Strange problems with string class

Hey Guys. This is a weird one. Basically what I am doing is reading a string from the file using getline(file, buffer) where buffer is a C++ string. For example: lets say the file contains the string "hello" without the quotes on one line. After I read the file and print out the buffer it says "hello" (cout << buffer) like it should..BUT when I do this:

buffer == "hello" , the answer is false. Why? I can't seem to see why. Furthermore, when I do cout << buffer << "." << endl;

I get this output:

.ello

Why is this happening? In case you are wondering the file does not contain any blank spaces or lines after the string "hello". Any ideas?

Can you post your code, it sounds like it is only a few lines long.

Here is my code. I am using g++ on ubuntu linux btw.

Code:

---

#include <iostream>
#include <string>
#include <fstream>
using namespace std;

int main()
{
  string buffer;
  ifstream file;
  file.open("tfile.txt");
  getline(file, buffer);
  cout << buffer<<endl;
  if(buffer == "hello")
  {
    cout << "match" <<endl;
  }
  return 0;
}

---

Change this:

Code:

---

  getline(file, buffer);
  cout << buffer<<endl;

---

to this:

Code:

---

  if (!file.is_open())
  {
    cout << "Error opening file.";
    return 1;
  }

  getline(file, buffer);
  cout << "-->" << buffer << "<--" << endl;

---

What's the output?

The output is:

<--hello

Your program returns "hello" and "match" for me, and appears to contain no errors.

I suspect tfile.txt does not contain what you think it contains. Your note about getting ".ello" for output makes it look like you've got a "\r" in the buffer somehow - check your file. It should be 5 bytes, exactly. (Use a hex editor to see what's really in the file.)

I think you are right. My file is 7 bytes somehow. I don't quite understand why, since hello is the only thing in it. I examined it in hex editor, and it appears to confirm it. I attached the file here too. I don't understand why its there, and how it gets there, very weird.

Yeah, thanks everybody I fixed it.

I think it's because you switched platforms, specifically, line endings. If you created tfile.txt under DOS or Windows and then ran your program under Linux (i.e., Ubuntu), that's exactly what would happen.

Here's why. From http://en.wikipedia.org/wiki/Newline

Code:

---

    * LF:    Multics, Unix and Unix-like systems (GNU/Linux, AIX, Xenix, Mac OS X, etc.), BeOS, Amiga, RISC OS, and others
    * CR+LF: DEC RT-11 and most other early non-Unix non-IBM OSes, CP/M, MP/M, MS-DOS, OS/2, Microsoft Windows
    * CR:    Commodore machines, Apple II family and Mac OS up to version 9

---

DOS line endings are CR+LF, so that's what your file would contain. Then, reading it under Linux, which has LF line endings, the CR would be part of the previous line. With Macs, the opposite problem would occur: each line (except the first) would start with a LF.

Quote:

---

Originally Posted by dwks

With Macs, the opposite problem would occur: each line (except the first) would start with a LF.

---

Not quite, with crusty old Macs the opposite problem would occur. Mac OS X is firmly in the Unix camp, as evidenced by your list.

Yes, you're right. I didn't know that . . . even though it was in my post. :D