Subversion Repositories SmartDukaan

'''
Created on 16-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 html2text.unescaping import *
from datastore.DataAccessor import *
from datastore.DataCodeAccessor import *



class indiaplaza_spider(BaseSpider):
   
    def __init__(self):
       initialize_table()
       da = DataHelper() 
       #INDIAPLAZA_DOMAINNAME = "indiaplaza"
       INDIAPLAZA_DOMAINNAME = get_code_word("INDIAPLAZA_DOMAINNAME")   
       self.domain_name = INDIAPLAZA_DOMAINNAME
       #INDIAPLAZA_CT = 18
       #INDIAPLAZA_CT = int(get_code_word("INDIAPLAZA_CT"))
       #INDIAPLAZA_NO = 1
       CT = int(da.get_extra_vars('indiaplaza_count'))
       NO = 1
       if CT>18:
           NO = CT
       #INDIAPLAZA_URL = "http://www.indiaplaza.in/mobile-phones-Mobiles-1.htm?PageNo="
       INDIAPLAZA_URL = get_code_word("INDIAPLAZA_URL")
       while(NO<=CT):
            url1 = INDIAPLAZA_URL + str(NO)
            self.start_urls.append(url1)
            NO=NO+1
        
    def start_requests(self):
        #adding entry for the supplier i.e its name and site
        #INDIAPLAZA_HOMEPAGE = "www.indiaplaza.com"
        INDIAPLAZA_HOMEPAGE = get_code_word("INDIAPLAZA_HOMEPAGE")
        da = DataHelper()
        da.add_supplier(self.domain_name, INDIAPLAZA_HOMEPAGE)
        listreq = []
        
        #for each request a referer has to be set
        #INDIAPLAZA_REFERER = "www.google.com/search"
        INDIAPLAZA_REFERER = get_code_word("INDIAPLAZA_REFERER")
        for url1 in self.start_urls:
            request = Request(url = str(url1), callback=self.parse, dont_filter=True)
            request.headers.setdefault("Referer", INDIAPLAZA_REFERER)
            listreq.append(request)
        return listreq
        
    def parse(self, response):
        da = DataHelper()
        #INDIAPLAZA_URL1 = "http://www.indiaplaza.in"
        INDIAPLAZA_URL1 = get_code_word("INDIAPLAZA_URL1")
        hxs = HtmlXPathSelector(response)
        #INDIAPLAZA_XPATH1 = '//tr/td/table[@id="browsesku"]'
        INDIAPLAZA_XPATH1 = get_code_word("INDIAPLAZA_XPATH1")
        phone_info = hxs.select(INDIAPLAZA_XPATH1)
        items = []
        #INDIAPLAZA_XPATH2 = './/div[@class="skuimg"]/a/@title'
        INDIAPLAZA_XPATH2 = get_code_word("INDIAPLAZA_XPATH2")
        #INDIAPLAZA_XPATH3 = './/div[@class="skuimg"]/a/@href'
        INDIAPLAZA_XPATH3 = get_code_word("INDIAPLAZA_XPATH3")
        if not phone_info:
            ct = int(da.get_extra_vars('indiaplaza_count'))
            if ct>18:
                fails = int(da.get_extra_vars('indiaplaza_fails'))
                fails = fails+1
                da.set_extra_vars('indiaplaza_fails',str(fails),'')
                if fails > 0:
                    da.set_extra_vars('indiaplaza_flag','FALSE','')
                da.set_extra_vars('indiaplaza_fails',str(fails),'')
        else:            
            for i in phone_info:
                item = {}
                item['title'] = i.select(INDIAPLAZA_XPATH2)[0].extract()
                item['url'] = i.select(INDIAPLAZA_XPATH3)[0].extract()
                items.append(item)
                
            for item in items:
                str2 = INDIAPLAZA_URL1 + str(item['url'])
                da.add_ipbasic(item['title'],unescape(str2))
                     
SPIDER = indiaplaza_spider()