BIND 10 trac1771-2, created. 930e5b9931e061578f77ac3acc9d796459c13633 [1771-2] clarified the semantics of FIND_GLUE_OK more.

[BIND 10 source code commits]

The branch, trac1771-2 has been created
        at  930e5b9931e061578f77ac3acc9d796459c13633 (commit)

- Log -----------------------------------------------------------------
commit 930e5b9931e061578f77ac3acc9d796459c13633
Author: JINMEI Tatuya <jinmei at isc.org>
Date:   Fri Jun 15 18:08:03 2012 -0700

    [1771-2] clarified the semantics of FIND_GLUE_OK more.
    
    that it's irrelevant to query type or whether the name is pointed by NS.

commit f3e904c386e2a0d93cba0c7df8698c5ba1b441ce
Author: JINMEI Tatuya <jinmei at isc.org>
Date:   Fri Jun 15 17:52:17 2012 -0700

    [1771-2] cleanup: DS RR doesn't have to be considered separately in getRRsets.

commit ed608bae6873c4597f2136804758d68da84754c2
Author: JINMEI Tatuya <jinmei at isc.org>
Date:   Fri Jun 15 17:49:01 2012 -0700

    [1771-2] cleanup: check_ns param for getRRsets is not needed now and removed.

commit ef571c297aabab0808ac3eecf3c457fdce55a73d
Author: JINMEI Tatuya <jinmei at isc.org>
Date:   Fri Jun 15 17:22:07 2012 -0700

    [1771-2] remove seen_ns&other check in getRRsets.
    
    on a closer look, it just doesn't seem to be necessary if we are okay
    with just "hiding" records under a delegation NS for normal (ie. non
    GLUE_OK) lookups, instead of throwing an exception.
    
    Essentially only one existing test case (which expects an exception
    in that setup) should be updated (and the "brokenns2" case just seems
    to be redundant and so was removed), and the disabled tests
    in zone_finder_context_unittest now pass.

-----------------------------------------------------------------------