BIND 10 #449: Implementation of the resolver cache

[BIND 10 Development]

#449: Implementation of the resolver cache
-------------------------------------+-------------------------------------
                 Reporter:           |                Owner:  jelte
  zhanglikun                         |               Status:  closed
                     Type:  task     |            Milestone:  R-Team-
                 Priority:  major    |  Sprint-20110222
                Component:           |           Resolution:  complete
  resolver                           |            Sensitive:  0
                 Keywords:           |  Add Hours to Ticket:  0
Estimated Number of Hours:  8.0      |          Total Hours:  0
                Billable?:  1        |
                Internal?:  0        |
-------------------------------------+-------------------------------------

Comment (by zhanglikun):

 Replying to [comment:9 jelte]:
 > Replying to [comment:6 stephen]:

 > >
 > > '''message_entry.h'''[[BR]]
 > > Comment: the fact that the RRsetRef object is used here makes me
 wonder whether this should be part of the basic RRset class hierarchy.
 The comments in src/lib/dns/rrset.h suggests that the design is not final
 and should be revisited.  Thoughts?
 > >
 >
 > add to discussion on copy?

 No, it's only used by message entry, doesn't make sense to make it be part
 of RRset.
 your wondering may be caused by the similirity between RRsetRef and
 RRsetPtr, :)

 >
 > > genMessage(): the parameter description in the header suggests that
 "response" is the first argument and time_now the second.
 > >
 > > Presumably the reason that a number of methods are "protected" instead
 of "private" is to allow for testing?  If so, a one-line comment in the
 header file explaining that would be helpful.
 > >
 >
 > we'll have to ask likun

 Done

-- 
Ticket URL: <http://bind10.isc.org/ticket/449#comment:14>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development