This apparently happens when gtest-1.5 is in use. Patch was reviewed by
Thomas Markwalder.

This has the side-effect of requiring additional copies of RRTTL
to be constructed and is less optimal. See the ticket on why
boost::optional is problematic.

Conflicts:
	ChangeLog

Using EXPECT_EQ on class D2Dhcid failed under gtest 1.5
due to lack of "<<" operator.  Adding it solves this problem.

Adds the initial implementation of b10-dhcp-ddns NameChangeTransaction class.
This class is an abstract, state-model driven class which provides the basic
mechanics for performing DDNS updates based on a NameChangeRequest.

This initializes the basic data structures of the memory manager and
starts loading data into the segments.

Conflicts:
	ChangeLog
	src/lib/dhcpsrv/memfile_lease_mgr.cc
	src/lib/dhcpsrv/memfile_lease_mgr.h
	src/lib/dhcpsrv/tests/mysql_lease_mgr_unittest.cc

... and don't generate isc/server_common/bind10_server.py from
isc/server_common/bind10_server.py.in. This fixes a builddir vs srcdir
path failure for the isc.server_common module.

- Patch by David Carlier

Added virtual method for validating state handler map, added additional unit tests and commentary.

By mistake, the previous merge was missing the changes which addressed
review comments.

Don't LDADD the xfr library for the resolver.
This was probably just a copy and paste mistake.
I brought this up on jabber a few days ago.

Added methods to base class for supporting server selection from a list
of DNS servers.

Interim commit for 3086 which includes the preliminary implementation
of the base class, NameChangeTransaction.  b10-dhcp-ddns module will use
derivations of this class to carry out NameChangeRequests.