[bind10-dev] revised in-memory NSEC task plans

Jelte Jansen jelte at isc.org
Wed May 9 09:00:16 UTC 2012


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 05/09/2012 10:23 AM, Michal 'vorner' Vaner wrote:
> 
>> - complete #1805 with a quick hack implementation.  that is,
>> maintain an internal table of names in the getClosestNSEC()
>> method (either by hardcoding the data or allowing the test to
>> build it), and use that table to identify the "previous name".
>> this way we can start other tasks depending on #1805 without
>> waiting for the completion of #1803 and #1804.  When these two
>> tasks are completed we'll complete getClosestNSEC() using the
>> getPreviousNode() interface.  That part will be deferred to a
>> separate task.
> 
> #1803/#1804 is in review. I think it is less work to use the
> interface there even if it was to change than implementing some
> kind of static table. And the chance is it could stay the same or
> with minor tweaks.
> 

And on a more general note, much of such work could already be done if
the interface is there (but not implemented, not even with static
data); we can write an initial version of the code and the tests for
it, which will obviously fail then. Then when the groundwork itself is
done you can simply merge it in an it should all work, since you know,
we never make mistakes. Or at least it would require less work to
hammer it all in place since a lot of surrounding work has already
been done.

Of course it may not always be possible to write it as a stub, and the
interface may change a little, but at least we can get started on
several tickets at the same time if the calls themselves exist in some
form.

You (jinmei) already kind of suggested such an approach by branching
from 1805-in-progress into 1806-9, I think that next time we should do
the APIs as stubs first, if possible.

Jelte
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk+qMiAACgkQ4nZCKsdOncUa4wCbB78t1fCa4glRwkMTAPQ65aD2
M7QAmwdYUYhYOPnz1NVfsfblJ6nlVb9t
=OZzt
-----END PGP SIGNATURE-----


More information about the bind10-dev mailing list