How to add new colum to Scrapy output from csv?

问题:

I parse websites and it works fine but I need to add new colum with IDs to output. That column is saved in csv with urls:

https://www.ceneo.pl/48523541, 1362
https://www.ceneo.pl/46374217, 2457

Code of my spider:

import scrapy
from ceneo.items import CeneoItem
import csv

class QuotesSpider(scrapy.Spider):
    name = "quotes" 

    def start_requests(self):
        start_urls = []
        f = open('urls.csv', 'r')
        for i in f:
            u = i.split(',')
            start_urls.append(u[0])
        for url in start_urls:
            yield scrapy.Request(url=url, callback=self.parse)

    def parse(self, response):
        all_prices = response.xpath('(//td[@class="cell-price"] /a/span/span/span[@class="value"]/text())[position() <= 10]').extract()
        all_sellers = response.xpath('(//tr/td/div/ul/li/a[@class="js_product-offer-link"]/text())[position()<=10]').extract()

        f = open('urls.csv', 'r')
        id = []
        for i in f:
            u = i.split(',')
            id.append(u[1])

        x = len(all_prices)     
        i = 0

        while (i < x):
            all_sellers[i] = all_sellers[i].replace('Opinie o ', '')
            i += 1

        for urlid, price, seller in zip(id, all_prices, all_sellers):
            yield {'urlid': urlid.strip(), 'price': price.strip(), 'seller': seller.strip()}

In the results I get wrong data because (zip funtion?) IDs are taken alternately:

urlid,price,seller
1362,109,eMAG
1457,116,electro.pl
1362,597,apollo.pl
1457,597,allegro.pl

And it should output:

urlid,price,seller
1362,109,eMAG
1362,116,electro.pl
1457,597,apollo.pl
1457,597,allegro.pl

回答1:

You can get ID in start_requests and assign to request using meta={'id': id_} and later in parse you can get ID using response.meta['id'].

This way you will have correct ID in parse.

I use string data instead of file to create working example.

#!/usr/bin/env python3

import scrapy

data = '''https://www.ceneo.pl/48523541, 1362
https://www.ceneo.pl/46374217, 2457'''

class QuotesSpider(scrapy.Spider):

    name = "quotes" 

    def start_requests(self):
        #f = open('urls.csv', 'r')

        f = data.split('\n')

        for row in f:
            url, id_ = row.split(',')

            url = url.strip()
            id_ = id_.strip()

            #print(url, id_)

            # use meta to assign value 
            yield scrapy.Request(url=url, callback=self.parse, meta={'id': id_})

    def parse(self, response):
        # use meta to receive value
        id_ = response.meta["id"]

        all_prices = response.xpath('(//td[@class="cell-price"] /a/span/span/span[@class="value"]/text())[position() <= 10]').extract()
        all_sellers = response.xpath('(//tr/td/div/ul/li/a[@class="js_product-offer-link"]/text())[position()<=10]').extract()

        all_sellers = [ item.replace('Opinie o ', '') for item in all_sellers ]

        for price, seller in zip(all_prices, all_sellers):
            yield {'urlid': id_, 'price': price.strip(), 'seller': seller.strip()}

# --- it runs without project and saves in `output.csv` ---

from scrapy.crawler import CrawlerProcess

c = CrawlerProcess({
    'USER_AGENT': 'Mozilla/5.0',
    'FEED_FORMAT': 'csv',
    'FEED_URI': 'output.csv', 
})
c.crawl(QuotesSpider)
c.start()

BTW: there is standard function id() so I use variable id_ instead of id