Hi Brian,
Reconsidering...
If the current line number starts at zero and is incremented only by CR/LF
then my example program is wrong. It called getLineNumber *after* reading
the line and it should have called it *before*.
So I can support your original patch (modulo not using AtomicBoolean).
(Though my suggestion about doing all the work of keeping track of the
line number in BufferedReader
is still a valid though separate suggestion).
Roger
On 9/9/19 5:02 PM, Brian Burkhalter wrote:
If we wanted to go the route of changing the spec to match the behavior, the
additional verbiage below is one possibility.
Thanks,
Brian
+++ b/src/java.base/share/classes/java/io/LineNumberReader.java
@@ -35,19 +35,20 @@
* <p> By default, line numbering begins at 0. This number increments at every
* <a href="#lt">line terminator</a> as the data is read, and can be changed
* with a call to {@code setLineNumber(int)}. Note however, that
* {@code setLineNumber(int)} does not actually change the current position in
* the stream; it only changes the value that will be returned by
* {@code getLineNumber()}.
*
* <p> A line is considered to be <a id="lt">terminated</a> by any one of a
* line feed ('\n'), a carriage return ('\r'), or a carriage return followed
- * immediately by a linefeed.
+ * immediately by a linefeed. For {@link #readLine()} only, the end of the
+ * stream is also considered to be a line terminator.
public class LineNumberReader extends BufferedReader {
/** The current line number */
private int lineNumber = 0;
@@ -183,19 +184,21 @@
}
}
return n;
}
}
/**
* Read a line of text. Whenever a <a href="#lt">line terminator</a> is
- * read the current line number is incremented.
+ * read the current line number is incremented. Unlike for the other
+ * {@code read} methods of this class, reaching the end of the stream
+ * will also increment the current line number.
*
* @return A String containing the contents of the line, not including
* any <a href="#lt">line termination characters</a>, or
* {@code null} if the end of the stream has been reached
*
* @throws IOException
* If an I/O error occurs
*/
public String readLine() throws IOException {
On Sep 9, 2019, at 12:02 PM, Brian Burkhalter <brian.burkhal...@oracle.com>
wrote:
On Sep 9, 2019, at 7:35 AM, Roger Riggs <roger.ri...@oracle.com
<mailto:roger.ri...@oracle.com>> wrote:
I would lean toward updating the spec to reflect the current implementation.
It seems strange however that if one read an entire stream using read() in one
case and readLine() in another that the results would differ.
