On Mon, Dec 22, 2003 at 10:50:05AM -0500, Carlos Moreno wrote:
> I will be moving to libpqxx for any new code that we
> write. (BTW, is libpqxx *officially* accepted as the
> C++ API?)
Yes. In fact this was what triggered the whole unbundling debate and
the ensuing decision to unbundle.
> So, I was hoping to get some comments regarding the
> issue of whether or not I should expect any trouble
> when moving to version 7.4 of the backend.
The code in libpq++ is simple enough that any problems _in the library_
are likely to be compilation problems, not runtime problems. OTOH it's
also simple enough that whether your libpq++-based code will continue
to work (from a runtime perspective) will depend largely on details of
your own code, such as how you check returned status codes or how deep
you get into details such as the character escaping done by COPY.
Other than that, I for one just don't know what's going to happen.
The 7.4 release cycle has proven pretty compatible for me, but since I
regularly updated from CVS the changes were spread out over a long
period from my perspective.
> PS: Jeroen, I have a couple of comments/feedback on
> libpqxx; should I e-mail you directly, or you
> prefer that I post them here?
Take a look at the libpqxx page on GBorg (or via http://pqxx.tk/); it
has a mailing list and bug/request lists managed by GBorg. See which
is most appropriate for what you want to say, or is the most convenient.
Given the unbundling, pgsql-interfaces is probably not the right venue
for this anymore.
Jeroen
