Develop Reference

Pass variable to test.py in spider folder using scrapy

I'm using Scrapy. The following is the code for test.py in spider folder.
from scrapy.spider import BaseSpider
from scrapy.selector import HtmlXPathSelector
from craigslist_sample.items import CraigslistSampleItem
class MySpider(BaseSpider):
name = "craig"
allowed_domains = ["craigslist.org"]
start_urls = ["http://seattle.craigslist.org/npo/"]
def parse(self, response):
hxs = HtmlXPathSelector(response)
titles = hxs.select("//span[#class='pl']")
items = []
for titles in titles:
item = CraigslistSampleItem()
item["title"] = titles.select("a/text()").extract()
item["link"] = titles.select("a/#href").extract()
items.append(item)
return items
Essentially, I want to iterate my url list and pass url into MySpider class for start_ulrs. Could you anyone give me suggestion on how to make this?

Instead of having "statically defined" start_urls you need to override start_requests() method:
from scrapy.http import Request
class MySpider(BaseSpider):
name = "craig"
allowed_domains = ["craigslist.org"]
def start_requests(self)
list_of_urls = [...] # reading urls from a text file, for example
for url in list_of_urls:
yield Request(url)
def parse(self, response):
...

Scrapy only scraping first result of each page

I'm currently trying to run the following code but it keeps scraping only the first result of each page. Any idea what the issue may be?
from scrapy.contrib.spiders import CrawlSpider, Rule
from scrapy.contrib.linkextractors.sgml import SgmlLinkExtractor
from scrapy.selector import HtmlXPathSelector
from firstproject.items import xyz123Item
import urlparse
from scrapy.http.request import Request
class MySpider(CrawlSpider):
name = "xyz123"
allowed_domains = ["www.xyz123.com.au"]
start_urls = ["http://www.xyz123.com.au/",]
rules = (Rule (SgmlLinkExtractor(allow=("",),restrict_xpaths=('//*[#id="1234headerPagination_hlNextLink"]',))
, callback="parse_xyz", follow=True),
)
def parse_xyz(self, response):
hxs = HtmlXPathSelector(response)
xyz = hxs.select('//div[#id="1234SearchResults"]//div/h2')
items = []
for xyz in xyz:
item = xyz123Item()
item ["title"] = xyz.select('a/text()').extract()[0]
item ["link"] = xyz.select('a/#href').extract()[0]
items.append(item)
return items
The Basespider version works well scraping ALL the required data on the first page:
from scrapy.spider import BaseSpider
from scrapy.selector import HtmlXPathSelector
from firstproject.items import xyz123
class MySpider(BaseSpider):
name = "xyz123test"
allowed_domains = ["xyz123.com.au"]
start_urls = ["http://www.xyz123.com.au/"]
def parse(self, response):
hxs = HtmlXPathSelector(response)
titles = hxs.select('//div[#id="1234SearchResults"]//div/h2')
items = []
for titles in titles:
item = xyz123Item()
item ["title"] = titles.select("a/text()").extract()
item ["link"] = titles.select("a/#href").extract()
items.append(item)
return items
Sorry for the censoring. I had to censor the website for privacy reasons.
The first code crawls through the pages well the way I'd like it to crawl, however it only pulls the first item title and link. NOTE: The XPath of the first title using "inspect element" in google is:
//*[#id="xyz123SearchResults"]/div[1]/h2/a,
second is //*[#id="xyz123SearchResults"]/div[2]/h2/a
third is //*[#id="xyz123SearchResults"]/div[3]/h2/a etc.
I'm not sure if the div[n] bit is what's killing it. I'm hoping it's an easy fix.
Thanks

for xyz in xyz:
item = xyz123Item()
item ["title"] = xyz.select('a/text()').extract()[0]
item ["link"] = xyz.select('a/#href').extract()[0]
items.append(item)
return items
Are you sure about the indentation of the return items ? It should be one less.

Scrapy project, scraping a schedule

So I'm trying to scrape the schedule at this page.. http://stats.swehockey.se/ScheduleAndResults/Schedule/3940
..with this code.
from scrapy.spider import BaseSpider
from scrapy.selector import HtmlXPathSelector
class SchemaSpider(BaseSpider):
name = "schema"
allowed_domains = ["http://stats.swehockey.se/"]
start_urls = [
"http://stats.swehockey.se/ScheduleAndResults/Schedule/3940"
]
def parse(self, response):
hxs = HtmlXPathSelector(response)
rows = hxs.select('//table[#class="tblContent"]/tbody/tr')
for row in rows:
date = row.select('/td[1]/div/span/text()').extract()
teams = row.select('/td[2]/text()').extract()
print date, teams
But I can't get it to work. What am I doing wrong? I've been trying to figure out myself for a couple of hours now but I have no idea why my XPath doesn't work properly.

