Scrapy Xpath在Shell中无法使用代码

Question

For a couple of hours I have been struggling with the following. I am trying to scrape https://www.upwork.com/jobs/_~0180b9eef40aafe057/ (and similar postings)

My xpath expression does work in the shell and in a xpath verifier but not in my code.

When I output the response into a text file using:

    with open('response.html','w+') as f:
        f.write(response.body)

and then test the xpath on the html code by using http://videlibri.sourceforge.net/cgi-bin/xidelcgi it is working fine.

This works in the Shell:

for item in response.xpath("//p[strong = 'About the Client']/following-sibling::p"):
    print " ".join(map(unicode.strip, item.xpath(".//text()").extract()))
    print 'Succes!'

But using it into my Scrapy Spider it returns nothing.

I've tried a lot of different solutions but nothing seems to work.

EDIT added complete code:

from scrapy.contrib.spiders.init import InitSpider
from scrapy.http import Request, FormRequest
import scrapy
from scrapy.spiders import CrawlSpider, Rule
from scrapy.linkextractors import LinkExtractor
from ypscrape.items import item1
from scrapy.loader.processors import Join, MapCompose, TakeFirst
from scrapy.loader import ItemLoader
import arrow

import logging
import re


class MySpider(CrawlSpider):

    # Login credentials for account, so more details are available
    #Rss Token for the RSS feed which pulls the new links
    rsstoken = 'REDACTED'
    user = 'REDACTED'
    password = 'REDACTED'

    name = 'dataup'
    allowed_domains = ['upwork.com']
    login_page = 'http://www.upwork.com/login'
    rssurl = 'https://www.upwork.com/ab/feed/jobs/rss?api_params=1&q=&securityToken='+ rsstoken

    # can probably be removed
    rules = (
    Rule(LinkExtractor(allow=r'upwork\.com\/jobs\/.*?_%'), callback='parse_item', follow=False),
    )

    #Called when Spider is started, initiates the login request
    def start_requests(self):
        self.log("start request started")
        yield Request(
            url=self.login_page,
            callback=self.login,
            dont_filter=True
        )

    # Use the RSS feed to gather the newest links
    def get_urls_from_rss(self, response):
        urllist = []
        content = response
        self.log("Get rss from url")
        #print str(content.body)
        gathered_urls = re.findall('(https\:\/\/.*?\/jobs\/.*?)source=rss', str(content.body))

        # Request the URLS and send them to parse_item
        for url in gathered_urls:
            if url not in urllist:
                #Check if URL has not been visited before ADD THIS
                urllist.append(url)
                yield scrapy.Request(url, callback=self.parse_item)

    def login(self, response):
        """Generate a login request."""

        self.log("login request started")
        return FormRequest.from_response(response,formname='login',
                    formdata={'login[username]': self.user, 'login[password]': self.password} 
                    , callback=self.check_login_response, method="POST")

    def check_login_response(self, response):
        """Check the response returned by a login request to see if we are
        successfully logged in.
        """
        self.log("check request started")

        if "<title>My Job Feed</title>" in response.body:
            self.log("Successfully logged in. Let's start crawling!")
            # Now the crawling can begin..
            yield scrapy.Request(self.rssurl, callback = self.get_urls_from_rss )
        else:
            self.log("Bad times :( Logging in failed")
            # Something went wrong, we couldn't log in, so nothing happens.
        #return self.initialized()

    def parse_item(self, response):
        self.logger.info('Crawling item page! %s', response.url)

        # Collect the data from the page

        with open('response.html','w+') as f:
            f.write(response.body)

        for item in response.xpath("//p[strong = 'About the Client']/following-sibling::p"):
            print " ".join(map(unicode.strip, item.xpath(".//text()").extract()))
            print 'Bingo'



        l = ItemLoader(item1(), response)


        l.add_value('timestamp', arrow.utcnow().format('YYYY-MM-DD HH:mm'))
        l.add_xpath('category1', '//*[@id="layout"]/div[2]/div[3]/div[1]/a/text()')



        return l.load_item()
        # Scrape data from page

EDIT 2: I think I found the solution. Replacing the xpath with

//p[strong]/strong

seems to solve the problem. What is the problem? I think that it has to do with the encoding. It can not find 'About the Client' because the response object it gets is something like ' About the Client ' with a couple of white spaces or something else encoding related. Thank you for the help

Answer 1

I've quickly put it into a spider:

import scrapy


class UpworkSpider(scrapy.Spider):
    name = "upwork"
    allowed_domains = ["upwork.com"]
    start_urls = [
        "https://www.upwork.com/jobs/_~0180b9eef40aafe057/",
    ]

    def parse(self, response):
        for item in response.xpath("//p[strong = 'About the Client']/following-sibling::p"):
            print " ".join(map(unicode.strip, item.xpath(".//text()").extract()))
            print 'Succes!'

Then, I run it as:

$ scrapy runspider spider.py

And getting:

Croatia  Kastel Sucurac
            04:56 PM 
Succes!
 4
        Jobs Posted   0% Hire Rate,
        4 Open Jobs 
Succes!

in the output.