Python Scrapy请求-处理一组又一组URL-我可以使用优先级吗?
如何让scrapy一组一组地处理URL?我有两个URL列表。我必须处理第一个列表,包括Python Scrapy请求-处理一组又一组URL-我可以使用优先级吗?,python,scrapy,scrapy-spider,Python,Scrapy,Scrapy Spider,如何让scrapy一组一组地处理URL?我有两个URL列表。我必须处理第一个列表,包括项目管道,然后才能处理第二个列表 两者都应由一个蜘蛛处理 我不确定优先级是否对我有帮助 优先级(int)–此请求的优先级(默认为0)。调度程序使用优先级定义用于处理请求的顺序。优先级值较高的请求将更早执行。允许负值,以指示相对较低的优先级 因为我不知道它是否只是根据优先级重新排序请求——它可能会在第一个列表的最后一个URL之前通过管道连接第二个列表的第一个URL 我能否确保第一个列表中的项将在第二个列表中的项之
项目管道
,然后才能处理第二个列表
两者都应由一个蜘蛛处理
我不确定优先级是否对我有帮助
优先级(int)–此请求的优先级(默认为0)。调度程序使用优先级定义用于处理请求的顺序。优先级值较高的请求将更早执行。允许负值,以指示相对较低的优先级
因为我不知道它是否只是根据优先级重新排序请求
——它可能会在第一个列表的最后一个URL之前通过管道连接第二个列表的第一个URL
我能否确保第一个列表中的项
将在第二个列表中的项之前导出为XML(我使用的是XMLItemExporter
)
编辑:
错误(@Wilfredo):
2017-11-23 20:12:16[scrapy.utils.signal]错误:发现错误
信号处理程序:>回溯(最近一次调用last):
文件
“/home/milano/.virtualenvs/eoilenv/local/lib/python2.7/site packages/scrapy/utils/signal.py”,
第30行,发送捕获日志
*参数,**命名)文件“/home/milano/.virtualenvs/eoilenv/local/lib/python2.7/site packages/pydispatch/robustapply.py”,
第55行,在robustApply中
return receiver(*参数,**命名)TypeError:spider_idle()正好接受2个参数(给定1个)2017-11-23 20:12:16
[scrapy.core.engine]信息:关闭十字轴(已完成)
编辑二:
# coding=utf-8
import scrapy
from bot.items import TestItem
from scrapy import Spider, Request, signals
from scrapy.exceptions import DontCloseSpider
class IndexSpider(Spider):
name = 'index_spider'
allowed_domains = ['www.scrape.com']
def start_requests(self):
for url in ["https://www.scrape.com/eshop"]:
# for url in ["https://www.scrape.com/search/getAjaxResult?categoryId=1&"]:
yield Request(url, callback=self.parse_main_page)
def parse_main_page(self, response):
# get subcategories and categories
self.categories = []
self.subcategories = []
parts = response.selector.xpath("//div[contains(@class,'side-nav') and not(contains(@class,'side-nav-'))]")
for part in parts:
part_name = part.xpath('.//h4/text()').extract_first().strip()
category_list = [part_name]
categories_ul = part.xpath('./ul')
categories_lis = categories_ul.xpath('./li')
for category_li in categories_lis:
category_list = category_list[:1]
category_name = category_li.xpath('./a/text()').extract_first().strip()
category_href = category_li.xpath('./a/@href').extract_first().strip()
categoryId = self._extract_categoryId_from_url(category_href)
category_list.append(category_name)
self.categories.append((categoryId, category_list))
subcategories_lis = category_li.xpath('.//li')
for subcategory_li in subcategories_lis:
category_list = category_list[:2]
subcategory_href = subcategory_li.xpath('./a/@href').extract_first()
subcategory_name = subcategory_li.xpath('./a/text()').extract_first().strip()
subcategoryId = self._extract_categoryId_from_url(subcategory_href)
category_list.append(subcategory_name)
self.subcategories.append((subcategoryId, category_list))
# Scrape all subcategories (then categories)
# for sub in self.subcategories:
# url = "https://www.scrape.com/search/getAjaxResult?categoryId={}".format(sub[0])
# yield Request(url,meta={'tup':sub,'priority':1,'type':'subcategory'},priority=1,callback=self.parse_category)
def parse_category(self, response):
