords.begin()->first;
    }
    else
    {
        // 随机选择一个后缀词
        random_device rd;
        ranlux48 engine(rd());
        uniform_int_distribution<> dist(0, this->count);//在类中定义了count,这里就省掉了遍历
        int random_number = dist(engine);//产生一个随机数

        std::string result;
        for (auto &it : this->sufwords)//抽奖
        {
            if (random_number < it.second)
            {
                result = it.first;
            }
            else
                random_number -= it.second;
        }
        return result;
    }
}

vector<string> charlist = splitchar(filestr);//先将从文件读到的字符串分割
    string preword = "", sufword = "";
    for (int i = 0; i < charlist.size() - len_suf - len_pre; i++)//每次向后移动一个字符，进行切割
    {
        preword = "", sufword = "";
        for (int j = i; j < i + len_pre + len_suf; j++)
        {
            if (j - i < len_pre)
            {
                preword += charlist[j];//从第i个字符开始，到第i+len_pre个字符连接起来作为前缀
            }
            else
            {
                sufword += charlist[j];//从第i+len_pre个到字符开始，到第i+len_pre+len_suf个字符连接作后缀
            }
        }
        this->addWordPair(preword, sufword);//添加进wordpairlist
    }

/*
startword——启动词
lenout——长度限制（避免无限循环）
*/
void createArticle::generateArticle(string startword, int lenout)
{
    this->article += startword;
    bool stop; // 加一个停止标志，当无法匹配到前缀时停止
    int prewordlen = this->wordpairlist.front().getPreword().length();
    int sufwordlen = this->wordpairlist.front().getSufwords().begin()->first.length();
    string lastword;
    for (int i = 0; i < lenout; ++i)
    {
        stop = true;
        if (this->article.length() >= prewordlen) // 如果文章长度大于词对中前缀词的长度，则直接拼接
        {
            lastword = this->article.substr(this->article.length() - prewordlen, prewordlen);//article最后的len_pre个字符，作为前缀
            for (auto &it : this->wordpairlist)
            {
                if (it.getPreword() == lastword)//通过lastword匹配词对
                {
                    this->article += it.chooseSufword();
                    stop = false;
                    break;
                }
            }
            if (stop)//遍历了一边词对的list没有匹配的词对时，退出循环
                break;
        }
        else//启动词长度小于词对前缀的情况，例如词对分割为3+2时，启动词长度为2,小于前缀长度3,无法正常拼接，于是走此处
        {
            lastword = this->article;
            for (auto &it : this->wordpairlist)//同上遍历
            {
                int position = it.getPreword().find(lastword);
                if (position != string::npos)
                {
                    this->article += (it.getPreword() + it.chooseSufword()).substr(position+lastword.length(), sufwordlen);//先将前后缀连接，再从匹配到的位置开始截取
                    stop = false;
                    break;
                }
            }
            if (stop)
                break;
        }
    }
}