RFR 8180410: ByteArrayOutputStream should not throw IOExceptions

Roger Riggs Roger.Riggs at Oracle.com
Thu Mar 15 14:33:45 UTC 2018

Hi Brian,


line 163: ok but I don't think I would keep using the expanded 'byte 
array output stream" phrase
instead of the proper noun: ByteArrayOutputStream. (except for consistency)

In the Write.java test:

line 60:  Probably "-" in the message should be "--" for consistency

line 68:  Add the "e" Throwable to the thrown RuntimeException so it 
gets printed in a stack trace/message

Line 100:..  If this was a testng test, the Assert.assertEquals messages 
would conveniently
print expected and actual values.  I'd suggest converting it to use @run 

(Though I expect after these are debugged, they will never fail).

73: some formatting cleanup of the old code might be useful. (spaces 
around "=")

Regards, Roger

On 3/14/2018 8:48 PM, Brian Burkhalter wrote:
> https://bugs.openjdk.java.net/browse/JDK-8180410
> http://cr.openjdk.java.net/~bpb/8180410/webrev.00/
> This proposed patch would make the following changes:
> 1. Add a new method writeBytes(byte[]) which writes all supplied bytes but does not throw IOE.
> 2. Document some previously undocumented exceptions which can be thrown by write(byte[],int,int) and writeTo(OutputStream).
> 3. s/@exception/@throws/.
> The test is renamed from WriteBounds to Write with the addition of a test of write() and writeBytes().
> A CSR will be filed later.
> Thanks,
> Brian

More information about the core-libs-dev mailing list