GitLab

Small fix, skipping review

as to copySection, it's now called appendSection, and works the other
way around (i.e. it copies from a source message into the object,
instead of copying from the object to a target message), which feels
more natural

Checks that we don't reject synthesized name of the maximum length by
mistake.

note: qtype_is_any was originally const, and was temporarily changed to
mutable.  it's now immutable again.

as discussed on jabber.

the actual bit position isn't a big deal right now, but the intent would be
a bit clearer than an odd intermediate position (and, in fact, using the
MSB was the original intent; the previous value was a typo)

We overflow the size of name by prolonging a name of full length by
little bit.

and updated some doxygens as per Jeremy's mail

it effectively makes this branch with trac517 (which has been merged to
master since then).  it will help merge this branch to master.

also resolved trivial conflicts in src/lib/datasrc/tests/rbtree_unittest.cc.

This reverts commit 1ab958cd.

As agreed, we don't provide the authority section for the original
DNAME/CNAME, only for the last target. As we don't chain, we don't
provide it at all.

also removed an unused initialization as a piggybacked cleanup.
okayed on jabber, behavior confirmed, too.