Subversion Repositories SmartDukaan

'''
Created on 11-May-2010

@author: gaurav
'''


from scrapy.spider import BaseSpider
from scrapy.selector import HtmlXPathSelector
from scrapy.http import Request

from demo.items import DemoItem
from scrapy.contrib.spidermiddleware import referer
from scrapy.http.headers import Headers
from scrapy.http.request.form import FormRequest
from scrapy.log import msg
from scrapy.http.response import Response
#from datastore.DataAccessor import add_new_phone

from datastore import DataAccessor
from datastore.DataAccessor import DataHelper

class scrapy_price(BaseSpider):
    
    def __init__(self):
       self.domain_name = "infibeam.com"
       da = DataHelper()
       for pitem in da.get_all_vendors():
            self.start_urls.append(pitem.v_url.strip())
    
    def start_requests(self):
        listreq = []
        for url1 in self.start_urls:
            request = Request(url = url1, callback=self.parse)
            request.headers.setdefault("Referer", "www.google.com/search")
            listreq.append(request)
        return listreq
       
    def parse(self, response):
        hxs = HtmlXPathSelector(response)
        sites = hxs.select('//p[@class="box"]')
        items = []
        for site in sites:
            item = {}
            tmp = site.select('.//a[@class="nocol"]/@name')
            item['title'] = tmp[0].extract()
            psite = site.select(".//a[3][@href]/@href")[0].extract()
            item['link'] = psite
            items.append(item)
        str1 = "http://www.infibeam.com"
        da = DataHelper() 
        da.set_all_crawled(False)     
        for i in items:
            str2 = str(i['link'])
            if str(i['link']).find("http://www.infibeam.com") == -1:
                str2 = str1 + str2  
            print "name"
            print i['title']
            print "site" 
            print str2
            da.add_new_phone(str2,i['title'],"vendor") 
            
        #lt = len(da.get_all_phones())  
        #print "length" + str(lt)
        #for ph in da.get_all_phones():
         #   print ph
            
        f = open('/home/gaurav/twopassscrapy/pricelinks.txt', 'w')
        for i in items:
            f.write(i['title'])
            f.write("\n")
            f.write(i['link'])
            f.write("\n")
        f.close()    
        
SPIDER = scrapy_price()