Subversion Repositories SmartDukaan

'''

Created on 27-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 naaptol_spider(BaseSpider):

    def __init__(self):

       initialize_table() 

       #NAAPTOL_DOMAINNAME = "naaptol"   

       NAAPTOL_DOMAINNAME = get_code_word("NAAPTOL_DOMAINNAME")

       self.domain_name = NAAPTOL_DOMAINNAME 

       #NAAPTOL_URL = "http://www.naaptol.com/sitemap.xml"

       NAAPTOL_URL = get_code_word("NAAPTOL_URL")

       self.start_urls.append(NAAPTOL_URL)

    def start_requests(self):

        #adding entry for the supplier i.e its name and site

        #NAAPTOL_HOMEPAGE = "http://www.naaptol.com"

        NAAPTOL_HOMEPAGE = get_code_word("NAAPTOL_HOMEPAGE")

        da = DataHelper()

        da.add_supplier(self.domain_name, NAAPTOL_HOMEPAGE)

        listreq = []

        #for each request a referer has to be set

        #NAAPTOL_REFERER = "http://www.google.com"

        NAAPTOL_REFERER = get_code_word("NAAPTOL_REFERER")

        for url1 in self.start_urls:

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

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

            listreq.append(request)

        return listreq

    def parse(self, response):

        da = DataHelper()

        hxs = HtmlXPathSelector(response)

        #NAAPTOL_XPATH1 = '//url/loc/text()'

        NAAPTOL_XPATH1 = get_code_word("NAAPTOL_XPATH1")

        phone_urls = hxs.select(NAAPTOL_XPATH1)

        #elements in chk_list are specific to this site for determining valid sites 

        #NAAPTOL_CHKLIST1 = ["mobile_phones/pdas_and_smartphones" ,"mobile_phones/gsm_handsets" ,"mobile_phones/cdma_handsets"]

        #list separeated by ';'

        NAAPTOL_CHKLIST1 = get_code_word("NAAPTOL_CHKLIST1")

        NAAPTOL_CHKLIST1 = NAAPTOL_CHKLIST1.split(';')

        for i in phone_urls:

            site = i.extract()

            site = unescape(site)

            pos1 = pos2 = pos3 = 0

            temp =""

            # temp contains string b/w 2nd last and 3rd last slash(/)

            pos1 = site.rfind('/')

            if pos1 != -1:

                pos2 = site.rfind('/',0,pos1-1)

            if pos2 != -1:    

                pos3 = site.rfind('/',0,pos2-1)

            if pos3 > 0:

                temp = site[pos3+1:pos1]

            for c in NAAPTOL_CHKLIST1:

                if temp == c:

                    da.add_naaptolurl(site)           

SPIDER = naaptol_spider()