Python實現線上程裡運行scrapy的方法

本文執行個體講述了Python實現線上程裡運行scrapy的方法。分享給大家供大家參考。具體如下：

如果你希望在一個寫好的程式裡調用scrapy，就可以通過下面的代碼，讓scrapy運行在一個線程裡。

"""Code to run Scrapy crawler in a thread - works on Scrapy 0.8"""import threading, Queuefrom twisted.internet import reactorfrom scrapy.xlib.pydispatch import dispatcherfrom scrapy.core.manager import scrapymanagerfrom scrapy.core.engine import scrapyenginefrom scrapy.core import signalsclass CrawlerThread(threading.Thread):  def __init__(self):    threading.Thread.__init__(self)    self.running = False  def run(self):    self.running = True    scrapymanager.configure(control_reactor=False)    scrapymanager.start()    reactor.run(installSignalHandlers=False)  def crawl(self, *args):    if not self.running:      raise RuntimeError("CrawlerThread not running")    self._call_and_block_until_signal(signals.spider_closed, \      scrapymanager.crawl, *args)  def stop(self):    reactor.callFromThread(scrapyengine.stop)  def _call_and_block_until_signal(self, signal, f, *a, **kw):    q = Queue.Queue()    def unblock():      q.put(None)    dispatcher.connect(unblock, signal=signal)    reactor.callFromThread(f, *a, **kw)    q.get()# Usage example below: import osos.environ.setdefault('SCRAPY_SETTINGS_MODULE', 'myproject.settings')from scrapy.xlib.pydispatch import dispatcherfrom scrapy.core import signalsfrom scrapy.conf import settingsfrom scrapy.crawler import CrawlerThreadsettings.overrides['LOG_ENABLED'] = False # avoid log noisedef item_passed(item):  print "Just scraped item:", itemdispatcher.connect(item_passed, signal=signals.item_passed)crawler = CrawlerThread()print "Starting crawler thread..."crawler.start()print "Crawling somedomain.com...."crawler.crawl('somedomain.com) # blocking callprint "Crawling anotherdomain.com..."crawler.crawl('anotherdomain.com') # blocking callprint "Stopping crawler thread..."crawler.stop()

希望本文所述對大家的Python程式設計有所協助。

