scrapy extention实战-空闲时关闭爬虫

scrapy extention实战

 

1.      空闲-关闭

使用扩展+spider_idle信号关闭爬虫。

启用扩展：settings.py

EXTENSIONS = {
    #'scrapy.extensions.telnet.TelnetConsole': None,
    'extention_my.RedisSpiderSmartIdleClosedExensions': 300,
}

 

额外配置参数：conf.py

MYEXT_ENABLED = True
IDLE_NUMBER = 5

 

扩展类：

extention_my.py

#coding:utf-8

"""
----------------------------------------
description:

author: sss

date:
----------------------------------------
change:
   
----------------------------------------

"""
__author__ = 'sss'

import time
from scrapy import signals
from scrapy.exceptions import NotConfigured

from utils.mylogger import mylogger

logger_c = mylogger(__name__)
logger_m = logger_c.logger

class RedisSpiderSmartIdleClosedExensions(object):

    def __init__(self, idle_number, crawler):
        self.crawler = crawler
        self.idle_number = idle_number
        self.idle_list = []
        self.idle_count = 0

    @classmethod
    def from_crawler(cls, crawler):
        # 首先检查是否应该启用和提升扩展
        # 不然不配置
        from conf import MYEXT_ENABLED
        if not MYEXT_ENABLED:
            raise NotConfigured

        # 获取配置中的时间片个数，默认为360个，30分钟
        from conf import IDLE_NUMBER as idle_number

        # 实例化扩展对象
        ext = cls(idle_number, crawler)

        # 将扩展对象链接到信号， 将signals.spider_idle 与 spider_idle() 方法关联起来。
        crawler.signals.connect(ext.spider_opened, signal=signals.spider_opened)
        crawler.signals.connect(ext.spider_closed, signal=signals.spider_closed)
        crawler.signals.connect(ext.spider_idle, signal=signals.spider_idle)

        # return the extension object
        return ext

    def spider_opened(self, spider):
        logger_m.info("opened spider %s redis spider Idle, Continuous idle limit： %d", spider.name, self.idle_number)

    def spider_closed(self, spider):
        logger_m.info("closed spider %s, idle count %d , Continuous idle count %d",
                    spider.name, self.idle_count, len(self.idle_list))

    def spider_idle(self, spider):
        self.idle_count += 1  # 空闲计数
        self.idle_list.append(time.time())  # 每次触发 spider_idle时，记录下触发时间戳
        idle_list_len = len(self.idle_list)  # 获取当前已经连续触发的次数
        print(self.idle_number, self.idle_count, self.idle_list)

        # 判断 当前触发时间与上次触发时间 之间的间隔是否大于5秒，若是大于5秒，说明redis 中还有key
        if idle_list_len > 2 and self.idle_list[-1] - self.idle_list[-2] > 6:
            self.idle_list = [self.idle_list[-1]]

        elif idle_list_len > self.idle_number:
            # 连续触发的次数达到配置次数后关闭爬虫
            logger_m.info('\n continued idle number exceed {} Times'
                        '\n meet the idle shutdown conditions, will close the reptile operation'
                        '\n idle start time: {},  close spider time: {}'.format(self.idle_number,
                                                                                self.idle_list[0], self.idle_list[0]))
            # 执行关闭爬虫操做
            self.crawler.engine.close_spider(spider, 'closespider_pagecount')

其它没有什么，主要是判断是否关闭条件的设计。