Re: In re: improving async support in psycopg

Alexander Sabourenkov ha scritto:
> Hello.
> 
> I have already done a similar work, though much more invasive and
> resulting in what amounts to a fork of psycopg2.
> 
> Most code changes were in pqpath.c. Although my coroutine/event-loop
> library is quite different from the gevent, general principles are same
> and I hope one can find some insights on how it's done.
> 

I think this is not how async support should be implemented.
It is too invasive.

Instead psycopg should just provide a *minimal* API (like libpq) to
allow external applications to take advantage of an asynchronous
environment.

If you want a connection that it is always async, you can write a Python
wrapper on the top of psycopg.
It should not be hard to write a custom SQLAlchemy dialect, too.

How a socket is checked for ready status should not be done in the
psycopg implementation, it should be done by external applications (and
using greenlets and an standard event/reactor API, you can write a
flexible and reusable wrapper).



Manlio