I have created to web spiders within the same file that scrape data from different websites. When I run each spider on its own, I can successfully pipeline the scraped data into a postgresql table from the models.py file. But when I run both spiders at the same time using the api, the tables for each spider are created, but no data can be scraped into them for some reason. I think this may have to do with the way that the defined Pipeline class is called each time a spider is run but it is hard to tell since the documentation doesn't really show the internals of the implementation. From the code I provide, do I have the correct setup for pipelining the data into the two different tables?
Models.py
from sqlalchemy.orm import sessionmaker
from models import Tickets, Tickets3, db_connect, create_vs_tickets_table, create_tc_tickets_table
class ComparatorPipeline(object):
"""Price comparison pipeline for storing scraped items in the database"""
def __init__(self):
"""
Initializes database connection and sessionmaker.
Creates deals table.
"""
engine = db_connect()
create_vs_tickets_table(engine)
create_tc_tickets_table(engine)
self.Session = sessionmaker(bind=engine)
def process_item(self, item, spider):
"""Save tickets in the database.
This method is called for every item pipeline component.
"""
if spider.name == "comparator":
session = self.Session()
ticket = Tickets(**item)
try:
session.add(ticket)
session.commit()
except:
session.rollback()
raise
finally:
session.close()
return item
elif spider.name == "comparator3":
session = self.Session()
ticket3 = Tickets3(**item)
try:
session.add(ticket3)
session.commit()
except:
session.rollback()
raise
finally:
session.close()
return item
spider definitions
import scrapy
import re
import json
from scrapy.crawler import CrawlerProcess
from scrapy import Request
from scrapy.contrib.spiders import CrawlSpider , Rule
from scrapy.selector import HtmlXPathSelector
from scrapy.selector import Selector
from scrapy.contrib.loader import ItemLoader
from scrapy.contrib.loader import XPathItemLoader
from scrapy.contrib.loader.processor import Join, MapCompose
from concert_comparator.items import ComparatorItem, ComparatorItem3
from urlparse import urljoin
from scrapy.crawler import CrawlerRunner
from twisted.internet import reactor, defer
from scrapy.utils.log import configure_logging
bandname = raw_input("Enter a bandname \n")
vs_url = "http://www.vividseats.com/concerts/" + bandname + "-tickets.html"
tc_url = "https://www.ticketcity.com/concerts/" + bandname + "-tickets.html"
class MySpider(CrawlSpider):
handle_httpstatus_list = [416]
name = 'comparator'
allowed_domains = ["www.vividseats.com"]
start_urls = [vs_url]
tickets_list_xpath = './/*[@itemtype="http://schema.org/Event"]'
def parse_json(self, response):
loader = response.meta['loader']
jsonresponse = json.loads(response.body_as_unicode())
ticket_info = jsonresponse.get('tickets')
price_list = [i.get('p') for i in ticket_info]
if len(price_list) > 0:
str_Price = str(price_list[0])
ticketPrice = unicode(str_Price, "utf-8")
loader.add_value('ticketPrice', ticketPrice)
else:
ticketPrice = unicode("sold out", "utf-8")
loader.add_value('ticketPrice', ticketPrice)
return loader.load_item()
def parse_price(self, response):
loader = response.meta['loader']
ticketsLink = loader.get_output_value("ticketsLink")
json_id_list= re.findall(r"(\d+)[^-]*$", ticketsLink)
json_id= "".join(json_id_list)
json_url = "http://www.vividseats.com/javascript/tickets.shtml?productionId=" + json_id
yield scrapy.Request(json_url, meta={'loader': loader}, callback = self.parse_json, dont_filter = True)
def parse(self, response):
"""
# """
selector = HtmlXPathSelector(response)
# iterate over tickets
for ticket in selector.select(self.tickets_list_xpath):
loader = XPathItemLoader(ComparatorItem(), selector=ticket)
# define loader
loader.default_input_processor = MapCompose(unicode.strip)
loader.default_output_processor = Join()
# iterate over fields and add xpaths to the loader
loader.add_xpath('eventName' , './/*[@class="productionsEvent"]/text()')
loader.add_xpath('eventLocation' , './/*[@class = "productionsVenue"]/span[@itemprop = "name"]/text()')
loader.add_xpath('ticketsLink' , './/*/a[@class = "btn btn-primary"]/@href')
