Thread: Telnet Email problem

I'm no expert, but I think it's better to create a batch file intead of a C++ Project

well, this is just my opinion, perhaps lame,
you can create a text file, with batch file commands,
make your program execute the batch file,
execute it,
then delete,
you'll need either fstream or io.h

I need it to be in C++ because I want the user to be able to edit what is in the batch file. i.e.: the address that the email is being sent to, and the message.

Ok, but how do I code that? I can make it a string, but I get an error message anyway I try to do it:

Code:

#include <iostream>
#include <string>
using namespace std;
int main()
{
    string cmd;
    cout<<"Enter the command: ";
    cin>>cmd;
    system(cmd); //or system('cmd'); error either way
    system("pause");
}

Or you can always do Shell Execute

Code:

//prototype
ShellExecute(NULL, "open", /*Your .bat here: */, NULL, NULL, SW_SHOWNORMAL);

Your original attempt fails because telnet does not work the way you seem to think it does. Creating a batch file to do it will result in the same problem.

When you start the telnet program, everything else stops and waits for the telnet program to finish before moving on. That means any other programs, including your program or the batch file you are trying to create, are waiting. When/if telnet ever ends, the rest of the commands will execute, but they won't go to telnet since that will have finished running already.

If you take your original program

Code:

#include <iostream>
using namespace std;
int main()
{
    system("telnet mail.mycbsi.com 25");
    system("helo x");
    system("mail from: <[email protected]>");
    system("rcpt to: <[email protected]>");
    system("data");
    system("Test Message");
    system(".");
    system("quit");
    system("pause");
}

and run it again, when the

Code:

220 pbrane.mycbsi.net ESMTP Postfix

shows up, start typing your comamnds again then quit. What you'll see is that telnet responds to your commands, and then you will get

Code:

'helo' is not recognized as an internal or external command, operable program or batch file.

(or whatever your dos box error is for the flavor of windows you are running) followed by the same error message for each command that follows until your C++ program is done.

If you need to make a program to do the e-mail over a telnet session, you're going to have to learn some sockets programming.

Also, using the system() function like that is asking for problems. It is a good habit to avoid getting into. The occasional system() call for debugging purposes when you are the only person running the program can sometimes prove useful, and there are other situations where the system() call might be warranted, but on the whole allowing another program to begin execution in the midst of your program running is opening things up for security vulnerabilities.

You could try open telnet throuh a pipe and pipe your commands to it. I don't know, never used telnet before. Should work if it's an executable.

Originally Posted by jafet

You could try open telnet throuh a pipe and pipe your commands to it. I don't know, never used telnet before. Should work if it's an executable.

Piping info into a program means to take some output and use it for some input. A common use is

Code:

type sometext.txt | more

(Note: the | is the vertical bar on the backslash key) In this example, the type program "types" all the information from the text file, and then the dosbox uses that as input for the More program. Find a rather large text file and type the above, replacing sometext.txt with the name of the text file, and then do it again without the | more and see what the difference is.

There are also other ways to pipe data, do a google search if you are interested.

jafet's suggestion is to take the list of commands that you want to go to the telnet program and pipe them in from the command line. To do that, you would do

Code:

type commands.txt | telnet mail.mycbsi.com 25

Assuming commands.txt are the commands you want to use, telnet would start and then process all the input from the commands.txt file. You could have your program write the commands to a file (filling in the particulars that will change each time, the answers to the questions the user will type in) and then type the commands.txt as above.

There are quite a number of gotchas that would accompany this kind of attempt to use telnet, there aren't many ways to handle errors and you wouldn't learn very much from this method in my opinion. Learning socket programming isn't much easier, but in the long run you would learn more about programming and be able to do more than just this one limited case.