如何在基于Django WebSocket的实时聊天机器人中向PostgreSQL添加数据？

Question

I am trying to create real-time chat using Django back-end and Angular 4 front-end with PostgreSQL database. Let's assume that I would like to create chat robot for instance like ALICE . It seems to me that the most optimal solution would be to use websockets? I create channel layer using Redis . Now, I would like to save all messages to my PostgreSQL database. I wonder how it should be done. Should I create models and use them in consumers.py or maybe there is an another more optimal way? In the future I would like to analyze data from database and return the answer.

consumers.py:

# In consumers.py
from channels import Group

# Connected to websocket.connect
def ws_connect(message):
    # Accept the connection
    message.reply_channel.send({"accept": True})
    # Add to the chat group
    Group("chat").add(message.reply_channel)

# Connected to websocket.receive
def ws_receive(message):
    Group("chat").send({
        "text": message.content['text'],
    })
    print(message.content['text'])

# Connected to websocket.disconnect
def ws_disconnect(message):
    Group("chat").discard(message.reply_channel)

routing.py:

from channels.routing import route
from backend.consumers import ws_connect, ws_receive, ws_disconnect

channel_routing = [
    route("websocket.connect", ws_connect),
    route("websocket.receive", ws_receive),
    route("websocket.disconnect", ws_disconnect),
]

settings.py:

CHANNEL_LAYERS = {
    "default": {
        # This example app uses the Redis channel layer implementation asgi_redis
        "BACKEND": "asgi_redis.RedisChannelLayer",
        "CONFIG": {
            "hosts": [(redis_host, 6379)],
        },
        "ROUTING": "backend.routing.channel_routing",
    },
}

Answer 1

I had the same issue, so create a model in your models.py for example Message model.

class Message(models.Model):
    message = models.TextField()

and import it to consumers.py

#consumers.py
from .models import Message

def ws_receive(message):
Group("chat").send({
    "text": message.content['text'],
})
msg = Message()
msg.message = message.content['text']
msg.save()

But if you will use this snippet in production, it will hit your database performance. But as I heard, you can make some kind of queue with python module multiprocessing and save messages from that queue. I haven't seen implementation with custom queues, so you should make your own.