Private-projects / Scrapy_project / fleurop / pipelines.py
pipelines.py
Raw
# Define your item pipelines here
#
# Don't forget to add your pipeline to the ITEM_PIPELINES setting
# See: https://docs.scrapy.org/en/latest/topics/item-pipeline.html


# useful for handling different item types with a single interface
from itemadapter import ItemAdapter
from bs4 import BeautifulSoup

class FleuropPipeline:

    def process_item(self, item, spider):
        adapter = ItemAdapter(item)

        if adapter.get('delivery_period'):
            adapter['delivery_period'] = adapter.get('delivery_period').strip()[10:]
        else:
            adapter['delivery_period'] = ''

        columns = ['product_description', 'delivery_info']
        for column in columns:
            col = adapter.get(column)
            if col:
                adapter[column] = ' '.join(list(BeautifulSoup(col, "html5lib").stripped_strings)).replace('\n', ' ')
            else:
                adapter[column] = ''

        return item