How can I compare the third token from each line using 'getline()'? - c++

Question

Input file: In each row, there's an entry that is a pair of ID - name - GPA. Values are tab-delimited.
20210001 Bill 3.61
20210002 Joe 3.21
20210003 Royce 4.32
20210004 Lucy 2.21
I have to rearrange this file sorted by the GPA (in decreasing order).
#include <iostream>
#include <fstream>
#include <sstream>
#include <string>
#include <vector>
using namespace std;
int main() {
ifstream inputfile("input.txt");
ofstream outputfile("output.txt");
if (inputfile.fail()) {
cout << "Cannot open inputfile" << endl;
}
if (outputfile.fail()) {
cout << "Cannot open outputfile" << endl;
}
if (inputfile.is_open()) {
string line;
while (getline(inputfile, line)) {
string token;
stringstream ss(line);
while (getline(ss, token, '\t')) {
}
}
}
inputfile.close();
outputfile.close();
return 0;
}
I'm not sure what to do next.

Answer 1

When doing I/O from/to streams (like file streams) it usually makes it easier to create a class to keep the data for each record in the file and to create overloads for operator>> (in) and operator<< (out).
Example:
#include <algorithm>
#include <fstream>
#include <iostream>
#include <sstream>
#include <string>
#include <vector>
// one line in the file could possibly have this representation in your program:
struct record {
std::uint32_t ID;
std::string name;
double GPA;
};
// an overload to read one line of data from an istream (like an ifstream) using getline
std::istream& operator>>(std::istream& is, record& r) {
if(std::string line; std::getline(is, line)) {
std::istringstream iss(line);
if(not (iss >> r.ID >> r.name>> r.GPA)) {
is.setstate(std::ios::failbit);
}
}
return is;
}
// an overload to write one line to an ostream (like an ofstream)
std::ostream& operator<<(std::ostream& os, const record& r) {
return os << r.ID<< '\t' << r.name << '\t' << r.GPA << '\n';
}
With that boilerplate in place, making the actual program becomes easy.
int main() {
std::ifstream inputfile("input.txt");
// read all records from the file into a vector
std::vector<record> records(
std::istream_iterator<record>(inputfile),
std::istream_iterator<record>{}
);
// sort the records according to GPA
// if you want a decending order, just make it return rhs.GPA < lhs.GPA;
std::sort(records.begin(), records.end(),
[](const record& lhs, const record& rhs) {
return lhs.GPA < rhs.GPA;
}
);
std::ofstream outputfile("output.txt");
// put the result in the output file
std::copy(records.begin(),
records.end(),
std::ostream_iterator<record>(outputfile));
}
There may be a few things in this answer that you haven't seen before. I'll list resources for those I anticipate may require some reading:
operator<< / operator>> overloading. See Stream extraction and insertion
std::istringstream - You put a std::string in it - and then it behaves like any other std::istream (like a std::ifstream).
std::istream_iterator/std::ostream_iterator
std::sort / std::copy are two of the many algorithms in the standard library.
std::vector - A class template acting as a dynamic array where you can add and remove data, like record in this answer.
Lambda expressions - In this answer it's the functor [](const record& lhs, const record& rhs) { ... } used to sort the records.

Answer 2

If you know exactly how many tokens are in a line:
You could simply getline() 3 times with the tab delimiter, and store the values separately.
string id;
string name;
string gpa;
getline(ss, id, '\t');
getline(ss, name, '\t');
getline(ss, gpa, '\t');
This logic would live within your loop that iterates over the lines in the file.

Answer 3

You could use a struct to contain all your fields:
struct user
{
int id; string name; double point;
};
Then insert all of them into a std::vector and finally use sort() with the comp parameter to sort by points.
Code:
#include <iostream>
#include <fstream>
#include <sstream>
#include <string>
#include <vector>
#include <algorithm>
#include <iomanip>
using namespace std;
struct user
{
int id; string name; double point;
};
vector<user> users;
void stripInfoFromString(string inp)
{
stringstream cur(inp);
user curUser;
cur >> curUser.id >> curUser.name >> curUser.point;
users.push_back(curUser);
}
bool compareUser(user x, user y)
{
return x.point < y.point;
}
int main()
{
string a1 = "20210001 Bill 3.61";
string a2 = "20210002 Joe 3.21";
string a3 = "20210003 Royce 4.32";
string a4 = "20210004 Lucy 2.21";
stripInfoFromString(a1);
stripInfoFromString(a2);
stripInfoFromString(a3);
stripInfoFromString(a4);
sort(users.begin(), users.end(), compareUser);
cout << fixed << setprecision(2);
for (user cur : users)
{
cout << cur.id << " " << cur.name << " " << cur.point << "\n";
}
}
Output:
20210004 Lucy 2.21
20210002 Joe 3.21
20210001 Bill 3.61
20210003 Royce 4.32
I used standard input/output to minimize the code, you can simply switch for file inputs easily.
More info:
struct : https://en.cppreference.com/w/c/language/struct
sort() : https://en.cppreference.com/w/cpp/algorithm/sort
Also, see here why is using namespace std; considered bad practice.

Answer 4

I would suggest defining a struct to hold the 3 tokens, and then create a std::vector holding instances of that struct for each line. You can then sort that vector on the 3rd token. You already have <vector> in your header includes.
#include <iostream>
#include <fstream>
#include <sstream>
#include <string>
#include <vector>
#include <algorithm>
using namespace std;
struct entry
{
int id;
string name;
double gpa;
};
int main() {
ifstream inputfile("input.txt");
ofstream outputfile("output.txt");
vector<entry> entries;
if (inputfile.fail()) {
cout << "Cannot open inputfile" << endl;
}
if (outputfile.fail()) {
cout << "Cannot open outputfile" << endl;
}
string line;
while (getline(inputfile, line)) {
istringstream iss(line);
entry e;
string token;
getline(iss, token, '\t');
e.id = stoi(token);
getline(iss, e.name, '\t');
getline(iss, token, '\t');
e.gpa = stod(token);
/* alternatively:
iss >> e.id >> e.name >> e.gpa;
*/
entries.push_back(e);
}
inputfile.close();
outputfile.close();
sort(entries.begin(), entries.end(),
[](const entry &e1, const entry &e2){
return e1.gpa > e2.gpa;
}
);
for (const entry &e : entries) {
outputfile << e.id << '\t' << e.name << '\t' << e.gpa << '\n';
}
return 0;
}
Demo