A functional reactive graph database backed by PostgreSQL

In layman's terms, OpenARC lets you build graphs of objects that react to one another.

OpenARC requires Python 3+ and PostgreSQL 10+.

Consult your operating system documentation for details on how to install these requirements.

Use pip to install the last release:

pip install openarc --user

Live on the edge by cloning our repository:

git clone https://www.github.com/kchoudhu/openarc
cd openarc
pip install . --user

If you don't already have a database to point to, you can use the makefile to spin up your own local instance to work with:

cd openarc
make dbmshardinit

OpenARC follows the XDG specification, and looks for openarc.conf in the following locations:

• XDG_CONFIG_HOME
• $HOME/.config/
• /usr/local/etc/

An annotated sample openarc.conf is available in the cfg directory of the project distribution.

Start by importing the library and defining an OAG class:

class OAG_User(OAG_RootNode):
    @staticproperty
    def context(cls): return "myproject"

    @staticproperty
    def streams(cls): return {
        'username' : [ 'text', str(), None ],
        'password' : [ 'text', str(), None ]
    }

new_user =\
    OAG_User(initprms={
        'username' : 'kamil',
        'password' : 'hunter2'
    })

print(new_user.username)

# Output: kamil

Persist to the database:

new_user.db.create()

This creates a table in the database:

> select * from myproject.user;

 _user_id | password | username
----------+----------+----------
        1 | hunter2  | kamil

Update the object:

new_user[0].username = 'hana'
new_user.db.update()

The database is updated accordingly:

> select * from myproject.user;

 _user_id | password | username
----------+----------+----------
        1 | hunter2  | hana

More documentation about the database capabilities of OpenARC is available here.

An OAG_Group is a collection of users:

class OAG_Group(OAG_RootNode):
    @staticproperty
    def context(cls): return "myproject"

    @staticproperty
    def streams(cls): return {
        'user' : [ OAG_User, False, "user_update_handler" ]
    }

    def user_update_handler(self):
        print("[user_update_handler] User [%s] has been updated" % self.user.username)


group = OAG_Group()
group.user = new_user

An event dependency is created between group and new_user by assigning the latter to the former under group's user node.

user_update_handler is an event handler that is invoked whenever the user node of the OAG_Group object is changed:

new_user.password = '*****'
# [user_update_handler] User [hana] has been updated

Nodes can be nested arbitrarily, permitting complex functional event chains. Think Excel, but for programmers.

If you have the project distribution, take a look under openarc/tests for a wide variety of use cases. Assuming your environment is set up, you can execute the tests by executing

make test

Interesting examples of OpenARC usage live under openarc/examples. Each example is thoroughly documented and comes with a README; the hope is that they will open your eyes as to what is possible with OpenARC.

Documentation is distributed with the code, and can be generated by executing

make docs