Daffi is fast, simple and lightweight library for inter process communication
Project description
Daffi
Daffi facilitates remote computing and enables remote procedure calls between multiple endpoints. It supports many-to-many relationships between endpoints, allowing for seamless communication between distributed systems. The library abstracts the complexities of remote computing and provides a user-friendly interface for initiating and managing remote procedure calls. It also offers various features such as fault tolerance, load balancing, streaming and security, to ensure reliable and secure communication between endpoints.
Daffi comprises three primary classes:
- Global - Initialization entrypoint. Once Global object is initialized application can respond on remote requests and trigger remote callbacks itself.
- Callback - Represents a collection of methods encapsulated in a class inherited from Callback or a standalone function decorated with the callback decorator. These functions/methods can be triggered from another process.
- Fetcher - Represents a collection of methods encapsulated in a class inherited from Fetcher or a standalone function decorated with the fetcher decorator. These functions/methods serve as triggers for the corresponding callbacks defined in another process.
Documentation
View full documentation at: https://600apples.github.io/dafi/
Basic example
You need to create two files shopping_service.py
and shopper.py
shopping_service.py
- represents a set of remote callbacks that can be triggered by a client application.
shopper.py
- represents shopping_service client (fetcher)
Class based approach
shopping_service.py
:
import logging
from daffi import Global
from daffi.registry import Callback
logging.basicConfig(level=logging.INFO)
class ShoppingService(Callback):
auto_init = True # class is automatically initialized, eliminating the need to manually create an object.
def __post_init__(self):
self.shopping_list = []
def get_items(self):
"""Return all items that are currently present in shopping list"""
return self.shopping_list
def add_item(self, item):
"""Add new item to shopping list"""
self.shopping_list.append(item)
def clear_items(self):
"""Clear shopping list"""
self.shopping_list.clear()
if __name__ == '__main__':
Global(init_controller=True, host="localhost", port=8888).join()
(This script is complete, it should run "as is")
shopper.py
:
import logging
from daffi import Global
from daffi.decorators import alias
from daffi.registry import Fetcher
logging.basicConfig(level=logging.INFO)
class Shopper(Fetcher):
"""
Note: Functions without a body are treated as proxies for remote callbacks.
All arguments provided to this function will be sent to the remote service as-is.
"""
def get_items(self):
"""Return all items that are currently present in shopping list."""
pass
def add_item(self, item):
"""Add new item to shopping list."""
pass
def clear_items(self):
"""Clear shopping list"""
pass
@alias("add_item")
def add_many_items(self, *items):
"""
Alias for `add_item` callback.
This function shows streaming capabilities for transferring data from one service to another.
"""
for item in items:
yield item
if __name__ == '__main__':
g = Global(host="localhost", port=8888)
shopper = Shopper()
items = shopper.get_items()
print(items)
shopper.add_item("orange")
items = shopper.get_items()
print(items)
shopper.add_many_items("bread", "cheese")
items = shopper.get_items()
print(items)
shopper.clear_items()
items = shopper.get_items()
print(items)
g.stop()
(This script is complete, it should run "as is")
To check the full example, you need to execute two scripts in separate terminals
python3 shopping_service.py
...
INFO 2023-03-27 19:49:45 | controller[0x91adb83e] | Controller has been started successfully. Process identificator: '0x91adb83e'. Connection info: tcp: [ host '[::]', port: 8888 ]
INFO 2023-03-27 19:49:45 | node[0x91adb83e] | Node has been started successfully. Process identificator: '0x91adb83e'. Connection info: tcp: [ host '[::]', port: 8888 ]
python3 shopper.py
...
INFO 2023-03-27 19:53:15 | node[0xd7e5d488] | Node has been started successfully. Process identificator: '0xd7e5d488'. Connection info: tcp: [ host '[::]', port: 8888 ]
[]
['orange']
['orange', 'bread', 'cheese']
[]
INFO 2023-03-27 19:53:15 | node[0xd7e5d488] | Node stopped.
Decorators base approach
shopping_service.py
:
import logging
from daffi import Global
from daffi.decorators import callback
logging.basicConfig(level=logging.INFO)
shopping_list = []
@callback
def get_items():
"""Return all items that are currently present in shopping list"""
return shopping_list
@callback
def add_item(item):
"""Add new item to shopping list"""
shopping_list.append(item)
@callback
def clear_items():
"""Clear shopping list"""
shopping_list.clear()
if __name__ == '__main__':
Global(init_controller=True, host="localhost", port=8888).join()
(This script is complete, it should run "as is")
shopper.py
:
"""
Note: Functions without a body are treated as proxies for remote callbacks.
All arguments provided to this function will be sent to the remote service as-is.
"""
import logging
from daffi import Global
from daffi.decorators import alias, fetcher
logging.basicConfig(level=logging.INFO)
@fetcher
def get_items():
"""Return all items that are currently present in shopping list."""
pass
@fetcher
def add_item(item):
"""Add new item to shopping list."""
pass
@fetcher
def clear_items():
"""Add new item to shopping list."""
pass
@alias("add_item")
@fetcher
def add_many_items(*items):
"""
Alias for `add_item` callback.
This function shows streaming capabilities for transferring data from one service to another.
"""
for item in items:
yield item
if __name__ == '__main__':
g = Global(host="localhost", port=8888)
items = get_items()
print(items)
add_item("orange")
items = get_items()
print(items)
add_many_items("bread", "cheese")
items = get_items()
print(items)
clear_items()
items = get_items()
print(items)
g.stop()
(This script is complete, it should run "as is")
To check the full example, you need to execute two scripts in separate terminals
python3 shopping_service.py
...
INFO 2023-03-27 20:31:27 | controller[0xbac16ef4] | Controller has been started successfully. Process identificator: '0xbac16ef4'. Connection info: tcp: [ host '[::]', port: 8888 ]
INFO 2023-03-27 20:31:27 | node[0xbac16ef4] | Node has been started successfully. Process identificator: '0xbac16ef4'. Connection info: tcp: [ host '[::]', port: 8888 ]
python3 shopper.py
...
INFO 2023-03-27 20:31:43 | node[0xb9e10444] | Node has been started successfully. Process identificator: '0xb9e10444'. Connection info: tcp: [ host '[::]', port: 8888 ]
[]
['orange']
['orange', 'bread', 'cheese']
[]
INFO 2023-03-27 20:31:44 | node[0xb9e10444] | Node stopped.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
File details
Details for the file daffi-2.2.1.tar.gz
.
File metadata
- Download URL: daffi-2.2.1.tar.gz
- Upload date:
- Size: 447.0 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: python-httpx/0.23.1
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 106bc20da31eb6da91710f26ece0de76dc091c19ddc89635539c84088d533ff4 |
|
MD5 | 93ed7e0b722e093e99d82e71b68f0c01 |
|
BLAKE2b-256 | dd823db855fb0e78d66a9166b01877dd0c16f4bf13d93758590a36f7b369f371 |
File details
Details for the file daffi-2.2.1-py3-none-any.whl
.
File metadata
- Download URL: daffi-2.2.1-py3-none-any.whl
- Upload date:
- Size: 74.4 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: python-httpx/0.23.1
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | d2d6855a681fd72f4755a0c75592e71c9f5fdbac40d875af0aeb0002dc6f1420 |
|
MD5 | f595b271eeee76f0275732a62ac6bd43 |
|
BLAKE2b-256 | 7a4b43d245b8075283d8d08035e27d706a5d5d8cf3e26ad222b6570a78dc95a9 |