You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Influenza_fund_linkage_system/app_test/stock_data_spider.py

66 lines
2.5 KiB

import os
import django
# 设置DJANGO_SETTINGS_MODULE环境变量
os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'test_Bootstrap.settings') # 将'test_Bootstrap'替换为你的项目名称
# 初始化Django设置
django.setup()
from app_test.user_agents_pool import agent_list
# 使用绝对导入来导入模型
from app_test.models import StockData
import pandas as pd
import requests
import time
import random
import re
from datetime import date
# from .models import StockData
class GetYaoqiGupiao(object):
user_Agent = random.choice(agent_list)
def __init__(self):
self.headers = {
'user-agent': self.user_Agent
}
def get_content(self,id,start_time,end_time):
url = f'https://q.stock.sohu.com/hisHq?code=cn_{id}&start={start_time}&end={end_time}&stat=1&order=D&period=d&callback=historySearchHandler&rt=jsonp&r=0.44942133747295787&0.3001773339487177'
data_all = []
# 爬取医药股票历史数据
res = requests.get(url, headers=self.headers)
# 设置3秒休眠时间
time.sleep(3)
html = res.content.decode("gb2312")
html = html.replace('historySearchHandler', '')
info_list = re.findall(r'\["202\d+.*?"\]', html, re.S)
datas = []
for info in info_list:
number_list = re.findall(r'"(.*?\d+.*?)"',info,re.S)
date = number_list[0]
shoupan = number_list[1]
datas.append([date, float(shoupan)])
df = pd.DataFrame(datas, columns=['date', '收盘'])
return df
def get_stock_data():
# 选用两只和布洛芬相关的股票数据
id_list = {'301075':'多瑞医药','301211':'亨迪药业'}
today = str(date.today()).replace('-', '')
getyaoqigupiao = GetYaoqiGupiao()
df1 = getyaoqigupiao.get_content('301075','20230201',today)
df2 = getyaoqigupiao.get_content('301211','20230201',today)
merged_df = df1.merge(df2,on = 'date',how = 'inner')#取交集
merged_df['stock_data'] = (merged_df['收盘_x']+merged_df['收盘_y'])/2
data = merged_df[['date','stock_data']]
data_list = data.values.tolist()
data_list = list(reversed(data_list))
print(data_list)
for data in data_list:
obj, created =StockData.objects.get_or_create(date=data[0], defaults={'shoupan': data[1]})
if created:
print(f"Added new record for date {data[0]} with infections {data[1]}")
else:
print(f"Record for date {data[0]} already exists.")
# get_stock_data()