using scrapy to scrape asp.net website with javascript buttons and ajax requests

Question

I'd been trying to scrape some date from as asp.net website, the start page should be the following one: http://www.e3050.com/Items.aspx?cat=SON

First, I want to display 50 item per page (from the select element) Second, I want to paginate through pages.

I tried the following code for 50 items per page, but didn't work:

start_urls = ["http://www.e3050.com/Items.aspx?cat=SON"]    
def parse(self, response):
        requests = []
        hxs = HtmlXPathSelector(response)

        # Check if there's more than 1 page         
        if len(hxs.select('//span[@id="ctl00_ctl00_ContentPlaceHolder1_ItemListPlaceHolder_lbl_PageSize"]/text()').extract()) > 0:
            # Get last page number
            last_page = hxs.select('//span[@id="ctl00_ctl00_ContentPlaceHolder1_ItemListPlaceHolder_lbl_PageSize"]/text()').extract()[0]
            i = 1

            # preparing requests for each page
            while i < (int(last_page) / 5)  + 1:
                requests.append(Request("http://www.e3050.com/Items.aspx?cat=SON", callback=self.parse_product))
                i +=1 

            # posting form date (50 items and next page button)
            requests.append(FormRequest.from_response(
                    response,
                    formdata={'ctl00$ctl00$ContentPlaceHolder1$ItemListPlaceHolder$pagesddl':'50',
                              '__EVENTTARGET':'ctl00$ctl00$ContentPlaceHolder1$ItemListPlaceHolder$pager1$ctl00$ctl01'},
                    callback=self.parse_product,
                    dont_click=True
                )
            )

            for request in requests:
                yield request

Answer 1

Check out this here is an exact solution..

in parse method selecting 50 products per page

in page_rs_50 handled pagination

start_urls = ['http://www.e3050.com/Items.aspx?cat=SON']
pro_urls = [] # all product Urls

def parse(self, response): # select 50 products on each page
    yield FormRequest.from_response(response,
        formdata={'ctl00$ctl00$ContentPlaceHolder1$ItemListPlaceHolder$pagesddl': '50',
                  'ctl00$ctl00$ContentPlaceHolder1$ItemListPlaceHolder$sortddl': 'Price(ASC)'},
        meta={'curr': 1, 'total': 0, 'flag': True},
        dont_click=True,
        callback=self.page_rs_50)

def page_rs_50(self, response): # paginate the pages
    hxs = HtmlXPathSelector(response)
    curr = int(response.request.meta['curr'])
    total = int(response.request.meta['total'])
    flag = response.request.meta['flag']
    self.pro_urls.extend(hxs.select(
        "//td[@class='name']//a[contains(@id,'ctl00_ctl00_ContentPlaceHolder1_ItemListPlaceHolder_itemslv_ctrl')]/@href"
    ).extract())
    if flag:
        total = hxs.select(
            "//span[@id='ctl00_ctl00_ContentPlaceHolder1_ItemListPlaceHolder_lbl_pagesizeBtm']/text()").re('\d+')[0]
    if curr < total:
        curr += 1
        yield FormRequest.from_response(response,
            formdata={'ctl00$ctl00$ContentPlaceHolder1$ItemListPlaceHolder$pagesddl': '50',
                      'ctl00$ctl00$ContentPlaceHolder1$ItemListPlaceHolder$sortddl': 'Price(ASC)',
                      'ctl00$ctl00$ScriptManager1': 'ctl00$ctl00$ScriptManager1|ctl00$ctl00$ContentPlaceHolder1$ItemListPlaceHolder$pager1$ctl00$ctl01'
                , '__EVENTTARGET': 'ctl00$ctl00$ContentPlaceHolder1$ItemListPlaceHolder$pager1$ctl00$ctl01',
                      'ctl00$ctl00$ContentPlaceHolder1$ItemListPlaceHolder$hfVSFileName': hxs.select(
                          ".//input[@id='ctl00_ctl00_ContentPlaceHolder1_ItemListPlaceHolder_hfVSFileName']/@value").extract()[
                                                                                          0]},
            meta={'curr': curr, 'total': total, 'flag': False},
            dont_click=True,
            callback=self.page_rs_50
        )
    else:
        for pro in self.pro_urls:
            yield Request("http://www.e3050.com/%s" % pro,
                callback=self.parse_product)


def parse_product(self, response):
    pass
    #TODO Implementation Required For Parsing

Answer 2

I did not extensively research your code, but i see something strange:

        # Get last page number
        last_page = hxs.select('//span[@id="ctl00_ctl00_ContentPlaceHolder1_ItemListPlaceHolder_lbl_PageSize"]/text()').extract()[0]
        i = 1

        # preparing requests for each page
        while i < (int(last_page) / 5)  + 1:
            requests.append(Request("http://www.e3050.com/Items.aspx?cat=SON", callback=self.parse_product))
            i +=1 

First, instead of these manipulations with i, you can do:

        for i in xrange(1, last_page // 5 + 1):

Then you do:

            requests.append(Request("http://www.e3050.com/Items.aspx?cat=SON", callback=self.parse_product))

Are you creating many requests to the same URL?