tup = response.meta['tup']
type = response.meta['type']
priority = response.meta['priority']
current_page = response.meta.get('page', 1)
categoryId = tup[0]
categories_list = tup[1]
number_of_pages_href = response.selector.xpath(u'//a[text()="Last"]/@href').extract_first()
try:
number_of_pages = int(number_of_pages_href.split('p=')[1].split('&')[0])
except:
number_of_pages = current_page
if current_page < number_of_pages:
url = "https://www.scrape.com/search/getAjaxResult/?categoryId={}&p={}".format(categoryId, current_page + 1)
yield Request(url, self.parse_category, meta={'tup': tup, 'page': current_page + 1,'priority':priority,'type':type}, priority=priority)
hrefs = self._extract_all_product_urls(response)
for href in hrefs:
yield Request(href, self.parse_product, meta={"categories_list": categories_list,'type':type}, priority=2 if priority==1 else -1)
def parse_product(self, response):
yield TestItem(url=response.url,type=response.meta['type'], category_text='|'.join(response.meta['categories_list']))
def _extract_categoryId_from_url(self, url):
categoryId = url.split('/')[-2]
return categoryId
def _extract_all_product_urls(self, response):
hrefs = response.selector.xpath("//a[contains(@class, 'shop-item-image')]/@href").extract()
return [u"https://www.scrape.com{}".format(x) for x in hrefs]
@classmethod
def from_crawler(cls, crawler, *args, **kwargs):
spider = super(IndexSpider, cls).from_crawler(crawler, *args, **kwargs)
crawler.signals.connect(spider.spider_idle,
signal=scrapy.signals.spider_idle)
return spider
def spider_idle(self):
self.crawler.signals.disconnect(self.spider_idle,
signal=scrapy.signals.spider_idle)
# yield a new group of urls
if self.categories:
for cat in self.categories:
url = "https://www.scrape.com/search/getAjaxResult?categoryId={}".format(cat[0])
yield Request(url, meta={'tup': cat, 'priority': 0, 'type': 'category'}, priority=0,
callback=self.parse_category)
self.categories = []
raise DontCloseSpider()
#编码=utf-8
进口羊瘙痒
从bot.items导入测试项
来自scrapy进口蜘蛛、请求、信号
从scrapy.exceptions导入DontCloseSpider
类索引Spider(Spider):
名称='index_spider'
允许的_域=['www.scrap.com']
def start_请求(自我):
对于[”中的urlhttps://www.scrape.com/eshop"]:
#对于[”中的urlhttps://www.scrape.com/search/getAjaxResult?categoryId=1&"]:
屈服请求(url,callback=self.parse\u主页)
def parse_主页面(自我,响应):
#获取子类别和类别
self.categories=[]
self.subcategories=[]
parts=response.selector.xpath(//div[contains(@class,'side-nav')和not(contains(@class,'side-nav-'))]))
对于部分中的部分:
part_name=part.xpath('.//h4/text()')。提取_first().strip()
类别列表=[零件名称]
categories_ul=part.xpath('./ul')
categories\u lis=categories\u ul.xpath('./li')
对于类别列表中的类别列表:
类别列表=类别列表[:1]
category_name=category_li.xpath('./a/text()')。extract_first().strip()
category_href=category_li.xpath('./a/@href').extract_first().strip()
categoryId=self.\u从\u url(category\u href)提取\u categoryId\u
类别列表。追加(类别名称)
self.categories.append((categoryId,category_list))
子类别\u lis=category\u li.xpath('.//li')
对于子类别列表中的子类别列表:
类别列表=类别列表[:2]
subcategory_href=subcategory_li.xpath('./a/@href')。extract_first()
