web2py is constantly tested with pypy as an environment, and I really don't see why web2py's implementation should be considered "not pypy friendly". That being said, the improvements on psycopg2-cffi vs psycopg2 are ISOLATED in that specific "corner". If, for example, psycopg2-ccfi turns to be 50% faster than psycopg2, that "splice" of performances will be noticed in regards of the driver itself and there's nothing that web2py can do to "alter" that 50%.
On Wednesday, May 28, 2014 7:50:09 PM UTC+2, Joe Barnhart wrote: > > I see there is a fairly new-ish implementation of psycopg2 which conforms > to the cffi calling convention, for use with pypy. Has anyone tried this > along with a fresh web2py to see if improvements have been made? > > I've always assumed pypy would not do much for web2py because of our > preference for "exec" instead of compiled code. But I noticed that my > web2py projects recently have been putting most of the heavy-hitting stuff > in modules, which would definitely be pypy-friendly. > > So I'm curious if anyone has dabbled with pypy and web2py (and Postgres) > recently... > > -- Joe > > -- Resources: - http://web2py.com - http://web2py.com/book (Documentation) - http://github.com/web2py/web2py (Source code) - https://code.google.com/p/web2py/issues/list (Report Issues) --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