loader.add_xpath('eventDate' , './/*[@class = "productionsDate"]/text()')
loader.add_xpath('eventCity' , './/*[@class = "productionsVenue"]/span[@itemprop = "address"]/span[@itemprop = "addressLocality"]/text()')
loader.add_xpath('eventState' , './/*[@class = "productionsVenue"]/span[@itemprop = "address"]/span[@itemprop = "addressRegion"]/text()')
loader.add_xpath('eventTime' , './/*[@class = "productionsTime"]/text()')
print "Here is ticket link \n" + loader.get_output_value("ticketsLink")
#sel.xpath("//span[@id='PractitionerDetails1_Label4']/text()").extract()
ticketsURL = "concerts/" + bandname + "-tickets/" + bandname + "-" + loader.get_output_value("ticketsLink")
ticketsURL = urljoin(response.url, ticketsURL)
yield scrapy.Request(ticketsURL, meta={'loader': loader}, callback = self.parse_price, dont_filter = True)
class MySpider3(CrawlSpider):
handle_httpstatus_list = [416]
name = 'comparator3'
allowed_domains = ["www.ticketcity.com"]
start_urls = [tc_url]
tickets_list_xpath = './/div[@class = "vevent"]'
def parse_json(self, response):
loader = response.meta['loader']
jsonresponse = json.loads(response.body_as_unicode())
ticket_info = jsonresponse.get('B')
price_list = [i.get('P') for i in ticket_info]
if len(price_list) > 0:
str_Price = str(price_list[0])
ticketPrice = unicode(str_Price, "utf-8")
loader.add_value('ticketPrice', ticketPrice)
else:
ticketPrice = unicode("sold out", "utf-8")
loader.add_value('ticketPrice', ticketPrice)
return loader.load_item()
def parse_price(self, response):
print "parse price function entered \n"
loader = response.meta['loader']
event_City = response.xpath('.//span[@itemprop="addressLocality"]/text()').extract()
eventCity = ''.join(event_City)
loader.add_value('eventCity' , eventCity)
event_State = response.xpath('.//span[@itemprop="addressRegion"]/text()').extract()
eventState = ''.join(event_State)
loader.add_value('eventState' , eventState)
event_Date = response.xpath('.//span[@class="event_datetime"]/text()').extract()
eventDate = ''.join(event_Date)
loader.add_value('eventDate' , eventDate)
ticketsLink = loader.get_output_value("ticketsLink")
json_id_list= re.findall(r"(\d+)[^-]*$", ticketsLink)
json_id= "".join(json_id_list)
json_url = "https://www.ticketcity.com/Catalog/public/v1/events/" + json_id + "/ticketblocks?P=0,99999999&q=0&per_page=250&page=1&sort=p.asc&f.t=s&_=1436642392938"
yield scrapy.Request(json_url, meta={'loader': loader}, callback = self.parse_json, dont_filter = True)
def parse(self, response):
"""
# """
selector = HtmlXPathSelector(response)
# iterate over tickets
for ticket in selector.select(self.tickets_list_xpath):
loader = XPathItemLoader(ComparatorItem(), selector=ticket)
# define loader
loader.default_input_processor = MapCompose(unicode.strip)
loader.default_output_processor = Join()
# iterate over fields and add xpaths to the loader
loader.add_xpath('eventName' , './/span[@class="summary listingEventName"]/text()')
loader.add_xpath('eventLocation' , './/div[@class="divVenue location"]/text()')
loader.add_xpath('ticketsLink' , './/a[@class="divEventDetails url"]/@href')
#loader.add_xpath('eventDateTime' , '//div[@id="divEventDate"]/@title') #datetime type
#loader.add_xpath('eventTime' , './/*[@class = "productionsTime"]/text()')
print "Here is ticket link \n" + loader.get_output_value("ticketsLink")
#sel.xpath("//span[@id='PractitionerDetails1_Label4']/text()").extract()
ticketsURL = "https://www.ticketcity.com/" + loader.get_output_value("ticketsLink")
ticketsURL = urljoin(response.url, ticketsURL)
yield scrapy.Request(ticketsURL, meta={'loader': loader}, callback = self.parse_price, dont_filter = True)
configure_logging()
runner = CrawlerRunner()
@defer.inlineCallbacks
def crawl():
yield runner.crawl(MySpider)
yield runner.crawl(MySpider3)
reactor.stop()
crawl()
reactor.run()
I don't quite understand your problem but as I see it you have two tables, two different classes representing a database entry (Tickets and Tickets3) and one pipeline to save them into the database.
And this single pipeline is called with results from both spiders. Why don't you use the spider variable in your process_item function to distinguish between your different elements? Spiders have a name so you have the tool to see which spider sent the item to process.