RFR: (small): JDK-8190307: SA: Sanity tests for the clhsdb commands: universe, intconstant, type

[David Holmes]

Hi Jini,

OutputAnalyzer has a bunch of methods that deal with checking return 
values and output contents, and throwing exceptions (not assertion 
failures!) so that you don't need to manually do all that. It also 
prints diagnostics to aid with debugging when things do not contain what 
is expected.

Looking at the individual tests:

TestIntConstant.java

  92 
Asserts.assertTrue(output.contains("CollectedHeap::G1CollectedHeap 2"));
   93         Asserts.assertTrue(output.contains("RUNNABLE 2"));
   94         Asserts.assertTrue(output.contains("_thread_uninitialized 
0"));
   95         Asserts.assertTrue(output.contains("intConstant 
_temp_constant 45"));

Apart from  the last line this expected output looks very strange. If 
this indeed what we expect from issuing those commands we should 
document/comment what corresponds to what.

Plus this assumes G1 is being used but Lingered App is started using the 
test run vmOptions AFAICS so it would use whatever GC were passed 
through, wouldn't it?

---

TestType.java

The expected output is not quite so strange, but still could do with 
some commentary.

   90         Asserts.assertTrue(output.contains("type G1CollectedHeap 
CollectedHeap"));

Same comment about assuming G1.

Thanks,
David
------

On 30/10/2017 1:44 PM, Jini George wrote:
> Hello,
> 
> We have been working on writing sanity tests for the various jhsdb 
> clhsdb commands of the SA to improve the robustness of the SA. As a part 
> of this, here is a webrev for sanity tests for 3 of the clhsdb commands:
> 
> Bug ID: https://bugs.openjdk.java.net/browse/JDK-8190307
> webrev: http://cr.openjdk.java.net/~jgeorge/8190307/webrev.00/
> 
> I would like to request for reviews for this simple addition.
> 
> Thank you,
> Jini.