Scrapy 数据存储在 csv 个动态文件名的文件中

Question

I am trying to scrape data from different urls and I want to save the data in csv files with filename as the top level domain of the scraped url.

For example if am scraping data from https://www.example.com/event/abc then the saved file name should be example.com . The data is scraping in proper way, but I was not successful in saving the file with proper filename

Code

class myCrawler(CrawlSpider):
    name = 'testing'
    rotate_user_agent = True
    base_url=''
    start_urls = []
    allowed_domains = ''
    handle_httpstatus_list = [404,403]

    custom_settings = {
            # in order to reduce the risk of getting blocked
            'DOWNLOADER_MIDDLEWARES': {'sitescrapper.sitescrapper.middlewares.RotateUserAgentMiddleware': 400,
                                       'sitescrapper.sitescrapper.middlewares.ProjectDownloaderMiddleware': 543, },
            'COOKIES_ENABLED': False,
            'CONCURRENT_REQUESTS': 6,
            'DOWNLOAD_DELAY': 2,
            'DEPTH_LIMIT' : 1,
            'CELERYD_MAX_TASKS_PER_CHILD' : 1,

            # Duplicates pipeline
            'ITEM_PIPELINES': {'sitescrapper.sitescrapper.pipelines.DuplicatesPipeline': 300},

            # In order to create a CSV file:
            'FEEDS': {'%(allowed_domains).csv': {'format': 'csv'}},
        }
    def __init__(self, category='', **kwargs):
        self.base_url = category
        self.allowed_domains = ['.'.join(urlparse(self.base_url).netloc.split('.')[-2:])]
        self.start_urls.append(self.base_url)
        print(f"Base url is {self.base_url} and allowed domain is {self.allowed_domains}")  

        self.rules = (
            Rule(
                LinkExtractor(allow_domains=self.allowed_domains),
                process_links=process_links,
                callback='parse_item',
                follow=True
            ),
        )   
        super().__init__(**kwargs)

Thanks in advance

Answer 1

We can specify the location of download and set the filename dynamically by using

'FEEDS': {"./scraped_urls/%(file_name)s" : {"format": "csv"}},

in custom_settings

Answer 2

If you're using the split("\") function on your url?