一、分布式爬虫简述
(一)分布式爬虫优势
1.充分利用多台机器的带宽速度
2.充分利用多台机器的ip地址
(二)Redis数据库
1.Redis是一个高性能的nosql数据库
2.Redis的所有操作都是原子性的
3.Redis的数据类型都是基于基本数据结构,无需额外的抽象
4.Redis五种数据类型:string、hash、list、set、zset(sorted set)
(三)python操作redis数据库
1.终端:pip install redis
2.代码如下
import redis
db = redis.Redis(host="localhost", port="6379", decode_responses=True)# 如果用到相同的key值,可以自动修改
db.set("name", "Sam")
db.set("name2", "张三")print(db.get("name2"))# 多个值
db.mset({"k1":"v1","k2":"v2"})
print(db.mget("k1","k2","name2"))# hash
db.hset("hash1","hkey1","hvalue1")
db.hset("hash1","hkey2","hvalue2")
db.hset("hash1","hkey3","hvalue3")print(db.hget("hash1","hkey2"))
print(db.hgetall("hash1"))db.lpush("list1",11,22,33)
print(db.llen("list1"))
print(db.lrange("list1",0,-1))db.sadd("set1", 55, 44 ,77)
print(db.scard("set1"))
print(db.smembers("set1"))db.zadd("zset1",{"item1":1,"item2":2,"item3":2})
print(db.zcard("zset1"))
print(db.zrange("zset1",0,-1))
print(db.zrange("zset1",0,-1,withscores=True))
(四)Redis数据保存至mongodb数据库
import redis
import pymongo
import jsondb_redis = redis.Redis(host="localhost", port="6379", decode_responses=True)client_mongo = pymongo.MongoClient("mongodb://localhost:27017")
db_mongo = client_mongo["RedisToMongo"]
col_mongo = db_mongo["C1"]for i in db_redis.lrange("app:items", 0 -1):page = {"title":json.loads(i)["title"]}res = col_mongo.insert_one(page)print(res.inserted_id)
二、分布式爬虫实战
实现一个简单的分布式:
1.创建爬虫项目和文件同scrapy一样的步骤
2.修改settings.py文件中的user-agent、robotstxt_obey、log_level、打开注释掉的item_piplines
3.终端安装scrapy-redis:pip install scrapy-redis
4.在app.py文件中修改如下代码:
import scrapy
from ..items import C07L07Item
from scrapy_redis.spiders import RedisSpiderclass AppSpider(RedisSpider):name = "app"redis_key = "app"# start_urls = ["http://127.0.0.1:5000/C07L07"]def __init__(self, *args, **kwargs):domain = kwargs.pop("domain","")self.allowed_domains = filter(None, domain.split(","))super(AppSpider, self).__init__(*args, **kwargs)def parse(self, response):links = response.xpath('//a/@href').getall()for link in links:link = "http://127.0.0.1:5000"+linkyield scrapy.Request(url=link,callback=self.parse_details, dont_filter=True)def parse_details(self, response):item = C07L07Item()item["title"] = response.textyield item
在items.py文件中修改数据结构
import scrapyclass C07L07Item(scrapy.Item):title = scrapy.Field()
在pipelines.py文件中修改代码
from itemdapter import ItemAdapterclass C07L07Pipeline:def process_item(self, item, spider):print(item["title"])return item
5.在settings.py文件中添加如下代码,修改ITEM_PIPELINES
DUPEFILTER_CLASS = "scrapy_redis.dupefilter.RFPDupeFilter"
SCHEDULER = "scrapy_redis.scheduler.Scheduler"
SCHEDULER_PERSIST = TrueREDIS_URL = "redis://127.0.0.1:6379"
DOWNLOAD_DELAY = 1ITEM_PIPELINES = {"C07LO7.pipelines.C07LO7Pipeline":300,"scrapy_redis.pipelines.RedisPipeline":400
}
6.在终端链接redis数据库:redis-cli
lpush app http://127.0.0.1:5000/C07L07
7.运行爬虫代码:scrapy crawl app(可以开多进程)