The reason for the TaskSpec class is so that I can easily re-add the same job back to the queue. I get a good number of httpx.ReadTimeout errors, presumably due to my slow connection, or perhaps because the actual parsing takes a while, and some of the requests spend too long being left waiting. So sometimes, I need to re-queue a webpage scraping job, and this was the easiest way I could come up with. I had also tried to create task objects and re-add the coroutine back to the queue using task.get_coro(), but this didn’t work since the coroutine was already executed. Also, this way, the actual parsing functions get_page_{1,3} don’t actually need to know about the queue.