Question

我不熟悉C ++文件阅读，但我通过pyspark完成了很多工作。所以现在我有一个txt文件，内容如下：

1   52  Hayden Smith        18:16   15  M   Berlin

2   54  Mark Puleo      18:25   15  M   Berlin

3   97  Peter Warrington    18:26   29  M   New haven

4   305 Matt Kasprzak       18:53   33  M   Falls Church

5   272 Kevin Solar     19:17   16  M   Sterling

6   394 Daniel Sullivan     19:35   26  M   Sterling

7   42  Kevan DuPont        19:58   18  M   Boylston

8   306 Chris Goethert      20:00   43  M   Falls Church

如您所见，有8列和351行（我只显示了8行），对于每一行，[0]是排名，[1]是BIB，[2]是名字，[3]是姓，[4]是时间，[5]是年龄，[6]是性，[7]是镇例如，第一行，1是排名，52是BIB，Hayden Smith是名字，18：16是时间，15是年龄，M是男性，柏林是城镇。

我有一个已排序的链接结构，名为：Class SortedLinked 和一个itemtype类，名为：Class Runner

您不必担心SortedLinked类。

Class Runner有四个私有属性：

string name, int age, int min, int sec

在我的驱动文件中，我可以这样做：

SortedLinked mylist                  // initialize a sorted list

Runner M("Jordan", 22, 20, 20)       // initialize a Runner called Jordan, who is 22 years old, and finished the race in 20 mins and 20 sec

mylist.add(M) //add Runner M into my sorted list

所以我需要阅读文本文件并创建一个Runner对象，其中包含跑步者的名字，年龄，分钟数和秒数。将该Runner插入已排序的链接列表。

所以，如果这是在pyspark，我可以这样做：

file=sc.textFile("hdfs")             //we usually use hdfs in pyspark

newfile = file.map(lambda line: line.split('\t')    //for each column, they are seperated by Tabs, except column[2][3] are separated by a space 

ColumnIneed = newfile.filter(lambda r: [r[2], r[3], r[4], r[5]]) // I only need the column [2][3][4][5]

mylist = ColumnIneed.collect()    // transform the RDD into a list

Then I can just transform every row into a Runner object.

但是，在C ++中我只知道这个：

ifstream, infile;

string s, sAll;

if(infile.is_open())
{

   while(getline(line, s))

   {

      s = s.rstrip('\n')     //does NOT work in C++
      name, age, time = s.split('\t')  // Does NOT work in C++ and I dont need all the columns

所以，问题：

1，我需要访问每一行，并删除换行符

2，我只需要列[2] [3] [4] [5] //每列用Tabs分隔

3，列[4]是时间，这是文本文件中的字符串，我需要拆分＆＃34;：＆＃34;并投入一分钟和几秒钟

4，列[2] [3]是名字和姓氏，我需要将它们组合成字符串名称

5，列[2] [3]用空格分隔

理想情况下，我想这样做：

while(I need a loop)
{

   eachline = access each line;

   eachline.strip('\n')  //strip newline

   eachline.split('\t')  //split Tabs

   string name = eachline[2][3];

   string time = eachline[4];

   int min;

   int sec;

   min, sec = time.split(':")

   int age = eachline[5];

   Runner M(name, age, min, sec)    //I don't know if this works, because it looks like you are overwriting the Runner M each time you access a new line. 

   mylist.add(M)      //add M into my linkedlist, this step you don't need to worry, I already finished. 

}

如果你有更好的方法，我会非常感激。

Answer 1

一些代码段

    std::ifstream in;
    in.open(/*path to file*/);
    std::string line;
    if(in.is_open())
    {
        while(std::getline(in, line)) //get 1 row as a string
        {
            std::istringstream iss(line); //put line into stringstream
            std::string word;
            while(iss >> word) //read word by word
            {
                std::cout << word << std::endl;
            }
            /*
            int row;
            int age;
            std::string name;
            iss >> row >> age >> name; // adopt to your input line
            Runner M(name, age, min, sec); //common agreement - variables shouldn't start with capital, you don't override M, each time u create new local variable type of Runner, then you put copy of M into some container, M gets destroyed at the end of the block, probably you could use movement semantic, but you need C++ basics first    
            mylist.add(M);
            */
        }
    }

C ++文件读取和拆分列

1 个答案: