使用istringstream丢失随机字符

Losing random characters with istringstream

本文关键字:随机 字符 istringstream 使用      更新时间:2023-10-16

我正在将文本文件的行读取为变量myline,然后我试图使用istringstream对这些行进行标记。然而,似乎我失去了随机字符从原始文本文件字符串。

cout<< myline << buff << flush; //print original text file line
istringstream iss(myline);
string sub;
while (iss >> sub) {
cout << "[" << sub << "]" << endl;
} 

如果您查看我的输出,您可以看到我从文本文件中得到了正确的字符串,但是当我使用istringstream然后打印单个标记(在[]括号中看到)时,一些标记被过早地截断了。

#include <iostream>
[#include]
[<iostream]
#include <sstream>
[#include]
[<sstream>]
using namespace std;
[using]
[namespace]
[st]
int main()
[int]
[main(]
{
    string str("   SOME  LONG    STRINGtwithnSPACES    ");
[string]
[str("]
[SOME]
[LONG]
[STRINGtwithnSPACES]
    istringstream iss(str);
[istringstream]
[iss(str);]
    string s;
[strin]
    while (iss >> s) {
[while]
[(iss]
[>>]
        cout << "[" << s << "]" << endl;
[cout]
[<<]
["["]
[<<]
[s]
[<<]
["]"]
[<<]
[e]
    }
    return 0;
[retur]
}
有谁知道我做错了什么吗?提前感谢!

编辑:这是一个可以完全编译的代码版本。您可以使用任何文本文件

运行它。
#include <cstring>
#include <cstdio>
#include <iostream>
#include <iomanip>
#include <string>
#include <sstream>
#include <vector>
#include <algorithm>
#include <iterator>
using namespace std;
class MyFileReader {
public:
    //constructor
    MyFileReader(const char* p);
    //destructor
    ~MyFileReader();
    //getLine()
    int getLine(char *buffer, int size);
    //getCurrentLineNumber()
    int getCurrentLineNumber();
    void tokenizeLine(vector<string>& vec);
    FILE * pFile;
};
    //constructor
    MyFileReader::MyFileReader(const char* p) {
        pFile = fopen(p, "r");
    }
    //destructor
    MyFileReader::~MyFileReader() {
        fclose(pFile);
    }
    //getLine()
    int MyFileReader::getLine(char *buffer, int size){
        char *out = fgets(buffer, size, pFile);
        if (out==NULL) {
            return -1;
        }
        char *pch = strpbrk(out,"n");
        if (pch != NULL) {
            return 1;
        }
        else {
            return 0;
        }
    }
    int MyFileReader::getCurrentLineNumber() {
        static int mynumber=2;
        return mynumber++;
    }
    //tokenizeLine
    void MyFileReader::tokenizeLine(vector<string>& vec) {
        string myline("");
        char buff[10];
        while (1) {
            int result = getLine(buff, sizeof(buff));
            if (result == -1 ) {
                if (myline.length() > 0) 
                    cout << myline << flush;
            break;
            }
            else if (result == 0) {
                myline += buff;
            }
            else if (result == 1) {
                cout<< myline << buff << flush;
                istringstream iss(myline);
                string sub;
                while (iss >> sub) {
                    cout << "[" << sub << "]" << endl;
                } 
                myline = "";
            }
            else {
                printf("PANIC");
            }
            }
            return;
        }
    int main(int argc, char **argv) {
    vector<string> v;
    const char *filename = argv[1];
    MyFileReader f(filename);
    f.tokenizeLine(v);
    return 0;
    }
为了生成上面的输出,我在下面运行它:
#include <iostream>
#include <sstream>
using namespace std;
int main()
{
    string str("   SOME  LONG    STRINGtwithnSPACES    ");
    istringstream iss(str);
    string s;
    while (iss >> s) {
        cout << "[" << s << "]" << endl;
    }
    return 0;
}

错误在这里:

else if (result == 1) {
            cout<< myline << buff << flush;
            istringstream iss(myline);
            string sub;
            while (iss >> sub) {
                cout << "[" << sub << "]" << endl;
            } 
            myline = "";
        }

如果result == 1,意味着buff 包含 n,并不意味着只包含 n。例如,如果包含n,则删除缓冲区。因此,如果该行恰好有n*10 (sizeof buffer)个字符,您的代码可以工作,否则,该行的最后一个字符不会复制到myline,而是删除。

快速修复方法是:

    else if (result == 1) {
            myline += buff; // copy the rest of the line into `myline`
            cout<< myline << flush; // buff now is part of myline
            istringstream iss(myline);
            string sub;
            while (iss >> sub) {
                cout << "[" << sub << "]" << endl;
            } 
            myline = "";
        }

尽管您可能想要考虑从缓冲区中删除n,如:

int MyFileReader::getLine(char *buffer, int size){
    char *out = fgets(buffer, size, pFile);
    if (out==NULL) {
        return -1;
    }
    //char *pch = strpbrk(out,"n");
    char *pch = strchr(out,'n'); // no need to search for a string
    if (pch != NULL) {
        *pch = ''; // drop the 'n'
        return 1;
    }
    else {
        return 0;
    }
}

您必须将cout<< myline << flush;更改为cout<< myline << endl;


除了这个错误,请考虑使用ifstream:

#include <iostream>
#include <fstream>
#include <sstream>
using namespace std;
int main()
{
    ifstream file("test.txt");
    if(!file)
    {
        /* error */
    }else
    {
        string line;
        while(getline(file, line))
        {
            istringstream iss(line);
            string s;
            while (iss >> s) {
                cout << "[" << s << "]" << endl;
            }
        }
    }
}

您的行缓冲区只有10字节长。

void MyFileReader::tokenizeLine(vector<string>& vec) {
        string myline("");
        char buff[10];// this is too short
....

编辑

正如Dyp正确指出的那样,当您在输入文件中检测到n时,追加逻辑是不正确的。