RFR: 8041331 Cleanup files for jtreg and windows

Lance Andersen lance.andersen at oracle.com
Tue Apr 22 11:22:45 UTC 2014


Hi Stuart
On Apr 21, 2014, at 9:26 PM, Stuart Marks <stuart.marks at oracle.com> wrote:

> 
> 
> On 4/21/14 1:58 PM, Lance Andersen wrote:
>> Need a reviewer for some additional updates of the SQLException tests I pushed on Friday so that they play nicer WRT clean up on windows with some tools.
>> 
>> Did some minor refactoring as part of this to address some of Roger's earlier comments seeing I had to touch the tests.
>> 
>> The webrev is at http://cr.openjdk.java.net/~lancea/8041331/webrev.00/
> 
> Hi Lance,
> 
> Thanks for pointing me at this webrev.
:-)
> 
> Extracting all the serialization/deserialization code into a helper method createSerializedException() makes good sense. Nice use of generics, too.

Thank you.
> 
> I'd go farther and avoid creating a file in the first place. You can serialize to a ByteArrayOutputStream, extract the byte array, and deserialize from that. It's probably necessary to flush between serialization and deserialization. This avoids having to do any file cleanup at all. (The IOExceptions still have to be handled or declared, though.)

Roger had suggested it earlier and thought I would leave it but obviously it was the wrong choice :-)

I have switched this to use a  byte array vs file IO
> 
> You could go back and remove the fname argument from all callers, or you could leave it in as a label to use in messages if any error occurs in this routine.

removed the extra argument, netbeans was my friend here ;-)
> 
> One tiny nitpick, at SQLNonTransientConnectionExceptionTests.java line 170, an extra semicolon was inserted.

Thanks for catching that,  fixed

The revised webrev is at http://cr.openjdk.java.net/~lancea/8041331/webrev.01/

Best
Lance
> 
> Thanks,
> 
> s'marks
> 



Lance Andersen| Principal Member of Technical Staff | +1.781.442.2037
Oracle Java Engineering 
1 Network Drive 
Burlington, MA 01803
Lance.Andersen at oracle.com






More information about the core-libs-dev mailing list