python - How does pyspider pass information to the next function, but not through the url

Question

Sometimes information is passed directly to the next function through pyspider, but there is no need to cral a new URL.

For example, a certain part of the information on a list page needs to be passed to the next function, but it does not want to be passed through the URL of the list.

I thought about using the send_message method, and also using self.crawl's response.save (just write a URL that is easier to crawl)
But these two methods are not particularly good.

Is there any good method?

Answer 1

The next function has finished executing and it no longer exists. How do you pass information to something that does not exist?

Answer 2

It’s nothing more than a jump callback, and the data that has come out yields, but you should still test it.

def detail(self, response):
    next_urls = [i.attr.href for i in response.doc('#fetch urls')]
    for url in next_urls:
        self.crawl(url, callback=self.list_page)
    for i in items = [
        # some result    
    ]
        yield i