Yes i can definately help with testing. Currently there seems to be 2 existing libs for using orientdb.
- Compass: https://github.com/emehrkay/Compass which ties directly into orientdb through its rest interface, and will eventually have binary support. - Bulbs with Rexter: http://bulbflow.com/http://bulbflow.com/ / https://github.com/tinkerpop/rexster/wiki/Rexster-Configuration bulbflow uses a system called blueprints, which is eddentially an orm for most graph databases. then there are a few layers that talk to the different db's. most of them use rexter (second link for configuration for orientdb). While the first option, Comnpass, could probably be easier to work into dal, if bulbflow/rexter is used, we can let people use whatever graphdb they want, as long as it supports blueprint. Which I believe most do. The main difference with them, is there is no relations like has many, belongs to, or many to many. They have edges, which are still relationships, but are represented in the form of fields. For an example, a user, and a post, might have an edge of date (published date), and author (who made the post). And bot hthe nodes themselves, and edges can be searched for. On Tue, Sep 11, 2012 at 7:32 AM, Massimo Di Pierro < massimo.dipie...@gmail.com> wrote: > Adding support for other SQL databases is easy. I have no data about > OrientDB but we can easily add an adapter if you help with testing. > > > On Tuesday, 11 September 2012 08:15:01 UTC-5, luckysmack wrote: > >> Well its starting to seem like a good fit. But yes I won't be using DAL, >> since I'll be using orientdb. I know I lose something with that decision, >> but its what I need. I'm also curious how people are using it here as well >> with web2py. >> On Sep 11, 2012 1:46 AM, "pbreit" <pbreit...@gmail.com> wrote: >> >>> It sounds like you know enough about these things to figure out if >>> Web2py is too bloated for you. My sense is that even if you think it might >>> be bloated, the extraneous stuff really doesn't get in the way too much. >>> Not using DAL means you are missing out on one of the core features. >>> Building a REST API should be fine either using what Web2py gives you out >>> of the bos or rolling your own. You can either build multiple Web2py apps >>> or build some web2py apps and some other apps. DAL doesn't really support >>> NoSQL much at this point so you're on your own there. I've used Web2py to >>> build a POS system and it works great. >>> >>> >>> >>> On Saturday, September 8, 2012 12:39:26 AM UTC-7, luckysmack wrote: >>>> >>>> I am somewhat new to python, and * shock * have an idea for a simple >>>> app I want to build. To start the app will be relatively light weight, but >>>> if it works out in my grand scheme could be far more complex.but the core >>>> will be fairly simple. something an experienced python dev could probably >>>> whip up in bottle in a few days. >>>> >>>> The core will a simply be an advanced rest based api. The other half >>>> dozen or so apps will all be built of this core. The either apps will >>>> likely be built as a cms like system to manage each other. And as different >>>> as a POS in store program. >>>> >>>> So as many cool things as I thing web2py has, do you guys think its the >>>> right system? I know w2p can do great APIs easily. But for that simple >>>> aspect, I don't need a milti-application admin interface, or a code editor, >>>> and I may not even use DAL. (For my project I may actually use something >>>> like neo4j/orientdb/titan. Not sure yet. Might use mongodb as instead). So >>>> for that simple part, all the other stuff seems a little bloated to me. >>>> Stuff that I won't need. >>>> >>>> Sure as a whole, all the apps will be built into somewhat of a cms >>>> (which I would like to build anyways), for that I'm not sure I would need >>>> the web2py admin part. As a cms I would probably have my own interface, >>>> even for the admins. How or would web2py admin ui fit in. I know you guys >>>> are biased towards web2py, but does it sound like it would be a right fit? >>>> Or would it be too complex? In comparison, I feel django is too bloated as >>>> well since I would be doing a similar thing, except it would be done quite >>>> a bit differently. The core of how it works doesn't seem to fit my ideas. >>>> >>>> If I don't use web2py, the next best things I see as a starting point >>>> are pyramid, or bottle/flask or even wheezy looks pretty cool. >>>> >>>> What do you guys think? The core great api would be the crux of the >>>> other apps. This core is what talks to the db. And each if the apps with >>>> build on it. The rest part will be made so they can all communicate with >>>> each other based on the URL. >>>> >>>> The either individual apps, were they to be on their own, I can totally >>>> see as a web2py app. So I'm curious how this idea as a whole, would fit >>>> into web2py. If it can. Since there are a handful of web2py featured I >>>> won't even use. I don't need them to be auto imported if I'm not using them >>>> (since I can't see what's being imported). So in a way it feels like bloat. >>>> A reason I don't like django. >>>> >>>> What do you guys think? Any input is greatly appreciated. >>>> >>> -- >>> >>> >>> >>> >> -- > > > > -- -- Shawn McElroy "Anything worth doing, is worth doing right" — Hunter S. Thompson "A mind troubled by doubt cannot focus on the course to victory" — A<http://www.goodreads.com/author/quotes/5237.Hunter_S_Thompson>rthur Golden --
