Scanning Text File For Email Addresses

Write a program that scans a text file for possible e-mail addresses. Addresses look like this:

someone@somewhere.net

Read tokens from the input file one by one using hasNext() and next(). With the default delimiters of Scanner, an entire e-mail address will be returned as one token. Examine each token using the indexOf() method of String. If a token contains an at sign @ followed some characters later by a period, regard it as a possible e-mail address and write it to the output file.

Programs such as this scan through web pages looking for e-mail addresses that become the targets of spam. Because of this, many web pages contain disguised e-mail addresses that can't easily be automatically extracted and above that is says to modify a program from our chapter which is this:

import java.util.Scanner;
import java.io.*;
class NamedFileInOut {
public static void main (String[] args) throws IOException {
int num, square;

[Code] ....

This is the first time hasnext() and next() have been introduced to me, so whereas scan.nextInt() looks for integers, does hasNext() looks for strings or characters? the wording in the exercise text is confusing to me. so basically i should create a text file with a ton of strings and within that jumble of text, stick a few email addresses; then when the program asks the user for the input file name, use that text file's name correct? pretty sure i have it up to that point, but using the indexOf() is what i'm having trouble wrapping my head around. do i look for the index of "@" and "."? i could conceptually see how i could say if the indexOf(".") is three spaces before the end then i know it's a .com or .net or .org or whatever. but how would i use indexOf("@") when the "user name" AND the "provider" (i.e. @yahoo or @google or @whatever) have an infinite number of lengths? we haven't had literally any discussion on input/output in classes so i am totally green to this.