pipeline补充

yield item 会执行 process_item

def process_item　　return item 交给下一个pipeline

　　　　　　　　　　raise DropItem　　丢弃掉，不给下一个pipeline

crawler.settings　　crawler封装了settings

配置文件必须大写

# -*- coding: utf-8 -*-

# Define your item pipelines here
#
# Don't forget to add your pipeline to the ITEM_PIPELINES setting
# See: http://doc.scrapy.org/en/latest/topics/item-pipeline.html

from scrapy.exceptions import DropItem

class Day96Pipeline(object):

    def __init__(self,conn_str):
        self.conn_str = conn_str

    @classmethod
    def from_crawler(cls, crawler):
        """
        初始化时候，用于创建pipeline对象
        :param crawler:
        :return:
        """
        conn_str = crawler.settings.get('DB')
        return cls(conn_str)

    def open_spider(self,spider):
        """
        爬虫开始执行时，调用
        :param spider:
        :return:
        """
        self.conn = open(self.conn_str, 'a')

    def close_spider(self,spider):
        """
        爬虫关闭时，被调用
        :param spider:
        :return:
        """
        self.conn.close()

    def process_item(self, item, spider):
        """
        每当数据需要持久化时，就会被调用
        :param item:
        :param spider:
        :return:
        """
        # if spider.name == 'chouti'
        tpl = "%s\n%s\n\n" %(item['title'],item['href'])
        self.conn.write(tpl)
        # 交给下一个pipeline处理
        return item
        # 丢弃item，不交给
        # raise DropItem()

class Day97Pipeline(object):

    def __init__(self,conn_str):
        self.conn_str = conn_str

    @classmethod
    def from_crawler(cls, crawler):
        """
        初始化时候，用于创建pipeline对象
        :param crawler:
        :return:
        """
        conn_str = crawler.settings.get('DB')
        return cls(conn_str)

    def open_spider(self,spider):
        """
        爬虫开始执行时，调用
        :param spider:
        :return:
        """
        self.conn = open(self.conn_str, 'a')

    def close_spider(self,spider):
        """
        爬虫关闭时，被调用
        :param spider:
        :return:
        """
        self.conn.close()

    def process_item(self, item, spider):
        """
        每当数据需要持久化时，就会被调用
        :param item:
        :param spider:
        :return:
        """
        # if spider.name == 'chouti'
        tpl = "%s\n%s\n\n" %(item['title'],item['href'])
        self.conn.write(tpl)

pipelines.py