如何按一个元素对JSON文件进行排序并从中删除重复项？

Question

我们假设使用 scrapy (python) 将 Billboard 100 艺术家提取到一个 json 文件中，并抓取每个页面中的前 5 个艺术家，并按字母顺序对它们进行排序并删除重复项。然后，将它们加载到新的谷歌工作表中。这是我到目前为止所做的：

import json
import scrapy
import datetime
from datetime import datetime
from datetime import timedelta, date


class BillboardWeeklySpider(scrapy.Spider):
    name = 'billboard-weekly'
    allowed_domains = ['www.billboard.com']
    start_urls = ['https://www.billboard.com/charts/artist-100/']

    def __init__(self):
        self.last_week_str = ""

    def parse(self, response):
        for i in range(4):
            string_date = response.css('#chart-date-picker::attr(data-date)').get()
            real_date = datetime.strptime(string_date, '%Y-%m-%d')
            day_delta = timedelta(weeks=1)
            last_week = real_date - i * day_delta
            self.last_week_str = last_week.strftime('%Y-%m-%d')
            next_page = f"https://www.billboard.com/charts/artist-100/{self.last_week_str}"
            if next_page:
                yield response.follow(next_page, callback=self.week_parse)

    def week_parse(self, response):

        for element in response.css('.o-chart-results-list-row-container'):
            name = element.css('#title-of-a-story::text').get()
            number = element.css(
                'span.c-label.a-font-primary-bold-l.u-font-size-32\@tablet.u-letter-spacing-0080\@tablet::text').get()
            clean_name = name.strip()
            clean_number = number.strip()
            if int(clean_number) > 5:
                break
            yield {
                'name': clean_name,
                'rank': clean_number,
                'date': response.url
            }

Answer 1

import scrapy
import json
import datetime
from datetime import datetime
from datetime import timedelta, date


class QuestionSpider(scrapy.Spider):
    global f
    name = 'billboard-weekly'
    allowed_domains = ['www.billboard.com']
    start_urls = ['https://www.billboard.com/charts/artist-100/']
    f = open("billboard.json","w")
    def __init__(self):
        self.last_week_str = ""

    def parse(self, response):
        for i in range(4):
            string_date = response.css('#chart-date-picker::attr(data-date)').get()
            real_date = datetime.strptime(string_date, '%Y-%m-%d')
            day_delta = timedelta(weeks=1)
            last_week = real_date - i * day_delta
            self.last_week_str = last_week.strftime('%Y-%m-%d')
            next_page = f"https://www.billboard.com/charts/artist-100/{self.last_week_str}"
            if next_page:
                yield response.follow(next_page, callback=self.week_parse)

    def week_parse(self, response):
        global f
        item = {}
        for element in response.css('.o-chart-results-list-row-container'):
            name = element.css('#title-of-a-story::text').get()
            number = element.css(
                'span.c-label.a-font-primary-bold-l.u-font-size-32\@tablet.u-letter-spacing-0080\@tablet::text').get()
            clean_name = name.strip()
            clean_number = number.strip()
            if int(clean_number) > 5:
                break
            item["name"] = clean_name
            item["rank"] = clean_number
            item["date"] = response.url
            jsonl = json.dumps(item)
            f.write(jsonl)
            print(item)