BIND 10 #2310: In-memory version of ZoneFinder::findAtOrigin()

[BIND 10 Development]

#2310: In-memory version of ZoneFinder::findAtOrigin()
-------------------------------------+-------------------------------------
            Reporter:  jinmei        |                        Owner:
                Type:  task          |  jinmei
            Priority:  medium        |                       Status:
           Component:  data source   |  reviewing
            Keywords:                |                    Milestone:
           Sensitive:  0             |  Sprint-20130205
         Sub-Project:  DNS           |                   Resolution:
Estimated Difficulty:  5             |                 CVSS Scoring:
         Total Hours:  2.30          |              Defect Severity:  N/A
                                     |  Feature Depending on Ticket:
                                     |          Add Hours to Ticket:  0
                                     |                    Internal?:  0
-------------------------------------+-------------------------------------

Comment (by jinmei):

 Replying to [comment:14 vorner]:

 > That doesn't sound really nice. Having our own is probably better for
 such a short function.
 >
 > > So, I'm personally okay with using the library functions, but that's
 > > not about this specific case, and we should unify the policy and be
 > > consistent throughout our code.  For now I didn't touch it.
 >
 > Can you add an explanation why we don't use the htonl, please?

 Added, hoping that's what you intended.

 > > Do you mean storing TTL in the form of RRTTL in `ZoneData`?  If so,
 > > for in-memory it doesn't make much sense anyway because we mainly use
 > > it only for wire-format rendering (`getTTL()` (and `toText()` which
 > > uses the getTTL) isn't expected to be called except for rare cases).
 >
 > I don't mean all the TTLs. I meant the min_ttl one, because we construct
 the RRTTL every time it is requested, to be compared. But it's probably
 marginal, so we can ignore it.

 Ah, so you are referring to `createTTLFromData()` in zone_finder.cc.
 Yes, that's a bit of overhead, but `RRTTL` constructors (including
 this one) is generally cheap, and in practice this is requested only
 for SOAs in negative responses (not so uncommon though, especially for
 TLDs), so I think it's acceptable.  In this particular case I'd rather
 keep `ZoneData` smaller (and if we store only in the form of `RRTTL`
 we need to convert its value to the net byte order in rendering).

 So I've kept the code as is.

 > So, except for the comment about htonl, it is OK to merge. I don't think
 I need to see such simple comment, so merge afterwards.

 Okay, thanks, merge done.  Closing.

-- 
Ticket URL: <http://bind10.isc.org/ticket/2310#comment:16>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development