BIND 10 #2751: add "subtract RDATA" interface to RdataSet

[BIND 10 Development]

#2751: add "subtract RDATA" interface to RdataSet
-------------------------------------+-------------------------------------
            Reporter:  jinmei        |                        Owner:  muks
                Type:  task          |                       Status:
            Priority:  medium        |  reviewing
           Component:  data source   |                    Milestone:
            Keywords:                |  Sprint-20130917
           Sensitive:  0             |                   Resolution:
         Sub-Project:  DNS           |                 CVSS Scoring:
Estimated Difficulty:  5             |              Defect Severity:  N/A
         Total Hours:  0             |  Feature Depending on Ticket:
                                     |  shared memory data source
                                     |          Add Hours to Ticket:  0
                                     |                    Internal?:  0
-------------------------------------+-------------------------------------
Changes (by vorner):

 * owner:  vorner => muks


Comment:

 Hello

 Replying to [comment:9 muks]:
 > * In the `RdataSetTest.subtract` unittest, why are TTLs of 1076895760
 >   used? I see the rest of the files use that TTL too. It's a distraction
 >   (we wonder why that TTL was picked), and something simple like 3600
 >   would be better.

 Well, part of the tests use that magic number to ensure all bytes of TTL
 are properly stored. Because the tests for merge and subtract use the same
 auxiliary functions, they share the constant.

 Do you think it is worth making the check functions more complex to have
 3600 in places we don't need the 4-bytes check?

 > * It's a shame that `subtractIterate()'ion` is an n^2 operation as for
 >   each RDATA to subtract, we have to go through the entire list of RDATA
 >   for the other rrset. It's not this branch's fault, but just an
 >   observation.

 It is small glitch on the elegance, I agree. But I don't think it is worth
 worrying about. The number of RRs in single RdataSet will probably be
 small, as well as the number of items we remove. And, the subtraction is
 not in performance-sensitive path.

-- 
Ticket URL: <http://bind10.isc.org/ticket/2751#comment:12>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development