Hi Ted:
Thanks for your reply.
The first 5 failing test is due to a null value in the top level (when
constructing the object), though
not in the argument place. The test case generated by the tool reveals a
corner case like:
Closure var0 = null;
Closure[] var1 = new Closure[]{var0};
ChainedClo
But it looks to me like you have null values embedded in the inputs.
I think the same would happen if you called Arrays.sort and some of the
objects in the array being sorted were null.
Why do you think that non-nullity of the top level arguments is all that
matters?
On Tue, Aug 17, 2010 at 1:30
Please find it in the following url:
http://docs.google.com/leaf?id=0B_sGdVRvWdcXOGRhYmZkZjctODJhZS00Nzg5LThkN2EtMzZjZGY5ZjIyNTRk&hl=en&authkey=COXhhb8F
It should be accessible.
thanks
-Sai
On Tue, Aug 17, 2010 at 4:30 PM, Ted Dunning wrote:
> Attachments are stripped. Can you put the repor
Attachments are stripped. Can you put the reports onto a web-site?
On Tue, Aug 17, 2010 at 12:04 PM, Sai Zhang wrote:
> Hi all:
>
> I am now writing an automated test generation tool for Java, and use Apache
> Common Collections
> version 3.2 (latest release) as one of the experimental benchmark
Hi all:
I am now writing an automated test generation tool for Java, and use Apache
Common Collections
version 3.2 (latest release) as one of the experimental benchmarks.
According to the tool design and its internal checking mechanism, it
generates many failing tests.
I attach 10 (executable fa
