GitLab

going to depend on libacl.

with fixing conflicts in src/bin/resolver/Makefile.am

This is from patch ticket #1053 from jinmei.
http://bind10.isc.org/attachment/ticket/1053/makefile.diff

It is a workaround to problem with isc/log name reuse.

I renumbered a new entry because of this.

Clang seems to be confused by too many templates and function
overloading, so we sacrifice little bit of convenience to make it happy.

Conflicts:
	src/lib/server_common/Makefile.am

placeholder for subsequent development.

I also noticed some of the comments in the tests are stale or not really
correct, so I fixed them, too.

instead of returning a dummy structure, expecting it to cause a subsequent
failure.

this branch)

typedef.  RequestACL is itself a shortcut, so it doesn't make sense to
define another one.

note that other members than the remote IP address were removed to avoid
defining things that "you ain't gonna need (it)".  But added comments
about possibility of these extensions.

an exception).
not directly related to the subject of this branch, but the case was
encountered during the resolver test.  IMO, the library shouldn't cause
crash due to broken input even from a buggy implementation, and even if
the validation is normally expected to be done at a higher level.

framework.  it now accepts "action only" rule, so the test test in that
case was updated accordingly.