Re: sqlite3

[Philip Kaludercic]

Lars Ingebrigtsen <larsi@gnus.org> writes:

> Philip Kaludercic <philipk@posteo.net> writes:
>
>> Yes, but do you want the persistent data interface to be bound to
>> SQLite, or should a separate backed be implemented, if required?
>
> I don't think a separate backend is necessary, but I won't stand in the
> way of somebody implementing it.  (I'm sure somebody has also
> implemented a GIF decoder in Emacs Lisp because they're against using C
> or something.)

I raise this question from the perspective of the forwards-compatibility
library (compat.el, https://sr.ht/~pkal/compat/).  If there were no
distinction between persistent values and SQLite, then it would become
considerably more difficult to provide compatibility definitions if is
expected that the persistent value library can be mixed together with
raw SQL queries at any time.

-- 
        Philip Kaludercic