Python自定义scrapy中间模块避免重复采集的方法_Python

本文实例讲述了Python自定义scrapy中间模块避免重复采集的方法。分享给大家供大家参考。具体如下：

				?

									from scrapy import log

									from scrapy.http import Request

									from scrapy.item import BaseItem

									from scrapy.utils.request import request_fingerprint

									from myproject.items import MyItem

									class IgnoreVisitedItems(object):

									  """Middleware to ignore re-visiting item pages if they

									  were already visited before. 

									  The requests to be filtered by have a meta['filter_visited']

									  flag enabled and optionally define an id to use 

									  for identifying them, which defaults the request fingerprint,

									  although you'd want to use the item id,

									  if you already have it beforehand to make it more robust.

									  """

									  FILTER_VISITED = 'filter_visited'

									  VISITED_ID = 'visited_id'

									  CONTEXT_KEY = 'visited_ids'

									  def process_spider_output(self, response, result, spider):

									    context = getattr(spider, 'context', {})

									    visited_ids = context.setdefault(self.CONTEXT_KEY, {})

									    ret = []

									    for x in result:

									      visited = False

									      if isinstance(x, Request):

									        if self.FILTER_VISITED in x.meta:

									          visit_id = self._visited_id(x)

									          if visit_id in visited_ids:

									            log.msg("Ignoring already visited: %s" % x.url,

									                level=log.INFO, spider=spider)

									            visited = True

									      elif isinstance(x, BaseItem):

									        visit_id = self._visited_id(response.request)

									        if visit_id:

									          visited_ids[visit_id] = True

									          x['visit_id'] = visit_id

									          x['visit_status'] = 'new'

									      if visited:

									        ret.append(MyItem(visit_id=visit_id, visit_status='old'))

									      else:

									        ret.append(x)

									    return ret

									  def _visited_id(self, request):

									    return request.meta.get(self.VISITED_ID) or request_fingerprint(request)

希望本文所述对大家的Python程序设计有所帮助。

Python自定义scrapy中间模块避免重复采集的方法

相关文章

热门资讯