On Tue, Aug 16, 2016 at 9:59 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> I think this might have advantages purely from the standpoint of new
> compilers possibly offering useful warnings we don't get now. But
> if we only go this far, I'm pretty dubious that it really helps people
> to develop extensions in C++. Almost invariably, if you ask *why* they
> want to do that, you'll get an answer involving C++ libraries that are
> not going to play very nice with our error handling or memory management
> conventions.
FWIW, it's not uncommon to opt-out of C++ exceptions entirely, for
various reasons. For example, the Google C++ style guide forbids it
(if only for historical reasons), as does the GCC style guide (since
GCC was a C program until several years ago [1]). Sometimes, these
third party libraries that mandate the use of exceptions do indeed
create significant headaches for Postgres, compatibility-wise, but
that isn't a given.
IMV, it would be useful to use C++ classes (and even template classes)
for a small number of data structures, while still largely adhering to
earlier practices (this is what GCC did). Specifically, a few modules
such as StringInfo, could be made to follow the RAII/scope bound
resource management usefully, which doesn't seem incompatible with
memory contexts. However, this doesn't seem terribly exciting to me.
[1] https://lwn.net/Articles/542457/
--
Peter Geoghegan
