I Pauls point is that the code in libdbmail would be well written and
scrutinized by many people, so it would be at least as efficient as
anything else. The extra layer in between might add some inefficiency,
but you could make that same argument for everything in software, might
it not be more efficient to write your compression libraries too? Not
to mention the fact that anything you write would have to be updated any
time the DB schema changes, the point of an abstraction layer is to give
you a consistent interface no matter what happens under the hood.
Matt
Joel Moss wrote:
I mean that if for example I wrote a webmail app, surely it would be
more efficient to take the data directly from the DB. If I wrote the app
in PHP, that would be very easy.
Paul J Stevens wrote:
Joel Moss wrote:
Sounds very interesting, but surely it would be more efficient to just
access the DB directly?
Of course not. Unless you assume the dbmail code is designed to be
inefficient.
Using libdbmail will give you full control, independent of the storage
backend, authentication backend, schema changes, etc.
_______________________________________________
DBmail mailing list
[email protected]
https://mailman.fastxs.nl/mailman/listinfo/dbmail
