[pylons-discuss] Re: Pyramid, traversal, SQLAlchemy, and ACLs

2015-10-21 Thread Jonathan Vanasco
This looks painful to me for a few reasons: 1. Hitting the db on everything is pricey. Building this with the foresight that a read-through cache will eventually be needed might be a good idea. 2. Some of the SqlAlchemy generation looks to be likely expensive (within Python). It would probably

Re: [pylons-discuss] Pyramid, traversal, SQLAlchemy, and ACLs

2015-10-21 Thread Paul Everitt
Yep, that looks interesting. Couple of comments: - You can certainly do traversal to a resource without doing location-awareness (__name__, __parent__ etc.) as that is a one-direction operation. But if you want to go back up for any reason, e.g. hierarchical ACLs, you’ll need that. Or, if you

Re: [pylons-discuss] Pyramid, traversal, SQLAlchemy, and ACLs

2015-10-21 Thread Steve Piercy
There was a discussion in #pyramid early this morning that might relate to this. https://botbot.me/freenode/pyramid/2015-10-21/?msg=52400847&page=2 --steve On 10/21/15 at 4:28 PM, paulwever...@gmail.com (Paul Everitt) pronounced: > tl;dr Complicated SQLAlchemy SQL expression for permission filt

[pylons-discuss] Pyramid, traversal, SQLAlchemy, and ACLs

2015-10-21 Thread Paul Everitt
tl;dr Complicated SQLAlchemy SQL expression for permission filtering Pyramid ACLs Hi all. I’ve been noodling around for a while on traversal/location-awareness/hierarchies in SQLAlchemy. During the coming month I’m going to work on a package for this: https://github.com/pauleveritt/pyramid_s