Using twisted deferreds in a scrapy spider!

gistfile1.txt

$ scrapy runspider txspider.py
2016-07-05 23:11:39 [scrapy] INFO: Scrapy 1.1.0 started (bot: scrapybot)
2016-07-05 23:11:39 [scrapy] INFO: Overridden settings: {}
2016-07-05 23:11:40 [scrapy] INFO: Enabled extensions:
['scrapy.extensions.corestats.CoreStats', 'scrapy.extensions.logstats.LogStats']
2016-07-05 23:11:40 [scrapy] INFO: Enabled downloader middlewares:
['scrapy.downloadermiddlewares.httpauth.HttpAuthMiddleware',
 'scrapy.downloadermiddlewares.downloadtimeout.DownloadTimeoutMiddleware',
 'scrapy.downloadermiddlewares.useragent.UserAgentMiddleware',
 'scrapy.downloadermiddlewares.retry.RetryMiddleware',
 'scrapy.downloadermiddlewares.defaultheaders.DefaultHeadersMiddleware',
 'scrapy.downloadermiddlewares.redirect.MetaRefreshMiddleware',
 'scrapy.downloadermiddlewares.httpcompression.HttpCompressionMiddleware',
 'scrapy.downloadermiddlewares.redirect.RedirectMiddleware',
 'scrapy.downloadermiddlewares.cookies.CookiesMiddleware',
 'scrapy.downloadermiddlewares.chunked.ChunkedTransferMiddleware',
 'scrapy.downloadermiddlewares.stats.DownloaderStats']
2016-07-05 23:11:40 [scrapy] INFO: Enabled spider middlewares:
['scrapy.spidermiddlewares.httperror.HttpErrorMiddleware',
 'scrapy.spidermiddlewares.offsite.OffsiteMiddleware',
 'scrapy.spidermiddlewares.referer.RefererMiddleware',
 'scrapy.spidermiddlewares.urllength.UrlLengthMiddleware',
 'scrapy.spidermiddlewares.depth.DepthMiddleware']
2016-07-05 23:11:40 [scrapy] INFO: Enabled item pipelines:
[]
2016-07-05 23:11:40 [scrapy] INFO: Spider opened
2016-07-05 23:11:40 [scrapy] INFO: Crawled 0 pages (at 0 pages/min), scraped 0 items (at 0 items/min)
2016-07-05 23:11:40 [scrapy] DEBUG: Crawled (200) <GET http://httpbin.org/get> (referer: None)
2016-07-05 23:11:40 [txspider] DEBUG: Logging from a thread 0! ~~~ (:
2016-07-05 23:11:40 [txspider] DEBUG: Sleeping for 5 secs
2016-07-05 23:11:40 [txspider] DEBUG: Logging from a thread 1! ~~~ (:
2016-07-05 23:11:45 [scrapy] DEBUG: Scraped from <200 http://httpbin.org/get>
{'thread': 0}
2016-07-05 23:11:45 [scrapy] DEBUG: Scraped from <200 http://httpbin.org/get>
{'thread': 1}
2016-07-05 23:11:45 [scrapy] DEBUG: Scraped from <200 http://httpbin.org/get>
{'inline': None}
2016-07-05 23:11:45 [scrapy] INFO: Closing spider (finished)
2016-07-05 23:11:45 [scrapy] INFO: Dumping Scrapy stats:
{'downloader/request_bytes': 212,
 'downloader/request_count': 1,
 'downloader/request_method_count/GET': 1,
 'downloader/response_bytes': 513,
 'downloader/response_count': 1,
 'downloader/response_status_count/200': 1,
 'finish_reason': 'finished',
 'finish_time': datetime.datetime(2016, 7, 6, 2, 11, 45, 697663),
 'item_scraped_count': 3,
 'log_count/DEBUG': 7,
 'log_count/INFO': 7,
 'response_received_count': 1,
 'scheduler/dequeued': 1,
 'scheduler/dequeued/memory': 1,
 'scheduler/enqueued': 1,
 'scheduler/enqueued/memory': 1,
 'start_time': datetime.datetime(2016, 7, 6, 2, 11, 40, 159804)}
2016-07-05 23:11:45 [scrapy] INFO: Spider closed (finished)

txspider.py

import scrapy

from twisted.internet import defer, threads, reactor


class TXSpider(scrapy.Spider):
    name = 'txspider'
    start_urls = [
        'http://httpbin.org/get',
    ]

    def parse(self, response, n_threads=2):
        d = defer.DeferredList([
            threads.deferToThread(self._process, n) for n in range(n_threads)
        ] + [
            # This method is decorated with defer.inlineCallbacks!
            self._inline_func(response)
        ])
        d.addCallback(self._make_output)
        # You must return a single deferred in a request callback,
        # and the output must complain with what scrapy expects, that is,
        # either an item, request or None.
        return d

    def _process(self, n):
        self.logger.debug("Logging from a thread %s! ~~~ (:", n)
        return {
            'thread': n,
        }

    @defer.inlineCallbacks
    def _inline_func(self, response, sleep=5):
        # You can decorate a request callback with 
        # inlineCallbacks.
        self.logger.debug("Sleeping for %s secs", sleep)
        d = defer.Deferred()
        reactor.callLater(sleep, d.callback, None)
        ret = yield d
        defer.returnValue({
            'inline': ret,
        })

    def _make_output(self, results):
        return [out for ok, out in results if ok and out]