BIND 10 #1793: update b10-auth command handler to reload in-memory/sqlite3

[BIND 10 Development]

#1793: update b10-auth command handler to reload in-memory/sqlite3
-------------------------------------+-------------------------------------
                   Reporter:         |                 Owner:  UnAssigned
  jinmei                             |                Status:  reviewing
                       Type:  task   |             Milestone:
                   Priority:         |  Sprint-20120417
  medium                             |            Resolution:
                  Component:         |             Sensitive:  0
  b10-auth                           |           Sub-Project:  DNS
                   Keywords:         |  Estimated Difficulty:  4
            Defect Severity:  N/A    |           Total Hours:  0
Feature Depending on Ticket:  xfr    |
  for in-memory                      |
        Add Hours to Ticket:  0      |
                  Internal?:  0      |
-------------------------------------+-------------------------------------
Changes (by vorner):

 * owner:  vorner => UnAssigned
 * status:  accepted => reviewing


Comment:

 It is ready for review. Instead of some kind of keeping the configuration
 with the data source, I look into the configuration in the loadzone
 command.

 But there are two FIXME comments added.

 One of them is a hack for the older tests to pass. This is because the
 tests don't provide the configuration I try to read (so I can decide it is
 not sqlite3). We can put it to a separate ticket or fix it in this branch,
 if the hack looks too ugly.

 The other is, the current configuration allows to have two IN in-memory
 data sources. I look for the first one only, so if the zone to be reloaded
 is in the second, it wouldn't work. But I think such configuration
 wouldn't work anyway, due to internal limitations.

 Anyway, I believe we'll want to modify the zone loading and configuration
 stuff anyway, because we'll be switching to the new datasources for
 database as well and we'll need a different way to configure them globally
 (so XfrOut and such can access them too). At that time, this'll get
 revisited.

 I think this doesn't need a separate changelog entry, this should be
 unified with the rest of loading-from-database stuff.

-- 
Ticket URL: <http://bind10.isc.org/ticket/1793#comment:7>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development