On Feb 5, 2010, at 1:34 PM, Marko Kreen wrote:
> py-postgresql seems to be more serious, but as it's python3 only
> which makes it irrelevant today.
Furthermore, if it did work on python2, it's *not* something that's going to appeal to mainstream users (Python heavy
webframeworks) as it *partially* suffers from the same problem that pg8000 does. It's mostly pure-Python, but it has
someC optimizations(notably, PQ message buffer). I have done some profiling, and *with a few tweaks* it's about 2x-3x
*slowerthan psycopg2* for the retrieval of a single int column. I think it could go faster, but I don't think it's
worththe work.
ISTM that the target audience are folk who are married to PG, and are generally unhappy with DB-API, but do not want to
buyinto a "big" abstraction layer like SQLAlchemy. Sure, it supports DB-API like other drivers so it *would be* usable
withframeworks, but why take the 3x *or greater* hit over a properly implemented libpq version?
Finally, I just don't see the existing (often PG specific) goals that I have in mind for it appealing to the majority
of[web framework/abstraction] users.
> Psycopg was the leader, especially in web-environments,
> but it has non-obvious license and with dead website it does not
> seem that attractive. Although it is well-maintained still.
>
> Best path forward would be to talk with Psycopg guys about
> license clarification/change.
Yep.