subcategory_name=subcategory_li.xpath('./a/text()')。extract_first().strip()
subcategoryId=self.\u从url中提取\u categoryId\u(subcategory\u href)
类别列表。追加(子类别名称)
self.subcategories.append((子类别ID,类别列表))
#刮除所有子类别(然后是类别)
#对于self.subcategories中的子类:
#url=”https://www.scrape.com/search/getAjaxResult?categoryId={}.格式(sub[0])
#产生请求(url,meta={'tup':sub,'priority':1,'type':'subcategory'},priority=1,callback=self.parse_category)
def parse_类别(自我、响应):
tup=response.meta['tup']
type=response.meta['type']
优先级=响应.meta['priority']
当前页面=response.meta.get('page',1)
categoryId=tup[0]
类别列表=tup[1]
页数\u href=response.selector.xpath(u'//a[text()=“Last”]/@href')。extract\u first()
尝试:
页数=int(页数href.split('p=')[1]。split('&')[0])
除:
页面数量=当前页面
如果当前页面<页面数量:
url=”https://www.scrape.com/search/getAjaxResult/?categoryId={}&p={}.格式(类别ID,当前页面+1)
产生请求(url,self.parse_category,meta={'tup':tup,'page':当前_页面+1,'priority':priority,'type':type},priority=priority)
hrefs=self.\u提取\u所有\u产品\u URL(响应)
对于hrefs中的href:
屈服请求(href,self.parse_product,meta={“categories_list”:categories_list,'type':type},如果优先级==1,优先级=2,否则-1)
def parse_产品(自我,响应):
屈服测试项(url=response.url,type=response.meta['type'],category_text='|'。加入(response.meta['categories_list']))
定义从url(自我,url)中提取类别ID:
categoryId=url.split('/')[-2]
返回类别ID
def_extract_all_product_URL(自我、响应):
hrefs=response.selector.xpath(//a[contains(@class,'shop item image')]/@href”).extract()
返回[u”https://www.scrape.com{}.hrefs中x的格式(x)]
@类方法
来自_爬虫程序的def(cls、爬虫程序、*args、**kwargs):
spider=super(IndexSpider,cls)。来自爬虫程序(爬虫程序,*args,**kwargs)
爬虫。信号。连接(spider.spider\u空闲,
信号=刮擦。信号。十字轴(空闲)
回程卡盘
def卡盘_怠速(自):
self.crawler.signals.disconnect(self.spider\u idle,
def start_requests(self):
urls = ['url1', 'url2']
yield Request(
url=urls[0],
callback=self.process_request,
meta={'urls': urls, 'current_index':0}
def process_request(self, response):
# do my thing
yield {} # yield item
current_index = response.meta['current_index'] + 1
if current_index < len(response.meta['urls']:
yield Request(
url=response.meta['urls'][current_index],
callback=self.process_request,
meta={'urls': response.meta['urls'], 'current_index': current_index})
# -*- coding: utf-8 -*-
import scrapy
from scrapy.exceptions import DontCloseSpider
class ExampleSpider(scrapy.Spider):
name = 'example'
allowed_domains = ['example.com']
first_group_of_urls = [
'http://quotes.toscrape.com/page/1/'
]
second_group_of_urls = [
'http://quotes.toscrape.com/page/2/'
]
def start_requests(self):
for url in self.first_group_of_urls:
yield scrapy.Request(url=url, callback=self.parse)
def parse(self, response):
self.logger.debug('In response from %s', response.url)
@classmethod
def from_crawler(cls, crawler, *args, **kwargs):
spider = super(ExampleSpider, cls).from_crawler(crawler, *args, **kwargs)
crawler.signals.connect(spider.spider_idle,
signal=scrapy.signals.spider_idle)
return spider
def spider_idle(self):
self.crawler.signals.disconnect(self.spider_idle,
signal=scrapy.signals.spider_idle)
for url in self.second_group_of_urls:
self.crawler.engine.crawl(scrapy.Request(url), self)
raise DontCloseSpider