BIND 10 #3264: Refactor d2 unit tests to use new new TransactionTest class

[BIND 10 Development]

#3264: Refactor d2 unit tests to use new new TransactionTest class
-------------------------------------+-------------------------------------
            Reporter:  tmark         |                        Owner:  tmark
                Type:  enhancement   |                       Status:
            Priority:  medium        |  reviewing
           Component:  dhcp-ddns     |                    Milestone:  DHCP-
            Keywords:                |  Kea1.0
           Sensitive:  0             |                   Resolution:
         Sub-Project:  DHCP          |                 CVSS Scoring:
Estimated Difficulty:  0             |              Defect Severity:  N/A
         Total Hours:  5.5           |  Feature Depending on Ticket:
                                     |          Add Hours to Ticket:  .5
                                     |                    Internal?:  0
-------------------------------------+-------------------------------------
Changes (by marcin):

 * owner:  marcin => tmark


Comment:

 Replying to [comment:9 tmark]:
 > Replying to [comment:8 marcin]:
 > > Reviewed commit faf86ad3a04fc7fca0af2b9d19abba0fa1d94387
 > >
 > > Changes are pretty straight forward. Just some minor things...
 > >
 > > '''nc_trans_unittests.cc'''
 > > doOneExchange should be documented
 > >
 >
 > Oops. Done.

 The parameters are not documented :). Sorry to be picky. If you ever
 decide to move this class out of the anonymous namespace the doxygen will
 complain about not documented parameters.
 >
 >
 > > sendUPdateTimeout: I have to admit I was not familiar with the
 !''ASSERT_NO_FATAL_FAILURE!'' macro. I greped the bind10 code and the only
 occurences I found were in nc_trans_unittests.cc. Since, it is the only
 place where it is used, it would be useful to have some short explanation
 why it has to be used. I am surprised that we never used it anywhere else.
 >
 > If method-A calls method-B and an ASSERT inside method-B fails, the
 failure will cause method-B to be aborted, method-A will continue to
 execute.  Using this macro around method-B inside method-A will cause
 method-A to abort also.   It is a bit like the difference between EXPECT
 and ASSERT.

 I was rather thinking about putting a comment into the code. But I leave
 it up to you.

 >
 > The method, doOneExchange, can suffer fatal assertions which invalidate
 not only it but the calling method or test as well.  In other words, if
 the doOneExchange blows up the rest of test is pointless.  I use
 ASSERT_NO_FATAL_FAILURE to abort the test immediately.
 >
 >
 > >
 > > Please check the alignment of parameters passed to ''doOneExchange''.
 > >
 >
 > Got it

 ok.

 >
 > > Why did you check in the src/bin/d2/tests/test_data_files_config.h.in
 file?
 > >
 > >
 >
 > I fixed the copyright.
 >

 ok.

 I don't need to see this ticket again. If you concur with comments above,
 please do appropriate modification before the merge.

-- 
Ticket URL: <http://bind10.isc.org/ticket/3264#comment:11>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development