app.py

from seleniumwire import webdriver
from selenium.webdriver.chrome.options import Options
from fastapi import FastAPI, Request
import uvicorn
import time
import json
from urllib.parse import unquote, urlparse

app = FastAPI()

def convert_cookies_to_dict(cookies):
    cookies = dict([l.split("=", 1) for l in cookies.split("; ")])
    return cookies

def get_root_domain(url):
    # 解析URL
    parsed_url = urlparse(url)
    # 获取域名部分
    domain = parsed_url.netloc
    
    # 分割域名部分以获取根域名
    # 假设根域名是域名的最后两个部分
    parts = domain.split('.')
    if len(parts) > 1:
        # 返回根域名部分
        return '.'.join(parts[-2:])
    else:
        # 如果域名部分少于两个部分，返回整个域名
        return domain
        
@app.get("/")
def main():
    return {"code": 200,"msg":"Success"}
    
@app.get("/chrome")
def chrome(url:str=None,wait:int=5,header:str=None,cookie:str=None):

    # 必须有目标url
    if type(url) == str:
        target_url = unquote(url)
    else:
        return {"code": 500,"msg":"No target URL"}

    # 等待时间必须在 0 到 30 之间
    if wait in range(0, 31):
        wait_time = wait
    else:
        return {"code": 500,"msg":"The waiting time must be between 0 and 30"}

    header_array = {}

    # header可以覆写，但必须传入json
    try:
        if type(header) == str:
            header_array.update(json.loads(unquote(header)))
    except Exception as e:
        return {"code": 500,"msg":"The header field is not JSON"}
        
    # 如果输入了cookie
    if type(cookie) == str:
        header_array.update({"cookie":unquote(cookie)})

    if 'cookie' in header_array:
        cookie_array = header_array['cookie']
        del header_array['cookie']
        cookie_domain = f'.{get_root_domain(target_url)}'
        for key, value in cookie_array.items():
            driver.execute_script(f'document.cookie = "{key}={value}; path=/; domain={cookie_domain}; expires=Thu, 01 Jan 2050 00:00:00 GMT;');

    options = Options()
    options.add_argument('--headless')
    
    driver = webdriver.Chrome(options=options)

    driver.header_overrides = header_array
    
    driver.get(target_url)

    print(driver.page_source)

    if wait_time > 0:
        time.sleep(wait_time)

    # 获取当前URL
    current_url = driver.current_url
    
    # 获取页面源代码
    page_source = driver.page_source
    
    # 获取cookie
    cookies = driver.get_cookies()

    # 是否有跳转过
    is_jump = (target_url != current_url)
    
    data = {
        "url": current_url,
        "page_source": page_source,
        "cookies": cookies,
        "is_jump": is_jump
    }
    
    driver.quit()
    return {"code": 200,"data":data}

if __name__ == '__main__':
    uvicorn.run(app='app:app', host="0.0.0.0", port=7860)