books_spider.py

from scrapy import Spider
import re
from items.BookItem import BookItem
from database.connectors.BookConnector import BookConnector


class BooksSpider(Spider, BookConnector):
    name = 'books'
    start_urls = ['http://books.toscrape.com/']
    page = 1

    def parse(self, response):
        self.logger.debug('Current page: {}'.format(self.page))

        # follow links to book pages
        for idx, href in enumerate(response.css('div.image_container a::attr(href)')):  # TODO delete enumerate
            yield response.follow(href, self.parse_book, meta={'idx': idx})

        # pagination
        next_page = response.css('li.next a::attr(href)').get()
        if next_page is not None:
            self.page += 1
            yield response.follow(next_page, callback=self.parse)

    def parse_book(self, response):
        self.logger.debug('Index book in page: {}'.format(response.meta.get('idx')))

        book = BookItem()

        table_data = response.css('table td::text').getall()
        book.update({
            'title': response.css('div.product_main h1::text').get(),
            'description': response.xpath("//*[@id='product_description']/following::p/text()").get(),
            'image_urls': [response.urljoin(response.css('div.active img::attr(src)').get())],
            'rating': self._get_rating(response.css('p.star-rating::attr(class)').get().split(' ')),
            'upc': table_data[0],
            'product_type': table_data[1],
            'price_excl_tax': table_data[2][1:],
            'price_incl_tax': table_data[3][1:],
            'tax': table_data[4][1:],
            'in_stock': int(re.search(r'\d+', table_data[5]).group()),
            'count_reviews': int(table_data[6]),
            'currency_type': table_data[2][0],
            'category': response.css('ul.breadcrumb li:nth-child(3) a::text').get()
        })
        yield book

    def _get_rating(self, class_all):
        if 'One' in class_all:
            return 1
        if 'Two' in class_all:
            return 2
        if 'Three' in class_all:
            return 3
        if 'Four' in class_all:
            return 4
        if 'Five' in class_all:
            return 5