Skip to content

t-mart/aqueue

Repository files navigation

aqueue

aqueue is an async task queue with live progress display.

You put items (tasks) in, and they get processed, possibly creating more items which get processed, and so on, until all items are completed. A typical use case would be to scrape a website.

Meanwhile, a nice visualization of the queue's goings-on is displayed in the terminal.

Demonstration of aqueue

Note

aqueue, or any asynchronous framework, is only going to be helpful if you're performing I/O-bound work.

Installation

aqueue is a Python package hosted on PyPI. The recommended installation method is pip-installing into a virtual environment:

pip install aqueue

Getting Started

There's two things you need to do to use aqueue:

  1. Implement your Item subclasses.
  2. Start your queue with one of those items.

Example

If you had a hierarchy of items like this...

Simple item hierarchy with one root item and many children items stemming from it.

Then, you might process it with aqueue like this...

import aqueue


class RootItem(aqueue.Item):
   async def process(self) -> aqueue.ProcessRetVal:
      # display what we're doing in the worker status panel
      self.set_worker_desc("Processing RootItem")

      # make an HTTP request, parse it, etc
      ...

      # when you discover more items you want to process, enqueue them by yield-ing
      # them:
      for _ in range(3):
            yield ChildItem()

   async def after_children_processed(self) -> None:
      # run this method when this Item and all other Items it enqueued are done
      print("All done!")


class ChildItem(aqueue.Item):

   # track the enqueueing and completion of these items in the overall panel
   track_overall: bool = True

   async def process(self) -> aqueue.ProcessRetVal:
      self.set_worker_desc("Processing ChildItem")
      # this child item has no further children to enqueue, so it doesn't yield
      # anything


if __name__ == "__main__":
   aqueue.run_queue(
      initial_items=[RootItem()],
      num_workers=2,
   )

Project Information

About

An async task queue with live progress display

Topics

Resources

License

Stars

Watchers

Forks

Sponsor this project

 

Packages

No packages published