标题王者农药英雄技能爬取

import requests,json
from lxml import etree
import pandas as pdherolist_json = requests.get('https://pvp.qq.com/web201605/js/herolist.json').text
herolist = json.loads(herolist_json)
headers = {'User-Agent':'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/81.0.4044.138 Safari/537.36'}
print(herolist)
heroes_info = []def herotype_text(hero):if hero['hero_type'] == 1:herotype = '战士'elif hero['hero_type'] == 2:herotype = '法师'elif hero['hero_type'] == 3:herotype = '坦克'elif hero['hero_type'] == 4:herotype = '刺客'elif hero['hero_type'] == 5:herotype = '射手'elif hero['hero_type'] == 6:herotype = '辅助'return herotypedef fromturltotype(p):pt1 = int(partners[0][0:3])pt2 = int(partners[1][0:3])for hero in herolist:if hero['ename'] == pt1:pt1_num = hero['hero_type']pt1_type = herotype_text(hero)if hero['ename'] == pt2:pt2_num = hero['hero_type']pt2_type = herotype_text(hero)return pt1_type,pt2_typefor hero in herolist:herolink = "https://pvp.qq.com/web201605/herodetail/"+str(hero['ename'])+".shtml"page =  requests.get(herolink,headers=headers)page_tree = etree.HTML(page.content)name = page_tree.xpath("//h2[@class='cover-name']")[0].textlife = page_tree.xpath('//ul[@class="cover-list"]/li[1]/span/i/@style')[0][6:-1]#不可以直接获取绝对路径xpath，会报错hurt = page_tree.xpath('//ul[@class="cover-list"]/li[2]/span/i/@style')[0][6:-1]skill = page_tree.xpath('//ul[@class="cover-list"]/li[3]/span/i/@style')[0][6:-1]    difficulty = page_tree.xpath('//ul[@class="cover-list"]/li[4]/span/i/@style')[0][6:-1]herotype = herotype_text(hero)partners = page_tree.xpath('//div[@class="hero-info-box"]/div[1]/div[1]/div[2]//a/@href') pt1,pt2 = fromturltotype(partners)pt = [pt1,pt2,herotype]hero_info = [name,herotype,pt,life,hurt,skill,difficulty]heroes_info.append(hero_info) 
print(heroes_info)
heroes_df = pd.DataFrame(heroes_info)
heroes_df.columns = ['英雄名','类型','搭档英雄类型','生存能力','攻击伤害','技能效果','上手难度']
heroes_df.sort_values(by='生存能力',ascending=False)  #这个方法有点问题,因为目前数据都不是数值数据类型，所以排序的时候会出现20反而比100排得靠前的情况
heroes_df.dtypes      #  查看当前数据类型
heroes_df[['生存能力','攻击伤害','技能效果','上手难度']] = heroes_df[['生存能力','攻击伤害','技能效果','上手难度']].apply(pd.to_numeric)#进行数据类型转换
heroes_df.dtypes     # 再看一次数据类型 ，已经变成浮点型了heroes_df.sort_values(by='生存能力',ascending=False)   #  重新排序，返回的是正确排序
# 值计数
heroes_df['搭档英雄类型'].astype('str').value_counts()# 画图
import pygal
line_chart = pygal.Bar()
line_chart.title = '王者荣耀英雄'
line_chart.x_labels = heroes_df['英雄名']
line_chart.add('生存能力', list(heroes_df['生存能力']))
line_chart.add('攻击伤害', list(heroes_df['攻击伤害']))
line_chart.add('技能效果', list(heroes_df['技能效果']))
line_chart.add('上手难度', list(heroes_df['上手难度']))
line_chart.render_to_file('hero.svg')###############heroes_df.to_csv('王者荣耀英雄数据.csv',encoding ='gbk',header = True, index = None)