Subversion Repositories SmartDukaan

'''

Created on 14-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.DataCodeAccessor import *

from datastore.DataAccessor import *

from html2text.unescaping import *

class univercell_price(BaseSpider):

    def __init__(self):

       initialize_table()

       #UNIVERCELL_DOMAINNAME1 = "univercell1"   

       UNIVERCELL_DOMAINNAME1 = get_code_word("UNIVERCELL_DOMAINNAME1")

       self.domain_name = UNIVERCELL_DOMAINNAME1 

       # get urls from the database and append them in the list for crawling

       da = DataHelper()

       for pitem in da.get_all_univervendors():

            self.start_urls.append(pitem.v_site.strip())

    def start_requests(self):

        #for each request a referer has to be set

        listreq = []

        #UNIVERCELL_REFERER = "www.google.com/search"

        UNIVERCELL_REFERER = get_code_word("UNIVERCELL_REFERER")

        for url1 in self.start_urls:

            request = Request(url = str(url1), callback=self.parse)

            request.headers.setdefault("Referer", UNIVERCELL_REFERER)

            listreq.append(request)

        return listreq

    def parse(self, response):

        da = DataHelper()

        #VATPLUSTAX = 0

        #removelist is used for converting price to decimal format containing only numbers and '.'

        #UNIVERCELL_REMOVELIST = ["Rs",",","-","/"]

        #list separated by ';'

        UNIVERCELL_REMOVELIST = get_code_word("UNIVERCELL_REMOVELIST")

        UNIVERCELL_REMOVELIST = UNIVERCELL_REMOVELIST.split(';')

        hxs = HtmlXPathSelector(response)

        #UNIVERCELL_XPATH4 = '//td[@class="gray-border"]' 

        UNIVERCELL_XPATH4 = get_code_word("UNIVERCELL_XPATH4")

        sites = hxs.select(UNIVERCELL_XPATH4)

        items = []

        for site in sites:

            item = {}

            #UNIVERCELL_XPATH5 = './/tr[2]/td/a/text()'

            UNIVERCELL_XPATH5 = get_code_word("UNIVERCELL_XPATH5")

            item['title'] = site.select(UNIVERCELL_XPATH5)[0].extract()

            #UNIVERCELL_XPATH6 = './/tr[3]/th/label/text()'

            UNIVERCELL_XPATH6 = get_code_word("UNIVERCELL_XPATH6")

            item['price'] =site.select(UNIVERCELL_XPATH6)[0].extract()

            items.append(item)

        for i in items:

            str1 = str(i['title']).strip()

            amnt = i['price']

            if amnt != '':        

                for r in UNIVERCELL_REMOVELIST: 

                    while amnt.find(r) != -1:

                        amnt = amnt.replace(r, "")

            amnt = amnt.strip() 

            # 4% additional vat is there on the price

            UNIVERCELL_VATPLUSTAX = 4*int(amnt)/100

            pr = int(amnt) + UNIVERCELL_VATPLUSTAX 

            #adding model-name,quotedprice and finalprice

            da.add_new_univerphone(unescape(str1),amnt,pr) 

SPIDER = univercell_price()