1
我想刮掉amazon.com的產品鏈接有超過800個評論,但我一直從下一頁獲得相同的頁面鏈接按鈕,它一直返回頁面2一遍又一遍我應該得到第3,4頁等Python Scrapy繼續從下一頁按鈕獲得相同的頁面鏈接
我已經設置了條件溢出和轉換審查字符像1,020整數和比較如果大於800或不以此爲基礎訪問頁
這裏是代碼
# -*- coding: utf-8 -*-
import scrapy
from amazon.items import AmazonItem
from urlparse import urljoin
class AmazonspiderSpider(scrapy.Spider):
name = "amazonspider"
DOWNLOAD_DELAY = 1
start_urls = ['https://www.amazon.com/s/ref=lp_165993011_nr_n_0?fst=as%3Aoff&rh=n%3A165793011%2Cn%3A%21165795011%2Cn%3A165993011%2Cn%3A2514571011&bbn=165993011&ie=UTF8&qid=1493778423&rnid=165993011']
def parse(self, response):
SET_SELECTOR = '.a-carousel-card.acswidget-carousel__card'
for attr in response.css(SET_SELECTOR):
#print '\n\n', attr
item = AmazonItem()
review_selector = './/*[@class="acs_product-rating__review-count"]/text()'
link_selector = './/*[@class="a-link-normal"]/@href'
if attr.xpath(review_selector).extract_first():
if int(''.join(attr.xpath(review_selector).extract_first().split(','))) >= 800:
url = urljoin(response.url, attr.xpath(link_selector).extract_first())
item['LINKS'] = url
if url:
yield scrapy.Request(url, callback=self.parse_link, meta={'item': item})
next_page = './/span[@class="pagnRA"]/a[@id="pagnNextLink"]/@href'
next_page = response.xpath(next_page).extract_first()
print '\n\n', urljoin(response.url, next_page)
if next_page:
yield scrapy.Request(
urljoin(response.url, next_page),
callback=self.parse
)
def parse_link(self, response):
item = AmazonItem(response.meta['item'])
catselector = '.cat-link ::text'
defaultcatselector = '.nav-search-label ::text'
cat = response.css(catselector).extract_first()
if cat:
item['CATAGORY'] = cat
else:
item['CATAGORY'] = response.css(defaultcatselector).extract_first()
return item
這裏是當我打印調用解析函數之前下一個頁面鏈接輸出遞歸