why BufferedReader.readLine can read a line which doesn't have a line separator

Question

Reads a line of text. A line is considered to be terminated by any one of a line feed ('\n'), a carriage return ('\r'), or a carriage return followed immediately by a linefeed.------javadoc 1.8

Then I have a text file like this:

the first line
the second line

note: the last character of the seond line is 'e' that is to say there dont exist carriage return.

then here is my demo code.

public void process() throws IOException{
    BufferedReader br = new BufferedReader(new FileReader("demo.txt"));
    String line;
    while((line=br.readLine())!=null){
        System.out.println(line);
    }
    br.close();
}

the real output:

 the first line
 the second line

then my question is that why the readLine method can get the second line for it doesnt have line-separator (\n or \r or \n\r).
I know there exist a end of file (EOF), but it seemed the javadoc dont tell the EOF is also the line-separator explicitly.

If I use Scanner instead of BufferedReader the code as below:

public void testScan() throws IOException{
    Scanner scan = new Scanner(new FileInputStream("demo.txt"));
    String line;
    while((line=scan.nextLine())!=null){
        System.out.println(line);
    }
    scan.close();
}

then the output would be:

the first line
the second line
Exception in thread "main" java.util.NoSuchElementException: No line found
    at java.util.Scanner.nextLine(Scanner.java:1540)
    at com.demo.Demo.testScan(Demo.java:39)
    at com.demo.Demo.main(Demo.java:49)

Answer 1

Because it's programmed that way.

Really, it's what the user of the method wants. If the last line is missing a line separator at the end, it will read until EOF so that no data is lost. You don't want to lose an entire line because of a missing line separator.

Practically all similar functions work in the same way. For example, if you're looking at the fgets() function in the C library, it will also work that way. So does f.readline() in Python.

Edit: the Scanner works also in the similar way, but the difference is that a Scanner throws an exception whereas BufferedReader returns null when all lines have been read.