Two problems:
tbody is a tag that is added by modern browsers. Scrapy simply doesn't see it in the html.
xpaths for data and teams weren't right: you should use relative xpath (.//), also td indexes was wrong, should be 2 and 3 instead of 1 and 2
Here's the whole code with some mofidications (working):
from scrapy.item import Item, Field
from scrapy.spider import BaseSpider
from scrapy.selector import HtmlXPathSelector
class SchemaItem(Item):
date = Field()
teams = Field()
class SchemaSpider(BaseSpider):
name = "schema"
allowed_domains = ["http://stats.swehockey.se/"]
start_urls = [
"http://stats.swehockey.se/ScheduleAndResults/Schedule/3940"
]
def parse(self, response):
hxs = HtmlXPathSelector(response)
rows = hxs.select('//table[#class="tblContent"]/tr')
for row in rows:
item = SchemaItem()
item['date'] = row.select('.//td[2]/div/span/text()').extract()
item['teams'] = row.select('.//td[3]/text()').extract()
yield item
Hope that helps.

Iterating through sites with Python Scrapy

How do I iterate through sites with Scrapy? I'd like to extract the body of all sites that match http://www.saylor.org/site/syllabus.php?cid=NUMBER, where NUMBER is 1 through 400 or so.
I've written this spider:
from scrapy.contrib.spiders import CrawlSpider, Rule
from scrapy.contrib.linkextractors.sgml import SgmlLinkExtractor
from scrapy.selector import HtmlXPathSelector
from syllabi.items import SyllabiItem
class SyllabiSpider(CrawlSpider):
name = 'saylor'
allowed_domains = ['saylor.org']
start_urls = ['http://www.saylor.org/site/syllabus.php?cid=']
rules = [Rule(SgmlLinkExtractor(allow=['\d+']), 'parse_syllabi')]
def parse_syllabi(self, response):
x = HtmlXPathSelector(response)
syllabi = SyllabiItem()
syllabi['url'] = response.url
syllabi['body'] = x.select("/html/body/text()").extract()
return syllabi
But it doesn't work. I understand it's looking for links in that start_url, which is not really what I want it to do. I want to iterate through the sites. Make sense?
Thanks for the help.

Try this:
from scrapy.spider import BaseSpider
from scrapy.http import Request
from syllabi.items import SyllabiItem
class SyllabiSpider(BaseSpider):
name = 'saylor'
allowed_domains = ['saylor.org']
max_cid = 400
def start_requests(self):
for i in range(self.max_cid):
yield Request('http://www.saylor.org/site/syllabus.php?cid=%d' % i,
callback=self.parse_syllabi)
def parse_syllabi(self, response):
syllabi = SyllabiItem()
syllabi['url'] = response.url
syllabi['body'] = response.body
return syllabi

scrapy: newbie attempting to debug code

Total newbie, trying to get scrapy to read a list of urls from csv and return the items in a csv.
Need some help to figure out where I'm going wrong here:
Spider code:
from scrapy.spider import BaseSpider
from scrapy.selector import HtmlXPathSelector
import random
class incyspider(BaseSpider):
name = "incyspider"
def __init__(self):
super(incyspider, self).__init__()
domain_name = "incyspider.co.uk"
f = open("urls.csv")
start_urls = [url.strip() for url in f.readlines()]
f.close
def parse(self, response):
hxs = HtmlXPathSelector(response)
sites = hxs.select('//div[#class="Product"]')
items = []
for site in sites:
item['title'] = hxs.select('//div[#class="Name"]/node()').extract()
item['hlink'] = hxs.select('//div[#class="Price"]/node()').extract()
item['price'] = hxs.select('//div[#class="Codes"]/node()').extract()
items.append(item)
return items
SPIDER = incyspider()
Here's the items.py code:
from scrapy.item import Item, Field
class incyspider(Item):
# define the fields for your item here like:
# name = Field()
title = Field()
hlink = Field()
price = Field()
pass
To run, I'm using
scrapy crawl incyspider -o items.csv -t csv
I would seriously appreciate any pointers.

I'm not exactly sure but after a quick look at your code I would say that at least you need to replace this line
sites = hxs.select('//div[#class="Product"]')
by this line
sites = hxs.select('//div[#class="Product"]').extract()

As a first punt at answering this, your spider code is missing an import for your incyspider item class. Also you're not creating an instance of any kind of item to store the title/hlink/price info, so the items.append(item) line might complain.
Since your spider is also called incyspider, you should rename the item to be something like incyspiderItem and then add the following line to your spider code
from scrapy.spider import BaseSpider
from scrapy.selector import HtmlXPathSelector
import random
from incyspider.items import incyspiderItem
class incyspider(BaseSpider):
name = "incyspider"
def __init__(self):
super(incyspider, self).__init__()
domain_name = "incyspider.co.uk"
f = open("urls.csv")
start_urls = [url.strip() for url in f.readlines()]
f.close
def parse(self, response):
hxs = HtmlXPathSelector(response)
sites = hxs.select('//div[#class="Product"]')
items = []
for site in sites:
item = incyspiderItem()
item['title'] = hxs.select('//div[#class="Name"]/node()').extract()
item['hlink'] = hxs.select('//div[#class="Price"]/node()').extract()
item['price'] = hxs.select('//div[#class="Codes"]/node()').extract()
items.append(item)
return items
If I'm wrong, then please edit the question to explain how you know there is a problem with the code eg: is the expected output different to the actual output? If so, how?