Python实现在线程里运行scrapy的方法
本文实例讲述了Python实现在线程里运行scrapy的方法。分享给大家供大家参考。具体如下：
如果你希望在一个写好的程序里调用scrapy，就可以通过下面的代码，让scrapy运行在一个线程里。
"""
CodetorunScrapycrawlerinathread-worksonScrapy0.8
"""
importthreading,Queue
fromtwisted.internetimportreactor
fromscrapy.xlib.pydispatchimportdispatcher
fromscrapy.core.managerimportscrapymanager
fromscrapy.core.engineimportscrapyengine
fromscrapy.coreimportsignals
classCrawlerThread(threading.Thread):
def__init__(self):
threading.Thread.__init__(self)
self.running=False
defrun(self):
self.running=True
scrapymanager.configure(control_reactor=False)
scrapymanager.start()
reactor.run(installSignalHandlers=False)
defcrawl(self,*args):
ifnotself.running:
raiseRuntimeError("CrawlerThreadnotrunning")
self._call_and_block_until_signal(signals.spider_closed,\
scrapymanager.crawl,*args)
defstop(self):
reactor.callFromThread(scrapyengine.stop)
def_call_and_block_until_signal(self,signal,f,*a,**kw):
q=Queue.Queue()
defunblock():
q.put(None)
dispatcher.connect(unblock,signal=signal)
reactor.callFromThread(f,*a,**kw)
q.get()
#Usageexamplebelow:
importos
os.environ.setdefault('SCRAPY_SETTINGS_MODULE','myproject.settings')
fromscrapy.xlib.pydispatchimportdispatcher
fromscrapy.coreimportsignals
fromscrapy.confimportsettings
fromscrapy.crawlerimportCrawlerThread
settings.overrides['LOG_ENABLED']=False#avoidlognoise
defitem_passed(item):
print"Justscrapeditem:",item
dispatcher.connect(item_passed,signal=signals.item_passed)
crawler=CrawlerThread()
print"Startingcrawlerthread..."
crawler.start()
print"Crawlingsomedomain.com...."
crawler.crawl('somedomain.com)#blockingcall
print"Crawlinganotherdomain.com..."
crawler.crawl('anotherdomain.com')#blockingcall
print"Stoppingcrawlerthread..."
crawler.stop()
希望本文所述对大家的Python程序设计有所帮助。