Skip to content

Implement a client  #48

Description

@ManiMozaffar

A client is only needed for the purpose of multi accounting in crawler.
for example you want to crawl a website, same url, but with 10 different accounts.

each should have its own pipeline, and logic. So we need a Client class where it can be injected to our script.
then each client, opens an Engine Client, and then use that to login, and then return all those as a list

from fastcrawler.engine import AioHttpEngine

async def get_clients():
    clients = AioHttpEngine.new_client() * 10

    clients = [
         asyncio.gather(
               client.get(...)
               for client in clients
     ]    
    return clients


class MySpider:
    clients = Depends(get_clients)

Simple case would be like that.

-> Note: this is WIP. It will not be included very soon, because right now we have decided this design is over engineered and only useful in few cases

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions