Python爬虫爬取王者荣耀英雄信息并保存到图数据库

爬取信息说明

英雄名称
英雄类型
英雄包含的所有皮肤名称

创建英雄类型节点

王者荣耀官方给出的英雄类型是以下几种：
在这里插入图片描述
直接准备好英雄类型词典

hero_type_dict = ['战士', '法师', '坦克', '刺客', '射手', '辅助'
]

添加到图数据库中

def create_hero_type_node():for hero_type in hero_type_dict:cypher = "MERGE (n:HeroType{label: '" + hero_type + "'})"graph.run(cypher).data()print('创建英雄类型节点成功')

创建英雄信息节点

获取英雄信息

def get_hero_info_list():# 英雄的全部信息的urlhero_info = 'https://pvp.qq.com/web201605/js/herolist.json'# 获取英雄的全部信息response = requests.get(hero_info)# 转为字典格式hero_info_dict = json.loads(response.text)return hero_info_dict

打印的内容如下：
在这里插入图片描述
这里需要注意的是，部分英雄包含两个英雄类别。

保存英雄信息

def create_hero_node():hero_info_dict = get_hero_info_list()# 1战士 2法师 3坦克 4刺客 5射手 6辅助for hero in hero_info_dict:# print(hero)# print(str(hero.get('cname')) + '===' + str(hero_type[hero.get('hero_type')-1]) + '===' + str(hero.get('skin_name')))hero_type_list = [str(hero_type_dict[hero.get('hero_type') - 1])]if '|' in str(hero.get('skin_name')):skin_name_list = hero.get('skin_name').split('|')else:skin_name_list = [hero.get('skin_name')]if 'hero_type2' in str(hero):hero_type_list.append(str(hero_type_dict[hero.get('hero_type2') - 1]))# 创建英雄信息节点hero_cypher = "MERGE (n:Hero{label: '" + str(hero.get('cname')) + "'})"graph.run(hero_cypher).data()# 创建英雄->类型关系for hero_type in hero_type_list:cypher_rel = "MATCH(h:Hero{label:'" + str(hero.get('cname')) + "'}),(t:HeroType{label:'" + hero_type + "'}) MERGE (h)-[r:类型]->(t) RETURN h,r,t"graph.run(cypher_rel).data()for skin_name in skin_name_list:# 创建英雄皮肤节点cypher = "MERGE (n:Skin{label:'" + skin_name + "'})"graph.run(cypher).data()# 创建英雄->皮肤关系cypher_rel = "MATCH(h:Hero{label:'" + str(hero.get('cname')) + "'}),(s:Skin{label:'" + skin_name + "'}) MERGE (h)-[r:皮肤]->(s) RETURN h,r,s"graph.run(cypher_rel).data()print(str(hero.get('cname')) + '===' + str(hero_type_list) + '===' + str(skin_name_list))

完整代码

import jsonimport requests
from bs4 import BeautifulSoup
from py2neo import Graph, RelationshipMatcher, NodeMatcherfrom dict import hero_type_dicturl = "bolt://localhost:7687"
username = "neo4j"
password = 'Suns3535'
graph = Graph(url, auth=(username, password), name="wzry")
node_matcher = NodeMatcher(graph=graph)
relationship_matcher = RelationshipMatcher(graph=graph)def get_hero_info_list():# 英雄的全部信息的urlhero_info = 'https://pvp.qq.com/web201605/js/herolist.json'# 获取英雄的全部信息response = requests.get(hero_info)# 转为字典格式hero_info_dict = json.loads(response.text)return hero_info_dictdef create_hero_type_node():for hero_type in hero_type_dict:cypher = "MERGE (n:HeroType{label: '" + hero_type + "'})"graph.run(cypher).data()print('创建英雄类型节点成功')def create_hero_node():hero_info_dict = get_hero_info_list()# 1战士 2法师 3坦克 4刺客 5射手 6辅助for hero in hero_info_dict:# print(hero)# print(str(hero.get('cname')) + '===' + str(hero_type[hero.get('hero_type')-1]) + '===' + str(hero.get('skin_name')))hero_type_list = [str(hero_type_dict[hero.get('hero_type') - 1])]if '|' in str(hero.get('skin_name')):skin_name_list = hero.get('skin_name').split('|')else:skin_name_list = [hero.get('skin_name')]if 'hero_type2' in str(hero):hero_type_list.append(str(hero_type_dict[hero.get('hero_type2') - 1]))# 创建英雄信息节点hero_cypher = "MERGE (n:Hero{label: '" + str(hero.get('cname')) + "'})"graph.run(hero_cypher).data()# 创建英雄->类型关系for hero_type in hero_type_list:cypher_rel = "MATCH(h:Hero{label:'" + str(hero.get('cname')) + "'}),(t:HeroType{label:'" + hero_type + "'}) MERGE (h)-[r:类型]->(t) RETURN h,r,t"graph.run(cypher_rel).data()for skin_name in skin_name_list:# 创建英雄皮肤节点cypher = "MERGE (n:Skin{label:'" + skin_name + "'})"graph.run(cypher).data()# 创建英雄->皮肤关系cypher_rel = "MATCH(h:Hero{label:'" + str(hero.get('cname')) + "'}),(s:Skin{label:'" + skin_name + "'}) MERGE (h)-[r:皮肤]->(s) RETURN h,r,s"graph.run(cypher_rel).data()print(str(hero.get('cname')) + '===' + str(hero_type_list) + '===' + str(skin_name_list))# 创建英雄类型节点
create_hero_type_node()
# 创建英雄信息
create_hero_node()