GitLab

That was a hangover from the log4cxx implementation and appears not to
be needed in log4cplus.

Note: still need to add tests for LoggerImpl class.

With log4cplus, the output format of messages had changed.  This
corrects the test messages so that the tests passes.

Debug logging levels implemented as new log4cplus levels.  With
this change, most of the tests pass.  The ones that fail are caused
by the different form of output.

The code compilers and produces basic output.  Still to be done is
1) Add idea of debug levels.
2) Modify tests to copy with new format output.

with fixing conflicts in src/bin/xfrin/tests/xfrin_test.py
Also made a couple of cleanups to prevent disruption:
 - initialize s_Name::position in constructor
 - add tsigctx_mock.py to EXTRA_DIST

with fixing (trivial) conflicts in:
	src/lib/dns/python/Makefile.am
	src/lib/dns/python/tsigerror_python.cc
Also moved tsigerror_python_inc.cc to EXTRA_DIST because it doesn't have
to be compiled separately.

the python API (failing this will cause build errors on some systems)

turns out botan had a check for exactly this, so we don't need to reinvent our own

Added separate checks for compatibility with 1.8 and 1.9, see configure.ac
cryptolink now has 3 special cases depending on version

case for None returned

so that we can merge this branch as discussed in the planning call today

added tests for get_default_value() in ConfigData
removed the now unnecessary self.name_str in ZoneInfo
downcase Name objects for storage in internal zones dict

Conflicts:
	configure.ac

case with an invalid tsigkey string.

I did this instead of upgrading boost or changing gcc.

We do the same thing in six other Makefiles.

This will fix failure noticed on NetBSD.

This minimises chances of subscribing and then throwing.