Scrapy selector not working on Splash response

Question

I'm trying to scrape some dynamic content using Scrapy. I have succesfully set up Splash to work along with it. However, the selectors of the following spider yield empty results:

# -*- coding: utf-8 -*- 

import scrapy
from scrapy.selector import Selector
from scrapy_splash import SplashRequest

class CartierSpider(scrapy.Spider):
  name = 'cartier'
  start_urls = ['http://www.cartier.co.uk/en-gb/collections/watches/mens-watches/ballon-bleu-de-cartier/w69017z4-ballon-bleu-de-cartier-watch.html']

  def start_requests(self):
    for url in self.start_urls:
      yield SplashRequest(url, self.parse, args={'wait': 0.5})

  def parse(self, response):
    yield {
      'title': response.xpath('//title').extract(),
      'link': response.url,
      'productID': Selector(text=response.body).xpath('//span[@itemprop="productID"]/text()').extract(),
      'model': Selector(text=response.body).xpath('//span[@itemprop="model"]/text()').extract(),
      'price': Selector(text=response.body).css('div.price-wrapper').xpath('.//span[@itemprop="price"]/text()').extract(),
    }

The selectors work just fine using the Scrapy shell, so I'm very confused about what is not working.

The only difference I can find among the two situations is that the encoding of the string response.body is treated differently: it's just gibberish if I try to print/decode it from within the parse function.

Any hint or reference would be greatly appreciated.

Answer 1

Your spider works fine with me, with Scrapy 1.1, Splash 2.1 and no modification of the code in your question, just using settings suggested in https://github.com/scrapy-plugins/scrapy-splash

As other have mentioned, your parse function can be simplified by using response.css() and response.xpath() directly, without needing to re-build a Selector from the response.

I tried with:

import scrapy
from scrapy.selector import Selector
from scrapy_splash import SplashRequest

class CartierSpider(scrapy.Spider):
  name = 'cartier'
  start_urls = ['http://www.cartier.co.uk/en-gb/collections/watches/mens-watches/ballon-bleu-de-cartier/w69017z4-ballon-bleu-de-cartier-watch.html']

  def start_requests(self):
    for url in self.start_urls:
      yield SplashRequest(url, self.parse, args={'wait': 0.5})

  def parse(self, response):
    yield {
      'title': response.xpath('//title/text()').extract_first(),
      'link': response.url,
      'productID': response.xpath('//span[@itemprop="productID"]/text()').extract_first(),
      'model': response.xpath('//span[@itemprop="model"]/text()').extract_first(),
      'price': response.css('div.price-wrapper').xpath('.//span[@itemprop="price"]/text()').extract_first(),
    }

and got this:

$ scrapy crawl cartier
2016-06-08 17:16:08 [scrapy] INFO: Scrapy 1.1.0 started (bot: stack37701774)
2016-06-08 17:16:08 [scrapy] INFO: Overridden settings: {'NEWSPIDER_MODULE': 'stack37701774.spiders', 'SPIDER_MODULES': ['stack37701774.spiders'], 'BOT_NAME': 'stack37701774'}
(...)
2016-06-08 17:16:08 [scrapy] INFO: Enabled downloader middlewares:
['scrapy.downloadermiddlewares.httpauth.HttpAuthMiddleware',
 'scrapy.downloadermiddlewares.downloadtimeout.DownloadTimeoutMiddleware',
 'scrapy.downloadermiddlewares.useragent.UserAgentMiddleware',
 'scrapy.downloadermiddlewares.retry.RetryMiddleware',
 'scrapy.downloadermiddlewares.defaultheaders.DefaultHeadersMiddleware',
 'scrapy.downloadermiddlewares.redirect.MetaRefreshMiddleware',
 'scrapy.downloadermiddlewares.redirect.RedirectMiddleware',
 'scrapy.downloadermiddlewares.cookies.CookiesMiddleware',
 'scrapy_splash.SplashCookiesMiddleware',
 'scrapy_splash.SplashMiddleware',
 'scrapy.downloadermiddlewares.httpcompression.HttpCompressionMiddleware',
 'scrapy.downloadermiddlewares.chunked.ChunkedTransferMiddleware',
 'scrapy.downloadermiddlewares.stats.DownloaderStats']
2016-06-08 17:16:08 [scrapy] INFO: Enabled spider middlewares:
['scrapy.spidermiddlewares.httperror.HttpErrorMiddleware',
 'scrapy.spidermiddlewares.offsite.OffsiteMiddleware',
 'scrapy.spidermiddlewares.referer.RefererMiddleware',
 'scrapy.spidermiddlewares.urllength.UrlLengthMiddleware',
 'scrapy.spidermiddlewares.depth.DepthMiddleware']
2016-06-08 17:16:08 [scrapy] INFO: Enabled item pipelines:
[]
2016-06-08 17:16:08 [scrapy] INFO: Spider opened
2016-06-08 17:16:08 [scrapy] INFO: Crawled 0 pages (at 0 pages/min), scraped 0 items (at 0 items/min)
2016-06-08 17:16:08 [scrapy] DEBUG: Telnet console listening on 127.0.0.1:6023
2016-06-08 17:16:11 [scrapy] DEBUG: Crawled (200) <GET http://www.cartier.co.uk/en-gb/collections/watches/mens-watches/ballon-bleu-de-cartier/w69017z4-ballon-bleu-de-cartier-watch.html via http://localhost:8050/render.html> (referer: None)
2016-06-08 17:16:11 [scrapy] DEBUG: Scraped from <200 http://www.cartier.co.uk/en-gb/collections/watches/mens-watches/ballon-bleu-de-cartier/w69017z4-ballon-bleu-de-cartier-watch.html>
{'model': u'Ballon Bleu de Cartier watch', 'productID': u'W69017Z4', 'link': 'http://www.cartier.co.uk/en-gb/collections/watches/mens-watches/ballon-bleu-de-cartier/w69017z4-ballon-bleu-de-cartier-watch.html', 'price': None, 'title': u'CRW69017Z4 - Ballon Bleu de Cartier watch - 36 mm, steel, leather - Cartier'}
2016-06-08 17:16:11 [scrapy] INFO: Closing spider (finished)
2016-06-08 17:16:11 [scrapy] INFO: Dumping Scrapy stats:
{'downloader/request_bytes': 618,
 'downloader/request_count': 1,
 'downloader/request_method_count/POST': 1,
 'downloader/response_bytes': 213006,
 'downloader/response_count': 1,
 'downloader/response_status_count/200': 1,
 'finish_reason': 'finished',
 'finish_time': datetime.datetime(2016, 6, 8, 15, 16, 11, 201281),
 'item_scraped_count': 1,
 'log_count/DEBUG': 3,
 'log_count/INFO': 7,
 'response_received_count': 1,
 'scheduler/dequeued': 2,
 'scheduler/dequeued/memory': 2,
 'scheduler/enqueued': 2,
 'scheduler/enqueued/memory': 2,
 'splash/render.html/request_count': 1,
 'splash/render.html/response_count/200': 1,
 'start_time': datetime.datetime(2016, 6, 8, 15, 16, 8, 545105)}
2016-06-08 17:16:11 [scrapy] INFO: Spider closed (finished)

Answer 2

I've tried that SplashRequest and ran into the same problem as you did. After messing around I tried executing a LUA script instead.

script = """
                function main(splash)
  local url = splash.args.url
  assert(splash:go(url))
  assert(splash:wait(0.5))
  return {
    html = splash:html(),
    png = splash:png(),
    har = splash:har(),
  }
end
"""

And then make the request using the script as an argument. You can mess around with the script. Test it on the shell at localhost:9200 or another port that you chose.

yield SplashRequest(
            url,
            self.parse, args={'lua_source': self.script}, endpoint='execute')

Oh and by the way, the way you yield info is just weird, use items instead.