比较简单的html语言识别 ，反正就暴力处理吧，进一步的话可以把标签都保存下来，判断匹配问题，题目没要求就没写
输入

<html>
i like the book <tag> this is a tree\n  abs </tag>like father like son
</html>

输出：
i like the book this is a tree
abs like father like son

#include <iostream>
#include <bits/stdc++.h>
using namespace std;
vector<string>ans;
string s,all;
vector<string>tag;
bool check(string ss)
{
    if(ss.find("\n"))
    {
        return true;
    }
    else return false;
}
int main()
{

    while(getline(cin,s))
    {
        all+=s;
    }
    //cout<<endl;
    //cout<<all<<endl;
    string k,t;
    int flag=1;
    for(int i=0;i<all.size();i++)
    {
        if(all[i]!='<'&&flag)
        {
            k+=all[i];
        }
        else if(all[i]=='<')
        {
            flag=0;
            if(k.size())
            {
               ans.push_back(k);
            }
        }
        else if(all[i]=='>')
        {
            flag=1;
            tag.push_back(t);
            t.clear();
            k.clear();
        }
        else if(all[i]!='<'&&!flag)
        {
            if(all[i]!='/')
                t+=all[i];
        }
    }

    for(int i=0;i<ans.size();i++)
    {
        if(check(ans[i]))
        {
            for(int j=0;j<ans[i].size();j++)
            {
                if(ans[i][j]=='\\')
                {
                    cout<<endl;
                    j++;
                }
                else cout<<ans[i][j];
            }
        }
        else cout<<ans[i];
    }
    return 0;
}