为列表中的每个 url 重新启动 scrapy

Question

我正在尝试运行一个scrapy bot，它将为列表中给出的每个url重复运行蜘蛛。我到现在写的代码如下

def run_spider(url_list,allowed_list):
    runner = CrawlerRunner(get_project_settings())
    d = runner.crawl('scraper',start_urls=url_list, allowed_domains=allowed_list)
    d.addBoth(lambda _: reactor.stop())
    reactor.run()



for start, allowed in zip(start_url,allowedUrl):
    url_list = []
    allowed_list = []
    url_list.append(start)
    allowed_list.append(allowed)
    print(type(url_list),type(allowed_list))
    run_spider(url_list,allowed_list) 

蜘蛛本身在第一个 url 上运行良好，但一旦循环命中它就会给出错误twisted.internet.error.ReactorNotRestartable，完整的回溯在这里：

Traceback (most recent call last):
  File "C:\brox\Crawler\main.py", line 34, in <module>
    run_spider(url_list,allowed_list)
  File "C:\brox\Crawler\main.py", line 24, in run_spider
    reactor.run()
  File "C:\brox\Crawler\venv\lib\site-packages\twisted\internet\base.py", line 1282, in run
    self.startRunning(installSignalHandlers=installSignalHandlers)
  File "C:\brox\Crawler\venv\lib\site-packages\twisted\internet\base.py", line 1262, in startRunning
    ReactorBase.startRunning(self)
  File "C:\brox\Crawler\venv\lib\site-packages\twisted\internet\base.py", line 765, in startRunning
    raise error.ReactorNotRestartable()
twisted.internet.error.ReactorNotRestartable

我遵循文档中描述的方法，但是如何为循环中的每个项目重新启动蜘蛛。任何建议都会非常有帮助。

P:S:: 当简单地传递允许的域并启动 url 时，蜘蛛机器人本身就可以正常工作

Answer 1

要使您的代码正常工作，您必须重新排列 reactor.run() 和 reactor.stop() 逻辑。以下是解决问题的方法示例：

from scrapy.crawler import CrawlerRunner
from twisted.internet.defer import gatherResults
from twisted.internet import reactor

def run_spider(url_list, allowed_list):
    """
    :returns: Deferred
    """
    runner = CrawlerRunner(get_project_settings())
    return runner.crawl('scraper', start_urls=url_list, allowed_domains=allowed_list)

d_list = []
for start, allowed in zip(start_url, allowedUrl):
    # ... your logic ...
    # Append the deferred into a list.
    d_list.append(run_spider(url_list, allowed_list))

# "Join"
results = gatherResults(d_list)
# Stop the reactor after all the sites are scraped or a failure occurs
results.addBoth(lambda _: reactor.stop())

reactor.run()

run_spider() 返回一个Deferred。在循环中，将Deferred 附加到列表中并“加入”所有列表或在发生故障时停止处理（阅读gatherResults）。一旦站点全部被刮掉，反应堆就会停止。

在网上搜索ReactorNotRestartable，这已经解释过很多次了。

Answer 2

我也是scrapy的新手。但是我会在不重新启动爬虫的情况下编写这个爬虫（这是有问题的，因为它需要重新启动反应器）。像这样的：

class MySpider(scrapy.Spider):
    name = 'spidery'
    allowed_domains = ['allowed_list']
    start_urls = url_list
    
    def parse(self, response):
       #here is what you want to scrape

crawler = CrawlerProcess()
crawler.crawl(MySpider)
crawler.start()

我希望应该这样做！