help-gnats
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: patch #2 - towards a generic backend datastore


From: Mel Hatzis
Subject: Re: patch #2 - towards a generic backend datastore
Date: Thu, 21 Apr 2005 16:46:18 -0700
User-agent: Mozilla Thunderbird 1.0.2 (Macintosh/20050317)

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

I have done some additional testing with my proposed patch and
everything appears to be in order.

*I plan on committing these changes to the mainline tomorrow.*

Note that there is a minor change introduced with this patch.
If the index is reloaded when gnatsd processes an 'rset'
command, gnatsd currently generates the following message:

     210 Reset state...reloaded the index.

After this patch is committed, the following message will
be issued (which is the same message that is generated
currently, whenever the index is not reloaded):

     210 Reset state.

This change is required in order to make the client/server
interaction datastore agnostic. Thankfully, the rset command
was the only one which referred to the index - and thus
needed to be modified.

- --
Mel Hatzis

Mel Hatzis wrote:
> Please review the attached patch which contains a number of
> modifications intended to restructure the code to support a
> generalized backend datastore.
> 
> This second patch is rather large - larger than I originally
> planned. I expect that all subsequent patches will be
> significantly smaller.
> 
> The primary goal in this patch was to completely isolate the
> index from the generic (i.e. non-datastore specific) GNATS
> functionality and introduce a flat-file datastore library with
> which all relevant binaries are linked. A small addition to
> configure was made to build the flat-file datastore library
> by default.
> 
> This is a significant leap forward in generalizing the backend
> datastore. The GNATS/datastore interface really starts to become
> more obvious with this patch.
> 
> I have attempted to adequately describe all changes in detail
> in the ChangeLog. Feel free to ask me to elaborate where
> necessary.
> 
> Once again, I've run these changes through my regression test
> suite and they appear to work well. Of course, I encourage
> more testing if people have the bandwidth.
> 
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (Darwin)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFCaDtKNF74HmYqaSERAqqaAJ4sKH+JurajgHwaTh6rJQDHx7yOZgCglfKh
afJu2EuD78wcvn26kWb4hnA=
=xffG
-----END PGP SIGNATURE-----




reply via email to

[Prev in Thread] Current Thread [Next in Thread]