There is however this problematic statement in the PrintStream class doc that neither of the two alternative versions of the fix addresses:
"Two other features are provided as well. Unlike other output streams, a PrintStream never throws an IOException; instead, exceptional situations merely set an internal flag that can be tested via the checkError method.” In both cases this would need to be changed as it is incorrect. Brian > On Jul 15, 2019, at 10:14 AM, Brian Burkhalter <brian.burkhal...@oracle.com> > wrote: > > Here is an alternative version which adds a writeBytes(byte[]) method instead > of overriding write(byte[]): > > http://cr.openjdk.java.net/~bpb/8187898/webrev-alt.00/ > <http://cr.openjdk.java.net/~bpb/8187898/webrev-alt.00/> > <http://cr.openjdk.java.net/~bpb/8187898/webrev-alt.00/ > <http://cr.openjdk.java.net/~bpb/8187898/webrev-alt.00/>> > > This has the advantage of allowing new code to call writeBytes() without a > try-catch block without introducing a compatibility issue for code which is > already calling write(byte[]) in a try-catch block.
