CODEWITHSUNDEEP
pythonscrapy
Mr. B
Mr. B

Reputation: 2686

Scrapy: How to manually insert a request from a spider_idle event callback?

I've created a spider, and have linked a method to the spider_idle event.

How do I add a request manually? I can't just return the item from parse -- parse is not running in this case, as all known URLs have been parsed. I have a method to generate new requests, and I would like to run it from the spider_idle callback to add the created request(s).

class FooSpider(BaseSpider):
    name = 'foo'

    def __init__(self):
        dispatcher.connect(self.dont_close_me, signals.spider_idle)

    def dont_close_me(self, spider):
        if spider != self:
            return
        # The engine instance will allow me to schedule requests, but
        # how do I get the engine object?
        engine = unknown_get_engine()
        engine.schedule(self.create_request())

        # afterward, ensure we stay alive by raising DontCloseSpider
        raise DontCloseSpider("..I prefer live spiders.")

UPDATE: I've determined that I probably need the ExecutionEngine object, but I don't exactly know how to get that from a spider, though it available from a Crawler instance.

UPDATE 2: ..thanks. ..crawler is attached as a property of the superclass, so I can just use self.crawler with no additional effort. >.>

Upvotes: 17

Views: 4940

Answers (1)

Steven Almeroth
Steven Almeroth

Reputation: 8192

class FooSpider(BaseSpider):
    def __init__(self, *args, **kwargs):
        super(FooSpider, self).__init__(*args, **kwargs)
        dispatcher.connect(self.dont_close_me, signals.spider_idle)

    def dont_close_me(self, spider):
        if spider != self:
            return

        self.crawler.engine.crawl(self.create_request(), spider)

        raise DontCloseSpider("..I prefer live spiders.")

Update 2016:

class FooSpider(BaseSpider):
    yet = False

    @classmethod
    def from_crawler(cls, crawler, *args, **kwargs):
        from_crawler = super(FooSpider, cls).from_crawler
        spider = from_crawler(crawler, *args, **kwargs)
        crawler.signals.connect(spider.idle, signal=scrapy.signals.spider_idle)
        return spider

    def idle(self):
        if not self.yet:
            self.crawler.engine.crawl(self.create_request(), self)
            self.yet = True

Upvotes: 27

Related Questions